Risks and Benefits of Second-Generation Rotavirus Vaccines

From Medscape Pediatrics > Viewpoints William T. Basco, Jr., MD Posted: 08/17/2011 Intussusception Risk and Health Benefits of Rotavirus Vaccination in Mexico and Brazil Patel MM, López-Collada VR, Bulhöes MM, et al N Engl J Med. 2011;364:2283-2292 Study Summary The 1999 rotavirus vaccine was associated with an increased risk for intussusceptions, with the highest risk at 3-7 days after the first dose, believed to be associated with the peak viral replication occurring during this period. This resulted in 1 case of vaccine-induced intussusception per 10,000 children who received the vaccine. Studies of the 2 new rotavirus preparations, conducted in more than 120,000 children, do not reveal an increased risk for intussusception. In 2006 and 2007, Brazil and Mexico began administering the new rotavirus vaccines, allowing for an extensive postmarketing evaluation of the potential effects of these new vaccines. Patel and colleagues used 2 statistical approaches to evaluate the data: a case-series approach (resulting in a rate ratio) and a case-control approach (resulting in an odds ratio). Data were collected from 2008 to 2010 in multiple locations in both countries. In general, children were immunized at 2 months (dose 1) and 4 months (dose 2) of age. Surveillance identified potential cases, and then investigators conducted reviews of medical records to verify and obtain additional clinical data. A verified case was any infant who had intussusception documented at surgery, by contrast enema, by ultrasound, or at autopsy. All cases were between 6 and 35 weeks old at the time of intussusception and all affected children were born after the new rotavirus vaccines were introduced into their respective countries. The investigators considered the period of 1-7 days after vaccination to be the time of principal risk. They adjusted for the background incidence of intussusception in each country in 2-week intervals. They also conducted a comparison of benefit and risk and estimated how many deaths might have occurred in each country without the rotavirus vaccine. They then compared this with the number of intussusceptions and deaths from intussusceptions. The study cohort included 615 case infants (with intussusception) and 2050 control infants (without intussusception). In the intussusception group, 19 children (3.1%) died. In Mexico, 91% of infants with intussusception were diagnosed after the first dose, compared with 44% in Brazil. In Mexican infants, a significant increase in intussusception occurred in the 1- to 7-day period after the first vaccination. In Brazil, a significant increase in intussusception occurred in the 1- to 7-day period after the second vaccination. Age of receipt of the first dose of vaccine did not appear to be related to rate of intussusception in either country. The investigators comment that in Brazil, the first dose of rotavirus vaccine is given along with oral polio vaccine, whereas in Mexico, the rotavirus vaccine is given with the inactivated polio vaccine. Other existing data suggest that coadministration of these 2 live virus vaccines reduces immune response to the first rotavirus vaccine. When applied to a hypothetical cohort of children in Mexico or Brazil, the rotavirus vaccine would avert 663 rotavirus deaths in Mexico and 640 rotavirus deaths in Brazil, compared with an estimated 2 deaths in Mexico and 3 deaths in Brazil caused by the vaccine-induced intussusception. Patel and colleagues concluded that receipt of rotavirus vaccine is associated with a short-term increase in risk for intussusception, but the overall risk-benefit ratio suggests that the benefits of reduced rotavirus deaths and hospitalization in the first 5 years of life far exceed the small number of excess deaths potentially caused by the vaccine. Viewpoint In an accompanying editorial, Greenberg points out that the rotavirus vaccine has reduced hospitalization and deaths in both developed nations and in nations in transition. Clearly, some increased risk for intussusception after vaccination exists with the newer vaccines, although this risk was not demonstrated in the US trials. Whether this risk is the same in fully developed nations as in developing nations is also difficult to determine and cannot be answered directly with these data. I am most interested in the modeling part of the study, which might be the most helpful data for clinicians in the United States. The assumptions would be different if the same study were conducted in the United States, and the findings would probably be biased toward a reduced benefit of the vaccine compared with that demonstrated in Mexico and Brazil. Nonetheless, with such huge benefit-risk ratio (eg, 663 deaths averted compared with 2 vaccine-associated deaths in Mexico), the risk-benefit ratio in the United States would still be great even if the benefit was halved. References