Development and validation of risk prediction model for sarcopenia in patients with colorectal cancer

Objectives: Sarcopenia is associated with a poor prognosis in patients with colorectal cancer. However, the clinical factors that lead to colorectal cancer patients with sarcopenia are still unclear. The objectives of the study are to develop and validate a nomogram that predicts the happen of sarcopenia and provide a reliable tool for healthcare providers to identify the high-risk population of colorectal cancer patients with sarcopenia early. Methods A total of 359 patients diagnosed with colorectal cancer from July 2021 to May 2022 were included. All patients were randomly divided into a training (n = 287) cohort and a validation cohort (n = 72) at the ratio of 80/20. Univariate and multivariate logistic analysis were performed to evaluate the factors associated with sarcopenia. The diagnostic nomogram of sarcopenia in patients with colorectal cancer was constructed in the training cohort and was validated in the validation cohort. AUC, calibration curve and Hosmer-Lemeshow test were used to evaluate the performance of the nomogram. Results Smoking history, drinking history, diabetes, TNM stage, nutritional status, and physical activity were included in the nomogram for the prediction of sarcopenia. The diagnostic nomograms exhibited good discrimination, the AUC value of the diagnostic nomogram is 0.971 and 0.922 in the training and verification cohort. Nomogram's calibration work is also excellent (H-L test:0.886). Conclusions The nomogram composed of preoperative factors can successfully predict the occurrence of sarcopenia in patients with colorectal cancer, which is helpful to early identify the high-risk population and employ an intervention for them timely.