Associations of height-standardized calf circumference and mid-arm circumference with overall survival in adults with cancer: A retrospective cohort study.

BACKGROUND

Body size may influence anthropometric measurements' performance in survival prediction. Our study aimed to develop standardized calf and mid-arm circumferences and compare their prognostic value with that of unstandardized forms in patients with cancer.

METHODS

This multicenter retrospective cohort study included 15,826 patients. Malnutrition was assessed using the Patient-Generated Subjective Global Assessment with a score ≥4. The concordance index (C-index) and receiver operating characteristic (ROC) curves were used to compare predictive power. Cox and logistic regression analyses were performed to identify the independent factors associated with survival and malnutrition, respectively.

RESULTS

The highest C-index and the area under the ROC curve (AUC) values were observed for height-standardized calf and mid-arm circumferences (C-index: 0.584 and 0.585; 1-year AUC values: 0.600 and 0.600; and 3-year AUC values: 0.597and 0.594, respectively). Compared with unstandardized calf circumference, the height-standardized measure improved the predictive power for survival (difference = 0.020; P < 0.001). Similarly, height-standardized mid-arm circumference showed superior predictive ability (difference = 0.015; P < 0.001). Additionally, height standardization enhanced the predictive accuracy of calf and mid-arm circumferences for malnutrition (calf circumferences: difference = 0.014, P < 0.001; mid-arm circumferences: difference = 0.008, P < 0.001). Height-standardized calf and mid-arm circumferences were independent prognostic factors of survival and risk factors for malnutrition.

CONCLUSION

Height-standardized calf and mid-arm circumferences were independently associated with survival and malnutrition in patients with cancer. Although the improvements in predictive performance were modest, height standardization may improve the clinical utility of anthropometric assessments by accounting for differences in body size.