Phase angle measured by bioelectrical impedance analysis as a novel predictor of exercise capacity in patients undergoing exercise stress echocardiography
Y Mochizuki, Y Kunimoto, S Gohbara, Y Yamamoto, S Ichikawa, R Hachiya, E Toyosaki, H Fukuoka, T ShinkeAbstract
Background/Introduction
With the progressive ageing of the population, the prevalence of heart failure and valvular heart disease continues to increase. Exercise stress echocardiography (ESE) plays a pivotal role in functional assessment; however, in frail or elderly patients, reduced physical performance often limits the ability to apply appropriate exercise load. Consequently, determining the optimal initial workload and predicting maximal exercise load (MEL) remain significant clinical challenges.
Purpose
We sought to investigate whether bioelectrical impedance analysis (BIA)–derived parameters, particularly phase angle (PhA), a novel marker reflecting cellular integrity and muscle quality, could predict exercise capacity during ESE beyond conventional body composition measures.
Methods
Between April 2023 and December 2025, 271 consecutive patients underwent ESE at our Medical University. Among them, 240 patients without implanted cardiac devices and eligible for BIA assessment were included in the analysis. Associations between MEL, achievement of ≥25 W and ≥50 W, and BIA-derived parameters were examined using correlation and multivariable regression analyses.
Results
The mean age was 69.9 years, and 43% were female. The median Clinical Frailty Scale score was 3. More than 90% of ESE referrals were for evaluation of valvular heart disease. Mean skeletal muscle mass, bone mineral content, body fat mass, and PhA were 39.2 ± 7.8 kg, 2.4 ± 0.4 kg, 14.6 ± 6.1 kg, and 4.62 ± 0.82, respectively. PhA correlated negatively with age and BNP levels (r = −0.57 and −0.58, both p < 0.0001), and positively with skeletal muscle mass and handgrip strength (r = 0.56 and 0.67, both p < 0.0001). Notably, PhA demonstrated a strong positive correlation with MEL (r = 0.72, p < 0.0001). In multivariable linear regression adjusted for age, skeletal muscle mass, body fat mass, and BNP, PhA remained an independent predictor of MEL (t = 3.60, p = 0.0004). Furthermore, multivariable logistic regression analysis revealed that PhA independently predicted achieving ≥25 W and ≥50 W during ESE, irrespective of age, skeletal muscle mass, body fat mass, and BNP (≥25 W: OR 4.57, p = 0.016; ≥50 W: OR 2.61, p = 0.015).
Conclusion
BIA-derived PhA may serve as a practical and objective tool for estimating appropriate exercise load prior to ESE, potentially facilitating individualized stress protocols in an ageing cardiovascular population.Fig 1Fig 2