【摘要】 目的 探讨精索静脉曲张(VC)术后疼痛不缓解的危险因素并构建预测模型。方法 回顾性分析贵州省人民医院2022年9月至2024年8月125例VC手术患者资料，按8∶2比例分为建模组(100例)和验证组(25例)。通过单因素、LASSO回归及多因素Logistic回归筛选独立预测因子，建立列线图模型，采用ROC曲线、校准曲线及Hosmer-Lemeshow检验评估模型效能。结果 左侧静脉内径≥3.3 mm(OR=4.106)、术前疼痛≥8个月(OR=0.707)、NRS评分≥6分(OR=2.126)及存在牵拉痛/刺痛(OR=19.573)是术后疼痛不缓解的独立预测因子(均P<0.05)。模型在建模组和验证组中AUC均优于单一指标，校准曲线显示预测值与实际值一致性良好(χ~2=4.794,P=0.797)。结论 显微镜下手术可有效缓解VC相关性疼痛；基于多参数构建的预测模型能量化手术获益概率，为个体化治疗提供循证支持，具有临床价值。更多还原

【Abstract】 Objective To analyze the risk factors for postoperative non-relief of varicocele-associated pain and establish a surgical benefit prediction model. Methods Clinical data of 125 patients, who were diagnosed with varicocele between September 2022 and August 2024 in Guizhou Provincial People’s Hospital were retrospectively analyzed.They were divided into the modelling group(80%,100 cases) and the validation group(20%, 25 cases) by random sampling method. Based on the results of pre-and post-surgery pain score(Numerical Rating Scale, NRS), they were subdivided into the valid group and the invalid group.The modeling group was subjected to one-way analysis and regression analysis using least absolute shrinkage and selection operator(LASSO)and multifactorial logistic regression analysis.The independent influencing factors that could predict the efficacy of the surgery were obtained and the prediction model was established. The predictive model was visualized by applying a column-line diagram. The efficacy of the model was evaluated using the receiver operating characteristic curve(ROC)and area under the curve(AUC), and the specificity and sensitivity of the model were determined by setting the cut off value from the ROC curve.The calibration curve was plotted to evaluate the calibration of the prediction model. Results The results of LASSO regression and multifactorial logistic regression analyses showed that the internal diameter of the left varicose vein, the duration of preoperative pain, the preoperative pain score, and the nature of the pain were the independent risk factors for the patients’ postoperative pain non-relief(P<0.05). A predictive model of surgical outcome was developed using the above indicators and it was visualized using a column-line diagram. In both the modelling and validation populations, ROC curve analysis showed that the AUC of the model for predicting surgical efficacy was higher than the AUC values corresponding to the other parameters, suggesting that the model had a better predictive assessment ability. Calibration curve analysis revealeda high consistency between the model predicted values and the actual observed values, suggesting that the model was well calibrated. The results of the hosmer and lemeshow test goodness-of-fit test further verified that the predictive model had a reliable calibration ability(χ~2=4.794,P=0.797>0.05). Conclusion The prediction model based on multidimensional preoperative parameters of varicocele-associated pain shows a significant advantage over traditional univariate prediction indicators in surgical benefit evaluation. This model can provide evidence-based decision support for implementation of individual surgical decision-making, and has good clinical application value.更多还原