Phase Angle, Skeletal Muscle Function, and Molecular Phenotype in HFpEF: A Subanalysis of the SAUNA-HFpEF Study

Background: The phase angle (PhA) derived from bioelectrical impedance analysis (BIA) is a non-invasive marker of cellular integrity that has been associated with nutritional status, functional status and prognosis in heart failure (HF). Whether PhA is related to the skeletal muscle molecular phenotype in HF with preserved ejection fraction (HFpEF) remains unknown. Methods: In this exploratory subanalysis of the SAUNA-HFpEF trial, 17 patients with HFpEF underwent BIA with assessment of body composition, cardiopulmonary exercise testing, isokinetic assessment of quadriceps muscle strength, and measurement of circulating growth differentiation factor-15 (GDF-15), as well as nutrition-related and metabolic laboratory parameters before and after a 10-week sauna intervention. Paired vastus lateralis biopsies (n = 12) were analyzed by quantitative real-time polymerase chain reaction (qRT-PCR) for catabolic (FBXO32, GSK-3β), fatty acid metabolism (ACADL, CPT1B, CD36), and anabolic (mTOR, IGF-1) markers. Associations with PhA were assessed using Spearman correlation and repeated-measures correlation (rmcorr), with false discovery rate (FDR) correction for multiple testing. Results: Cross-sectionally, lower baseline PhA correlated with higher GDF-15 (r = −0.758; 95% CI −0.936 to −0.412), higher extracellular water to total body water (ECW/TBW) ratio (r = −0.744; 95% CI −0.941 to −0.350), higher expression of GSK-3β, ACADL, FBXO32, CD36, and CPT1B (all r ≤ −0.620), and lower isokinetic quadriceps strength (r = 0.523) (all FDR-adjusted p < 0.05). No significant associations were found for IGF-1, mTOR, or muscle mass. An exploratory comparison using a previously published PhA cut-off of 4.65° showed consistent differences. Longitudinally, PhA did not change significantly over the intervention, and longitudinal changes in PhA were unrelated to changes in molecular markers but correlated with changes in the ECW/TBW ratio (rmcorr r = −0.708; 95% CI −0.883 to −0.360). Body composition, waist circumference, the lipid profile, HbA1c and iron status were also assessed at both time points without any observed changes (all FDR-adjusted p > 0.05). Conclusions: In this small exploratory analysis, lower PhA was associated with a more catabolic and metabolically altered skeletal muscle gene expression profile in HFpEF. PhA may provide complementary, non-invasive information on muscle quality and cellular and fluid status. Its potential role in identifying patients who warrant a structured nutritional assessment deserves further investigation. These hypothesis-generating findings require validation in larger prospective studies.