Desempenho do método Ogawa-Kudoh em cultura de Mycobacterium tuberculosis de amostras paucibacilares

Abstract

Objective: To assess the performance of the Ogawa–Kudoh (OK) method for tuberculosis diagnosis in pulmonary and extrapulmonary paucibacillary sample materials, relative to Lowenstein-Jensen (LJ) and Mycobacteria Growth Indicator Tube (MGIT) processing. Methods: This cross-sectional study employed extrapulmonary and pulmonary samples (other than sputum) received at the Mato Grosso do Sul Central Public Health Laboratory between February 2017 and March 2018. The samples were subjected to OK, LJ, and MGIT processing as per the 2008 National Manual for Laboratory Surveillance of Tuberculosis and Other Mycobacteria. Results: A total of 347 samples (83.3% extrapulmonary, 16.7% pulmonary) were seeded in at least one of three media (OK: 330; LJ: 314; MGIT: 113). Most extrapulmonary samples consisted of cerebrospinal fluid (28.2%), bone marrow (20.5%), and tissue fragments (16.7%). Pulmonary samples were predominantly alveolar bronchial lavage (6.9%) and tracheal secretion (9.8%). Adopting MGIT as a reference, OK proved 75% sensitive and 100% specific. With LJ as the reference, OK sensitivity was 57.1% and specificity 96.3%. Agreement categories were as follows: OK vs. LJ, moderate (κ = 0.485); OK vs. MGIT, almost perfect (κ = 0.851); LJ vs. MGIT, substantial (κ = 0.655). The OK method proved accurate and specific for diagnosing tuberculosis in pulmonary and extrapulmonary paucibacillary sample materials, providing the same accuracy as LJ and MGIT, characterizing the approach as a valuable laboratory tool for investigating cases of suspected tuberculosis in laboratory facilities with limited resources.