Abstract
INTRODUCTION: It remains controversial which frailty score correlates with adverse outcomes. Instead of these subjective and time consuming scores, we studied whether ultrasound guided lower extremity thickness measurements more closely are associated with adverse outcomes. METHOD: Patients undergoing gastrointestinal malignancy surgery were included as study subjects. Frailty was identified using the FRAIL scale assessment. Ultrasound measured the muscle thickness and echo intensity of the patient's upper and lower limbs. ANOVA was used to analyze the relationship between muscle data and frailty assessment. Significant indicators from the one-way analysis were included in the multivariate logistic regression equation. RESULTS: A total of 160 study subjects were included, 52 in the normal group, 78 in the prefrailty group, and 30 in the frailty group. The ANOVA showed that there were significant differences in age, hemoglobin, albumin, history of hypertension, history of coronary artery disease, the history of cerebrovascular disease, rate of postoperative complications, rate of transferring to intensive care unit, time out of bed, length of hospitalization, thickness and echo intensity of quadriceps femoris the vastus lateralis and the tibialis anterior, echo intensity of the medial gastrocnemius among the three groups (P < 0.05). Pearson's correlation analysis showed FRAIL score was correlated with muscle thickness and echo intensity of the lower limbs. Multifactor logistic regression analysis showed that the prefrailty group was positively correlated with age（β = 0.146, P = 0.004）, echo intensity of the medial gastrocnemius (β = 0.055, P = 0.031), and rate of postoperative complications（β = 1.439, P = 0.021）, also negatively correlated with muscle thickness of the tibialis anterior（β = -2.124, P = 0.007）. The frailty group was positively correlated with age（β = 0.22, P = 0.005）, tibialis anterior echo intensity (β = 0.082, P = 0.015), medial gastrocnemius echo intensity (β = 0.089, P = 0.026), cerebrovascular disease history (β = 2.311, P = 0.04), and postoperative complication rate (β = 2.684, P = 0.003). It was negatively correlated with albumin (β = -0.26, P = 0.017), quadriceps muscle thickness (β = -2.257, P = 0.017), and tibialis anterior muscle thickness (β = -5.408, P = 0.001). CONCLUSION: Ultrasound measurement of lower (not upper) extremity muscle thickness and echo intensity was significantly associated with discriminating severity of frailty and postoperative outcomes than frailty scores in elderly patients.