Improving dengue case confirmation by combining rapid diagnostic test, clinical, and laboratory variables

Abstract

Background Dengue is the most widely distributed arboviral disease in tropical and subtropical countries. Early diagnosis is difficult, and most of the suspected cases are diagnosed according to clinical criteria. In underdeveloped countries, laboratory tests are done on a proportion of dengue with warning signs or severe dengue suspect cases. This study aimed to design a diagnostic algorithm using rapid diagnostic tests (RDT), ELISA tests together with clinical and hematology variables to confirm dengue cases in febrile patients from an endemic area in Colombia. Methods and results. A total of 505 samples were collected from patients with the acute febrile syndrome (<7 days) assisted to the Municipal Hospital in Girardot (Colombia). Serum samples were evaluated by rapid diagnostic tests -RDT- (IgM and IgG antibodies and NS1 antigen immunochromatographic assay), capture ELISAs (IgM, IgG, and NS1 antigen), and by RT-PCR. We analyzed individual test performance to determine which were the most useful to confirm dengue cases. Individual results for IgM, IgG, and NS1 RDT yield low sensitivity and specificity values than the reference standard. A high sensitivity (96.3%) and specificity (96.4%) were obtained after combining the IgM and NS1 ELISAs results. The analysis using the combined NS1 RDT and IgM ELISA results showed 90.3% sensitivity and 96.2% specificity. Adjusted odd ratios (aOR) were calculated including data from symptoms, signs, and diagnostic laboratory tests to differentiate dengue from other febrile illnesses (OFI). Myalgia (aOR: 1.87, CI95%: 1.04-3.38), abdominal tenderness (aOR: 1.89, CI95%: 1.14-3.10), platelets count <140.000/mm3 (aOR: 2.19 CI95%: 1.31-3.67). The analysis using the results of the diagnostic test yields significant ratios for IgM RDT (aOR:2.63 CI95%: 1.59-4.33) and NS1 RDT also differentiate dengue cases from OFI. Combined positive IgM or NS1 RDT and the one that combined positive NS1 RDT or IgM ELISA detect dengue cases in 81.6% and 90.6%, respectively (p<0.001). Conclusion. Our findings showed that only clinical diagnosis does not confirm true dengue cases dengue and needs to be complemented by laboratory diagnostic tests to establish the diagnosis. We also demonstrate the usefulness of rapid tests in diagnosis, suggesting their implementation with IgM ELISA test to better confirm dengue cases.