Diversity of rotavirus strains circulating in children under five years of age who presented with acute gastroenteritis before and after rotavirus vaccine introduction, University Teaching Hospital, Lusaka, Zambia, 2008–2015

Background: Following the introduction of rotavirus vaccine into the routine immunization schedule, the burden of rotavirus disease has significantly reduced in Zambia. Although rotavirus vaccines appear to confer good cross-protection against both vaccine and non-vaccine strains, concerns about strain replacement following vaccine implementation remain. We describe the diversity of the circulating rotavirus strains before and after the Rotarix® vaccine was introduced in Lusaka from January 2012. Methods: Under five children were enrolled through active surveillance at University Teaching Hospital using a standardized WHO case investigation form. Stool samples were collected from children who presented with ≥3 loose stool in 24 h and were admitted to the hospital for acute gastroenteritis as a primary illness. Samples were tested for group A rotavirus antigen enzyme-linked immunosorbent assay. Randomly selected rotavirus positive samples were analysed by reverse transcription polymerase chain reaction for G and P genotyping and and Nucleotide sequencing was used to confirm some mixed infections. Results: A total of 4150 cases were enrolled and stool samples were collected from 4066 (98%) children between 2008 and 2011, before the vaccine was introduced. Rotavirus antigen was detected in 1561/4066 (38%). After vaccine introduction (2012 to 2015), 3168 cases were enrolled, 3092 (98%) samples were collected, and 977/3092 (32%) were positive for rotavirus. The most common G and P genotype combinations before vaccine introduction were G1P[8] (49%) in 2008; G12P[6] (24%) and G9P[8] (22%) in 2009; mixed rotavirus infections (32%) and G9P[8] (20%) in 2010, and G1P[6] (46%), G9P[6] (16%) and mixed infections (20%) in 2011. The predominant strains after vaccine introduction were G1P[8] (25%), G2P[4] (28%) and G2P[6] (23%) in 2012; G2P[4] (36%) and G2P[6] (44%) in 2013; G1P[8] (43%), G2P[4] (9%), and G2P[6] (24%) in 2014, while G2P[4] (54%) and G2P[6] (20%) continued to circulate in 2015. Conclusion: These continual changes in the predominant strains suggest natural secular variation in circulating rotavirus strains post-vaccine introduction. These findings highlight the need for ongoing surveillance to continue monitoring how vaccine use affects strain evolution over a longer period of time and assess any normal seasonal fluctuations of the rotavirus strains.