Development of hazards prediction model for lower limb lymphedema in postoperative cervical cancer patients

LI Sen; LIU Gaoming; ZHAO Guorui; ZHOU Ying

doi:10.12102/j.issn.1009-6493.2025.14.004

Abstract：Objective:To explore the influencing factors of lower limb lymphedema in postoperative cervical cancer patients and to construct a Cox proportional-hazards prediction model.Methods:A total of 303 postoperative cervical cancer patients admitted to a cancer hospital in Hunan province were selected for study subjects from July 2020 to June 2021.Predictors were screened using LASSO regression and Cox regression analysis.The Cox proportional-hazards prediction model was constructed.Then the Nomogram was constructed.The model's discriminative ability and predictive performance were evaluated using the C-index,calibration curve,and area under the curve of receiver operator characteristic.Results:The result of Cox regression analysis showed that 5 variables were retained,namely hypertension,postoperative concurrent chemoradiotherapy,prolonged static posture after surgery,postoperative physical labor intensity and whether exercise regularly after surgery.Based on literature review and expert consultation,2 additional variables(number of pelvic lymph nodes removed and postoperative radiation dose)were incorporated to construct the Cox proportional-hazards prediction model.The Nomogram was also constructed.The C-index was 0.858(95%CI 0.784-0.932)in the test set.The area under the curve of receiver operator characteristic at 1 year and 2 years time points were 0.920 and 0.881,respectively.And the calibration curve demonstrated good agreement between predicted and observed outcomes.Conclusions:The Cox proportional-hazards prediction model for lower limb lymphedema in postoperative cervical cancer patients had good predictive performance.It could provide reference for early identification of high-risk patients and prevention and treatment of lower limb lymphedema in postoperative cervical cancer patients.