RESUMEN
INTRODUCTION: Laboratory diagnosis of measles can be challenging, and the reintroduction of the measles virus in Brazil has brought about new issues. The aim of this study was to analyze the qPCR results of swab and urine samples and compare them with those of immunological methods for the diagnosis of measles. METHODS: This was a cross-sectional study based on a retrospective analysis of 3,451 suspected cases using laboratory test surveillance databases for qPCR (respiratory swabs and urine) and serologic tests for IgM and paired IgG. Sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and agreement through kappa and adjusted kappa coefficients (PABAK) were calculated using different diagnostic strategies. RESULTS: The swab and urine samples obtained using real-time qPCR were equivalent. Samples collected simultaneously and the combined samples showed moderate agreement between IgM ELISA and real-time qPCR; however, 48.9 % of the IgM ELISA analyses did not demonstrate detectable qPCR concentrations during simultaneous collections and 43.9 % of combined collections. The paired analysis of IgG showed an accuracy of 67.5 % for IgM and 90.7 % for real-time qPCR. CONCLUSIONS: Diagnosis based on IgM presents detection delimitation in samples collected early (1-5 days), suggesting that these individuals satisfy at least two criteria. In addition to qPCR, paired analysis of IgG using ELISA can be used to increase the sensitivity and specificity of laboratory diagnoses.