Perioperative Symptom Trajectories and a Risk Prediction Model for Cervical Cancer: A Prospective Longitudinal Study.

BACKGROUND

This prospective longitudinal quantitative study aimed to clarify perioperative symptom cluster characteristics in cervical cancer patients, identify influencing factors and hematological indicators, and develop a clinically applicable risk prediction model.

METHODS

A prospective longitudinal study was conducted at a tertiary cancer hospital in Northwest China (April 2024-June 2025). Symptoms were assessed using the MDASI-PeriOp-GYN scale at preoperative, postoperative, and pre-discharge time points. Exploratory PCA identified symptom cluster structures. Based on Linear Mixed Model we explore heterogeneous symptom trajectories. Independent predictors were screened using logistic regression and restricted cubic splines. Model performance was evaluated using ROC curves, calibration curves, and decision curve analysis.

RESULTS

Five stable symptom clusters were determined: a general somatic symptom cluster, a psycho-neurological symptom cluster, a gastrointestinal motility and energy cluster, an emotional and eating disorder cluster, and a consciousness and sedation-related cluster; and patients were classified into three trajectories: well-recovered, moderately persistent and severely distressed types. Age, operative duration, fibrinogen, neutrophil-to-lymphocyte ratio and D-dimer were risk factors, while serum albumin and serum potassium were protective factors. The seven-indicator combined model had an AUC of 0.863, with sensitivity 68.4%, specificity 87.8% and accuracy 82.0%, possessing superior predictive effect compared with single indexes.

CONCLUSION

Patients in the severe distress trajectory showed rapid symptom changes around discharge, representing a high-risk group for perioperative management. The Nomogram prediction model effectively stratifies patients into high- and low-risk groups for severe symptom distress, enabling precise interventions, improving transitional care, and enhancing perioperative symptom management.