Background: In a hospital-based study by Martinello (2002), specific G gene genotypes of respiratory syncytial virus subgroup A virus were associated with an increased severity of illness.
Aim: We sought to confirm the association of G genotypes with disease severity in a population-based study.
Material and methods: Ninety-one type A respiratory syncytial viruses (identified in the 1999/2000 season by polymerase chain reaction and cell culture), collected in a German multicenter study (PRI.DE) were analyzed for G gene diversity (amino acids 1-165). Disease severity was classified according to World Health Organization criteria for pneumonia in outpatients and by a bronchiolitis score (Rodriguez, 1997) in inpatients. Multiple regression analysis was used to explain disease severity.
Results: Three clusters were identified (cluster 1, n = 35; cluster 2,n = 35; cluster 3, n = 21). Sixty-seven patients had severe disease. After controlling for other variables, illness severity was significantly greater for cluster 2 viruses (odds ratio, 7.0; 95% confidence interval, 1.6 49), compared with viruses in other clusters. Other known risk factors (male gender, age) were not associated with disease severity. Our cluster 2 is genetically distinct from the virulent genotype in Martinello's study.
Discussion: Previously reported associations between G genotypes and disease severity in hospitalized patients can be generalized across the spectrum of illnesses including outpatients. The association seems not to be linked to a specific G gene structure. Rather G gene diversity in combination with the susceptibility of the host cohort may form the basis of such associations. Because of the magnitude of the effect, the underlying mechanisms warrant further investigation.