Résumé
BACKGROUND: GeneXpert MTB/RIF is a real-time PCR assay with established diagnostic performance in pulmonary and extra-pulmonary forms of tuberculosis. The aim of this study was to assess the contribution of GeneXpert MTB/RIF assay to the management of patients with any form of active tuberculosis in a single large tertiary center in Saudi Arabia, with a special focus on the impact on time to start of antituberculous therapy compared with Ziehl-Neelsen (ZN) smears and mycobacterial cultures. MATERIALS AND METHODS: Clinical, radiological and laboratory records for all patients who were commenced on antituberculous therapy between March 2011 and February 2013 were retrospectively reviewed. RESULTS: A total of 140 patients were included, 38.6% of which had pulmonary tuberculosis. GeneXpert MTB/RIF was requested for only 39.2% of patients and was the only reason for starting antituberculous therapy for only 12.1%. The median time to a positive GeneXpert MTB/RIF result was 0 days (IQR 3) compared with 0 day (IQR 1) for smear microscopy (P > 0.999) and 22 days (IQR 21) for mycobacterial cultures (P < 0.001). No patients discontinued antituberculous therapy because of a negative GeneXpert MTB/RIF result. CONCLUSIONS: In a setting wherein physicians are highly experienced in the diagnosis and treatment of tuberculosis, GeneXpert MTB/RIF was remarkably under-utilized and had only a limited impact on decisions related to starting or stopping antituberculous therapy. Cost-effectiveness and clinical utility of routine testing of all smear-negative clinical samples submitted for tuberculosis investigations by GeneXpert MTB/RIF warrant further study.