Evaluation of screening strategies for pulmonary tuberculosis among hospitalized patients in a low-burden setting: cost-effectiveness of GeneXpert MTB/RIF compared to smear microscopy

Objective: Hospitalized patients undergoing evaluation for pulmonary tuberculosis (TB) require airborne isolation while testing for Mycobacterium tuberculosis (MTB) to reduce risk of nosocomial transmission. GeneXpert MTB/RIF (Xpert) is more rapid and accurate than sputum smear microscopy, but it is not routinely used to 'rule out' infectious pulmonary TB among hospitalized patients in the United States. We sought to evaluate the diagnostic performance and cost-effectiveness of Xpert-based TB evaluation. Methods: We conducted a retrospective cohort study of hospitalized adults evaluated for pulmonary TB at a large academic medical center in New York from 2010 to 2017. Using propensity score matching, we compared hospital length-of-stay among patients undergoing conventional smear-based TB evaluation to a control group with non-TB pneumonia. We performed a probabilistic cost-effectiveness analysis to compare Xpert-based versus conventional TB evaluation. Results: In total 1,421 patients were evaluated for TB with airborne isolation and sputum testing; mycobacterial culture was positive for MTB in 49 (3.4%). Conventional TB evaluation was associated with an increase of 4.4 hospital days compared to propensity-matched controls. Xpert-based testing strategies dominated conventional TB evaluation with a cost savings of $5,947 (95% CI, $1,156-$12,540) and $4,445 (95% CI, $696-$9,526) per patient depending on the number of Xpert tests performed (1 vs 2, respectively) and assumptions about the reduction of length of stay achieved. Conclusions: In the evaluation of hospitalized patients for pulmonary TB, Xpert-based testing has superior diagnostic performance and is likely cost-effective compared to smear microscopy due to reduced hospital length-of-stay associated with more rapid test results.