摘要
目的构建老年前列腺增生(BPH)合并糖尿病(DM)患者经尿道前列腺电切术(TURP)术后尿道狭窄(US)的预测模型,并进行外部验证,旨在为临床防治提供参考依据。方法回顾性选取2021年3月至2022年8月于新乡医学院第一附属医院接受TURP治疗的老年BPH合并DM患者365例作为建模样本,根据术后3个月是否发生US分为狭窄组42例与未狭窄组323例。比较两组患者的临床资料,采用Lasso-Logistic回归分析术后US的独立影响因素,构建列线预测模型,采用校正曲线、受试者工作特征曲线(ROC)和决策曲线分析(DCA)评估预测模型的预测价值。另选取2022年9月至2023年9月于新乡医学院第一附属医院接受TURP治疗的BPH合并DM患者127例作为验证样本,分为狭窄组14例与未狭窄组113例,采用ROC和DCA曲线验证预测模型在验证样本中的预测价值。结果建模样本、验证样本中两组的前列腺体积、DM病程、国际前列腺症状(IPSS)评分、空腹血糖(FBG)、糖化血红蛋白(HbA1c)、术前合并尿路感染、既往导尿史、手术时间、术中出血量、术后导尿管留置时间、术后导尿管牵引时间比较,差异均有统计学意义(P<0.05);经Lasso-Logistic回归分析结果显示,前列腺体积、DM病程、IPSS评分、FBG、HbA1c、术前合并尿路感染、手术时间、术后导尿管留置时间均是老年BPH合并DM患者TURP术后US的独立危险因素(P<0.05);构建列线预测模型,校正曲线显示,该模型一致性指数(C-index)为0.879,校准度为0.902,ROC显示,该模型预测建模样本术后US的AUC为0.855(95%CI:0.782~0.928),预测验证样本术后US的AUC为0.935(95%CI:0.852~1.000),DCA曲线显示,该模型净获益优于两种极端情况,具有较高临床正向净获益。结论前列腺体积、DM病程、IPSS评分、FBG、HbA1c、术前合并尿路感染、手术时间、术后导尿管留置时间是老年BPH合并DM患者TURP术后发生US的独立危险因素,基于上述因素构建列线预测模型具有较高预测效能及临床适用性,可为临床预测TURP术后US的发生风险提供可靠预测方式。
Objective To construct a prediction model of urethral stricture(US)after transurethral resection of prostate(TURP)in elderly patients with benign prostatic hyperplasia(BPH)combined with diabetes(DM),and to conduct external validation,in order to provide a reference for clinical prevention and treatment.Methods A retrospective study was conducted to select 365 elderly patients with BPH and DM who underwent TURP treatment at the First Affiliated Hospital of Xinxiang Medical University from March 2021 to August 2022 as modeling samples.According to whether US occurred 3 months after surgery,the patients were divided into a stenosis group of 42 cases and a non-stenosis group of 323 cases.The clinical data of the two groups of patients were compared,and Lasso-Logistic regression analysis was used to identify independent factors affecting postoperative US.A nomogram prediction model was constructed,and calibration curves,receiver operating characteristic curves(ROC),and decision curve analysis(DCA)were used to evaluate the predictive value of the prediction model.In addition,127 patients with BPH and DM who underwent TURP treatment at the First Affiliated Hospital of Xinxiang Medical University from September 2022 to September 2023 were selected as validation samples,and they were divided into a stenosis group of 14 cases and a non-stenosis group of 113 cases.ROC and DCA curves were used to verify the predictive value of the prediction model in the validation samples.Results The differences in prostate volume,duration of diabetes,International Prostate Symptom Score(IPSS),fasting blood glucose(FBG),glycosylated hemoglobin(HbA1c),preoperative urinary tract infection,previous catheterization history,operation time,intraoperative blood loss,postoperative catheter indwelling time,and postoperative catheter traction time between the two groups in the modeling sample and validation sample were statistically significant(P<0.05).The results of Lasso-Logistic regression analysis showed that prostate volume,duration of DM,IPSS score,FBG,HbA1c,preoperative urinary tract infection,operation time,and postoperative catheterization time were independent risk factors for US in elderly BPH patients with DM after TURP(P<0.05).The linear prediction model was constructed,and the calibration curve showed that the consistency index(C-index)of the model was 0.879 and the calibration degree was 0.902.The ROC curve showed that the AUC of US predicted by the model after operation was 0.855(95%CI:0.782 to 0.928)for modeling samples and 0.935(95%CI:0.852 to 1.000)for validation samples.DCA curve showed that the net benefit of this model was better than that of the two extreme cases,with a high clinical positive net benefit.Conclusion Prostate volume,DM course,IPSS score,FBG,HbA1c,preoperative urinary tract infection,operative time,and postoperative catheter retention time are independent risk factors for US after TURP in elderly patients with BPH and DM.The linear prediction model constructed based on these factors has high predictive efficacy and clinical applicability,and it can provide a reliable way to predict the risk of US after TURP.
作者
张冠英
徐云
马阔
吴春磊
余沁楠
ZHANG Guan-ying;XU Yun;MA Kuo;WU Chun-lei;YU Qin-nan(Department of Urology,the First Affiliated Hospital of Xinxiang Medical University,Xinxiang 453100,Henan,CHINA;Department of Endocrinology,the First Affiliated Hospital of Xinxiang Medical University,Xinxiang 453100,Henan,CHINA)
出处
《海南医学》
CAS
2024年第11期1604-1610,共7页
Hainan Medical Journal
基金
新乡医学院第一附属医院博士科研基金(编号:xyyfy2014BS-003)。
关键词
前列腺增生
糖尿病
经尿道前列腺电切术
尿道狭窄
预测模型
危险因素
临床适用性
Prostatic hyperplasia
Diabetes mellitus
Transurethral resection of prostate
Urethral stenosis
Prediction model
Risk factors
Clinical applicability
