BACKGROUND: Estimation of performance status (PS) has been assessed as a tool to determine which patients with newly diagnosed pulmonary TB (PTB) are most at risk of dying. This simple prediction rule has not been validated in patients with PTB with different background characteristics and from different geographic areas. METHODS: A retrospective cohort study was conducted in two Japanese hospitals in different regions and included 432 inpatients with newly diagnosed smear-positive non-multidrug-resistant lung TB without HIV infection. The patients had a mean ± SD age of 64.9 ± 19.7 years and 135 were female (31.3%). Detailed nursing charts were reviewed to estimate PS, which was graded 0 (best condition), 1, 2, 3 or 4 (worst condition), for each patient. RESULTS: Single variable and multivariable Cox regression analyses models revealed that a one-point increase in PS was associated with a 2.8-fold (95% CI 2.2-3.6) and 2.3-fold (95% CI 1.8-3.0), adjusted for age, gender, comorbidities and treatment regimen, increase in the HR for death (p < 0.001 for both models). Kaplan-Meier curves also showed a significant difference in mortality among different PS groups (p < 0.001). CONCLUSION: PS was strongly associated with mortality from PTB in the study cohort. Estimation of PS at the start of treatment for newly diagnosed PTB patients could be a useful tool in case management in resource-limited countries.
BACKGROUND: Estimation of performance status (PS) has been assessed as a tool to determine which patients with newly diagnosed pulmonary TB (PTB) are most at risk of dying. This simple prediction rule has not been validated in patients with PTB with different background characteristics and from different geographic areas. METHODS: A retrospective cohort study was conducted in two Japanese hospitals in different regions and included 432 inpatients with newly diagnosed smear-positive non-multidrug-resistant lung TB without HIV infection. The patients had a mean ± SD age of 64.9 ± 19.7 years and 135 were female (31.3%). Detailed nursing charts were reviewed to estimate PS, which was graded 0 (best condition), 1, 2, 3 or 4 (worst condition), for each patient. RESULTS: Single variable and multivariable Cox regression analyses models revealed that a one-point increase in PS was associated with a 2.8-fold (95% CI 2.2-3.6) and 2.3-fold (95% CI 1.8-3.0), adjusted for age, gender, comorbidities and treatment regimen, increase in the HR for death (p < 0.001 for both models). Kaplan-Meier curves also showed a significant difference in mortality among different PS groups (p < 0.001). CONCLUSION: PS was strongly associated with mortality from PTB in the study cohort. Estimation of PS at the start of treatment for newly diagnosed PTB patients could be a useful tool in case management in resource-limited countries.
Authors: Lauren S Peetluk; Felipe M Ridolfi; Peter F Rebeiro; Dandan Liu; Valeria C Rolla; Timothy R Sterling Journal: BMJ Open Date: 2021-03-02 Impact factor: 2.692