Rotavirus (RV) infection remains a significant global public health concern, particularly among infants and young children, accounting for substantial morbidity and mortality worldwide (1). In China, the burden of RV-related diseases has been a notable challenge, with RV-associated hospitalizations accounting for 32%–50% of all hospitalizations for diarrhea among infants and children<5 years of age (2, 3).

RVs are known for their genetic diversity, characterized by distinct combinations of G (glycoprotein) and P (protease-sensitive) genotypes. These variations in genotypes have implications for both the severity of illness and vaccine effectiveness, as certain RV genotypes may exhibit differential susceptibility to vaccine-induced immunity.

Vaccination is one of the most effective ways to prevent RV gastroenteritis (RVGE). In 2009, WHO Strategic Advisory Group of Experts recommended the employment of routine RV vaccination in expanded program on immunization (EPI) program for infants worldwide, with a particular emphasis in developing countries with high disease burden (4, 5).

Currently, two RV vaccines have been licensed in mainland of China: the pentavalent RotaTeq (RV5, Merck & Co., USA) and the monovalent RV vaccine (Lanzhou Institute of Biological Products Co., China). RV5, approved by the National Drug Administration on April 12, 2018, is indicated for infants aged 6 to 32 weeks. This vaccine is designed to protect against RVGE caused by serotypes G1, G2, G3, G4, and P1A[8] (6). The monovalent RV vaccine, manufactured by Lanzhou Institute, was introduced in China in 2001.However, the post-vaccination prevalence and genetic diversity of RV among children under 5 years in China continue to be of substantial scientific interest.

To address this, we performed a comprehensive meta-analysis of studies published from 2019 to 2023. This analysis was aimed at monitoring the shifts in RV genotypes in the Chinese pediatric population following the introduction of RV vaccine. Our hypothesis was that the introduction of RV vaccine would lead to a shift in the predominant RV genotypes.

To the best of our knowledge, this represents the first meta-analytical exploration of RV prevalence and genotypic diversity in the post-RV vaccine era among children under the age of five in China. Our analysis reveals a RV prevalence of 19.00% in this demographic, with a noted predilection towards male children in RV-associated diarrhea cases. A gradation in positivity rates was observed across age groups, with the highest in the 12-23 month age group, followed by 24-35 months and 6-11 months. The serotype G9 emerged as the predominant RV genotype, comprising nearly 90% of infections, while P[8] was the principal P type, accounting for 64.02% of cases. In the context of G-P genotype combinations, G9P[8] was implicated in approximately 80% of RV cases, superseding other combinations such as G8P[8] and G3P[8].

Our study’s finding of a 19.00% prevalence of RV post-RV vaccine implementation contrasts with higher prevalence rates reported in previous studies. A comprehensive analysis encompassing 69 studies between 2011 and 2018 indicated an RV positivity rate of 34.0% (95%CI: 31.3%-36.8%) in inpatient settings and 23.9% (95%CI: 18.1%-30.9%) in outpatient settings, respectively (32). This is higher compared to our findings and even exceeds the 24% RV prevalence rate (95%CI: 22%-26%) observed in a South African meta-analysis of 43 studies conducted over almost four decades (1982-2020) (33). Similar trends of higher RV infection rates have been reported elsewhere. A meta-analysis in Ethiopia, comprising 11 studies from 1981 to 2018, noted a pooled prevalence of 23% (34). In Japan, a 33.0% RV positivity rate was reported three years post-RV vaccine introduction (2014-2015) (35). Additionally, a study in Central Kenya, pre-RV vaccine introduction (2009-2014), detected RV in 27.5% of fecal specimens from children under five with acute gastroenteritis (36). This discrepancy could stem from several factors, including our study period (2019-2023) coinciding with the COVID-19 pandemic and its associated quarantine measures, which likely impacted RV transmission dynamics. Furthermore, varying disease burdens, diagnostic assay sensitivity, and population selection and characterization across studies could contribute to these differences. Additionally, the escalating RV vaccine coverage, as evidenced in a birth cohort study from Shanghai’s Minhang district, might have played a crucial role in diminishing RV prevalence. This study demonstrated a rise in three-dose RV vaccine coverage from 9.0% in the 2013 birth cohort to 29.8% in the 2020 cohort (trend χ² = 37.18, P=0.001) (37).

Our findings underscore a heightened incidence of RV-associated diarrhea predominantly in children aged between 6 and 35 months, with a peak positivity in the 12-23 month age group and a notably lower incidence in infants under 6 months. These findings align with prior meta-analyses conducted in China, encompassing studies from 2011 to 2018 (32), and earlier research before 2011 (38). The earlier study delineated the highest RV positivity rate at 40.1% (95%CI: 33.4%-47.2%) in children aged 12 to 23 months, and the lowest at 17.7% in those aged 0 to 6 months (38). Similarly, the subsequent study reported a peak RV positivity rate of 42.7% (95% CI: 39.3%-46.2%) in the 12-23 month cohort, contrasted with the lowest rate of 17.3% (95% CI: 14.6%-20.1%) in children aged 48-59 months (38). Our results corroborates the WHO recommendation for administering the RV vaccine prior to the age of two, reflecting the age-specific vulnerability to RV infection (39).

In this meta-analysis, the G9 genotype emerged as the predominant strain in RV infections, accounting for approximately 90% of cases. This contrasts with earlier meta-analyses that included studies from China prior to 2019 (32, 40). Our 2014 meta-analysis, incorporating 93 studies published between 1994 and 2012, aimed to evaluate the diversity and temporal shifts in RV strains in Chinese children under 5 years (40). In that analysis, 22,112 and 10,660 RV samples were examined for G and P types, respectively (40). The most frequent G types identified were G1 (39.5%) and G3 (35.6%), with G2 (1.3%) and G9 (0.1%) being less prevalent (40). Another meta-analysis covering 2011-2018, reported G3 (26.1%), G9 (17.5%), and G1 (12.8%) as the most common G types in a similar demographic (32).

These trends suggest a shift in the dominant RV strains over time. For instance, the proportion of G9 escalated from 0.1% during 1994-2012 to 17.5% between 2011-2018, and further to 85.48% in 2019-2020. Conversely, the prevalence of G1 reduced from 39.5% in 1994-2012 to 12.85% in 2011-2018, and then to 4.86% in 2019-2020. Despite these variations, the serotypes G1, G2, G3, G4, and G9 have remained the most common RV strains in recent decades. Notably, our meta-analysis did not identify any G4 strains, as only two studies (22, 29) reported G-P combinations including G4, but without segregating G4 data for separate analysis. Hence, the prevalence of the G4 serotype post-RV vaccine implementation remains unanalyzed.

This genotypic shift in RV prevalence is not unique to China. A similar trend was observed in South Africa, as detailed by Omatola, CA, et al. (41), who analyzed 43 studies from 1982 to 2020 to assess RV infection prevalence and genotypic changes pre- and post-vaccine introduction in children under 5. Pre-vaccine, the predominant G genotypes were G1 (48%), followed by G2 (19%) and G3 (12%). Post-vaccine introduction, G2 (27%) and G9 (25%) emerged as the dominant strains, with G12 (15%) following (41). Comparable shifts towards G2, G9, and G12 genotypes in the post-vaccine era have also been reported in Australia (42) and Zambia (43).

Recent studies have observed an increased prevalence of RV genotypes not represented in vaccine formulations following the integration of RV vaccines into national immunization programs (44, 45). A meta-analysis conducted in South Africa (41) highlighted a significant decline in the G1P[8] genotype (from 43.13% to 11.53%) and a concurrent rise in the G9P[8] (23.22%), G2P[4] (21.2%), and G12P[8] (14.3%) genotypes in the vaccine era. These findings suggest that the introduction of the RV5 vaccine may exert selective pressure on circulating strains, potentially favoring the emergence of less dominant or mutant strains not effectively neutralized by the vaccine. A parallel increase in G2P[4], G9P[8], and G12P[8] genotypes, along with other previously rare genotypes not covered by the monovalent Rotarix vaccine (which targets the G1P[8] strain), has been noted in meta-analyses from Ethiopia and Europe (34, 46).

While some researchers attribute the shifting genotype distribution to inadequate protection against heterologous and newly emerging RV strains, thereby facilitating strain selection due to vaccine-induced immunological pressures (42, 47), others argue that natural fluctuations in strains or gene reassortment events are more likely contributors to the emergence and epidemiological viability of these variants, particularly in contexts of limited herd immunity (48, 49). Despite evidence indicating that the RV vaccine provides both homotypic and heterotypic immunity (50), ongoing national epidemiological surveillance of RV strains in China remains imperative. This is particularly relevant considering the increasing prevalence of the G9 strain, which is not included in RV5, and the observed reduction in genotypes covered by RV5 (G1, G2, G3, and G4) post-implementation. Future research should aim to elucidate the impact of these shifts in RV genotype distribution on the effectiveness of RV5 against RVGE.

This meta-analysis is subject to several limitations. Firstly, the diversity in study design, geographical regions, and sample size among the included studies may contribute to heterogeneity in our results. Additionally, variability in serotyping and genotyping methodologies across these studies could introduce heterogeneity and potentially affect the precision of our findings, albeit less likely the general trends. Secondly, many of the studies predominantly focus on the more common G and P types, specifically G1-G3, G8, G9, P[4], P[6], and P[8]. This focus might lead to misclassification or inability to type less common types like G8 and G12, thereby limiting our comprehensive understanding of the RV strain landscape in mainland China. Thirdly, the analysis encompasses data from studies published between 2019 and 2023. Consequently, ongoing surveillance will be crucial for assessing the long-term variations in RV strains in the post-RV vaccine era.

In conclusion, while this meta-analysis offers an extensive evaluation of RV prevalence and genotypic diversity in Chinese children following the implementation of the RV vaccine, it underscores the necessity for continued national epidemiological monitoring. Such sustained surveillance is pivotal not only for enhancing future vaccine effectiveness but also for understanding the reasons behind any observed vaccine failures.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

YL: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. SW: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Supervision, Writing – original draft. FL: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Supervision, Writing – original draft. ST: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – review & editing. FW: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Supervision, Validation, Writing – original draft, Writing – review & editing.