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.

Recently, two rotavirus vaccines have been recommended for routine immunization of infants worldwide. These vaccines proved efficacious during clinical trials and field use in both developing and developed countries, and appear to provide good protection against a range of rotavirus genotypes, including some that are not included in the vaccines. However, since conclusive data that the vaccines will protect against a wide variety of rotavirus strains are still lacking and since vaccines may exert some selection pressure, a detailed picture of global strain prevalence from the pre-rotavirus vaccine era is important to evaluate any potential changes in circulating strains observed after widespread introduction of rotavirus vaccines. Thus, we systematically reviewed rotavirus genotyping studies spanning a 12-year period from 1996 to 2007. In total, ~110,000 strains were genotyped from 100 reporting countries. Five genotypes (G1-G4, and G9) accounted for 88% of all strains, although extensive geographic and temporal differences were observed. For example, the prevalence of G1 strains declined from 2000 onward, while G3 strains re-emerged, and G9 and G12 strains emerged during the same period. When crude strain prevalence data were weighted by region based on the region's contribution to global rotavirus mortality, the importance of genotypes G1 and G9 strains that were more prevalent in regions with low mortality was reduced and conversely the importance of G8 strains that were more prevalent in African settings with greater contribution to global rotavirus mortality was increased. This study provides the most comprehensive, up-to-date information on rotavirus strain surveillance in the pre-rotavirus vaccine era and will provide useful background to examine the impact of rotavirus vaccine introduction on future strain prevalence.