Evaluation of a rapid dipstick (Crystal VC) for the diagnosis of cholera in Zanzibar and a comparison with previous studies

• Published in: PloS one 2012-05, Vol.7 (5), p.e36930-e36930
• Main Authors: Ley, Benedikt, Khatib, Ahmed M, Thriemer, Kamala, von Seidlein, Lorenz, Deen, Jacqueline, Mukhopadyay, Asish, Chang, Na-Yoon, Hashim, Ramadhan, Schmied, Wolfgang, Busch, Clara J-L, Reyburn, Rita, Wierzba, Thomas, Clemens, John D, Wilfing, Harald, Enwere, Godwin, Aguado, Theresa, Jiddawi, Mohammad S, Sack, David, Ali, Said M
• Format: Article
• Language: English
• Subjects: Antibiotics; Biology; Care and treatment; Cholera; Cholera - diagnosis; Comparative analysis; Crystals; Culture; Diagnosis; Diagnostic systems; Diagnostic tests; Diarrhea; Epidemics; Ethics; Humans; Infections; Infectious diseases; Intravenous administration; Medical diagnosis; Medical laboratories; Medical research; Medicine; Outbreaks; Patients; Public health; Reagent Kits, Diagnostic; Reproducibility of Results; Sensitivity; Sensitivity analysis; Sensitivity and Specificity; Tanzania; Venture capital; Vibrio cholerae; Waterborne diseases; Workers
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0036930

The gold standard for the diagnosis of cholera is stool culture, but this requires laboratory facilities and takes at least 24 hours. A rapid diagnostic test (RDT) that can be used by minimally trained staff at treatment centers could potentially improve the reporting and management of cholera outbreaks. We evaluated the Crystal VC™ RDT under field conditions in Zanzibar in 2009. Patients presenting to treatment centers with watery diarrhea provided a stool sample for rapid diagnostic testing. Results were compared to stool culture performed in a reference laboratory. We assessed the overall performance of the RDT and evaluated whether previous intake of antibiotics, intravenous fluids, location of testing, and skill level of the technician affected the RDT results. We included stool samples from 624 patients. Compared to culture, the overall sensitivity of the RDT was 93.1% (95%CI: 88.7 to 96.2%), specificity was 49.2% (95%CI: 44.3 to 54.1%), the positive predictive value was 47.0% (95%CI: 42.1 to 52.0%) and the negative predictive value was 93.6% (95%CI: 89.6 to 96.5%). The overall false positivity rate was 50.8% (213/419); fieldworkers frequently misread very faint test lines as positive. The observed sensitivity of the Crystal VC RDT evaluated was similar compared to earlier versions, while specificity was poorer. The current version of the RDT could potentially be used as a screening tool in the field. Because of the high proportion of false positive results when field workers test stool specimens, positive results will need to be confirmed with stool culture.