前列腺穿刺活檢術術后感染的危險因素分析及預測模型建立

Analysis of risk factors and establishment of a prediction model for infection after prostate biopsy

SUN Penghao，SONG Wei

（Department of Urology，Shandong Provincial Hospital Affiliated to Shandong First Medical University，Jinan 250000，China）

ABSTRACT：Objective To analyze the risk factors leading to infection after prostate biopsy，establish a nomogram prediction model and verify it.Methods Clinical data of 523 patients who underwent ultrasound-guided prostate biopsy at our hospital during Jan.2023 and Jul.2024 were retrospectively analyzed.Patients were divided into an infection group and a non-infection group.Independent risk factors for infection after prostate biopsy were identified with univariate and multivariate binary logistic regression analyses，and a nomogram prediction model was constructed，which was validated with receiver operating characteristic （ROC） curve，calibration curve，and decision curve analysis （DCA）.Results Infection occurred in 54 cases （10.3%）.Univariate and multivariate logistic regression analyses showed that age gt;65 years （OR=3.535，P=0.003），diabetes （OR=5.693，Plt;0.001），hypoproteinemia （OR=8.936，Plt;0.001），preoperative urinary tract infection （OR=6.153，Plt;0.001），puncture needles gt;12 （OR=4.347，Plt;0.001），and transrectal puncture （OR=3.701，Plt;0.001） were independent risk factors for infection.Based on the multivariate logistic analysis results，a risk prediction nomogram model was constructed，with an area under the ROC curve （AUC） of 0.894.The calibration curve and DCA both indicated that the model had high predictive accuracy and clinical decision-making efficiency.Conclusion Age gt;65 years，diabetes，hypoproteinemia，preoperative urinary tract infection，puncture needles gt;12，and transrectal puncture are independent risk factors for infection after prostate biopsy.The nomogram prediction model based on these factors helps identify high-risk patients，thereby enabling individualized treatment plans to reduce the incidence of infection.

KEY WORDS：prostate cancer; prostatic biopsy; postoperative infection; nomogram prediction model

摘要：目的 分析前列腺穿刺活檢術（PB）患者術后發生感染的危險因素，建立列線圖預測模型并進行驗證。方法 收集2023年1月—2024年7月于山東第一醫科大學附屬省立醫院泌尿外科行超聲引導下PB的523例患者的臨床資料。根據術后是否出現感染將患者分為感染組和未感染組。采用單因素及多因素二元logistic回歸分析篩選影響PB術后感染的獨立危險因素，并構建列線圖預測模型。采用受試者工作特征（ROC）曲線、校準曲線和決策曲線分析（DCA）評估與驗證該列線圖模型的預測效能。結果 523例行PB的患者中有54例（10.3%）發生術后感染。單因素和多因素logistic回歸分析顯示，＞65歲（OR=3.535，P＝0.003）、糖尿病（OR=5.693，P＜0.001）、低蛋白血癥（OR=8.936，P＜0.001）、術前尿路感染（OR=6.153，P＜0.001）、穿刺針數＞12針（OR=4.347，P＜0.001）、經直腸穿刺（OR=3.701，P＜0.001）是PB術后感染的獨立危險因素?；诙嘁蛩豯ogistic回歸分析結果構建PB術后感染的風險預測列線圖模型，該模型ROC曲線下面積（AUC）為0.894，校準曲線和DCA均表明該模型具有較高的預測精度和臨床決策效率。結論 ＞65歲、有糖尿病、低蛋白血癥、術前尿路感染、穿刺針數＞12針、經直腸穿刺是PB術后感染的獨立危險因素?；谏鲜鲆蛩貥嫿ǖ牧芯€圖預測模型有助于篩選出PB術后感染的高?；颊撸瑥亩贫▊€體化治療方案，降低PB術后感染的發生率。

關鍵詞：前列腺癌；前列腺穿刺活檢術；術后感染；列線圖預測模型

中圖分類號：R619.3 文獻標志碼：ADOI：10.3969/j.issn.1009-8291.2025.02.006

收稿日期：2024-08-10 修回日期：2024-10-28

基金項目：山東省自然科學基金項目（No.ZR2021MH283）

通信作者：宋偉，主任醫師。E-mail：sss16273@126.com

作者簡介：孫朋浩，碩士研究生在讀。研究方向：前列腺癌診治。

E-mail：sph1627@163.com

前列……