Safety monitoring of ROTAVAC vaccine and etiological investigation of intussusception in India: study protocol

The objectives of intussusception surveillance are:

To assess the risk of intussusception following ROTAVAC administration using the self-controlled case-series method.

To describe potential infectious etiologies of intussusception by testing for a wide range of enteric pathogens in stool samples of intussusception cases and matched controls using sensitive molecular methods.

The project is managed by the Christian Medical College (CMC), Vellore, and the Centers for Disease Control and Prevention (CDC), Atlanta, USA in collaboration with the Translational Health Science and Technology Institute (THSTI), Faridabad and the Indian Council of Medical Research (ICMR). CMC, Vellore is responsible for all administrative arrangements, while monitoring the surveillance is jointly done by CMC, THSTI and ICMR.

The criteria for sentinel hospitals to participate in surveillance include the ability to diagnose and manage cases of intussusception (availability of pediatric surgeon, radiologist, equipment, and facilities to manage intussusception). We selected large tertiary care hospitals in states introducing the vaccine as part of the UIP, and requested their participation. A meeting was organised for site representatives of potential sentinel hospitals. Representative from each site were requested to provide details on the facilities available in their hospitals for managing intussusception cases and retrospective data on intussusception admissions among children aged less than two years for a period of one year (Additional file 1: Table S1). Sites with appropriate expertise and infrastructure to manage intussusception cases were selected to participate in the surveillance program. Active surveillance was initiated at 28 sentinel hospitals including two hospitals from states/union territories without rotavirus vaccine as part of the UIP, but which serve as referral centres admitting cases of intussuception from states with rotavirus vaccination (Table 1). A memorandum of understanding (MoU) was signed between each sentinel hospital and CMC, Vellore. During surveillance initiation, training sessions were conducted to the surveillance staff at each hospital, which includes one or more pediatric surgeons, radiologists, pediatricians/community health physicians, and a field research assistant.

The multi-centric evaluation for intussusception is at 28 sentinel hospitals in nine states of India for four years (Fig. 1). Based on the recent figures, the sentinel hospitals should be able to admit 40–50 cases per state per year (Additional file 1: Table S1). For the primary objective, the self-controlled case-series methodology will investigate the temporal association between the transient exposure and outcome, in which the individual with the outcome of interest act as his/her own control [22]. The secondary objective of evaluating potential infectious etiologies will be performed through a matched case-control design.

For the secondary objective, stool specimens are collected from both cases and controls. A bulk stool specimen (~ 5 ml) is obtained from each enrolled child, preferably on the day of presentation to hospital. In case of any delay, attempts are made to obtain a stool specimen within 48 h of hospital admission to rule out nosocomial infection. The stool specimen is collected in a sterile screw-top container labelled with a unique identification number and date of collection. At the sentinel hospitals, stool specimens are stored at − 20 ^ͦ C until shipment to referral laboratory at CMC, Vellore. Once in a month, stool specimens are sent to the referral laboratory in a vaccine carrier with frozen gel packs.

For the secondary objective, testing for the presence of multiple enteropathogens in stool samples from intussusception cases and their matched controls is by using custom made Taqman array card (TAC) assays [24]. Briefly, total nucleic acid is extracted from stool samples and tested for enteropathogen targets including enteric viruses, bacteria and parasites using arrayed singleplex real-time polymerase chain reaction (qPCR) assays. Figure 2 shows all the enteropathogen targets included on these assays for this evaluation.

Once a month, completed CRFs of all the children enrolled into surveillance (cases and controls) along with supportive documents and stool specimens are sent to the referral laboratory at CMC, Vellore. On receipt of CRFs and stool samples, an acknowledgement e-mail communication along with any issues found with the quality and quantity of the samples is sent to the sentinel hospitals. At the central data processing centre, all CRFs are stored in a secure, locked cabinet. Trained personnel do data entry into an electronic database using structured query language. Forms with missing data are reviewed and if the program coordinator determines that it is possible to recover the data, the site investigator will attempt to obtain the missing information. Data quality for the surveillance is checked through on-going assessment of sentinel hospital performance.

After initiation of surveillance, sentinel sites are visited once in 3 months. At each visit, the sites are evaluated using a monitoring checklist (Additional file 2: Table S2), which recorded performance in terms of enrollment of children with intussusception, collection of ultrasound reports along with images, collection of procedure/treatment notes, enrolling the matched controls, collection of adequate stool samples and obtaining copies of the vaccination cards. Every year, a collaborators’ meeting is organized to discuss the work done by each site and to enable collaborators to provide feedback to individual sites. The hospitals not meeting performance criteria are excluded from subsequent surveillance based on monitoring reports.

Initial reluctance in participation was overcome by site investigators’ meetings before initiation, which helped to explain the project design, surveillance activities and significance of the project. During the surveillance, determining the vaccination status of the child and obtaining a photocopy of the vaccine card is challenging, as parents do not always carry the vaccine cards. In such cases, parents are asked to send a copy of the vaccination card by post or by email. In certain cases, the surveillance staff travels to the child’s home to obtain copies. For subjects with no immunization cards available, the surveillance staff contact the auxiliary nurse midwife at health sub-centers [29] who maintain government immunization records to verify receipt of rotavirus vaccine. For children vaccinated at private hospitals, attempts are made to identify the manufacturer by contacting the health facility where the child received the rotavirus vaccine. For the etiology evaluation, enrolling controls satisfying all the matching criteria was challenging, especially at big referral hospitals having cases from two or more states. Surveillance physicians had to carefully scrutinize all the potential controls before enrolling a matched control.

Monitoring and demonstrating the safety of the vaccine in terms of intussusception after introduction of the vaccine into the routine immunization programme is important to ensure continued support and commitment for the rotavirus vaccination program. Anticipated public health benefits from this surveillance include the generation of information useful to public health officials and to national governments in making decisions through future risk-benefit analysis of the vaccine. This evaluation will also generate evidence for other low- and middle- income countries (LMIC) planning to introduce rotavirus vaccine.

The knowledge on the etiology of intussusception is limited worldwide [30]. Evaluating the infectious agents present in stool samples from intussusception cases and matched controls may help to determine potential infectious etiologies of intussusception, and will assist in further understanding the association, if any, between rotavirus vaccination and intussusception.