Diagnostic yield of sputum microbiological analysis in the diagnosis of pulmonary tuberculosis in a period of 10 years.

INTRODUCTION Pulmonary tuberculosis (TB) requires an early diagnosis for prompt introduction of treatment and prevention of transmission. Definitive diagnosis is obtained by microbiological culture and identification of Mycobacterium tuberculosis in respiratory specimens, mostly sputum samples. MATERIALS AND METHODS Retrospective data analysis of all patients suspected of pulmonary TB that submitted three consecutive sputum samples to the Pulmonology Diagnostic Center (PDC) Laboratory between 2004 and 2013. Extrapulmonary TB cases were excluded. Four microbiological analyses were executed on each specimen: two smears with Ziehl-Neelsen staining, direct and concentrate; and two culture examinations, one in liquid and one in solid medium. Statistical analysis was performed by SPSS. RESULTS A total of 694 patients were enrolled in this study (65% men, mean age 48.5±18.6 years, 97% Portuguese), most of them exhibiting TB-related complaints. Pulmonary TB was diagnosed in 41% of the patients; 54% had non-specific radiological changes and 34% had pulmonary cavitation. The cumulative sensitivity rates of each of the three smears were 24.6%, 27.7% and 28.8% for concentrated samples and 19.3%, 20.4% and 22.5% for direct samples. The cumulative sensitivities of sputum culture were 33.3%, 37.9% and 41.8% for solid medium, and 43.9%, 51.6% and 55.4% for liquid medium. Pondering all forms of microbiological analysis, the cumulative sensitivities of each sample were 51.2%, 59.6% and 63.2%. There was an incremental yield of 8.4% for the second specimen and 3.5% for the third specimen. All sensitivity rates were higher among patients with pulmonary cavitation. CONCLUSIONS This study showed an incremental yield with more than one sputum sample. However, overall sensitivity remained low, suggesting a need for new diagnostic strategies and novel and better diagnostic tools.