Phase angle is associated with sarcopenic obesity in post-stroke patients

Evidence is limited concerning the association between bioimpedance analysis-derived phase angle and sarcopenic obesity. This study examined this association in patients who underwent convalescent rehabilitation after stroke.

This cross-sectional study included hospitalized patients with post-acute stroke. The phase angle was measured using a multifrequency bioimpedance analysis. Sarcopenic obesity was diagnosed according to the European Society of Clinical Nutrition and Metabolism/European Association for the Study of Obesity criteria. A multiple linear regression analysis was used to examine the association between the phase angle and sarcopenic obesity after adjusting for confounding factors. A receiver operating characteristic curve was used to calculate the optimal phase angle cutoff value for predicting sarcopenic obesity.

A total of 760 patients (median age 71 years; 352 women) were analyzed. The median (interquartile range, 25th and 75th percentiles) phase angle was 4.45° (4.10°, 4.88°). Sarcopenic obesity was diagnosed in 16 (3.9%) men, 18 (5.1%) women, and 34 (4.5%) patients overall. The multiple linear regression analysis showed that sarcopenic obesity (β = -0.185, p = 0.017) and sarcopenia (β = -0.121, p = 0.012) were significantly and negatively associated with the phase angle. Obesity was not significantly associated with the phase angle (β = -0.078, p = 0.094). The optimal cutoff value of the phase angle for diagnosing sarcopenic obesity was 4.29° (sensitivity 0.65, specificity 0.80, area under the curve [AUC] 0.79, 95% CI 0.77-0.87, p < 0.001) in men and 3.84° (sensitivity 0.67, specificity 0.84, AUC 0.81, 95% CI 0.79-0.86, p < 0.001) in women.

The phase angle was associated with sarcopenic obesity, and the cutoff values of the phase angle that could predict sarcopenic obesity were 4.29° for men and 3.84° for women. This simple and practical phase angle-based prediction of sarcopenic obesity can be useful in clinical practice.