Rotavirus is the leading cause of severe diarrhea in infants and children worldwide, leading to more than half a million deaths each year in children under the age of 5. Before the introduction of the rotavirus vaccine, almost all children in the US were infected by the age of 5, and rotavirus infections were responsible for 400,000 doctor visits, 200,000 ER visits, 55,000-70,000 hospitalizations, and 20-60 deaths each year. The first rotavirus vaccine, RotaShield, was found to have an association with intussusception. Two newer vaccines, Rotateq and Rotarix, were thought not to carry that risk, but two new trials have shown that they do. Still, the risk is small and the benefits of the vaccines are great.
What is intussusception?
Intussusception is a “telescoping” of the intestine where one section slides inside another section. This can cut off the blood supply, block the intestine, and cause tears, infections, and death. Most cases are in young children. They have severe abdominal pain (intermittent at first), and pass blood in the stool, typically mixed with mucus and having the appearance of currant jelly. A barium enema can confirm the diagnosis and simultaneously treat it. Sometimes surgery is needed.
The baseline incidence of intussusception in children is on the order of 1-4 per 1,000. Most cases have no identified cause, but the most plausible candidate is hypertrophied lymphoid tissue resulting from viral illnesses, especially rotavirus infections.
Vaccine studies
The first rotavirus vaccine, Rotashield, was introduced in 1999. It was voluntarily withdrawn from the market within a year because post-marketing surveillance found 1-2 excess cases of intussusception per 10,000 recipients. Newer vaccines, Rotateq and Rotarix, were licensed only after testing (in over 60,000 infants each) failed to find any association with intussusception. Those trials were designed to have enough statistical power to detect a risk similar to that of RotaShield. Both new vaccines contain live, attenuated strains of virus and are given orally. Rotateq is a pentavalent (prepared from 5 strains) vaccine given in 3 doses at age 2, 4, and 6 months. Rotarix is monovalent (prepared from 1 strain) and is given in 2 doses at age 2 and 4 months. Either is recommended, but about 10 times more doses of the pentavalent vaccine have been administered.
After the new vaccines came into common use, studies in other countries pointed to a small increase in intussusception with the newer vaccines, although at a much lower rate than with Rotashield. The FDA responded by initiating a new study of both vaccines in the Post-Licensure Rapid Immunization Safety Monitoring (PRISM) program, a component of the Mini-Sentinel pilot program that was developed by the FDA to conduct active surveillance of the safety of medical products.
That study, by Yih et al., was published in the February 6, 2014 issue of The New England Journal of Medicine simultaneously with another study by Weintraub et al. that was based on data from the Vaccine Safety Datalink project, a collaborative effort between the CDC and nine large managed care organizations including Kaiser Permanente and Group Health Cooperative.
Results
The Yih study found a significant increase in the risk of intussusception after the first dose of Rotateq, with 1.5 excess cases per 100,000 vaccine recipients. They found no association with subsequent doses, but couldn’t rule that out.
For Rotarix, the findings were not significant, but a relatively small number of children got that vaccine (an order of magnitude lower), and the authors concluded that their risk estimates were imprecise and should be considered in the context of findings of increased risk in other studies.
The Weintraub study identified 6 cases of intussusception after Rotarix, compared to 0.72 cases expected from historical data. They found no association with Rotateq.
The two studies appear to have found different results, but they are based on only a handful of cases of intussusception. Small variations in case detection, confirmation of vaccine status, and chance alone could have affected the outcomes.
Their results should be considered along with other studies from the US and other countries. It is reasonable to conclude that intussusception can occur with either vaccine, but that the risk is low, on the order of 1-5 cases per 100,000 infants.
Vaccine benefits
As always, risks must be weighed against benefits, and the benefits of these vaccines are indisputable. An accompanying editorial in the NEJM estimated that each year in the US the vaccines have prevented 53,000 hospitalizations and 170,000 ER visits at the cost of 45 to 213 cases of intussusception. The ACIP estimates that 14 infant deaths are prevented each year in the US, and in Mexico deaths from diarrhea decreased by 40% after the vaccine program was implemented. Even the unvaccinated may benefit as they are exposed to fewer rotavirus infections in the community. The greatest benefits will be seen in Third World countries, but there is greater risk there too. If intussusception is promptly diagnosed and treated, it is relatively benign and has no lasting consequences; but adequate medical care may not be readily available in Third World countries. Another consideration: it’s possible that some infants may be protected from intussusceptions caused by the rotavirus itself.
Recommendations
A rotavirus vaccine (either one) is recommended by the CDC, the American Pediatric Association, and other professional groups as part of the routine immunization schedule in the US. Parents should be informed of the signs of intussusception and should monitor their infants especially in the first 7 days after vaccination; and since intussusception can recur, caution is advised in children who have a history of intussusception.
Science-based medicine in action
This is a success story. Pretty much everything about the rotavirus vaccines was done right. A serious potentially preventable health problem was identified, and the responsible strains of virus were pinpointed. An effective vaccine was developed and extensively tested. The first vaccine was voluntarily withdrawn by the manufacturer within a year because of a problem recognized during post-marketing surveillance. Government programs provided even more effective post-marketing surveillance for the newer vaccines. A possible side effect was recognized. Two (not one) new studies were funded by government agencies, not by Big Pharma. Even VAERS (the Vaccine Adverse Event Reporting System) was used appropriately, not to blame vaccines for random events following vaccination, but to compare the numbers of spontaneously reported cases at different time periods following vaccination and to give researchers a clue that the first dose of vaccine might be implicated, aiding them in designing appropriate studies. Vaccines work, and the scientific process generally works pretty well in spite of the flawed humans who carry it out.