Abstract:
Objective
To use data-driven method for feature screening and establish a moderate to severe pain risk prediction model after microwave ablation of liver cancer.
Methods
A total of 239 patients undergoing microwave ablation of hepatoma under local anesthesia who were hospitalized in the Department of Interventional Oncology Affiliated to Guangxi Medical University from January 2018 to December 2023 were selected as the study objects, and were divided into mild pain group and moderate to severe pain group according to postoperative pain scale (VAS).LASSO model was used to screen the characteristics of pain after microwave ablation of liver cancer, and a prediction model was constructed.The prediction model was internally verified by bootstrap resampling method.
Results
126 of the 239 patients had moderate to severe pain after surgery, and the incidence of moderate to severe pain was 52.72%.Based on the data driven feature screening results, Age, sex, cirrhosis, primary disease, Child-Pugh grade, previous surgery,lesion diameter, number of lesions, distance from liver envelope, ablation time, pain disaster, anxiety and depression were included as predictors of moderate to severe pain after microwave ablation of liver cancer.Internal verification by Bootstrap method showed that the average areaunder ROC curve (AUC) was 0.978,C-Index was 0.978, AUC was 0.882, model specificity was 0.941,and sensitivity was 0.792.Based on the above results, the nomogram of moderate and severe postoperative pain was constructed.
Conclusion
The verification results show that the calibration and differentiation of the model are good,and the moderate and severe pain nomogram based on this model has good predictive efficacy and clinical application value.
Key words:
Liver cancer,
Microwave ablation,
Pain,
A nomogram,
Prediction model
Qiaoling Wei, Yan Huang, Chang Zhao, Qingfeng Song, Zuyi Chen, Ying Huang, Chang Meng, Jing Huang. Construction and verification of moderate to severe pain risk prediction model after microwave ablation of liver cancer[J]. Chinese Journal of Clinicians(Electronic Edition), 2024, 18(08): 717-723.