Do childhood vaccines cause thrombocytopenia?

An increasing body of evidence has been gathered since the mid-1960s to support a link between vaccinations, particularly the measles-mumps-rubella (MMR) vaccine and idiopathic thrombocytopenic purpura (ITP). The incidence rate is estimated to be between one in 25,000 to one in 40,000 doses of MMR (1,2); this is much less frequent than after natural infection with measles (common), rubella (one in 3000 cases) and varicella. The purpose of the present commentary is to review vaccine-associated thrombocytopenia (VATP).

Vaccine-preventable diseases are becoming rare in Canada, with an average of 10 reported cases of measles per year between 2002 and 2006; although there was a large outbreak in Quebec in 2007, with 95 confirmed cases (3). With widespread vaccination and the near disappearance of vaccine-preventable diseases in Canada, there is less societal tolerance for adverse events following immunization (AEFI). One of the challenges in assessing AEFIs is distinguishing events that are causally linked with vaccination from those that are only temporally associated. Cases of thrombocytopenia in the first month after vaccination often have a history of recent viral infection or coadministration of medications that may also lead to thrombocytopenia, making the actual cause of VATP difficult to identify.

Since 1992, the Immunization Monitoring Program, ACTive (IMPACT), conducted by the Canadian Paediatric Society, has performed active surveillance for children who are hospitalized with AEFIs, including VATP. Trained nurse monitors at each of the 12 IMPACT centres review all admissions for children admitted for VATP (children with a platelet count of less than 100×10⁹/L and no obvious other cause, such as cancer chemotherapy) within one month of documented receipt of any vaccine. Jadavji et al (4) reported on the first nine years of surveillance (with 61 cases) for VATP in 2003. One of the limitations in the IMPACT data is that detecting a case requires the treating physician to document administration of a vaccine in the previous month (the nurse monitors do not interview the parents). A recent American study (1) found that treating physicians had asked only two of 13 children with VATP about recent vaccination.

The IMPACT data on VATP are similar to reports from other countries, including the United Kingdom, France and the United States (1,2,5,6). In Canada, 103 cases of VATP have been documented by IMPACT since 1992 (7). The median age was 13 months, and 61% of those affected were boys. Petechial rash and bruising were the typical presenting signs. Most (73%) cases were treated with intravenous immunoglobulin. Most children did quite well, with rapid recovery; only six of 95 children with follow-up data still had abnormal platelet counts after three months. However, two children had severe bleeding-related complications, one had a gastrointestinal bleed requiring intensive care and one had post-traumatic intracranial bleeding leading to death.

Most cases of VATP are associated with MMR or measles vaccine, including 72% of the cases reported to IMPACT (25 of these 74 children had received one or more additional vaccines, including 10 children who also received the diphtheria, pertussis and tetanus vaccine, and 10 children who had received the varicella vaccine). Of the children who had MMR associated with thrombocytopenia in the IMPACT study, nine (12%) had a previous recorded dose of the vaccine without known thrombocytopenia. When all of the VATP cases were considered, 31% of the VATP episodes occurred after the second or third exposure to a vaccine.

It is not known how often VATP recurs after repeat vaccination; there have been at least two cases of recurrent measles VATP reported in the literature (8,9), but in a small number of children with previous ITP (57 in three studies), MMR vaccination did not precipitate a recurrence (1,2,6). A data linkage study (10) examining the second dose of MMR did not find any increased risk of thrombocytopenia. Some authors believe that due to the existence of the two recurrence case reports and the low risk of measles in North America, VATP should be considered to be a contraindication to subsequent measles vaccination (9), but that view is not widely held (1,2,6).

Two doses of the measles vaccine are used in Canada because after the first dose at 12 to 15 months of age, only 85% to 95% of children are protected because the remaining maternal antibody can interfere with vaccine efficacy. The role of the second dose of the measles vaccine is to protect the 5% to 15% of children who were unprotected after the first dose (rather than as a booster); after the second dose of the measles vaccine, nearly 100% of children are immune to measles (11). Thus, a practical alternative for children with VATP after a first MMR dose is to test for measles antibody in serum. A positive test result is an acceptable alternative to revaccination. In the setting of measles or rubella outbreaks, the risk of thrombocytopenia after the illness is at least 10-fold greater than VATP, so vaccination should be considered.