The link between rotavirus vaccination and intussusception: implications for vaccine strategies.

By the age of five years, between 1 in 40 and 1 in 77 children in Europe and the USA will have been hospitalised for rotavirus diarrhoea. Although health care and economic costs of the illness are high, mortality in developed countries is very low. In contrast with the developing world, and despite considerable efforts to educate carers in simple oral rehydration therapy, it is estimated that 600 000–800 000 children, or 1 in 40 children in the first five years of life, die annually from rotavirus infection. The advantages of an oral rotavirus vaccine thus seem self evident: economic benefits to the developed world and significantly reduced infant mortality in the developing world. Candidate rotavirus vaccines were first developed by tissue culture adaptation and attenuation of bovine and rhesus rotaviruses. Subsequently, such heterologous rotaviruses were improved for use as human vaccines by reassortment with human rotaviruses. After favourable results of prelicensing trials in the USA and Finland showing vaccine efficacies of 68% and 91% for reduction of all and serious rotavirus infections, respectively, the US food and drug administration licensed an oral tetravalent rhesushuman reassortant rotavirus vaccine (RRV-TV) in August 1998. Distribution of the vaccine for incorporation into infant schedules commenced in the USA in October 1998. By the end of May 1999, when approximately 1.5 million doses of RRV-TV vaccine had been distributed, nine cases of intussusception in infants who had received RRV-TV had been reported to the Vaccine Adverse Events Reporting System, as compared with only four reports to that organisation in the preceding seven years. As a result of these data, the vaccination programme was initially suspended and the company withdrew the vaccine from the market in October 1999. Three questions arise as a result of these actions: •was this a true effect? •what was the mechanism of the effect? •what are the ethical implications? The first of these questions was carefully addressed in the recently published investigation by the Rotavirus Intussusception Investigation team of the Centres for Disease Control (CDC) in the USA. The study was based on retrospective case controlled and case series analyses of infants aged 1–12 months, hospitalised with radiologically or operatively confirmed intussusception between 1 November 1998 and 30 June 1999 across the 19 states of the USA where 80% of the administered RRV-TV had been distributed. Of the 446 infants identified with intussusception, 429 were included in the study (96%). A total of 17.2% of the patients with intussusception and 12.8% of the controls had received RRV-TV (p=0.02). The severity of intussusception was not significantly different in control and vaccinated patients. In the RRV-TV group, the risk of intussusception was greatest in the 3–14 days after any dose of vaccine, with an odds ratio of 10.6. This translated to an odds ratio of 21.7 after the first dose when the majority (64%) of cases occurred. The authors estimated that if the RRV-TV programme were fully implemented across the USA, between 4670 and 9474 infants would be vaccinated for each case of intussusception attributable to RRV-TV. They concluded that their data provided evidence of a strong temporal and specific link between RRV-TV and intussusception. The CDC’s retrospective study was not designed to address the aetiology of the link between RRV-TV vaccine and intussusception. Indeed, the pathogenesis of intussusception in infants is poorly understood. The observation of mesenteric lymphadenopathy or inflamed Peyer’s patches in a significant proportion of cases points to an inflammatory process. This has been linked to a number of viral infections: commonly adenovirus, but also human herpes virus (HHV)-6, HHV-7, Ebstein-Barr virus, and cytomegalovirus. Although uncommon, wild rotavirus infection has also been Murphy TV, Gargiullo PM, Massoudi MS, et al. Intussusception among infants given an oral rotavirus vaccine. N Engl J Med 2001;344:564–72.