Our initial doubts about the overall risk of this vaccine arose from evaluation of published data from CDC which failed to reveal a striking increase in the overall number of cases of intussusception that would have been anticipated from the extremely high odds ratios, noted above, that were presented for the first week following vaccination. For example, CDC reported the occurrence of three cases of intussusception in 9802 vaccinees in a California managed care study, which translated to an overall incidence of 30 cases per 100 000 over an unspecified period of time. However, because one of the three cases occurred during the first week after vaccination, CDC reported that the incidence of intussusception was 314 per 100 000 individuals during that period (CDC 1999b). However, the overall risk of the vaccine in this large cohort of children was not presented.
Similarly, in a Minnesota study, five cases were reported in recipients of 53 479 doses, an incidence of ~9 per 100 000 doses over an unspecified period of time; however because three of the infants developed intussusception within 1 week of vaccination, the rate of intussusception was calculated to be 292 per 100 000 infant years (CDC 1999b).
In addition, the CDC reported that 102 cases of intussusception were detected by the VAERS in the ~ 1 million vaccinated infants (ACIP 1999). This translates to an incidence of ~ 10 cases per 100 000 vaccinees over an unspecified period of time. Although VAERS is a passive reporting system, even assuming a 50% sensitivity of detection, the maximum incidence from these data would translate to ~ 20 cases per 100 000 vaccinees, a figure well below the background rate of 51 cases per 100 000 infants during the first year of life. Because of the low number of cases overall, we considered that there was a 'disconnect' between the extremely high odds ratios calculated from case/control, case series, and cohort studies, and the relative risks of 1.8, 1.6 and 1.7, noted earlier, with the actual number of cases observed.
The results of the 10 MCOs study are of particular interest (CDC-EIS Conference 2000). Of the 61 371 infants who received at least a single dose of the RRV-TV vaccine, six infants developed intussusception during the 3 week period after vaccination, which yields an incidence of 9.8 cases per 100 000 vaccinees (i.e. 1 per 10 229 infants) over an overall unspecified period of time, a figure well below the predicted 1 per ~ 2000 infants during the first year of life as baseline from the NY State study. It is clear to us from the further expanded cohort study and from the figures cited above, that an 'epidemic' of intussusception did not occur following administration of the rotavirus vaccine in the USA, confirming our conclusion that there was a 'disconnect' or major inconsistency between the odds ratios during the first week after vaccination and the attributable risk figure of 1 case per 12 274 vaccinees derived from the 10 MCOs cohort study.
Was this article helpful?