急性心肌梗死患者直接经皮冠状动脉介入治疗成功后住院期间心力衰竭的预测因素

Predictors of in-hospital heart failure in patients undergoing successful primary percutaneous coronary intervention for acute myocardial infarction

目的：本研究旨在明确ST段抬高心肌梗死（STEMI）患者成功行直接经皮冠状动脉介入治疗（PCI）后住院期间发生心力衰竭（HF）的预测因素。方法回顾性分析接受直接PCI成功治疗的初发STEMI患者的临床和冠状动脉造影资料，根据住院期间是否发生HF将患者分为HF组和无HF组。确定住院期间HF的发生率、预测因素及其对预后的影响。结果共入选患者834例，男662例（79.4%），年龄（62.9±12.9）岁。其中，HF组94例（11.3%），无HF组740例（88.7%）。HF组的30 d全因死亡率显著高于无HF组（24.5%比1.5%，P＜0.001）。Cox回归分析显示，犯罪血管为前降支（HR 2.173，95% CI 1.12～4.212，P=0.022）、ln 24 h N末端B型利钠肽原（NT-proBNP）（HR 1.904，95% CI 1.479～2.452，P＜0.001）、24 h超敏C反应蛋白（hsCRP）≥11.0 mg/L（中位数）（HR 2.901，95% CI 1.309～6.430，P=0.009）和基线血糖（HR 1.022，95% CI 1.000～1.044，P=0.046）是住院期间发生HF的独立预测因素。受试者工作曲线显示，以24 h NT-proBNP≥1171 pg/ml为阈值诊断住院期间HF的敏感性和特异性分别为92.5%和76.8%（c=0.883， P＜0.001），以24 h hsCRP≥13.5 mg/L为阈值诊断住院期间HF的敏感性和特异性分别为86.0%和77.0%（c=0.829，P＜0.001）。在犯罪血管为前降支的患者中，24 h NT-proBNP＜1171 pg/ml且24 h hsCRP＜13.5 mg/L的患者住院期间HF的发生率为0.4%，而24 h NT-proBNP≥1171 pg/ml且24 h hsCRP≥13.5 mg/L的患者住院期间HF的发生率为60.9%，两者差异有统计学意义（P＜0.001）。结论 STEMI患者即使接受直接PCI成功治疗，其住院期间HF的发生率仍然较高，发生HF者预后差。犯罪血管为前降支、hsCRP、NT-proBNP和基线血糖是住院期间发生HF的独立预测因素。检测并联合应用不同的血清生物标记物是预测STEMI患者直接PCI术后住院期间发生HF的有效方法。

Objectives We sought to determine the factors that predicted in-hospital heart failure（HF）in patients undergoing successful primary percutaneous coronary intervention（PCI）for ST-segment elevation myocardial infarction（STEMI）. Methods The clinical and angiographic data were retrospectively reviewed in patients undergoing successful primary PCI for their ifrst STEMI. According to the occurrence of in-hospital HF, patients were divided into HF group and non-HF group. The incidence and predictors of in-hospital HF and its impact on prognosis were determined. Results A total of 834 patients were included, among them 94 patients （11.3%） were in the HF group and 740 patients（88.7%） were in the non-HF group. The mean age was （62.9±12.9） years and 662 patients （79.4%） were male. All-cause mortality at 30 days was signiifcantly higher in the HF group than in the non-HF group （24.5%vs. 1.5%, P〈0.001）. In Cox regression analysis, left anterior descending artery （LAD） as the culprit vessel （HR 2.173, 95% CI 1.12~4.212, P=0.022）, ln 24 h NT-proBNP （HR 1.904, 95%CI 1.479~2.452, P〈0.001）, 24 h hsCRP≥11.0 mg/L （median） （HR 2.901, 95%CI 1.309~6.430, P=0.009） and baseline serum glucose （HR 1.022, 95%CI 1.000 ~ 1.044, P=0.046） were independent predictors of in-hospital HF. Receiver operator characteristic analysis identiifed 24 h NT-proBNP ≥ 1171 pg/ml （c=0.883, P 〈 0.001） and 24 h hsCRP ≥ 13.5 mg/L （c=0.829, P 〈 0.001） were the best cut-off values in discriminating in-hospital HF with a sensitivity and speciifcity of 92.5%and 76.8%for 24 h NT-proBNP, 86.0%and 77.0%for 24 h hsCRP, respectively. Even among patients with LAD as the culprit vessel, the incidence of in-hospital HF was only 0.4%in patients whose 24 h NT-proBNP was〈1171 pg/ml and 24 h hsCRP was〈13.5 mg/L;while the incidence of in-hospital HF was up to 60.9%in patients whose 24 h NT-proBNP≥1171 pg/ml and 24 h hsCRP≥13.5 mg/L （P〈0.001）. Conclusions The incidence of in-hospital HF was still high in STEMI patients even after successful primary PCI. Patients with in-hospital HF had poor prognosis. LAD as the culprit vessel, hsCRP, NT-proBNP and baseline serum glucose were independent predictors of in-hospital HF. Assessment and combined use of different serum biomarkers were effective methods to estimate the risk of in-hospital HF in STEMI patients undergoing primary PCI.