Abstract

Madagascar is the country that reports the most cases of plague in the world. This is an evaluation study of plague diagnosis over a 2-year period from January 2017 to December 2018. For all suspected plague cases, peripheral RDT (performed locally), central RDT (performed in the laboratory) and PCR were performed. Parameters of the diagnostic tests used, year of study, performance and study of concordance between tests were studied. 2981 cases were collected, 21.6% were confirmed, 28.4% probable and 50% suspected. The sensitivities of peripheral RDT, central RDT and PCR were 96.55, 100 and 97.41%, respectively; the specificities were 41.43, 51.17 and 89.24%. Cohen's kappa was 0.12 between peripheral RDT and culture; 0.17 between culture and central RDT and 0.64 between CRP and culture. For pneumonic plague (PP) patient samples, sensitivities were 80.00, 66.66 and 93.33%, Cohen's kappa was 0.017 between peripheral RDT and culture; 0.013 between central RDT and culture and 0.371 between PCR and culture; sensitivities of peripheral RDT and central RDT were 75.00 and 62.50% for 2017, respectively. The null hypothesis between diagnostic tests was rejected, discordance between tests was found. Sensitivity is lowered during lung sampling and during 2017.

Key words: Plague, polymerase chain reaction, technology assessment.