Discriminative Evaluation of Sarcopenic Dysphagia Using Handgrip Strength or Calf Circumference in Patients with Dysphagia Using the Area under the Receiver Operating Characteristic Curve

Abstract

This multicenter cross-sectional study aimed to evaluate the discriminative ability of sarcopenic dysphagia (SD) using handgrip strength (HGS) or calf circumference (CC) in patients with dysphagia. Patients aged 20 years or older who were registered in a database at acute, rehabilitation, long-term care hospitals and home health care facilities were included. Logistic regression analysis was performed using SD as the outcome and HGS, CC, and other confounding factors as covariates, separately by sex. Algorithm-based SD diagnosis and HGS or CC were used as the reference and index tests, respectively. Their accuracy was evaluated using the area under the receiver operating characteristic curve (AUC), and cutoff values were calculated. Of the 460 patients, 285 (126 males) were diagnosed with SD. Logistic regression analysis showed that HGS (odds ratio [OR], 0.909; 95% confidence interval [CI], 0.873–0.947) in males and CC (OR, 0.767; 95% CI, 0.668–0.880) in females were independently associated with SD. The AUC for HGS in males was 0.735 (p < 0.001) and CC in females was 0.681 (p < 0.001). The cutoff values were 19.7 kg for HGS in males (sensitivity, 0.75; specificity, 0.63) and 29.5 cm for CC in females (sensitivity, 0.86; specificity, 0.48). HGS in males and CC in females provided statistically significant information to discriminate SD from dysphagia.