Integration of poliovirus and enteropathogen sewage surveillance in Dhaka Bangladesh: a longitudinal surveillance study, June 2019–June 2020

BACKGROUND: The Global Polio Eradication Initiative (GPEI) uses environmental surveillance to monitor circulation of poliovirus. After the certification of poliovirus eradication, environmental surveillance continues, generally through its integration into other infectious disease surveillance programmes. In this study, we evaluated the programmatic utility of TaqMan Array Cards (TACs) to detect poliovirus in sewage while simultaneously detecting enteric pathogens and markers of antimicrobial resistance (AMR). METHODS: This longitudinal surveillance study was conducted across 12 sites in three wards (8, 9, and 10) in Dhaka, Bangladesh, from June, 2019, to June, 2020. Demographic data of children younger than 5 years, including vaccination history, sanitation, and household characteristics, were obtained from the Health and Demographic Surveillance System. Sewage samples were collected between June, 2019, and June, 2020, using a bag-mediated filtration system. We used TACs to detect poliovirus in sewage and simultaneously tested for other enteric pathogens and markers of AMR. Sites were selected after mapping of the informal sewage network and a demographic survey of the population. Samples were collected before and after a bivalent oral poliovirus vaccine (bOPV) campaign. We used a water-quality probe to assess physicochemical properties of the sewage. A multivariable mixed-effects gamma hurdle regression model was used to examine the association of enterovirus detection and concentration with site properties. FINDINGS: The Health and Demographic Surveillance System recorded 19 236 children younger than 5 years across 12 sites in the three wards in Dhaka. The bOPV campaign reached 17 282 children. 372 sewage samples were collected over 379 days. The highest concentration of Sabin 1 and Sabin 3 polioviruses was detected 2 weeks after the bOPV campaign (mean viral copies per litre of sewage were 0·83 SD 2·13 for Sabin 1 and 0·84 1·96 for Sabin 3 vs baseline values of 0·05 0·21 for Sabin 1 and 0·11 0·50 for Sabin 3; p=0·004 for Sabin 1 and p=0·005 for Sabin 3). Detection of enteroviruses was more likely with increasing levels of total dissolved solids (adjusted odds ratio per absolute increase of 100 mg/L 1·39 95% CI 1·17-1·61). The median environmental surveillance viral load of rotavirus was 0·567 (IQR 0·202-0·839), and rotavirus had the strongest correlation with respective concurrent clinical case incidence (r=0·828; p=0·0017). 31 AMR genes of clinical significance were detected. INTERPRETATION: When the GPEI dissolves, poliovirus surveillance would need to be integrated into other surveillance programmes. TACs could be used to screen suitable pathogens for integrated sewage surveillance alongside poliovirus. Further validation is required across different geographies and poliovirus prevalence, and data interpretation requires an understanding of site sensitivity. FUNDING: Gates Foundation.