Detection of group C rotavirus in children with acute gastroenteritis in Blantyre, Malawi

Among 606 children who were treated for acute gastroenteritis at the Queen Elizabeth Central Hospital in Blantyre, Malawi, Group C rotavirus (Gp C RV) was detected by enzyme-linked immunosorbent assay in fecal specimens from 16 (3.9%) of 408 inpatients and in 4 (2.0%) of 198 outpatients. Thirteen (65%) children excreting Gp C RV were coinfected with Group A rotavirus. Rotaviruses are important agents of gastroenteritis in humans and animals. They are segmented, double stranded, RNA viruses that belong to the family Reoviridae. Seven rotavirus groups (A to G) are described, of which only Groups A, B and C have been identified in humans. 1 Group A rotaviruses (Gp A RV) are established as the major cause of severe diarrhea in infants and young children worldwide. 1 In contrast Group B rotaviruses have been associated with epidemics of diarrhea in adults. 2 Group C rotaviruses (Gp C RV) cause gastroenteritis in children and adults and have a worldwide distribution. 2 Recently Gp C RV have gained increasing recognition as significant causes of diarrhea in children, with regard to endemic childhood diarrhea 3 and nosocomial outbreaks of gastroenteritis. 4 There is a lack of information regarding Gp C RV infections in children in sub-Saharan Africa. A single case recently reported from South Africa involving a 3-month-old child represents the first confirmed detection of Gp C RV in Africa. 5 However, a seroprevalence study from South Africa found that 29% of children younger than 5 years of age had demonstrable antibody to Gp C RV, suggesting that this infection may be more common than previously appreciated. 6 Between July, 1997, and June, 1999, we conducted a hospital-based study of gastroenteritis caused by infection with Gp A RV among children with and without HIV infection in Blantyre, Malawi. 7 We have now examined for Gp C RV fecal samples that were collected during the second year of our study. Methods. Full details of the study protocol and HIV testing procedures have been described for the parent study. 7 Briefly fecal and blood samples were collected from children <5 years of age with acute, dehydrating gastroenteritis who either attended the Under-5 Rehydration Clinic at the Queen Elizabeth Central Hospital in Blantyre (outpatients) or were admitted to the pediatric wards of the same hospital (inpatients). A maximum 20-point severity score was calculated for each inpatient, which took account of the duration and frequency of diarrhea and vomiting, maximum axillary temperature, use of intravenous fluids and survival. 7 Written consent was obtained from the parents or guardians of all children enrolled in the study. Fecal samples collected between July, 1998, and June, 1999, were stored at −80°C until tested for Gp C RV. An enzyme-linked immunosorbent assay (ELISA) antigen detection method was used, based on the Cowden strain of porcine Gp C RV. 3, 8 Reverse transcription-polymerase chain reaction (RT-PCR) and Southern hybridization were used to confirm the presence of Gp C RV in ELISA-positive specimens as described previously. 3, 9 Briefly the 5′ half of gene 8 (encoding VP7) was amplified with primers BMJ14 (5′-GGCATTTAAAAAAGAAGA-3′) and BMJ30 (5′-GTACCAAATGCAGCCCAT-3′). For some specimens a nested PCR was performed with a pair of internal primers, BMJ147 (5′-GAAGCTGTCTGACAAACTGG-3′) and BMJ31 (5′-ATTACCATCATTCAAATGTC-3′). Southern hybridization was performed using an oligonucleotide probe. 9 Results. Fecal samples from 606 children were examined for Gp C RV, comprising 408 inpatients and 198 outpatients. Group C rotavirus was detected in 20 (3.3%) of the children, including 16 inpatients (3.9%) and 4 outpatients (2.0%) (Table 1). Sufficient fecal material remained to enable confirmation by RT-PCR and Southern hybridization of 11 of 13 ELISA-positive specimens; 2 samples gave equivocal results. Four children (3 inpatients and 1 outpatient) were coinfected with HIV. Among inpatients the age of children excreting Gp C RV (median 6 months, range 1 to 16 months) did not differ from those of children without Gp C RV in the stool (median 8.5 months, range 1 to 54 months). Similarly among outpatients the age of the 4 Gp C RV-positive children (median 7 months, range 5 to 9 months) was similar to those of the Gp C RV-negative children (median 11 months, range 1 to 33 months). The severity score of children with Gp C RV infection (median 7, range 2 to 10) did not differ significantly from the score of children without Gp C RV infection (median 5, range 2 to 17). Overall 13 (65%) of the children who excreted Gp C RV were coinfected with Gp A RV. The severity score did not differ significantly between Gp C RV-positive children who were coinfected with Gp A RV (median 8, range 2 to 10) and those who were not coinfected (median 5.5, range 2 to 8).Table 1: Characteristics of children with Group C rotavirus infectionsDiscussion. This study has documented the presence of Gp C RV for the first time in sub-Saharan Africa, outside of South Africa. This is also the first description of Gp C RV in HIV-infected children, after an earlier report in an HIV-infected adult. 10 The low rate of detection in all patients suggests that Gp C RV is an uncommon cause of gastroenteritis in this group of children. Furthermore the frequent association of Gp C RV with Gp A RV casts doubt on the pathogenic role of Gp C RV in some children. However, in a recent study involving gnotobiotic calves, dual infection with bovine Group A and porcine Group C rotaviruses was associated with enhanced group C viral shedding and with more extensive histopathologic lesions in the small intestine than was observed in those with Gp C RV alone. 11 This raises the theoretical possibility that Gp A RV infections could increase the severity of Gp C RV disease in children. Although this could not be demonstrated in the current study, numbers were small. We were unable to amplify by RT-PCR 2 of 13 ELISA-positive specimens. This could be attributed to low concentrations of fecal RNA or alternatively could be explained by sequence mismatches between the template and the primers used in this study. An increasing nucleic acid sequence database for Gp C RV will facilitate attempts to refine molecular diagnostics, including RT-PCR-based assays. The application of more sensitive screening tests, including RT-PCR, and ELISA-based assays that use recombinant antigens derived from human Gp C RV 12 will allow a more complete assessment of the importance of Gp C RV in pediatric gastroenteritis. Acknowledgments. This work received financial support from the Wellcome Trust of Great Britain (Grant 049485/Z/96) and forms part of the Malawi-Liverpool-Wellcome Trust Programme of Research in Clinical Tropical Medicine. Nigel A. Cunliffe, M.R.C.P. Benson D. M. Thindwa Cert, M.L.T. Malcolm E. Molyneux, F.R.C.P. Robin L. Broadhead, F.R.C.P.C.H. Wellcome Trust Research Laboratory and Department ofPaediatricsUniversity of Malawi College of MedicineBlantyre, Malawi Baoming Jiang, Ph.D. Viral Gastroenteritis SectionCenters for Disease Control and PreventionAtlanta, GA (BJ) Nigel A. Cunliffe, M.R.C.P. Benson D. M. Thindwa Cert, M.L.T. Malcolm E. Molyneux, F.R.C.P. Winifred Dove, B.Sc. (Hons) C. Anthony Hart, F.R.C.Path. Wellcome Trust Research Laboratory and Department of MedicalMicrobiology and Genito-Urinary MedicineUniversity of LiverpoolLiverpool, UK