Association between rotavirus vaccination and intussusception in Australian children: A record linkage study.

BACKGROUND

Post-licensure surveillance studies have shown a small but increased risk of intussusception among infants in the days following rotavirus vaccination (RV).

OBJECTIVES

We assessed the temporal trends of intussusception-coded hospitalisations before and after the commencement of a universal rotavirus vaccination programme in Western Australia (WA) in 2007. We also assessed the perinatal factors and pathogens associated with these hospitalisations.

METHODS

Intussusception-coded hospitalisations occurring in a cohort of 367 476 WA-born children (2000-2012) aged <5 years were probabilistically linked to perinatal and pathology records. Age-specific incidence rates for overall and pathogen-specific intussusception-coded hospitalisations were calculated before (2000-2006) and after (2008-2012) RV introduction. Adjusted Cox proportional hazards models were used to assess perinatal risk factors for intussusception.

RESULTS

The overall rate of intussusception-coded hospitalisation was 26.4 per 100 000 child-years (95% confidence interval [CI] 24.0, 29.0) among children aged <5 years, with rates being 70% higher (95% CI 39, 107) in the RV period than in the pre-RV period. Compared with the pre-RV period, rates were higher among those aged 12-23 months (by 55%, 95% CI 5, 127) and 2-4 years (by 84%, 95% CI 20, 182) in the RV period. However, the risk of intussusception-coded hospitalisations associated with intussusception management-related procedure code(s) was similar among all age groups in both birth periods. Among infants aged <12 months, male sex, non-Aboriginal status, birth to multiparous mothers, and birth in RV era were independent risk factors associated with intussusception-coded hospitalisations. Adenovirus was strongly associated with intussusception (6.7 per 100 000 child-years, 95% CI 5.3, 9.3).

CONCLUSIONS

The risk of intussusception-coded hospitalisations was higher post-RV introduction, but not for intussusception-coded hospitalisations associated with procedure code(s). The increase was no higher in the vaccine-eligible age group than in older age groups, suggesting that the apparent increase is likely to be attributable to causes other than vaccination.