LASSO Logistic Regression for Predicting Postoperative Severe Pain After Hepatic Hemangioma Ablation
Ruize Gao1, Fei Xu2, Yuntang Song2
1Department of Interventional Radiology, Beijing Chaoyang Hospital Affiliated to Capital Medical University, Beijing, 100043, People's Republic of China.
Purpose:
To develop a least absolute shrinkage and selection operator (LASSO) logistic regression to predict postoperative severe pain after thermal ablation of hepatic hemangioma (HH).
Patients And Methods:
From January 2014 to March 2024, 285 patients with HH treated by thermal ablation were retrospectively recruited. Forty-seven patients with postoperative severe pain [visual analogue scale (VAS) score ≥ 5] were matched 1:2 with 94 patients with mild pain (VAS score < 5). The LASSO and multivariate logistic regression identified independent risk factors for severe pain after thermal ablation for HH. The model's performance was evaluated using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA). Internal validation was performed using the Bootstrap method.
Results:
The ablation time (OR = 1.070, p = 0.046), postoperative levels of aspartate aminotransferase (AST) (OR = 1.012, p < 0.001), lactate dehydrogenase (LDH) (OR = 1.009, p = 0.001), neutrophil to lymphocyte ratio (NLR) (OR = 1.266, p = 0.034) were independent risk factors of severe pain. The model's area under the curve (AUC) = 0.985 (95% CI, 0.971-0.998). After internal verification by the Bootstrap method, the model still had a high discriminative ability (AUC = 0.979, 95% CI, 0.971-0.985). The calibration curve illustrated good agreement between the predicted and observed probability of severe pain. DCA verified that the model possesses significant predictive value.
Conclusion:
Our nomogram predicts postoperative severe pain for HH with good discrimination and calibration based on the easily available risk factors.
