Severity of rotavirus gastroenteritis in Indian children requiring hospitalization.

INTRODUCTION The burden of rotavirus gastroenteritis is greatest in India and other developing countries. With the availability of two licensed vaccines and a number of additional vaccines in various stages of development and trial, analysis of detailed clinical information is essential for the development of a uniform method of severity assessment. METHODS Diarrhoeal stool samples from 1001 children <5 years of age hospitalized with gastroenteritis were screened for rotavirus using a commercial enzyme immunoassay. Positive samples were confirmed by genotyping using hemi nested multiplex RT-PCR. Detailed clinical data was collected for gastroenteritis assessment for 934 children and extraintestinal presentations were analyzed in 470 children. Severity scoring was carried out for all children using the Vesikari score and in a subset by Clark's scoring system. RESULTS Rotavirus was detected in 35.4% of samples tested between December 2005 and November 2008. Clark's and Vesikari scores showed moderate correlation but varied greatly in the categorization of severe disease. Using Clark's scoring, only 1.6% were categorized as presenting with severe disease in comparison to 66.1% by the Vesikari score. Association of extraintestinal symptoms with rotavirus gastroenteritis was not documented in this study. CONCLUSION The assessment of disease severity using two common severity scoring systems highlights the difference in the categorization of "severe" disease. This underscores the need for a robust scoring system which is needed for vaccine trial and in post-licensure surveillance, because vaccine efficacy is estimated for protection against severe rotavirus gastroenteritis.