Background: : Antigenemia is common among children with rotavirus disease. Because obtaining stool specimens is cumbersome, we evaluated whether detection of antigenemia in sera obtained during routine clinical practice could augment rotavirus surveillance to assess the effect of vaccination.
Methods: : We determined the sensitivity, specificity, and positive and negative predictive values of serum/plasma rotavirus antigen detection using fecal antigen positivity as the gold standard. Fecal specimens obtained by active surveillance and residual serum/plasma specimens obtained during routine clinical testing from children 15 days to 23 months of age presenting with acute gastroenteritis (AGE) to a children's hospital in Houston were tested for rotavirus using a commercially available enzyme immunoassay. Using case-control methods, we compared vaccine effectiveness (VE) using cases identified through serum/plasma testing versus stool testing.
Results: : Of the 205 AGE patients with fecal specimens, 71 (35%) had a serum/plasma sample available. Among these 71 children, antigenemia was detected in 22 of 29 with rotavirus-positive fecal specimens (sensitivity = 75%; 95% confidence interval [CI] = 60%-91%) versus 2 of 42 children with rotavirus-negative fecal specimens (specificity = 95%; 95% CI = 89%-100%). The positive and negative predictive values of rotavirus antigenemia were 92% (95% CI = 81%-100%) and 85% (95% CI = 75%-95%), respectively. Thirty-four of 195 children with AGE without fecal specimens had serum/plasma available; 10 (29%) had rotavirus antigenemia. Three-dose VE using cases identified through serum/plasma testing was similar (VE = 84%; 95% CI = 25%-96%) to that using cases identified though fecal testing (VE = 85%; 95% CI = 55%-95%).
Conclusions: : Detection of antigenemia in routinely collected serum/plasma could augment identification of rotavirus disease for postlicensure evaluation of impact and effectiveness of rotavirus vaccination.