Measles vaccination and Guillain-Barré syndrome.

BACKGROUND: Guillain-Barré syndrome (GBS) has been associated with several infectious agents, and the possibility that the disorder may be caused by vaccination has been raised. We compared the numbers of cases of GBS observed immediately after mass measles vaccination campaigns with the numbers that would be expected from baseline rates, to assess whether there is a causal relation between measles vaccination and GBS. METHODS: We analysed data on 2296 cases of GBS reported to the Poliomyelitis Eradication Surveillance System of the Pan American Health Organization as cases of suspected poliomyelitis. These cases occurred among 73 million immunised children aged 9 months to 15 years in Argentina, Brazil, Chile, and Colombia, between January, 1990, and December, 1994. These children were targeted for mass measles vaccination campaigns (each lasting 1 month) in 1992 and 1993. The frequency of GBS cases observed during the vaccination campaigns or the next 42 days (the latent period) was compared with that during the rest of the study period, with the assumption of a Poisson distribution. FINDINGS: The average annual incidence of GBS was 0.62 per 100000 children aged 1-14 years. The number of cases that would be expected within any 72-day period would therefore be 92. The average observed number of cases during latent periods after measles vaccination was 97. The probability that 97 or more cases would occur during a period with an expected number of 92 was 0.31. INTERPRETATION: The average annual rates of GBS by age-group for the 5 years analysed were consistent with previous data; thus we are confident that the surveillance system is sufficiently sensitive. There was no statistically significant association between measles vaccination and GBS. If there is any causal relation, the number of GBS cases due to measles vaccination was so small that data from the vaccination of more than 70 million children were not sufficient to detect a rise in the number of observed GBS cases beyond the expected number.