Prevalence and genotypes of group A rotavirus among outpatient children under five years old with diarrhea in Beijing, China, 2011–2016

Infection with group A rotavirus is the main cause of severe diarrheal disease in children worldwide [1], and one of the common causes of death in children aged under 5 years old [2, 3]. According to World Health Organization (WHO), approximately 215,000 children aged under 5 years old died from rotavirus infection in 2013 [4]. In China, 40% of diarrhea-related hospitalizations and 30% of diarrhea-related outpatient visits among children younger than 5 years old from 1994 through 2014 were caused by group A rotavirus [5]. From 2003 to 2012, rotavirus infection contributed to 42% of diarrhea-associated death among Chinese children under 5 years old [6].

According to the sequence diversity of VP7 and VP4 genes, group A rotavirus can be divided into several G and P genotypes [7, 8]. So far, there have been 27 G genotypes and 37 P genotypes identified. Other new G and P genotypes are expected to be identified in the future [9‐12]. G3P [8], G1P [8], G2P [4] and G4P [8] have been the four dominant strains from 1989 through 2004 globally [13‐15]. Since 2004, G9P [8] was detected as one of the major epidemic strains, gradually increasing worldwide [16]. However, the major epidemic strains of group A rotavirus varies by countries and regions. Moreover, the major epidemic strains may change in different years in the same area [7, 17, 18]. In China, G1P [8] was the dominant strain from1983–2000 [19, 20]. The proportion of G3 increased dramatically from 2000 to 2007, and G3P [8] became the dominant strain [21‐23]. In 2008, G9 was reported to be a highly prevalent genotype in Xinjiang in the northwest part of China like other countries [24‐26].

There is no specific drug for the rotavirus diarrhea, but effective vaccines are available and can protect children from illness and death caused by rotavirus. In China, only the LLR (Lanzhou Institute of Biological Products, Lanzhou, China) vaccine is licensed for use, which is sourced from G10P [12] of sheep. However, this vaccine has not yet been included in the national immunization program, and there are no available clinical trial data so far. Further understanding the distribution of rotavirus strains may provide evidence for the use of vaccines in China. Previous data in China, usually based on a short-term observational studies, are limited. However, in this study we presented the results of 6-years of surveillance data to further elucidate the changing epidemiology, and thus inform targeted control measures and vaccine development.

Understanding the distribution and antigenic variation of rotavirus infection among children under 5 years old in China can inform vaccination policy in the future. Our study showed that rotavirus was responsible for 20.8% of diarrhea in outpatient children in Beijing, ranging from 16.4 to 27.3%. Children aged 1–2 years old had the highest rotavirus-positive rate (29.0%), followed by those aged 6–11 months old (24.0%), and children aged 2–3 years old (17.9%). The circulation of rotavirus was seasonal, peaking from November to January each year in Beijing. Urban children had higher rotavirus-positive rates compared with children who lived in rural areas, probably reflecting crowding and sanitation. The data also shows significantly different rotavirus-positive rates in different years. G9P [8] had become the dominant genotype in recent years. Some lineages of VP7 and VP4 gene sequences were discovered in previous years, but not in recent years.

Previous studies showed that, the rotavirus-positive rate varied among different regions and years in China. These studies showed that the rotavirus-positive rates of outpatient service ranged from 14.1 to 40.8%, with the rotavirus-positive of hospitalization was even higher in previous studies [30‐33]. Our rate was lower than in Gansu Province (66/229, 28.8%) [26], but higher than Shanghai (138/1479, 9.3%) [25].

Most children infected with rotavirus at least once before the age of 5 years old [1, 34]. A previous study [28] also presented similar results which showed children aged 6 months to 3 years old were the most vulnerable to rotavirus infection. Rotavirus-positive rate of children under 6 months is lower (9.0%), possibly due to maternal antibodies remained and less contact with the outside world.

There was a strong association between rotavirus infection in childhood and climate factors, with highest rotavirus-positive rates during the cold season [35]. For every 1 °C increase in mean temperature, the rotavirus incidence reduced by 10% [36]. Data from 11 sentinel hospitals in China from August 2003 through July 2007 [23] showed rotavirus-positive peaks in winter in all regions. Our result was consistent with this finding.

Our research found that children who lived in rural areas had lower rotavirus-positive rates compared with urban children. This result was similar to a study in Northern Cameroon [37]. Less crowding and better sanitation in rural areas may explain this difference. However, medical consultations were the highest in Beijing hospitals, two of which were in the city, one in the suburbs. This might result in sampling bias. Further, children in rural areas may have less access to hospitals or to clinics.

Epidemic strains vary over time and place in both developed and developing countries. G12P [8] was the dominant genotype in children in the United States during 2011 to 2013 [38]. G1P [8] was the most predominant genotype in Japan during 2014 to 2015 [39]. However, G12P [8], G1P [8] and G12P [6] were the most common strains in Myanmar, Sir Lanka and Nepal respectively of south-east Asian from 2009 to 2015 [40]. In China, from 2000 through 2013, G3P [8] was the major stain in Wuhan, and there was genetic evolution [14]. In our study, G3P [8] and G9P [8] were both the dominant genotypes, accounting for 24.2 (24/99) and 22.2% (22/99) respectively in 2011. However, G9 became the main genotype through these years in China [41], accounting for 56.1% (101/180) in Beijing in 2012. Our results confirmed that G9P [8] was the dominant genotype from 2013 to 2016, accounting for more than 70% of samples. Many factors could influence the rotavirus-positive rate in surveillance studies, such as socioeconomic factors [42], access to health facilities, sampling, serotype [43], density of population, vaccination rate [3], crowding, education [44], malnutrition [45], sanitation, geographic and behavioral factors [35]. Our longstanding hospital-based surveillance provided insight into trends in rotavirus infection. Some uncommon strains such as G8P [8], G2P [8], G3P [4] and G9P [4] were detected in this study, but not G12 [46, 47].

Evolutionary trees in this study were used to verify that the G and P strains were divided into different genotypes correctly. There were at least ten branches of VP7 gene sequences of G9 type, originating from people or pigs. Our study showed that G9 strains clustered into two lineages. The main lineage of G9 strains were similar to that in Japan, the United States, China and other countries. In another lineage, there were only four strains. These were similar to strains found in USA (F45) in the past. But in recent 3 years, this strain has not been circulating. The VP4 gene sequences of P [8] clustered into most P[8]b and P[8]c lineages. Only fifteen strains were P[8]c mostly before 2014, similar to the strain found in Russia (Nov10-N709). However, P[8]b was widely prevalent in Beijing in recent years. The P[8]b in Beijing was similar to that in Japan (UR14–26, To14–18), the United States (CNMC3), China (X1302, Z1557) and other countries of the world in recent years. According to previous studies [48, 49], some P[8]b strains might have occurred through reassortment between P[8]a strains and unidentified strains with the P[8]b-VP4 gene. However, P[8]a were not detected in our study.

In conclusion, our study highlights that rotavirus causes a high disease burden in children under 5 years in Beijing, with a change in predominant strains over time. At present, G9P [8] is the dominant strain. Further research is needed to explore the related factors influencing genotypes selection, and correlation between genotypes with clinical illness severity in children. The economic burden of rotavirus-positive and the cost-effectiveness of a vaccination program are also should be considered. Currently there is no specific drug for the rotavirus diarrhea, but vaccines are available and can protect children from illness and death caused by rotavirus. In 2009, WHO recommended that all countries should promote the rotavirus vaccine [50]. RotaTeq® (RV5) and Rotarix® (RV1) were included in the national immunization program in 84 countries by the end of 2015 [4], and 33 African countries by 2016 [51], but not in China. Understanding the local and national distribution of rotavirus strains is vital for targeted research and development of vaccine for China. Available rotavirus vaccination could be considered in China to prevent rotavirus infection and reduce the burden of disease [52, 53].

Rotavirus was a major cause of childhood diarrhea in Beijing during 2011 to 2016. The rotavirus-positive rate of 2012 and 2013 were higher than 2011, the dominant genotype changed from G3P [8] to G9P [8] in that 2 years. At present, G9P [8] is the dominant strain. Targeted measures, such as available rotavirus vaccination could be considered in China to prevent rotavirus infection and reduce the morbidity and mortality.