ISSN 1155-5645

Pediatric
Anesthesia
Volume 28 Number 1 2018

@medicinejournal
An international journal published by

Affiliated with the Association of Paediatric Anaesthetists of Great

Britain and Ireland, the European Society of Paediatric Anaesthesia,
Association des Anesthésistes Réanimateurs Pédiatriques
d’Expression Française (the French-Language Society of Paediatric
Anaesthesia), the Canadian Pediatric Anesthesia Society, the
Chinese Society of Pediatric Anesthesiology, the Italian Society of
Paediatric Anaesthesia, the Japanese Society of Pediatric
Anesthesiology, the Paediatric Anaesthesia Community of South
Africa, the Scientific Working Group of Pediatric Anesthesia of the
German Society of Anaesthesiology, the Society of Pediatric
Anesthesia and the Society for Paediatric Anaesthesia in New
Zealand and Australia

PAN_28_1_Issueinformation-online.indd i 05-12-2017 16:17:31

Pediatric Anesthesia
E D I TO R - I N - C H I E F
A. Davidson (Melbourne, Australia)

A S S O C I AT E E D I TO R - I N - C H I E F E D U CAT I O N E D I TO R
Brian Anderson (Auckland, New Zealand) Mark Thomas (London, UK)

C O R R E S P O N D E N C E E D I TO R
David Polaner (Denver, USA)

SECTION EDITORS Britta Regli-von Ungern-Sternberg STATISTICS EDITOR

Adrian Bosenberg (Seattle, USA) (Perth, Australia) Katherine Lee (Melbourne, Australia)
Joe Cravero (Boston, USA) Francis Veyckemans (Lille, France) SOCIAL MEDIA EDITOR
Susan Goobie (Boston, USA) Laszlo Vutskits (Geneva, Switzerland) Hannah Gill (Bristol, UK)
Dean Kurth (Philadelphia, USA) Suellen Walker (London, UK)
Chandra Ramamoorthy (Stanford, USA)

EDITORIAL BOARD Tom Hansen (Odense, Denmark) Steven Shafer (Stanford, USA)
Dean Andropoulos (Houston, USA) Pablo Ingelmo (Montreal, Canada)
Mark Ansermino (Vancouver, Canada) Qingquan Lian (Wenzhou, China)
Victor Baum (Charlottesvile, USA) Andreas Loepke (Cincinnati, USA)
Thomas Engelhardt (Aberdeen, UK) Allison Ross (Durham, USA)

ASSOCIATE EDITORS Narasimhan Jagannathan (Chicago, USA) Santhanam Suresh (Chicago, USA)
Nicole Almenrader (Rome, Italy) Peter Larsson (Stockholm, Sweden) Shoichi Uezono (Tokyo, Japan)
Karin Becke (Nurnberg, Germany) Christine Mai (Boston, USA) Isabeau Walker (London, UK)
Patrick Birmingham (Chicago, USA) Shobha Malviya (Michigan, USA) Gina Whitney (Denver, USA)
Robin Cox (Calgary, Canada) Steven Roberts (Liverpool, UK) Ma-Zhong Zhang (Shanghai, China)
Jurgen de Graaff (Rotterdam, The Netherlands) Nada Sabourdin (Paris, France) Yunxia Zuo (Chengdu, China)
Geoff Frawley (Melbourne, Australia) Justin Skowno (Sydney, Australia)

PA S T E D I TO R S - I N - C H I E F
Gordon Bush
Ted Sumner
Neil Morton

Devoted to the publication of information of interest and importance to Publisher. Pediatric Anesthesia is published by John Wiley & Sons Ltd: 9600
practising anesthetists everywhere, Pediatric Anesthesia aims to be truly Garsington Road, Oxford OX4 2DQ, UK. Tel: +44 (0)1865 776868, Fax: +44
international in policy and flavour. The scientific and clinical content of the (0)1865 714591.
journal covers a wide selection of medical disciplines in all areas relevant to Journal Customer Services: For ordering information, claims and any
anesthesia and intensive care in newborns, infants and children. Pediatric enquiry concerning your journal subscription please go to http://www.
Anesthesia has an intercontinental Editorial Board, which is supported by wileycustomerhelp.com/ask or contact your nearest office:
Associate Editors from around the world, to ensure that the journal is pub- Americas: Email [email protected]; Tel: +1 781 388 8598 or 1 800 835 6770
lishing the best work from the front line of research in the field. In addition (Toll free in the USA & Canada).
to top quality scientific research papers and thorough and up to date reviews, Europe, Middle East and Africa: Email: [email protected]; Tel: +44 (0)
there are thought provoking editorials and sections covering case presenta- 1865 778315
tions; and correspondence. Asia Pacific: Email: [email protected]; Tel: +65 6511 8000
This journal is affiliated with the following societies: The European Society for Japan: For Japanese speaking support, Email: [email protected]; Tel (tollfree):
Paediatric Anaesthesia, the Association of Paediatric Anaesthetists of Great 005 316 50 480.
Britain and Ireland, Association des Anesthésistes Réanimateurs Pédiatriques Visit our Online Customer Help available in 7 languages at www.wileycustom-
d’Expression Françfaise (the French-Language Society of Paediatric erhelp.com/ask
Anaesthesia), the Italian Society of Paediatric Anaesthesia, the Japanese Society Editorial Office: Louise Whelan (email: [email protected]).
of Pediatric Anesthesiology, the Society for Pediatric Anesthesia, the Society for Production Editor: Chandhini Mohan (email: [email protected]).
Paediatric Anaesthesia in New Zealand and Australia, Paediatric Anaesthesia Advertising: Sarah Young, Pediatric Anesthesia, (email: [email protected]).
Community of South Africa, and Chinese Society of Pediatric Anesthesiology. Commercial reprints: Claire Porter (email: [email protected]).
Pediatric Anesthesia is indexed by: Index Medicus, MEDLINE, Current For further details of submission instructions, subscription and all other
Contents/Clinical Medicine, EMBASE, SciSearch, Research Alert, Ad information visit: http://www.wileyonlinelibrary.com/journal/pan
Referendum Anaesthesiology, SUBIS, Current Opinion in Anaesthesiology
Copyright © 2018 John Wiley & Sons Ltd. All rights reserved. No part of this
Editorial correspondence should be directed to: A. Davidson, care of Louise publication may be reproduced, stored or transmitted in any form or by any
Whelan, Pediatric Anesthesia, John Wiley & Sons Ltd, 9600 Garsington Road, means without the prior permission in writing from the copyright holder.
Oxford OX4 2DQ, UK. Email: [email protected]; Tel: +44 (0)1243 Authorization to copy items for internal and personal use is granted by the
770686; Fax: +44 (0)1243 770460 copyright holder for libraries and other users registered with their local
Reproduction Rights Organisation (RRO), e.g. Copyright Clearance Center
Information for subscribers. Pediatric Anesthesia is published in 12 issues per (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright.
year. Institutional subscription prices for 2018 are: Print & Online: US$3228 com), provided the appropriate fee is paid directly to the RRO. This consent
(The Americas), US$3764 (Rest of World), €2219 (Europe), £1748 (UK). Prices does not extend to other kinds of copying such as copying for general
are exclusive of tax. Asia-Pacific GST, Canadian GST/HST and European VAT distribution, for advertising or promotional purposes, for creating new
will be applied at the appropriate rates. For more information on current tax collective works or for resale. Special requests should be addressed to
rates, please go to www.wileyonlinelibrary.com/tax-vat. The price includes [email protected]
online access to the current and all online back files to January 1st 2013,
where available. For other pricing options, including access information and The Publisher, affiliated societies and Editors cannot be held responsible for
terms and conditions, please visit www.wileyonlinelibrary.com/access errors or any consequences arising from the use of information contained in
this journal; the views and opinions expressed do not necessarily reflect those
Delivery Terms and Legal Title: Where the subscription price includes print of the Publisher, affiliated societies and Editors, neither does the publication
issues and delivery is to the recipient’s address, delivery terms are Delivered at of advertisements constitute any endorsement by the Publisher, affiliated
Place (DAP); the recipient is responsible for paying any import duty or taxes. societies and Editors of the products advertised.
Title to all issues transfers FOB our shipping point, freight prepaid. We will
endeavour to fulfil claims for missing or damaged copies within six months of Wiley’s Corporate Citizenship initiative seeks to address the environmental, social,
publication, within our reasonable discretion and subject to availability. economic, and ethical challenges faced in our business and which are important
to our diverse stakeholder groups. Since launching the initiative, we have focused
Back issues: Single issues from current and recent volumes are available at on sharing our content with those in need, enhancing community philanthropy,
the current single issue price from [email protected]. Earlier issues may reducing our carbon impact, creating global guidelines and best practices for paper
be obtained from Periodicals Service Company, 351 Fairview Avenue – Ste 300, use, establishing a vendor code of ethics, and engaging our colleagues and other
Hudson, NY 12534, USA. Tel: +1 518 822-9300, Fax: +1 518 822-9305, Email: psc@ stakeholders in our efforts. Follow our progress at www.wiley.com/go/citizenship
periodicals.com Access to this journal is available free online within institutions in the developing
PEDIATRIC ANESTHESIA, (1155-5645), is published monthly. US mailing world through the HINARI initiative with the WHO. For information, visit www.
agent: Mercury Media Processing, LLC, 1850 Elizabeth Avenue, Suite #C, Rah- who.int/hinari/en/
way, NJ 07065, USA. Periodical postage paid at Rahway, NJ. Postmaster: Send ISSN 1155-5645 (Print)
all address changes to PEDIATRIC ANESTHESIA, John Wiley & Sons Inc., C/O ISSN 1460-9592 (Online)
The Sheridan Press, PO Box 465, Hanover, PA 17331.
Printed in Singapore by C.O.S Printers Pte Ltd.

PAN_28_1_Issueinformation-online.indd ii 05-12-2017 16:17:32

Instructions to authors dards_for_reporting_in_Pediatric_Anesthesia.pdf. Clinical Implications, please add 1-2 sentences an-
swering the following questions: a. What is already known about the topic b. What new information this
study adds. A structured abstract of no more than 300 words should include the following: Background;
GENERAL Aims; Methods; Results; Conclusions. Maximum words – 3500; maximum figures and tables – 6; maxi-
mum references – 25. Word counts include all text from the introduction to the end of the text after the
Pediatric Anesthesia’s mission is to advance the science and clinical practice of paediatric anaesthe- disclosures.
sia, pain management and peri-operative medicine through dissemination of research, education Educational Reviews - Educational reviews should have the following structure: Introduction, Main Ar-
and quality improvement. ticle, Summary, Reflective questions, References. Present 3-4 reflective questions that the reader should
Priority is given to high-quality original research that advances knowledge, safety, organisation or ponder upon when they have assimilated the knowledge within the article. Summary of no more than
methodology applicable to other settings and countries. We would particularly like to encourage the 300 words. Maximum words – 5000-6000; maximum references – 20. Word counts include all text and
reporting of randomised controlled trials. references.
We will support our authors by posting the accepted version of articles by NIH grant-holders to Systematic reviews – Systematic reviews are encouraged and should include a clear aim and search
PubMed Central upon acceptance by the journal. Authors must ensure that manuscripts are clearly strategy. If the review is a meta-analysis it should be submitted and structured as a research report
indicated as NIH-funded using the guidelines below. with an abstract, background methods, results discussion, and a clearly articulated aim, search
We accept articles with audio and video files (Podcasts). The published article links to the corre- strategy etc. Summary of no more than 300 words giving information on methods of selecting
sponding Podcast hosted on YouTube or iTunes. See our special feature page and submission policy the publications cited. Maximum words – 4000; maximum figures and tables – 6; maximum refer-
below for more details. ences – no limit. Word counts include all text from the introduction to the end of the text after the
Please read the instructions below for brief details on the Journal’s requirements for manuscripts. disclosures.
Please visit the journal webpage http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1460-9592/ Focused reviews – Concise reviews are encouraged and should cover a topic which is of relevance
homepage/ForAuthors.html for full and updated Author Guidelines. to pediatric anesthesia, conveying a small number of key points that must be based on solid
Ethics evidence. Summary of no more than 200 words. Maximum words – 1500; maximum figures and
The journal’s ethical policies are outlined in the separate document, Ethical Policies of tables – 2; maximum references – 10 Word counts include all text from the introduction to the end
Pediatric Anesthesia (http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1460-9592/homepage/ of the text after the disclosures.
PAN_Ethics_Policy_March_2014.pdf ). These guidelines must be followed in full. Special interest articles – Novel papers that are neither research reports nor reviews on specific
Upon acceptance of a manuscript, the author identified as the formal corresponding author for topics will be considered if they have a great and broad interest to the specialty. Summary of no
the paper will receive an email prompting them to login into Author Services; where via the Wiley more than 300 words. Maximum words – 4000; maximum figures and tables – 6; maximum refer-
Author Licensing Service (WALS) they will be able to complete the license agreement on behalf of all ences – no limit. Word counts include all text from the introduction to the end of the text after the
authors on the paper. Your article cannot be published until this has been done. disclosures.
Plagiarism Checking Case reports – only exceptional reports that have important education or safety messages will be
We request that authors screen their manuscripts for plagiarism before submission. We do NOT considered. Our current rejection rate is 90%. Conclude with 3 learning points for our readers. All case
question the honesty of our authors. Almost all plagiarism is accidental, and we make this request reports require parental/ patient consent for publication. Summary of no more than 100 words.Maxi-
to protect our authors. Please indicate how you screened your manuscript for plagiarism at submis- mum words – 1000; maximum figures or tables – 1; maximum references – 5 Word counts include all
sion: Article checker (www.articlechecker.com), CrossCheck (www.ithenticate.com/), CrossRefMe text and references.
(www.crossrefme.com), Doc Cop (www.doccop.com), Dupli Checker (www.duplichecker.com), Editorials - Editorials are usually by invitation. They should be less than 1500 words and should refer to
Plagiarisma (www.plagiarisma.net), Plagiarism Checker (www.searchenginereports.net/articlech- a paper in the issue within the first two sentences. They should have less than 6 references and no tables
eck.aspx), The Plagiarism Checker (www.Dustball.com/cs/plagiarism.checker), Other, Please add or figures. Usually they should have 3 or fewer authors.
details or I did not screen for plagiarism. Correspondence – Letters to the editor are encouraged, particularly if they comment, question or criticize
research reports that have been published in the journal. Such letters MUST refer to the research reports
SUBMISSION REQUIREMENTS in the first paragraph, and list that paper as reference 1. Letters that describe cases are only considered if
All submissions to Pediatric Anesthesia should conform to the uniform requirements http://www. they have an important safety message and require parental/ patient consent for publication. Maximum
icmje.org/ for manuscripts submitted to biomedical journals, drawn up by the International Com- words – 800; maximum figures and tables – 1; maximum references – 5. Word counts include all text
mittee of Medical Journal Editors (ICMJE) see http://www.icmje.org/. and references.
General requirements Guidelines for specific papers:
All submissions should include the following: Randomised clinical trials (RCTs) must conform to the CONSORT statement (http://www.consort-
Main document statement.org) on the reporting of RCTs. A flow diagram of subjects, the trial protocol, and the registra-
1. Title page; tion details of the trial must be included in the paper, along with and a numbered checklist provided
2. Title; as supplementary material. Trials must have been registered at a publicly accessible trial registration
3. Article category; website prior to commencement of the trial.
4. First name, middle initial (if any) and family name of all authors – no degree/titles/ positions of Diagnostic studies must conform to the STARD statement http://www.stard-statement.org/. A flow
responsibility. All those listed as authors must fulfil the ICMJE criteria – see above; diagram of subjects, the trial protocol, and the registration details of the trial must be included in the
5. Affiliations should be written after the authors list as follows and linked to authors with corre- paper along with and a checklist provided as supplementary material.
sponding superscript number. Only include: Department, division or unit name in English, (if any), Qualitative research – authors should refer to the EQUATOR NETWORK resource guidance on good
affiliation name, city(without state), country; research reporting (http://www.equator-network.org), which has the full suite of reporting guidelines
6. Corresponding author details should be written after the affiliations list as follows: title (Mr/Mrs/ (both quantitative and qualitative).
Ms/Dr/Prof), first name(s) written with initials only, and followed by the last name – e.g. Dr. J. E. Observational studies (Epidemiology): Please follow the STROBE guidelines http://www.strobe-
Smith; add Department, division or unit name in English, (if any), affiliation name, street address, statement.org/ and submit the study protocol as supplementary material.
city, postal code, country. Email address; Systematic reviews/ meta-analysis of randomised trials and other evaluation studies must conform
7. For research reports only, add 1-2 sentences answering the following questions in bullet points: a. to PRISMA guidelines http://www.prisma-statement.org (these have superseded the QUOROM
What is already known b. What this article adds; guidelines) and submit the study protocol as supplementary material.
8. A structured abstract for research reports with a clearly stated background, aim, method, results
and conclusion; SUBMISSION
9. A summary for educational reviews, systematic reviews special interest articles and case reports; Submission must have been read and approved by all authors. Submission of a manuscript implies
10. Six MeSH-compliant keywords (http://www.nlm.nih.gov/mesh) that do not replicate the title; that it reports unpublished work and that it is not under active consideration for publication else-
11. Disclosures: Indicate at the end of the text before references: 1. Any necessary ethical approval(s); where, nor been accepted for publication, nor been published in full or in part (except in abstract
2. The source of funding for the study with grant numbers; and 3. Any conflict of interest. You are form).
required to make a statement, even if the answer is ‘none’. All material to be considered for publication in Pediatric Anesthesia should be submitted via the jour-
12. A reference list in AMA style (number/author) endnotes in the order made in the text. Example: nal’s online submission system at ScholarOne Manuscripts http://mc.manuscriptcentral.com/pan.
confirmed by other studies.23 / 23 Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting Full instructions and support are available on the site and an user ID and password can be
items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009; 339: b2535. For obtained on the first visit. If you have any queries please contact Louise Whelan, Pediatric Anesthe-
books, names and initials of all authors, the full title, place of publication, publisher, year of publica- sia Editorial Office: [email protected].
tion and page number should be given.
13. Tables – if any, in tabulate text at the end of the main document, following the references. DO NOT
Proofs
submit tables as separate files. Tables submitted as pictures cannot be used.
One copy of the proofs will be sent to the first named or designated author and one should
14. Figure captions – if any, in a list following the references/ tables. (Figures must be uploaded ad-
be returned within 48 hours of receipt to avoid delay in publication. Overseas contributors
ditionally as individual graphic files. Please do not embed figures.) Figures embedded in word cannot
be used by the publisher. Figures should be submitted as single separate image files: either resolution should ensure that a rapid airmail service is used. Authors are encouraged to use E-annota-
independent EPS files, or high resolution (600dpi at print size) TIFF files. tion tools available in Adobe Acrobat Professional or Acrobat Reader (version 7.0 or above)
to e-annotate the PDF copy of their proofs which can be returned to the Corrections Team
Supporting information/additional files if appropriate electronically.
1. Figures – prepared and labelled as advised in ‘PRE-SUBMISSION ADVICE AND PREPARATION’
above. 2. Audio/ Video- prepared as advised in ‘PRE-SUBMISSION ADVICE AND PREPARATION’ Offprints
above. Please use the file designation ‘audio/video data’. 3. Supplementary data - prepared and labelled Authors will be provided with electronic offprints of their paper (as a PDF). Paper offprints may
as advised in ‘PRE-SUBMISSION ADVICE AND PREPARATION’ above. Please use the file designa- be ordered at prices quoted on the order form that accompanies proofs, provided that the form is
tion ‘supporting data’. 4. Study protocol – the appropriate study protocol (see ‘Guidelines on specific returned with the proofs.
papers’ below) 5. Consent for publication– A completed / signed parental/patient consent form should
be uploaded onto S1M as file designation ‘Publication consent’. Download the links given in ‘POLI- Tracking your manuscript: Author Services
CIES’, above. 6. Online video broadcast release – a completed form should be uploaded onto S1M as file Online production tracking is available for your article through Wiley Author Services:
designation ‘online broadcast release’. Download the links given in ‘POLICIES’, above. 7. Permission – if http://authorservices.wiley.com/bauthor/.
reproducing others work (see ‘PRE-SUBMISSION ADVICE’ above). A full version of the Author Guidelines can be found online at http://onlinelibrary. wiley.com/
Cost of colour reproduction of artwork journal/10.1111/(ISSN)1460-9592/homepage/ForAuthors.html.
Authors to pay the full cost for the reproduction of their colour artwork. If you would like for your
figures or tables to appear in colour please sign and complete the Colour Work Agreement Form Online Open
(CWAF) when your paper has been accepted: http://otis.wiley.com/otis/journal/overview/ If the OnlineOpen option is selected the corresponding author will have a choice of the fol-
en/6657/SN_Sub2000_P_CoW.pdf. Manuscripts received with colour files WILL NOT be repro- lowing Creative Commons License Open Access Agreements (OAA): Creative Commons At-
duced in colour unless a CWAF has been received. Once completed, please return the form to Cus- tribution License OAA Creative Commons AttributionNon-Commercial License OAA Creative
tomer Services (OPI), John Wiley & Sons Ltd, European Distribution Centre, New Era Estate, Old- Commons Attribution Non-Commercial-NoDerivs License OAA.
lands Way, Bognor Regis, West Sussex PO22 9NQ. To preview the terms and conditions of these open access agreements please visit the Copyright
Pre-acceptance English-language editing services FAQs hosted on Wiley Author Services http://exchanges.wiley.com/authors/faqs---copy-
Pediatric Anesthesia is pleased to recommend to authors pre-acceptance English-language editing right-_301.html and visit http://www.wileyopenaccess.com/details/content/12f25db4c87/Copy-
services: http://authorservices.wiley.com/bauthor/english_language.asp. The Editor may rec- right–License.html.
ommend an English-language editing service to the author as a condition of acceptance. If you select the OnlineOpen option and your research is funded by certain funders [e.g. The
Wellcome Trust and members of the Research Councils UK (RCUK) or the Austrian Science Fund
PREPARATION OF TYPESCRIPTS (FWF)] you will be given the opportunity to publish your article under a CC-BY license sup-
Accepted article types porting you in complying with your Funder requirements. For more information on this policy
Research reports – Research reports should follow the Minimal standards for reporting in Pediatric An- and the Journal’s compliant self-archiving policy please visit: http://www.wiley.com/go/funder-
esthesia” http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1460-9592/homepage/Minimal_stan- statement.

PAN_28_1_Issueinformation-online.indd iii 05-12-2017 16:17:32

Pediatric Anesthesia
N U M B E R 2 8 , I S S U E 1 , J A N U A RY 2 0 1 8

CONTENTS
In this issue
3 In this issue: January 2018 A. Davidson

Editorials
4 What can we learn (or not) from in vitro airway studies for clinical applications in children? N. Jagannathan and
B.S. von Ungern-Sternberg
6 We need personalized assessments for implicit bias in higher risk situations D.B. Waisel

Educational Review
8 New solutions to reduce wrong route medication errors R.S. Litman, V.I. Smith and P. Mainland

Systematic Reviews
13 The anatomy of the pediatric airway: Has our knowledge changed in 120 years? A review of historic and recent
investigations of the anatomy of the pediatric larynx J. Holzki, K.A. Brown, R.G. Carroll and C.J. Coté
23 Donohue syndrome: A review of literature, case series, and anesthetic considerations A. Kirkwood, G. Stuart and
L. Harding
28 Assessment of sedation level prior to neonatal intubation: A systematic review E.H.M. de Kort, N.M. Halbmeijer,
I.K.M. Reiss and S.H.P. Simons

Research Reports
37 Pressure-flow characteristics of breathing systems and their components for pediatric and adult patients C. Wenzel,
S. Schumann and J. Spaeth
46 Introducing the 6-4-0 fasting regimen and the incidence of prolonged preoperative fasting in children H. Andersson,
P.M. Hellström and P. Frykholm
53 Ultrasound-guided pudendal nerve block in children: A new technique of ultrasound-guided transperineal
approach I. Gaudet-Ferrand, P. De La Arena, S. Bringuier, O. Raux, L. Hertz, N. Kalfa, C. Sola and C. Dadure
59 Clinical implications of de Barsy syndrome L.L. Warner, D.A. Olsen and H.M. Smith
63 Incidence and predictors of 30-day postoperative readmission in children D. Vo, D. Zurakowski and D. Faraoni

Case Report
71 Nitroglycerin: An unusual solution to intraoperative hypothermia in a 4-year-old burn patient E.J. Goenaga Díaz,
J.E. Carter and T.W. Templeton

Correspondence
73 In response to “Interventions designed using quality improvement methods reduce the incidence of serious airway
events and airway cardiac arrests during pediatric anesthesia” B.H. Cloyd
74 Continuous Erector Spinae Plane block for thoracic surgery in a pediatric patient J.C. De la Cuadra-Fontaine, M. Concha,
F. Vuletin and H. Arancibia
75 Mispositioning the end of a cuff inflating line in long-axis ultrasound imaging of the pediatric larynx and
trachea K. Kayashima and K. Imai

Pediatric Anesthesia is indexed in Index Medicus, MEDLINE, Current Contents/Clinical Medicine, EMBASE/
Excerpta Medica, Sci Search, Research Alert, Ad Referendum Anaesthesiology,
SUBIS and Current Opinion in Anaesthesiology

PAN_28_1_Issueinformation-online.indd 1 05-12-2017 16:17:32

DOI: 10.1111/pan.13299

In this issue: January 2018

NEW SOLUTIONS TO REDUCE WRONG ROUTE MEDICATION ERRORS

New connectors will be coming your way soon. This fascinating review covers the history and new developments of the tubing connectors.

INCIDENCE AND PREDICTORS OF 30-DAY POSTOPERATIVE READMISSION IN CHILDREN

Developing quality outcome measures is important. For better or worse, hospital readmission is one that is frequently used. This large study
identifies children at risk of readmission.

INTRODUCING THE 6-4-0 FASTING REGIMEN AND THE INCIDENCE OF PROLONGED PREOPERATIVE FASTING IN
CHILDREN

Another study that will hopefully help children being less grumpy and thirsty.

THE ANATOMY OF THE PEDIATRIC AIRWAY: HAS OUR KNOWLEDGE CHANGED IN 120 YEARS? A REVIEW OF
HISTORIC AND RECENT INVESTIGATIONS OF THE ANATOMY OF THE PEDIATRIC LARYNX

There is still argument over the basic shape of a child’s larynx. This paper presents a strong case for the funnel.

ASSESSMENT OF SEDATION LEVEL PRIOR TO NEONATAL INTUBATION: A SYSTEMATIC REVIEW

In the neonatal intensive care, endotracheal intubation is a not infrequent and a stressful event for the infant. Proving adequate sedation
beforehand is important but evidence to guide best practice is scarce. This paper reviews how adequate sedation is assessed.

Andrew Davidson
Editor-in-Chief

Pediatric Anesthesia. 2018;28:3. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 3
DOI: 10.1111/pan.13280

EDITORIAL

What can we learn (or not) from in vitro airway studies for
clinical applications in children?
Albert Einstein once said, “if you can’t explain it simply, you don’t Mohr et al., aimed to assess the impact of tracheal tube (TT) short-
understand it well enough”, implying that thorough knowledge and ening in infants, a not uncommon clinical practice to reduce the
comprehension of a topic is a prerequisite for its useful application. work of breathing in neonates and infants.1 The authors used an
Baseline knowledge of many topics in pediatric anesthesia, however, infant lung model to test this hypothesis for the existing practice.
are so limited that they even preclude safe research in human sub- However, the study showed that shortening the TT did not reduce
jects, let alone provide simple explanation and clinical application. In work of breathing in this infant lung model. Although this in vitro
vitro research studies aim to bridge this knowledge gap by integrat- study environment was not identical to in vivo settings, these results
ing laboratory findings to clinical trials, in an attempt to bring novel may discourage clinicians from shortening the TT, and potentially
concepts from “bench-to-bedside”. The main advantages of an help design future clinical studies to further explore the impact of
in vitro study are the ability to identify potential suitability for shortening TTs when caring for infants.
human subjects testing, or to assess the underlying human physiol- The second study is a quality improvement project assessing
ogy or material properties of certain equipment. Alternatively, the breathing circuit compliance in infants.2 Glenski et al. performed an
results of an in vitro study may be convincing enough to pursue in vitro study on infant test lungs by performing preuse self-tests on
human subjects testing. However, their inability to fully emulate nonexpanded, and then subsequently expanded breathing circuits.
real-life clinical conditions remains a major limitation. The authors found that this led to falsely elevated tidal volumes in
Einstein also gave us further advice: “A theory can be proved by infants less than 10 kg during pressure-controlled ventilation. The
experiment; but no path leads from experiment to the birth of a the- results obtained from their study were then applied clinically in real
ory.” Within the context of pediatric anesthesia research, this quote infants and led to a practice change in the authors’ institution after
reminds us that just because a model is available, it should not mean department-wide education and implementation.
that one should proceed to perform in vitro trial. Similarly, the 19th Both of these studies are fine examples of the use of laboratory
century French physiologist, Claude Bernard remarked, “The experi- testing to explore common clinical practices and challenge the valid-
menter who does not know what he is looking for will never under- ity of existing clinical paradigms to better define patient care without
stand what he finds.” The design of an in vitro study should risking patient morbidity.
foremost have a translational application in mind. Herein this edito-
rial, we discuss the potential advantages and limitations associated
with recent in vitro research in pediatric anesthesia. 2 | MANIKIN STUDIES
When embarking on a new in vitro airway study, a few questions
should be considered as a litmus test for whether to proceed with a As the Chemistry Nobel Laureate put it, “A theory has only the alter-
particular design. Failure to resoundingly answer “yes” to one of the native of being right or wrong. A model has a third possibility: it
following questions, should be a flag for revisiting the study question may be right, but irrelevant.” The latter is unfortunately true for
its merits, and/or its methodological aspects: many medical manikin models—as there are few high-quality pedi-
atric manikins that can truly emulate the anatomic/physiologic condi-
• Is an in vitro study the only way we can study this topic because tions of a real-life patient, thereby limiting their relevance and
human subjects study is impossible or harmful? generalizability. While manikin studies may be suitable to teach,
• Will the results of this in vitro study have potential to help design train, acquire, and assess skills by emulating rarely encountered clini-
better in vivo studies? cal situations, manikins may be useful in evidence-based training in
• Will the results of this study have potential to alter current prac- the simulated management of the difficult airway. For example, how
tices or improve patient care? to assess various airway device performance in a Pierre Robin mani-
kin.3 Although this type of manikin may be useful for teaching and
training purposes, the results observed from such a manikin study
1 | EQUIPMENT RESEARCH may not be translated to real-life clinical conditions. Certain vari-
ables, such as secretions, dynamic changes (reflex activation of the
Two recent interesting in vitro studies with potential clinical implica- airway, light anesthesia, oxygen desaturations), are difficult to emu-
tions involve the use of infant lung models. The first, conducted by late. Even high fidelity simulation with better features cannot

4 | © 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/pan Pediatric Anesthesia. 2018;28:4–5.
EDITORIAL | 5

completely remove the sense of an artificial environment and may Princess Margaret Hospital Foundation and the Stan Perron Charita-
not be reflective of real clinical practice/behavior. The time pressure ble Trust.
and stress experienced in actual clinical practice cannot be replicated
and experienced during performances on a manikin. Although these DISCLOSURES
limitations exist, manikins have been a valuable tool to train on, and
enhances clinical education. Situations exist where it may be unethi- Britta S von Ungern-Sternberg is a Section Editor of Pediatric Anes-
cal to perform a clinical trial, or test experimental equipment on thesia. Narasimhan Jagannathan is an Associate Editor of Pediatric
human subjects. In these situations, the use of manikins and/or ani- Anesthesia.
mal models (see below) become a more suitable study model, how-
ever the translational value of a manikin is highly questionable.
ORCID

Britta S. von Ungern-Sternberg http://orcid.org/0000-0002-8043-

3 | ANIMAL STUDIES 8541

Animal models often serve as a reasonable alternative to human sub-

Narasimhan Jagannathan1,2
jects for rarely encountered clinical situations. The Cannot Intubate,
Britta S. von Ungern-Sternberg3,4
Cannot Oxygenate (CICO) scenario in children is rarely encountered 1
Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, IL, USA
in real practice, with limited data regarding best practices, and 2
Department of Pediatric Anesthesia, Northwestern University Feinberg
anatomical differences in infants make extrapolation of adult evi-
School of Medicine, Chicago, IL, USA
dence problematic.4 Moreover, the assessment of various 3
Department of Anesthesia and Pain Management, Princess Margaret
approaches and/or techniques cannot be ethically and feasibly stud-
Hospital for Children, Perth, WA, Australia
ied in human subjects. Therefore, animal models have served as 4
Medical School, Division of Anaesthesiology and Pain Medicine, The
alternative in vitro study models for these purposes. The use of rab-
University of Western Australia, Perth, WA, Australia
bit and/or piglet models may provide helpful information on efficacy
Email: [email protected]
of certain oxygenation and percutaneous techniques and maybe a
more realistic approach compared to manikin models.5 Similar to
manikin studies, animal models may not exactly duplicate real-life
clinical conditions, or necessarily translate to a transfer of skills into REFERENCES
clinical practice, they may provide the only alternative environment/
1. Mohr R, Thomas J, Cannizzaro V, et al. Impact of endotracheal tube
condition to emulate rare clinical conditions, and therefore help
shortening on work of breathing in neonatal and pediatric in vitro
guide practice. lung models. Pediatr Anesth. 2017;27:942-948.
In conclusion, in vitro models can be useful (or not) to help clini- 2. Glenski TA, Diehl C, Clopton RG, Friesen R. Breathing circuit compli-
cians guide clinical practice by possibly emulating extremely rare clin- ance and accuracy of displayed tidal volume during pressure-con-
trolled ventilation of infants: a quality improvement project. Pediatr
ical situations, act as a preliminary model prior to human subjects
Anesth. 2017;27:935-941.
testing, or offer information that would preclude progression to 3. Hippard HK, Kalyani G, Olutoye OA, Mann DG, Watcha MF. A com-
human trials. Investigators should carefully consider which models parison of the Truview PCD and the GlideScope Cobalt AVL video-
and study designs may fulfill these criteria in order to ‘separate the laryngoscopes to the Miller blade for successfully intubating manikins
simulating normal and difficult pediatric airways. Pediatr Anesth.
wheat from the chaff’ to help us guide better practice choices for
2016;26:613-620.
clinicians. 4. Sabato SC, Long E. An institutional approach to the management of
the ‘Can’t Intubate, Can’t Oxygenate’ emergency in children. Pediatr
Anesth. 2016;26:784-793.
FUNDING 5. Prunty SL, Aranda-Palacios A, Heard AM, et al. The ‘Can’t intubate
can’t oxygenate’ scenario in pediatric anesthesia: a comparison of the
Britta S von Ungern-Sternberg holds the Callahan Chair in Paediatric Melker cricothyroidotomy kit with a scalpel bougie technique. Pediatr
Anaesthesia and is partly funded by the Late Frank Callahan, the Anesth. 2015;25:400-404.
DOI: 10.1111/pan.13289

EDITORIAL

We need personalized assessments for implicit bias in higher

risk situations
In a previous issue of Pediatric Anesthesia, Rosenbloom et al1 and Genomes differ within the same socially constructed race category
2
Maher et al present studies looking at different approaches to (eg, Black, White, Native American, and Pacific Islander).4 Self-assess-
assessing health and healthcare disparities. ment of race does not indicate genetic structure.5 Biological differ-
Healthcare disparities affect racial, ethnic, gender, socioeconomic, ences are better off determined at the genetic level, unless it is well
geographic, and other cohorts in areas such as insurance coverage, shown that the relevant genes are consistent with self-assessed race.
access to care, quality of care, and clinician actions. In a “Limited
Evaluation of the Association of Race and Anesthetic Medication
Administrations”, Rosenbloom et al1 examined whether there were 2 | RISKS WITH RACE OR GENETIC-BASED
healthcare disparities between patient race and administration of ASSESSMENTS OF PATIENTS
medication. In their study, race was self-reported as either Black or
White. They found no racial disparities in the use of preoperative or In the medical system in the United States, costly health insurance
intraoperative medications. for “high-risk” patients often makes good insurance unaffordable,
Health (as compared to healthcare) disparities are variations in mor- and inexpensive insurance may limit the care options offered, creat-
bidity and mortality among groups. Maher et al2 found that African- ing additional disparities. Genetic data already affect the costs and
American children undergoing spinal fusion bled more and had an availability of life insurance and long-term care insurance.6 There are
increased rate and volume of blood transfusion. Limited data availability loopholes in the Genetic Information Nondiscrimination Act, which
in their retrospective study makes it difficult to determine etiologies for was designed to limit the harms of genetic testing. For example, a
this disparity. Studies that indicate increased bleeding in African-Ameri- client that knows but does not share a genetic predisposition to a
can patients in other procedures suggests a physiologic component. disease may lose the benefits of insurance later because of “guilt by
Physiologic causes may be due to genetics or may arise from inadequate omission.” Considering Maher et al’s study, it is disquieting to add
living conditions, prior healthcare, and optimization for surgery. In this genetic discrimination to a cohort associated with health and health-
study, race was self-identified at registration from a prepopulated list care disparities.
that included: 2 or more races; American Indian or Alaska Native; Black To a cynic like me, the harms of the potential consequences of
or African-American; Caucasian/White; Native Hawaiian or Other Paci- genetic data will only increase; while you may think me “crying
fic Islander; Not reported/declined; or Other (free text). wolf”, who would have thought 20 years ago that body mass index
Studying disparities in populations who have a history of unequal and tobacco use would either limit getting insurance or require pay-
access, inadequate care and poorer outcomes increases awareness ing higher premiums.
of the general and specific issues. That said, we need to approach Despite these risks, these studies need to be done.
these studies with some caution.

3 | CONCLUSION
1 | BLACK IS NOT BLACK, WHITE IS NOT
WHITE Clinicians scorn interpersonal healthcare disparity data: “It is not rel-
evant to me, because I am not prejudice.” But implicit bias is ubiqui-
Darker-skinned Black people are more discriminated against than tous. A systematic review7 of papers found that implicit bias—
lighter-skinned Black people in both Black and White communities, unconscious and automatic—was negatively correlated with indica-
although there is some evidence of recent improvement in that area.3 tors of quality care. Implicit bias occurs by race, ethnicity, socioeco-
Colorism, which is bias attached to skin shade, is present across soci- nomic status, gender, age, body habitus, disease, disability, and
eties and races. In studies assessing conscious or unconscious bias of whether the patient contributed to their disease. We need to recog-
clinicians, it would be nice to categorize the patients not only by race nize higher risk situations where unconscious bias may affect patient
but also by prejudices within that race, such as categorizing patients care, have the wherewithal to acknowledge the possibility of implic-
as either lighter skinned or darker skinned. And because the outcome itly biased behavior, and take overt actions to minimize it.
is a function of an interpersonal relationship, the race and skin shade Studies like these help us recognize higher risk situations, such
of anesthesiologists should also be categorized. as variation in preoperative drug administration when anesthetizing

6 | © 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/pan Pediatric Anesthesia. 2018;28:6–7.
EDITORIAL | 7

children of color. In addition to the previously described cohorts at David B. Waisel

risk of harm from implicit bias, disparity research is needed in other Department of Anesthesiology, Perioperative and Pain Medicine, Boston
populations of patients receiving anesthesia, including patients who Children’s Hospital, Harvard Medical School, Boston, MA, USA
require interpreters, are having surgical treatment for gender dys- Email: [email protected]
phoria, have a nonbinary gender or are agender, are obese, or with
whom we have a subpar relationship (the pejoratively termed “diffi-
cult” patient). Situations in which disparities may occur include the REFERENCES
quality of preoperative discussions, the extent of trainee supervision,
and the performance of postoperative visits. 1. Rosenbloom J, Senthil K, Long A, et al. A limited evaluation of the
association of race and anesthetic medication administration: a single-
Future studies should incorporate technically more complex fac-
center experience with appendectomies. Pediatr Anesth. 2017;27:
tors that affect performance, such as numbers of rooms covered, 1142-1147.
complexity of cases, times of day, extent of trainee supervision, and 2. Maher K, Owusu-Akyaw K, Zhou J, et al. African-American race is
other sources of production pressure. Studying additional factors associated with increased blood transfusion in pediatric scoliosis sur-
gery. 2016. http://www.asaabstracts.com/strands/asaabstracts/abstract.
may identify opportunities for system-level improvements, or at least
htm?year=2016&index=16&absnum=4342. Accessed December 5,
identify higher risk situations for disparate management. 2017.
To improve care, we need personalized assessments of our 3. Hunter M. The consequences of colorism. In: Hall R, ed. Melanin Mil-
behaviors to identify individual higher risk situations. Unfortunately, lenn. Skin Color as 21st Century International Discourse. Dordrecht,
Netherlands: Springer; 2013:247-256.
a congenital and justified distrust of organizational responses to
4. Solberg OD, Mack SJ, Lancaster AK, Single RM, Sanchez-mazas A,
personalized data makes these assessments unwelcome in our Thomson G. Balancing selection and heterogeneity across the classical
community, highlighting another (discounted) systemic barrier to human leukocyte antigen loci: a meta-analytic review of 497 popula-
serving our patients. tion studies. Hum Immunol. 2008;69:443-464.
5. Boquett J, Schu€ler-faccini L, Jobim LF, et al. Self-assessment of color
categories and its relationship with HLA profiling in Brazilian bone
CONFLICT OF INTEREST marrow donors. Biol Blood Marrow Transplant. 2015;21:1140-1144.
6. Peikoff K. Fearing punishment for bad genes. New York Times [Inter-
No conflict of interest. net]. New York City; April 8, 2014; D1. https://www.nytimes.com/
2014/04/08/science/fearing-punishment-for-bad-genes.html. Accessed
September 8, 2017.
ORCID 7. FitzGerald C, Hurst S. Implicit bias in healthcare professionals: a sys-
tematic review. BMC Med Ethics. BMC Med Ethics. 2017;18:19.
David B. Waisel http://orcid.org/0000-0002-5307-4092
Accepted: 28 September 2017

DOI: 10.1111/pan.13279

EDUCATIONAL REVIEW

New solutions to reduce wrong route medication errors

Ronald S. Litman1,2 | Victoria I. Smith3 | Phoebe Mainland4

1
Department of Anesthesiology & Critical
Care, The Children’s Hospital of Summary
Philadelphia, Philadelphia, PA, USA Wrong route medication errors due to tubing misconnections are potentially life-
2
Institute for Safe Medication Practices,
threatening complications that have been made possible by the universal use of the
Horsham, PA, USA
3
Drexel University, Philadelphia, PA, USA Luer connector. The new International Organization of Standardization standards for
4
Department of Anaesthesia and small bore connectors, ISO 80369 series, have been developed to reduce the risk of
Perioperative Medicine, AlfredHealth,
these types of erroneous connections. Tubing connectors for different routes of
Melbourne, VIC, Australia
clinical application will contain differently designed connectors that are physically
Correspondence
incompatible. However, design and manufacturing standards have progressed slowly,
Dr Ronald S. Litman, Department of
Anesthesiology & Critical Care, The and clinical roll-outs have been delayed, despite the implementation of California
Children’s Hospital of Philadelphia,
laws to promote their use. We are currently in a state of transition where new con-
Philadelphia, PA, USA.
Email: [email protected] nectors for enteral and neuraxial use will be entering the clinical market in the next
few years, and increasing use will identify additional patient safety concerns.
Funding information
The preparation of this review was funded
by departmental resources. KEYWORDS
Medication errors, small bore connectors, tubing misconnections
Section Editor: Mark Thomas

1 | INTRODUCTION flames in the Ramstein Air Base air show disaster.4 Some authors
have speculated that during the rescue effort, the incompatibility of
Much of the practice of hospital-based medicine can be traced to intravascular tubing between the USA army hospital and the German
the use of a universal tubing connector that allows connection paramedics contributed to the inability to resuscitate many of the
between medical devices from different manufacturers throughout injured spectators.1 In 1995, the International Organization of Stan-
the world. This connector was originally created in 1896 by Karl dardization (ISO) published an international standard specifying the
€er Company of
Schneider, an instrument maker for the H. Wulfing Lu consistent use of the Luer-tipped syringe.5 This standard has since
Paris. It consisted of a glass cylinder and plunger with a “male” inser- been used throughout the world.
tion tip that was constructed with a 6% taper to snugly fit into the There are always unintended consequences when standardization
corresponding “female” ports of various types of medical instillation processes occur. Because of the universal nature of Luer connectors,
devices, such as intravenous, urethral, or intrauterine catheters.1 delivery systems for intravascular (IV), enteral, respiratory, and neu-
€ er traveled to the United States, where he sold the patent of his
Lu raxial* medical devices have been accidentally interconnected to
universal syringe to Maxwell W. Becton and Fairleigh S. Dickinson, each other, either directly, or indirectly using adaptors that were not
who used it to start their fledgling company they called Becton Dick- intended for that specific use.6-8 These accidental misconnections
2
inson. And as they say, the rest is history. In the 1930s, Becton result in wrong route medication errors, and have resulted in patient
Dickinson added an external cuff locking mechanism to prevent acci- harm. As not all substances are safe when administered via different
TM
dental disconnections, the eponymous Luer-Lok we all still use routes, wrong route administrations can have severe consequences,
today.3 Over the ensuing century throughout the 1900s, the Luer- such as when an enteral feed is administered by the IV route, or a
tipped syringe became the USA industry standard for nearly all types drug that is ordinarily safe to administer by vein becomes toxic when
of medical tubing connections. Until the 1980s, many syringe manu- administered into one of the neuraxial spaces, such as intrathecal,
facturers in Europe continued to use the similar but incompatible
Rekord type syringe tip. But, in 1988, 3 pilots and 67 spectators lost *The term “neuraxial” describes the route of administration into the intrathecal, epidural,
and perineural spaces, as well as other areas of regional infiltration of local anesthetic to
their lives, and many hundreds were injured from the spread of
achieve pain control.

8 | © 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/pan Pediatric Anesthesia. 2018;28:8–12.
LITMAN ET AL. | 9

epidural, or perineural. Although published cases of wrong route route connections by preventing physical compatibility between
medication errors are likely widely underreported, published reports delivery routes. The Luer connector dimensions will remain for
describe such accidental connections that resulted in patient harm. intravascular and hypodermic applications only (Table 1 and Fig-
A 2011 systematic review of enteral-to-IV misconnections ure 1). The new standards were subsequently endorsed by the Asso-
described 116 cases in 34 publications, which resulted in the death ciation for the Advancement of Medical Instrumentation (AAMI),26
9
of 21 patients. Despite widespread recognition of the severity of and The Joint Commission released a follow up Sentinel Event in
this type of medication error,10 new reports continue to appear, 2014 urging healthcare providers to begin preparing for the new
such as a 6-week-old infant who was accidently administered breast connectors and to continue to vigilantly avoid misconnections until
milk intravenously.11 the new connectors debut.27
The accidental administration of intravascular medications into The only regulatory or legislative solution to wrong route errors
the neuraxial spaces has been amply reported.12-17 Some of these occurred in 2015 when the California legislature passed Assembly
reports describe devastating consequences. For example, a 59-year- Bill 444,28 which contains 3 main provisions:
old male was left permanently disabled after he was accidentally
administered high-dose gadolinium into his external ventricular drain 1. Beginning January 1, 2016, a health facility is prohibited from
during an imaging study.18 The continued occurrence of similar cases using an intravenous connector that would fit into a connector
prompted an impassioned editorial pleading for the urgent imple- other than the type it was intended for.
mentation of preventative measures.19,20 2. Beginning July 1, 2016, a health facility is prohibited from using
Although administration of toxic substances into the intravascular an enteral feeding connector that would fit into a connector
and neuraxial spaces are the most commonly reported wrong route other than the type it was intended for.
medication errors, additional types of calamities involving tubing 3. Beginning January 1, 2017, a health facility is California from
with the universal Luer connection have occurred with disastrous using an epidural connector that would fit into a connector other
outcomes. These include the accidental administration of enteral than the type it was intended for.
feed into the suction port of a tracheostomy, and fatal air emboli,
one after the connection of high-flow oxygen into an intravascular All of the above prohibitions stipulate that an exception can be
port, another after the connection of the tubing of an automated made when an emergency or urgent situation exists or if the prohibi-
noninvasive blood pressure cuff that was accidently connected to an tion would impair the ability to provide appropriate health care.
intravenous port.21 The successful implementation of the changes will not be possi-
ble without the participation of medical device manufacturers, who
will now be required to produce and distribute devices with connec-
1.1 | The new small bore connector standards
tors that comply with the new design specifications. The design for
Over the past several decades, there has been a concerted effort the connector for enteral use was the first to be confirmed, in
across the globe to devise ways to reduce the wrong route errors response to the California legislation deadline. A consortium of man-
22-24
that resulted from the use of universal Luer connectors. Organi- ufacturers of enteral devices and enteral feed manufacturers was
zations in the USA including the Food and Drug Administration,6 formed, known as the Global Enteral Device Supplier Association
Emergency Care Research Institute,7 and the Joint Commission on (GEDSA, www.gedsa.org), a nonprofit trade association to promote
Healthcare8 have issued warnings and sentinel events regarding the adoption of devices compliant with ISO 80369-3, enteral and other
danger of universal Luer connectors and the need for caution. Similar standards in the series, and to address operational and safety issues
statements and warnings have been issued in other countries, such in implementing products. GEDSA launched the Stay Connected initia-
as Patient Alerts from the National Patient Safety Agency in Eng- tive (stayconnected.org), a communications program with the goal to
land.25 In response to increasing awareness of this problem, a pro- facilitate transition to devices compliant with the ISO 80369 series
ject to develop unique connectors for medical devices used in of Small Bore Connectors and begin implementation of these
different applications started initially in 1997 by a Forum Task Group changes throughout the world.
of the European Committee for Standardization (CEN). Later this
was handed over to the ISO to create new standards for tubing con- T A B L E 1 New International Organization of Standardization (ISO)
nectors that would prevent interconnectivity. After extensive collab- 80369 connector series
oration between biomedical engineers, computer design experts, ISO standard Delivery route
manufacturers, consultants, and clinicians, it resulted in the ISO 80369-2 Respiratory
80369 series of standards for Small Bore Connectors. The first stan- 80369-3 Enteral
dard of the series, General Principles, was published in 2010. These
80369-5 Limb cuff inflation
new standards consist of design specifications of male and female
80369-6 Neuraxial
connectors that are unique for different routes of application into
80369-7 Intravascular (6% tapered Luer)
which medications or other substances may be administered. The
goal of these new specifications is to reduce the possibility of wrong
10 | LITMAN ET AL.

F I G U R E 1 The new small bore

connectors International Organization of
Standardization (ISO) 80369 series
(Courtesy of, and used with permission
from Phoebe Mainland, MBBS, FANZCA
and Kyle Steele)

In collaboration with the ISO Committee Working Group respon- With the introduction of these new connector systems into clini-
sible for the ISO 80369 series of standards and GEDSA, specific cal practice, previously unidentified problems have arisen.30 For
‘clinical advisory groups’ of international experts (from the United example, use of the new EnFitâ connectors resulted in dosing inac-
States and Europe, and for neuraxial, Australia) have been formed to curacies with the lowest volume (1 mL, 3 mL, and 5 mL) ENFitâ syr-
give input to clinical requirements for implementation of enteral and inges. After considerable effort and study, a new low-dose ENFitâ
neuraxial devices. tip syringe has been designed, with the intention of limiting the dose
The first new ISO small bore connectors to be introduced into inaccuracy to less than 10% (NeoMedâ Inc. “Low Dose Tip”).31 How-
clinical use will be the enteral (ISO 80369-3) connectors and the ever, one major syringe manufacturer, Becton Dickinson, made the
neuraxial (ISO 80369-6) connectors. Each will have a unique size decision to abandon the ENFitâ design, and launch a new type of
and shape and will be physically incompatible with Luer connectors enteral syringe design in its place (http://www.bd.com/hypodermic/
and the other new small bore connectors. Many manufacturers have products/enteral/). At the time of this writing, all other enteral syr-
informally agreed upon (ie, it is not standardized) a name and color inge manufacturers produce devices with the ENFitâ connector,
allocation for these new connections. The enteral (eg, nasogastric, or although proprietary non-Luer devices are also on the market. It will
percutaneous gastric or jejunal feeding tubes) connectors may be remain to be seen whether the Becton Dickinson enteral syringe will
known as ENFitâ, and devices may have purple coloring of the syr- retain compatibility with other manufacturers’ products.
inges and tubing (Figure 2). These have been introduced throughout Similar problems are expected to arise with clinical introduction
the USA and Europe beginning in 2016.29 The neuraxial connectors of the NRFitTM connector for neuraxial devices. For example, as
(eg, perineural infiltration, epidural, and intrathecal administration) the NRFitTM has a smaller internal diameter lumen than its Luer
may be known as NRFitTM, may have yellow coloring of syringes and counterpart, flow rates of cerebrospinal fluid during diagnostic or
tubing, and are expected to be introduced into clinical practice in therapeutic (eg, intrathecal chemotherapy) lumbar punctures will
2017. As the use of colors purple and yellow is not standardized, decrease, thus possibly affecting clinical practice.32,33 These proce-
there are on the market devices using these colors that are not nec- dures may take longer to perform and could result in patient
essarily compliant with ISO 80369 series, and may be used for other safety concerns if clinicians do not wait long enough to collect
clinical applications. adequate amounts of fluid, or falsely abandon the procedure think-
Because some patients have an indwelling enteral device such as ing the correctly placed needle has been misplaced into another
a percutaneous feeding tube, manufacturers have also made avail- space.
able adaptors for enteral devices that facilitate connection between The ISO 80369 standard for the connector for limb cuff inflation
old and new systems during the transitional phase. The use of transi- in adult blood pressure measurement has been published, and is
tion sets or adaptors is not planned for neuraxial devices. based on a currently available design with devices. Compliant cuffs
and tubing are already on the market. There are plans for additional
designs in ISO 80369-5 for inflation tourniquets for limbs and for
neonatal/pediatric blood pressure cuffs. In the coming years, addi-
tional connectors in the ISO 80369 series will also be introduced for
respiratory equipment, such as oxygen tubing with different connec-
tors for high pressures (eg, oxygen to facemask) and low pressures
(eg, gas sampling in anesthesia systems).
Despite the legislation in California mandating use of different
connectors for enteral and neuraxial tubing by the beginning of
2017, to date rollout has been inconsistent between centers, yet
F I G U R E 2 New ENFitâ connector to be used for connecting
slow and steady transitions are occurring. The most consistent road-
enteral feeds (Courtesy of, and used with permission from GEDSA)
block to implementation is finding consistent supply of devices with
LITMAN ET AL. | 11

the new connectors from manufacturers, especially in larger hospital 3. Milamed DR, Brown K, Murphy E. Luer’s lure: from an international
systems that may have constraints on equipment suppliers (Personal standards perspective. Anesthesiology. 2012;117:1358-1363.
4. Seletz JM. Flugtag-88 (Ramstein Air Show Disaster): an army
communication, Rory Jaffe MD, MBA, Executive Director, California
response to a MASCAL. Mil Med. 1990;155:152-155.
Hospital Patient Safety Organization). 5. ISO 10555-1:1995 (and amendments 1999, 2004): Sterile, Single-
Use Intravascular Catheters –Part 1: General Requirements. In: Inter-
national Organization of Standardization (ISO), ed. Geneva, 1995.
6. U.S. Food and Drug Administration. Preventing Tubing and Luer Mis-
2 | SUMMARY
connections. 2015. https://www.fda.gov/MedicalDevices/Productsa
ndMedicalProcedures/GeneralHospitalDevicesandSupplies/Tubinga
Wrong route medication errors due to tubing misconnections are ndLuerMisconnections/ucm313322.htm. Accessed October 23,
potentially life-threatening complications that have been made pos- 2017. Accessed June 16, 2016.
7. ECRI. Top 10 health technology hazards for 2011. Health Dev.
sible by the universal use of the Luer connector. The new ISO
2010;39:393-394.
80369 series of standards for small bore connectors have been 8. The Joint Commission. Tubing misconctions—A persistent and
developed to reduce the risk of these types of erroneous connec- potentially deadly occurence. Sentinel Event Alert. 2014;53:1-6.
tions. Tubing connectors for different routes of clinical application 9. Simmons D, Symes L, Guenter P, Graves K. Tubing misconnections:
normalization of deviance. Nutr Clin Pract. 2011;26:286-293.
will contain differently designed connectors that are physically
10. Harris G.U.S. Inaction Lets Look-Alike Tubes Kill Patients. The New
incompatible. However, design and manufacturing standards have
York Times. 2010. http://www.nytimes.com/2010/08/21/health/pol
progressed slowly, and clinical roll-outs have been slow, despite icy/21tubes.html?pagewanted=all&_r = 1. Accessed October 23,
the implementation of California laws to promote their use. We 2017.
11. Do€ ring M, Brenner B, Handgretinger R, Hofbeck M, Kerst G. Inad-
are currently in a state of transition where new connectors for
vertent intravenous administration of maternal breast milk in a six-
enteral and neuraxial use will be entering the clinical market in
week-old infant: a case report and review of the literature. BMC Res
the next few years, and increasing use will identify additional Notes. 2014;7:1-4.
patient safety concerns. A supplementary podcast that provides an 12. Ajmal M. Accidental intrathecal injection of aminophylline in spinal
overview of this article as well as discussion of the below reflec- anesthesia. Anesthesiology. 2011;114:998-1000.
13. Mohseni K, Jafari A, Nobahar MR, Arami A. Polymyoclonus seizure
tive questions is available here: [https://youtu.be/5PIF9JYqZOE].
resulting from accidental injection of tranexamic acid in spinal anes-
thesia. Anesth Analg. 2009;108:1984-1986.
14. Reddy GK, Brown B, Nanda A. Fatal consequences of a simple mis-
REFLECTIVE QUESTIONS
take: how can a patient be saved from inadvertent intrathecal vin-
cristine? Clin Neurol Neurosurg. 2011;113:68-71.
1. Does use of the universal Luer connector predispose to wrong
15. D’Addario A, Galuppo J, Navari C, et al. Accidental intrathecal admin-
route medication errors? istration of vincristine. Am J Forensic Med Pathol. 2010;31:83-84.
2. What are the new ISO 80369 small bore connector standards? 16. Townley KR, Lane J, Packer R, Gupta RK. Unintentional infusion of
3. How will you monitor patient safety concerns when new small phenylephrine into the epidural space. A&A Case Rep. 2016;6:124-
126.
bore connectors are used clinically?
17. Beckers A, Verelst P, van Zundert A. Inadvertent epidural injection
of drugs for intravenous use. A review. Acta Anaesthesiol Belg.
2012;63:75-79.
ETHICAL APPROVAL
18. Singh S, Rejai S, Antongiorgi Z, Gonzalez N, Stelzner M. Misconnec-
tions in the critically Ill: injection of high-dose gadolinium into an
No ethical board approval was required for this review article.
external ventricular drain. A&A Case Rep. 2016;6:121-123.
19. Birnbach DJ, Brull SJ, Prielipp RC. J’Accuse! failure to prevent epidural
and spinal catheter misconnections. A&A Case Rep. 2016;6:107-110.
CONFLICT OF INTEREST
20. Gelb AW, Mainland P, Phillips P, et al. J’agis! Anesth Analg.
2016;123:798-799.
The authors claim no conflicts of interest pertaining to the material
21. U.S. Food and Drug Administration. Medical Device Safety: Examples
discussed within this review. of Tubing and Luer Misconnections. https://www.fda.gov/medicalde
vices/productsandmedicalprocedures/generalhospitaldevicesandsupplie
s/tubingandluermisconnections/ucm313275.htm. Accessed October
ORCID 23, 2017.
22. NHS. Safer spinal (intrathecal), epidural and regional devices. 2011.
Ronald S. Litman http://orcid.org/0000-0001-8626-8267
http://www.nrls.npsa.nhs.uk/resources/?entryid45=94529. Accessed
October 23, 2017
23. Update on ISO 80369 series of Small Bore Connectors in medical
REFERENCES devices. The Association of Anaesthetists of Great Britain and Ire-
land (AAGBI). https://www.aagbi.org/sites/default/files/Neuraxial-
1. Brown J. Life and Death of the Luer. 2012. https://www.mddionli FAQ-Version-2017-03-23.pdf. Accessed October 23, 2017.
ne.com/life-and-death-luer. Accessed October 23, 2017. 24. AAMI/FDA. Infusion Device Summit. 2010. http://s3.amazonaws.c
2. Dickinson B. BD History. 2017. http://www.bd.com/en-us/compa om/rdcms-aami/files/production/public/FileDownloads/Summits/AAM
ny/about-bd. Accessed October 23, 2017. I_FDA_Summit_Report.pdf. Accessed October 23, 2017
12 | LITMAN ET AL.

25. Sentinel Event Alert. 2014. http://www.jointcommission.org/assets/ 31. Kinsella SM. The shock of the ‘Nuer’: the UK experience with new
1/6/SEA_53_Connectors_8_19_14_final.pdf. Accessed October 23, non-Luer neuraxial equipment and implications for obstetric anaes-
2017. thesia. Int J Obstet Anesth. 2013;22:1-5.
26. California So. Assembly Bill No. 444, Chapter 198. An act to amend 32. Kinsella SM, Goswami A, Laxton C, Kirkham L, Wharton N, Bowen
Section 1279.7 of the Health and Safety Code, relating to health M. A clinical evaluation of four non-Luer spinal needle and syringe
facilities. 2015. systems. Anaesthesia. 2012;67:1217-1224.
27. ISMP. Update on Implementation of the New ENFit Enteral Connec- 33. Sharpe P, Scott S, Gross JM. An evaluation of non-Luer safety
tor. Acute Care ISMP Medication Safety Alert. 2015. http://www. connectors for neuraxial procedures. Anaesthesia. 2013;68:562-
ismp.org/newsletters/acutecare/showarticle.aspx?id=114. Accessed 570.
October 23, 2017.
28. ISMP. Another twist in the ENFit implementation process. Acute
Care ISMP Medication Safety Alert. 2016;21:4.
29. Bledsoe S, Oschman A, Hampson K, Hoebing W, Pabst D. Transition How to cite this article: Litman RS, Smith VI, Mainland P. New
to ENFit enteral devices: special challenges for pediatric institutions. solutions to reduce wrong route medication errors. Pediatr
Am J Health Syst Pharm. 2016;73:1781-1783.
Anesth. 2018;28:8–12. https://doi.org/10.1111/pan.13279
30. Monteiro RS, Pillai A, Choi SW, Bogod D, Yentis SM. Flow charac-
teristics of Luer and non-Luer spinal needles. Anaesthesia.
2017;72:749-754.
Accepted: 11 October 2017

DOI: 10.1111/pan.13281

SYSTEMATIC REVIEW

The anatomy of the pediatric airway: Has our knowledge

changed in 120 years? A review of historic and recent
investigations of the anatomy of the pediatric larynx

Josef Holzki1 | Karen A. Brown2 | Robert G. Carroll3† | Charles J. Cote

4

1
Department of Pediatrics, Centre
ge, Che
Hospitaliere de Lie ^ ne
e, Belgium Summary
2
Department of Anesthesia, McGill Background: There is disagreement regarding the anatomy of the pediatric airway,
University Health Center, The Montreal
particularly regarding the shape of the cricoid cartilage and the location of the nar-
Children0 s Hospital, Queen Elizabeth
Hospital Foundation of Montreal Chair in rowest portion of the larynx.
Pediatric Anesthesia, Montreal, QC, Canada
Aims: The aim of this review is to clarify the origin and the science behind these
3
Radiology & Diagnostics, Quantitative
Imaging Inc., Largo, FL, USA differing views.
4
Harvard Medical School, MassGeneral Methods: We undertook a review of published literature, University Libraries, and
Hospital for Children, The Massachusetts
authoritative textbooks with key search words and phrases.
General Hospital, Boston, MA, USA
Results: In vivo observations suggest that the narrowest portion of the airway is
Correspondence
more proximal than the cricoid cartilage. However, in vitro studies of autopsy speci-
Josef Holzki, Department of Pediatrics,
ge,
Centre Hospitaliere Universitaire de Lie mens measured with rods or calipers, confirm that the nondistensible and circular or
Che^ ne
e, Belgium.
near circular cricoid outlet is the narrowest level. These anatomic studies confirmed
Email: [email protected]
the classic “funnel” shape of the pediatric larynx. In vivo studies are potentially mis-
Section Editor: Mark Thomas
leading as the aryepiglottic, vestibular, and true vocal folds are in constant motion
with respiration. These studies also do not consider the effects of normal sleep,
inhalation agents, and comorbidities such as adenoid or tonsil hypertrophy that
cause some degree of pharyngeal collapse and alter the normal movement of the
laryngeal tissues. Thus, the radiologic studies suggesting that the narrowest portion
of the airway is not the cricoid cartilage may be the result of an artifact depending
upon which phase of respiration was imaged.
Conclusion: In vivo studies do not take into account the motion of the highly pliable
laryngeal upper airway structures (aryepiglottic, vestibular, and vocal folds). Maximal
abduction of these structures with tracheal tubes or bronchoscopes always demonstrates
a larger opening of the glottis compared to the outlet of the cricoid ring. Injury to the lar-
ynx depends upon ease of tracheal tube or endoscope passage past the cricoid cartilage
and not passage through the readily distensible more proximal structures. The infant
larynx is funnel shaped with the narrowest portion the circular or near circular cricoid car-
tilage confirmed by multiple in vitro autopsy specimens carried out over the past century.

KEYWORDS
age, airway, child, infant, neonate, otolaryngology, techniques

†
Robert Carroll MD passed away unexpectedly on August 13, 2016.

Pediatric Anesthesia. 2018;28:13–22. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 13
14 | HOLZKI ET AL.

1 | INTRODUCTION vocal cords is larger than the cricoid ring. This observation supported
the notion of a funnel shape. He concluded that “the cricoid cartilage
The lumen of the infant larynx is described as oval, cylindrical, spher- is unquestionably the narrowest part of the pediatric larynx”, noting
ical, conical, and funnel shaped with the narrowest portion being the that this configuration could persist into adolescence. Peter11
nondistensible cricoid cartilage. Following publication of Eckenhoff’s reported additional anatomical details: (i) The cricothyroid membrane
classic paper, the descriptor “funnel shaped” was widely adopted.1 (conus elasticus) slants posteriorly impinging on the laryngeal lumen
However, recent in vivo investigations and reviews have questioned narrowing the anterior aspect of the subglottic space Figure 2A). (ii)
the shape and location of the narrowest portion of the infant In infancy, the laminae of the cricoid tilt posteriorly. As the length of
larynx,2-8 however, one in vivo study documented a circular outlet the cricoid cartilage in infants is 8.4 1.4 mm, this posterior angula-
of the cricoid ring.9 We were intrigued by these assertions that con- tion should confer an ellipsoid shape to the subglottic space.9
tradict traditional anatomic concepts.1,10-15 As a correct understand- Indeed, Peter reported that the A-P dimension of the glottis is larger
ing of airway anatomy is fundamental to the safe use of tracheal than the near circular cricoid outlet (funnel shape). During develop-
tubes in infants and children, these contradictory observations ment, the posterior laminae gradually assume an upright position
prompted our review. (Figure 2B).
In his classic paper, Eckenhoff compared the investigations of
Bayeux and Peter with his clinical observations during direct laryn-
2 | MATERIALS AND METHODS goscopy.1 With the vocal cords widely abducted, the outlet of the
cricoid ring appeared smaller than the glottis. He concluded that “in
Using the key words and phrases: “pediatric/infant anatomy of the the infant, the lamina is inclined posteriorly at its superior aspect, so
airway, larynx, dimensions, narrowest part of, pediatric laryngoscopy— that the larynx is funnel shaped with the narrowest point of the funnel
bronchoscopy, difference from adults”, we searched PubMed, the at the laryngeal exit” (Figures 2B, 3A-D). This report brought scien-
ge, 4000 Lie
University Library Lie ge, Belgium, for Historical French tific descriptions of the pediatric larynx to the attention of anesthesi-
language articles not found on the Internet, Springer Book archives, ologists and introduced the descriptor “funnel shaped” and the
the University Library Cologne, Kerpenerstrasse 62, 50937, Ger- concept of a round cricoid.
€ r Medizin, Gleuelerstrasse
many, and Deutsche Zentralbibliothek fu Butz12 examined 24 fresh unpreserved laryngeal autopsy speci-
€ln, Germany (Table 1), original research and textbooks.
60, 50931 Ko mens (premature infants to adolescents). Using dilators in 7 speci-
We investigated in vitro anatomical descriptions dating back to 1897 mens, he found that the glottis accepted sizes 4 Fr larger than the
and in vivo endoscopic and radiologic airway assessments. We con- tracheal origin; the larynx became consistently narrower from the
centrated on dimensions of the laryngeal lumen, the narrowest part glottis to the cricoid outlet (funnel shape). The cricoid ring in all
of the larynx, and the shape of the cricoid outlet. Only anatomy specimens tended to be circular in cross-sections.* Too-Chung and
reports presenting at least 2 methods of investigation (measure- Green16 investigated the rate of growth of the cricoid cartilages.
ments, calibrations, and photographs) were selected. They examined larynges (only 5/68 preserved) from 15 newborns
(1 day-1 week), 44 infants (>1 week-1 year), and 9 children (2-
15 years). They measured the interior dimension of the cricoid ring
3 | RESULTS at its lower border with a Vernier caliper. They observed a slightly
larger transverse cricoid diameter in neonates and a gradual cross-
Our search revealed autopsy data from hundreds of pediatric laryn- over to a larger A-P diameter in children until 2 years. At 5 months,
ges10-14,16-18 and one textbook.15 In vivo studies also reported data they reported a circular outlet and at 15 years, they observed aver-
from hundreds of children.2,3,6-9 All investigations examined children age A-P/transverse diameters of 1:0.97. The ratio of all groups was
from preterm to 15 years (Table 1). between 1:0.91 and 1:1.1, near circular. Tucker et al17 published
serial cross-sections of larynges from 21 children (3 months-4 years)
and demonstrated the “V”-shape of the backward tilted cricoid lami-
3.1 | Chronological in vitro autopsy reports
nae (Figure 4A). This observation explains both the funnel shape of
In 1897, Bayeux10 reported autopsy data from 28 infants and chil- the cricoid lumen and the posterolateral endocricoid ulcerations
dren. After abandoning plaster casts which distorted the anatomy, commonly associated with prolonged tracheal intubation.15,19 Holin-
he used wax castings and calibration rods which proved to be more ger et al15 expanded upon these observations based upon combined
accurate; as wax liquefies at 40°C it molds to surfaces without dis- endoscopic experiences from GF Tucker and B. Benjamin.17,19 They
tortion or dilatation. In all specimens, the A-P dimension of the glot- concluded that: “The lumen of the glottis is somewhat pentagonal in
tis was larger than the cricoid ring (Figure 1A). Calibration rods shape, as the vocal folds taper inferiorly into the subglottic larynx,
demonstrated that the nondistensible cricoid outlet was the narrow- where the lumen is elliptical with the greater diameter in the A-P
est part of the larynx and that it was nearly circular. To further
assess the caliber of the glottis, he divided the arch of the cricoid
*In anatomy, there never exist geometrical structures like rings, funnels, or ellipses, only
ring (Figure 1B), confirming that the transverse dimension of the similarities to these geometrical forms.
HOLZKI ET AL. | 15

T A B L E 1 Summary of in vitro and in vivo studies with comments and conclusions

Cricoid-A-P/transverse diameter
Author (y) N Age ratio (r) and conclusions Comments
In vitro studies
Bayeux (1897)10 28 4 mo-14 y Calibration, splitting of cricoid arch, “The cricoid cartilage is
caliper measurements. unquestionably the narrowest
part of the pediatric larynx.”
Peter (1936)11 15 1 d-17 y r = 1/1.1 The cricoid outlet has a near circular shape
(moderately larger in the transverse dimension).
He confirmed that the A-P dimension of the
glottis is larger than the cricoid outlet (funnel shape).
Butz (1968)12 18 1 d-1 y “In all cases the glottis accepted
a dilator at least 4 Fr larger that
the tracheal origin” “the larynx
seems to consistently funnel
down to the opening diameter
of the trachea.” (7 infant larynges)
Too-Chung & 15 1 d-1 wk r = 1/1.08 The cricoid is near circular, described “It was noticeable immediately
Green (1974)16 with a slightly larger transverse cricoid diameter that the child’s cricoid is elliptical
in neonates with crossover at 2 y to near circular at 15 y in shape, the coronal diameter
being greater.”
52 2 wk-1 y r = 1/0.91 Slightly larger A-P diameter
9 2-15 y r = 1/1.02 (circular diameter)
Tucker et al (1977) 17
21 3 mo-4 y “V” shape of backward tilted lamina
Eckel et al (1999)20 43 1-60 mo They measured glottic length in mm; subglottic
cartilaginous cross-section, subglottic airway
and tracheal airway in mm2. All cross-sections
were shorter than glottic length, tracheal diameters
always larger than the subglottic airway, showing the
subglottic airway as the narrowest part of the
pediatric larynx
Sellars & Keen 21 1 d-2 wk r = 1:1.13 in 16/21 r = 1:0.87 in 5/21 This large contrast
(1990)18 between A-P/transverse diameters has never been
reported in the literature
Holinger (1997)15 Large experience over more than 20 y of airway “When viewed from the side, the
endoscopy and teaching together with Tucker laryngeal lumen is slightly larger
an Benjamin superiorly at the glottic level and
narrower at the inferior aspect
of the cricoid cartilage” “The
cricoid ring is a smooth round circle”
Fayoux et al (2006)13 150 Preterm-3 mo “The narrowest part of the airway was
the cricoid area. . .in each age group.”
Fayoux et al (2008)14 274a 15-41 wk There was no significant difference between “The diameter of the cricoid lumen
gestation the A-P and lateral diameter of the cricoid (circular) was significantly less than that of
the trachea and glottic lumen.”
26a 1 d-1 m Same ratios as in the premature group
Total examinations 672
In vivo studies
Litman et al (2003)2 99 2 mo-13 y “The narrowest portion of the larynx
was the transverse dimension at the
level of the vocal cords.” “Transverse
dimensions were narrower than A-P
dimensions at all levels of the larynx
above the cricoid ring and in most
children at the cricoid ring.” “The
cricoid ring is functionally the
narrowest portion of the larynx.”

(Continues)
16 | HOLZKI ET AL.

TABLE 1 (Continued)
Cricoid-A-P/transverse diameter
Author (y) N Age ratio (r) and conclusions Comments
Dalal et al (2009) 3
128 6 mo-13 y Videos at the glottis and superior aspect of cricoid “Our study reveals the Cricoid-A-P
ring were obtained. Pictures were plotted on graph dimension is larger than its
paper. Mean cross-sectional areas of cricoid and Cricoid-transverse dimension
paralyzed glottis were compared (cricoid ellipsoid). suggesting that the cricoid is ellipsoid
rather than round.” “in anesthetized
paralyzed children, the glottis is
narrower than the cricoid from
infancy to adolescence.”
Wani et al (2016)6 130 1 mo-9 y r = 1/1.02 (practically circular) “The cone shaped airway characteristic,
which has been historically proposed,
was not observed. Given that the
subglottic transverse diameter is the
smallest area dimension, one must
assume this is the most likely area of
resistance for the passage of an
endotracheal tube rather than only
the cricoid.” “The subglottic area and
the cricoid change from an elliptical
to a round (circular) shape.”
Wani et al (2017)7 40 1 d-12 moc r = 1/1.36 at subglottic region. r = 1/1.098 “Increase in transverse dimensions
at cricoid level (near circular) observed from subglottis to
cricoid. . . A-P dimensions showed a
decrease from subglottis to the
cricoid.” “The mean cross-sectional
area at the 2 levels were similar.”
“The cricoid is not round as has been
observed in older children” “The ratio
between the transverse and A-P
diameters at the cricoid was 0.89.”
Wani et al (2017)b,8 54 2 mo-8 y Air volumes in the subglottic, cricoid to the tracheal “An increase in airway volumes was
regions were examined. The coaxial view of the observed from the subglottis
airway from the caudad end shows a funnel shape (0.17 mm3) to the cricoid (0.19 mm3).”
Wani (2016)b,21 102 1 mo-10 y r = 1:1.0 overall but ranged from 0.91 to 1.06.
(9 age Age groups 2-3, 6-8 y had a moderately larger
groups) transverse than A-P diameter of the cricoid
(near circular)
Total 553
a
Note that 150/300 larynges were also reported in their 2008 publication (personal communication).
b
It is not known if some of these patients were included in prior publications.
c
Note that 11 patients were reported in a prior study.

dimension. At the inferior aspect of the cricoid cartilage the lumen and in 5 specimen transverse diameters were considerably larger than
the cartilage are a smooth round shape (Figure 3B,C,D)”. The authors the A-P diameters (different from Too-Chung). Thus, the A-P/trans-
illustrated the 2-dimensional inverted funnel shape in coronal sec- verse ratios in this age group may vary. Eckel et al20 examined 43
tions as typical for adducted vocal cords in autopsy specimens. They larynges (<5 years) by the method of plastination which preserves
further described the glottic level as pliable and larger than the rigid the larynx as it appears in situ and found that cricoids tended to be
cricoid outlet. To have a holistic view of the lumen of the larynx, A- circular. Mid-larynx cross-sections proximal to the cricoid arch
P and coronal sections of the larynx should be combined with the demonstrated the transverse diameter to be wider than the glottis
cross-sections of the larynx at different levels (Figure 3); the authors (Figure 4B). They also found that the transverse diameter of the
confirmed the common site of injury from tracheal tubes and the cadaveric glottis was narrower than the A-P dimension, resulting in
autopsy work of previous investigators.10-12,16,17 an almost triangular shape (Figure 4B). This is consistent with what
Sellars and Keen18 reported A-P/transverse diameters of 21 is expected as in death, the soft tissues of the upper airway revert
autopsy specimens (newborn-2 weeks) at the inner aspect of the cri- to their neutral position (Figure 3B) similar to several in vivo reports
coid ring (measurements higher in the larynx were not made). In 16/ (see further).6,7 More recently, Fayoux et al13 examined serial sec-
21 specimens, the A-P diameter exceeded the transverse, however, tions from 150 nonpreserved infant larynges (preterm-3 months)
HOLZKI ET AL. | 17

(A) (B) the cricoid outlet and that the cricoid arch and posterior lamina have
a 4-fold height discrepancy which creates an oblique entrance to the
cricoid ring (Figure 4C). Therefore, methods which identify the cri-
coid cartilage by the tip of the lamina vs the superior aspect of the
anterior cricoid arch are likely to demonstrate different measure-
ments (Figure 3). All outlets of the investigated larynges were round
with A-P/transverse ratios between 1/0.99 and 1/1.05.

3.2 | In vivo clinical reports

Litman et al2 reviewed laryngeal MRIs in 99 anesthetized, sponta-
neously breathing children (2 months-13 years). The A-P/transverse
diameters of the larynx were measured at the vocal cords and cri-
coid level. Transverse diameters increased linearly in a caudad direc-
tion through the larynx, while A-P diameters did not change much
F I G U R E 1 A, Soot print from a wax casting (Bayeux) showing a relative to the different laryngeal levels. The larynx was described as
lateral view of pediatric larynx from a 5-year-old child. The A-P
conical in the transverse (coronal) section with the apex at the vocal
diameter at the level of the cricoid outlet is about 44% smaller than
at the level of the vocal cords. B, Depiction of the anterior view of cords and as cylindrical in the A-P dimension. The narrowest por-
pediatric larynx from another 5-year-old child by Bayeux. A tions of the larynx were at the glottic opening and the immediate
measuring rod could not exit the outlet of the cricoid ring until the sub-vocal cord level but functionally, the cricoid ring was described
arch of the cricoid ring was divided, demonstrating that the outlet of as the narrowest portion of the larynx. In the horizontal view, they
the cricoid ring is the narrowest part of the pediatric larynx
described the subglottic superior aspect of the cricoid ring as ellip-
soid. They also commented that contraction of the laryngeal muscles
within 6 hours after death (near physiologic conditions). They used 3 influenced the dimension of the larynx above the cricoid ring. Dalal
methods to evaluate the laryngeal lumen: (i) tracheal tubes (analo- et al3 photographically estimated the cross-sectional areas of the lar-
0 ynx at the glottis and superior aspect of cricoid ring in 128 children
gous to Bayeux s calibration rods); (ii) nondilatable, cylindrical bal-
loons, recording the change in pressure as the balloon was moved under anesthesia and neuromuscular blockade. They reported that
along the length of the larynx; and (iii) determination of the A-P/ the relaxed glottis was narrower than the cricoid ring. They also
transverse diameters with calipers of the cricoid ring and the inter- found that the infant larynx was cylindrical in the A-P dimension but
arytenoid distance (IAD) at the vocal processes at the posterior glot- conical in the transverse dimension. Interestingly, the authors stated:
tis. The narrowest diameters in anteroposterior and transverse “Further studies are needed to determine whether these static airway
laryngeal plane (28 weeks gestational age (GA) until the 3rd month measurements in anesthetized and paralyzed children reflect the
of life) was the cricoid ring (Fig. S1) (funnel shape). Building upon dynamic characteristics of the glottis and cricoid in children”, acknowl-
other investigations,12,15 these measurements provide important edging differences between in vivo and in vitro studies.
insights regarding the mechanism of injuries during airway instru- Wani et al6 have reported several studies. They initially reported
mentation: (i) the risk for glottic injury increases if tracheal tubes observations from 130 laryngeal CT scans in children (1 month-
with an OD larger than the IAD are introduced in the larynx; and (ii) 9 years); a description of the glottis was not part of that study. All
the risk for injury at the level of the posterior cricoid cartilage children were under natural sleep, sedation, or sevoflurane anesthe-
increases if a tracheal tube with an OD smaller than the IAD, but lar- sia. They estimated the A-P/transverse dimensions from immediately
ger than the outlet of the cricoid ring is introduced.19 In a second below the vocal cords and at the level of the cricoid ring; the lumen
series, they examined cricoid and tracheal diameters from 274 non- of the larynx was ellipsoid immediately below the vocal cords (A-P
fixed larynges (15-41 weeks GA) and 26 infants (0-1 month) within 9.2 mm vs transverse 7.5 mm) changing to a circular shape at the
6 hours after death.14 They concluded that “Cricoid diameters were level of the cricoid outlet (A-P 8.5 mm vs transverse 8.3 mm, a cir-
always significantly less than tracheal diameters and IAD” and “there cle). Thus, the A-P dimension at the subglottic level was greater than
was no significant difference between the A-P and lateral diameter of that of the cricoid but the transverse diameter of the cricoid was lar-
the cricoid.” This study confirmed that throughout intrauterine life ger than at the subglottic level; the cross-sectional areas at the 2
until the first month of life that the cricoid ring is narrower than the levels were not different. They concluded: “The cone shaped airway
glottis (funnel shape). They further documented the triangular- characteristic, which has been historically proposed, was not observed.
shaped lumen of the glottis in all specimens. Given that the subglottic transverse diameter is the smallest area
‡
Unpublished observations from 2 authors (JH and MHR) of 16 dimension, one must assume this is the most likely area of resistance
autopsy specimens (0-7 months), demonstrated a ratio of 1/1.02 at for the passage of an endotracheal tube rather than only the cricoid.”
These authors published a similar study examining the ratio of the
‡
MA Rothschild, Institution of Legal Medicine, University of Cologne, Germany. cricoid cartilage to the left mainstem bronchus from 102 CT scans
18 | HOLZKI ET AL.

(A) (B)

F I G U R E 2 A, Endoscopic picture (2.2 mm Hopkins lens, JH) of the glottis of a 1-year-old child, spontaneously breathing under inhalation
anesthesia. Note that the vocal cords are in an almost parallel (adducted) position. The anterior subglottic narrowing is caused by the
posteriorly slanting thyroid cartilage and the cricothyroid membrane, narrowing the subglottic space anteriorly. B, Drawing showing the lateral
view of an A-P section through a neonatal larynx (autopsy specimen by Peter).11 The backward tilted lamina results in a cricoid lumen which
has a large, slanting entrance and an approximately 46% smaller circular outlet. The height ratio between arch and lamina is 1:4 (see also
Figure 4C)

(B)

(A)

(C)

(D)

F I G U R E 3 Influence of the level of cross-sections on the configuration of the laryngeal lumen, regardless of the method of investigation. A,
A-P section of the larynx of an 8-month-old infant specimen (JH). B, Cross-sections at the vocal cord level from autopsy specimens (Holinger
et al, 8-month-old infant, and Eckel et al, 3-month-old infant, with permission by Wolters Kluwer and Springer).15,20 “Tear-drop”-like entrance
to the glottic opening (CT-scan by Sirisopana et al,9 child <3 y, with permission of Wolters Kluwer) showing a clear tissue-air interface but no
cartilaginous structures. C, Cross-sections at mid-level of the larynx show an oval shape. The posterior V-shape of the lamina is visible in these
anatomical cross-sections (Holinger,15 Eckel, courtesy of the author).20 The hyperdense lining of the lumen in the MRI (not depicting
cartilages!) represents the intensely perfused laryngeal mucosa (Litman et al2 with permission by Wolters Kluwer). D, Cross-sections through
the outlet of larynx showing a circular configuration (Holinger et al15 Eckel, courtesy of the author.20 CT-scan (Sirisopana et al9 with
permission by Wolters Kluwer)

from children 1 month to 10 years of age.21 This report did not transverse diameter 1.02/1) similar to their previous study.6 Another
assess subglottic dimensions but confirmed the near circular cricoid study7 reported observations of 40 laryngeal CT scans of infants (0-
with the A-P diameter slightly greater than the transverse (A-P/ 12 months) during similar conditions as the studies above (11 had
HOLZKI ET AL. | 19

(a) (b) (c)

F I G U R E 4 Influence of the level of cross-section on the configuration of the laryngeal lumen, regardless of the method of investigation. A,
Cross-section above the arch of the cricoid ring of an infant larynx at the level of the anterior ligamentous cricothyroid membrane (Tucker
et al17 with permission from Rights Link). The V-shaped lamina of the cricoid, bounding the oval lumen of the mid-larynx by two-thirds from
posterior, demonstrates an “elliptical” mucosal lumen at this level, obscuring the underlying surface” of the cricoid. B, Cross-section at the level
of the vocal cords (2-month-old infant, plastination technique, Eckel et al20 with permission of Springer). The cranial part of lamina and vocal
processes of the arytenoid cartilages are clearly visible, demonstrating the tight connection between the vocal cords, the vocal process, and
the superior lamina. C, Photo of a freshly trimmed cricoid ring of a neonate (autopsy specimen, courtesy MA Rothschild, Institution of Legal
Medicine, University Cologne, Germany, from unpublished data). The high lamina, the short arch, the oblique entrance of the cricoid ring, and
of the plane level of the circular cricoid outlet (see Figures 2B and 3A) demonstrate the difficulty of precisely defining the anatomy of the
pediatric cricoid ring by cross-sections only

been previously reported to have a circular outlet5). They found that nondistensible cricoid cartilage (funnelists).10-14,16-18 How can we
the cricoid had a larger A-P than transverse diameter (6.7 vs reconcile these apparent contradictory observations? Is the larynx
6.1 mm) but at the subglottic level, they again found that the trans- funnel-shaped or cylindrical? Is the narrowest portion at the level of
verse diameter was less than the cricoid and the A-P diameter was the glottis or the cricoid ring? We believe that these divergent views
greater (7.2 A-P vs 5.3 mm transverse). They concluded that the air- reflect the methods of evaluation used (in vivo vs in vitro). Autopsy
way in neonates and infants between the subglottic area and the cri- data confirm the nondistensible cricoid ring10-15,20 as the narrowest
coid remains ellipsoid, “that the airway is wider anteroposteriorly and portion, however, the laryngeal lumen varies in form and size at dif-
narrows in the transverse dimension from the subglottis to the cricoid in ferent levels (predominantly cylindrical in transverse but conical in
infants.” However, the mean cross-sectional area at the 2 levels was A-P dimensions). As the pediatric larynx consists of pliable (supra-
similar (29.9 vs 32.1 mm3). It is possible that this ellipsoid shape is glottis, glottis, and proximal subglottis) and rigid parts (the cricoid
secondary to the slices not being exactly perpendicular to the axis of ring) (Figures 3, 4C), describing laryngeal anatomy depends upon the
the cricoid. They also acknowledged that they did not standardize methodology used for assessment. Studies which employ horizontal
the phase of respirations which might also have impacted their mea- cross-sections to identify the level of the cricoid outlet by CT scans
surements by the variable position of the vocal cords. A fourth study or MRI are less precise than autopsy specimens for several reasons:
reported 3D CT scan reconstructions of the upper airway at 3 mm a slice thickness of 2.5-3.0 mm, the angulation of the cricoid, and
intervals from 54 children 2 months to 8 years8 but only 4 were inability to precisely show the cartilaginous structures.10-12,15,16
infants. The airway volumes in subglottic and cricoid levels were More importantly, these measurements may vary greatly in vivo
0.17 0.06 vs 0.19 0.07 mm3, respectively. The authors com- compared with in vitro because the laryngeal tissue folds (vocal,
mented that the narrowest portion of the airway was subglottic but vestibular, and aryepiglottic folds) are dynamic, pliable structures13
“we cannot comment whether this region or others such as the cricoid that change position and shape with the phase of respirations
represent the most rigid aspect of the airway.” (change in transverse but minimal in the A-P dimension).7,8,22 With
inspiration, the laryngeal muscles contract, stretching open the
aryepiglottic, vestibular (false cords), and vocal folds, so that the dis-
4 | DISCUSSION tance between these structures increases (Figure 5A). With expira-
tion or neuromuscular blockade (Figure 5B), these soft tissue folds
Our literature review revealed 2 views of the infant larynx: the tradi- return to a neutral position by the stability of the cuneiform and cor-
tional “funnelists” and the “nonfunnelists”. The in vivo studies2,3,6-8 niculate cartilages.23 With voluntary glottic closure (lifting heavy
all suggest that the narrowest portion of the pediatric larynx in spon- weights), or involuntary glottic closure (laryngospasm), contraction of
taneously breathing sleeping or anesthetized infants and children is the intrinsic laryngeal muscles closes the larynx at 3 levels
at the laryngeal inlet (nonfunnelists), whereas the in vitro studies (aryepiglottic, vestibular, and vocal folds). Thus, the various tissue
describe the narrowest portion of the pediatric larynx as the folds of the larynx are in constant motion and the distance between
20 | HOLZKI ET AL.

them depends upon the phase of respiration during which the exami- tracheal tubes)10,13,14; the maximum IADs were always larger than the
23
nation was made (or death). Although the paralyzed vocal cords are rigid cricoid outlet (Figure 6A,B). When studying the laryngeal anat-
always in a near cadaveric position (Figure 5B), they never impede the omy from the viewpoint of intubation, the only meaningful way to
advancement of tracheal tubes. A further consideration is what hap- compare the lumen at various levels of the larynx, is by gently intro-
pens to these structures during sleep or anesthesia compared to the ducing measuring rods (or tracheal tubes)10,12,13 which maximally dis-
awake state. It is well known that there is collapse of pharyngeal tis- tend the soft tissues of the laryngeal folds without injury.
sues during natural sleep and during anesthesia. Such narrowing of the Wani et al6-8 have presented several studies with possible overlap-
upper airway facilitates apposition of the laryngeal tissue folds further ping patient populations. Given that several hundred CT scans were
narrowing the upper airway above the cricoid cartilage, particularly if retrospectively reviewed and only 40 selected for one study (11 previ-
there is accompanying tonsillar or arytenoid hypertrophy.24,25 The ously reported) and 56 for another, it is possible that observer bias
effects of sleep, anesthetic medications, or arytenoid/tonsillar hyper- could have influenced their conclusions regarding the shape of the cri-
trophy were not considered in the in vivo studies. coid and the subglottic region. It is also unclear if the CT cuts were
The maximum interarytenoid diameter (IAD) can be measured obtained at sufficiently thin intervals to allow accurate assessment the
in vivo, but in vitro measurements may be more accurate with mea- cricoid outlet. Wani et al8 did not cite the 300 investigations by Fay-
13
suring rods or calipers as these tissue folds are easily distended. oux et al who documented the opposite of their assertions (Fig. S1).
After neuromuscular blockade, the vocal cords relax and assume a Furthermore, in their most recent study of 3D CT-imaging, they found
parallel position (Figure 5B). Transverse interarytenoid diameters a mean airway volume difference between the subglottic area and the
(IAD) vary from zero (laryngospasm) to a maximum diameter when cricoid outlet of only 0.02 mm3 which is of uncertain clinical rele-
13
fully abducted. In autopsy specimens, the vocal cords adopt a vance. Furthermore, two of their other studies found no difference in
cadaveric position; in vivo they are adducted to different degrees cross-sectional areas of these same 2 levels.6,7 These differences in
depending upon the phase of respiration but rarely maximally the transverse measures between the subglottis and the cricoid are
abducted as seen in Figure 5A. likely clinically unimportant in contrast to the A-P dimensions (Fig-
Litman et al,2 Dalal et al,3 and Wani et al6-8 presented age related ure 6B). It is unclear to us why the resistance to tracheal tube passage
but seemingly constant A-P diameters of glottis and proximal subglot- would differ with similar cross-sectional areas. Wani et al’s8 observa-
tis. Litman specifically stated that “active contraction of the laryngeal tions of the cricoid cartilage are similar to Peter0 s; there appear to be
muscles influenced the dimension of the larynx”. Wani et al commented minute differences of cricoid dimensions from study to study that
that they did not control for the phase of respirations; the patients in overall indicate that the cricoid outlet is nearly circular.
the Dalal et al study were paralyzed. These investigators reported
upper transverse laryngeal diameters narrower than those from
autopsy specimens. We assume that this is because their methodolo- 5 | CONCLUSION
gies did not consider the conditions at the time of evaluation, ie, the
pliability of the tissues in vivo and respiratory related motion of glottis We reviewed 9 autopsy studies published over the last century
and proximal subglottic structures. In contradistinction, autopsy stud- that documented the narrowest portion of the infant larynx as the
ies measured A-P diameters of the glottis and cricoid outlet in hun- nondistensible cricoid cartilage, supporting the concept that the
dreds of specimens with calipers or rods providing the most accurate infant larynx is in fact funnel-shaped.10-16,18,20 Six in vitro studies
means for determining the narrowest level of the larynx (similar to described the cricoid cartilage as circular or nearly circular10-15; two

(A) (B)

F I G U R E 5 A, Fully abducted vocal cords during inspiration under inhalation anesthesia for functional checkup. Note the pronounced
cricothyroid membrane and the tapering down to the narrower outlet of the cricoid ring (JH). B, Adducted vocal cords after neuromuscular
blockade (JH), ie, return to neutral position from the stabilizing effect of cuneiform and corniculate cartilages
HOLZKI ET AL. | 21

(A) (B)

F I G U R E 6 A, Reconstruction of an infant larynx based on fresh autopsy specimen (see Figure 4C) and investigations by Bayeux, Peter, and
Eckel. An oval-shaped lumen of the cricoid can be described only above the cricoid arch (=oblique, oval entrance to the cricoid ring) which
tapers down to the narrower, circular outlet, being narrower than the glottic and cricoid inlet. Thus, a tracheal tube which is passed through
the cricoid is always located posteriorly within the pediatric larynx because of the funnel shape (always a smaller outlet at the cricoid
compared with the more proximal, larger, and distensible structures) which forces the tracheal tube posteriorly and limits the amount of leak
behind the tube (see Fig. S2). B, Lateral neck xerogram of a 2-day-old term infant (CJC). The image demonstrates clearly the more posteriorly
located position of the cricoid outlet and the overall funnel shape

in vitro studies reported an ellipsoid cricoid outlet.16,18 Thus, there We therefore submit that the original conclusions of Bayeux in
is some disagreement about the actual shape of the cricoid ring. 1897 supporting the “funnelist viewpoint” remain valid today10:
Two studies described the cricoid ring as V-shaped in the posterior “If the intubating hand feels a small resistance against the passing
part and a near circular cricoid outlet without comparing it with of a tube, it is not caused by the vocal cords but the cricoid ring. If one
the glottic level15,17 but providing important information about the wanted to grant the active vocal cords within a surrounding of pliable
cricoid structure. Two in vivo studies described the cricoid as ellip- muscles a greater importance for tube selections than the non-dilatable
3,21
soid in shape but 4 in vivo studies reported it as circular or cricoid ring, it would mean to support a theory that the resistance of
nearly circular2,8,9,21; most in vivo studies describe the cricoid as the perineum is larger (for the passing newborn head) than that of the
either circular or near circular which permits safe intubation with entrance of the pelvis)”.10
round tracheal tubes which contrary to one review, permit a mini-
mal leak during mechanical ventilation (Fig. S2).4 Five in vivo stud-
ACKNOWLEDGMENTS
ies reported that the narrowest portion of the larynx was proximal
to the cricoid cartilage2,3,6-8 but there was no difference in cross- Hans Hoeve MD, pediatric otorhinolaryngologist (Sophia Children0 s
6-8
sectional area or volume when the 2 levels were compared. Hospital Rotterdam) contributed with his knowledge of the anatomy,
It is clear that the nonfunnelists believe that the cricoid outlet is physiology, and pathology of the pediatric airway to this article.
larger than the subglottic region; however, they did not fully con-
sider in vivo laryngeal dynamics or the known effects of sedatives
ETHICAL APPROVAL
and sleep upon upper airway tone and secondary mild obstruction.
As there is enormous pliability of the laryngeal tissue folds, the resis- The photos of the endoscopic pictures and the autopsy specimen
tance to passing a tracheal tube through the vocal cords and struc- were anonymously archived according to the ethical standards of
tures above the cricoid arch is very low. Thus, these structures can the respective institutions.
be actively opened (Figure 5A), whereas the cricoid ring is rigid and
nondistensible. The real issue is not what appears to narrow the lar-
CONFLICT OF INTEREST
ynx by continuous motion of soft tissues above the cricoid but
rather the actual narrowest nondistensible portion of the larynx The authors report no conflict of interest.
where intubation injuries are most likely if too large tracheal tubes
are placed. Overall, all studies that we reviewed are in agreement
ORCID
despite differing conditions at the time of evaluation (natural sleep,
anesthesia, and autopsy). Josef Holzki http://orcid.org/0000-0002-3415-859X
22 | HOLZKI ET AL.

17. Tucker GF, Tucker JA, Vidic B. Anatomy and development of the cri-
REFERENCES
coid: serial-section whole organ study of perinatal larynges. Ann Otol
Rhinol Laryngol. 1977;86:766-769.
1. Eckenhoff JE. Some anatomic considerations of the infant larynx influ-
18. Sellars I, Keen EN. Laryngeal growth in infancy. J Laryngol Otol.
encing endotracheal anesthesia. Anesthesiology. 1951;12:401-410.
1990;104:622-625.
2. Litman RS, Weissend EE, Shibata D, Westesson PL. Developmental
19. Benjamin B. Prolonged intubation injuries of the larynx: endoscopic
changes of laryngeal dimensions in unparalyzed, sedated children.
diagnosis, classification, and treatment. Ann Otol Rhinol Laryngol
Anesthesiology. 2003;98:41-45.
Suppl. 1993;160:1-15.
3. Dalal PG, Murray D, Messner AH, Feng A, McAllister J, Molter D.
20. Eckel HE, Koebke J, Sittel C, Sprinzl GM, Pototschnig C, Stennert E.
Pediatric laryngeal dimensions: an age-based analysis. Anesth Analg.
Morphology of the human larynx during the first five years of life
2009;108:1475-1479.
studied on whole organ serial sections. Ann Otol Rhinol Laryngol.
4. Tobias JD. Pediatric airway anatomy may not be what we thought:
1999;108:232-238.
implications for clinical practice and the use of cuffed endotracheal
21. Wani TM, Rafiq M, Terkawi R, Moore-Clingenpeel M, AlSohaibani
tubes. Pediatr Anesth. 2015;25:9-19.
M, Tobias JD. Cricoid and left bronchial diameter in the pediatric
5. Motoyama EK. The shape of the pediatric larynx: cylindrical or fun-
population. Pediatr Anesth. 2016;26:608-612.
nel shaped? Anesth Analg. 2009;108:1379-1381.
22. England SJ, Bartlett Jr D, Daubenspeck JA. Influence of human vocal
6. Wani TM, Bissonnette B, Rafiq Malik M, et al. Age-based analysis of
cord movements on airflow and resistance during eupnea. J Appl
pediatric upper airway dimensions using computed tomography
Physiol Respir Environ Exerc Physiol. 1982;52:773-779.
imaging. Pediatr Pulmonol. 2016;51:267-271.
23. Fink BR, Demarest RJ. Laryngeal Biomechanics. Cambridge, MA: Har-
7. Wani TM, Rafiq M, Akhter N, AlGhamdi FS, Tobias JD. Upper airway
vard University Press; 1978:1-14.
in infants-a computed tomography-based analysis. Pediatr Anesth.
24. Litman RS, Kottra JA, Berkowitz RJ, Ward DS. Upper airway
2017;27:501-505.
obstruction during midazolam/nitrous oxide sedation in children with
8. Wani TM, Rafiq M, Talpur S, Soualmi L, Tobias JD. Pediatric upper
enlarged tonsils. Pediatr Dent. 1998;20:318-320.
airway dimensions using three-dimensional computed tomography
25. Donnelly LF, Casper KA, Chen B. Correlation on cine MR imaging of
imaging. Pediatr Anesth. 2017;27:604-608.
size of adenoid and palatine tonsils with degree of upper airway
9. Sirisopana M, Saint-Martin C, Wang NN, Manoukian J, Nguyen LH,
motion in asymptomatic sedated children. Am J Roentgenol.
Brown KA. Novel measurements of the length of the subglottic air-
2002;179:503-508.
way in infants and young children. Anesth Analg. 2013;117:462-470.
10. Bayeux R. Tubage du larynx dans le croup. Presse Medicale.
1897;6:29-33.
11. Peter K. Handbuch der Anatomie des Kindes [Handbook of the Anat- SUPPORTING INFORMATION
omy of the Child], Peter KW, Heidreich F, eds. pp. 525-590. Berlin,
Germany: Springer; 1936. Additional Supporting Information may be found online in the sup-
12. Butz Jr RO. Length and cross-section growth patterns in the human porting information tab for this article.
trachea. Pediatrics. 1968;42:336-341.
13. Fayoux P, Devisme L, Merrot O, Marciniak B. Determination of
endotracheal tube size in a perinatal population: an anatomical and
experimental study. Anesthesiology. 2006;104:954-960.
How to cite this article: Holzki J, Brown KA, Carroll RG, Cote
14. Fayoux P, Marciniak B, Devisme L, Storme L. Prenatal and early
CJ. The anatomy of the pediatric airway: Has our knowledge
postnatal morphogenesis and growth of human laryngotracheal
structures. J Anat. 2008;213:86-92. changed in 120 years? A review of historic and recent
15. Holinger LDG, Green CG. Anatomy. In: Holinger LDL, Lusk RP, investigations of the anatomy of the pediatric larynx. Pediatr
Green CG, eds. Pediatric Larnygology and Bronchoesophagology. Anaesth. 2018;28:13–22. https://doi.org/10.1111/pan.13281
Philadelphia, PA: Lippincott-Raven; 1997:16-33.
16. Too-Chung MA, Green JR. The rate of growth of the cricoid carti-
lage. J Laryngol Otol. 1974;88:65-70.
Accepted: 28 September 2017

DOI: 10.1111/pan.13273

SYSTEMATIC REVIEW

Donohue syndrome: A review of literature, case series, and

anesthetic considerations

Alana Kirkwood | Grant Stuart | Louise Harding

Department of Anaesthesia, Great Ormond

Street Hospital, London, UK Summary
Background: Donohue syndrome is a rare autosomal recessive disorder of insulin
Correspondence
Dr Alana Kirkwood, Department of resistance, causing a functional defect in insulin receptor function, and affecting the
Anaesthesia, Great Ormond Street Hospital, ability of the insulin to bind the receptor. Features include severe hyperinsulinism
London, UK.
Email: [email protected] and fasting hypoglycemia, along with severe failure to thrive despite feeding. An
accelerated fasting state results in muscle wasting, decreased subcutaneous fat, and
Section Editor: Francis Veyckemans
an excess of thick skin. A reduced thoracic diameter is accentuated by increased
abdominal distension, which impacts on respiratory reserve.
Cardiac disease develops early in life, with progressive hypertrophic cardiomyopathy
as a result of hyperinsulinism. Prognosis is poor with the majority of patients dying
in infancy of sepsis. The aim of this review is to report our experience of providing
anesthesia for patients with Donohue syndrome, and inform guidance for safe
management of these children, based on a comprehensive literature review.
Methods: A literature search was carried out using PubMed, Medline, and the
Cochrane Library, and using the MESH search terms detailed below. Patients were
identified by formal request to the department of pediatric endocrinology at Great
Ormond Street Hospital. Each patient’s notes were searched manually and electroni-
cally for both clinical presentation and outcome, and anesthesia records.
Discussion: There is currently no published literature relating to anesthetic manage-
ment of Donohue syndrome. We report a case series of 5 patients with Donohue
syndrome who have presented to our institution. This small series of children with
this complex disorder demonstrates a clearly increased risk of general anesthesia.
Many of the risks relate to restrictive lung disease and abdominal distension which
worsens with bag valve mask ventilation and limited respiratory reserve which leads
to precipitous desaturation. During induction, a spontaneously breathing technique
is recommended. If required, bag valve mask ventilation should be accompanied by
constant gastric aspiration. Intubation is challenging, and a difficult airway plan,
including a second experienced anesthetist and ENT support, should be in place.
These children are predisposed to developing cardiomyopathy and therefore at risk
of cardiovascular collapse under anesthesia.

KEYWORDS
Donohue syndrome, insulin, insulin resistance, leprechaunism, Mendenhall syndrome, receptor

Pediatric Anesthesia. 2018;28:23–27. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 23
24 | KIRKWOOD ET AL.

1 | INTRODUCTION

Donohue first described a condition he termed dysendocrinism in a What is already known

white female infant born to first cousins in 1948.1 She was born
with intrauterine growth retardation (IUGR), weighing only 1.75 kg
• There is as yet no published literature regarding anes-
thetic risk and management of children with Donohue
and presented with further weight loss on day 28 of life. She was
syndrome.
underdeveloped and poorly nourished with abundant hair, loose skin,
and obvious muscle wasting. A systolic murmur was noted. She had
What this article adds
marked abdominal distension and multiple subcutaneous abscesses
across the scalp, back, and forehead. She deteriorated rapidly and • This case series, alongside a comprehensive literature
died aged 8 weeks. review, details our institutions experience and suggests
2
Donohue and Uchida noted the same clinical features and phe- guidance for management of these children who pose
notype in the second child of the same family in 1954. The term significant anesthetic risk.
3
Leprechaunism was coined to describe this syndrome as those
affected had the appearance of these caricatures. To avoid this pejo-
rative term, it is now more widely referred to as Donohue syn- in these children is due to an accelerated fasting state, and they
drome. have a blunted response to exogenous insulin.7 These biochemical
findings are associated with hypertrophy of the pancreatic islets of
Langerhans. In addition, there is hypertrophy of other organs, includ-
2 | EPIDEMIOLOGY/INCIDENCE ing splenomegaly and hepatomegaly. Excess glycogen and iron depo-
sition with bile duct cholestasis have all been described in these
Donohue syndrome is a rare disorder with an incidence of 1 in 4 patients. They often have enlarged kidneys and nephrocalcinosis,
million live births,4,5 although this may be underreported. There is a with atrophic adrenal glands.9 There are frequent lymphoid hypopla-
strong association with parental consanguinity. It is an autosomal sia and cystic gonadal appearances.
recessive disorder caused by a functional defect in the Insulin Cardiac disease may be a feature of Donohue syndrome,
Receptor gene on chromosome 19p13. This causes insulin resistance manifesting after 1-2 months of life with significant hypertrophic
by a variety of mechanisms, and there are currently 50 described cardiomyopathy.10,11 The cardiomyopathy is attributed to supraphys-
6
mutations of the insulin receptor gene. Having been studied exten- iological hyperinsulinism and is mediated via IGF1. This can also be
7,8
sively, it is now thought the disorder results from a defect in insu-
lin receptor function, affecting the ability of the receptor to bind
insulin. There are 2 milder forms of insulin resistance with the same
mechanism: Rabson Mendenhall Syndrome and Type A Insulin Resis- BOX 1 Characteristic Features of Donohue Syndrome

tance syndrome. The prognosis is dependent on the degree of insu- Facial features Elfin like pointed chin
lin resistance; children with classic Donohue syndrome die in
Microcephaly
infancy, the majority from bronchopneumonia. In other milder forms
Low set prominent ears
of insulin resistance, patients can live into adulthood. In all syn-
Orbital hypertelorism
dromes of insulin resistance, survival may be improved by Insulin-like
Broad nose
growth factor 1 (IGF1) therapy, although its efficacy in Donohue
Thick lips
syndrome is still unclear.
Facial hair
Skin features Hypertrichosis

3 | CLINICAL AND PATHOLOGICAL Acanthosis nigricans

FEATURES Excessive thick, hyperelastic skin
Decreased subcutaneous fat
Children with Donohue syndrome present with severe intra- and Muscle wasting
extrauterine growth retardation. Intrauterine growth retardation is Other features Large hands and feet (relative to body)
typically seen acutely from 7 months gestation. Birth weights are Low body weight for age
typically <0.4th percentile. Postnatally there is severe failure to Hypotonia
thrive with marasmus and malnutrition despite feeding.
Abdominal distension
Severe hyperinsulinism combined with fasting hypoglycemia and
Reduced lateral thoracic dimensions
postprandial hyperglycemia are the hallmarks of Donohue syndrome
Hyperplasia of nipples, genitals, other organs
(Box 1). A combination of this biochemical triad and the presence of
clinical features are sufficient to make the diagnosis. Hypoglycemia
KIRKWOOD ET AL. | 25

seen transiently in other forms of hyperinsulinism such as infants of 6 | CASE SERIES

diabetic mothers where the in utero exposure to elevated insulin
levels causes fetal cardiomyopathy which resolves spontaneously The above literature search yielded no results with relevance to
after birth. anesthetic management.
Bone age is severely delayed and abnormalities of the meta- We, therefore, report our experience of 5 children with Donohue
physics and epiphysis are evident on pathological examination.12 syndrome undergoing 12 general anesthetics and suggest anesthetic
considerations in this rare population.
Of the 5 children, all presented in infancy with intrauterine
4 | CLINICAL COURSE AND TREATMENT growth retardation (IUGR) and low birth weights, 2 being less than
the 0.4th percentile. Four of the five were born to consanguineous
Due to their accelerated fasting state and disordered homeostatic parents. Three presented in the first week of life with episodes of
response to hypoglycemia, these children fail to gain weight or hyperglycemia and hypoglycemia and were found to be hyperinsu-
length despite adequate feeding. They lose muscle mass and are at linemic.
risk of sepsis due to immunodeficiency. Four of these children have undergone 12 general anesthetics
The only current treatment option in Donohue syndrome is IGF- for indications including insertion of percutaneous gastrostomy,
1 therapy. The aim of therapy is to prevent compensatory hyperin- insertion of central venous access, endoscopy, MRI, and laparotomy.
sulinemia, which is the cause of many of the features of the syn- A 5th child was assessed as having too high a risk for general anes-
drome, including the cardiomyopathy. Clinically, the aim is to thesia.
support linear growth and development and improve life expectancy. For 7 of the 12 anesthetics (3 patients), there was documented
IGF has approximately 6% of the hypoglycemic effect of insulin.13 It difficulty with bag mask ventilation. These usually required a 2 per-
is known to be structurally similar to insulin and can bind in vivo to son technique and manual decompression of abdominal distension
cause increased peripheral glucose uptake, increased glycogen syn- by placement of a nasogastric tube into the stomach. A laryn-
thesis, and inhibition of protein catabolism. It is hypothesized that goscopy grade of IIb or above was noted on 5 occasions (2
recombinant IGF-1 therapy may have these advantageous anabolic patients) with ENT assistance and rigid bronchoscopy being
effects in insulin resistance. There is scant evidence for its use in required on one occasion. A previous intubation in this patient had
insulin resistance syndromes14 with the majority of literature being been uneventful (Grade II laryngoscopy). One induction associated
in the form of case studies.15,16 In the small number of case studies with intermittent desaturations was marked by postoperative cere-
available, some found IGF-1 therapy ineffective in these patients, bral edema and seizures requiring prolonged care on the pediatric
while others found modest reductions in levels of hyperinsulinemia, intensive care unit.
along with improved glucose utilization. Although combined numbers Two anesthetics (in 2 patients) were marked by cardiac arrests
in these studies are small,15-18 IGF-1 therapy appears to be both bio- secondary to difficult manual ventilation, intubation, and hypoxia. On
chemically and clinically effective. As well as having long-term both occasions, circulation was restored with CPR and both were
improvements in glucose control and secondary effects of hyperinsu- extubated postoperatively on PICU. Other adverse events included
linemia, there also appears to be an acute reduction in both glucose accidental perioperative extubation with subsequent difficult mask
and insulin levels following a bolus dose of IGF-1. Long-term IGF-1 ventilation and significant desaturation, and multiple failed intuba-
therapy necessitates regular fundoscopy14,17 to assess for the devel- tions requiring ENT support (described above).
opment of retinopathy. None of the children were noted to have particularly difficult
venous access.
Of the 5 children, 2 have subsequently died, one is being palli-
5 | MATERIALS AND METHODS ated and 2 are being managed on IGF-1 therapy. Where significant
cardiac disease became a feature, the risk vs benefit ratio was re-
A literature search was carried out using PubMed, Medline, and evaluated and on occasion surgery was declined.
Cochrane library, and using the following (MESH) search terms: Table 1 represents airway difficulties and adverse events noted
Donohue syndrome, Leprechaunism, Mendenhall syndrome, insulin in the above case series.
resistance, receptor, insulin.
Patients were identified by a formal request made to the
department of pediatric endocrinology at Great Ormond Street 7 | DISCUSSION
Hospital. Each patient’s notes were searched both electronically
and manually, looking for initial presentation and antenatal/birth There is currently no published literature or guidelines on the anes-
history, clinical diagnosis and progress, and outcomes. The anes- thetic management of this complex condition. This small case series
thesia records were examined for documentation of procedure, shows clearly that these children have a high anesthetic risk. It is
induction technique, grade of laryngoscopy, and any adverse important to weigh the risks of anesthesia against the benefits of
events. the surgery. Where surgery is necessary, difficulty can be reasonably
26 | KIRKWOOD ET AL.

T A B L E 1 Tabular summary of 5 patient case series—anesthesia experience, airway management, and adverse events
Mask
Anesthetic ventilation Adverse Cardiac
Pt episode Weight difficulty Laryngoscopy events arrest Details
1 1 <0.4th N 1 N N
percentile
2 N 1 N N
3 N Laryngeal N N
mask airway
2 1 3 kg Y 1 (periarrest) Y Y Failed intubation, difficult mask ventilation
leading to cardiac arrest. Grade 1 periarrest intubation
2 Y 1 N N
3 Y 2B Y N Accidental perioperative extubation. Difficult mask
ventilation. Prolonged desaturation.
4 Y 2B N N
5 Y 3 N N
6 Y Laryngeal N N
mask airway
3 1 1.09 kg N 2 Y Y Cardiac arrest postextubation, followed by difficult mask
ventilation and prolonged desaturation
4 1 2.7 kg N 2 N N
2 Y 3 Y N Difficult mask ventilation requiring 2 person technique,
multiple failed intubations, ENT rigid bronchoscope
required. Prolonged hypoxic period

predicted and a difficult intubation/ventilation plan must be in place where possible. Intubation in Donohue syndrome can be challenging.
with senior support available. Laryngoscopy is hindered by reduced mouth size and rapid desatura-
Because of their low birth weight and failure to thrive they are tion. An uncomplicated anesthetic does not preclude difficulties in
prone to many of the usual risks of neonatal anesthesia including diffi- future cases. We would recommend a spontaneously breathing gas
cult intravenous access, hypothermia, and hypoglycemia. Patients with induction technique in these patients, to minimize both bag valve
Donohue syndrome can develop hypoglycemia with minimal fasting mask ventilation and apnea time. Intravenous access should be
and should, therefore, receive intravenous maintenance with dextrose placed prior to induction or as soon as possible thereafter. A difficult
preoperatively. They require close glycemic control in the periopera- airway plan should also be in place including senior help, indirect
tive period. Studies show an intact hypothalamo-pituitary axis in these laryngoscopy, and if possible, the involvement of the ENT team.
children, and it can, therefore, be assumed that children with Donohue Children with Donohue syndrome are at risk of cardiovascular col-
syndrome will mount a normal stress response to surgery. lapse either due to hypoxia or to an associated cardiomyopathy, and
Cardiomyopathy and biventricular hypertrophy are not present at therefore, we would suggest that all resuscitation equipment and
birth, but are almost universal within 1-2 months of age. This may drugs are readily available throughout general anesthesia.
not be clinically evident, and therefore, a preoperative echocardio- We note that this review is significantly weakened by its retro-
graph should be carried out. Anesthesia in a patient with Donohue spective nature. In many cases, information was not available and
syndrome and hypertrophic obstructive cardiomyopathy carries con- therefore not able to be included. Despite attempts to contact all fami-
siderably more risk and should be avoided whenever possible. lies involved, no consent was obtained meaning that our case descrip-
The small thoracic diameter and significant abdominal distension tions have necessarily been vague in order for individuals to remain
in these patients leads to a reduction in chest compliance and makes unidentifiable. Nonetheless we feel that the high-risk nature of this
bag valve mask ventilation problematic. They have minimal func- condition and the significant anesthetic challenges involved make this
tional reserve and even brief periods of apnea cause profound desat- message a valuable addition to the literature on the subject.
urations that can lead to hypoxic cardiac arrest. Manual ventilation
and consequent gastric insufflation with preexisting abdominal dis-
ETHICAL APPROVAL
tension exacerbate this situation, and therefore, we would suggest
insertion of a large bore nasogastric tube prior to induction or as Discussed with Great Ormond Street Ethics Committee Chair.
soon after induction as possible. Airway adjuncts (including a laryn- Despite attempts to contact parents, we were unable to gain paren-
geal mask airway) for difficult mask ventilation should be readily tal consent. As discussed with the journal editor, the article has been
available, and we would suggest an experienced second anesthetist significantly modified to ensure individuals are unidentifiable.
KIRKWOOD ET AL. | 27

CONFLICT OF INTEREST syndrome with decreased insulin receptor autophosphorylation due to

a novel INSR mutation. Eur J Pediatr. 2013;172:1125-1129.
The authors report no conflict of interest. 10. Planchenaulta D, Martin-Coignardb D, Rugemintwazac D. Le syn-
drome de Donohue ou leprechaunisme: a propos d’un cas. Arch Pedi-
atr. 2014;21:206-210.
11. Baykan A, Cansever M, Konuskan B. Hypertrophic cardiomyopathy with
ORCID
leprechaunism. J Pediatr Endocrinol Metab. 2008;21:317-318.
Alana Kirkwood http://orcid.org/0000-0003-1867-7186 12. Patterson J, Watkins WL. Leprechaunism in a male infant. J Pediatr.
1962;60:731.
13. Quin JD, Fisher BM, Paterson KR, Inoue A, Beastall GH, MacCuish
AC. Acute response to recombinant insulin-like growth factor I in a
REFERENCES patient with Mendenhall’s syndrome. N Engl J Med. 1990;323:1425-
1426.
1. Donohue WL. Dysendocrinism. J Pediatr. 1948;32:739-748. 14. McDonald A, Williams RM, Regan FM, Semple RK. IGF-I treatment
2. Donohue WL, Uchida I. Leprechaunism: euphuism for a rare familial of insulin resistance. Eur J Endocrinol. 2007;157:51-56.
disorder. J Pediatr. 1954;45:505-519. 15. de Kerdanet M, Caron-Debarle M, Nivot S, et al. Ten-year improve-
3. Fernhoff PM. 50 years ago in the journal of pediatrics. J Pediatr. ment of insulin resistance and growth with recombinant human insu-
2004;145:697. lin-like growth factor 1 in a patient with insulin receptor mutations
4. Mohamed S. An infant with Leprechaunism, ambiguous genitalia and resulting in leprechaunism. Diabetes Metab. 2014;41:331-337.
poor glycemic control: a management challenge. Acta Endocrinologica 16. Nakae J, Kato M, Kato M, Murashita M, et al. Long term effects of
(Buc). 2014;10:134-139. rhIGF-1 on metabolism and growth control. J Clin Endocrinol Metab.
5. Martınez-Frıas M, Bermejo E, Rodrıguez-Pinilla E, Martınez-Fernan- 1998;83:542-549.
dez ML. Sındrome de Donohue (Leprechaunismo). ECEMC. 2010;24: 17. Semple RK, Williams RM, Dunger DB. What is the best management
1-2. strategy for patients with severe insulin resistance? Clin Endocrinol.
6. Kosztolanyi G. Leprechaunsism/Donohue syndrome/insulin receptor 2010;73:286-290.
gene mutations: a syndrome delineation story from clinicopathologi- 18. Kuzuya H, Matsuura N. Trial of insulin-like growth factor I therapy
cal description to molecular understanding. Eur J Pediatr. 1997;156: for patients with extreme insulin resistance syndromes. Diabetes.
253-255. 1993;42:696-705.
7. Kobayashi M, Olefsky JM, Elders J, et al. Insulin resistance due to a
defect distal to the insulin receptor: demonstration in a patient with
leprechaunism. Proc Natl Acad Sci. 1978;75:3469-3473.
8. Taylor SI, Roth J, Blizzard RM, Elder MJ. Qualitative abnormalities in How to cite this article: Kirkwood A, Stuart G, Harding L.
insulin binding in a patient with extreme insulin resistance: decreased Donohue syndrome: A review of literature, case series, and
sensitivity to alterations in temperature and pH. Proc Natl Acad Sci.
anesthetic considerations. Pediatr Anesth. 2018;28:23–27.
1981;78:7157-7161.
9. Tinka Hovnik T, Bratanic N, Podkrajsek KT. Severe progressive
https://doi.org/10.1111/pan.13273
obstructive cardiomyopathy and renal tubular dysfunction in Donohue
Accepted: 17 October 2017

DOI: 10.1111/pan.13285

SYSTEMATIC REVIEW

Assessment of sedation level prior to neonatal intubation:

A systematic review

Ellen H. M. de Kort1,2 | Nienke M. Halbmeijer2 | Irwin K. M. Reiss2 |

Sinno H. P. Simons2

1
Department of Pediatrics and
Neonatology, Maxima Medical Center, Summary
Veldhoven, The Netherlands Background: Adequate premedication before neonatal endotracheal intubation
2
Division of Neonatology, Department of
reduces pain, stress, and adverse physiological responses, diminishes duration and
Pediatrics, Erasmus MC Sophia Children’s
Hospital, Rotterdam, The Netherlands number of attempts at intubation, and prevents traumatic airway injury. Therefore,
intubation should not be started until an adequate level of sedation is reached. It is
Correspondence
Ms. Ellen H. M. de Kort, Department of not clear how this should be measured in the clinical situation.
Neonatology, Maxima Medical Center,
Objectives: The aim of this study is to provide a systematic review of the usability
Veldhoven, The Netherlands.
Email: [email protected] and validity of scoring systems or other objective parameters to evaluate the level
of sedation before intubation in neonates. Secondary aims were to describe parame-
Funding information
Fonds NutsOhra; Netherlands Organisation ters that are used to determine the level of sedation and criteria on which the deci-
for Health Research and Development,
sion to proceed with intubation is based.
Grant/Award Number: 90713494
Methods: Literature was searched (January 2017) in the following electronic data-
Section Editor: Francis Veyckemans
bases: Embase, Medline, Web of Science, Cochrane Central Registrar of Controlled
Trials, Pubmed Publisher, and Google Scholar.
Results: From 1653 hits, 20 studies were finally included in the systematic review.
In 7 studies, intubation was started after a predefined time period; in 1 study, pre-
oxygenation was the criterion to start with intubation; and in 12 studies, intubation
was started in case of adequate sedation and/or relaxation. Only 4 studies
described the use of 3 different objective scoring system, all in the neonatal inten-
sive care unit, which are not validated.
Conclusion: No validated scoring systems to assess the level of sedation prior to
intubation in newborns are available in the literature. Three objective sedation
assessment tools seem promising but need further validation before they can be
implemented in research and clinical settings.

KEYWORDS
intubation, newborn infant, premature, premedication, sedation

1 | INTRODUCTION morbidity, sepsis, or necrotizing enterocolitis, and the administration

of surfactant, to surgical procedures such as bowel surgery, surgical
A significant proportion of preterm and critically ill newborn infants closure of a patent ductus arteriosus, laser coagulopathy in case of
needs endotracheal intubation at some point during their admission retinopathy of prematurity, or placement of a surgical central venous
to the Neonatal Intensive Care Unit (NICU) or because of anesthesia. catheter. Endotracheal intubation is a painful and stressful procedure
Indications range from respiratory insufficiency in case of pulmonary and often is associated with adverse physiological responses such as

28 | © 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/pan Pediatric Anesthesia. 2018;28:28–36.
DE KORT ET AL. | 29

bradycardia, hypoxemia, systemic hypertension, and intracranial

3.2.3 | Types of interventions
hypertension.1-7 The use of premedication before intubation reduces
the risk of these adverse events, and also reduces the duration and For the primary objective, studies evaluating an objective score or
number of attempts needed for successful intubation and prevents parameter to assess the level of sedation after administration of
traumatic injury to face, eyes, gums, tongue, and glottic structures.6-11 premedication were included. For the secondary objectives, studies
Neonatal intubation should therefore be preceded by sedative addressing any type of intervention were included.
premedication.
The goal of administering premedication is to achieve a proper
3.2.4 | Types of outcome measures
level of sedation before intubation is started or muscle relaxants
are administered. Still the question is how this level of sedation For the primary objective, studies with degree of sedation during
should be defined and how this should be measured in the clinical intubation as outcome measure were included. For the secondary
situation. objective, studies addressing any type of outcome measure were
included.

2 | OBJECTIVES
3.3 | Data collection and analysis
Our primary objective was to provide a systematic review of studies
3.3.1 | Electronic searches
assessing the usability and validity of objective scoring systems or
parameters to evaluate the level of sedation before intubation in The following electronic databases were searched: Embase, Medline,
newborn infants. Because we presumed that very few studies have Web of Science, Cochrane Central Registrar of Controlled Trials,
addressed this issue, our secondary objective was to describe com- Pubmed Publisher, and Google Scholar. The search strategies are
mon practice, addressing the following 2 questions: outlined in Table 1. There were no restrictions on the basis of publi-
cation date or publication status. The search strategy also had no
1. On what parameter(s) and/or criteria is the decision to proceed restriction in language. The search was performed on January 27,
with intubation after the administration of premedication based? 2017.
2. Which (objective) scores or other parameters are used to deter-
mine the level of sedation after administration of premedication?
3.3.2 | Selection of studies
Two authors (EdK and NH) independently assessed titles and
3 | MATERIALS AND METHODS
abstracts of the search results. Full copies of all potentially relevant
studies were obtained. Both authors made a decision on final inclu-
3.1 | Study design
sion after retrieval of full copies independently. Disagreements were
This systematic review was conducted according to the guidelines of resolved by consensus; if necessary a third author was consulted
the Preferred Reporting Items for Systematic Reviews and Meta- (SS). Only articles written in English, Dutch, German, and French
Analyses (PRISMA).12 were selected.

3.2 | Criteria for considering studies for this review 3.3.3 | Data extraction
Two authors (EdK and NH) extracted details of the included studies
3.2.1 | Types of studies
independently. The following data were extracted: use of premedica-
For our primary objective, eligible for inclusion were randomized tion, used drugs or combination of drugs, scores or parameters
controlled trials (RCTs), quasi-RCTs, controlled clinical trials, and determining the decision to proceed with intubation, sedation as
observational studies evaluating the use of objective scoring systems parameter to proceed with intubation, actions undertaken in case of
or parameters to assess level of sedation. For our secondary objec- insufficient sedation.
tives, also single-case studies, poster presentations, editorials, and
reviews were eligible for inclusion, provided these extensively
3.3.4 | Quality appraisal individual studies
reported the complete procedure from administration of premedica-
tion to start of the intubation. Availability of full text was imperative. Because no studies were found reporting on the usability and
validity of scoring systems or other parameters to assess the
level of sedation, and we thus only report on common practice
3.2.2 | Types of study population
as described in the methods sections of the reports, evaluation
The target population consisted of preterm and term neonates who of the methodological quality and the risk of bias was not
needed endotracheal intubation. relevant.
30 | DE KORT ET AL.

T A B L E 1 Search strategies
Electronic database Search strategy
Embase (intubation/de OR “respiratory tract intubation”/exp OR (intubat*):ab,ti) AND (newborn/de OR “newborn intensive care”/exp
OR prematurity/de OR “low birth weight”/exp OR “newborn care”/de OR (newborn* OR (new* NEXT/1 born*) OR neonat*
OR nicu OR nicus OR prematur* OR preterm* OR “low birth weight” OR lbw OR vlbw OR elbw):ab,ti) AND (premedication/
exp OR “hypnotic sedative agent”/exp OR “sedative agent”/exp OR “conscious sedation”/de OR “deep sedation”/de OR
“anesthetic agent”/exp OR “muscle relaxation”/de OR “muscle relaxant agent”/exp OR “neuromuscular blocking agent”/exp
OR (premedicat* OR sedat* OR anesthetic* OR anaesthetic* OR ((musc* OR neuromusc*) NEAR/3 (relax* OR block*))):ab,
ti) NOT ([animals]/lim NOT [humans]/lim) NOT (“cesarean section”/de OR pregnancy/exp OR “pregnant woman”/de OR
(cesarean* OR caesarean* OR pregnan*):ab,ti)
Medline (intubation/OR “Intubation, Intratracheal”/OR (intubat*).ab,ti.) AND (exp infant, newborn/OR “Intensive Care Units,
Neonatal”/OR “Intensive Care, Neonatal”/OR (newborn* OR (new* ADJ born*) OR neonat* OR nicu OR nicus OR
prematur* OR preterm* OR “low birth weight” OR lbw OR vlbw OR elbw).ab,ti.) AND (premedication/OR exp “Hypnotics
and Sedatives”/OR “conscious sedation”/OR “deep sedation”/OR exp “anesthetics”/OR “muscle relaxation”/OR exp
“Neuromuscular Agents”/OR (premedicat* OR sedat* OR anesthetic* OR anaesthetic* OR ((musc* OR neuromusc*) ADJ3
(relax* OR block*))).ab,ti.) NOT (exp animals/NOT humans/) NOT (“cesarean section”/OR exp pregnancy/OR “pregnant
women”/OR (cesarean* OR caesarean* OR pregnan*).ab,ti.)
Cochrane ((intubat*):ab,ti) AND ((newborn* OR (new* NEXT/1 born*) OR neonat* OR nicu OR nicus OR prematur* OR preterm* OR
“low birth weight” OR lbw OR vlbw OR elbw):ab,ti) AND ((premedicat* OR sedat* OR anesthetic* OR anaesthetic* OR
((musc* OR neuromusc*) NEAR/3 (relax* OR block*))):ab,ti) NOT ((cesarean* OR caesarean* OR pregnan*):ab,ti)
Web of Science TS=(((intubat*)) AND ((newborn* OR (new* NEAR/1 born*) OR neonat* OR nicu OR nicus OR prematur* OR preterm* OR
“low birth weight” OR lbw OR vlbw OR elbw)) AND ((premedicat* OR sedat* OR anesthetic* OR anaesthetic* OR ((musc*
OR neuromusc*) NEAR/2 (relax* OR block*)))) NOT ((cesarean* OR caesarean* OR pregnan*)) NOT ((animal* OR rat OR
rats OR mouse OR mice OR murine) NOT (human* OR child* OR patient*)))
Google Scolar Intubation|intubated newborn|newborns|neonates|neonatal|nicu|nicus|premature|preterm|”low birth weight”|lbw|vlbw|elbw
premedication|sedation|anesthetic|anaesthetic|”muscular|neuromuscular relaxation|blockers” -cesarean -caesarean -pregnancy
-pregnant

relaxation.7,13-23 Nevertheless, no more than 4 of these 12 studies

4 | RESULTS
used an objective scoring system.13-16 In one other of the 12 stud-
ies, disappearance of the eyelash reflex was considered an indicator
4.1 | Study selection
of hypnosis and therefore as the criterion to start with intubation.21
A flowchart of study selection is provided in Figure 1. The initial In the control group, loss of muscle tone was the criterion to start
electronic database search yielded 2597 records, which number was with intubation.21 The methods section did not describe, however,
reduced to 1652 after duplicates were removed. One additional how muscle relaxation was rated. In the remaining 7 studies that
record found in reference lists was added. Thus, the titles and based the start of the intubation on the degree of sedation and/or
abstracts of in total 1653 records were screened on relevance to the relaxation, it was not clear if sedation and/or relaxation were rated
primary and secondary objectives. None of the studies appeared rel- either in an objective or a rather subjective way.7,17-20,22,23
evant to the primary objective. Regarding the secondary objective, In 8 of the 12 studies, the intubating clinician’s observation of
full text was obtained from 75 records and 1578 records were insufficient sedation led to the administration of more premedication
excluded. Of these 75 records, 5 were excluded because of language to achieve an adequate level of sedation and/or relaxation.7,13,14,17-
19,21,22
restriction (1 record) or unavailability (4 records). Of the remaining In the remaining 4 studies, this was not mentioned at
70 records, full text was read. The studies’ methods sections were all.15,16,20,23
screened for information about parameters on which the decision to In 7 of all 20 included studies, intubation was started after a cer-
start with intubation was based. Twenty studies provided this tain period of time had elapsed after the administration of premedi-
information and these were included in the final analysis of our cation.5,6,24-28 The methods section of these studies neither made
secondary objective. clear if level of sedation also played a role, nor if all patients were
adequately sedated upon start of the intubation procedure. In one
study, preoxygenation was the leading parameter to start with intu-
4.2 | Study characteristics and results of individual
bation.29
studies
Only 4 of the 20 studies used an objective scoring system to
Table 2 shows the characteristics of the 20 included studies. In 12 assess the level of sedation before proceeding with intubation.
studies, the decision to proceed with intubation after the administra- These objective scoring systems are outlined in Table 3. The first
tion of premedication was based on the degree of sedation or scoring system is a sedation score used in 2 different reports.
DE KORT ET AL. | 31

F I G U R E 1 Flowchart of included and

excluded studies

Smits et al used this score in their dose-finding study for propofol administered if sedation and relaxation were found unsatisfactory,
in newborns and de Kort et al used the score in a study on the until a satisfactory condition was achieved with no predefined maxi-
use of remifentanil in preterm infants undergoing the INSURE pro- mum dose of propofol.13 In the report of de Kort et al,14 in case of
13,14
cedure. The score is adopted from the study of Naulaers et al ineffective sedation, additional remifentanil was administered up to a
into the effectiveness of methohexital as premedication for elec- predefined maximum, with a conversion to propofol in case of
15
tive intubation. The level of sedation is assessed from the motor persisting inadequate sedation.
response to a firm stimulus by rubbing of the feet (1 = sponta- The third score is from Thall et al, in their report describing how
neous movement, 2 = moves on touch, 3 = moves on stimulus, to perform dose-finding studies for premedication in neonatal intu-
4 = no movement). In both studies, sedation was presumed to be bation. For this purpose they used the Neonatal Pain, Agitation and
effective with scores 3 and 4, upon which intubation was started. Sedation Scale (N-PASS) first described by Hummel et al.30 The
Another scoring system is the relaxation score also used in the score consists of 5 variables that should all be scored within 5 min-
report of Smits et al, combining this score with the above-mentioned utes of the first sedative administration: crying/irritability, behavior
sedation score. After the administration of propofol, level of sedation state, facial expression, extremity tone, and vital parameters. Each
and muscle relaxation were determined. For relaxation, the muscle item is scored on a 5-point scale from 2 to +2, with 2 corre-
tone of the extremities was assessed (1 = hypertonic, 2 = normal sponding to highest sedation and +2 corresponding to highest infant
tone, 3 = mildly hypotonic, and 4 = hypotonic). Also, scores 3 and 4 discomfort. For the purpose of endotracheal intubation, Thall et al16
reflect effective relaxation. Both relaxation and sedation were considered a score between 7 and 3 to reflect a good sedative
assessed by the intubating neonatologist. Additional propofol was state.
32

T A B L E 2 Characteristics of included studies

|

Parameter for Consequences insufficient seda-

First author, year, country Study design Premedication start intubation Start intubation tion/relaxation
Smits A, 2016, Belgium Prospective observational Propofol Sedation and relaxation Good sedation and relaxation Additional propofol in case of
scores unsatisfactory relaxation and
sedation according to physician
De Kort, 2016, Prospective observational Remifentanil Sedation Good sedation score Additional remifentanil until good
The Netherlands sedation score
Avino D, 2014, Belgium Noninferiority randomized trial SG: remifentanil Time SG: after complete remifentanil
CG: morphine, atropine infusion; CG: 3 min after
and midazolam midazolam infusion
Baleine J, 2014, France Prospective observational Nasal midazolam Sedation At hypnosis, muscle relaxation or Second dose in case of arousal at
apnea introduction of tube or
excessively awake at 5 min
Durrmeyer X, 2014, France Prospective observational Atropine, sufentanil Relaxation At onset of paralysis Extra atracurium (up to a
and atracurium predefined maximum)
Thall PF, 2014, USA Not applicable Not applicable Sedation Good sedation score Not applicable
Simons S, 2013, Prospective observational Propofol Sedation When sedation was scored yes by Extra propofol (up to a maximum
The Netherlands treating physician of 3 doses)
Penido MG, 2011, Brazil Double-blinded randomized trial SG: propofol, remifentanil Time 2 min after remifentanil infusion
CG: midazolam, remifentanil
Choong K, 2010, Canada Double-blinded randomized SG: remifentanil, atropine Time 30 s after last drug administration
controlled trial and placebo
CG: fentanyl, atropine and
succinyl-choline
Welzing L, 2010, Germany Prospective observational Propofol Time 75 s after propofol administration
Lemyre B, 2009, Canada Prospective observational Atropine, fentanyl and Relaxation Cessation of spontaneous None mentioned
Succinyl-choline movements (succinyl-choline
(the latter only in GA ≥ 34 wks) group) or sufficient relaxation
Van Looy, 2008, USA Prospective observational Midazolam and fentanyl Time 1-2 min after medication Another dose of fentanyl in case
administration of 3 failed attempts
Ghanta S, 2007, Australia Randomized, open-label SG: propofol Hypnosis and relaxation SG: disappearance of eyelash reflex SG: 1 extra dose of propofol, after
controlled trial CG: morphine, atropine and CG: muscle relaxation that conversion to
suxamethonium CG: repeating doses of
suxamethonium
Dempsey E, 2006, Ireland Prospective observational Mivacurium and fentanyl Preoxygenation Saturation > 95%, typically 2 min
after fentanyl infusion
Milesi C, 2006, France Prospective observational Nitrous oxide Relaxation At suppression of muscle tone
(cessation of movements and
DE

hypotonia limbs and jaw)

KORT

(Continues)
ET AL.
DE KORT ET AL. | 33

5 | DISCUSSION
Consequences insufficient seda-

This systematic review was performed to provide insight into the

availability of validated objective scoring systems to assess newborn

Up to 4 extra doses of
infants’ level of sedation prior to intubation. The literature as per

suxamethonium

None mentioned
None mentioned

January 27, 2017 does not provide such a validated scoring system.
tion/relaxation

The decision to proceed with the intubation procedure after

administering premedication should be based on the preintubation
sedation level. However, studies reporting on this issue show that
level of sedation is not always the key factor to proceed with intu-
bation. In several studies intubation proceeded when a certain per-
fasciculations and begin of muscle
SG: once spontaneous movements

iod of time had elapsed after the administration of premedication,

1-2 min after preoxygenation

assuming this amount of time being sufficient for drug effect. How-
No reaction to heel rubbing
CG: after fentanyl infusion

At disappearance of muscle

ever, as drug pharmacokinetics and pharmacodynamics can differ

considerably depending on gestational age and postnatal age, suffi-
1 min after infusion

cient drug effect may need a different period of time in individual

Start intubation

patients. Using the same time frame in all neonates may very well
had ceased

flaccidity

lead to a proportion of patients not being adequately sedated

for the procedure. To guide individualized dosing, we strongly
recommend basing the start of the intubation on the actual level of
sedation and administering extra doses if necessary.
SG: relaxation CG: time

In the majority of studies included in this review, quality of seda-

tion is indeed the key factor to guide intubation. However, this is
start intubation
Parameter for

assessed in very diverse ways, often rather subjective or not further

Relaxation

Sedation

specified. Validated scoring systems are lacking, especially prior to

Time

Time

elective intubation in the operating room. However, the literature

describes 3 objective assessment tools used in the NICU: the easy-
to-perform sedation score first described by Naulaers et al, the also
easy-to-perform relaxation score first used by Smits et al, and the
SG: morphine, atropine and
SG: atropine, fentanyl, and

more extensive scoring system suggested by Thall et al which is

CG: atropine, fentanyl

based on the N-PASS described by Hummel et al.13,14,16,30

The purpose of validated objective scoring systems is to be able
suxamethonium
Premedication

SG: thiopental
SG: morphine

to predict effective sedation during the intubation procedure before

CG: placebo

CG: placebo
Methohexital
mivacurium

CG: none

actually starting the procedure, thereby preventing neonates from

intubation without effective sedation. Validation of such scoring sys-
tems can be done by comparing the scoring system with the actual
level of sedation during the intubation procedure in a larger group of
Randomized placebo controlled

patients. In neonatal intubation studies, the actual level of sedation

Randomized controlled trial

Prospective observational

during an intubation procedure is frequently assessed with the vali-

Randomized controlled

dated intubation score adopted from Viby-Mogensen et al.31 In this

Randomized study

non blinded trial

intubation score, the items ease of laryngoscopy, position of the

nonblinded trial

vocal cords, coughing, jaw relaxation, and movement of the extremi-

Study design

ties are all judged on a 4-point scale. A score of 2 or less on each

item reflects effective sedation.13,14,24,25
CG, control group; SG, study group.

Although both using the Viby-Mogenson intubation score to

qualify level of sedation during the intubation procedure, the reports
Naulaers G, 1997, Belgium
First author, year, country
(Continued)

Lemyre B, 2004, Canada

of Smits et al and De Kort et al do not provide enough information

Buthada A, 2000, USA
Roberts K, 2006, USA

Oei J, 2002, Australia

to draw any conclusions about the accurateness of the sedation and

relaxation scores. In the sedation score, the absence of a motor reac-
tion or only a slight motor reaction to a firm stimulus is presumed to
TABLE 2

indicate the neonate will tolerate the insertion of the endotracheal

tube into the supraglottic airway. However, in this assumption, the
used stimulus should be stronger than the act of inserting an
34
|

T A B L E 3 Objective scoring systems

Highest sedation Lowest sedation

Level of sedation
Sedation Score 4 3 2 1
Motor response No movements Movement on firm stimulus Movement on slight touch Spontaneous movements
to firm stimulus
Definition Effective sedation = score 3 or 4
Relaxation score 4 3 2 1
Extremities tone Hypotonic Mildly hypotonic Normal tone Hypertonic
Definition Effective relaxation = score 3 or 4
Good sedative state 2 1 0 1 2
Crying/irritability No cry with painful stimuli Moans or cries minimally No sedation/no pain signs Irritable or crying at High-pitched or silent
with painful stimuli intervals, consolable continuous cry, inconsolable
Behavior state No arousal to any stimuli, Arouses minimally to stimuli, little No sedation/no pain signs Restless, squirming, Arching, kicking, constantly awake of
no spontaneous movements spontaneous movement awakens frequently arouses minimally/no movement
Facial expression Mouth is lax, no expression Minimal expression with stimuli No sedation/no pain signs Any pain expression intermittent Any pain expression continual
Extremities tone No grasp reflex, flaccid tone Weak grasp reflex, ↓ muscle tone No sedation/no pain signs Intermittent clenched toes, fists, Continual clenched fists, toes, finger
finger splay, body is not tense splay, body is tense
Vital signs No variability with stimuli, <1% variability from No sedation/no pain signs ↑ 10%-20% from baseline, ↑ 20% from baseline, SaO2 < 75%
hypoventilation or apnea baseline with stimuli SaO2 76%-85% with with stimulation—slow recovery
stimulation—quick recovery
Definition Good sedative state = total
score on all 5 items 7 to 3
DE
KORT
ET AL.
DE KORT ET AL. | 35

endotracheal tube. The question rises if the act of inserting an endo- Limitation of this review is that we used the description of the
tracheal tube, mainly via the nasal route, requires a stronger stimulus intubation procedure in the methods section to answer the ques-
than heel rubbing. For example, by anesthesiologists, the much fir- tions of our secondary objective. Because describing the entire pro-
mer stimulus of pinching the trapezius muscle is frequently used cedure from administration of premedication to intubation was not
before inserting a supraglottic airway. Validation of that stimulus in a the primary goal of the included studies, it is possible that these
clinical study has not been reported. In summary, although both descriptions were not complete and that more scoring systems or
sedation and relaxation scores seem potentially useful scores in the other parameters were used in practice.
neonatal population, validation of these scores is mandatory before
further use.
The scoring system suggested by Thall et al concerns a compre- 6 | CONCLUSION
hensive and precise scoring system, which is likely to adequately
reflect the level of sedation. Also, different definitions can be used Validated scoring systems to assess the level of sedation prior to
in case of different needs of sedation level according to its purpose. intubation in newborns are not available in the literature. Three
However, to the best of our knowledge, clinical trials using this scor- objective sedation assessment tools seem promising but need further
ing system are lacking. Therefore, any conclusion about its accuracy validation before they can be implemented in research and clinical
and usability is not possible and validation is needed before future settings. Future research is necessary to find a physical stimulus that
use. Even if validated, the scoring system must prove its suitability in enables clinicians to better anticipate the response to insertion of an
daily practice. In neonatal intubation, fast performance of the proce- endotracheal tube and laryngoscopy and to find the best premedica-
dure is mandatory, especially when fast acting agents are used. The tion regimen to achieve the preset goals.
extensiveness of the score could possibly make it time consuming
and therefore less suitable.
ACKNOWLEDGMENTS
Despite the lack of validated objective scoring systems, several
reports do have shown that premedication before neonatal intuba- We greatly thank Wichor M. Bramer of the Erasmus Medical Center
tion has become standard practice in the majority of neonatal library for performing the electronic database search for this system-
10,11,32-38
units. There is however much debate about which atic review and Ko Hagoort for editing the manuscript. This research
premedication or premedication regimen is best. Studies evaluating was supported by a grant from Fonds NutsOhra. S. Simons was
certain premedication strategies are mainly focused on using 1 dos- financially supported by a personal grant of the Netherlands Organi-
ing strategy for the entire neonatal population. However, pharma- sation for Health Research and Development (ZonMw nr
codynamics and pharmacokinetics are influenced by factors such as 90713494).
gestational age, postnatal age and morbidity. For example, Smits
et al13 found that neonates of different gestational ages needed
DISCLOSURES
different doses of propofol for adequate sedation. Most important
in our opinion is that the used premedication achieves effective The authors report no conflict of interest.
sedation. It should also have a quick recovery to allow for fast
extubation, and no significant side effects. The search for the most
ORCID
suitable premedication strategy should be directed toward personal-
ized medicine and focus on administering just enough premedica- Ellen H. M. de Kort http://orcid.org/0000-0003-0469-0128
tion to achieve adequate sedation in the individual patient. In this
point of view, a scoring system that adequately indicates the level
of sedation is mandatory. REFERENCES
The work of field and the techniques used differ between pedi-
1. Maheshwari R, Tracy M, Badawi N, Hinder M. Neonatal endotracheal
atric anesthesiologists and neonatologists. For example, they have intubation: how to make it more baby friendly. J Paediatr Child
different drug choices, pursue different levels of anesthesia (deep vs Health. 2016;52:480-486.
superficial) and require different duration of sedation (hours vs min- 2. Marshall TA, Deeder R, Pai S, Berkowitz GP, Austin TL. Physiologic
changes associated with endotracheal intubation in preterm infants.
utes). However, the goals of administering premedication during
Crit Care Med. 1984;12:501-503.
induction in the operating room and before intubation in the NICU 3. Kelly M, Finer NN. Nasotracheal intubation in the neonate. Physio-
are comparable and in both situations the level of sedation should logic responses and effects of atropine and pancuronium. J Pediatr.
be objectively assessed before continuing the procedure. With this 1984;105:303-309.
4. Millar C, Bissonnette B. Awake intubation increases intracranial pres-
in mind, neonatologists and anesthesiologists can possibly share
sure without affecting cerebral blood flow velocity in infants. Can J
valuable knowledge in this area of the field. With the availability of
Anaesth. 1994;41:281-287.
a validated scoring system that can make a distinction between dif- 5. Lemyre B, Doucette J, Kalyn A, Gray S, Marrin ML. Morphine for
ferent levels of sedation, it should even be possible to use a single elective endotracheal intubation in neonates: a randomized trial.
scoring system for both settings. BMC Pediatr. 2004;4:20.
36 | DE KORT ET AL.

6. Buthada A, Sahni R, Rastogi S, Wung JT. Randomised controlled trial 24. Avino D, Zhang WH, De Ville A, Johansson AB. Remifentanil versus
of thiopental for intubation in neonates. Arch Dis Child Fetal Neonatal morphine-midazolam premedciaton on the quality of endotracheal
Ed. 2000;82:F34-F37. intubation in neonates: a noninferiority randomized trial. J Pediatr.
7. Oei J, Hari R, Butha T, Lui K. Facilitation of neonatal nasotracheal 2014;164:1032-1037.
intubation with premedication: a randomized controlled trial. J Paedi- 25. Penido MG, De Oliveira Silva DF, Tavares EC, Silva YP. Propofol ver-
atr Child Health. 2002;38:146-150. sus midazolam for intubating preterm neonates: a randomized con-
8. Barrington KJ, Finer NN, Etches PC. Succinylcholine and atropine for trolled trial. J Perinatol. 2011;31:356-360.
premedication of the newborn infant before nasotracheal intubation: 26. Choong K, Alfaleh K, Doucette J, et al. Remifentanil for endotracheal
a randomized, controlled trial. Crit Care Med. 1989;17:1293-1296. intubation in neonates: a randomized controlled trial. Arch Dis Child
9. Carbajal R, Eble B, Anand KJS. Premedication for tracheal intubation Fetal Neonatal Ed. 2010;95:F80-F84.
in neonates: confusion or controversy? Semin Perinatol. 2007;31:30- 27. Welzing L, Kribs A, Eifinger F, Huenseler C, Oberthuer A, Roth B.
317. Propfol as an inducation agent for endotracheal intubation can cause
10. Kumar P, Denson SE, Mancuso TJ, et al. Premedication for nonemer- significant arterial hypotension in preterm neonates. Pediatr Anesth.
gency endotracheal intubation in the neonate. Pediatrics. 2010;125: 2010;20:605-611.
608-615. 28. Van Looy JW, Schumacher RE, Bhatt-Mehta V. Efficacy of a
11. Byrne E, MacKinnon R. Should premedication be used for semi- premedication algorithm for nonemergent intubation in a neonatal
urgent or elective intubation in neonates? Arch Dis Child. intensive care unit. Ann Pharmacother. 2008;43:947-955.
2006;91:79-83. 29. Dempsey EM, Hazzani FA, Faucher D, Barrington KJ. Facilitation of
12. Moher D, LIberati A, Tetzlaff J, Altman GG; PRISMA Group. Pre- neonatal endotracheal intubation with mivacurium and fentanyl in
ferred reporting items for systematic reviews and meta-analysis: the the neonatal intensive care unit. Arch Dis Child Fetal Neonatal Ed.
PRISMA statement. J Clin Epidemiol. 2009;62:1006-1012. 2006;91:F279-F282.
13. Smits A, Thewissen L, Caicedo A, Naulaers G, Allegaert K. Propofol 30. Hummel P, Puchalski M, Creech SD, Weis MG. Clinical reliability and
dose-finding to reacht optimal effect for (semi-)elective intubation in validity of the N-PASS: neonatal pain, agitation and sedation scale
neonates. J Pediatr. 2016;179:54-60. with prolonged pain. J Perinatol. 2008;28:55-60.
14. De Kort EHM, Hanff L, Roofthooft D, Reiss IKM, Simons SHP. Insuf- 31. Viby-Mogensen J, Engbaek J, Eriksson LI, et al. Good clincial
ficient sedation and severe side effects after fast administration of research practice (GCRP) in pharmacodynamics studies of neuromus-
remifentanil during INSURE in preterm newborns. Neonatology. cular blocking agents. Acta Anaesthesiol Scand. 1996;40:59-74.
2017;111:172-176. 32. Whyte S, Birrell G, Wyllie J. Premedication before intubation in UK
15. Naulaers G, Deloof E, Vanhole C, Kola E, Devlieger H. Use of neonatal units. Arch Dis Child Fetal Neonatal Ed. 2000;82:F38-F41.
methohexital for elective intubation in neonates. Arch Dis Child. 33. Kelleher J, Mallya P, Wyllie J. Premedication before intubation in UK
1997;77:F61-F64. neonatal units: a decade of change? Arch Dis Child Fetal Neonatal Ed.
16. Thall PF, Nguyen HQ, Zohar S, Maton P. Optimizing sedative dose 2009;94:F332-F335.
in preterm infants undergoing treatment for respiratory distress syn- 34. Chaudhary R, Chonat S, Gowda H, Clarke P, Curley A. Use of
drome. J Am Stat Assoc. 2014;109:931-943. premedication for intubation in teritiary neonatal uints in the United
17. Baleine J, Milesi C, Mesnage R, et al. Intubation in the delivery Kingdom. Pediatr Anesth. 2009;19:653-658.
room: experience with nasal midazolam. Early Hum Dev. 2014;90:39- 35. Wheeler B, Broadbent R, Reith D. Premedication for neonatal intu-
43. bation in Autralia and New Zealand: a survey of current practice. J
18. Durrmeyer X, Dahan S, Delorme P, et al. Assessment of atropine- Paediatr Child Health. 2012;48:997-1000.
sufentanil-atracurium anaesthesia for endotracheal intubation: an 36. Durrmeyer X, Daoud P, Decobert F, et al. Premedication for neona-
observational study in very premature infants. BMC Pediatr. tal endotracheal intubation: results from the epidemiology of proce-
2014;14:120. dural pain in neonates study. Pediatr Crit Care Med. 2013;14:e169-
19. Simons SHP, Van der Lee R, Reiss IKM, van Weissenbruch MM. e175.
Clinical evaluation of propofol as sedative for endotracheal intuba- 37. Simon L, Trifa M, Mokhtari M, Hamza J, Treluyer JM. Premedication
tion in neonates. Acta Paediatr. 2013;1002:e487-e492. for tracheal intubation: a prospective survey in 75 neonatal and
20. Lemyre B, Cheng R, Gaboury I. Atropine, fentanyl and succinyl- pediatric intensive care units. Crit Care Med. 2004;32:565-568.
choline for non-urgent intubations in newbors. Arch Dis Child Fetal 38. Sarkar S, Schumacher RE, Baumgart S, Donn SM. Are newborns
Neonatal Ed. 2009;94:F436-F442. receiving premedication before elective intubation. J Perinatol.
21. Ghanta S, Abdel-Latif ME, Lui K, Ravindranathan H, Awas J, Oie J. 2006;26:286-289.
Propofol compared with the morphine, atropine and suxamethonium
regimen as induction agents for neonatal endotracheal intubation: a
randomized, controlled trial. Pediatrics. 2007;119:1248-1255.
22. Milesi C, Pidoux O, Sabatier E, Badr M, Cambonie G, Picaud JC. How to cite this article: de Kort EHM, Halbmeijer NM, Reiss
Nitrous oxide analgesia for intubating preterm neonates: a pilot IKM, Simons SHP. Assessment of sedation level prior to
study. Acta Paediatr. 2006;95:1104-1108.
neonatal intubation: A systematic review. Pediatr Anesth.
23. Roberts KD, Leone TA, Edwards WH, Rich WD, Finer NN. Premedi-
cation for nonemergent neonatal intubations: a randomized, con- 2018;28:28–36. https://doi.org/10.1111/pan.13285
trolled trial comparing atropine and fentanyl to atropine, fentanyl,
and mivacurium. Pediatrics. 2006;118:1583-1591.
Accepted: 17 October 2017

DOI: 10.1111/pan.13284

RESEARCH REPORT

Pressure-flow characteristics of breathing systems and their

components for pediatric and adult patients

Christin Wenzel | Stefan Schumann | Johannes Spaeth

Department of Anesthesiology and Critical

Care, Faculty of Medicine, Medical Center, Summary
University of Freiburg, Freiburg, Germany Background: Breathing circuits connect the ventilator to the patients’ respiratory
Correspondence system. Breathing tubes, connectors, and sensors contribute to artificial airway resis-
Johannes Spaeth, Department of tance to a varying extent. We hypothesized that the flow-dependent resistance is
Anesthesiology and Critical Care, University
Medical Center Freiburg, Freiburg, Germany. higher in pediatric breathing systems and their components compared to respective
Email: [email protected] types for adults.
Funding information Aims: We aimed to characterize the resistance of representative breathing systems
This work was supported by institutional and their components used in pediatric patients (including devices for adults) by
funding.
their nonlinear pressure-flow relationship.
Section Editor: Prof Britta von Ungern- Methods: We used a physical model to measure the flow-dependent pressure gradi-
Sternberg
ent (ΔP) across breathing tubes, breathing tube extensions, 90°- and Y-connectors,
flow- and carbon dioxide sensors, water traps and reusable, disposable and coaxial
breathing systems for pediatric and for adult patients. ΔP was analyzed for usual flow
ranges and statistically compared at a representative flow rate of 300 mL∙s1 (ΔP300).
Results: ΔP across pediatric devices always exceeded ΔP across the corresponding
devices for adult patients (all P < .001 [no 95% CI includes 0]). ΔP300 across breath-
ing system components for adults was always below 0.2 cmH2O but reached up to
4.6 cmH2O in a flow sensor for pediatric patients. ΔP300 was considerably higher
across the reusable compared to the disposable pediatric breathing systems (1.9 vs
0.3 cmH2O, P < .001, [95% CI 1.59 to 1.56]).
Conclusion: The resistances of pediatric breathing systems and their components
result in pressure gradients exceeding those for adults several fold. Considering the
resistance of individual components is crucial for composing a breathing system
matching the patient’s needs. Compensation of the additional resistance should be
considered if a large composed resistance is unavoidable.

KEYWORDS
airway—resistance, anesthesia—pediatric, device—equipment, respiratory—mechanics,
ventilation—mechanical

1 | BACKGROUND airway resistance may lead to increased work of breathing1 or emer-

gence of intrinsic positive end-expiratory pressure.2 At any rate, the
During mechanical ventilation, the breathing circuit connects the pressure gradient across artificial airways causes the monitored air-
ventilator to the patient’s respiratory system. Each component of way pressure to differ from the pressure that is actually applied to
the breathing circuit contributes to total airway resistance. A large the patient’s lungs.3 It has been demonstrated that the endotracheal

Pediatric Anesthesia. 2018;28:37–45. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 37
38 | WENZEL ET AL.

tube is the individual component with the highest contribution to

airway resistance.4 Also, breathing circuit filters were shown to con-
tribute—to a lesser extent—to the artificial resistance.5 However, What is already known
research on other components of the breathing circuit is lacking or
outdated6 considering the continuously growing product range.
• The endotracheal tube is the pivotal individual compo-
nent determining the total airway resistance in mechani-
Pediatric anesthesiologists have to deal with devices for both
cally ventilated pediatric patients.
pediatric and adult patients—due to the large variability of their
patients’ individual needs. We hypothesized that breathing systems
• Rohrer’s coefficients characterize sufficiently the nonlin-
ear pressure-flow relationships of endotracheal tubes.
and components for pediatric patients in particular are associated
They can be used to calculate tracheal pressure from the
with high resistance. We therefore investigated the pressure-flow
pressure measured at the airway opening and moreover
characteristics of representative breathing systems and their compo-
to compensate for the endotracheal tube’s resistance.
nents for pediatric and adult patients. In this regard, the flow-depen-
dent pressure gradients (ΔP) across the investigated devices were
What this article adds
determined in a laboratory environment. In order to enable calcula-
tion of ΔP for an arbitrary flow rate, we additionally determined • Specific breathing system components (connectors, sen-
resistance coefficients for each component. sors, breathing tubes) for pediatric patients contribute to
total airway resistance by a considerable amount. This
should be considered when composing a breathing circuit
2 | MATERIALS AND METHODS which matches the patient’s needs best.
• The flow-dependent pressure gradients for a wide range
Breathing systems, silicone breathing tubes of different lengths, tube of breathing system components used in pediatric venti-
extensions, water traps, connectors (Y-piece, 90°-connector and lation therapy (including components for adult patients)
combined), and sensors (flow sensor, flow sensor combined with were quantified. Rohrer’s coefficients for the breathing
Y-piece and 90°-connector and cuvette for CO2 sensor) were inves- system components, enabling calculation of the pressure
tigated. Table 1 lists specifications of all investigated devices. Fig- gradient across the components, are provided.
ure 1 illustrates the designs of the investigated connectors and
sensors (not including breathing systems and breathing tubes). Con-
ventional reusable breathing systems, conventional disposable
2.2 | Calculation of flow-dependent pressure
breathing systems, and coaxial breathing systems were investigated.
gradient
Conventional breathing systems were investigated with and without
a water trap. In order to characterize the devices’ nonlinear resistances, the coeffi-
cients describing the pressure-flow relationship of the respective
device, ie, Rohrer coefficients (K1 and K2) and inertance (I), were
2.1 | Setup for measurements _ and
determined from the continuously measured ΔP, flow rate (V),
In order to measure the pressure-flow characteristics of the individu- € via multiple linear regression analysis
derived volume acceleration (V)
ally investigated devices, they were streamed with a sinusoidal gas of Equation 18:
flow generated by a piston pump. For including maximal flow rates
DP ¼ K1 V_ þ K2 V_ 2 þ I V
€ (1)
that may typically occur in adult and pediatric patients during
mechanical ventilation,7 we chose flow rate amplitudes of In order to facilitate the comparison between components, ΔP
1
500 mL∙s (devices for pediatrics) or 700 mL∙s1 (devices for was calculated from K1 and K2, following Equation 1, at a represen-
adults). Flow rate was measured using a pneumotachograph (Fleisch tative flow rate of 300 mL∙s1 (ΔP300). As volume acceleration in
Type 1, Dr. Fenyves & Gut, Hechingen, Germany). Pressures were such calculation would be arbitrary, we assumed a constant flow
measured at a distance of 1.5 cm from both ends of the respectively € = 0) implying that the inertance term of Equation 1 is negligible.
(V
investigated device via piezoresistive pressure sensors (Type 4, SI-
€ rdlingen, Germany). Pressure measure-
Special Instruments GmbH, No
2.3 | Statistical analyses
ment sites at both ends were of identical inner diameters to avoid
the Bernoulli effect. Data are presented as mean SD if not indicated otherwise. For
For measurements across breathing tubes and breathing systems, ΔP300, Rohrer’s coefficients and inertance, no SD is given if below
these were arranged with a defined sagging of 33% of their total 0.02. Pressure and flow measurements were recorded using self-
length, in order to simulate clinical conditions. Flow rate and pres- written acquisition software, developed in LabView (ver. 7.1,
sure signals were sampled at 200 Hz. Twelve sinusoidal cycles were National Instruments, Austin, TX, USA) and analyzed using MATLAB
recorded for each condition. All measurements were repeated 6 (ver. 2014 R1, The MathWorks, Natick, MA, USA). For comparison
times after dismantling and reassembly of the setup. of ΔP300, ANOVA were calculated followed by post hoc t-tests if
WENZEL ET AL. | 39

T A B L E 1 Names, reference numbers

Pediatric Adult
(REF), and dimensions of breathing
systems and components Length ID Length ID
REF [cm] [mm] REF [cm] [mm]
Breathing tubes
Silicon 2165848 60 10 2165627 60 19
Silicon 66-10-110a 110 10 2166038 110 19
Silicon 66-10-150a 150 10 2166046 150 19
Tube extensions MP01860 10 10 MP01845 15 12
Water trap 8404760 n.a. 5 8404985 n.a. 15
Connectors
Y straight M27077 n.a. 15-5 M25650 n.a. 15
a
90° angled 60-65-002 n.a. 15 8412235 n.a. 15
Combined Y and 8403077 n.a. 15-5 M25682 n.a. 15
90°
Sensors
Straight flow 8411130 n.a. 10-5 n.a. n.a. n.a.
sensor
Angled flow 8410185 n.a. 15-5 n.a. n.a. n.a.
sensor
Cuvette CO2 6870280 n.a. 15-5 6870279 n.a. 15-12
sensor
Breathing systems
Reusable Assembledb 150 10 Assembledb 150 19
Disposable MP00331 150 15 MP00300 150 22
Assembled + WT 66-10- 180 10 66-12- 180 22
085a + WT 085a + WT
Readymade + WT MP00362 180 15 MP00337 180 22
Coaxial inspiratory MP00379 150 11 MP00315 150 16
Coaxial expiratory 170 12c 190 15c

ID, inner diameter; WT, water trap.

Reference numbers refer to Dr€ager Medical (Lu €beck, Germany) or if marked (a) to VBM Medizintech-
nik GmbH (Sulz am Neckar, Germany).
b
Breathings system manually assembled from: silicon breathing tube 150 cm, Y straight connector,
and 90° angled connector.
c
Listed is the actual diameter of the expiratory hose of the coaxial tube.

appropriate (GraphPad PRISM, ver. 6.02, GraphPad Software Inc., La devices with low inner diameter but large volume, such as breathing
Jolla, CA, USA). A P-value < .05 was considered significant. For each tubes and breathing systems. ΔP across pediatric components was
statistical comparison, the 95% confidence interval of the difference always higher compared to ΔP across the corresponding components
[95% CI] is given. for adults (Figure 2; all P < .001, [no 95% CI includes 0]).
ΔP300 was ≤ 0.2 cmH2O in all breathing system components for
adults. In pediatric components ΔP300 was larger compared to the
3 | RESULTS components for adults and ranged up to 4.6 cmH2O in a combined
flow sensor for pediatric patients (Table 3). In breathing tubes for
In all investigated breathing system components ΔP increased dispro- adults, ΔP300 did not depend significantly on tube length but
portionately with increasing flow and their pressure-flow characteris- increased in pediatric breathing tubes with increasing length. ΔP300
tics could be reliably described by Rohrer’s coefficients and inertance varied depending on the direction of flow in pediatric components
(Table 2). K1 and K2 were higher in pediatric devices compared to with asymmetric geometry.
respective values found in devices used in adults. This difference In conventional breathing systems for adults, ΔP300 was < 0.2 cmH2O.
was more pronounced for K2. Inertance reached highest values in In pediatric breathing systems, ΔP300 was considerably higher
40 | WENZEL ET AL.

(Figure 3). In the reusable breathing system, ΔP300 amounted up to

1.9 cmH2O, whereas ΔP300 was only 0.3 cmH2O in the disposable
breathing system. Inclusion of a water trap into a conventional pedi-
atric breathing system led to an increase of ΔP300 in both the reusable
(1.9 to 2.1 cmH2O; P < .001, [95% CI 0.23 to 0.27]) and the dispos-
able type (0.3 to 0.4 cmH2O; P < .001, [95% CI 0.13 to 0.14];
Figure 3). In coaxial breathing systems, ΔP300 differed between
inspiratory and expiratory flow in both systems for pediatric patients
(0.8 vs 0.6 cmH2O; P < .001, [95% CI 0.30 to 0.22]) and systems
for adults (0.3 vs 0.5 cmH2O; P < .001, [95% CI 0.20 to 0.21]).

4 | DISCUSSION

The main findings of the present study can be summarized as follows:

(i) the pressure gradient across breathing system components
depends nonlinearly on flow and can be described by Rohrer’s
approach; (ii) at flow rates relevant for the ventilation of pediatric
patients, the pressure gradient across pediatric breathing systems and
their components can be considerable, whereas it is consistently low
across devices for adults; and (iii) reusable pediatric breathing systems
exhibit a higher resistance compared to disposable and coaxial ones.
The higher resistance of breathing system components for pedi-
atric patients can be explained by the generally smaller physical
dimensions of the pediatric components (Table 1). It is well known
that a streamed tube’s inner diameter is the pivotal determinant for
resistance to gas flow.9 Moreover, decreasing the inner diameter
promotes higher gas velocities and thus turbulent flow. This leads to
a nonlinear resistance that increases with the flow rate. This nonlin-
earity also becomes apparent in the quadratic coefficient (K2) of the
pressure-flow relationship we determined in the present study.
Breathing systems’ components for pediatric patients are gener-
ally designed to fit small dimensions in order to decrease additional
dead space. However, small dimensions impose a greater artificial
airway resistance. The tolerable dead space volume of a breathing
circuit depends on the required tidal volume. Limiting dead space to
the smallest volume possible is particularly relevant in neonatal and
infant patients but may be less important in adolescent pediatric
patients. While total airway resistance may be less important during
controlled mechanical ventilation, it is quite relevant during sponta-
neous breathing support where it contributes to the patient’s work
of breathing. In that case, adult patients may easily compensate for
the comparatively low additional airway resistance of their breathing
systems. By contrast, pediatric patients, particularly when critically ill,
may fail to overcome the high resistance imposed by pediatric
breathing systems. Composing a breathing circuit is therefore a
F I G U R E 1 Technical schemes of connectors and sensors trade-off between dead space volume and airway resistance.
included in the study. Schemes refer to types and dimensions given Differences between resistances of adult and pediatric breathing
in Table 1. Left: components used in pediatric patients; right:
system components may be of particular relevance for the pediatric
components used in adult patients. TE = tube extension,
anesthesiologist, who deals with both depending on the patient’s
WT = water trap, C1 = connector Y straight, C2 = connector 90°
angled, C3 = combined Y and 90°, S1 = straight flow sensor, individual needs. Knowledge of the respective resistances may be
S2 = combined flow sensor, S3 = CO2 sensor. Please note that the crucial to avoid ventilation problems related to increased additional
components S1 and S2 for adult patients are not applicable (n.a.) airway resistance. The present study aims at providing these facts.
WENZEL
ET AL.

T A B L E 2 Rohrer’s coefficients (K1, K2) and inertance (I) for breathing systems and components
Pediatric Adult

Inspiration Expiration Inspiration Expiration

K1 [cmH2O/ K2 [cmH2O/ K1 [cmH2O/ K2 [cmH2O/ I [cmH2O/ K1 [cmH2O/ K2 [cmH2O/ K1 [cmH2O/ K2 [cmH2O/ I [cmH2O/
s/L] L2/s2] s/L] L2/s2] L*s2] s/L] L2/s2] s/L] L2/s2] L*s2]
Breathing tubes
Silicon (60 cm) 0.90 2.26 0.93 2.36 0.11 0.05 0.06 0.01 0.19 0.03
Silicon (110 cm) 1.30 3.41 1.26 3.24 0.19 0.11 0.11 0.07 0.24 0.06
Silicon (150 cm) 1.76 4.34 1.80 3.88 0.28 0.12 0.08 0.06 0.02 0.07
Tube extensions 0.23 4.33 0.17 5.22 0.05 0.00 1.92 0.19 1.62 0.03
Water trap 0.13 4.13 0.14 4.25 0.02 0.01 0.13 0.03 0.16 0.01
Connectors
Y straight 0.10 2.27 0.11 2.88 0.02 0.01 0.52 0.03 0.62 0.01
90° angle 0.07 1.07 0.00 1.88 0.02 0.02 0.47 0.06 0.76 0.01
Combined Y and 90° 0.22 5.34 0.20 8.31 0.02 0.04 0.71 0.18 0.66 0.01
Sensors
Straight flow sensor 1.53 30.18 1.18 35.05 0.07 n.a. n.a. n.a. n.a. n.a.
Angled flow sensor 0.23 50.29 0.53 37.73 0.08 n.a. n.a. n.a. n.a. n.a.
Cuvette CO2 sensor 1.78 7.79 0.90 14.52 0.05 0.15 0.26 0.03 0.68 0.01
Systems
Reusable 1.24 16.15 1.21 16.66 0.35 0.10 1.02 0.14 0.76 0.08
Disposable 0.26 2.33 0.19 2.56 0.11 0.01 1.45 0.02 1.44 0.07
Reusable + WT 1.78 17.10 1.82 17.45 0.39 0.12 1.40 0.20 1.02 0.10
Disposable + WT 0.34 3.59 0.26 3.77 0.17 0.01 1.71 0.01 1.68 0.07
Coaxial inspiratory 0.12 8.85 n.a. n.a. 0.14 0.08 2.46 n.a. n.a. 0.11
Coaxial expiratory n.a. n.a. 0.41 5.01 0.11 n.a. n.a. 0.33 3.91 0.10

n.a., not applicable; WT, water trap.

SD was < 0.02 cmH2O for all measurements and is therefore not given.
|
41
42 | WENZEL ET AL.

F I G U R E 2 Inspiratory flow-dependent pressure gradient (ΔP) across breathing tubes (upper left panel), tube extensions and water traps
(middle left panel), connectors (lower left panel), and sensors (upper right panel) used for pediatric (solid lines) and adult patients (dashed lines).
Please note the reference line at a flow rate of 300 mL s1, referring to ΔP300 (see also Table 3). The lower right panel shows the pressure
gradients at a flow rate of 300 mL s1 (ΔP300) depending on the devices’ lowest inner diameters (open symbols = devices used in pediatric
patients; filled symbols = devices used in adult patients)

Typically, a breathing system includes tubes, a Y-piece, and a 90° Adding components for specific purposes, such as a carbon diox-
connector. We investigated such basic combined systems (below ide sensor or a tube extension, would further increase ΔP to a rele-
referred to as reusable) and found that ΔP300 reached almost vant extent. In this regard, the routine use of breathing circuit filters
2 cmH2O in the pediatric model. Referring to Lowe et al10 such might be questioned. The filter intended for use in pediatric patients
basic systems account for about two-third of specific airway would additionally provide approximately 0.5 cmH2O at a flow rate
resistance found in 3-year-old children. of 300 mL∙s1.5 Furthermore, our results demonstrate that straight
WENZEL ET AL. | 43

T A B L E 3 Pressure gradients at a flow rate of 300 ml∙s1 (ΔP300)

across breathing systems and components for inspiration (insp.) and
expiration (exp.)
Pediatric Adult

ΔP300 ΔP300

Insp. Exp. Insp. Exp.

[cmH2O] [cmH2O] [cmH2O] [cmH2O]
Breathing tubes
Silicon (60 cm) 0.48 0.49 0.02 0.02
Silicon (110 cm) 0.70 0.67 0.04 0.04
Silicon (150 cm) 0.92 0.89 0.04 0.04
Tube extension 0.46 0.52 0.17 0.21
Water trap 0.41 0.43 0.00 0.02
Connectors
Y straight 0.24 0.29 0.05 0.06
90° angle 0.12 0.17 0.05 0.09
Combined Y and 0.55 0.81 0.08 0.11
90°
Sensors
Straight flow 3.17 3.51 n.a. n.a.
sensor
Angled flow 4.60 3.55 n.a. n.a.
sensor
Cuvette CO2 1.24 1.58 0.07 0.07
sensor
Breathing systems
Reusable 1.83 1.86 0.10 0.11
Disposable 0.29 0.29 0.13 0.14
F I G U R E 3 Flow-dependent pressure gradient (ΔP) across
Reusable + WT 2.07 2.12 0.16 0.15
different breathing systems used for pediatric patients for inspiration
Disposable + WT 0.42 0.42 0.14 0.15 and expiration. Please note the reference line at a flow rate of
Coaxial 0.84 n.a. 0.25 n.a. 300 mL s1, referring to ΔP300. WT = water trap
inspiratory
Coaxial n.a. 0.57 n.a. 0.45
increased disproportionally with increasing flow rate and decreased
expiratory
disproportionally with decreasing flow rate. As already discussed,
WT, water trap; n.a., not applicable. ΔP300 reached almost 2 cmH2O in the pediatric reusable breathing
SD was < 0.02 cmH2O for all measurements and is therefore not given.
systems. By contrast, it was comparably low in the disposable
and the coaxial types. Again, this is attributable to the different
connectors generally provided a lower resistance compared to inner diameters, being lower in the reusable system. It is notewor-
angled ones; with regard to DP the use of straight connectors thy that ΔP across the reusable breathing system overestimated
appears preferable. the sum of ΔP across the respective isolated components. This
The pediatric flow sensors represent significant sources of resis- may be caused by flow phenomena at the connection sites
tance. These sensors measure the flow rate by hot-wire anemome- promoting additional turbulences.12 Such flow phenomena also
try, a technique which usually comes with a minimal resistance. 11
account for the differences of ΔP between inspiration and
However, the hour-glass shape of the investigated sensor’s case in expiration that became evident in some devices with asymmetric
question caused an additional resistance comparable to that of an geometry.
endotracheal tube with 4.5 mm inner diameter.3 Compared to these Including a water trap in the breathing systems increased ΔP to a
sensors the resistance of the other pediatric components was rather minor extent (≤0.2 cmH2O) which we consider to be of subordinate
low. However, it is the sum of resistances that accounts for total air- clinical relevance. Without a water trap, liquid may accumulate in
way resistance. the tubing system. The narrowing of the tube at the accumulation
The flow-dependent pressure gradient differed considerably site would increase resistance in a nonpredictable manner and more-
among breathing systems for pediatric patients, whereas it was over lead to self-triggering of the ventilator.13 Therefore, we would
consistently low in breathing systems for adults. In all systems, ΔP not suggest omitting the use of water traps.
44 | WENZEL ET AL.

Coaxial breathing systems incorporate the inspiratory tube within flow rate of 300 mL s1. This flow rate may appear disproportionally
the expiratory one, thus facilitating clinical handling at the expense high for small infants.18 However, we consider this flow rate repre-
of the lumen’s cross-sectional area. In a recent study, we compared sentative for a wide range of spontaneously breathing pediatric
14
coaxial and conventional breathing systems for adults. The coaxial patients.
breathing systems exhibited particularly high resistances compared
to conventional ones. Interestingly, in the present study, the pedi-
atric coaxial breathing system showed a comparably low resistance. 6 | CONCLUSION
In contrast to the coaxial breathing systems for adults, the effective
cross-sectional areas of the pediatric coaxial tubing system were The resistances of breathing systems and breathing system compo-
even larger than those of the respective conventional ones. In both nents for pediatric patients result in considerable pressure gradients
types of coaxial tubing systems, resistance differed between inspira- exceeding those of corresponding devices for adults several fold.
tion and expiration, which can be attributed to differences in the Taking into consideration the resistance of individual components is
effective cross-sectional areas of the inspiratory and the expiratory crucial for composing a breathing system matching the patient’s
tube. needs. Compensation for the additional work of breathing should be
By addressing the flow-dependent pressure gradient we aim to considered if a large composed resistance is unavoidable. Resistance
facilitate the transfer of our results into clinical routine. From a clini- coefficients for several breathing system components for pediatric
cal point of view, artificial airway resistance implies a diagnostic and and adult patients are provided.
a therapeutic problem. The diagnostic problem is that airway pres-
sure but not pulmonary pressure is the standard measure of routine
CONFLICT OF INTEREST
clinical monitoring. However, airway pressure differs from pulmonary
pressure by the extent of ΔP across all artificial airway components. None declared.
Knowledge of ΔP would allow for estimating pulmonary pressure
from airway pressure measurement and thus give a better estimation
ACKNOWLEDGMENTS
of the pressure actually loading the lung. The therapeutic problem
concerns especially spontaneous breathing during which breathing The authors gratefully acknowledge the technical support of Mat-
system resistance adds to the patient’s work of breathing. In order thias Schneider, an outstanding mechanical engineer.
to compensate for the additional work of breathing caused by the
breathing circuit, one may increase the inspiratory pressure support
ORCID
by the amount of ΔP estimated for the peak flow rate. However, this
would lead to overcompensation at the beginning of the inspiration Johannes Spaeth http://orcid.org/0000-0003-3931-6400
and would not apply to expiration.15 An adequate compensation
needs to dynamically adapt airway pressure to ΔP resulting from the
current flow rate. Some ventilators do compensate for the resistance REFERENCES
16
of the endotracheal tube (automatic tube compensation). This can
1. Brochard L, Rua F, Lorino H, et al. Inspiratory pressure support com-
effectively reduce the additional work of breathing and particularly pensates for the additional work of breathing caused by the endotra-
relieve patients with high ventilatory demand from the effort cheal tube. Anesthesiology. 1991;75:739-745.
required to overcome the artificial airways.17 Compensating ΔP of 2. Rossi A, Polese G, Brandi G, et al. Intrinsic positive end-expiratory
pressure (PEEPi). Intensive Care Med. 1995;21:522-536.
the other components of the breathing circuit would allow further
3. Spaeth J, Steinmann D, Kaltofen H, et al. The pressure drop across
relief for the patient. The present study provides resistive coeffi- the endotracheal tube in mechanically ventilated pediatric patients.
cients for these additional components, being essential for such an Pediatr Anesth. 2015;25:413-420.
approach. 4. Dennison FH, Taft AA, Mishoe SC, et al. Analysis of resistance to
gas flow in nine adult ventilator circuits. Chest. 1989;96:1374-
1379.
5. Wilkinson KA, Cranston A, Hatch DJ, et al. Assessment of a hygro-
5 | LIMITATIONS scopic heat and moisture exchanger for paediatric use. Anaesthesia.
1991;46:296-299.
6. Hatch DJ. Tracheal tubes and connectors used in neonates–dimen-
The investigated artificial airway components represent a limited
sions and resistance to breathing. Br J Anaesth. 1978;50:959-964.
sample of the available commercial products. Our aim was to include 7. Newth CJ, Venkataraman S, Willson DF, et al. Weaning and extubation
a representative collection of components. Furthermore, it has to be readiness in pediatric patients. Pediatr Crit Care Med. 2009;10:1-11.
conceded that other types of the same categories may show differ- 8. Guttmann J, Kessler V, Mols G, et al. Continuous calculation of intra-
ing characteristics and that the investigated devices are subject to tracheal pressure in the presence of pediatric endotracheal tubes.
Crit Care Med. 2000;28:1018-1026.
technical advancement.
9. Guttmann J, Eberhard L, Fabry B, et al. Continuous calculation of
In order to allow for easier comparison of the resistances of the intratracheal pressure in tracheally intubated patients. Anesthesiology.
investigated airway components we arbitrarily chose an exemplary 1993;79:503-513.
WENZEL ET AL. | 45

10. Lowe L, Murray CS, Custovic A, et al. Specific airway resistance in 16. Elsasser S, Guttmann J, Stocker R, et al. Accuracy of automatic tube
3-year-old children: a prospective cohort study. Lancet. compensation in new-generation mechanical ventilators. Crit Care
2002;359:1904-1908. Med. 2003;31:2619-2626.
11. Berry JM, Rothmann BS, Eisenkraft JB. Monitoring ventilation. In: 17. Guttmann J, Haberthur C, Mols G, et al. Automatic tube compensa-
Ehrenwerth J, Eisenkraft JB, Berry JM, eds. Anesthesia Equipment: Prin- tion (ATC). Minerva Anestesiol. 2002;68:369-377.
ciples and Applications. Philadelphia: Elsevier-Saunders; 2013:233-234. 18. Bennett NR, Bingham R. Breathing system filters for paediatric prac-
12. Loring E, Drazen JM. Kinetic energy loss and cenvective acceleration tice. Anaesthesia. 2007;62:192-193.
in respiratory resistance measurements. Lung. 1979;156:33-42.
13. Sansome AJ. Inappropriate triggering. Anaesthesia. 1988;43:1065-
1066.
14. Wenzel C, Schumann S, Spaeth J. Coaxial tubing systems increase How to cite this article: Wenzel C, Schumann S, Spaeth J.
artificial airway resistance and work of breathing. Respir Care. Pressure-flow characteristics of breathing systems and their
2017;62:1171-1177.
components for pediatric and adult patients. Pediatr Anesth.
15. Guttmann J, Bernhard H, Mols G, et al. Respiratory comfort of auto-
matic tube compensation and inspiratory pressure support in con- 2018;28:37–45. https://doi.org/10.1111/pan.13284
scious humans. Intensive Care Med. 1997;23:1119-1124.
Accepted: 11 October 2017

DOI: 10.1111/pan.13282

RESEARCH REPORT

Introducing the 6-4-0 fasting regimen and the incidence of

prolonged preoperative fasting in children

Hanna Andersson1 | Per M. Hellstrӧm2 | Peter Frykholm1

1
Department of Surgical Sciences,
Section of Anaesthesia and Intensive Care, Summary
Uppsala University Hospital, Uppsala, Background: Children often starve for longer than recommended by current preop-
Sweden
2 erative fasting guidelines.
Department of Medical Sciences,
Gastroenterology, Uppsala University Aims: We studied the effects of implementing a more lenient fasting regimen on
Hospital, Uppsala, Sweden
the duration of clear fluid fasting, as well as the incidence of extended fasting in
Correspondence children.
Hanna Andersson, Department of Surgical
Methods: Preoperative duration of clear fluid fasting was recorded for patients
Sciences, Section of Anesthesiology and
Intensive Care Medicine, Uppsala Universitet scheduled for procedures in a unit applying the standard 6-4-2 fasting regimen. This
Medicinska och farmaceutiska
group was compared with a cohort in the same unit 1 year after transitioning to a
vetenskapsomradet – Uppsala University
Hospital ing 70, Uppsala, Sweden. 6-4-0 fasting regimen. The latter includes no limitations on clear fluid intake until
Email: [email protected]
the child is called to theater. A third cohort from a unit in which the 6-4-0 fasting
Section Editor: Francis Veyckemans regimen has been implemented for over a decade was also studied for comparison.
Results: Patients fasting according to the 6-4-2 fasting regimen (n = 66) had a med-
ian fasting time for clear fluids of 4.0 h and a 33.3% incidence of fasting more than
6 h. After transitioning to the 6-4-0 fasting regimen (n = 64), median duration of
fasting for clear fluids decreased to 1.0 h, and the incidence of fasting more than
6 h decreased to 6.3%. In the second unit (n = 73), median fasting time was 2.2 h
and the proportion of patients fasting more than 6 h was 21.9%.
Conclusion: The introduction and implementation of the 6-4-0 fasting regimen
reduces median fluid fasting duration and the number of children subjected to
extended fasting.

KEYWORDS
anesthesia, children, fasting, fluids, preoperative

1 | INTRODUCTION At our center, we were troubled by the logistic problems of

ensuring reasonable fasting times while following the Scandina-
In most countries, similar fasting guidelines are applied to chil- vian guidelines.6 We questioned the rationale for clear fluid fast-
1,2
dren and adults; 6 h for solids, and 2 h for clear fluids. ing and thus implemented a regimen in which the latter was
Despite clear recommendations of 2-h fasting for clear liquids, omitted from the year 2000 and onwards. When another unit in
children are still being subjected to unnecessarily long preopera- our hospital planned to adopt the 6-4-0 regimen, we decided to
3-5
tive starvation intervals in routine practice. The most common study the effects of the transition on actual fasting times. We
reasons for prolonged fasting are incorrect preoperative instruc- hypothesized that this more liberal regimen would be easier to
tions, changes in the surgical schedule and parents not following follow and hence would reduce the incidence of prolonged
instructions.4,5 fasting in daily clinical practice.

46 | © 2017 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/pan Pediatric Anesthesia. 2018;28:46–52.
ANDERSSON ET AL. | 47

The aim of the study was to compare the duration of fluid fast-
ing as well as the incidence of extended fasting in daily clinical prac- What is already known
tice, with the standard 6-4-2 regimen vs the 6-4-0 regimen in
children scheduled for elective surgery. • Children are fasted for longer than recommended by
international preoperative fasting guidelines.

What new information does this study add

2 | MATERIALS AND METHODS
• Removing the time limit for clear fluids reduces median
Following local ethics committee approval (Ethics Committee of fasting times as well as the incidence of extended preop-
Uppsala, Sweden Dnr. 2014/487/1) children and parents received erative fasting.
information about the study at the preoperative assessment and oral
consent was obtained from all parents or legal guardians. The
requirement for written informed consent was waived by the ethics
committee. The study was not preregistered as a clinical trial as
T A B L E 1 The 6-4-0 fasting regimen
there was no randomized intervention and it may thus be regarded
as an audit of a change in practice. • For pediatric patients, 0- to 16-year old
• Allows for clear fluids up until the patient is called to surgery (rou-
All patients under 17 years of age, scheduled for elective pro-
tinely 30 min before induction)
cedures, were asked to participate in the study. Emergency cases
• Clear fluids are defined as: water, fruit punch, fruit juice without
and patients that were already intubated on arrival at the operation pulp, coffee or tea without milk, and ice lollies/popsicles. We do
theater were excluded. On arrival at the operation theater, the not allow carbonated drinks, milk, or yoghurt
accompanying nurse and parents were asked when the child last • Four-hour fasting for breast milk, infant formula, and milk-based
products like yoghurt. Parents are encouraged to wake up and feed
had had anything to eat or drink and the reported respective times
babies under 12 months at 4 AM if they are scheduled as first proce-
were documented using a standardized case report form. A few
dure in the morning
sips of water with the premedication did not count as if the • Patients instructed to fast for solids from midnight, but 6-h fasting
patient had had a drink. In total, 203 children were included in the accepted when appropriate
study. • Most patients are individually assessed by a pediatric anesthetist
and the prescribed fasting interval may thus be customized if
Three different groups were analyzed. Group 1 consisted of chil-
increased risk of aspiration is suspected
dren undergoing surgery at the departments of ear-nose-throat
(ENT), oral surgery, or plastic surgery when applying the 6-4-2 fast-
ing regimen (ENT_2 h, n = 66). These parents were informed to fast as first case, in the morning and in the afternoon, and between inpa-
their child 6 h for solids, 4 h for breast milk and formula, and 2 h tients and ambulatory surgery patients.
for clear fluids. Group 2 consisted of children in that same depart-
ment, 1 year after introducing the 6-4-0 fasting regimen (ENT_0 h,
2.1 | Anesthesia methods
n = 64). The latter includes fasting for solids for 6 h, milk-containing
fluids for 4 h, but no definite limitation on clear fluid intake until the Individual anesthesia methods were not registered in this study. In
child is called to theater (Table 1). Group 3 consisted of children in the ENT department, the routine anesthesia induction in children is
the main pediatric operation unit where the 6-4-0 fasting regimen intravenous with propofol or mask induction with sevoflurane.
has been implemented since the year 2000 (MP_0 h, n = 73). The Remifentanil is given to facilitate intubation without muscle relax-
shift of fasting regimen at the ENT unit was made during the sum- ants. Sevoflurane is used for maintenance. In the pediatric unit, most
mer season and registration was paused during the first year after children receive a laryngeal mask airway after i.v. induction with
the shift. Before changing fasting regimen, ward nurses received oral propofol, and continue with sevoflurane and spontaneous breathing.
information and local standard preoperative procedures were chan- For intubation, the routine i.v. induction is with thiopenthone, fen-
ged. No changes were made in how information was given to tanyl, and atracurium. Intubated children receive an oro-gastric tube,
patients and caregivers. The 6-4-0 fasting regimen allows almost all but the stomach is not routinely suctioned in elective cases. Classic
elective pediatric patients to drink up until called to theater. How- deep extubation is not practiced, but LMAs (and occasionally ETTs,)
ever, individual assessments are made for each patient and proce- are often removed when the child is still asleep after a bolus of
dure by both anesthetist and surgeon. These individual assessments propofol.
were not registered in this study.
Measured outcomes were actual total fluid fasting times, ie, time
2.2 | Information about fasting times
from last drink to induction. For post hoc analysis, cut-offs for pro-
longed fasting were set at 4, 6, and 12 hours. Children under Most children and caregivers are informed about fasting times by a
36 months of age were defined as toddlers and analyzed as a sub- nurse at the preoperative assessment visit. Inpatients admitted for
group. We also studied the difference between children scheduled elective surgery are usually seen by a pediatric anesthetist the day
48 | ANDERSSON ET AL.

before surgery. Some patients are only assessed through their medi- T A B L E 2 Patient characteristics
cal records. In those cases, outpatients receive a telephone call from ENT 6-4-2 ENT 6-4-0 MP 6-4-0
the outpatient nurse, in which information about fasting is included. (n = 66) (n = 64) (n = 73)
The final operation list is set in the afternoon the day before sur- Age (mean, 5 (0-15) 6 (0-15) 6 (0-15)
gery, after which ward and outpatient nurses give the final informa- range)
tion about when to stop (or not to stop) eating and drinking, <3 years, n (p) 18 (27.3%) 16 (25.0%) 26 (35.6%)
respectively. Some of the children included in this study may have 3-5 years, n (p) 22 (33.3%) 17 (26.6%) 14 (19.2%)
been recommended longer fasting intervals than 0 or 2 h due to an >6 years, n (p) 26 (39.4%) 31 (48.4%) 33 (45.2%)
increased assessed risk for aspiration at the preoperative assessment ASA
or by the surgeon due to the planned procedure. However, we did 1, n (p) 41 (62.1%) 43 (67.2%) 32 (43.8%)
not register what information the patients and caregivers were given
2, n (p) 23 (34.8%) 20 (31.3%) 34 (46.6%)
about fasting times, and hence these children will also have been
3, n (p) 2 (3%) 1 (1.6%) 7 (9.6%)
included in this study.
Time of surgery
First case, n 24 (36.4%) 16 (25.0%) 32 (43.8%)
2.2.1 | Statistics and sample size calculation (p)
Morning, n 28 (42.4%) 33 (51.6%) 25 (34.2%)
Power analysis concluded that 36 patients per group were needed to (p)
detect a difference in mean fasting time of 4.5 h, with a significance Afternoon, n 14 (21.2%) 15 (23.4%) 16 (21.9%)
level of 0.05 and a power of 80%. To ensure sufficient size for the sub- (p)
group analysis, recruitment of at least 60 patients was planned for each In/outpatient status
study group. All reported data were analyzed using Microsoft Excel Inpatients, n 24 (36.4%) 13 (20.3%) 38 (52.1%)
2016MSO (Redmond, WA, USA), Statistica Statistical Software (Tulsa, (p)
OK, USA) and the R package “rcmdr” (Rcmdr: R Commander. R package Outpatients, 42 (63.6%) 51 (79.7%) 35 (47.9%)
version 2.3-2.). Total fasting times were not normally distributed and n (p)
hence presented as median values with 95% confidence intervals. The
95% confidence interval for medians was calculated using boot-strap-
ping. Due to unequal variance, comparison of total fasting times 34.8% to 6.2% (OR = 0.13, 95% CI: 0.03-0.35, P < .001), and
between groups was performed using Kruskal-Wallis ANOVA. Fre- patients fasting 12 h or more decreased from 15.2% to 3.1%
quencies of patients fasting for extended times were reported as per- (OR = 0.18, 95% CI: 0.03-0.72, P = .032).
centages. The effect of the predictors “group”, “age”, “time of day for
surgery”, and “in/outpatient status” on the outcome “fasting 6 hours or
3.2 | Duration of fasting in the main pediatric
more” were quantified by binary logistic regression analyses. In crude
anesthesia unit
analyses, the effect of each predictor was separately evaluated against
the outcome. Adjusted analyses were not performed, due to the small Median fasting duration for clear fluids in the main pediatric anes-
sample size. Odd ratios (OR) and their 95% confidence intervals (CI) are thesia unit (n = 73) was 2.3 h (95% CI 1.7-3.0 h). This was signifi-
presented. The level of significance was set to a = 0.05. cantly shorter than group 1 (P < .001), and also significantly longer
than group 2 (P < .01). The distribution of fluid fasting duration at
the main pediatric anesthesia unit is shown in Figure 1.
3 | RESULTS The incidence of prolonged fasting was lower in the main pedi-
atric anesthesia unit, compared to the ENT unit before the transition,
Patients characteristics are shown in Table 2. Fasting times for solids with 34.2% of children fasting 4 h or more (OR = 0.41, 95% CI:
did not differ between the groups. The results presented below only 0.20-0.80, P = .01), 23.3% fasting 6 h or more (OR = 0.57, 95% CI:
describe the differences in clear fluid fasting. 0.27-1.19, P = .135), and 8.2% of children fasting 12 h or more
(OR = 0.5, 95% CI: 0.16-1.44, P = .207).

3.1 | Fasting duration decreased when shifting

fasting regimen in the ENT unit 3.3 | Other variables affecting the outcome of
fasting 6 h or more
In group 1 (n = 66), the median fasting time for clear fluids of 4.0 h
(95% CI 3.1-4.5 h) was higher than median fasting time in group 2 Older children had lower odds of fasting 6 h or more, compared to
(n = 64) of 1.0 h (95% CI 0.9-1.5 h) (P < .0001), see Figure 1. children under 3 years of age (children aged 3-5 years: OR = 0.52,
After shifting fasting regimen in the ENT unit, patients fasting 95% CI: 0.23-1.15, P = .135; children over 6 years: OR = 0.39, 95%
4 h or more decreased from 56.1% to 18.8% (OR=0.18, 95% CI: CI 0.17-0.91, P = .031). The proportions of children fasting for more
0.08-0.39, P < .001), patients fasting 6 h or more decreased from than 6 h stratified by age and group are displayed in Figure 2.
ANDERSSON ET AL. | 49

4 | DISCUSSION

After shifting from 2- to 0-hour fasting for clear fluids, the median fast-
ing duration was reduced from 4 to 1 hour and the proportion of
patients fasting for more than 6 h was reduced from 33.3% to 6.3%.
The main benefit of the 6-4-0 fasting regimen is that it simplifies the
logistics of planning for minimized preoperative fasting. Fasting times
in the main pediatric unit was longer than in the ENT unit when apply-
ing the same fasting regimen. This may in part be described by the
slightly different clientele. The proportion of outpatients in the ENT
and the main pediatric unit was 80% and 50%, respectively. We have
seen in earlier pilot studies that our outpatients have a better compli-
ance to the fasting regimen and that inpatients are more frequently
recommended longer fasting intervals, due to co-morbidity and perhaps
more complex surgery. Furthermore, compliance to the fasting regimen
by medical personnel could be higher 1 year after a big transition.
Regardless of fasting regimen, there is a need for continuing education
of medical personnel on current fasting routines and for clarifying infor-
mation to patients and parents, to avoid prolonged fasting.
Although increasingly liberal guidelines have been published over
the last 2 decades, old routines and habits are deeply set. Previous
studies report mean fluid fasting duration to be 6-9 hour when
applying a 6-4-2 fasting regimen, even with an actively managed
list.3,4,7 In clinical practice, guidelines cannot be expected to be fol-
lowed to the letter. Rearrangements of the surgery schedule often
occur due to emergency cases or other unforeseen events. It is
nearly impossible to set an exact time for induction in children not
scheduled as first case in the morning. The recently published study
by Newton et al showed that an ambitious quality improvement pro-
ject with education of staff, more frequent communication by both
anesthetists and nurses and updated documents could reduce the
fasting time in pediatric outpatients. The single biggest improvement
came from implementation of 1-hour fasting for clear fluids.7
The 6-4-0 regimen allows for water, fruit punch, or ice lollies up
until the child is called to the operation theater, which makes life easier
for both parents and the anesthesiologist running the operating room
schedule. In the present study, shifting from 2- to 0-hour fasting for
clear liquids reduced fasting time in all patient categories, but had the
greatest effect among afternoon patients in whom the proportion of
children fasting more than 6 hour decreased from 36% to 0%.
In our institution, all infants and toddlers that breastfeed or drink
infant formula from the bottle are actively encouraged to feed at
4 hour before their planned procedure, to avoid prolonged fasting in
F I G U R E 1 Distribution of fasting duration before and after the our youngest patients. Older children often sleep through the night,
shift to 0-h fasting for clear liquids in the ENT unit, and in the main
sometimes resulting in extended fasting times. Allowing a light
pediatric anesthesia unit
breakfast for children scheduled later in the day might be safe, but
further research on this matter is necessary.
Children scheduled in the morning or afternoon had lower odds Young age seemed to entail longer mean duration of fasting. The
of fasting 6 h or more, compared to children scheduled as first case duration of fasting was reduced with the shift of fasting regimen,
(morning patients OR = 0.50, 95% CI 0.30-0.81, P = .014; afternoon but we could still observe a trend for toddlers and infants to be
patients OR = 0.31, 95% CI 0.11-0.79, P = .020). The proportions of fasted for longer intervals than older children. Small children under
children fasting for more than 6 h stratified by the time of day for 1 year of age sometimes only drink breast milk or formula and hence
surgery and group are displayed in Figure 3. are fasted for 4 hour. This could perhaps explain the longer mean
50 | ANDERSSON ET AL.

FIGURE 2 Proportion of patients fasting more than 6 h in different age groups

FIGURE 3 Proportion of patients fasting more than 6 h dependent on time of day for procedure

fasting duration in this subgroup. Introducing the 6-4-0 fasting regi- T A B L E 3 Group, age, time of surgery, and in/outpatient status in
men reduced the incidence of prolonged fasting for the youngest relation to the outcome fasting 6 h or more. Results from crude
children under 36 months of age. Reducing preoperative starvation regression analysisa

has greatest significance in these youngest children, as initially sug- ORb 95% CIc P
8
gested by the seminal study of Thomas. Group
Pediatric patients are more sensitive to fasting than adults due to ENT_2h (1) Reference
their significantly smaller glycogen stores in liver and skeletal muscles, ENT_0h (2) 0.13 0.03-0.35 <.001
and the younger the child, the faster they will develop hypoglycemia MP_0h (3) 0.57 0.27-1.19 .135
and ketogenesis.9 Shortened preoperative fasting times reduce the
Age
concentration of ketone bodies, osmolality, and anion gap, diminish
<3 y Reference
postoperative insulin resistance, and improve hemodynamic condi-
3-6 y 0.52 0.23-1.15 .109
tions.10-12 Another potential problem with fasting among pediatric
>6 y 0.39 0.17-0.91 .031
patients is preoperative discomfort. Children allowed to drink close to
Time of surgery
surgery experience less thirst, hunger, and anxiety, show better behav-
3,13,14 First case Reference
ior, and are more comfortable. Some children may fast for many
hours without experiencing symptoms of hypoglycemia, while others Morning 0.50 0.30-0.81 .014

are more vulnerable. Moreover, 0-hour fasting for clear liquids is no Afternoon 0.31 0.11-0.79 .020

guarantee against hypoglycemia or dehydration in every patient, but it In/outpatient status

provides the opportunity for parents to intervene and ameliorate symp- Inpatients Reference
toms when their children are thirsty or grumpy, possibly because of Outpatients 0.79 0.40-1.57 .488
developing hypoglycemia (Table 3). a
Logistic regression analyses with each predictor at a time against the
outcome.
b
Odds ratio.
4.1 | Is this really safe? c
95% confidence interval for odds ratio.

The goal of preoperative fasting is to reduce the risk of perioperative

pulmonary acid aspiration. The fear of perioperative pulmonary aspi- emptying of clear fluids follows an exponential curve with a half-
ration is the biggest holdback for the implementation of more liberal time of 10-27 min.6,17,18 An empty stomach can be expected, but
fasting regimens. However, clear fluids ingested 1 or 2 h prior to not guaranteed, regardless of fasting regimen. The incidence of peri-
anesthesia do not seem to affect gastric volume or pH.13-16 Gastric operative pulmonary aspiration in children ranges from 1 to 10 in
ANDERSSON ET AL. | 51

10 000,19-25 and was not increased when applying 0-h fasting for ORCID
clear liquids.25 In the current study, there was no case of pulmonary
Hanna Andersson http://orcid.org/0000-0002-1462-0513
aspiration in any of the groups, but the study was not powered to
detect a difference in pulmonary aspiration.
Several conditions are associated with a higher risk of pulmonary
REFERENCES
aspiration, the majority due to delayed gastric emptying.20,21,23,24 Safe
anesthesia technique is the most important factor for minimizing the 1. Smith I, Kranke P, Murat I, et al. Perioperative fasting in adults and
risk of pulmonary aspiration. At our center, children with American children: guidelines from the European Society of Anaesthesiology.
Eur J Anaesthesiol. 2011;28:556-569.
Society of Anesthesiologists (ASA) physical classification class III or IV,
2. Practice Guidelines for Preoperative Fasting and the Use of Pharma-
and those with anatomical deviations or disorders leading to increased cologic Agents to Reduce the Risk of Pulmonary Aspiration: Applica-
risk for regurgitation and/or gastric motility disorders, as well as chil- tion to Healthy Patients Undergoing Elective Procedures: An
dren reporting problems with regurgitation, may sometimes be pre- Updated Report by the American Society of Anesthesiologists Task
Force on Preoperative Fasting and the Use of Pharmacologic Agents
scribed longer fasting intervals. We emphasize that it is always up to
to Reduce the Risk of Pulmonary Aspiration. Anesthesiology. 2017;
the attending anesthetist to judge if there is an increased risk of 126: 376-393.
regurgitation and thence adapt the prescribed preoperative fasting. 3. Engelhardt T, Wilson G, Horne L, et al. Are you hungry? Are you
thirsty? –– fasting times in elective outpatient pediatric patients.
Pediatr Anesth. 2011;21:964-968.
4.2 | Limitations 4. Arun BG, Korula G. Preoperative fasting in children: an audit and its
implications in a tertiary care hospital. J Anaesthesiol Clin Pharmacol.
As the 6-4-0 regimen had only been implemented in the ENT unit 2013;29:88-91.
for a year, we could expect high compliance to the regimen by ward 5. Cantellow S, Lightfoot J, Bould H, et al. Parents’ understanding of
and compliance with fasting instruction for pediatric day case sur-
nurses. If our follow-up survey had been conducted after more than
gery. Pediatr Anesth. 2012;22:897-900.
1 year, it is possible that the effect would have been less obvious, 6. Søreide E, Eriksson LI, Hirlekar G, et al. Pre-operative fasting guideli-
as suggested by the longer median fasting duration in the patients nes: an update. Acta Anaesthesiol Scand. 2005;49:1041-1047.
anesthetized in the main pediatric unit, where the 6-4-0 fasting regi- 7. Newton RJG, Stuart GM, Willdridge DJ, et al. Using quality improve-
ment methods to reduce clear fluid fasting times in children on a
men has been implemented for much longer. Furthermore, this study
preoperative ward. Pediatr Anesth. 2017;27:793-800.
was not powered for analysis of confounding factors such as age, 8. Thomas DK. Hypoglycaemia in children before operation: its inci-
out/inpatient status, and type of surgery, even though these factors dence and prevention. Br J Anaesth. 1974;46:66-68.
affect fasting intervals. 9. van Veen MR, van Hasselt PM, de Sain-van der Velden MG, et al.
Metabolic profiles in children during fasting. Pediatrics. 2011;127:
The information about the study at the preoperative assessment
e1021-e1027.
might also have influenced how the parents and patients complied
10. Dennhardt N, Beck C, Huber D, et al. Impact of preoperative fasting
with the fasting regimen. In addition, we cannot assume that all par- times on blood glucose concentration, ketone bodies and acid-base
ents reported fasting times correctly. Children may eat or drink with- balance in children younger than 36 months: a prospective observa-
out parents reporting it, or even knowing about it. However, this tional study. Eur J Anaesthesiol. 2015;32:857-861.
11. Dennhardt N, Beck C, Huber D, et al. Optimized preoperative fast-
effect should have been similar for all groups.
ing times decrease ketone body concentration and stabilize mean
Finally, the study was neither randomized nor blinded, thus arterial blood pressure during induction of anesthesia in children
observer bias was possible. younger than 36 months: a prospective observational cohort study.
Pediatr Anesth. 2016;26:838-843.
12. Nygren J. The metabolic effects of fasting and surgery. Best Pract
ACKNOWLEDGMENTS Res Clin Anaesthesiol. 2006;20:429-438.
13. Schreiner MS, Triebwasser A, Keon TP. Ingestion of liquids com-
The authors thank Albin Wiklund for his help with patient recruit- pared with preoperative fasting in pediatric outpatients. Anesthesiol-
ment and data collection and statistician Katarina Selling for assis- ogy. 1990;72:593-597.
€ 14. Brady M, Kinn S, Ness V, et al. Preoperative fasting for preventing
tance with statistical analysis. All medical personnel at the OPO
perioperative complications in children. Cochrane Database Syst Rev.
anesthesia unit and main pediatric anesthesia unit are acknowledged 2009;4:CD005285.
for their assistance with data collection. 15. Splinter WM, Schaefer JD. Unlimited clear fluid ingestion two hours
before surgery in children does not affect volume or pH of stomach
contents. Anaesth Intensive Care. 1990;18:522-526.
ETHICAL APPROVAL 16. Schmidt AR, Buehler P, Seglias L, et al. Gastric pH and residual vol-
ume after 1 and 2 h fasting time for clear fluids in children†. Br J
This study was approved by the Ethics Committee of Uppsala, Swe- Anaesth. 2015;114:477-482.
den, Dnr. 2014/487/1. 17. Schmitz A, Kellenberger CJ, Liamlahi R, et al. Gastric emptying after
overnight fasting and clear fluid intake: a prospective investigation
using serial magnetic resonance imaging in healthy children. Br J
CONFLICT OF INTEREST Anaesth. 2011;107:425-429.
18. Schmitz A, Kellenberger CJ, Lochbuehler N, et al. Effect of different
The authors have no conflicts of interest. quantities of a sugared clear fluid on gastric emptying and residual
52 | ANDERSSON ET AL.

volume in children: a crossover study using magnetic resonance 24. Borland LM, Sereika SM, Woelfel SK, et al. Pulmonary aspiration in
imaging. Br J Anaesth. 2012;108:644-647. pediatric patients during general anesthesia: incidence and outcome.
19. Kelly CJ, Walker RW. Perioperative pulmonary aspiration is infre- J Clin Anesth. 1998;10:95-102.
quent and low risk in pediatric anesthetic practice. Pediatr Anesth. 25. Andersson H, Zare n B, Frykholm P. Low incidence of pulmonary
2015;25:36-43. aspiration in children allowed intake of clear fluids until called to the
20. Olsson GL, Hallen B, Hambraeus-Jonzon K. Aspiration during anaes- operating suite. Pediatr Anesth. 2015;25:770-777.
thesia: a computer-aided study of 185,358 anaesthetics. Acta Anaes-
thesiol Scand. 1986;30:84-92.
21. Walker RW. Pulmonary aspiration in pediatric anesthetic practice in
the UK: a prospective survey of specialist pediatric centers over a How to cite this article: Andersson H, Hellstrӧm PM,
one-year period. Pediatr Anesth. 2013;23:702-711. Frykholm P. Introducing the 6-4-0 fasting regimen
22. Warner MA, Warner ME, Warner DO, et al. Perioperative pulmonary
and the incidence of prolonged preoperative
aspiration in infants and children. Anesthesiology. 1999;90:66-71.
23. Murat I, Constant I, Maud’huy H. Perioperative anaesthetic morbid- fasting in children. Pediatr Anesth. 2018;28:46-52.
ity in children: a database of 24,165 anaesthetics over a 30-month https://doi.org/10.1111/pan.13282
period. Pediatr Anesth. 2004;14:158-166.
Accepted: 17 October 2017

DOI: 10.1111/pan.13286

RESEARCH REPORT

Ultrasound-guided pudendal nerve block in children: A new

technique of ultrasound-guided transperineal approach

Isabelle Gaudet-Ferrand1 | Pablo De La Arena1 | Sophie Bringuier2 | Olivier Raux1 |

Laurent Hertz1 | Nicolas Kalfa3 | Chrystelle Sola1 | Christophe Dadure4,5

1
Department of Pediatric Anesthesia,
Lapeyronie University Hospital, Montpellier Summary
University, Montpellier, France Background: Transperineal pudendal nerve block guided by nerve stimulator is used
2
Department of Anesthesiology and Critical
in pediatric anesthesia as an alternative to caudal analgesia in perineal surgery. The
Care Medicine, Lapeyronie University
Hospital, Montpellier, France risk of rectal puncture or intravascular injection is inherent to this blinded technique.
3
Department of Pediatric Surgery, We described a new technique of transperineal pudendal nerve block, with ultra-
Lapeyronie University Hospital, Montpellier,
France sound guidance, to improve safety of the technique.
4
Department of Pediatric Anesthesia, Aims: The first goal of this study was to describe this new technique and to test its
Lapeyronie University Hospital, Montpellier,
feasibility. The second objective was to evaluate intra operative effectiveness and
France
5
Inserm Unit 1051 INM, Montpellier,
postoperative pain control.
France Methods: After parental and children consent, this prospective descriptive study

Correspondence
included children aged 1-15 years, ASA status I-III, scheduled for general anesthesia
Isabelle Gaudet-Ferrand, Department of associated with bilateral pudendal nerve block for an elective perineal surgery. After
Pediatric Anesthesia, Lapeyronie University
Hospital, Montpellier, France.
standardized general anesthesia, the anesthesiologist performed pudendal nerve
Email: [email protected] block under ultrasound guidance with “out of plane” approach and evaluated the

Funding information
visualization of anatomical structures (ischial tuberosity, rectum, and pudendal
Fundings were obtained from Montpellier artery), of the needle and of the local anesthetic spread. Pudendal nerve block
University Hospital.
failure was defined as an increase in mean arterial blood pressure or heart rate more
Section Editor: Adrian Bosenberg than 20% compared to baseline values after surgical incision. In the postoperative
period, the need for rescue analgesia was noted.
Results: During the study period, 120 blocks were performed in 60 patients, includ-
ing 59 boys. Quality of the ultrasonographic image was good in 81% of blocks, with
easy visualization of ischium and rectum in more than 95% of cases. Localization of
the tip of the needle was possible for all pudendal nerve blocks, directly or indi-
rectly. The spread of local anesthetic was seen in 79% of cases. The block was
effective in 88% of cases.
Conclusion: The new technique of ultrasound-guided pudendal nerve block, described
in this study, seems to be easy to perform with a good success rate, and probably
improves safety of the puncture and of the injection by real-time visualization of
anatomical structures and local anesthetic spread.

KEYWORDS
anesthesia, child, pain, pudendal nerve, regional, ultrasound

Pediatric Anesthesia. 2018;28:53–58. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 53
54 | GAUDET-FERRAND ET AL.

1 | INTRODUCTION
What is already known
The pudendal nerve is a mixed nerve, formed from the anterior rami
of the second, third, and fourth nerves of the sacral plexus (S2-S4).
• Transperineal pudendal nerve block guided by nerve
stimulator is an efficient analgesia technique for perineal
Its terminal branches, that leave the Alcock’s canal and travel
surgery in children.
through the ischiorectal fossa, ensure the innervation of a big part
What new information this study adds
of the perineum (forming a triangle with base located behind the
anus and tip in front of the penis or clitoris). The efficacy of puden- • Using a similar transperineal approach, the pudendal
dal nerve block is well known in clinical practice. In adults, this tech- nerve block can be easily performed under real-time
nique is widely used, but only few studies are referenced in pediatric ultrasound guidance, avoiding occult rectal puncture and
anesthesia. In children, it is considered as an efficient alternative of intravascular injection.
caudal analgesia in hypospadias surgery with better postoperative
pain management.1,2 The transperineal approach guided by nerve
stimulator is considered as the reference in efficacy studies despite this new technique, and the occurrence of complication during
lack of data comparing the technique with or without use of nerve its performance. Evaluation of efficiency on pain relief (morphi-
1-6 nomimetic consumption in the intraoperative period, and rescue
stimulator. Different types of motor responses are possible with
nerve stimulation: rectal (anal sphincter contraction) or perineal (con- analgesia requirement in the postoperative period) was the second
traction of the transversalis perineum superficial muscle and move- objective.
ment of the penis/clitoris). Usually, the local anesthetic is injected
when at least one of these responses is present.3 A variant of this
2.2 | Study protocol
technique requires 2 injection points (anterior and posterior).4-6 Even
if no complication related to these transperineal “blinded” approach Perioperative anesthetic management was standardized. Propofol
was described in the literature, the risk of rectal or vascular puncture and/or sevoflurane were used for induction, in association with a
is inherent to the technique. In children, ultrasound (US) guidance continuous infusion of remifentanil at 1 lg/kg/min for 2 minutes.
has revolutionized the practice of regional anesthesia, with increas- Airway control was assured through insertion of a tracheal tube or a
ing use regularly this last decade.7 Advantages of US guidance in laryngeal mask. The maintenance of anesthesia was provided by
pediatric regional anesthesia are the real-time visualization of sevoflurane (1.2 MAC) in a equal mix of oxygen and nitrous oxide,
patient’s anatomy, the precise positioning of the needle, and the and remifentanil 0.05 lg/kg/min. Mechanical ventilation was
direct visualization of local anesthetic spread. New techniques based adapted to obtain a normal range of EtCO2 (30-40 mm Hg). Then,
on US guidance have also appeared. The aim of this study was to US-guided PNB was performed bilaterally in standard aseptic condi-
evaluate the feasibility and efficacy of a new transperineal technique tion following local protocol, by a senior anesthesiologist or a resi-
of US-guided pudendal nerve block in children, in order to improve dent trained in the technique. If an inguinal incision was required for
safety of the puncture by visualizing ischiorectal fossa and the adja- surgery, an US-guided ilio-inguinal or transversus abdominis plane
cent anatomical structures. block was added to pudendal nerve block. Baseline values of blood
pressure and heart rate were defined under general anesthesia with-
out nociceptive stimulation and after regional anesthesia procedure.
2 | MATERIALS AND METHODS
In all patients, perineal skin incision was allowed at least 15 minutes
after performing the PNB. The return to baseline infusion rate of
2.1 | Study design and participants
remifentanil (0.05 lg/kg/min) was required before perineal incision.
This prospective descriptive study was conducted in the Pediatric An increase in mean arterial blood pressure or heart rate more than
Anaesthesia Unit in Lapeyronie University hospital (Montpellier, 20% compared to baseline values was considered as an unsuccessful
France). Institutional Review Board approval was received from the PNB. In case of block failure, incremental remifentanil titration was
South Mediterranean IV Ethical Board. Written informed parental started. Fifteen minutes before the end of surgery, all patients
consent was obtained from all participants and assent of children received systemic analgesia, using paracetamol (Perfalganâ; Bristol-
was checked after explanations appropriate to their age. All chil- Myers Squibb, Rueil-Malmaison, France) 15 mg/kg and ketoprofen
dren aged 1-15 years, ASA status class I-III, scheduled for general (Profenidâ; Sanofi-Aventis, Paris, France) 1 mg/kg if there was no
anesthesia associated with bilateral pudendal nerve block for an contraindication. Remifentanil was stopped after closure of the skin.
elective perineal surgery, were candidate to participate. Exclusion Rescue analgesia by IV nalbuphine (Nubainâ; Dupont Pharma, Paris,
criteria were any contraindication to regional anesthesia, ASA France) 0.2 mg/kg was administered in the postanesthetics care unit
physical status IV-V, and refusal by the parents or the child. Data (PACU) if necessary. Postoperative analgesia was also standardized
were collected from induction of anesthesia to H24 postopera- such as systematic oral paracetamol (15 mg/kg) every 6 hours asso-
tively after admission in recovery room. The primary objective of ciated with oral ibuprofen (10 mg/kg) every 8 hours. Oral tramadol
this descriptive study was to evaluate the feasibility and ease of (1 mg/kg) was prescribed as rescue analgesia.
GAUDET-FERRAND ET AL. | 55

2.3 | Ultrasound-guided pudendal nerve block

procedure
After general anesthesia induction and airway control, children were
positioned supine, hips in abduction, flexed legs, and soles of the
feet joined (“Frog position”). Ultrasound images were obtained using
a portable US unit (LOGIQe ; GE Healthcare, Wauwatosa, WI, USA)
and an 8-13 MHz linear array probe. The long axis of the US probe
was positioned on a horizontal line connecting the ischial tuberosity,
previously palpated, to the anus (Figure 1). Ischial tuberosity is read-
ily identified such as a hypoechoic area, bounded superiorly by a
hyperechoic line. The probe was then moved medially on the same
axis, until the rectum appeared as another hypoechoic area. The
ischiorectal fossa is located between the ischium, laterally and the
rectum, medially. Pudendal artery can be searched using color Dop-
pler (Figure 2). After careful skin preparation with alcohol povidone- FIGURE 2 Pudendal artery with color Doppler, in the ischiorectal
iodine according to our local hygiene protocol, a 22 Gauge 50 mm fossa
insulated needle (Sonoplex stimâ, Pajunk, Geisingen, Germany) was
then introduced at the middle of the superior edge of the probe, in
an out-of-plane approach, with an inclination of 15° in the sagittal
plane, in an anterior-posterior direction (Figure 3). Position of the tip 2.4 | Measurement and data collected
of the needle was identified by direct vision, through movement of For each block, feasibility and ease of technique were as assessed
adjacent anatomical structures or with saline injection. Nerve stimu- by the following data: sonographic identification of anatomical
lation (0.5-1 mA) was started only after the operator considered the structures and visualization of needle and local anesthetic spread.
tip of needle in correct placement in the ischiorectal fossa, and the The sonographic identification of ischium and rectum was evaluated
potential motor response was noted. After a negative aspiration test, as easy, hard, or impossible. Artery identification, with and without
0.2 mL/kg of a solution of 0.2% Ropivacaine (Naropeineâ, Astra Doppler, was noted. The occurrence of response to nerve stimula-
Zeneca, Rueil Malmaison, France) added to 1 lg/mL of clonidine tion (2 Hz, 0.5-1 mA) and its specificity (rectal, perineal), puncture
(Catapressanâ, Bohringer Ingelheim, Reims, France), was injected in complications, and duration of the procedure were also collected.
the ischiorectal fossa under real-time US scanning (Figure 4). Intraoperative pain was assessed by hemodynamic parameters and
the maximum rate of remifentanil infusion required during surgery.
The efficacy of postoperative analgesia was assessed using a
FLACC scale8 in children up to 6 years, and VAS (visual analog
scale) in older children. This assessment was repeated every
15 minutes in the PACU and every 4 hours until discharge. If the
FLACC or VAS score was >3, the child received nalbuphine in
PACU or tramadol in hospitalization unit as rescue analgesia. Post-
operative analgesia was evaluated during 24 hours through the
need of rescue analgesia.

2.5 | Statistical analysis

Patients’ characteristics were presented using mean and SD for con-
tinuous variables and frequencies and proportions for categorical
variables. Comparisons between successful and unsuccessful puden-
dal nerve block were done using Student or Wilcoxon rank test for
continuous variables and chi-square or Fisher test for categorical
ones. Correlations between continuous variables were performed
using Pearson correlation coefficient. Statistical bilateral significance
threshold was set at 5%. Statistical analyses were performed using
F I G U R E 1 Anatomical landmarks of ultrasound-guided pudendal SAS version 9.1 (SAS Institute, Cary, NC, USA).
nerve block: the ischiorectal fossa
56 | GAUDET-FERRAND ET AL.

T A B L E 1 Surgery indication and type of regional anesthesia

Surgery indication (regional anesthesia) n (%)
Perineal incision (PNB) 46 (76.67%)
Hypospadias 19 (31.67%)
Phimosis (posthectomy/circumcision) 20 (33.33%)
Other penis surgery 4 (6.66%)
Testicle surgery 2 (3.33%)
Hypertrophic minor labia 1 (1.66%)
Inguino-scrotal incision (PNB+TAPB or II/IHB) 14 (23.3%)
Orchidopexia 13 (21.66%)
Penis+testicle surgery 1 (1.66%)

Bold values indicate the sum of the perineal incision and the sum of the
ilioinguinal-scrotal incision. PNB, pudendal nerve block; TAPB, transver-
sus abdominis plane block; II/IHB, ilio-inguinal/ilio-hypogastric block.
F I G U R E 3 Puncture of transperineal pudendal nerve block under
ultrasound guidance
call of the ambulatory department and 2 families were missed to call.
Data on postoperative pain control were lacking in these 6 patients.

3.2 | Feasibility of ultrasound-guided PNB

The US-guided blocks were performed by residents in 78.33% of case
and in 21.67% by senior practitioners. The quality of the picture was
judged “good” by the practitioner in 81.51% of cases, “intermediate” in
18.49%, and was never considered “bad”. The ischium and the rectum
were easily identified in more than 95% of cases (Table 2). Using Dop-
pler color, the pudendal artery identification was possible in 81.74% of
pictures. Localization of the tip of the needle was possible for all
blocks: directly in 17.8% of punctures or indirectly, by the movement
of adjacent anatomic structures, in 98.31% of cases. The mean depth
of injection was 20.3 9 mm, with no significant correlation to
patient’s weight (r = .3). The spread of local anesthetic was seen in
79.3% of cases. For 116 blocks (96.6%), a single puncture was neces-
sary. A second puncture was necessary in 4 cases. The first needle
F I G U R E 4 Ultrasound-guided pudendal nerve block: visualization
placement was good in 68.7% of cases, whereas 1 reorientation
of local anesthetic spread in the ischiorectal fossa
was noted in 24.35%, and more than 1 in 6.95% (maximum 3). The
mean duration of the technique from ultrasonographic identification
of anatomical structures to withdrawal of the needle was
3 | RESULTS
139.6 78 seconds. Motor response was found in 23.7% of cases.
There was anal sphincter contraction (63% of cases), contraction of
3.1 | Demographic data
the bulbospongious muscle (22%), and contraction of the deep
During the study period, 120 US-guided pudendal nerve blocks were transverse perineal muscle (15%). Only one side effect was noted
performed on 60 patients (59 boys, 1 girl) consecutively included. (blood during aspiration test, despite visualization of pudendal artery
No refusal was registered. The mean weight was 20.8 12.4 kg. with Doppler), and no injection was performed. The needle was
Forty-eight (40%) children were younger than 3 years, 40 (33.3%) relocated and injection of the local anesthetic was done without
were 3-6 years old and 32 (26.7%) were older than 6 years. complication.
Surgical indications are detailed in Table 1. Most patients had
surgery for hypospadias (31.67%) or phimosis (33.33%). An additional
3.3 | Efficacy of ultrasound-guided pudendal nerve
inguinal incision was indicated in 25% of children (Table 1).
block
Data on feasibility of the technique were extensively collected,
as well as data on intraoperative effectiveness. Day-case surgeries The surgical incision was performed an average of 39 21 minutes
were performed in 35 (58.3%) children. In these patients, 6 were lost after injection of the local anesthetic. The block was considered
to the exit of the hospital: 4 families did not answer to the phone effective in 88.33% of cases, with no increase in remifentanil
GAUDET-FERRAND ET AL. | 57

T A B L E 2 Identification of anatomical landmarks on 120 practice of the technique, and better adaptation of needle inclination.
ultrasound-guided pudendal nerve blocks The aspiration test before local anesthetic injection remains essential
Visualization n (%) to avoid intravascular injection with real-time US. The use of

Anatomical structures Easy Difficult Impossible epinephrine is always contraindicated in pudendal nerve block due to
the risk of vasoconstriction on a terminal vascularization.
Ischium 116 (98.31%) 2 (1.69%) 0 (0%)
A surprising ascertainment in this study is the low rate of
Rectum 114 (97.44%) 3 (2.56%) 0 (0%)
responses to neurostimulation, possibly explained by the position of
the needle in the middle of the ischiorectal fossa, far from the emer-
infusion rate during perineal incision. Mean maximum dose of gence of pudendal nerve from Alcock’s canal. The systematic motor
remifentanil was 0.06 0.03 lg/kg/min. response was not mandatory for the injection of local anesthetic in
No relationship was found between successful pudendal nerve our study. Also, we noted that the lack of response was not corre-
block and visualization of anatomical structures, visualization of lated with block failure. This can be explained because pudendal
spread of local anesthetic, or presence of a motor response to nerve nerve block is a spread block. Anatomical studies showed great vari-
stimulation. ations of the pudendal nerve throughout its path, but the way of
In the postoperative period, 60% of children received nalbuphine terminal branches always goes through the ischiorectal fossa, which
in PACU. No rescue analgesia was required in 80% of hospitalized is a fatty space allowing good diffusion of local anesthetic.10,11
children after PACU during the first 24 hours. In ambulatory patient, Pudendal nerve and its branches in the ischiorectal fossa have a
acetaminophen and ketoprofen were sufficient in 76% of cases. No small diameter.9 It is also difficult to identify them with US. Never-
relationship was found between successful PNB and need of rescue theless, neurostimulation use in association with US guidance can be
analgesia in the postoperative period. safe to avoid direct nerve injury.
Alternative techniques are not desirable in children due to the
need for prone or lateral decubitus position. A transgluteal approach
4 | DISCUSSION for US-guided pudendal nerve block has been previously described
in adults.12-14 Tip of the needle was positioned in the interligamen-
The results of this prospective descriptive study show that the per- tous plane between sacrospinous and sacrotuberous ligaments, medi-
formance of real-time US-guided transperineal approach of pudendal ally to the pudendal internal artery, at the level of ischial spine. The
nerve block is feasible and easy in children under general anesthesia. risk of associated blockade of sciatic nerve/sacral plexus is about
This US-guided technique allows an efficient postoperative pain 13%,14 which limits its use in ambulatory surgery. Another technique
management with low rate of patient requiring rescue analgesia is described with a pudendal nerve approach at the entrance of the
beyond the PACU. Alcock’s canal.15
As noted by our anesthesiologist team, the technique is easy to The efficacy of US-guided PNB, as described in the present
learn because the puncture site using the US-guided approach was study, is good in the perioperative period. Mean dose of remifentanil
sensibly unchanged in comparison with landmark technique.1-3 In was low.
addition, the majority of blocks were performed by residents, under Surprisingly, we found a discrepancy between the efficacy of the
the supervision of a senior anesthesiologist. The US visualization of nerve block intraoperatively and the use of nalbuphine in PACU. This
ischium and rectum was always possible. This is a crucial point in result could be partly explained by the high incidence of postopera-
improving safety. Also, the choice of puncture site and needle orien- tive delirium in children, which was not specifically distinguished
tation can be more suitable as to avoid a puncture of the rectum, from pain. This discrepancy suggests inappropriate use of nalbuphine
which is a septic risk situation. for its sedative effects. Moreover, in 25% of children, an inguinal
This safety is enhanced by the visualization of the tip of the nee- incision was associated with the perineal incision, and evaluation of
dle, which was indirect (movement of adjacent anatomic structures) in the only perineal pain was not possible in PACU. At last, insufficient
a large part of blocks. The difficulties in the visualization of pudendal analgesia can be underestimated under general anesthesia, due to
artery are probably explained by its small diameter and sinuous path, depression of the hemodynamic response to nociception.
as described in anatomical studies. Even in case of artery visualization, Efficacy of pudendal nerve block in children was demonstrated
this sinuous path and the eventuality of multiple arteries in the in hypospadias repair. In comparison with caudal analgesia, pudendal
ischiorectal fossa,9 make its avoidance uncertain in this “out of plane” nerve block provided more effective postoperative analgesia during
approach. Aspiration test and visualization of the local anesthetic the first 24 hours, improving parent satisfaction.1,2 For pediatric cir-
spread are better ways to confirm lack of intravascular injection. This cumcision, lower pain scores were noted with pudendal nerve block
spread has been seen in almost 80% of cases. The lack of spread visu- compared to dorsal penile nerve block during the first 12 hours
alization can possibly be explained by the fat density that makes it postoperatively.6
almost as anechoic as that of the local anesthetic, or by a too deep There are a few limitations in this study, which examined the
injection. In our personal experience, direct visualization of the needle feasibility of US-guided PNB in a transperineal approach. The spread
and of local anesthetic spread seem to be improved with regular of local anesthetic in the ischiorectal fossa and its potential effect
58 | GAUDET-FERRAND ET AL.

on success of pudendal nerve block were not evaluated, not allowing 2. Kendigelen P, Tutuncu AC, Emre S, Altindas F, Kaya G. Pudendal
to know if an injection too deep entailed a failure of the block. Our Versus Caudal Block in Children Undergoing Hypospadias Surgery: A
Randomized Controlled Trial. Reg Anesth Pain Med. 2016;41:610-
results concerning postoperative pain should be the subject of fur-
615.
ther investigations. Efficacy of the technique was not the primary 3. Aissaoui Y, Bruye re R, Mustapha H, Bry D, Kamili ND, Miller C. A
objective, and pain evaluation was based only on analgesic consump- randomized controlled trial of pudendal nerve block for pain relief
tion, and not on pain scores. The use of postemergence delirium after episiotomy. Anesth Analg. 2008;107:625-629.
4. Naja Z, El-Rajab M, Al-Tannir M, et al. Nerve stimulator guided
score would be necessary to better evaluate pain level in PACU. The
pudendal nerve block versus general anesthesia for hemorrhoidec-
optimal volume of local anesthetic in the ischiorectal fossa remains tomy. Can J Anaesth. 2006;53:579-585.
to be determined in children. Unfortunately, the study did not have 5. Naja Z, Ziade MF, Lo € nnqvist PA. Nerve stimulator guided pudendal
any control group, difficult to determine in view of the diversity of nerve block decreases posthemorrhoidectomy pain. Can J Anaesth.
2005;52:62-68.
surgical procedures. So, the efficacy, in each indication, of this new
6. Naja Z, Al-Tannir MA, Faysal W, Daoud N, Ziade F, El-Rajab M. A
US-guided technique has to be demonstrated in randomized studies. comparison of pudendal block vs dorsal penile nerve block for cir-
cumcision in children: a randomised controlled trial. Anaesthesia.
2011;66:802-807.
5 | CONCLUSION 7. Polaner DM, Taenzer AH, Walker BJ, et al. Pediatric Regional Anes-
thesia Network (PRAN): a multi-institutional study of the use and
incidence of complications of pediatric regional anesthesia. Anesth
The aim of this study was to evaluate the feasibility of a new approach Analg. 2012;115:1353-1364.
of transperineal pudendal nerve block in children, under US guidance. 8. Merkel SI, Voepel-Lewis T, Shayevitz JR, Malviya S. The FLACC: a
This technique is easy to learn and may improve safety of the puncture behavioral scale for scoring postoperative pain in young children.
Pediatr Nurs. 1997;23:293-297.
by direct visualization of anatomical landmarks and of local anesthetic
9. Fronho€ fer G, Niernberger T, Anderhuber F. Variations in the branch-
spread, thus avoiding rectal puncture or intravascular injection. Bene- ing of the internal pudendal artery in the ischioanal fossa. Ann Anat.
fits of the technique about efficiency and safety, compared with 1996;178:133-135.
blinded techniques, must be further evaluated in randomized studies. 10. Mahakkanukrauh P, Surin P, Vaidhayakarn P. Anatomical study of
the pudendal nerve adjacent to the sacrospinous ligament. Clin Anat.
2005;18:200-205.
11. Schraffordt SE, Tjandra JJ, Eizenberg N, Dwyer PL. Anatomy of the
ETHICAL APPROVAL
pudendal nerve and its terminal branches: a cadaver study. ANZ J
Institutional Review Board approval was received from the South Surg. 2004;74:23-26.
12. Kovacs P, Gruber H, Piegger J, Bodner G. New, simple, ultrasound-
diter-
Mediterranean IV Ethical Board on July 08, 2014 (CPP Sud Me
guided infiltration of the pudendal nerve: ultrasonographic tech-
ranee IV, registration number: 2014-A00499-38). nique. Dis Colon Rectum. 2001;44:1381-1385.
13. Rofaeel A, Peng P, Louis I, Chan V. Feasibility of real-time ultrasound
for pudendal nerve block in patients with chronic perineal pain. Reg
DISCLOSURES Anesth Pain Med. 2008;33:139-145.
14. Bellingham GA, Bhatia A, Chan CW, Peng PW. Randomized con-
No authors have any conflicts of interest or disclosures with regard trolled trial comparing pudendal nerve block under ultrasound and
to this study. fluoroscopic guidance. Reg Anesth Pain Med. 2012;37:262-266.
15. Bendtsen TF, Parras T, Moriggl B, et al. Ultrasound-Guided Pudendal
Nerve Block at the Entrance of the Pudendal (Alcock) Canal:
ORCID Description of Anatomy and Clinical Technique. Reg Anesth Pain
Med. 2016;41:140-145.
Isabelle Gaudet-Ferrand http://orcid.org/0000-0002-3567-6460
Laurent Hertz http://orcid.org/0000-0001-9501-3434
Chrystelle Sola http://orcid.org/0000-0002-8375-9121
How to cite this article: Gaudet-Ferrand I, De La Arena P,
Bringuier S, et al. Ultrasound-guided pudendal nerve block in
children: A new technique of ultrasound-guided transperineal
REFERENCES approach. Pediatr Anesth. 2018;28:53-58. https://doi.org/
10.1111/pan.13286
1. Naja ZM, Ziade FM, Kamel R, El-Kayali S, Daoud N, El-Rajab MA. The
effectiveness of pudendal nerve block versus caudal block anesthesia
for hypospadias in children. Anesth Analg. 2013;117:1401-1407.
Accepted: 13 October 2017

DOI: 10.1111/pan.13283

RESEARCH REPORT

Clinical implications of de Barsy syndrome

Lindsay L. Warner | David A. Olsen | Hugh M. Smith

Department of Anesthesiology, Mayo Clinic,

Rochester, MN, USA Summary
Background: De Barsy syndrome is a rare, autosomal recessive syndrome character-
Correspondence
Hugh M. Smith, Department of ized by cutis laxa, progeroid appearance, ophthalmic opacification, skeletal malfor-
Anesthesiology, Mayo Clinic, Rochester, MN, mations, growth delays, and intellectual disability.
USA.
Email: [email protected] Aims: The aim of this case series is to identify the anesthetic considerations in the
clinical management of patients with de Barsy syndrome.
Funding information
The study was funded by departmental Methods: A retrospective case review from 1968 to 2016 was performed at a
resources. single tertiary medical center to identify patients with de Barsy syndrome who
Section Editor: Dr Francis Veyckemans underwent anesthesia for diagnostic and surgical procedures. We collected and
analyzed the perioperative records and following data: age, sex, American Society
of Anesthesiologists physical status, relevant comorbidities, surgical procedures,
anesthesia management, and observed complications.
Results: Three patients underwent 64 unique anesthetics for a diverse collection
of diagnostic and surgical procedures. An array of anesthetics and techniques
were successfully used. Observations of the perioperative period found 7 epi-
sodes of intraoperative hyperthermia (>38.3°), a single difficult airway requiring
fiberoptic bronchoscopic-guided intubation, and repeatedly difficult intravenous
access.
Conclusion: This expanded case series suggests that providers caring for patients
with de Barsy syndrome should be aware of potential challenges with airway man-
agement, vascular access, and temperature monitoring.

KEYWORDS
cutis laxa, fever, genetic diseases, rare diseases, syndrome

1 | INTRODUCTION The constellation of clinical manifestations of de Barsy syndrome

include intrauterine growth retardation, postnatal growth delay,
De Barsy syndrome is a rare genetic syndrome characterized by cutis motor delay, cognitive impairment, hypotonia, athetoid movements,
laxa III (ARCL3 OMIM# 219150, 614438), progeroid appearance, microcephaly, wormian bones, large fontanelles, facial dysmor-
ophthalmic opacification, skeletal malformations, growth delays, and phism, cataracts, corneal clouding, thin/wrinkled skin, easy bruis-
1-7
intellectual disability. Given the diverse array of ophthalmic and ing, sparse hair, joint laxity, osteopenia, and inguinal hernias.1-14 A
skeletal malformations, de Barsy patients require frequent surgical number of de Barsy syndrome manifestations are directly relevant
interventions.8 First described by de Barsy in 1967,1 only 100 cases to perioperative patient management.3 Craniofacial abnormalities,
of the syndrome have been documented in the medical literature. cognitive delays, and skin fragility (ie, “paper thin”) are all issues
Several cases of de Barsy syndrome have been linked to mutations requiring specific perioperative patient care consideration.
of the PYCR1 or ALDH18A1 genes.9 The syndrome is inherited in A single investigation relating to anesthetic management of de
an autosomal recessive pattern. Barsy syndrome patients was published in 2010 by authors of the

Pediatric Anesthesia. 2018;28:59–62. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 59
60 | WARNER ET AL.

present study.3 This article reported on 2 patients and 35 anesthet-

ics associated with surgical and diagnostic procedures between 1968 What is known
and 2007. The present study identifies another patient with de
Barsy syndrome and 29 additional anesthetics performed between
• In a previously published article on anesthetic manage-
ment of patients with de Barsy syndrome, no major peri-
2007 and 2016. Given the rarity of this condition, and limited pub-
operative adverse events were found in 35 separate
lished data, this case series provides important information to provi-
anesthetic cases.
ders encountering de Barsy syndrome patients for the first time or
What this article adds
during emergent situations.
• This article expands upon the previous series with 29
additional anesthetic cases, and offers management con-
2 | MATERIALS AND METHODS siderations with respect to airway management, vascular
access, temperature monitoring and other issues unique
After receiving Institutional Review Board (Rochester, MN, USA) to de Barsy syndrome.
study approval, medical records prior to September 2016 were
searched to identify patients with de Barsy syndrome. As no single
longitudinal comprehensive medical index at our institution spans agents, propofol, or a combined technique were used in 54%, 39%,
this time frame, several medical databases were queried including and 8%, respectively. For analgesia, 2 caudal blocks and 1 peripheral
the Berkson Medical and Surgical Indexing Systema [1968-1975], the extremity block were performed. Fentanyl was the dominant opioid
Mayo Clinic Life Science System database [1996-2007] and the administered and used in 59% of all anesthetics. No significant medi-
Mayo Clinic Medical Index [1995-2016]. No exclusionary criteria cation reactions or complications were observed. A comprehensive
were applied. All available anesthetic and surgical records of de list of medications used (number of times) is as follows:
Barsy syndrome patients were analyzed. Data collected included age,
gender, ASA physical status, relevant comorbid conditions, surgical • Induction: Propofol (25), sodium pentothal (2)
procedure, airway management, vascular access, monitoring, anes- • Opioids: Fentanyl (38), oxymorphone (3), morphine (2), oxycodone
thetic induction and maintenance, intravenous and oral analgesics, (2), sufentanil (1)
muscle relaxants, antibiotics, and blood product transfusions, PACU • Antibiotics: Cefazolin (25), piperacillin sodium/tazobactam sodium
information, and postoperative complications. Clinical observations (1), gentamicin (1), vancomycin (1)
made by anesthesia providers were also recorded. • Antiemetics: Ondansetron (25), dexamethasone (10), droperidol
(3), granisetron (1)
• Vasoactive: Atropine (7), labetalol (1), phenylephrine (1), racemic
3 | RESULTS epinephrine (1)
• Other: Glycopyrolate (14), neostigmine (12), acetaminophen (11),
In 3 different pediatric patients with genetically confirmed de Barsy midazolam (6), albuterol (3), lidocaine (3), calcium gluconate (1),
syndrome, a total of 64 anesthetics were reviewed. One female and 2 clonidine (1), diazepam (1), heparin (1), ibuprofen (1), papaverine
male patients had each undergone 14, 11, and 39 anesthetics, respec- (1), tranexamic acid (1)
tively. With respect to airway management, 7 oral/nasal airways, 6 • Neuraxial: Bupivacaine (2), hydromorphone(1)
laryngeal mask airways, and 48 endotracheal tubes were placed. Of the • Regional: Bupivacaine(1)
latter, 23 were placed by direct laryngoscopy, 24 by videolaryngoscopy,
and 1 by fiberoptic intubation for unanticipated difficult airway. In this In 7 distinct cases, intraoperative hyperthermia (tempera-
patient, although providers had been able to secure his airway easily by ture > 38°C) was noted. This represents 11% of total cases, and
direct laryngoscopy prior to 2012, subsequent airway management occurred in more than 1 child. Difficulty was reported with peripheral
required videolaryngoscopy with a maximum of 2 attempts for success- venous line placement due to the cutis laxa component of the syn-
ful intubation. This same patient, by the age of 15, also began requiring drome. Eleven arterial lines (9 radial, 2 femoral) and 4 central lines (2
2-handed mask ventilation. In the other 2 patients, no difficulties were right internal jugular, 2 femoral) were placed without difficulty. Blood
reported with mask ventilation or endotracheal intubation. A single inci- products, albumin, neuromonitoring, central line placement, and periph-
dent of vomiting during preoxygenation was documented which eral blocks were all administered without recognized complications.
required rapid sequence intubation. The remaining airways were man-
aged with a mask or spontaneous breathing.
A wide range of inhalational agents and intravenous medication 4 | DISCUSSION
were used throughout the 64 anesthetics. For induction, inhalational
Although basic science has begun to unravel the genetic basis of de
a
An early epidemiologic database established by Dr. Joseph Berkson in 1935. Searchable
Barsy syndrome, the literature describing clinical management of
years: 1968-1975. these patients remains exceedingly limited. Several known genetic
WARNER ET AL. | 61

mutations are thought to explain the physical manifestations of de 10.3% of cases were associated with hyperthermia, all of which
Barsy syndrome. These include a mutation in the pyrroline-5-carbox- responded to active cooling alone. Iatrogenic intraoperative over-
ylate reductase gene (PYCR-1, known as type B) and the ALDH18A1 warming remains the most likely explanation of hyperthermia in our
13,15
gene (type A). Malfunctioning of these genes leads to mito- case series. This explanation may be seen in the context of limited
chondrial fragmentation with high apoptosis rates during oxidative subcutaneous insulation due to cutis laxa, and relatively low body
stress.4 A third mutation, ATP6V0A2-CDG, is also thought to be mass indexes (range: 11-19), both of which would increase suscepti-
possibly related to de Barsy syndrome.16 Two recent publications bility to room warming, heating lamps, or forced air blankets.
have described additional clinical features in de Barsy syndrome Accordingly, close attention to temperature monitoring and the use
patients. In 2016, Dutta et al reported slowly progressive dilatation of warming devices constitutes prudent clinical care of de Barsy syn-
of the aortic root, suggesting the need for cardiac evaluation and drome patients.
longitudinal monitoring.4 Lin et al described glaucoma and idio- Providers should also consider having advanced airway manage-
pathic hypertrophic pyloric stenosis in patients with de Barsy syn- ment equipment readily available. Although craniofacial abnormalities
drome.15 are present in this syndrome, they did not present severe airway
With regard to perioperative management, little is known about management challenges in the overwhelming majority of anesthetics.
Barsy syndrome patients. Extrapolating from known physical abnor- Difficulty with both mask ventilation and endotracheal intubation
malities, features such as joint hyperlaxity may pose positioning were observed in our series but limited to clinical experience with 1
issues, while pectus excavatum could present difficulty with ventila- patient. In the era of videolaryngoscopy, safe airway management
tion.15 Other features, such as developmental delay, present univer- would be anticipated.
sal concern with regard to patient coping and cooperation during the While patient safety cannot be definitively established by any
delivery of anesthesia. Given the rarity of this condition, and limited case series, given the extreme rarity of the syndrome, this report
published data, making available institutional experience pertaining attempts to provide successful management options to potential
to clinical management offers providers of de Barsy syndrome anesthesia providers of patients with de Barsy syndrome. And as
patients a basis of care. This study extends the published anesthesia widespread genetic testing is capable of confirming new cases, fur-
experience with de Barsy syndrome, including various anesthetic ther research will enhance our understanding of the risks surgery
techniques and pharmacologic agents safely utilized. Moreover, we and anesthesia pose to these unique patients.
have highlighted clinical considerations and management concerns
associated with de Barsy syndrome including airway management,
ETHICAL APPROVAL
vascular access, and monitoring for hyperpyrexia.
In this case series, 3 major concerns for de Barsy patients were This study was approved by Mayo Clinic IRB (16-007092).
identified, ie, difficult vascular access, hyperthermia, and potential
for difficult airway. With cutis laxa, thin skin and minimal subcuta-
CONFLICT OF INTEREST
neous tissue contribute not only to challenging intravenous access
but also concerns for skin fragility and breakdown.7,10 Ultrasound or Nothing to disclose.
other advanced imaging techniques may be useful in facilitating vas-
cular access in these patients.
ORCID
In 11% of our cases, intraoperative hyperthermia occurred. A
review of these cases did not identify any direct etiology for hyper- Lindsay L. Warner http://orcid.org/0000-0002-1473-7084
thermia from such causes as infection, neuroendocrine disorder,
anti-cholinergic use, or neuroleptic medication. Moreover, dantrolene
administration was not required in any case and no additional meta- REFERENCES
bolic derangements were reported. Previously published reports of
1. De Barsy AM, Moens E, Dierckx L. Nanism, oligophrenia, and hypo-
nonmalignant hyperthermia in such conditions as Costello syndrome, plasia of the elastic tissue in the skin and the cornea. Arch Belg Der-
Osteogenesis Imperfecta, and Arthrogryposis, have speculated that matol Syphiligr. 1967;23:411-412.
central temperature control dysregulation occurs during general 2. Aldave AJ, Eagle Jr RC, Streeten BW, Qi J, Raber IM. Congenital cor-
neal opacification in de Barsy syndrome. Arch Ophthalmol.
anesthesia.16-18 To date, however, no defect has been identified and
2001;119:285-288.
no direct evidence supports such a mechanism. In fact, a recent 3. Aponte EP, Smith HM, Wanek BJ, Rettke SR. Anesthesia considera-
investigation at our institution specifically examined intraoperative tions for patients with de Barsy syndrome. J Clin Anesth.
hyperthermia in 369 discrete anesthetic cases involving patients with 2010;22:499-504.
4. Dutta AK, Ekbote AV, Thomas N, Omprakash S, Danda S. De Barsy
arthrogryposis multiplex congenital and distal arthrogryposis syn-
syndrome type B presenting with cardiac and genitourinary abnor-
drome.19 In contrast to prior studies, this investigation found no
malities. Clin Dysmorphol. 2016;25:190-191.
increased risk of intraoperative hyperthermia or hypermetabolic 5. Dutta A, Ghosh SK, Ghosh A, Roy S. A 5-year journey with cutis laxa
events compared to control subjects without arthrogryposis. Similar in an Indian child: the de Barsy syndrome revisited. Indian J Dermatol.
to our intraoperative hyperthermia rate of 11%, Gleich et al reported 2016;61:81-84.
62 | WARNER ET AL.

6. Bartsocas CS, Dimitriou J, Kavadias A, Kyrzopoulos D. De Barsy syn- 14. Guerra D, Fornieri C, Bacchelli B, et al. The de Barsy syndrome.
drome. Progress Clin Biological Res. 1982;104:157-160. J Cutaneous Path. 2004;31:616-624.
7. Arazi M, Kapicioglu MI, Mutlu M. The de Barsy syndrome. Turkish J 15. Lin DS, Chang JH, Liu HL, et al. Compound heterozygous mutations
Pediatrics. 2001;43:79-84. in PYCR1 further expand the phenotypic spectrum of De Barsy syn-
8. Al-Owain M, Alanazi S, Khalifa O, et al. A case of de Barsy syndrome drome. Am J Med Genet A. 2011;155A:3095-3099.
with a severe eye phenotype. Am J Med Genet. 2012;158:2364- 16. Leao-Teles E, Quelhas D, Vilarinho L, Jaeken J. De Barsy syndrome
2366. and ATP6V0A2-CDG. Eur J Hum Genet. 2010;18:526.
9. Bicknell LS, Pitt J, Aftimos S, Ramadas R, Maw MA, Robertson SP. A 17. Stanton RP, Rao N, Scott Jr CI. Orthopaedic manifestations in de
missense mutation in ALDH18A1, encoding delta 1-pyrroline-5-car- Barsy syndrome. J Pediatr Orthop. 1994;14:60-62.
boxylate synthase (P5CS), causes an autosomal recessive neurocuta- 18. Benni F, Leoni T, Iacobucci T, D’Avolio S, de Francisci G. Anaesthesi-
neous syndrome. Eur J Hum Genet. 2008;16:1176-1186. ological considerations in Costello Syndrome. Paediatr Anaesth.
10. Hoekx J, Smeitink J, Brunner H, Monnens L. The de Barsy syndrome. 2002;12:376-378.
Tijdschrift Voor Kindergeneeskunde. 1989;57:53-57. 19. Dearlove O, Harper N. Costello syndrome. Paediatr Anaesth. 1997;7:
11. Pontz BF, Zepp F, Stoss H. Biochemical, morphological and immuno- 476-477.
logical findings in a patient with a cutis laxa-associated inborn disor-
der (de Barsy syndrome). Eur J Pediatrics. 1986;145:428-434.
12. Karnes PS, Shamban AT, Olsen DR, Fazio MJ, Falk RE. De Barsy syn-
drome: report of a case, literature review, and elastin gene expres- How to cite this article: Warner LL, Olsen DA, Smith HM.
sion studies of the skin. Am J Med Genet. 1992;42:29-34.
Clinical implications of de Barsy syndrome. Pediatr Anesth.
13. Zampatti S, Castori M, Fischer B, et al. De Barsy syndrome: a geneti-
cally heterogeneous autosomal recessive cutis laxa syndrome related 2018;28:59–62. https://doi.org/10.1111/pan.13283
to P5CS and PYCR1 dysfunction. Am J Med Genet. 2012;158:927-931.
Accepted: 24 October 2017

DOI: 10.1111/pan.13290

RESEARCH REPORT

Incidence and predictors of 30-day postoperative readmission

in children

Daniel Vo1 | David Zurakowski1 | David Faraoni2

1
Department of Anesthesiology,
Perioperative and Pain Medicine, Boston Summary
Children’s Hospital, Harvard Medical Background: Hospital readmissions are being used as a quality metric for hospital reim-
School, Boston, MA, USA
2 bursement without a clear understanding of the factors that contribute to readmission.
Department of Anesthesia and Pain
Medicine, The Hospital for Sick Children, Objective: The objective of this study was to report the incidence of 30-day post-
University of Toronto, Toronto, ON,
surgical readmission in children, identify the predictors for readmission, and create
Canada
an algorithm to identify high-risk children.
Correspondence
Methods: Data from the 2012-2014 Pediatric database of the American College of
Daniel Vo, Department of Anesthesiology,
Perioperative and Pain Medicine, Boston Surgeons National Surgical Quality Improvement Program were analyzed using uni-
Children’s Hospital, Boston, MA, USA.
variable and multivariable logistical regression analysis.
Email: [email protected]
Results: Among 182 589 children included in the 2012-2014 American College of
Section Editor: Francis Veyckemans
Surgeons National Surgical Quality Improvement Program Pediatric database, 4.8%
(8815/182 589) experienced a readmission within 30 days. Four significant predic-
tors were retained in the multivariable logistic regression model: American Society
of Anesthesiologists physical status ≥ 3 (OR: 1.9, 95% CI: 1.8-2.0), presence of
congenital heart disease (OR: 1.66, 95% CI: 1.31-2.11), inpatient status at time of
surgery (OR: 3.5, 95% CI: 3.3-3.7), and at least 1 postoperative complication (neuro-
logic, renal, wound, cardiac, bleeding, or pulmonary) (OR: 3.14, 95% CI: 2.92-3.34).
The multivariable logistic regression model showed reasonably good discrimination
in predicting 30-day readmissions with receiver operating characteristic area under
the curve of 0.747 (95% CI: 0.73-0.75) and good calibration (Brier score: 0.044). We
created a predictive algorithm of 30-day readmission based on the 4 significant
predictors.
Conclusion: Children with congenital heart disease, high American Society of Anes-
thesiologist physical class, inpatient status, and at least 1 postoperative complication
of any kind are at high risk for postsurgical readmissions. We provide an algorithm
for quantifying this risk with the goal of reducing the number of readmissions,
improving the care of patients with complex chronic illnesses, and reducing hospital
costs.

KEYWORDS
congenital heart disease, hospital patient readmission, pediatrics, risk assessment, risk
management, specialties surgical

Pediatric Anesthesia. 2018;28:63–70. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 63
64 | VO ET AL.

1 | INTRODUCTION

Hospital readmission rates are currently being used as quality indica- What’s known on this subject
tors by the Centers for Medicare & Medicaid Services (CMS). They
• Hospital readmissions are responsible for an additional
are being made public and tied to hospital reimbursements.1 As such,
$12 billion dollars in expenses to the healthcare system.
they represent a potential target for hospital administrators and poli-
• Pediatric, unplanned postsurgical readmissions are poorly
cymakers for improving the value returned from healthcare dollars.
characterized but often thought to be preventable.
Studies have estimated 9% to 48%2 of all readmissions were pre-
ventable at a cost of $12 billion annually.3 Not only do hospital
What this study adds
readmissions have economic consequences but they also place a
burden on the healthcare system, patients, and families and further • Using the 2012-2014 American College of Surgeons

fragment a patient’s disease course. Reducing readmission is a means National Surgical Quality Improvement Program Pediatric
for improving patient satisfaction and quality of life. It is unknown database, we identified preoperative, intraoperative, and

whether these metrics truly represent substandard care or variations postoperative risk factors for 30-day, unplanned readmis-
in patient populations with differing disease progressions. Such sion.

uncertainties only become more complex in pediatric populations • We provide a novel risk stratification algorithm for identi-
where the health of children is dependent on an “ecology of care,” fying pediatric patients at highest risk for postsurgical

involving the social, economic, medical, and perioperative homes of readmission.

4
these patients. While reimbursement penalties have not been
applied to pediatric hospitals yet, the Pediatric Quality Measures
Program has begun to look at readmissions at children’s hospitals demographics, preoperative characteristics, 30-day readmission, and
nationally and the Children’s Health Insurance Reauthorization Act5 mortality in both the inpatient and outpatient settings.10 A system-
foreshadows that such penalties may be in the future for pediatric atic sampling strategy is used to avoid bias in case selection and to
hospitals. ensure a diverse surgical case mix. Data are entered by a combina-
Previous studies looking at postsurgical readmissions have largely tion of automatic extraction from the patient’s existing EMR and
been restricted to specific populations such as patients with congen- data entered manually by a trained and certified Surgical Clinical
ital heart disease or specific procedures such as appendectomies7
6 Reviewer. To ensure the quality of the data collected, the ACS
8 NSQIP Pediatric conducts interrater reliability audits of selected par-
and ventriculo-peritoneal shunt revisions. These studies relied lar-
gely on internal, administrative databases intended for billing pur- ticipating sites, and sites with an interrater reliability audit disagree-
poses. This analysis of the American College of Surgeons National ment rate > 5% are excluded from the data. The audits reveal an
Surgical Quality Improvement Program Pediatric Database (ACS overall disagreement rate of approximately 2% for all assessed pro-
NSQIP Peds) is the largest retrospective database analysis looking at gram variables. Additionally, cases coming from hospitals with a 30-
hospital readmission for all pediatric surgical patients coming for all day follow-up rate < 80% were excluded. Exclusion criteria for the
types of noncardiac surgery. We looked at the incidence and predic- ACS NSQIP Pediatric database include: age > 18 years old, trauma
tors of unplanned 30-day hospital readmission in postsurgical cases, brain-dead organ donors, and solid organ transplantation. In
patients. Our objective was to identify demographic, preoperative, order to comply with the Hospital Participation Agreement between
operative, and postoperative risk factors for 30-day readmission and ACS and participating hospitals, facility identifiers as well as
to develop a predictive algorithm for assessing readmission risk geographic information were removed for use in the ACS NSQIP
based on these findings. The pediatric surgical population is growing Pediatric.
9
in volume and medical complexity. Unplanned postsurgical readmis-
sions remain a poorly characterized morbidity and emphasize that
2.2 | Variable selection
the procedure is not the primary endpoint of the perioperative per-
iod but an episode of care in the lifecycle of a disease process. We included all children undergoing pediatric surgery recorded in
the 2012-2014 ACS NSQIP Pediatric database. Out of over 150
variables, 4 demographic, 6 preoperative variables, and 6 postopera-
2 | MATERIALS AND METHODS
tive complications were selected for univariable analysis. We
selected age, gender, race, BMI, American Society of Anesthesiolo-
2.1 | Data source
gists physical status (ASA), elective vs emergent surgery, inpatient vs
This study was performed using data from the 2012, 2013, and outpatient admission status, surgical specialty (cardiothoracic, gen-
2014 ACS NSQIP Pediatric database. The ACS NSQIP Pediatric col- eral, orthopedic, ENT, plastics, and urogynecologic), congenital heart
lects deidentified data on children less than 18 years of age under- disease, and prematurity. While there are many more variables
going noncardiac surgery, and includes over 150 variables, including included in the database, many of the variables show correlation and
VO ET AL. | 65

our objective was to construct a simple model consisting of a bal- are expressed as regression coefficient (B), the odds ratio (OR) as a
ance of demographic, preoperative, intraoperative, and postoperative measure of risk, the 95% confidence interval (CI), and P values were
variables chosen for their clinical face-value and regular availability obtained from the Wald test.
before the procedures using Katz’s recommendations for multivari- Two metrics to assess model performance were used. The AUC,
able logistic regression analysis in medical research.11 Many of the also known as the c-statistic, was constructed to assess the discrimi-
variables selected have been shown in prior studies using smaller native ability of the set of multivariable predictors for 30-day read-
datasets and differing methodologies to impact morbidity, mortality, mission. In order to evaluate the internal validity of our predictive
12-16
and readmission. algorithm of 30-day readmission, we applied the bootstrapping
To standardize the univariable analysis (eg, variables are either method on the NSQIP 2012-2014 data. Bootstrap resampling draws
categorical or ordinal), children were stratified into 5 clinically rele- random samples with replacement from the entire sample (repeti-
vant age groups: <6 months, 6-12 months, 1-6 years, 6-12 years, tion = 500) and computes bias-corrected measures for model assess-
and >12 years old. Congenital heart disease was classified as mild, ment. Calibration was assessed graphically by plotting the observed
major, and severe according to an ACS NSQIP classification system. outcome against the predicted readmission rate. A smooth nonpara-
Minor CHD is defined as a lesion that does not require medical metric calibration line was created with the locally weighted scatter-
maintenance or has been repaired resulting in normal cardiovascular plot-smoothing algorithm to estimate the observed probabilities in
function. Major CHD is defined as a repaired lesion with a residual relation to the predicted probabilities.17
hemodynamic abnormality with or without medications. Severe CHD Statistical analysis was performed with IBM SPSS Statistics (ver-
is defined as uncorrected cyanotic heart disease, pulmonary hyper- sion 23.0, IBM Corporation, Armonk, NY). Bootstrapping was per-
tension, ventricular dysfunction requiring medications, or being listed formed with the lrm function in the R Design Package (R Core
for a heart transplant. Gestational age was classified as full term Team, R: A language and environment for statistical computing. R
(>37 weeks), premature (≤37 weeks but greater than 24 weeks), and Foundation for Statistical Computing, Vienna, Austria, 2013) based
extremely premature (<24 weeks). ASA class was divided into a high on a data-splitting (test and training set) approach.18
acuity (ASA ≥ 3) and low acuity group (ASA 1-2).
Postoperative complications were aggregated by system into pul-
monary complications (pneumonia, pulmonary embolism, and re-intu- 3 | RESULTS
bation), renal complications (acute renal failure, oliguria, and urinary
tract infection), wound complications (superficial and deep wound The 2012-2014 ACS NSQIP Pediatric database includes 182 749
infections and dehiscence), cardiac complications (ventricular fibrilla- procedures. There were 160 patients missing data on readmission
tion, cardiac arrest, and cardiopulmonary resuscitation), bleeding who were excluded from our analyses. Of the patients remaining,
requiring transfusion, and neurological complications (cerebral vascu- there were 77 907 female (42.7%) and 104 682 male patients
lar accident, seizures, and stroke). (57.3%). Neonates and infants under 1 year of age were the smallest
The primary outcome was unplanned readmission to the same or group (21%). All other age groups were equally represented. The
another institution within 30 days of a primary surgery with the majority of procedures were performed by General Surgery (27.8%)
cause for readmission suspected to be related to the procedure. The followed by ENT (20.3%) and Orthopedics (13.2%). The overall
patient had to be classified as an inpatient on his or her readmission. unplanned 30-day readmission rate was 4.8% (8815/182 589).
Patients missing data on readmission were excluded from analysis. Univariable analysis (Table 1) showed that inpatient status
Patients who met our readmission criteria but had missing data on (P < .001), extreme prematurity (P < .001), CHD (P < .001), higher
any of our 16 variables selected for regression were included in our ASA physical status class ≥ 3 (P < .001), and the surgical specialty
analysis as we chose to tolerate underestimates of the discriminative (P < .001) were associated with a higher incidence of 30-day read-
ability of the model. mission (Table 1). After adjusting for surgical specialties, multivariable
logistic regression analysis shown in Table 2 revealed that ASA phys-
ical status class ≥ 3 (OR: 1.9, 95% CI: 1.8-2.0), inpatient admission
2.3 | Statistical analysis
status (OR: 3.5, 95% CI: 3.3-3.7), and the presence of minor (OR:
Continuous data that were not normally distributed are expressed as 1.2, 95% CI: 1.1-1.3), major (OR: 1.2, 95% CI: 1.1-1.3), or severe
median and interquartile ranges (IQR), and categorical variables as (OR: 1.3, 95% CI: 1.2-1.7) CHD were significant and independent
number and percentage. Demographic, preoperative risk factors, and predictors of 30-day hospital readmission. There was no difference
postoperative complications were compared using Wilcoxon rank- in rates of readmission observed based on patient gender, age, or
sum test for continuous variables and the Pearson chi-square test emergent vs elective surgery.
for categorical variables and proportions. To control for possible con- Postoperative complications were aggregated into 6 categories
founding among variables, we used univariable and multivariable by system: pulmonary complications, wound complications, renal
logistic regression using backward selection to determine the inde- complications, neurological complications, cardiac complications, and
pendent predictors for 30-day readmission rates using a univariable bleeding complications requiring transfusion. The overall complica-
cutoff of P < .10 for inclusion and P > .05 for removal. The results tions rate was 9.8% (17 909/182 589). The most frequent
66 | VO ET AL.

T A B L E 1 Univariable analysis of preoperative risk factors T A B L E 2 Multivariable analysis of preoperative and postoperative
risk factors
Not readmitted Readmitted
Variable N = 173 774 (%) N = 8815 (%) P value Variable Beta coeff. OR 95% CI P value
Age ASA physical class
0-6 mo 7770 (4.5) 378 (4.3) <.001 ASA 1 or 2 Ref.
6-12 mo 28 841 (16.6) 1629 (18.5) ASA 3, 4, or 5 0.63 1.9 1.8-2.0 <.001*
1-6 y 46 383 (26.7) 2566 (29.1) CHD
6-12 y 45 709 (26.3) 1930 (21.9)
None Ref.
>12 y 45 071 (25.9) 2312 (26.2)
Mild 0.16 1.2 1.1-1.3 <.001*
ASA class
Major 0.14 1.2 1.1-1.3 .022*
ASA 1, 2 110 070 (63.3) 3549 (40.3) <.001
Severe 0.26 1.3 1.2-1.7 .011*
ASA 3, 4, 5 63 547 (36.6) 5263 (59.7)
Inpatient vs outpatient 1.25 3.5 3.3-3.7 <.001*
Gender
Bleeding complication 0.01 1.0 0.9-1.1 .864
Male 99 788 (57.4) 4894 (55.5) .04
Wound complication 2.13 8.4 7.8-9.1 <.001*
Female 73 986 (42.6) 3921 (44.5)
Pulmonary complication 0.62 1.9 1.6-2.2 <.001*
Race
Renal complication 1.79 6.0 5.2-6.9 <.001*
White 111 802 (64.3) 5592 (63.3) <.001
Neurologic complication 0.90 2.5 1.9-3.2 <.001*
Black 22 231 (12.8) 1090 (12.4)
Asian 5538 (3.2) 306 (3.5) ASA, American Society of Anesthesiologists; CHD, congenital heart dis-
ease; OR, odds ratio; CI, confidence interval.
Native American 858 (0.5) 65 (0.01)
Results are based on multivariable logistic regression, adjusted for age,
Other/Unknown 24 647 (14.2) 1087 (12.3) gender, prematurity, emergent vs elective procedure, and surgical spe-
Surgical specialty cialty.
Cardiothoracic 15 876 (9.1) 497 (5.6) <.001 *Statistically significant.

General surgery 47 132 (27.1) 3559 (40.5)

readmission to 10.3%. The frequency of a complication had no bear-
Neurosurgery 20 778 (12.0) 1567 (17.8)
ing on its impact on readmission. For example, bleeding was the
Orthopedics 23 597 (13.6) 578 (6.6)
most frequent complication, however, only 7% of patients experienc-
ENT 35 377 (20.4) 1753 (19.9)
ing a bleed required readmission. However, only 0.66% of patients
Urology/Gynecology 17 687 (10.2) 394 (4.5) experienced a renal complication, but 31.3% of those patients
Plastics 13 315 (7.7) 467 (5.3) required readmission. Figure 1 highlights the absence of relationship
Admission status between impact on readmission and frequency of a complication.
Inpatient 95 911 (55.2) 7456 (84.6) <.001 The incidence of readmission for each type of complication as well
Outpatient 77 863 (44.8) 1359 (15.4) as for patients experiencing no complications and at least 1 compli-
Urgency cation is represented as a bar graph. The number of patients experi-
Emergent 46 368 (26.7) 2607 (29.6) <.001 encing the complication is presented below the x-axis, and the
Elective 127 406 (73.3) 6208 (70.4) number of those patients readmitted is provided as “Events.”
Prematurity In Table 3, we combine the demographic variables, preoperative
Full term 133 172 (76.6) 6486 (73.6) <.001 risk factors, and postoperative complications retained in the multi-
28-38 wk 39 730 (22.9) 2240 (25.4) variable logistic regression model into a predictive algorithm. Risk of
<28 wk 872 (0.5) 89 (1.0) 30-day postoperative readmission is stratified according to the pres-
CHD ence or absence of any of the 4 variables. If a patient had a low
None 158 741 (91.3) 7450 (84.5) <.001 ASA physical status (1,2), no CHD, was an outpatient at the time of
Mild 8596 (5.0) 762 (8.6) surgery, and experienced no postoperative complications, the
Major 5281 (3.0) 474 (5.4) incidence of 30-day hospital readmission was 1.4% (95% CI: 1.3%-

Severe 1166 (0.7) 129 (1.5) 1.5%). The risk of readmission increased with the number of
variables retained in the final model. If a patient had all 4 variables
ASA class, American Society of Anesthesiology physical status; CHD,
of ASA physical status ≥ 3, CHD, inpatient status, and a postopera-
congenital heart disease.
Data are presented as number and percentages (%). tive complication, the incidence of readmission increased dramati-
cally to 20% (95% CI: 18.5%-21.0%).
complications were bleeding requiring transfusion (6.2%) and wound Validation results for our predictive model of 30-day readmission
complications (2.9%). If a patient had no complications, the absolute using the bootstrapping technique with 500 resamples indicated a
risk of 30-day readmission was reduced to 1.1%. However, the bias-corrected Somers’ D rank correlation measure of .454, corre-
occurrence of 1 or more complication more than doubled the risk of sponding to a c-statistic or area under the curve (AUC) of 0.747.
VO ET AL. | 67

4 | DISCUSSION

Our results suggest several important risk factors associated with

unplanned 30-day postoperative readmission in children undergoing
noncardiac surgery. Children with congenital heart disease, ASA
physical status class ≥ 3, inpatient status at the time of surgery, and
at least 1 postoperative complication of any system were at greatest
risk for 30-day postsurgical readmission. These findings are consis-
tent with previously known perioperative risk factors for morbidity
and mortality and help to complete a picture of the at-risk popula-
tion for unplanned pediatric hospital readmissions.19
Not surprisingly, medical complexity is a recurring theme in the
literature regarding postsurgical morbidity, mortality, and readmis-
sion. Looking at medical readmission to pediatric hospitals, Berry
et al examined 317 643 patients hospitalized and found an overall
readmission rate of 20% within 1 year. Characteristics of readmitted
F I G U R E 1 Frequency of postoperative complications and their
children were older age, public insurance use, Black race, and com-
bearing on readmission. Each postoperative complication is listed as
a bar column. Patients experiencing no complications are listed as plex chronic conditions (CCCs) particularly dependence on medical
“None” and patients experiencing 1 or more complications are devices (g-tube, ventriculo-peritoneal shunts, and tracheostomies).20
aggregated in the “At least 1” column. The number of patients Similarly, Burgonrappa et al21 looked at 2217 procedures performed
experiencing a complication are listed below each complication, and over 5 years at 2 tertiary care children’s hospitals within a single
the number of those patients readmitted are listed as “Events.” The
health care system and found an unplanned postsurgical readmission
height of the bar graph is the number of patients experiencing a
complication who were readmitted expressed as a percentage rate of 6.5%. The risk factors that they identified were patients with
CCCs and age less than 1 year.
T A B L E 3 Multivariable predictive algorithm of 30-d readmission Developed in 2000, the Pediatric Complex Chronic Conditions
among surgical patients: NSQIP 2012-2014 Classification System is a comprehensive set of ICD-10 codes and

Without With CPT codes. The working definition of a CCC is any medical condition
postoperative postoperative that can be reasonably expected to last at least 12 months (unless
ASA 3, 4, complication complication death intervenes) and to involve either several different organ sys-
or 5 CHD Inpatient (%) (%)
tems or 1 system severely enough to require specialty pediatric care
No No No 1.4 (1.3-1.5) 3.7 (3.4-4.1) and probably some period of hospitalization in a tertiary care cen-
No Yes No 1.5 (1.4-1.7) 4.2 (3.7-4.6) ter.22 Children with CCCs are thought to make up 2.9% of the pedi-
Yes No No 2.4 (2.2-2.6) 6.4 (5.9-7.0) atric patient population but account for 18.8% of admissions, 23.4%
Yes Yes No 2.7 (2.4-2.9) 7.2 (6.4-7.8) of bed-days, and 23.2% ($3.42 billion) of aggregate charges.20
No No Yes 4.3 (4.0-4.6) 11.0 (10.3-12.0) Static definitions of medical complexity or comorbidity such as
No Yes Yes 4.8 (4.3-5.2) 12.2 (11.1-13.3) the CCC classification system or ASA physical status by themselves
Yes No Yes 7.4 (6.9-7.8) 18.0 (17.0-19.1) are inadequate for assessing postoperative risk of readmission. Peri-

Yes Yes Yes 8.1 (7.5-8.7) 20.0 (18.5-21.0) operative risk is an interaction between procedure, surgeon, disease,
and patient. The medical population and the surgical population dif-
ASA, American Society of Anesthesiologists; CHD, congenital heart dis-
fer in that the majority of surgeries are elective and ideally under-
ease.
Data represent probability of 30-d readmission, with 95% confidence taken when a patient is medically optimized. Thus, it is not only the
intervals based on logistic regression. number of diagnoses or organ systems involved but also the func-
tional status of the patient that has bearing on risk of readmission.
The Nagelkerke R2 measure was .295. The intercept and slope of an Inpatient admission status indirectly captures aspects of medical
overall logistic calibration equation were 0.002 and 1.00. The maxi- complexity, clinical function, and perioperative optimization at the
mum absolute difference in predicted and calibrated probabilities, or time of surgery missed by both the CCC classification and ASA phys-
Emax, was .0006. The Brier quadratic probability score B was .044, ical class. It was in fact one of our most heavily weighted prognosti-
which is good evidence of prediction as a lower Brier score is better. cators (OR: 3.5, 95% CI: 3.3-3.7) and increased the risk of 30-day
Thus, our internal validation results suggest that our predictive postoperative readmission across all analyses.
model have good internal validity with a c-statistic approaching As an example, children with congenital heart disease are known
0.750, and a bias-corrected Somers’ D rank correlation of .454. Our have higher rates of morbidity and mortality in the immediate intra-
multivariable algorithm performs reasonably well at predicting 30- and postoperative period.23,24 Our evidence suggests this vulnerabil-
day readmission. ity extends to 30-day postsurgical readmissions. As shown in
68 | VO ET AL.

Figure 2, the rate of readmission increased with lesion severity (Mild variable significantly associated with higher readmission rates.28 This
3.8%, Major 5.1%, and Severe 6.0%). However, the impact of CHD led the authors to question whether pediatric surgical readmissions
on readmission is still amplified by admission status. Compared to are truly preventable. The dataset used in their study was an admin-
outpatients with the same CHD classification, inpatients experienced istrative dataset which likely lacked the granularity to model the
almost double the rate of readmission (Mild 6.9%, Major 9.0%, and depth of preoperative and postoperative variables that the ACS
Severe 11.1%). Our interpretation of this finding is that the ACS NSQIP Peds offers, leading to a paucity of identified risk factors and
NSQIP Peds categories of mild, moderate, or severe CHD are static counterintuitive findings.
definitions that fail to capture the variety of CHD lesions and an Modifying postoperative complications is an ongoing challenge and
individual’s performance. However, in-hospital admission declares the subject of much research. Evidence for many proposed interven-
the functional status of a patient with CHD irrespective of disease tions is lacking and beyond the scope of this manuscript. However,
classification. Another possible explanation for why CHD by itself while many individualized interventions have failed to show clear bene-
was not more impactful on readmission is that often this population fit, the readmission risk model we propose in Table 3 can help identify
is cared for by separate cardiac anesthesiologists and cardiac inten- “red flags” early in the postoperative course which could triage at-risk
sive care units postoperatively which may make them perform more patients to more intensive discharge planning (eg, dedicated discharge
like the general population. teams to ensure stability of therapy, assessment of home supports sys-
Of the 4 variables identified, postoperative complications may be tems, care transition coaching, or postdischarge telephonic monitoring).
most actionable. Studies using the ACS NSQIP data in the adult pop- For example as seen in Figure 2, renal complications and wound infec-
ulation have consistently found postoperative complications to be a tions while having varying incidence, both increase the risk of readmis-
driver of hospital readmission.25,26 Tahiri et al used the ACS NSQIP sion above 30%. Currently, there is a lack of evidence regarding
Peds to look at 30-day readmission in 5367 plastic surgery patients. intraoperative reno-protective or “rescue” strategies including fluid
They found a lower overall readmission rate of 2.4% and the most management or the use of vasoactive agents, diuretics, or sodium
significant independent predicators of readmission were postopera- bicarbonate.29 However, Goldstein et al describe using automated
27
tive medical and surgical complications. Other risk factors identi- EMR screening and a decision support process to identify patients
fied were medical comorbidities, contaminated wound classification, exposed to nephrotoxic drugs and at risk for acute kidney injury leading
higher ASA class, longer operative times, and longer length of stay. to daily serum creatinine monitoring decreasing both the exposure rate
In a preliminary analysis, we examined relative value units assigned and acute renal insufficiency rate by 64%.30 Similarly with wound
to procedures as a surrogate for procedural length and complexity infections, while advances have been made in the operating room ven-
and did not find an association with increased readmissions. The dif- tilation, sterilization, barriers, surgical technique, and best practices for
ference in overall readmission rate and readmission risk factors perioperative antibiotics,31 the key to preventing readmissions may lie
found in our 2 studies likely represents a different risk profile for in identifying the at-risk population and a team approach to plans of
pediatric plastic surgery patients and the general cohort. In their care and community-based postdischarge surveillance.32-34
study of pediatric surgical readmission, Brown et al failed to find a This study has several limitations. Retrospective database analy-
link between postoperative complications or length of stay and read- ses such as ours are inherently limited by the variables chosen to be
mission. In an analysis of over 260 000 pediatric cases performed included in the database. Data may have been missing or incorrectly
nationwide, being a high volume surgical center was the only entered. Hospital data in the ACS NSQIP Peds are deidentified so it
is unknown where and in what type of facility these surgeries were
performed. Geographic location has been shown to impact rates of
readmission.35 Surgical outcomes have been found to vary with vol-
ume and a hospital’s designation as “stand-alone children’s hospi-
tal.”36 As with any database, this sample may not be representative
of the national population of pediatric surgical patients. Despite
these limitations, the ACS NSQIP Peds represents a large dataset
which is case mix and risk adjusted from a variety of hospital set-
tings including but not limited to stand-alone children’s hospitals.
The database has rigorous quality control metrics and specifically
examines the surgical population using clinical data extracted from
the medical record as opposed to administrative/billing data.

5 | CONCLUSION
F I G U R E 2 Thirty-day readmission rates for patients with mild,
moderate, severe congenital heart disease (CHD) stratified according
to inpatient or outpatient status. Error bars represent 95% Children who have high ASA status, congenital heart disease, expe-
confidence intervals rience a postoperative complication, and are inpatients at the time
VO ET AL. | 69

of surgery are at increased risk for 30-day postsurgical readmission. 10. Anon. User Guide for the 2014 ACS NSQIP Pediatric Participant
We present a simplified prognostic model consisting of these 4 Use Data File. 2014. http://www.pediatric.acsnsqip.org/acsNsqipDa
taPed/jsp/pdf/PEDS.ACS.NSQIP.PUF.UserGuide.2014.pdf. Accessed
variables which shows good predictive discrimination and internal
November 10, 2015.
calibration using the bootstrapping method. The next steps are to 11. Katz M. Multivariable analysis: a primer for readers of medical
validate these findings using the 2015, 2016 ACS NSQIP Pediatric research. Ann Intern Med. 2003;138:644-650.
datasets. Preventing hospital readmission may prove to be a combi- 12. Faraoni D, Zurakowski D, Vo D, et al. Post-operative outcomes in
children with and without congenital heart disease undergoing non-
nation of improvements in perioperative care and the early identifi-
cardiac surgery. J Am Coll Cardiol. 2016;67:793-801.
cation of patients at risk for complications, readmission, and 13. Ramamoorthy C, Haberkern CM, Bhananker SM, et al. Anesthesia-
hospital-resource consumption. Further research is required to related cardiac arrest in children with heart disease: data from the
determine whether triaging these patients to separate predischarge Pediatric Perioperative Cardiac Arrest (POCA) registry. Anest Analg.
2010;110:1376-1382.
care and postdischarge surveillance reduces unintended postopera-
14. Berry J, Toomey S, Zaslavsky A, et al. Pediatric readmission preva-
tive readmissions. lence and variability across hospitals. JAMA. 2013;309:372-380.
15. Tsai T, Joynt K, Orav E, et al. Variation in surgical-readmission rates
and quality of hospital care. N Engl J Med. 2013;369:1134-1142.
ACKNOWLEDGMENTS 16. Davenport D, Bowe E, Henderson W, et al. National Surgical Qual-
ity Improvement Program (NSQIP) risk factors can be used to vali-
The authors thank the American College of Surgeons National Surgi-
date American Society of Anesthesiologists Physical Status
cal Quality Improvement Program and the hospitals participating in classification (ASA PS) levels. Ann Surg. 2006;243:636-641. discus-
the ACS NSQIP, source of the data used herein. ACS NSQIP has not sion 641–4.
verified and is not responsible for the statistical validity of the data 17. Le Manach Y, Collins G, Rodseth R, et al. Preoperative score to pre-
dict postoperative mortality (POSPOM): derivation and validation.
analysis or the conclusions derived by the authors.
Anesthesiology. 2016;124:570-579.
18. Miller M, Hui S, Tierney W. Validation techniques for logistic regres-
sion models. Stat Med. 1991;10:1213-1226.
CONFLICT OF INTEREST 19. Nakamura M, Toomey S, Zaslavsky A, et al. Measuring pediatric hos-
pital readmission rates to drive quality improvement. Acad Pediatr.
The authors report no conflict of interest.
2016;14(5 suppl):S39-S246.
20. Berry J, Hall D, Kuo D, et al. Hospital utilization and characteristics
of patients experiencing recurrent readmissions within children’s
ORCID
hospitals. JAMA. 2011;305:682-690.
21. Burjonrappa S, Theodorus A, Shah A, et al. Pediatric surgery
Daniel Vo http://orcid.org/0000-0001-9176-9875
readmissions: a root cause analysis. Pediatr Surg Int. 2015;31:
551-555.
22. Simon T, Berry J, Feudtner C, et al. Children with complex chronic
REFERENCES conditions in inpatient hospital settings in the United States. Pedi-
atrics. 2010;126:647-655.
1. Kansagara D, Englander H, Salanitro A, et al. Risk prediction models 23. Baum V, Barton D, Gutgesell H. Influence of congenital heart disease
for hospital readmission: a systematic review. JAMA. on mortality after noncardiac surgery in hospitalized children. Pedi-
2011;306:1688-1698. atrics. 2000;105:332-335.
2. Benbassat J, Taragin M. Hospital readmissions as a measure of qual- 24. Islam S, Yasui Y, Kaul P, Mackie A. Hospital readmission of patients
ity of health care: advantages and limitations. Arch Intern Med. with congenital heart disease in Canada. Can J Cardiol. 2016;32:987.
2000;160:1074-1081. 25. Kassin M, Owen R, Perez S, et al. Risk factors for 30-day hospital
3. Axon RN, Williams MV. Hospital readmission as an accountability readmission among general surgery patients. J Am Coll Surg.
measure. JAMA. 2011;305:504-505. 2012;215:322-330.
4. Feudtner C, Levin J, Srivastava R, et al. How well can hospital read- 26. Glance L, Kellermann A, Osler T, et al. Hospital readmission after
mission be predicted in a cohort of hospitalized children? A retro- noncardiac surgery: the role of major complications. JAMA Surgery.
spective, multicenter study. Pediatrics. 2009;123:286-293. 2014;149:439-445.
5. Dougherty D, Schiff J, Mangione-Smith R. The Children’s Health 27. Tahiri Y, Fischer J, Wink J, et al. Analysis of risk factors associated
Insurance Program Reauthorization Act quality measures initiatives: with 30-day readmissions following pediatric plastic surgery: a
moving forward to improve measurement, care, and child and adoles- review of 5376 procedures. Plast Reconstr Surg. 2015;135:521-
cent outcomes. Acad Pediatr. 2011;11:S1-S10. 529.
6. Mackie A, Ionescu-Ittu R, Pilote L, et al. Hospital readmissions in 28. Brown E, Anderson J, Burgess D, et al. Pediatric surgical readmis-
children with congenital heart disease: a population-based study. Am sions: are they truly preventable? J Pediatr Surg. 2017;52:161-165.
Heart J. 2008;155:577-584. 29. Sear J. Kidney dysfunction in the postoperative period. Br J Anaesth.
7. Newman K, Ponsky T, Kittle K, et al. Appendicitis 2000: variability in 2005;95:20-32.
practice, outcomes, and resource utilization at thirty pediatric hospi- 30. Goldstein S, Mottes T, Simpson K, et al. A sustained quality
tals. J Pediatr Surg. 2003;38:372-379. improvement program reduces nephrotoxic medication-associated
8. Tejerdor-Sojo J, Singleton L, McCormick K, et al. Preventability of acute kidney injury. Kidney Int. 2016;90:212-221.
pediatric 30-day readmissions following ventricular shunt surgery. J 31. Centers for Disease Control and Prevention (CDC). Surgical Site
Peds. 2015;167:1327-1333. Infection (SSI) Event: Procedure-associated Module. 2016. www.cdc.
9. Burns K, Casey P, Lyle R, et al. Increasing prevalence of medically gov/nhsn/pdfs/pscmanual/9pscssicurrent.pdf. Accessed July 19,
complex children in US hospitals. Pediatrics. 2010;126:638-646. 2017.
70 | VO ET AL.

32. McNaughton V, Orsted H. Surgical site infections in community care 36. Pasquali S, Gaies M, Jacobs J, et al. Centre variation in cost and out-
clients. Early detection and rational care through recognition of cli- comes for congenital heart surgery. Cardiol Young. 2012;22:796-799.
ent-specific risk factors. Wound Care Canada. 2005;3:10-13.
33. Huenger F, Schmachtenberg A, Haefner H, et al. Evaluation of post-
discharge surveillance of surgical site infections after total hip and
knee arthroplasty. AJIC. 2005;33:455-462. How to cite this article: Vo D, Zurakowski D, Faraoni D.
34. Kripalani S, Theobald C, Anctil B, et al. Reducing hospital readmis- Incidence and predictors of 30-day postoperative readmission
sion rates: current strategies and future directions. Annu Rev Med.
in children. Pediatr Anesth. 2018;28:63–70.
2014;2014:471-485.
35. Anderson G, Steinberg E. Hospital readmissions in the medicare pop- https://doi.org/10.1111/pan.13290
ulation. N Engl J Med. 1984;311:1349-1353.
Accepted: 18 October 2017

DOI: 10.1111/pan.13292

CASE REPORT

Nitroglycerin: An unusual solution to intraoperative

hypothermia in a 4-year-old burn patient

Eduardo J. Goenaga Dıaz1 | Jeffrey E. Carter2 | T. Wesley Templeton1

1
Department of Anesthesiology, Wake
Forest School of Medicine, Winston-Salem, Summary
NC, USA We describe the case of a 4-year-old child undergoing extensive burn surgery with
2
Department of Trauma Surgery, Wake
refractory intraoperative hypothermia. A low-dose nitroglycerin infusion was initi-
Forest School of Medicine, Winston-Salem,
NC, USA ated to reverse vasoconstriction and improve heat absorption, after which the
child’s temperature steadily improved. In hypothermic burn patients, topical vaso-
Correspondence
Eduardo J. Goenaga Dıaz, Department of constrictors may hinder surface warming efforts. A vasodilator infusion may aid in
Anesthesiology, Wake Forest School of
warming the pediatric patient undergoing extensive excision and grafting.
Medicine, Winston-Salem, NC, USA.
Email: [email protected]
KEYWORDS
Funding information burns, hypothermia, nitroglycerin, pediatric, surgery, vasodilation
The study was funded by departmental
resources.

Section Editor: David Polaner

1 | INTRODUCTION Franklin Lakes, NJ, USA) fluid warming system. Anesthesia was main-
tained with isoflurane, 0.7%-1.2%, rocuronium, and hydromorphone.
Burn patients are at risk for intraoperative hypothermia as a result One hour elapsed from OR arrival to draping of the patient for sur-
of increased heat loss through injured skin, increased evaporative gery. Excision of full-thickness burns was performed using a combi-
heat loss, and loss of normal thermal homeostatic mechanisms dur- nation of a Weck blade and the Versajet IIâ hydrosurgery system
1
ing general anesthesia. Pediatric patients are even more susceptible (Smith & Nephew, Andover, MA, USA).
1
due to their increased body surface area to volume ratio. In adults, All burns were excised and covered with TelfaTM nonadherent
perioperative hypothermia has been associated with acute lung dressings (Covidien, Dublin, Ireland) soaked in a 1:10 000 epinephrine
injury, coagulopathy, and impairments in the immune response and solution for hemostasis before grafting with allograft (AlloDerm, Life-
wound healing.2,3 Aggressive warming and maintenance of normoth- Cell Corp-Allergan, Dublin, Ireland). During debridement, the patient’s
ermia are essential. temperature by Foley probe dropped from 37°C to 35.7°C despite
warming of all fluids to 40°C, maintenance of the room temperature
at 103°F, and continuous forced air warming of all surfaces not
2 | CASE actively being debrided (Bair HuggerTM, 3M, St. Paul, MN)(Figure 1). At
this point, the operation was paused to allow the patient to rewarm.
A 4-year-old, 16.8 kg child with 49% total body surface area full- He was completely covered with a sterile forced air warming blanket
thickness burns sustained in a house fire 5 days prior presented to and wrapped in plastic for approximately 45 minutes. Despite these
the operating room for initial excision and grafting. On arrival to the measures, we observed no change in the patient’s core temperature.
operating room, his temperature was 37.7°C, saturating 97% on A low-dose nitroglycerin infusion was begun at 0.2 mcgkg 1
min 1

ambient air. He underwent an IV induction with propofol and to promote venodilation and improve peripheral perfusion. No
rocuronium and was easily intubated via direct laryngoscopy. IV change was made to the maintenance anesthetic. Over the next
catheters (16 Gauge) were placed in his right saphenous and right 30 minutes, a gradual decrease in blood pressure was observed as
external jugular veins, and a 3 Fr radial arterial catheter was placed. the temperature increased from 35.7°C to 36.3°C. At this point, the
Temperature was measured with a Foley catheter probe. All IV fluid operation was resumed with continued improvement in the patient’s
solutions were warmed using the enFlowâ (Becton Dikinson and Co, temperature. The nitroglycerin infusion was stopped when the

Pediatric Anesthesia. 2018;28:71–72. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 71
72 | GOENAGA DIAZ ET AL.

4 | CONCLUSION

We present a case of refractory intraoperative hypothermia in a

pediatric burn patient who responded to vasodilation with low-dose
nitroglycerin. In this setting, great vigilance must be maintained to
appropriately manage the patient’s volume status as these patients
remain at high risk for hypovolemia and circulatory shock.

ETHICAL APPROVAL

Received written, signed consent for publication from patient’s mother.

CONFLICT OF INTEREST

No conflicts of interest.
F I G U R E 1 Patient’s intraoperative temperature and mean arterial
pressure (MAP). Lowest temperature is reached while surgery is
paused for warming (green arrow), and begins to rise during
nitroglycerin infusion (shaded box), while MAP gradually decreases
Learning Points

patient’s temperature reached 37°C, approximately 70 minutes after • Children undergoing burn excision surgery are at high

initiation. The operation was completed with all surfaces grafted. risk for perioperative hypothermia.

Postoperatively, he remained intubated and was taken to the pedi- • Topical vasoconstrictors may hinder heat transfer to

atric intensive care unit. He was extubated without event on the deeper tissues from surface warming techniques.

second postoperative day. • Low-dose nitroglycerin may result in vasodilation and

improved surface heat absorption and distribution.

3 | DISCUSSION

Despite aggressive warming efforts, this patient suffered signifi- ORCID

cant ambient heat loss during anesthesia induction, preparation,
Eduardo J. Goenaga Dıaz http://orcid.org/0000-0002-3479-7180
and surgical excision of his burn wounds. Conventional warming
methods proved insufficient to maintain and eventually reestablish
normothermia. Given the total area excised and exposed to epi-
REFERENCES
nephrine-soaked dressings, we considered the possibility that poor
skin and surface tissue perfusion was impeding the transfer of sur- 1. Fuzaylov G, Fidkowski CW. Anesthetic considerations for major burn
face heat to the patient’s core. Once the nitroglycerin infusion was injury in pediatric patients. Pediatr Anesth. 2009;19:202-211.
2. Oda J, Kasai K, Noborio M, Ueyama M, Yukioka T. Hypothermia dur-
begun, vasodilation and improved perfusion of these exposed
ing burn surgery and postoperative acute lung injury in extensively
areas allowed for the absorption and transfer of more heat from burned patients. J Trauma. 2009;66:1525-1529.
the skin to deeper tissues throughout the body, increasing the 3. Rizzo JA, Rowan MP, Driscoll IR, Chan RK, Chung KK. Perioperative
patient’s core temperature. temperature management during burn care. J Burn Care Res. 2017;38:
It is possible that the recovery of the patient’s temperature could e277-e283.

potentially have been due to other factors. Volatile anesthetics can

cause peripheral vasodilation as well, but in this patient, the endtidal
concentration of isoflurane remained essentially unchanged through- How to cite this article: Goenaga Dıaz EJ, Carter JE,
out. Also, there was temporal correlation between the nitroglycerin Templeton TW. Nitroglycerin: An unusual solution to
infusion, the rise in temperature, and the modest drop in mean arte- intraoperative hypothermia in a 4-year-old burn patient. Pediatr
rial pressure which would be expected in a child receiving a low- Anesth. 2018;28:71-72. https://doi.org/10.1111/pan.13292
dose nitroglycerin infusion (Figure 1).
DOI: 10.1111/pan.13287

CORRESPONDENCE

In response to “Interventions designed using quality

improvement methods reduce the incidence of serious airway
events and airway cardiac arrests during pediatric anesthesia”
Sir—As an anesthesia resident I read the article “Interventions Also, though they were observed to be immediately available
designed using quality improvement methods reduce the incidence more frequently following the intervention, did the use of succinyl-
of serious airway events and airway cardiac arrests during pediatric choline and atropine actually increase?
anesthesia” by James Spaeth et al from Cincinnati Children’s Hospi- Finally, and this may be difficult to quantify, what role do you
tal with great interest. believe having compensation tied to success of the quality improve-
While the article was a report on the successful implementation ment measures played in achieving the positive outcome?
of a quality improvement method, and I commend your group on
reducing these serious events, I am interested in the data surround-
ETHICAL APPROVAL
ing the decreased number of airway events and airway cardiac
arrests and have some questions regarding the methodology. First, No ethics approvals were necessary for this article.
how did you define whether an event was airway related or not?
Was a particular time frame used or was it classified based on CONFLICT OF INTEREST
the report of the provider involved? Also, did the rate of events
directly associated with airway management (direct laryngoscopy, The authors report no conflict of interest.
mask ventilation, etc.) in particular decrease with the described
interventions?
ORCID
What was the timing and location of occurrence of the serious
airway events and airway-related cardiac events, especially as Benjamin H. Cloyd http://orcid.org/0000-0003-3495-5190
regards the institutional preintervention practice for adenotonsillec-
tomy patients of extubation and airway management by post opera-
tive care unit nurses?1 These patients are at great risk of both Benjamin H. Cloyd

serious airway and cardiac events, and the presence of a trained Department of Anesthesiology, University of Michigan, Ann Arbor, MI,

anesthesia provider is required during their emergence, whether in USA

the operating room or post operative care unit at the Mott Chil- Email: [email protected]

dren’s Hospital at the University of Michigan.

The implementation of the use of neuromuscular blocking
agents in all cases longer than 30 minutes for patients 2 years of REFERENCES
age and under sparked significant discussion regarding clinical prac-
1. Spaeth JP, Kreeger R, Varughese AM, Wittkugel E. Interventions
tice here at the Mott Children’s Hospital. I was surprised to find
designed using quality improvement methods reduce the incidence of
that the incidence of residual neuromuscular blockade did not
serious airway events and airway cardiac arrests during pediatric
change during the study period, despite the significantly increased anesthesia. Pediatr Anesth. 2016;26:164-172.
use of paralytics. The available evidence in both pediatric and adult 2. Naguib M, Kopman A, Ensor J. Neuromuscular monitoring and post-
patients suggests that increased use of paralytics is associated with operative residual curarisation: a meta-analysis. Br J Anaesth.
2007;98:302-316.
increased residual neuromuscular blockade, approximately 40% of
3. Ledowski T, O’Dea B, Meyerkort L, Hegarty M, von Ungern-Stern-
the time in adults and 28% of the time in children.2,3 What was berg BS. Postoperative residual neuromuscular paralysis at an Aus-
your institution’s rate of residual neuromuscular blockade and how tralian tertiary children’s hospital. Anesthesiol Res Pract. 2015;2015:
did you assess it? 1-4.

Pediatric Anesthesia. 2018;28:73–76. wileyonlinelibrary.com/journal/pan © 2017 John Wiley & Sons Ltd | 73
74 | CORRESPONDENCE

DOI: 10.1111/pan.13277

Continuous Erector Spinae Plane block for thoracic surgery in

a pediatric patient

Sir—Pediatric regional anesthesia has been gaining popularity since physiotherapy, SpO2 ranging from 80% to 90% (considered ade-
the introduction of ultrasound techniques,1 adding precision and quate for his cardiac pathology).
new blocks. Open thoracic surgery is considered a very painful procedure,
One newly described ultrasound-guided block in the adult popu- and thoracic epidural analgesia has been widely used with excellent
lation is the Erector Spinae Plane (ESP) block.2 results. In this patient, an alternative to epidural analgesia was
It is a facial plane block where local anesthetic is injected sought, because of the unknown effects of his chronic heart condi-
between the erector spinae muscle and the underlying transverse tion on epidural vessels. We believe, ESP block will become a useful
process, and was initially described at the T5 level for costal pain tool in regional anesthesia, potentially replacing epidural analgesia in
2
and thoracic surgery. many conditions. Obviously, prospective randomized controlled trials
There is a recent report of a single shot ESP block on a 7-year- comparing ESP block with current standard analgesic techniques are
old child.3 We would like to report the use of a continuous ESP block needed.
in a 3-year-old, 13 kg male patient for postoperative pain after open
thoracotomy.
The patient was a full-term newborn, with a hypoplastic left
heart syndrome. He already had several surgeries for his cardiac
defect. A postsurgical diaphragmatic paralysis after a Fontan proce-
dure was diagnosed and surgical treatment was indicated. A thora-
coscopic diaphragm plication attempt was unsuccessful and open
thoracotomy surgery was then scheduled. Balanced general anes-
thesia was given. Standard monitoring was placed and a radial arte-
rial line was inserted. Sevoflurane, fentanyl boluses, and
neuromuscular blocker were used. At the end of surgery, at the
surgical incision level (T9), an ESP block was done with the patient
still lying on his left side. A 6-13 MHz linear probe was used
(SonoSite Edge; SonoSite, Inc., Bothell, WA, USA). After skin clean-
ing, using a right sagittal paramedian probe orientation, transverse
processes were identified. A 5 cm 20G Tuohy needle was inserted
in-plane, caudal-cranial direction until the tip of the transverse pro-
cess was contacted (Figure 1). After negative aspiration, 8 mL of
0.25% levo-bupivacaine (Chirocaineâ 5 mg/mL; Abbvie, Elverum,
Norway) was injected. Immediately after injection, a 22G epidural
catheter was threaded 2 cm beyond the needle, fixed to the skin
and covered with a transparent dressing. The patient was extu-
bated and transferred to the pediatric intensive care unit. At PICU
arrival, the patient was awake, breathing smoothly, stable vital
signs, and 0/10 point CHIPPS4 score was measured. Two hours
later, a PCA pump with 0.1% levo-bupivacaine 3 mL/h continuous
infusion was started. A 1.5 mL rescue bolus, 30-minute lockout,
was programmed (to be used by a nurse or parent caring for the
child). Scheduled oral acetaminophen and intravenous ketorolac
were prescribed. On the 4th postoperative day, after chest tubes
were removed, the local anesthetic infusion was stopped and the
catheter was removed. During the postoperative period, only twice
F I G U R E 1 Patients’ position and US probe (upper); actual US
was a PCA bolus given. No other rescue medication was needed. image during block (middle), and anatomical sketch to understand US
The child never complained about pain and easily allowed chest image
CORRESPONDENCE | 75

We hope this report will encourage others to follow on the use 2

Department of Pediatric Surgery, Pontificia Universidad Catolica de
and study of this block in the pediatric population. Chile, Santiago, Chile
Email: [email protected]

CONFLICT OF INTEREST

Consent for report was sign by a parent. No funding involved.

REFERENCES
Authors have no conflict of interest to declare.
1. Polaner DM, Taenzer AH, Walker BJ, et al. Pediatric Regional Anes-
thesia Network (PRAN): a multi-institutional study of the use and inci-
ORCID dence of complications of pediatric regional anesthesia. Anesth Analg.
2012;115:1353-1364.
Juan Carlos De la Cuadra-Fontaine http://orcid.org/0000-0002- 2. Forero M, Adhikary SD, Lopez H, Tsui C, Chin KJ. The erector spinae
3868-0764 plane block: a novel analgesic technique in thoracic neuropathic pain.
Reg Anesth Pain Med. 2016;41:621-627.
3. Mun ~oz F, Cubillos J, Bonilla AJ, Chin KJ. Erector spinae plane block
for postoperative analgesia in pediatric oncological thoracic surgery.
Juan Carlos De la Cuadra-Fontaine1
Can J Anaesth. 2017;64:880-882.
Mario Concha1 4. Bu€ttner W, Finke W. Analysis of behavioural and physiological param-
Fernando Vuletin2 eters for the assessment of postoperative analgesic demand in new-
Hernan Arancibia1 borns, infants and young children: a comprehensive report on seven
consecutive studies. Paediatr Anaesth. 2000;10:303-318.
1
lica de Chile,
Department of Anesthesia, Pontificia Universidad Cato
Santiago, Chile

DOI: 10.1111/pan.13291

Mispositioning the end of a cuff inflating line in long-axis

ultrasound imaging of the pediatric larynx and trachea
Dear Editor, Postoperative investigation revealed that the end of the cuff
When using pediatric cuffed endotracheal tubes (ETTs), cuff position inflating line was positioned 4 mm from the tip of the ETT, and
should be confirmed using safe anesthesia practice.1-3 Placing the the longitudinal cuff diameter was about 12 mm (Figure 1C). The
ETT tip in the optimal position may be difficult because of varying cephalad edge of the cuff appeared to be positioned approximately
1 4 2
tracheal lengths. Long-axis, in addition to short-axis, ultrasound 2 mm from the caudal edge of the cricoid cartilage, judging by the
imaging of saline-inflated cuffs may help to confirm correct position- distance between the end of the cuff inflating line and the cuff
ing of pediatric ETTs. In this letter, we describe long-axis ultrasound cephalad edge at an intubation depth of 12 cm (Figure 1A). The
imaging features of a saline-infiltrated cuff inflating line that was ETT, however, appeared to be positioned too shallow at an intuba-
mispositioned in the trachea or larynx. Informed written consent was tion depth of 11 cm (Figure 1B). Misunderstanding the structure of
obtained from the patient’s parent for publication of anonymized the ETT could have resulted in intralaryngeal malpositioning of the
case details. ETT.
A 1-year-old boy (height, 66 cm; weight, 7.5 kg) was scheduled The midtracheal level is helpful in determining the optimal inser-
for bilateral laparoscopic, total extraperitoneal inguinal hernia repair. tion depth of the ETT in pediatric patients.1 The positions of the tra-
â
A 3.5 mm internal diameter cuffed ETT (Mallinckrodt , Dublin, Ire- chea, cricoid cartilage, thyroid cartilage, and the cuff inflating line
land) was successfully intubated to an oral depth of 12 cm from the may be better visualized in the long-axis than the short-axis view.4
upper incisors. A long-axis view of the saline-infiltrated cuff inflating Long-axis ultrasound images of the saline-inflated cuff in this case
line (Figure 1A) was obtained using an ultrasound apparatus could not be observed, probably because of hyperechoic shadows
â
equipped with an L12-2 MHz probe (L441, Noblus ; Hitachi Aloka created by cartilage formation in the perichondrium.5 However, we
Medical Ltd., Tokyo, Japan). We mistakenly assumed that the orifice were able to obtain images of the cuff inflating line where hypere-
of the cuff inflating line opened and ended in the middle of the cuff: choic shadows did not exist. Thus, long-axis ultrasound imaging of a
it was believed that the cuff was positioned too deep and, accord- saline-infiltrated cuff inflating line may facilitate correct positioning of
ingly, the depth of the ETT was adjusted to 11 cm (Figure 1B). The a pediatric ETT cuff below the caudal edge of the cricoid cartilage.
operation concluded uneventfully and the patient exhibited no Information regarding the position of the end of the cuff inflating line
hoarseness. is important for correct positioning. Many cuffed ETT designs are
CORRESPONDENCE | 75

We hope this report will encourage others to follow on the use 2

Department of Pediatric Surgery, Pontificia Universidad Catolica de
and study of this block in the pediatric population. Chile, Santiago, Chile
Email: [email protected]

CONFLICT OF INTEREST

Consent for report was sign by a parent. No funding involved.

REFERENCES
Authors have no conflict of interest to declare.
1. Polaner DM, Taenzer AH, Walker BJ, et al. Pediatric Regional Anes-
thesia Network (PRAN): a multi-institutional study of the use and inci-
ORCID dence of complications of pediatric regional anesthesia. Anesth Analg.
2012;115:1353-1364.
Juan Carlos De la Cuadra-Fontaine http://orcid.org/0000-0002- 2. Forero M, Adhikary SD, Lopez H, Tsui C, Chin KJ. The erector spinae
3868-0764 plane block: a novel analgesic technique in thoracic neuropathic pain.
Reg Anesth Pain Med. 2016;41:621-627.
3. Mun ~oz F, Cubillos J, Bonilla AJ, Chin KJ. Erector spinae plane block
for postoperative analgesia in pediatric oncological thoracic surgery.
Juan Carlos De la Cuadra-Fontaine1
Can J Anaesth. 2017;64:880-882.
Mario Concha1 4. Bu€ttner W, Finke W. Analysis of behavioural and physiological param-
Fernando Vuletin2 eters for the assessment of postoperative analgesic demand in new-
Hernan Arancibia1 borns, infants and young children: a comprehensive report on seven
consecutive studies. Paediatr Anaesth. 2000;10:303-318.
1
lica de Chile,
Department of Anesthesia, Pontificia Universidad Cato
Santiago, Chile

DOI: 10.1111/pan.13291

Mispositioning the end of a cuff inflating line in long-axis

ultrasound imaging of the pediatric larynx and trachea
Dear Editor, Postoperative investigation revealed that the end of the cuff
When using pediatric cuffed endotracheal tubes (ETTs), cuff position inflating line was positioned 4 mm from the tip of the ETT, and
should be confirmed using safe anesthesia practice.1-3 Placing the the longitudinal cuff diameter was about 12 mm (Figure 1C). The
ETT tip in the optimal position may be difficult because of varying cephalad edge of the cuff appeared to be positioned approximately
1 4 2
tracheal lengths. Long-axis, in addition to short-axis, ultrasound 2 mm from the caudal edge of the cricoid cartilage, judging by the
imaging of saline-inflated cuffs may help to confirm correct position- distance between the end of the cuff inflating line and the cuff
ing of pediatric ETTs. In this letter, we describe long-axis ultrasound cephalad edge at an intubation depth of 12 cm (Figure 1A). The
imaging features of a saline-infiltrated cuff inflating line that was ETT, however, appeared to be positioned too shallow at an intuba-
mispositioned in the trachea or larynx. Informed written consent was tion depth of 11 cm (Figure 1B). Misunderstanding the structure of
obtained from the patient’s parent for publication of anonymized the ETT could have resulted in intralaryngeal malpositioning of the
case details. ETT.
A 1-year-old boy (height, 66 cm; weight, 7.5 kg) was scheduled The midtracheal level is helpful in determining the optimal inser-
for bilateral laparoscopic, total extraperitoneal inguinal hernia repair. tion depth of the ETT in pediatric patients.1 The positions of the tra-
â
A 3.5 mm internal diameter cuffed ETT (Mallinckrodt , Dublin, Ire- chea, cricoid cartilage, thyroid cartilage, and the cuff inflating line
land) was successfully intubated to an oral depth of 12 cm from the may be better visualized in the long-axis than the short-axis view.4
upper incisors. A long-axis view of the saline-infiltrated cuff inflating Long-axis ultrasound images of the saline-inflated cuff in this case
line (Figure 1A) was obtained using an ultrasound apparatus could not be observed, probably because of hyperechoic shadows
â
equipped with an L12-2 MHz probe (L441, Noblus ; Hitachi Aloka created by cartilage formation in the perichondrium.5 However, we
Medical Ltd., Tokyo, Japan). We mistakenly assumed that the orifice were able to obtain images of the cuff inflating line where hypere-
of the cuff inflating line opened and ended in the middle of the cuff: choic shadows did not exist. Thus, long-axis ultrasound imaging of a
it was believed that the cuff was positioned too deep and, accord- saline-infiltrated cuff inflating line may facilitate correct positioning of
ingly, the depth of the ETT was adjusted to 11 cm (Figure 1B). The a pediatric ETT cuff below the caudal edge of the cricoid cartilage.
operation concluded uneventfully and the patient exhibited no Information regarding the position of the end of the cuff inflating line
hoarseness. is important for correct positioning. Many cuffed ETT designs are
76 | CORRESPONDENCE

(A)

F I G U R E 1 Long-axis ultrasound images

(B)
of the larynx and trachea, and a 3.5 mm
internal diameter cuffed Mallinckrodtâ
endotracheal tube. A, A long-axis
ultrasound image of the cuff inflating line
at a depth of 12 cm. B, The end of the
cuff inflating line moved as the ETT was
withdrawn to a depth of 11 cm. C, The
end of the cuff inflating line positioned
4 mm from the tip of the cuffed ETT; the
cuff inflating line opened 17 mm from the
(C)
tip

poorly suited for use in small children. Both manufacturers and REFERENCES
clinicians must consider the potential dangers of these designs.
1. Song IK, Kim SH, Ryu J, et al. Prediction of the midtracheal level
based on external anatomical landmarks: implication of the optimal
insertion depth of endotracheal tubes in pediatric patients. Pediatr
CONFLICT OF INTEREST
Anesth. 2016;26:1142-1147.
No conflicts of interest declared. 2. Weiss M, Dullenkopf A, Gysin C, Dillier CM, Gerber AC. Shortcomings
of cuffed paediatric tracheal tubes. Br J Anaesth. 2004;92:78-88.
3. Tessaro MO, Salant EP, Arroyo AC, Haines LE, Dickman E. Tracheal
rapid ultrasound saline test (T.R.U.S.T.) for confirming correct endotra-
ORCID
cheal tube depth in children. Resuscitation. 2015;89:8-12.
4. Kayashima K, Doi T, Yamasaki R, Imai K. Long-axis ultrasonic images
Kenji Kayashima http://orcid.org/0000-0003-3727-8746
of the pediatric larynx and trachea with a cuffed endotracheal tube.
Anesthesiology. 2017;127:1016.
5. Kraft M, Mende S, Arnoux A, Arens C. Anatomical landmarks for
Kenji Kayashima endosonography of the larynx. Head Neck. 2010;32:326-332.
Keiko Imai
Department of Anesthesiology, Japan Community Health Care
Organization Kyushu Hospital, Kitakyushu, Fukuoka, Japan
Email: [email protected]