The recent Global Enteric Multicenter Study (GEMS) showed that rotavirus is the leading cause of diarrhea in The Gambia as well as other parts of sub-Saharan Africa and South Asia.1 Half a million children die annually due to acute diarrhea caused by rotavirus and most of these deaths occur in low-income countries.2,3 Six of 7 of the countries with the highest rotavirus-related mortality are in Africa.4 There are 2 oral live-attenuated vaccines recommended for routine immunization of all infants by the World Health Organization. RotaTeq (Merck, Whitehouse Station, NJ) is a pentavalent reassortmant vaccine developed from human-bovine strains providing heterotypic protection and Rotarix (GlaxoSmithKline Biologicals, Brussels, Belgium) is a monovalent vaccine providing heterotypic protection based on the human G1P strain.5,6 Both vaccines have the potential to reduce the burden of rotavirus disease; however, implementation globally has been slow, particularly in sub-Saharan Africa.1 One of the impediments to wider vaccine implementation in the region has consistently been the lack of local data on the rotavirus disease burden and circulating genotype diversity to support decision making.
Rotavirus has 2 surface proteins, VP7 and VP4, which appear to induce both homotypic and heterotypic neutralizing antibody responses. Genotyping of VP7 which is a glycoprotein (G-type) and VP4 which is a protease-cleaved protein (P-type) is a useful tool for characterizing rotavirus strains.7 At least 12 G-types (including G1-G6, G8-G10 and G12) and 15 P-types (including P-P, P-P, P and P) have been characterized in humans.7,8 It is also reported that G1-G4 genotypes are the most prevalent globally with G1P, G2P, G3P and G4P representing 88.5% of rotavirus infections in children.8 Results from clinical trials show that the efficacies of the 2 available vaccines in Africa are generally low (39–77%) compared with developed countries.7 The available data from different African countries suggests that the genetic diversity of rotavirus circulating in sub-Saharan Africa is greater than in any other continents,9 which warrants continued investigation to forecast and monitor vaccine effectiveness as vaccines are widely introduced in Africa.
RotaTeq was introduced into The Gambia Expanded Program on Immunization in August 2013. This study provides 3-year baseline prevaccine implementation data on circulating rotavirus genotypes in The Gambia and highlights the need for postvaccine genotype monitoring.
MATERIALS AND METHODS
Subjects were recruited as part of GEMS, a 3-year prospective case-control study that was conducted in the Upper River Region of The Gambia to estimate the burden and microbiologic etiology of moderate-to-severe diarrhea (MSD) in children aged 0–59 months.1,10 The MSD case definition as well as the selection of age-sex and area-matched controls are defined elsewhere.11,12 After receiving informed consent from the parents/guardians, stool samples were collected from children presenting with MSD in health centres and their matched controls. Whole stool samples were stored at −70°C.
Ethical approval was obtained from the Joint Ethics Committee of The Gambia Government and Medical Research Council Unit, The Gambia and the Institutional Review Board of University of Maryland at Baltimore, MD.
Detection of Rotavirus and Other Pathogens
The detection of rotavirus was conducted using a commercially available immunoassay kit, the ProSpecT Rotavirus EZ Microplate Assay (REMEL Inc., Lenexa, KS) as previously described.13 The isolation of bacterial pathogens (Campylobacter, Salmonella, Shigella), diarrhogenic (Escherichia coli), the detection of protozoan (Cryptosporidium spp., Entamoeba histolytica, Giardia lamblia) and viral pathogens (Adenovirus, Norovirus, Sapovirus and Astrovirus) in this study have also been described previously.13–15
Genotyping of Rotavirus
In preparation for nucleic acid extraction, 500 µL of nuclease-free water was added to 500 µL of Vertrel XF (Miller Stephenson, Sylmar, CA) before adding 100 µL of liquid stool or 100 mg of solid stool. Tubes were then vortexed for 1 minute followed by centrifugation at 13,000 rpm for 10 minutes. Then, 200 µL of supernatant was transferred to a fresh 2 ml tube and nucleic acids were extracted using the off-board lysis work flow on the automated NucliSENS easyMag system (bioMérieux, Marcy l'Etoile, France) following manufacturer’s protocol.
For reverse transcriptase polymerase chain reaction (RT-PCR), the OneStep RT-PCR kit was used (Qiagen, West Sussex, United Kingdom) following manufacturer’s instructions. Briefly, 5 µL of dsRNA was added to 20 µL of OneStep RT-PCR master mix. We used primer pairs con2/con3 and sBeg9/End9 or sBeg9/EndA for the amplification of VP4 and VP7 genes, respectively.16,17 The RT-PCR comprised a reverse transcription step at 50°C for 30 minutes, a denaturation step of 15 minutes at 95°C, followed by 35 cycles of 94°C for 30 seconds, 52°C for 45 seconds, 72°C for 45 seconds and a final extension at 72°C for 10 minutes. Amplified cDNA was separated on a 1% agarose gel stained with 500 ng/µL ethidium bromide and using Tris-borate-EDTA as running buffer at 100 V for 60 minutes.
VP4 and VP7 genotyping were performed with primer cocktails as previously described.16–18 For, VP4 the following genotypes were screened: P, P, P, P, P, P and P. For VP7, the genotypes screened were G1, G2, G3, G4, G8, G9, G10 and G12. The master mix for all reactions contained final concentrations of 0.4 µM of primers, 2 mM MgCl2, 0.75 U of Taq polymerase (Qiagen), 1X Buffer (Qiagen), 0.4 mM of each deoxynucleoside triphosphate (dNTP) and nuclease-free water made up to a total volume of 22.5 µL. To this master mix, 2.5 µL of cDNA was added per reaction. The amplification cycle involved initial denaturation at 94°C for 10 minutes, followed by 35 cycles of 94°C for 1 minute, 42°C for 1 minute and 72°C for 1 minute and a final extension at 72°C for 10 minutes. Amplicons were separated by 1.5% agarose gel electrophoresis, stained with 500 ng/µL ethidium bromide and using Tris-borate-EDTA as running buffer. Genotypes were deduced based on amplicon size (bp).16–18 This process was repeated at least twice for rotavirus-positive samples for which a genotype could not be determined.
Data Management and Statistical Analysis
GEMS data collection and management have been described previously.19 The primary outcome of the analysis was the overall prevalence of rotavirus for all the enrolled children regardless of their MSD status. Specific analyses were also conducted on rotavirus genotypes.
Two-way cross tabulations with χ2 measure of association tests were applied to describe the distribution of the outcomes across different levels of categorical variables. The rich set of demographic, epidemiological, socio-economic and clinical data collected in GEMS have been previously described.12
Further, mixed effects logistic regression (random intercept) analyses were applied to estimate odds ratios (ORs) and their 95% confidence intervals (CIs) for the associations of outcomes and potential risk factors; test for interactions; control for potential confounders; to account for correlation between matched cases and controls as well as the recruitment of subjects more than once during the study period. Age and sex were used as predefined potential confounders for all the associations tested. The associations between rotavirus and other pathogens, which were found to have attributable cases of MSD in The Gambia [Cryptosporidium, enterotoxigenic Escherichia coli producing heat stable toxin, Norovirus GII, Adenovirus and Shigella], were also investigated using this analytic method. All analyses were carried out using Stata 12.1 (StataCorp., College Station, TX) and graphs were produced in GraphPad Prism 5 (GraphPad Software, San Diego, CA). P < 0.05 has been taken to indicate statistical significance.
A total of 2598 stool samples from children 0 to 59 months of age presenting with MSD and age, sex and area-matched healthy controls were assayed for rotavirus using enzyme-linked immunosorbent assay. Overall, rotavirus was detected in 10% (247) of the children. The detection rate of rotavirus among the cases was 20% (205/1029), much higher than 3% (42/1569) among controls, and this difference was significant (P < 0.01).
Of the 247 stool samples in which rotavirus was detected by enzyme-linked immunosorbent assay, 17% (43/247) were not processed either because the stool sample was missing or there was insufficient stool for genotyping analysis. Hence, RT-PCR and multiplex genotyping were conducted on 83% (204/247) samples, complete genotypes (G-type and P-type) were obtained for 187 samples and the rest 17 had nontypable (NT) P-types and/or G-types. It is important to note that 66% (28/42) and 85% (176/205) of the rotavirus-positive stool samples were genotyped from controls and MSD cases, respectively.
Circulating Rotavirus Genotypes
The distribution of G-types and P-types detected in stool samples of children aged 0–59 months are shown in Table 1. The most prevalent G-types were G1 and G2 which together made up 84% (171/204) of the rotavirus genotypes. G4 and G9 were found in 6% (12/204) and 4% (8/204) of the rotavirus genotypes, respectively. Other less common G-types were G3 and G8 that together comprised <2% of the genotypes found. The predominant P-Types were P and P which constituted 71% (144/204) of the circulating rotavirus typed. P and P were also found frequently with detection rates of 9% (18/204) and 11% (22/204), respectively. Rare P-types P and P together had a detection rate of 1% (2/204). Overall, mixed genotypes accounted for 4% (8/204) of the rotavirus infections typed (Table 1). NT genotypes were also detected, G-type 2% (5/204) and P-type 7% (14/204); this includes 2 samples for which a genotype could not be determined, even after 2 repeats.
The most prevalent rotavirus genotype found was G2P 28% (57/204) followed by G1P 26% (52/204). An unusual genotype G1P was the third most common genotype accounting for 10% (21/204) of the genotypes. Rare genotypes were also found; G2[P14], G8P, G9P and G4P each had a detection rate of <1%. The mixed infections included G1G10P, G1G2G4P, G1G9P, G1PP, G1PP, G2G9G10P, G2PP and G4PP. G10 was not included in Table 1 as it was only detected in mixed infections.
Distribution of Rotavirus Genotypes
Although the distribution of the predominant G-types among cases and controls was similar (Fig. 1), there were some important differences in the distribution of P-types. For instance, P accounted for 30% (53/176) and 11% (3/28) of rotavirus genotypes found in cases and controls, respectively, and this difference was significant (P = 0.018). In contrast, P accounted for 25% (7/28) of the rotavirus genotypes found in controls but was less common among cases 8% (15/176), (P = 0.02). All the mixed (n = 4) and 86% (12/14) of the nontypable P-type infections were found in cases (Fig. 1).
Rotavirus detection was 25% (100/300) among children 0–11 months of age with MSD, much higher than 18% (80/375) and 14% (25/149) found among older children 12–23 and 24–59 months of age. Overall, the rotavirus detection rates were 15% (125/860), 8% (90/1004) and 6% (32/487) among the 0–11, 12–23 and 24–59 months of age categories. G1 and G2 were the dominant G-types across all age groups; accounting for >70% of the rotavirus strains genotyped Fig. 1. The detection rate of G4 was highest among older children 24–59 months, accounting for 16% (4/25) rotavirus infections compared with <10% among younger children (Fig. 1). P, P and P accounted for over 80% of the P-types found in all age groups, and P was the predominant genotype across all age groups.
Nearly two-thirds (152/247) of rotavirus detections were in January and February. Overall, 96% (237/247) of the rotavirus detections occurred during the dry months (November to May). There were no rotavirus detections in September and October that are wet months in The Gambia. G1 was the dominant G-type detected during most of the dry months (December to March; Fig. 2). However, a switch appeared to happen whereby G2 became the dominant G-type from April through to the wet months of July and August when rotavirus detections were uncommon. Similarly, P was replaced by P as the predominant and eventually the only P-type with the arrival of the wet season (Fig. 2). G4 and P that were the third most common G-types and P-types were only found in the dry months, showing a similar pattern to G1 and P.
Associations Between Rotavirus Genotypes and Demographic, Epidemiological and Socio-economic Factors
The associations between rotavirus genotypes and demographic, epidemiological and socio-economic factors were investigated using a random intercept logistic regression model and the findings are summarized in Table 2. Due to space, only factors with significant associations with rotavirus detection and/or 1 of the 2 dominant genotypes (G1P) and G2P) are shown in Table 2. The odds of rotavirus detection were significantly higher among children with MSD than among the healthy controls (OR: 18.53, 95% CI: 10.99–31.24, P < 0.01). Likewise, the dominant genotypes also had strong associations with MSD, G1P8 (OR: 37.92, 95% CI: 9.04–159.16, P < 0.01) and G2P6 (OR: 11.21, 95% CI: 4.74–26.51, P < 0.01). Season also appeared to have a strong association with rotavirus detection, with the odds increasing nearly 10-fold in the dry season (November to May) compared with the wet season (June to October; Table 2), likewise, there was a significant 5-fold increase in the odds of G2P detection in the dry season compared with the wet season.
Although rotavirus detection did not have a strong association with gender, the odds of having G1P infection was significantly higher among female children than among male children (Table 2). The odds of rotavirus detection also increased significantly with the presence of cats, cows or rodents in the compounds where the children lived (Table 2). At the genotype level, the strong association with the presence of cats and rodents in the household was consistent with the G1P but not the G2P genotype. The use of untreated drinking water and stored water were also significant factors that appeared to increase the odds of rotavirus detection (2 weeks before the diarrheal episode); these findings were also consistent with the G1P genotype specifically.
Associations Between Rotavirus and Other MSD Pathogens
Significant negative associations between rotavirus infections and 3 of the 5 pathogens that had attributable cases of MSD in The Gambia were found. There were 160 Shigella spp. infections, of which 5 were coinfections with rotavirus (OR: 0.36, 95% CI: 0.14–0.92, P = 0.03); 193 Cryptospsoridium spp. of which 9 were coinfections with rotavirus (OR: 0.40, 95% CI: 0.19–0.83, P = 0.01); and of 195 Norovirus GII infections, there were 4 coinfections with rotavirus (OR: 0.16, 95% CI: 0.06–0.47, P = 0.01). Weak negative associations with rotavirus infections were found with Adenovirus (OR: 0.39, 95% CI: 0.12–1.34, P = 0.14) and enterotoxigenic E. coli (OR: 0.63, 95% CI: 0.33–1.20, P = 0.16). Overall, it appeared that the odds of rotavirus infection decreased with the detection of another diarrhea pathogen among Gambian children.
Genotyping of over 200 rotavirus strains detected from Gambian children 0–59 months of age showed that there were at least 18 different genotypes and 17 nontypable rotavirus strains in circulation during the 3-year course of the GEMS study. Many of these genotypes are rare or unusual, consistent with previous findings across sub-Saharan Africa.5,20–32 This study not only showed that G1P and G2P are the major genotypes circulating among children in The Gambia, but also that these 2 genotypes may also have different associations with various demographic, socio-economic and epidemiological factors. The globally common genotype G4 and the recently emerged strain G933 accounted for a small proportion of the genotypes found (6% and 4%, respectively) in The Gambia. In contrast, an unusual genotype, G1 [P10] was the third most prevalent genotype which accounted for 10% of the strains found. G1P accounts for >70% of rotavirus infections in Europe, North America and Australia/Oceania, but less than a quarter of infections in Africa is attributable to this genotype.8,33
The African Rotavirus Surveillance Network (ARSN) initiated in 2006 operates in 10 countries and conducts sentinel-based surveillance of rotavirus gastroenteritis.29 The ARSN-typed 622 rotavirus strains collected from 8 sub-Sahara African countries between 2006 and 2008, and reported that G1, the globally dominant strain accounted for 22% of infections.29 A quarter of the circulating genotypes in The Gambia were G1P, consistent with the ARSN findings. G2P is classified as a globally uncommon human strain, found in 2 of 8 countries in ARSN and accounted for 4% of the genotypes. They are most likely generated by reassortment.33 However, G2P was the most common genotype (27.9%) found in The Gambia. G2P was also reported as the second most common genotype in Ghana, accounting for 13% of genotypes found.34 P was initially detected in asymptomatic neonates and thus was thought to cause natural attenuation of rotavirus genotypes.35 However, recent studies have shown a high detection rate of P in symptomatic children; 30% of genotypes in Lagos, Nigeria24 and 75% in Accra, Ghana34 and accounted for more than a quarter of infections in Africa.32,33,36,37
Understanding the genetic epidemiology of rotavirus is of great importance for disease surveillance and monitoring postvaccine implementation. Following the introduction of Rotarix in the Brazilian Expanded Program on Immunization, there were notable declines in the rates of severe rotavirus gastroenteritis but there was also a marked changed in genotype diversity.38,39 The pre-Rotarix period appeared to be characterized by high diversity of co-circulating genotypes that were then replaced by a single dominant genotype in the post-Rotarix era. In Rio de Janeiro, 1 year after vaccine implementation, 96% of rotavirus genotypes in children up to 5 years of age with gastroenteritis and dehydration were attributable to the G2P compared with 1.4% before vaccine implementation.38–41 Interestingly, another case-control study showed that Rotarix was effective against severe G2P rotavirus gastroenteritis among infants aged 6–11 months of age but not in older children, suggesting that other factors such as waning immunity may have contributed to this trend.42 Although the emergence of G2P as the predominant genotype may also reflect natural cyclic fluctuations,43,44 these findings raise concerns about the use of vaccines of limited valency and the possible vaccine-driven selection of some strains.
The detection of unusual genotypes and their occurrence in mixed infections with more common genotypes suggests that reassortment could be an important driver of increased genotypic diversity in The Gambia.45 G1P, G1P and G4P were also found co-circulating in this population and 1 child had a mixed G1, G2 and G4 infection with P specificity. Three genotypes had P or P mixed infections with the globally common P genotype. The G10 genotype frequently found in cattle and infrequently among swine, horses and sheep46 was found in 2 mixed infections with G9 and G1. A P strain with G2 specificity was also detected in a diarrheic child in this study and although this genotype has been found in children with diarrhea elsewhere, it has also been found in rabbits and goats.8 Interestingly, there is evidence from this study that the presence of a cat, cow and/or rodents in the homestead significantly increases the risk of rotavirus infection, particularly G1P8 infections. Therefore, future studies in The Gambia should investigate interspecies (animal to human and vice versa) transmission and animal reassortment which could be important drivers of the generation of high rotavirus genotypic diversity.
Besides gene reassortment and interspecies transmission, the surge in diversity and unconventional genotypes may also be explained by the acquisition of point mutations leading to genetic drift or rearrangement of the genes into coding or noncoding regions.47 This phenomenon may also account for the genotypes classified as nontypable. The application of sequencing techniques and advanced primer design can help to characterize these genotypes.48
Rotarix that is a monovalent vaccine theoretically covers more than half of the genotypes found among Gambian children, while RotaTeq theoretically has over 90% coverage. However, both vaccines confer heterotypic protection, thus the actual effectiveness of these vaccines in this region of West Africa would need to be evaluated.7 Nonetheless, both live oral vaccines have had relatively low efficacies in sub-Saharan Africa (39–77%) which could be associated with high genotypic diversity.7 RotaTeq was introduced into The Gambian Expanded Program on Immunization in August 2013 and this study provides key baseline data for the continued monitoring of rotavirus infections among children. Although RotaTeq has theoretical coverage exceeding 90% of the rotavirus genotypes found in the Upper River Region of The Gambia, the effectiveness against rotavirus gastroenteritis will need to be evaluated. Continued active surveillance and monitoring for replacement disease are critical with the application of vaccines of limited valency.
Many thanks go to the Bill and Melinda Gates Foundation for sponsoring the GEMS projects and the Medical Research Council Unit, The Gambia for facilitating the study. We acknowledge the immense contributions of World Health Organization Regional Office for Africa and the GEMS Core team. We would also like to appreciate the field, clinical and laboratory teams that worked hard on this project. Finally, we would like to thank the study participants and Beate Kampmann for critical reading of the manuscript.
1. Kotloff KL, Nataro JP, Blackwelder WC, et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case-control study. Lancet. 2013;382:209–222
2. WHO. . Rotavirus vaccines. Wkly Epidemiol Rec. 2007;82:285–296
3. Parashar UD, Burton A, Lanata C, et al. Global mortality associated with rotavirus disease among children in 2004. J Infect Dis. 2009;200(Suppl 1):S9––S15
4. Tate JE, Burton AH, Boschi-Pinto C, et al.WHO-coordinated Global Rotavirus Surveillance Network. 2008 estimate of worldwide rotavirus-associated mortality in children younger than 5 years before the introduction of universal rotavirus vaccination programmes: a systematic review and meta-analysis. Lancet Infect Dis. 2012;12:136–141
5. Armah GE, Hoshino Y, Santos N, et al. The global spread of rotavirus G10 strains: Detection in Ghanaian children hospitalized with diarrhea. J Infect Dis. 2010;202(suppl):S231–S238
6. Glass RI, Parashar UD, Bresee JS, et al. Rotavirus vaccines: current prospects and future challenges. Lancet. 2006;368:323–332
7. Patel MM, Steele D, Gentsch JR, et al. Real-world impact of rotavirus vaccination. Pediatr Infect Dis J. 2011;30(1 suppl):S1–S5
8. Santos N, Hoshino Y. Global distribution of rotavirus serotypes/genotypes and its implication for the development and implementation of an effective rotavirus vaccine. Rev Med Virol. 2005;15:29–56
9. Bányai K, László B, Duque J, et al. Systematic review of regional and temporal trends in global rotavirus strain diversity in the pre rotavirus vaccine era: insights for understanding the impact of rotavirus vaccination programs. Vaccine. 2012;30(Suppl 1):A122–A130
10. Dione MM, Ikumapayi U, Saha D, et al. Antimicrobial resistance and virulence genes of non-typhoidal Salmonella isolates in The Gambia and Senegal. J Infect Dev Ctries. 2011;5:765–775
11. Farag TH, Nasrin D, Wu Y, et al. Some epidemiologic, clinical, microbiologic, and organizational assumptions that influenced the design and performance of the Global Enteric Multicenter Study (GEMS). Clin Infect Dis. 2012;55(suppl 4):S225–S231
12. Kotloff KL, Blackwelder WC, Nasrin D, et al. The Global Enteric Multicenter Study (GEMS) of diarrheal disease in infants and young children in developing countries: epidemiologic and clinical methods of the case/control study. Clin Infect Dis. 2012;55(suppl 4):S232–S245
13. Panchalingam S, Antonio M, Hossain A, et al. Diagnostic microbiologic methods in the GEMS-1 case/control study. Clin Infect Dis. 2012;55(Suppl 4):S294–S302
14. Lindsay B, Ochieng JB, Ikumapayi UN, et al. Quantitative PCR for detection of Shigella improves ascertainment of Shigella burden in children with moderate-to-severe diarrhea in low-income countries. J Clin Microbiol. 2013;51:1740–1746
15. Lindsay B, Pop M, Antonio M, et al. Alternative Methods of Bacterial Pathogen Detection: Culture, GoldenGate(R), Universal Biosensor(R), 16S rRNA-Gene Survey. Journal of clinical microbiology. 2013
16. Gentsch JR, Glass RI, Woods P, et al. Identification of group A rotavirus gene 4 types by polymerase chain reaction. J Clin Microbiol. 1992;30:1365–1373
17. Gouvea V, Glass RI, Woods P, et al. Polymerase chain reaction amplification and typing of rotavirus nucleic acid from stool specimens. J Clin Microbiol. 1990;28:276–282
18. Iturriza-Gomara M, Green J, Brown DW, et al. Comparison of specific and random priming in the reverse transcriptase polymerase chain reaction for genotyping group A rotaviruses. J Virol Methods. 1999;78:93–103
19. Biswas K, Carty C, Horney R, et al. Data management and other logistical challenges for the GEMS: the data coordinating center perspective. Clin Infect Dis. 2012;55(suppl 4):S254–S261
20. Akran V, Peenze I, Akoua-Koffi C, et al. Molecular characterization and genotyping of human rotavirus strains in Abidjan, Cote d’Ivoire. J Infect Dis. 2010;202(suppl):S220–S224
21. Aminu M, Page NA, Ahmad AA, et al. Diversity of rotavirus VP7 and VP4 genotypes in Northwestern Nigeria. J Infect Dis. 2010;202(suppl):S198–S204
22. Armah GE, Pager CT, Asmah RH, et al. Prevalence of unusual human rotavirus strains in Ghanaian children. J Med Virol. 2001;63:67–71
23. Armah GE, Steele AD, Esona MD, et al. Diversity of rotavirus strains circulating in west Africa from 1996 to 2000. J Infect Dis. 2010;202(suppl):S64–S71
24. Audu R, Omilabu SA, de Beer M, et al. Diversity of human rotavirus VP6, VP7, and VP4 in Lagos State, Nigeria. J Health Popul Nutr. 2002;20:59–64
25. Cunliffe NA, Gondwe JS, Broadhead RL, et al. Rotavirus G and P types in children with acute diarrhea in Blantyre, Malawi, from 1997 to 1998: predominance of novel PG8 strains. J Med Virol. 1999;57:308–312
26. Cunliffe NA, Kilgore PE, Bresee JS, et al. Epidemiology of rotavirus diarrhoea in Africa: a review to assess the need for rotavirus immunization. Bull World Health Organ. 1998;76:525–537
27. Esona MD, Armah GE, Steele AD. Rotavirus VP4 and VP7 genotypes circulating in Cameroon: Identification of unusual types. J Infect Dis. 2010;202(suppl):S205–S211
28. Kiulia NM, Kamenwa R, Irimu G, et al. The epidemiology of human rotavirus associated with diarrhoea in Kenyan children: a review. J Trop Pediatr. 2008;54:401–405
29. Mwenda JM, Ntoto KM, Abebe A, et al. Burden and epidemiology of rotavirus diarrhea in selected African countries: preliminary results from the African Rotavirus Surveillance Network. J Infect Dis. 2010;202(suppl):S5–S11
30. Nokes DJ, Peenze I, Netshifhefhe L, et al. Rotavirus genetic diversity, disease association, and temporal change in hospitalized rural Kenyan children. J Infect Dis. 2010;202(suppl):S180–S186
31. Potgieter N, de Beer MC, Taylor MB, et al. Prevalence and diversity of rotavirus strains in children with acute diarrhea from rural communities in the Limpopo Province, South Africa, from 1998 to 2000. J Infect Dis. 2010;202(suppl):S148–S155
32. Sanchez-Padilla E, Grais RF, Guerin PJ, et al. Burden of disease and circulating serotypes of rotavirus infection in sub-Saharan Africa: systematic review and meta-analysis. Lancet Infect Dis. 2009;9:567–576
33. Todd S, Page NA, Duncan Steele A, et al. Rotavirus strain types circulating in Africa: review of studies published during 1997-2006. J Infect Dis. 2010;202(suppl):S34–S42
34. Binka E, Vermund SH, Armah GE. Rotavirus diarrhea among children less than 5 years of age in urban Ghana. Pediatr Infect Dis J. 2011;30:716–718
35. Flores J, Midthun K, Hoshino Y, et al. Conservation of the fourth gene among rotaviruses recovered from asymptomatic newborn infants and its possible role in attenuation. J Virol. 1986;60:972–979
36. Steele AD, Ivanoff B. Rotavirus strains circulating in Africa during 1996-1999: emergence of G9 strains and P strains. Vaccine. 2003;21:361–367
37. Cunliffe NA, Rogerson S, Dove W, et al. Detection and characterization of rotaviruses in hospitalized neonates in Blantyre, Malawi. J Clin Microbiol. 2002;40:1534–1537
38. Gurgel RQ, Cuevas LE, Vieira SC, et al. Predominance of rotavirus PG2 in a vaccinated population, Brazil. Emerg Infect Dis. 2007;13:1571–1573
39. Gurgel RQ, Correia JB, Cuevas LE. Effect of rotavirus vaccination on circulating virus strains. Lancet. 2008;371:301–302
40. Dulgheroff AC, Figueiredo EF, Moreira LP, et al. Distribution of rotavirus genotypes after vaccine introduction in the Triângulo Mineiro region of Brazil: 4-Year follow-up study. J Clin Virol. 2012;55:67–71
41. Carvalho-Costa FA, Araújo IT, Santos de Assis RM, et al. Rotavirus genotype distribution after vaccine introduction, Rio de Janeiro, Brazil. Emerg Infect Dis. 2009;15:95–97
42. Correia JB, Patel MM, Nakagomi O, et al. Effectiveness of monovalent rotavirus vaccine (Rotarix) against severe diarrhea caused by serotypically unrelated G2P strains in Brazil. J Infect Dis. 2010;201:363–369
43. Rahman M, Sultana R, Ahmed G, et al. Prevalence of G2P and G12P rotavirus, Bangladesh. Emerg Infect Dis. 2007;13:18–24
44. Amarilla A, Espinola EE, Galeano ME, et al. Rotavirus infection in the Paraguayan population from 2004 to 2005: high incidence of rotavirus strains with short electropherotype in children and adults. Med Sci monit. 2007;13:CR333–337
45. Gentsch JR, Laird AR, Bielfelt B, et al. Serotype diversity and reassortment between human and animal rotavirus strains: implications for rotavirus vaccine programs. J Infect Dis. 2005;192(suppl 1):S146–S159
46. Kapikian AZ, Hoshino Y, Chanock RM.Knipe DM, Howley R.M, Griffin D.E, et al. Rotaviruses. Fields Virology. 2001 Philadelphia Lippincott, Williams and Willkins:1787–1825
47. Kirkwood CD. Genetic and antigenic diversity of human rotaviruses: potential impact on vaccination programs. J Infect Dis. 2010;202(Suppl):S43–S48
48. van Doorn LJ, Kleter B, Hoefnagel E, et al. Detection and genotyping of human rotavirus VP4 and VP7 genes by reverse transcriptase PCR and reverse hybridization. J Clin Microbiol. 2009;47:2704–2712