ABSTRACT
OBJECTIVES: The aim of this study was to investigate the predictive value of bioimpedance phase angle (PA) on selected clinical outcomes in patients hospitalized in internal-medicine wards. METHODS: This was a retrospective observational study of 168 patients admitted to the internalmedicine service (52.9% women, 47.1% men), with a mean (± SD) age of 73.9 ± 15.9 y. Anthropometric examination, laboratory tests, and bioelectrical impedance analysis were performed. Bioimpedance-derived PA was the study's parameter. Length of hospital stay, prospective all-cause hospital readmission, mortality, and falls were the clinical endpoints. RESULTS: Across the four PA quartile groups, age was incrementally higher (P ≤ 0.001). Multivariate linear regression models showed that PA quartile 1 was significantly associated with length of hospital stay (ß, SE) in both crude and adjusted models-respectively, ß (SE) = 6.199 (1.625), P ≤ 0.001, and ß = 2.193 (1.355), P = 0.033. Over a 9-mo follow-up period, the hazard ratios for readmission, in-hospital falls, and mortality were associated with the lowest phase angle (PA quartile 1 versus quartiles 2-4)-respectively, 2.07 (95% confidence interval [CI], 1.28-3.35), 2.36 (95% CI, 1.05-5.33), and 2.85 (95% CI, 1.01-7.39). Associations between narrow PA and outcomes continued to be significant after adjustments for various confounders. CONCLUSIONS: In internal-medicine wards, bioimpedance-derived PA emerged as a predictor of length of hospital stay, hospital readmission, falls, and mortality. The present findings suggest that in the hospital setting, PA assessment could be useful in identifying patients at higher risk who need specific nutritional support.
