| dc.description.abstract | A total of 784 faecal samples collected during the 7 months period, January to July 2010 from children under 5 years old with diarrhoea who were admitted at the 3 referral hospitals; Parirenyatwa Hospital, Harare Hospital and Chitungwiza hospital, were tested for presence of rotavirus antigen using enzyme immune-assay (EIA). Fifty faecal samples from children without diarrhoea were also tested for rotavirus antigen. Sixteen (32 %) of 50 non-diarrhoeal samples and 515 (65.7 %) diarrhoea samples were rotavirus positive. The association between diarrhoea and detection of rotavirus in faecal samples was statistically significant with an overall odds ratio of diarrhoea patients of 4.08 (p<0.0001). The rotavirus diarrhoea prevalence (59.4 %) was high in children ≤ 18 months old diarrhoea patients. The high prevalence of rotavirus diarrhoea was found during the dry cool season in diarrhoea patients < 59 months of age. Fifty rotavirus positive isolates from diarrhoea patients were genotyped using reverse-transcription polymerase chain reaction. A large proportion of samples could not be genotyped; 28.6 % did not produce G genotype result, and 43.2 % did not produce P genotype result. Of the strains that could be genotyped, G9 (24.3 %) was more predominant, followed by G2 (10.9 %), and P[6] (31 %) was more predominant followed by P[4] (14.3 %. The G and P combinations which were more predominant were G1P[6], G9P[6] and G8/G12P[4]. The RNA electrophoresis of the rotavirus genome was performed on 30 rotavirus positive samples, 90 % produced migration patterns typical of the group A rotavirus: 70.4 % were long electropherotypes and 22.2 % were short electropherotypes. The identification of unusual P and G combinations in Zimbabwe may affect the efficacy of currently available rotavirus vaccine formulations and may contribute to the design and development of a broadly reactive rotavirus vaccine for use in African countries. | |
