Diagnostic Performance of Genexpert MTB/RIF for Early Detection of Pulmonary Tuberculosis

Tuberculosis (TB) is one of the key health challenges in the world, particularly in low- and middle-income countries. Conventional diagnostic tools, such as microscopy of smears, are low in sensitivity, whereas culture tools, though sensitive, are slow. The GeneXpert MTB/RIF assay is a rapid method of detection; however, its local diagnostic capabilities require evaluation. Objective: To determine the diagnostic quality of the GeneXpert MTB/RIF test versus sputum smear microscopy for early Diagnosis of pulmonary tuberculosis in DG Khan. Methodology: The study was a cross-sectional descriptive study conducted over a six-month period (January 1, 2025, to June 30, 2025) at the Medicine and Pulmonology departments of Allama Iqbal Teaching Hospital, Dera Ghazi Khan. One hundred ninety clinically suspected pulmonary TB patients were recruited. Sputum samples were analyzed using both smear microscopy and GeneXpert MTB/RIF. Diagnostic performance indices, including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy, were calculated using culture results as the reference standard. Data were analyzed with SPSS version 25. Results: Of the 190 patients, the mean age was 37.2 ± 16.8 years, with 51.6% males and 48.4% females. The majority were from rural areas (72.6%) and farmers by occupation (62.1%). GeneXpert detected 156 positive cases compared with 108 by smear microscopy, including 54 additional smear-negative but GeneXpert-positive cases. Overall, the assay's accuracy was 71.6%, with a sensitivity of 71.2%, a specificity of 72.0%, a PPV of 91.8%, and an NPV of 35.5%. A moderate correlation was observed between smear microscopy results and the study's findings (r = 0.48, p < 0.001). Conclusion: GeneXpert MTB/RIF demonstrated a superior diagnostic yield compared to smear microscopy, particularly in smear-negative cases, making it a valuable tool for enhancing TB detection in high-burden settings.