Josephine van Dongen

8 Summary and general discussion 187 and had the most favorable risk-benefit profile with regard to intussusception. Based on these results, selective vaccination against rotavirus AGE can be favorable on all criteria, provided that the vaccine is effective in a population with MRC. We subsequently evaluated whether vaccination indeed offered protection against rotavirus disease in RIVAR eligible infants. Based on efficacy trials among healthy term and preterm infants, a conservative vaccine effectiveness of 60-80% was anticipated. In the RIVAR project the human rotavirus vaccine (HRV, Rotarix, GSK Biological SA, Belgium) was used. By comparing the vaccinated versus unvaccinated RIVAR cohor t we were able to assess rotavirus vaccine effectiveness among MRC infants ( chapter 4 ). In contrast to the vaccine efficacy trials, HRV was not significantly protective against severe rotavirus AGE (vaccine effectiveness 30%; 95%CI -36;65%) in this study, and no impact on rotavirus hospitalizations was observed. As the numbers were however small, a post-hoc analysis was added for all-cause AGE and rotavirus AGE of any severity, but comparable and non–significant effects were observed, suggesting that rotavirus vaccine performance is minimal in infants with MRC.The upper 95% confidence limit of rotavirus vaccine effectiveness was 65% in our study, as opposed to a mean estimate of 90% for healthy infants in high-income settings. 5,6 In addition, lower 95% bound vaccine effectiveness estimates for healthy infants are 76-84%. 5,6 We therefore conclude that HRV effectiveness is significantly lower in MRC infants and this finding signifies the need for risk-group specific research. Moreover, we found that serious adverse events that were possible vaccine reactions 7 occurred more frequently than previously described in the population with MRC, at a rate of 0.25 per 100 vaccine doses ( chapter 5 ). The majority of adverse events were gastrointestinal. Possible causality was based on the temporal relationship with HRV administration and biological plausibility in terms of their pathophysiology. Overall, more (gastrointestinal) adverse events were repor ted among vaccinated premature infants as opposed to non-rotavirus vaccinated (RR 1.07, 95%CI 1.04;1.10). In the group of infants with a gestational age below 27 weeks, who were vaccinated off-label, rotavirus vaccination was generally well tolerated and no safety signals were observed. Acknowledging the combined effectiveness and safety results, clinicians should weigh the risks and benefits of rotavirus vaccination for the individual infant with MRC. Fur thermore, there was no evidence of benefit due to non-specific effects of rotavirus vaccination ( chapter 6 ). Live-attenuated vaccines, like HRV, have been associated with beneficial non-specific effects. This phenomenon is attributed to trained immunity, a vaccine induced modification of the innate immune system that reduces the risk and/or severity of disease by non-target infections. In our study, HRV was not protective against acute hospitalization when excluding admissions for AGE (hazard ratio: 0.91, 95%CI 0.76;1.16). In addition, we did not