Acute gastroenteritis hospitalisations peaked during March to May, an autumn–winter pattern corresponding SB431542 supplier to the typical

rotavirus season months in South Africa. This was particularly evident in the HIV-uninfected children. There seemed to be a less seasonal pattern among admissions in HIV-infected compared to HIV-uninfected children, possibly reflecting a greater diversity of pathogens associated with acute diarrheal disease in HIV-infected children and a proportionally lesser role of rotavirus. Efficacy of the rotavirus vaccine against severe rotavirus gastroenteritis was 77% in South Africa and there was a 30% reduction in all-cause severe gastroenteritis in an efficacy trial conducted in South Africa and Malawi [15]. In South African infants, rotavirus vaccine was shown to be both safe and immunogenic in a group of HIV-infected children [16] and use of the vaccine in the routine immunisation program is expected to reduce the burden of rotavirus disease in these children. Rotavirus vaccine was introduced into the EPI in South Africa in August 2009 and is expected RG7204 mouse to provide considerable public health benefits in South Africa.

Efficacy of the rotavirus vaccines is greatest against severe disease and the impact of vaccination will be greatest on the more severe outcomes, for example hospitalisation. Postlicensure data from the United States shows that the rates of all-cause diarrhoea hospitalisations in children under 5 years of age declined following introduction GBA3 of the rotavirus vaccine [17]. This was not only in vaccine-eligible children and raises the possibility of indirect protection for unvaccinated persons in the community. The decrease in prevalence of rotavirus disease may thus be greater than expected following vaccine introduction in South Africa. However, in considering the findings of this study there are several limitations to consider. HIV results were not available for the participants

in the cohort who were not hospitalised, and an estimated HIV prevalence was used based on assumptions of maternal HIV prevalence and mother-to-child transmission of HIV. These assumptions may have led to an inaccurate estimate of the true incidence of acute gastroenteritis based on HIV infection status. For incidence calculations, those with an unknown HIV result were considered to be HIV-uninfected. There was thus a risk of misclassification as some of these may actually have been HIV-infected. However, any misclassification of children as HIV-uninfected who were truly HIV-infected would have led to an underestimation of the true incidence of acute gastroenteritis in the HIV-infected cohort. All the infants in this study were on average 6 weeks old on enrolment, so disease in neonates and preterm infants could not be investigated.