PLoS ONE (Jan 2017)

Laboratory evaluation of immunochromatographic rapid diagnostic tests for cholera in Haiti.

  • Wilfredo R Matias,
  • Fabrice E Julceus,
  • Cademil Abelard,
  • Leslie M Mayo-Smith,
  • Molly F Franke,
  • Jason B Harris,
  • Louise C Ivers

DOI
https://doi.org/10.1371/journal.pone.0186710
Journal volume & issue
Vol. 12, no. 11
p. e0186710

Abstract

Read online

BACKGROUND:Rapid diagnostic tests (RDT) for cholera are promising tools for detecting cholera in areas with limited laboratory infrastructure. However, evidence on the characteristics of the many available RDTs is scarce, and their use has been limited by suboptimal performance. We evaluated the performance characteristics of three cholera RDTs from Span Diagnostics, Artron Laboratories, and Standard Diagnostics in a regional laboratory in Haiti. METHODOLOGY/PRINCIPAL FINDINGS:We retrospectively reviewed records from May 2014 to October 2015 of a laboratory-based surveillance program for Vibrio cholerae at Hôpital Saint-Nicolas in Saint-Marc, Haiti. We compared the results of 511 Crystal VC, 129 Artron and 451 SD Bioline RDTs to bacterial culture as the gold standard. Of 905 cultures, 477 (52.7%) were positive for V. cholerae O1, of which 27.7% were serotype Inaba. No cultures grew V. cholerae O139. Sensitivity and specificity of Crystal VC were 98.6% (95%CI: 96.5%-99.6%) and 71.1% (95%CI: 64.7%-76.9%), respectively. Artron demonstrated a sensitivity of 98.6% (95%CI: 92.7%-100%) and specificity of 69.1% (95%CI: 55.2%-80.9%). SD Bioline demonstrated a sensitivity of 81.1% (95%CI: 75.6%-85.8%) and specificity of 92.8% (95%CI: 88.4%-95.9%). Crystal VC and Artron frequently showed false positive O139 bands, whereas none were seen with SD Bioline. CONCLUSIONS/SIGNIFICANCE:There is significant variation in the performance of different cholera diagnostic RDTs. Artron and Crystal VC RDTs have high sensitivity and low specificity, while SD Bioline RDT has low to moderate sensitivity and high specificity when performed by laboratory technicians in Haiti. Study limitations included its retrospective design. The suboptimal characteristics of these tests limit their use as clinical point-of-care tests; however, they may be useful in outbreak response, surveillance, and research in resource-limited settings.