Bioelectrical Impedance Analysis–Derived Phase Angle in Initial Medical and Nursing Assessment for Cardiac Rehabilitation: Association with Cardiac Rehabilitation Risk Category and Longitudinal Change During the Program—An Exploratory Prospective Obs

Background: Cardiac rehabilitation (CR) reduces cardiovascular mortality and improves quality of life in coronary patients. Risk stratification at program onset is essential for individualizing exercise prescription. Phase angle (PhA), derived from bioelectrical impedance analysis, is an index of cell membrane integrity that has emerged as a functional biomarker in various clinical populations. Its utility at the onset of a phase II CR program has not been explored in the Spanish context. This study evaluated the association of PhA with CR risk category, functional capacity and quality of life, its change during the program, and its incremental value over the Duke Treadmill Score (DTS) and clinical variables for risk classification in patients beginning phase II CR. Methods: This was an exploratory prospective longitudinal observational study with repeated measures. A total of 109 patients with ischemic heart disease (86.2% male; mean age: 60.2 ± 7.6 years) enrolled in an 8-week phase II CR program (Badajoz, Spain) were included. Phase angle was measured with a multifrequency bioimpedance analyzer (TANITA MC-780MA) under standardized conditions at baseline and program completion. Non-parametric tests (Wilcoxon, Mann–Whitney U, and Spearman correlation), multivariable models adjusted for potential confounders, receiver operating characteristic (ROC) analysis with incremental-value metrics, and a classification and regression tree (CART) evaluated with standard, repeated and nested stratified cross-validation were applied. Results: Baseline PhA was significantly higher in the low-risk group than in the high-risk group (5.43 ± 0.65° vs. 4.98 ± 0.81°; U = 1568.5; p = 0.002; rank-biserial r = 0.38). PhA correlated positively with the Duke Treadmill Score (r_s = +0.268; p = 0.005), physical functioning (r_s = +0.291; p = 0.002), vitality (r_s = +0.265; p = 0.005) and grip strength (r_s = +0.196; p = 0.041), and inversely with age (r_s = −0.288; p = 0.002) and the comorbidity index (r_s = −0.198; p = 0.039). High-risk status remained associated with lower PhA after adjustment for age, sex, BMI, comorbidity, LVEF, diabetes, revascularization type and time since the event (β = −0.49°; p = 0.002). PhA alone showed moderate discrimination (AUC = 0.689; 95% CI: 0.574–0.803); adding PhA to the DTS did not improve discrimination (ΔAUC = +0.004; p = 0.748); and a CART model combining DTS, PhA and grip strength (cross-validated accuracy: 85.2%) classified patients identically to the DTS alone. Without the DTS, adding PhA to a clinical model improved model fit and reclassification (likelihood-ratio p = 0.003; NRI = 0.64) without a significant AUC gain (+0.067; p = 0.106). After 8 weeks, PhA increased slightly (5.27 → 5.33°; p = 0.016); the mean change was small (95% CI: 0.01–0.10°) and was not associated with functional changes, and its clinical relevance is uncertain because repeatability data were not available. Conclusions: Phase angle is lower in high-risk CR patients independently of age, sex, body composition and comorbidity and correlates weakly to moderately with functional capacity and quality of life. However, it did not add discriminative value to the DTS, and the clinical relevance of its small mean change during the program is uncertain. These exploratory findings show associations only: neither the standalone use of PhA nor its added value over the DTS for risk stratification has been established, and its use for monitoring the response to rehabilitation has not been validated.