Abstract

OBJECTIVE: To investigate the risk of intussusception after monovalent rotavirus vaccine (RV1) given to infants aged 2 and 3 months in England. METHODS: Hospital Episode Statistics (HES) were used to identify infants aged 48-183 days admitted between 11/03/2013 and 31/10/2014 with intussusception. Diagnosis was confirmed from medical records and HES procedure codes. Vaccination status was obtained from general practitioners. The risk of admission within 1-7 and 8-21 days of vaccination was analysed using the self-controlled case-series (SCCS) method with age effect adjustment by including historical data before RVI introduction in July 2013. RESULTS: A total of 119 cases were identified during the study period and intussusception confirmed in 95 of whom 39 were vaccinated 1-21 days before onset. An increased relative incidence (RI) in this period was found, 4.53 (95% confidence interval 2.34-8.58) and 2.60 (1.43-4.81) respectively after the 1st and 2nd doses with an attributable risk of 1.91 and 1.49 per 100,000 doses respectively. The peak risk was 1-7 days after the first dose, RI 13.81 (6.44-28.32), with an estimated 93% of the 15 cases being vaccine-attributable. Mean interval between onset and admission, and clinical features were similar between vaccine-associated and background cases. Despite intussusception being a contraindication to rotavirus vaccination, 10 infants received a further dose; none had a recurrence. The RIs in a meta-analysis combing our results with Australia, Mexico, Brazil and Singapore using RV1, a 2, 4 month schedule and SCCS gave pooled RI estimates of 2.35 (1.45-3.8) and 1.77 (1.29-2.43) in the 21 day period after the 1st and 2nd doses, respectively. The earlier age at the 2nd dose in England did not affect the risk. CONCLUSION: We estimate that the RVI programme causes around 21 intussusception admissions annually in England but, since it prevents around 25,000 gastro-intestinal infection admissions, its benefit/risk profile remains strongly positive.