Objective Inflammation and malnutrition play pivotal roles in disease progression. This study evaluated the associations between inflammation/nutrition-based composite indices and treatment outcomes in hospitalized adults with previously untreated pulmonary tuberculosis (PTB) complicated by hypertension. Methods We included hospitalized adults with untreated PTB and concomitant hypertension admitted to Changsha Central Hospital (from January 2019 to December 2021). Inflammation/nutrition-based composite indices including Prognostic Nutritional Index (PNI), Hemoglobin-Albumin-Lymphocyte-Platelet (HALP) score, Monocyte-to-Albumin Ratio (MAR), Neutrophil-to-Albumin Ratio (NAR), and Red Cell Distribution Width-to-Albumin Ratio (RAR) were calculated. Associations between these indices and PTB treatment outcomes were evaluated using multivariable logistic regression, restricted cubic splines, and subgroup analysis. Results Among 1,012 hospitalized patients with untreated PTB complicated by hypertension, 166 individuals (16.40%) experienced unfavorable treatment outcomes. After multivariate adjustment, logistic regression revealed that RAR (aOR 0.75, 95% CI 0.64–0.89, p = 0.001) and NAR (aOR 0.40, 95% CI 0.19–0.84, p = 0.015) were inversely associated with favorable outcomes, while PNI was positively associated (aOR 1.03, 95% CI 1.01–1.06, p = 0.011). Restricted cubic splines revealed a negative linear association for NAR ( p = 0.030) and RAR ( p = 0.013), and a positive linear association for PNI ( p = 0.002). Machine learning algorithms including GLM, RF, and SVM ranked the relative contributions of inflammation/nutrition-based indices and their components to treatment outcomes, with RAR identified as the top-contributing indicators. Conclusion Routine laboratory-derived composite indices, particularly RAR and PNI, were associated with treatment outcomes in previously untreated PTB patients with hypertension, underscoring the importance of baseline nutritional and inflammatory status in the management of this patient population.
Xiao et al. (Wed,) studied this question.