Rotavirus is reported to be the most common cause of severe childhood diarrhea worldwide.1,2 Each year, in developing nations, rotavirus infections are estimated to result in approximately 500,000 deaths in children under 5 years of age.2 By the age of 3 years, almost every child worldwide has been infected by rotavirus, although in Africa and Asia, most children have had their first rotavirus infection by 18 months of age.3,4
Infections with human rotavirus are common in the first few years of life in both industrialized and developing countries,5–7 with a spectrum of illness ranging from asymptomatic infection to severe dehydration and death. Mortality is highest in the very young, due to the rapid onset of severe dehydration in this vulnerable population.1 The incidence rates and severity of disease decline rapidly with increasing age presumably because of acquired immunity, and this has formed the basis for the live-attenuated vaccine approach.8 However, immunity due to natural rotavirus infection is incomplete, as evidenced by the common finding of a number of repeat infections and illness in infancy and early childhood.5,6 As rotaviruses are commonly found in children in industrialized countries, and essential control measures such as water, sanitation and hygiene have not reduced infection in these settings, the major control measure likely to have a significant impact on the incidence of severe disease is considered to be vaccination.1
Two rotavirus vaccines are available commercially and are being introduced into countries worldwide. Rotarix (GSK Biologicals, Rixensart, Belgium) is a monovalent, attenuated human rotavirus strain that has shown good vaccine efficacy and safety protection in Latin America, Europe and Africa.9,10 RotaTeq (Merck & Co. Whitehouse, PA) is a pentavalent, bovine human rotavirus reassortant vaccine that has demonstrated good vaccine efficacy and safety in the Unite States, Europe and Africa.11–13
Circulating rotaviruses are antigenically and genetically diverse both geographically and temporarily,14 and concerns exist whether the vaccine may not prevent all types responsible for infection and disease in children. Certainly, the strain diversity observed in Africa and in Asia is high and more variability is seen than in Europe or the United States.15,16 Thus, knowledge of the genotypic diversity of circulating rotavirus strains in many countries is an important consideration as the rotavirus vaccines are being introduced globally, and it is important to assess how effective these vaccines are against the diversity of strains in circulation in countries such as Egypt.17 It would be of most benefit to consider the circulating rotavirus strains before, during and after the introduction of the vaccine into populations.
The objectives of the present study were to determine the genotypes of circulating rotavirus strains in a cohort of Egyptian children followed from birth to 2 years of age and also to investigate if a specific genotype identified among infants in their first year of life will reduce their risk to subsequent infection of rotavirus at older ages.
MATERIALS AND METHODS
Study Subjects/Sample Collection
The study population consisted of a cohort of children living in a rural district located in the Nile Delta in northern Egypt (Abu Homos), followed from birth to 2 years of age with twice-weekly home visits to detect diarrheal illness. In January 2004, following a house-to-house census by social workers, enrollment of neonates commenced after the mother gave birth and was invited to participate in the study. Before enrollment, written informed consent was obtained from each child’s parent or guardian, and the human use guidelines of the US Department of Defense and US Department of Health and Human Services were followed throughout the study.
Active surveillance of the cohort with twice-weekly home visits began in January 2004. If the child had diarrhea, the health social workers obtained a rectal swab, which was placed in Cary-Blair transport medium, and also a stool specimen if available. Upon clinical examination by a study physician, dehydration status was classified according to the World Health Organization criteria as “none,” “some” or “severe”.18 When it was deemed necessary by the physician, symptomatic children were referred to a village clinic for further evaluation and rehydration therapy. Deaths and other losses to follow up were recorded at the twice-weekly visits. All individuals in the cohort, irrespective of symptoms, were surveyed once every 2 weeks, at which time a rectal swab placed in Cary-Blair medium and a new stool specimen were collected. At each home visit, information regarding the child’s diet and breastfeeding status was obtained.
The study period was January 2004 through April 2007. Crude diarrhea incidence rates and age-specific incidence rates were calculated by dividing the number of episodes by the number of person- years at risk (total person-years of follow up minus the duration of diarrheal episodes after the first day of each episode).
A “diarrheal day” was defined as the occurrence of at least 3 non-formed stools (or at least 1, if bloody) in a 24-hour period. A “diarrheal episode” was defined as beginning on the first diarrheal day after at least 3 consecutive nondiarrheal days and ending on the last diarrheal day to be followed by at least 3 consecutive nondiarrheal days. A “rotavirus diarrheal episode” was defined as a diarrheal episode in which rotavirus solely was detected in a fecal specimen collected during the episode with the absence of other enteric pathogens. All rectal swabs and fecal specimens were transported in ice packs to the Abu Homos field laboratory, where they were refrigerated. Twice weekly, the rectal swabs and fecal specimens were transported in ice packs to the laboratories of US Naval Medical Research Unit 3 in Cairo, where the rectal swabs were evaluated with standard microbiologic methods for bacterial and parasitic pathogens.19
For rotavirus, the fecal specimens were kept frozen at −70°C until processed for enzyme immunoassay and polymerase chain reaction (PCR). For enzyme immunoassay, stool suspensions were prepared as 10% (weight/volume) phosphate-buffered saline (pH 7.2) extracts and were tested for rotavirus using a commercial enzyme-linked, immune-sorbent assay kit (Rotaclone; Meridian Diagnostics, Inc., Cincinnati, OH).
Rotavirus RNA Extraction/Genotyping
Viral genomic RNA of stool specimens positive for rotavirus was extracted from a 10% stool suspension (0.1 g or 100 µL stool in 2 mL of a 1:1 Vertrel/Water solution, v/v) using the QIAamp Viral RNA mini kit according to the manufacturer’s instruction (QIAgen, Valencia, VA). Following RNA extraction, VP4 and VP7 genes were amplified by classical reverse-transcriptase, nested, multiplex procedures (reverse transcriptase PCR) as described elsewhere,17 using well-established primers sets specific for the different G-type or P-type specific primers.20,21 Agarose gel electrophoresis and ethidium bromide staining were performed to visualize resulting bands.17
For untypeable strains (UTs), the VP7 and VP4 genes were amplified by reverse-transcription polymerase chain reaction (reverse transcriptase PCR) and sequencing of each strand of the 2 genes using the same primer as those used for reverse transcriptase PCR. PCR-amplified products of both VP7 and VP4 genes were purified with the QIAquick PCR purification kit following the manufacturer’s protocol. Direct sequencing of each amplicon was carried out using the dideoxynucleotide chain termination method with the ABI Prism Big Dye Terminator cycle sequencing Reaction kit (Applied Biosystems, Inc., Foster city, CA) on an ABI prism 3100 automated sequencer. Nucleotides sequences were aligned against rotavirus sequences obtained from Genbank using BioEdit software, version 5.1 to ascribe a genotype.
Study Population and Rotavirus Disease
During the study period, 348 newborns enrolled in the cohort experienced 4001 diarrheal episodes with 511 persons-years of follow up. A rectal swab was obtained from every diarrheal episode. Of the 4001 episodes of diarrhea observed in the children during the study, 226 episodes were due to rotavirus; of which 169 episodes, rotavirus was the sole etiologic agent. Additionally, 57 children that were not included for genotyping in this study had rotavirus plus 1 or 2 enteric bacterial pathogen (eg, enterotoxigenic E coli, Shigella spp. or Campylobacter spp.). The overall incidence rate of rotavirus was 0.30 episodes per person-year, while the age-specific incidence rates of rotavirus diarrheal episodes per person-years were 0.22 for infants <6 months, 0.26 for those aged 6–11 months, 0.39 for those aged 12–17 months and 0.32 for children >18 months (Table 1).
Overall, 40% (140/348) of children suffered from a symptomatic rotavirus infection where rotavirus was the sole pathogen in this community setting. A total of 169 symptomatic rotavirus diarrhea episodes (RDE) were identified in 140 children. One hundred and twelve children (80%) experienced a single RDE at a median age of 9 months of age [interquartile range (IQR); 5–15]; while 27 infants (19%) had their second RDE at a median age of 15 months (IQR; 10–20) and 1 infant suffered from 3 RDE at the age of 2, 16 and 22 months. Forty-three episodes were detected in 42 infants < 6 months of age, 56 episodes were from 49 children in the next 6 months of life and the remaining 70 episodes were from 49 children with age >1 year (Table 1). In the 28 cases of sequential rotavirus infections, 71% (20/27) of these episodes occurred in children >1 year, while 25% (7/28) and 4% (1/28) were identified among children in the age group from 6 to 11 months and infants <6 month, respectively. The duration between the first and second episode of rotavirus infection ranged from 1 to 17 months, with a mean of 8 months (data not shown).
Rotavirus Genotyping of the Strains
Of the 164 rotavirus diarrheal episodes evaluated, 135 (82%) could be assigned a G-type, while the remaining 29 (18%) could not be typed (5 samples were not assayed for G-type due to lack of material; Table 1). Only 95 (58%) of the 164 strains could be P-type and the remaining 69 strains (42%) were untypeable for the P genotype. Overall, 17 samples were untypeable for both G- and P-types.
Of the typed strains, G2 was the most frequently identified G genotype (32%), followed by G1 strains (24%) and G9 strains (19%; Table 2). Each of these 3 predominant strains were associated with differing P-types (Table 2). Less common types (G3, G2+G3, G2+G9 and G12) represented 7% of all G genotypes.
Rotavirus Strain and Clinical Features
There was a trend in the age distribution of the common genotypes. For instance, 26% of infants <6 months were exposed to G2 rotavirus genotype infection, while a smaller proportion of cases with G1 (10%) and G9 (14%) type-associated rotavirus were observed in this age group. In the second year of life, 43% of strains among children 12–17 months were G9 (Table 3).
While fever and vomiting characterized most of the episodes regardless of the rotavirus genotype, 76% of the G2P rotavirus episodes were significantly associated with fever (P = 0.03) and 71% were associated with vomiting (P = 0.06) when compared with G1P genotypes (Table 4). No dehydration was reported for any rotavirus cases, except for 2 cases with G2P (2/21 cases) in the age group of 6–11 and 12–17 months.
There was no significant difference in the maximum number of stools per child/day associated with G1-type episode (median=7, IQR: 4–9) and those caused by G2 (median = 5, IQR: 4–7) or G9 episodes (median= 7; IQR: 5–8).
The circulation of rotavirus was year-round, with a higher number of monthly cases of rotavirus diarrhea occurring around the fall season (September to November; Fig. 1). Forty percent (n = 16) of G1 episodes occurred in the month of September, while 31% of G2 and 33% of G9 episodes peaked at the month of November (Fig. 1).
Reinfection by Strain
Rotavirus G2 genotype was the predominant genotype causing 50% (14/28) of the second rotavirus episodes among children with median age 18 months (IQR; 9–20); 43% (6/14) of these episodes were associated with fever and vomiting. No difference in the duration of illness or associated symptoms between the first and second episodes was related to any specific genotype. Twenty-five percentage (7/28) of the second episodes were due to G9 genotype, and together with the G2 genotype, they comprised 75% of sequential RDE, 14% (4/28) were due to G1 and 11% (2/28) were of UT type. Twenty one percent (3/14) of G2, 25% (1/4) of G1 and 17% (1/7) of G9 cases had the same genotype in their first episode (Table 5). The mean duration between the first and second episodes due to G1 and G9 was 1 and 7 months, respectively, while the duration was 10 months between those caused by G2. Overall, 57% (16/28) of the second episode presented with fever, 39% with vomiting, 36% with both fever and vomiting and 4% with dehydration. There was no difference in the disease symptoms associated with the first and second episodes.
This is second largest reported prospective study of rotavirus disease from Egypt and the middle-East region in young children followed from birth, where 20% (28/140) of children presented with 2 rotavirus illnesses as opposed to 3% (12/363) reported in the early study from Egypt.7 In this study, we report the distribution of rotavirus genotypes isolated from diarrheal episodes in a cohort of 348 Egyptian children followed from birth to 2 years of age during 2004–2007. Forty percentage of the cohort suffered at least 1 rotavirus diarrhea episode with higher frequencies (65%) in children <1 year when compared with those older (35%). This figure is comparable with the 40% reported in another Egyptian study conducted 2 decades ago in infants <1 year of age,22 but higher than the 23% reported in the Abu Homos clinic-based surveillance conducted during 2000–2002.23 The high distribution of rotavirus infection in the younger age group in this study, predominantly in infants <12 months of age, is consistent with that reported earlier in Egypt22 and those reported in other Mediterranean areas and countries, such as Sicily24 and Italy.25
Rotavirus-associated diarrhea peaked in late summer (September) to late fall (November), similar to the trend reported in a population-based cohort study conducted in Abu Homos, Egypt, during 1995–1998, where 90% of rotavirus diarrheal episodes occurred between July and November.26 In a much earlier community-based study conducted in Egypt 2 decades before, rotavirus infection peaked from November to April,27 which is similar to the pattern seen in other North African countries during the same time period, such as Morocco,28 Algeria29 and Egypt.30 This apparent change in seasonality is interesting and may reflect subtle changes in various factors such as economic conditions, urbanization and population growth or other factors.
In the current cohort from Abu Homos, genotyping of the VP7 gene showed the presence of diverse rotavirus genotypes with G2 as the most frequently genotype (32%), followed by G1 (24%) and G9 (18%). In this population, G2 rotavirus infection was found to occur in 26% (12/47) in infants during their first 6 months of life, while G1 was prevalent [45% (18/40)] in the next 6 months of life and G9 [43% (12/28)] dominates in the 12–17 months of life. The predominance of G2 genotype in the current study demonstrates the variability of rotavirus strains in Egypt over the course of time when compared with earlier studies in Cairo22 and Abu Homos.26 During 1992–1993, Radwan22 identified G1 (17%), G4 (18%) and mixed genotypes (16%) as the most common genotypes in Cairo. After 2 years, Naficy26 reported genotypes G1 and G2 with a combined prevalence of 89% in Abu Homos.
This study confirms the earlier report that 17% of 272 children in Egypt experienced >1 symptomatic rotavirus episode.26 In addition, this study clearly identifies the introduction and emergence of G9 strains as an important type circulating in Egypt, and the absence of G4 strains that circulated previously—both consistent with global trends of strain diversity and distribution.
Human G9 strains have been identified on all human inhabited continents after being first described in 1983 in children with gastroenteritis in Philadelphia.31 The first report of the G9 strains causing disease in older children and severe disease was documented in the United Kingdom and Nigeria.32,33 Most strain surveillance, however, is hospital based, with few studies describing the distribution of strains in the community. A prospective and community-based study conducted on children from Latin America indicated the high incidence of G9 strains in the community with a broad spectrum of rotavirus disease severity as measured by wide range of numerical scores from 2 to 20 on the Vesikari clinical score.34 As a newly emerging human pathogen, better understanding of the association between disease severity and G9 strains should be elucidated from hospital-based studies.
To our knowledge, G9 was not detected in a recent hospital-based study examining enteric viruses conducted in Cairo,35 nor in an earlier cohort of children assembled from 2 villages near Abu Homos during 1995–1996.30 The first description of G9 strains in Egypt in 2002 indicate that this strain has also emerged in Egypt as in other locations.17 The emergence of new rotavirus strains exhibits the dynamic evolution of human rotaviruses and may provide challenges for the efficacy and design of new rotavirus vaccines worldwide.
In this study, a few apparently “reassortant” viruses were identified, which carried a G1P background. Most human rotaviruses show a strong association of the G1 serotype with a P VP4 genotype, and the P genotype is normally determined in viruses with a G2 VP7 serotype.36 Reassortments between naturally cocirculating strains that coinfect 1 host, while apparently rare, have been recorded to occur and drive the molecular evolution of “new” strains.37 The high rate of mixed rotavirus infection identified in the current study is considered a prerequisite for the occurrence of rotavirus reassortment events, that may lead to the appearance of new strains and new variants. The increased diversity in rotavirus strains and the high rate of mixed infections have defined a profile characteristic of developing countries, as previously described in studies conducted in Tunisia, Brazil and Bangladesh.38–40
It was possible to determine age-specific genotype variation in the second episode of rotavirus diarrhea, where 50% of those attacking children > 1 year. Age-specific variation in the pathogenesis has been reported.41 The observation that the same genotype was only present in few numbers of the second episodes differs from the first study conducted from Egypt7 and indicates a possible role of genotype-specific natural immunity.
Some limitations of this study include the lack of serum antibody data to help indicate genotype-specific immunity developed from the first rotavirus episodes. In addition, rotavirus cases were detected based on testing of stool specimens by enzyme immunoassay, which although widely used, may have missed low levels of rotavirus excretion. The fact that we only used passive recruitment of diarrhea cases may mean that we missed occurrences of rotavirus excretion as was observed in a recent study in India.42 Finally, the current study did not include and genotype the RDE that were coinfected with other enteric copathogen (in the first and second RDE) to avoid confusion with causality when another pathogen was involved. We had a large number of P untypeable strains (42%). However, this was not surprising since the big bulk of these strains (n = 52) have identified G type and most probably we expect that these strains may possess variants that were not picked due to primer mismatch with the majority belonging to G2 and G9.
In view of the common occurrence of multiple infections caused by G2 and G9 types in sequential rotavirus infection, G1 infections may be associated with the development of a greater degree of immunity to subsequent other rotavirus strains. The high rate of occurrence of these globally spread genotype infections early in life supports the high value of introducing current vaccines that are composed of these genotypes into the national immunization schedule.
The authors are grateful to CAPT Jesse Monestersky for his review and critique of this article.
1. WHO. . Rotavirus vaccine position paper. Wkly Epidemiol Rec. 2007:285–96
2. 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
3. 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
4. Kang G, Arora R, Chitambar SD, et al.Indian Rotavirus Strain Surveillance Network. Multicenter, hospital-based surveillance of rotavirus disease and strains among indian children aged <5 years. J Infect Dis. 2009;200(suppl 1):S147–S153
5. CDC. . Rotavirus surveillance - worldwide 2009. MMWR Morb Mortal Wkly Rep. 2011;60:514–516
6. Velázquez FR, Matson DO, Calva JJ, et al. Rotavirus infections in infants as protection against subsequent infections. N Engl J Med. 1996;335:1022–1028
7. Reves RR, Hossain MM, Midthun K, et al. An observational study of naturally acquired immunity to rotaviral diarrhea in a cohort of 363 Egyptian children. Calculation of risk for second episodes using age-specific person-years of observation. Am J Epidemiol. 1989;130:981–988
8. Kapikian AZ, Hoshino Y, Chanock RFields BN, Knipe DM, Howley PM. Rotaviruses. Fields Virology. 19964th ed Philadelphia Lippincott-Raven Press:1787–833. –1
9. Ruiz-Palacios GM, Pérez-Schael I, Velázquez FR, et al.Human Rotavirus Vaccine Study Group. Safety and efficacy of an attenuated vaccine against severe rotavirus gastroenteritis. N Engl J Med. 2006;354:11–22
10. Madhi SA, Cunliffe NA, Steele D, et al. Effect of human rotavirus vaccine on severe diarrhea in African infants. N Engl J Med. 2010;362:289–298
11. Vesikari T, Matson DO, Dennehy P, et al.Rotavirus Efficacy and Safety Trial (REST) Study Team. Safety and efficacy of a pentavalent human-bovine (WC3) reassortant rotavirus vaccine. N Engl J Med. 2006;354:23–33
12. Armah GE, Sow SO, Breiman RF, et al. Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants in developing countries in sub-Saharan Africa: a randomised, double-blinded, placebo-controlled trial. Lancet. 2011;9741:606–14
13. Zaman K, Dang DA, Victor JC, et al. Efficacy of pentavalent rotavirus vaccine against severe rotavirus gastroenteritis in infants in developing countries in Asia: a randomised, double-blind, placebo-controlled trial. Lancet. 2010;376:615–623
14. 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
15. 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
16. Miles M, Lewis K, Kang G, Parashar UD, Steele AD. A systematic review of rotavirus strain diversity in India, Bangladesh, and Pakistan. Vaccine. 2012;13(suppl 1):A131–139
17. Matson DO, Abdel-Messih IA, Schlett CD, et al. Rotavirus genotypes among hospitalized children in Egypt, 2000-2002. J Infect Dis. 2010;202(suppl):S263–S265
18. WHO. Programme for Control of Diarrhoeal Diseases. A Manual for the Treatment of Diarrhea for Use by Physicians and Other Senior Health Workers. 1990 Geneva, Switzerland WHO/CDD/SER/80.2 rev. 2
19. Murray PR, Baron EJ, Pfaller MA, et al. Manual of clinical microbiology. 19956th ed Washington, DC American Society for Microbiology Press
20. 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
21. 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
22. Radwan SF, Gabr MK, El-Maraghi S, et al. Serotyping of group A rotaviruses in Egyptian neonates and infants less than 1 year old with acute diarrhea. J Clin Microbiol. 1997;35:2996–2998
23. Wierzba TF, Abdel-Messih IA, Abu-Elyazeed R, et al. Clinic-based surveillance for bacterial- and rotavirus-associated diarrhea in Egyptian children. Am J Trop Med Hyg. 2006;74:148–153
24. Arista S, Vizzi E, Ferraro D, et al. Distribution of VP7 serotypes and VP4 genotypes among rotavirus strains recovered from Italian children with diarrhea. Arch Virol. 1997;142:2065–2071
25. Cascio A, Vizzi E, Alaimo C, et al. Rotavirus gastroenteritis in Italian children: can severity of symptoms be related to the infecting virus? Clin Infect Dis. 2001;32:1126–1132
26. Naficy AB, Abu-Elyazeed R, Holmes JL, et al. Epidemiology of rotavirus diarrhea in Egyptian children and implications for disease control. Am J Epidemiol. 1999;150:770–777
27. Zaki AM, DuPont HL, el Alamy MA, et al. The detection of enteropathogens in acute diarrhea in a family cohort population in rural Egypt. Am J Trop Med Hyg. 1986;35:1013–1022
28. Tazi-Lakhsassi L, Garbarg-Chenon A, Nicolas JC, et al. Epidemiological and clinical study and electrophoretyping survey of rotavirus acute diarrhoea in a children’s infectious disease unit in Casablanca, Morocco. Ann Inst Pasteur Virol. 1988;139:205–215
29. Tchambaz M, Messaoudi Z, Meziane O, et al. [Detection of rotavirus in the stools of infants aged 0-3 yr (study performed from July 1987 to May 1989)]. Arch Inst Pasteur Alger. 1989;57:83–103
30. el-Mougi M, Amer A, el-Abhar A, et al. Epidemiological and clinical features of rotavirus associated acute infantile diarrhoea in Cairo, Egypt. J Trop Pediatr. 1989;35:230–233
31. Clark HF, Hoshino Y, Bell LM, et al. Rotavirus isolate WI61 representing a presumptive new human serotype. J Clin Microbiol. 1987;25:1757–1762
32. Cubitt WD, Steele AD, Iturriza M. Characterisation of rotaviruses from children treated at a London hospital during 1996: emergence of strains G9P2A and G3P2A. J Med Virol. 2000;61:150–154
33. Steele AD, Nimzing L, Peenze I, et al. Circulation of the novel G9 and G8 rotavirus strains in Nigeria in 1998/1999. J Med Virol. 2002;67:608–612
34. Ruuska and Vesikari T. Rotavirus disease in Finnish children: use of numerical scores for clinical severity of diarrhoeal episodes. Scand J Infect Dis. 1990;22:259–267
35. Kamel AH, Ali MA, El-Nady HG, et al. Predominance and circulation of enteric viruses in the region of greater Cairo, Egypt. J Clin Microbiol. 2009;47:1037–45
36. Gentsch JR, Woods PA, Ramachandran M, et al. Review of G and P typing results from a global collection of rotavirus strains: implications for vaccine development. J Infect Dis. 1996;174(suppl 1):S30–S36
37. Iturriza-Gómara M, Isherwood B, Desselberger U, et al. Reassortment in vivo: driving force for diversity of human rotavirus strains isolated in the United Kingdom between 1995 and 1999. J Virol. 2001;75:3696–3705
38. Chouikha A, Fodha I, Bouslama L, et al. Emergence and characterization of human rotavirus g9 strains in Tunisia. J Infect Dis. 2009;200(suppl 1):S239–S243
39. Leite JP, Alfieri AA, Woods PA, et al. Rotavirus G and P types circulating in Brazil: characterization by RT-PCR, probe hybridization, and sequence analysis. Arch Virol. 1996;141:2365–2374
40. Rahman M, Matthijnssens J, Goegebuer T, et al. Predominance of rotavirus G9 genotype in children hospitalized for rotavirus gastroenteritis in Belgium during 1999-2003. J Clin Virol. 2005;33:1–6
41. Mata L, Simhon A, Urrutia JJ, et al. Epidemiology of rotaviruses in a cohort of 45 Guatamalan Mayan Indian children observed from birth to the age of three years. J Infect Dis. 1983;148:452–461
42. Beryl P, Gladstone BP, Sasirekha R, et al. Protective effect of natural rotavirus infection in an Indian birth cohort. NEJM. 2011;365:337–346.