摘要目的:探究腹部大手术后肺部并发症（postoperative pulmonary complications, PPC）的相关危险因素，建立一个风险预测评分模型。方法:回顾性纳入1 538例于2016年1月至2019年6月间接受腹部大手术患者的临床资料，通过单因素分析及多元logistic回归分析从人口统计学、实验室检查指标、手术信息等相关变量中筛选出PPC的独立危险因素，建立一个风险预测评分模型并验证。结果:1 538例患者在PPC分级中，分级≥3级（严重并发症）患者332例（21.6%），分级≤2级（非严重并发症）患者1 206例（78.4%）。单因素分析确认相关变量中性别、年龄、ASA分级、卒中史、术前SpO 2、急诊手术、上腹部手术、手术时间、术中输液量、术中输血浆量、术中出血量、术中尿量、血小板水平、ALT水平、AST水平、肌钙蛋白水平有意义（ P<0.05）。多元logistic回归分析发现ASA分级、卒中史、SpO 2≤97%、急诊手术、上腹部手术是腹部大手术患者PPC的独立危险因素，根据β值对各因素赋值，建立风险预测评分并验证，绘制受试者工作特性曲线（receiver operating characteristic curve, ROC），计算其曲线下面积（area under the cure，AUC）为0.60（95%CI 0.57~0.62）。 结论:ASA分级、卒中史、SpO 2≤97%、急诊手术、上腹部手术是腹部大手术患者发生PPC的独立危险因素，本研究建立的风险预测评分有助于预测PPC的发生，并需要进一步完善。

abstractsObjective:To investigate the risk factors of postoperative pulmonary complications (PPC) after major abdominal surgery.Methods:Clinical data were collected from 1 538 patients who underwent major abdominal surgery from January 2016 to June 2019 for retrospective analysis. Logistic regression analyses were performed to screen out the independent risk factors for PPC from relevant variables in demographical information, laboratory indicators and surgical information, to establish and verify a simple prediction model.Results:Among the 1 538 patients, there were 332 (21.6%) patients at ≥ grade 3 (major pulmonary complications), and 1 206 (78.4%) patients at grade ≤2 (without pulmonary complications). Univariate analysis confirmed that sex, age, American Society of Anesthesiologists (ASA) grade, stroke history, preoperative pulse oxygen saturation (SpO 2), emergency operation, upper abdominal surgery, operation duration; intraoperative infusion volume, plasma transfusion, blood loss, and urine volume; levels of platelet, alanine aminotransferase (ALT), aspartate aminotransferase (AST), and troponin were significant factors ( P<0.05). Multivariate logistic regression analysis showed that ASA grade, stroke history, SpO 2≤97%, emergency surgery, and upper abdominal surgery were the independent risks of PPC in patients with major abdominal surgery. A risk scoring system was established through assigning values to each factor according to the β value. A receiver operating characteristic curve (ROC) was plotted and the area under the cure curve (AUC) was calculated. The result was 0.60 (95%CI 0.57, 0.62). Conclusions:ASA grade, stroke history, SpO 2≤97%, emergency surgery, and upper abdominal surgery are the independent risk factors for PPC in patients after major abdominal surgery. The risk scoring system established in this study is helpful to predict the occurrence of PPC and needs further improvement.