Discordance between genotypic and phenotypic susceptibility test for tuberculosis in Cali, Colombia: a challenge in the clinical practice

Abstract

Background: The detection of Mycobacterium tuberculosis resistance patterns is based on phenotypic and genotypic susceptibility tests. The discordant results between them are a clinical challenge for the management of patients with drug-resistant tuberculosis. Aim: To evaluate the concordance between phenotypic and molecular tests in patients with drug-resistant tuberculosis treated in an institution in Cali, Colombia. Methods: A cross-sectional study was conducted. A phenotypic sensitivity profile was obtained from mycobacterial cultures. The genotypic susceptibility was obtained with Xpert-MTB/RIFâ or Genotype-MDRTBplusâ. The percentage of resistance and percentage of agreement between the results of the phenotypic and genotypic tests were evaluated. A Cohen's kappa coefficient (κ) was estimated for each type of resistance according to the test used. Results: A total of 30 cases with both genotypic and phenotypic testing were included. The phenotypic tests detected resistance to first-line drugs in 29/30 cases, while the molecular tests detected resistance in all the cases evaluated. The percentage of resistance detected between Genotype-MDRTBplusâ and the phenotypic test for rifampicin was 61.5% (overall agreement 41.1%, κ=0.40, p=0.96), while the percentage of resistance detected with Xpert-MTB/RIFâ was 100% (overall agreement 81.82%, κ: 0.00, p<0.001) for this same drug. Resistance to isoniazid detected by both types of tests was 94.4% (overall agreement 89.47%, κ: -0.055 p=0.59). Conclusions: Discordance between the results of genotypic and phenotypic tests is possible, so it is important to use and interpret both types of tests in a complementary way in the diagnosis of resistance to first-line drugs in M. tuberculosis infection.