作者：卞文鑫，赵继红

单位：

210000江苏省南京市，南京医科大学第一附属医院妇幼心血管内科

Units：

Department of Maternal and Child Cardiovascular Medicine, the First Affiliated Hospital of Nanjing Medical University,Nanjing 210000, China

关键词：

急性冠脉综合征；心房颤动；经皮冠状动脉介入治疗；血管并发症；影响因素分析；列线图

Keywords：

Acute coronary syndrome; Atrial fibrillation; Percutaneous coronary intervention; Vascular complications;Root cause analysis; Nomograms

CLC：

R 741.4 R 541.75

DOI：

10.12114/j.issn.1008-5971.2023.00.272

Funds：

2020年江苏省干部保健科研课题项目（BJ20014）

摘要：

目的　探讨急性冠脉综合征（ACS）合并心房颤动患者PCI后穿刺部位发生血管并发症的影响因素，构建其风险预测列线图模型并进行验证。方法　选取2020—2022年于南京医科大学第一附属医院行PCI的ACS合并心房颤动患者为研究对象，根据PCI后3 d内穿刺部位是否发生血管并发症将患者分为发生组和未发生组。收集患者的临床资料，采用多因素Logistic回归分析探讨ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的影响因素；采用R 3.5.3软件构建ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的风险预测列线图模型；采用Bootstrap法（重复抽样1 000次）进行内部验证，计算一致性指数；采用Hosmer-Lemeshoe拟合优度检验和校准曲线评价该列线图模型的拟合程度；采用ROC曲线分析该列线图模型对ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的预测价值；绘制决策曲线以评价该列线图模型的临床有效性。结果　共入组269例患者，发生血管并发症64例（23.8%）。两组手术时机、手术时间、压迫止血方式、压迫止血时间、术后抗血栓治疗方式和术后HAS-BLED出血评分比较，差异有统计学意义（P＜0.05）。多因素Logistic回归分析结果显示，手术时机、手术时间、压迫止血方式、压迫止血时间、术后抗血栓治疗方式和术后HAS-BLED出血评分是ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的独立影响因素（P＜0.05）。基于多因素Logistic回归分析结果，构建ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的风险预测列线图模型。该列线图模型的一致性指数为0.761〔95%CI（0.729，0.794）〕；Hosmer-Lemeshoe拟合优度检验结果显示，该列线图模型拟合较好（χ2=4.247，P=0.213）。ROC曲线分析结果显示，该列线图模型预测ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的AUC为0.767〔95%CI（0.730，0.803）〕。决策曲线分析结果显示，当该列线图模型预测ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的风险阈值在4%～72%范围时，患者的净获益率＞0。结论　急诊手术、手术时间≥2 h、弹力加压绷带止血、压迫止血时间≥3 h、术后三联抗血栓治疗和术后HAS-BLED出血评分≥3分是ACS合并心房颤动患者PCI后穿刺部位发生血管并发症的危险因素，基于上述危险因素构建的列线图模型对ACS合并心房颤动患者PCI后穿刺部位发生血管并发症具有一定预测价值。

Abstract：

Objective To explore the influencing factors of vascular complications at puncture site in acute coronarysyndrome (ACS) patients complicated with atrial fibrillation after PCI, and to construct and validate the nomogram model forpredicting its risk. Methods ACS patients complicated with atrial fibrillation who underwent PCI at the First Affiliated Hospitalof Nanjing Medical University from 2020 to 2022 were selected as the research subjects. The patients were divided into theoccurrence group and the non-occurrence group according to whether vascular complications occurred at puncture site within3 d after PCI. The clinical data of the patients were collected, and multivariate Logistic regression analysis was used to analyzethe influencing factors of vascular complications at puncture site in ACS patients complicated with atrial fibrillation after PCI.The nomogram model for predicting the risk of vascular complications at puncture site in ACS patients complicated with atrial fibrillation after PCI was constructed by using the R 3.5.3 software. The internal validation was performed by the Bootstrap method(1 000 repetitive samples) , and the consistency index was calculated. Hosmer-Lemeshoe goodness of fit test and calibrationcurve were used to evaluate the fitting degree of the nomogram model. ROC curve was used to analyze the predictive value ofthe nomogram model for vascular complications at puncture site in ACS patients complicated with atrial fibrillation after PCI.The decision curve was drawn to evaluate the clinical effectiveness of the nomogram model. Results A total of 269 patientswere enrolled, and 64 cases (23.8%) had vascular complications. There were significant differences in operation opportunity,operative time, compression hemostasis method, compression hemostasis time, postoperative anti-thrombotic treatment methodand postoperative HAS-BLED bleeding score between the two groups (P < 0.05) . Multivariate Logistic regression analysis showedthat operation opportunity, operative time, compression hemostasis method, compression hemostasis time, postoperative antithrombotic treatment method and postoperative HAS-BLED bleeding score were the influencing factors of vascular complicationsat puncture site in ACS patients complicated with atrial fibrillation after PCI (P < 0.05) . The nomogram model for predictingvascular complications at puncture site in ACS patients complicated with atrial fibrillation after PCI was constructed based onthe multivariate Logistic regression analysis results. The consistency index of the nomogram model was 0.761 [95%CI (0.729,0.794) ] . The results of Hosmer-Lemeshoe goodness of fit test showed that the nomogram model fit well (χ2=4.247, P=0.213) .The results of ROC curve analysis showed that the AUC of the nomogram model for predicting vascular complications at puncturesite in ACS patients complicated with atrial fibrillation after PCI was 0.767 [95%CI (0.730, 0.803) ] . The results of decision curveanalysis showed that when the risk threshold of the nomogram model for predicting vascular complications at puncture site in ACSpatients complicated with atrial fibrillation after PCI was 4%-72%, the net benefit rate of patients was greater than 0. ConclusionEmergency operation, operative time ≥ 2 h, elastic compression bandage for hemostasis, compression hemostasis time ≥ 3 h,postoperative triple anti-thrombotic therapy, and postoperative HAS-BLED bleeding score ≥ 3 are the risk factors for vascularcomplications at puncture site in ACS patients complicated with atrial fibrillation after PCI. The nomogram model constructedbased on the above risk factors has a certain predictive value for vascular complications at puncture site in ACS patientscomplicated with atrial fibrillation after PCI.

ReferenceList：