All authors: No reported conflicts. children without CRS gave 100% specificity with 65% sensitivity. Conclusions This study was the first to establish classification rules for identifying CRS in school-aged children, using laboratory biomarkers. These biomarkers should allow improved burden of disease estimates and monitoring of CRS control programs. values of .05 were considered statistically significant. To evaluate the ability of different biomarkers to differentiate children with from children without CRS, we calculated receiver operating characteristic (ROC) curves [26], which plot sensitivity against [1 C specificity] for every cutoff in the range of the observed biomarker values. We calculated the area under the ROC curve (AUC), which is considered a measure of diagnostic accuracy. For biomarkers with high AUCs, we defined a case-classification rule based on a marker-specific cutoff. We then estimated sensitivity and specificity and the corresponding 95% Wilson confidence interval for each rule. Ethical Issues The study protocol was approved by ethical review boards at participating institutions in S?o Paulo state, the S?o Paulo State Health Department, the Brazilian National Committee for Ethics in Research, the Pan American Health Organization, and the CDC. Informed consent was obtained from all study participants. RESULTS After reviewing medical records for 317 children aged 6C14 years detected through participating institutions, we identified 30 clinically compatible CRS cases, of which 13 were laboratory confirmed. Follow-up investigation of 225 suspected CRS cases reported to the S?o Paulo State Health Department and 88 suspected cases identified from laboratory records of the state measles/rubella reference laboratory at the Adolfo Lutz Institute identified 2 clinically compatible, laboratory-confirmed CRS cases. Thirty-one mothers of children with CRS and 62 age-matched children without CRS were enrolled in the comparison groups. Among the 32 children with CRS, 11 (34.4%) were female, the median age was 10.5 years (range, 6C14 years), 28 (87.5%) had a history of at least 1 postnatal Dacarbazine dose of rubella vaccine, and 15 (46.9%) had laboratory-confirmed CRS. Deafness was the most common disability (87.5%), followed by cardiac defects (71.9%) and cataracts (43.8%; Table 1). We compared bio-marker distributions Dacarbazine of laboratory-confirmed and clinically confirmed cases and found no differences (Table 2). Twenty-one enrolled mothers (67.0%) reported clinical symptoms of rubella or a diagnosis of rubella during pregnancy; 16 (51.6%) reported a history of 1 1 rubella vaccine dose after the CRS-affected pregnancy. Among 62 children in the comparison group, 33 (53.2%) were female, the median age was 10 years (range, 5C13 years), and 58 (93.6%) had history of receiving 1 rubella vaccine dose. Children with and children without CRS were not Dacarbazine significantly different when comparing sex (= .08, by the 2 2 test), age (= .52, by the Wilcoxon rank sum test), and history of reported rubella vaccination (= .44, by the Fisher exact Rabbit Polyclonal to Gab2 (phospho-Tyr452) test). Table 1 Characteristics of 32 Children With Congenital Rubella Syndrome (CRS), S?o Paulo, Brazil, 2008C2011 Valuea .029). The E1 signal strength among children with CRS may have been lower than that among their mothers (the difference was not statistically significant; = .071), whereas the ratio of the E1 signal strength to the total IgG antibody concentration was significantly lower in children with CRS (= .001; Table 3). Table 3 Comparison of Total Rubella Virus (RUBV)-Specific Immunoglobulin G (IgG) Concentration, RUBV-Specific IgG Avidity, and IgG Signal Strength to Individual RUBV Antigens for Sera From Children With Congenital Rubella Syndrome (CRS), Mothers of Children, and Age-Matched Children Without CRS, S?o Paulo, Brazil, 2008C2011 ValueaValueb .001); however, antibody avidity was comparable in both groups (= .07; Table 3 and Physique 2). Immune responses among children with CRS exhibited higher signals to C protein ( .001) and to E2 protein ( .001) than did those among children in the comparison group. Ratios of the antigen-specific signal strength to the RUBV-specific IgG antibody concentration among case patients were significantly higher than those among children without CRS for E1/IgG ( .001), E2/IgG ( .001), and C/IgG (.