Urology Unit, Azienda Ospedaliero-Universitaria delle Marche, Università Politecnica delle Marche, Via Conca 71, Ancona, 60126, Italy.
Urology Unit, ULSS 7 Pedemontana, Bassano del Grappa, Vicenza, Italy.
World J Urol. 2024 Nov 1;42(1):612. doi: 10.1007/s00345-024-05314-5.
To create a machine-learning model for estimating the likelihood of post-retrograde intrarenal surgery (RIRS) sepsis.
All consecutive patients with kidney stone(s) only undergoing RIRS in 16 centers were prospectively included (January 2022-August 2023).
adult, renal stone(s) only, CT scan (within three months), mid-stream urine culture (within 10 days).
concomitant ureteral stone, bilateral procedures. In case of symptomatic infection/asymptomatic bacteriuria, patients were given six days of antibiotics according to susceptibility profiles. All patients had antibiotics prophylaxis. Variables selected for the model: age, gender, age-adjusted Charlson Comorbidity Index, stone volume, indwelling preoperative bladder catheter, urine culture, single/multiple stones, indwelling preoperative stent/nephrostomy, ureteric access sheath, surgical time. Analysis was conducted using Python programming language, with Pandas library and machine learning models implemented using the Scikit-learn library. Machine learning algorithms tested: Decision Tree, Random Forest, Gradient Boosting. Overall performance was accurately estimated by K-Fold cross-validation with three folds.
1552 patients were included. There were 20 (1.3%) sepsis cases, 16 (1.0%) septic shock cases, and three more cases (0.2%) of sepsis-related deaths. Random Forest model showed the best performance (precision = 1.00; recall = 0.86; F1 score = 0.92; accuracy = 0.92). A web-based interface of the predictive model was built and is available at https://emabal.pythonanywhere.com/ CONCLUSIONS: Our model can predict post-RIRS sepsis with high accuracy and might facilitate patient selection for day-surgery procedures and identify patients at higher risk of sepsis who deserve extreme attention for prompt identification and treatment.
建立一种用于评估逆行肾内手术后（RIRS）发生脓毒症可能性的机器学习模型。
16 个中心连续前瞻性纳入仅接受 RIRS 治疗的肾结石患者（2022 年 1 月至 2023 年 8 月）。
成人，仅存在肾结石，CT 扫描（三个月内），中段尿培养（10 天内）。
同时存在输尿管结石，双侧手术。如果存在症状性感染/无症状菌尿，根据药敏谱给予患者六天的抗生素治疗。所有患者均接受抗生素预防。为模型选择的变量：年龄、性别、年龄调整 Charlson 合并症指数、结石体积、术前留置膀胱导管、尿液培养、单发/多发结石、术前留置支架/经皮肾造瘘管、输尿管进入鞘、手术时间。使用 Python 编程语言、Pandas 库和 Scikit-learn 库中的机器学习模型进行分析。测试的机器学习算法：决策树、随机森林、梯度提升。使用三折 K 折交叉验证准确估计整体性能。
共纳入 1552 例患者。发生脓毒症 20 例（1.3%），脓毒性休克 16 例（1.0%），另有 3 例（0.2%）与脓毒症相关的死亡。随机森林模型显示出最佳性能（精度=1.00；召回率=0.86；F1 评分=0.92；准确率=0.92）。已构建预测模型的基于网络的界面，并可在 https://emabal.pythonanywhere.com/ 上访问。
我们的模型可以高度准确地预测 RIRS 后脓毒症，可能有助于患者选择日间手术程序，并识别出更容易发生脓毒症的患者，需要对其进行密切关注，以便及时识别和治疗。
