Severity of Rotavirus-Vaccine-Associated Intussusception: Prospective Hospital-Based Surveillance, Australia, 2007-2018

Background: Multiple studies have shown an association between intussusception (IS) and receipt of monovalent or pentavalent rotavirus vaccine (RV) in the previous 21 days. Disease severity is an important consideration for risk-benefit evaluations of RV, but no studies have compared the severity of IS within 21 days of vaccination (vaccine-associated, VA) and later (not temporally-associated, VNA). Methods: We used active hospital-based surveillance in the Australian Paediatric Active Enhanced Disease Surveillance (PAEDS) network (July 2007 to February 2018) to identify infants ≤9 months of age meeting Brighton level 1 criteria for IS. We used five severity levels: (1) no surgery and length of stay (LOS) ≤1 day, (2) no surgery and LOS ≥2 days, (3) surgery, no bowel resection, (4) bowel resection, and (5) ICU admission. Results: Of 323 eligible cases, 87 (26.9%) were VA and 236 (73.1%) VNA. VA-IS cases (median 21 weeks; 24.1% ≤14 weeks) were significantly younger than VNA-IS cases (median 28 weeks, 7.2% ≤14 weeks). Cases 0-≤14 weeks of age were significantly more likely than cases ≥25 weeks to require bowel resection (relative risk ratio 4.6, 95% CI, 1.48-14.3). This effect was not associated with RV. After adjustment for age and sex, VA-IS was not significantly overrepresented in severity levels 2-5; adjusted RRR of 1.37 (95% CI: 0.61-3.11) for bowel resection in cases 0-≤14 weeks of age. Conclusions: IS was uncommon but significantly more severe under 14 weeks of age. After adjustment for age and sex, IS severity was not related to RV.