Association of reduced BMI, length of hospital stay, mortality, and malnutrition diagnosis in patients with acute exacerbation COPD: A secondary analysis of a cohort study

Background Body mass index (BMI) presents prognostic value in chronic obstructive pulmonary disease (COPD), and despite its limitations in capturing malnutrition, its use is common to assess nutritional status. We aimed to confirm the association between BMI and in-hospital outcomes in acute exacerbation of COPD (AECOPD) and its inaccuracy in diagnosing malnutrition. Methods We diagnosed malnutrition using the Subjective global assessment (SGA), Academy of Nutrition and Dietetics-American Society for Parenteral and Enteral Nutrition (AND-ASPEN), and two cutoff values for reduced BMI (age-related and <= 21.0). BMI accuracy was assessed using the area under the receiver operating characteristic (AUC-ROC) curve and SGA and AND-ASPEN as references. We evaluated in-hospital mortality and hospital stay outcomes and constructed logistic regression models. Results The median hospital stay was 11 (7-18) days, and 7.5% of patients died. Malnutrition prevalence according to BMI, SGA, and AND-ASPEN was 21.4% (mean of both cutoff values), 50%, and 54%, respectively. Reduced BMI presented low agreement (kappa = 0.315-0.383) and unsatisfactory accuracy (AUC-ROC curve = 0.333-0.679) with reference methods for malnutrition diagnosis. Age-related reduced BMI (odds ratio [OR] = 2.11; 95% CI, 1.10-4.04) and BMI <= 21.0 (OR = 2.25; 95% CI, 1.13-4.48) were associated with hospital stays longer than the median in adjusted models, but not in-hospital mortality. Conclusion BMI was inaccurate in identifying malnutrition in hospitalized patients with AECOPD and was associated with hospital stays longer than ten days.