成年胰腺癌患者全麻术后肺部并发症的危险因素

Risk factors of postoperative pulmonary complication in adult patients with pancreatic cancer under general anesthesia

目的:探讨影响成年胰腺癌患者术后肺部并发症(postoperative pulmonary complication, PPC)发生的相关危险因素。方法:根据纳入排除标准有效纳入1 162例患者,通过临床电子系统数据库进行回顾性数据检索,收集患者围手术期相关资料。计算全身炎症指数(systemic inflammatory index, SII)、预后营养指数(prognostic nutritional index, PNI)、中性粒细胞-淋巴细胞比值(neutrophil-lymphocyte ratio, NLR)、血小板-淋巴细胞比值(platelet-lymphocyte ratio, PLR)和淋巴细胞-单核细胞比值(lymphocyte-monocyte ratio, LMR),以描述术前患者炎症状态及营养情况。采用Logistic回归分析确定危险因素,进行受试者操作特征(receiver operating characteristic, ROC)曲线分析模型预测价值,并采用多因素Cox回归探讨患者术后30 d生存情况。结果:根据是否发生PPC将患者分为PPC组[39例(3.4%)]和非PPC组[Non-PPC组,1 123例(96.6%)]。与Non-PPC组患者比较:PPC组患者年龄较大( P<0.05),且男性患者比例较高( P<0.05);麻醉时间、手术时间、术后拔除胃管时间、术后进食时间、总住院天数、ICU治疗天数、术后总治疗天数较长( P<0.05),胃管拔除后再插比例、术后进食后再停比例、放弃治疗(院内未死亡)比例及院内死亡比例较高( P<0.05),术后2 d内排气排便比例、康复出院比例和术后30 d生存率较低( P<0.05);胶体液使用量较多( P<0.05),术中出血量较多( P<0.05),且有较大比例的患者进行输血( P<0.05);术前前白蛋白及白蛋白、低密度脂蛋白胆固醇、LMR、PNI降低( P<0.05),中性粒细胞百分比、中性粒细胞计数、单核细胞计数、淋巴细胞计数、红细胞分布宽度、血清降钙素原、CEA、SII、NLR、碱性磷酸酶、胱抑素C及Cr升高( P<0.05)。其他指标两组差异无统计学意义( P>0.05)。采用Logistic回归分析进行PPC危险因素建模,结果显示出血量、SII、NLR、PLR、中性粒细胞计数及百分比为独立危险因素。ROC曲线分析显示:曲线下面积(area under curve, AUC)为0.948(95%CI 0.919~0.978),敏感度为0.846,特异度为0.914,约登指数为0.760。Cox生存分析发现出血量[比值比(odds ratio, OR)1.616,95%CI 1.045~2.500, P=0.031]、SII( OR 1.004,95%CI 1.000~1.007, P=0.047)、PLR( OR 0.961,95%CI 0.931~0.993, P=0.016)及PPC( OR 0.018,95%CI 0.002~0.138, P<0.001)与术后30 d患者病死率呈正相关。 结论:术中出血量、术前SII、术前NLR、术前PLR及术前中性粒细胞计数和百分比是PPC发生的独立危险因素。

Objective To explore the risk factors of postoperative pulmonary complication(PPC)in adult patients with pancre‑atic cancer.Method A total of 1162 patients were selected according to the inclusion and exclusion criteria.Retrospective retriev‑al was conducted using the clinical electronic system database to collect their clinical data during the perioperative period.The system‑ic inflammatory index(SII),prognostic nutritional index(PNI),neutrophil‑lymphocyte ratio(NLR),platelet‑lymphocyte ratio(PLR)and lymphocyte‑monocyte ratio(LMR)were calculated to determine the inflammatory and nutritional conditions of patients before opera‑tion.Logistic regression analysis was performed to determine risk factors.A receiver operating characteristic(ROC)curve was plotted to evaluate its predicting value.Multivariate Cox regression was used to explore the survival of patients 30 days after surgery.Result According to the presence of PPC,the patients were divided into two groups:a PPC group(n=39,3.4%)and a Non‑PPC group(n=1123,96.6%).Compared with the Non‑PPC group,patients in the PPC group had increased age(P<0.05),with an elevated propor‑tion of male patients(P<0.05).The PPC group presented increases in the duration of anesthesia and operation,the time to withdraw gastric tube after surgery,postoperative feeding time,the total length of hospitalization stay,the length of intensive care unit(ICU)treat‑ment and the total duration of postoperative treatment(P<0.05).The PPC group presented increases the proportion of re‑insertion after gastric tube withdrawal,the proportion of re‑stop after postoperative feeding,the proportion of abandonment of treatment(no death in hospital)and the proportion of in‑hospital death(P<0.05),as well as decreases in the proportion of exhaust and defecation within 2 days after surgery,and the proportion of recovery and discharge and the survival rate 30 days after surgery(P<0.05).The PPC group con‑sumed increased volumes of colloidal liquid(P<0.05),with elevated blood loss(P<0.05),and an increased proportion of patients under‑went blood transfusion(P<0.05).The PPC group showed decreases in pre‑operative albumin and albumin,low density lipoprotein cho‑lesterol,LMR and PNI(P<0.05),as well as increases in neutrophil percentage,neutrophil count,monocyte count,lymphocyte count,erythrocyte distribution width,serum procalcitonin,CEA,SII,NLR,alkaline phosphatase,cystatin C and creatinine(Cr)(P<0.05).There was no statistical difference in other indexes between the two groups(P>0.05).The risk factors of PPC were modeled through lo‑gistic regression analysis.The results indicated that blood loss,SII,NLR,PLR,neutrophils count and percentage were independent risk factors.ROC curve analysis showed that the area under curve(AUC)was 0.948[95%confidence interval(CI)0.919,0.978],the sensi‑tivity was 0.846,the specificity was 0.914,and the Yoden index was 0.760.Cox survival analysis showed that blood loss[odds ratio(OR)1.616(95%CI 1.045,2.500),P=0.031],SII[OR 1.004(95%CI 1.000,1.007),P=0.047],and PLR[OR 0.961(95%CI 0.931,0.993),P=0.016]and postoperative pulmonary complications[OR 0.018(95%CI 0.002,0.138),P<0.001]were positively correlated with the mortality 30 days after surgery.Conclusions Intraoperative blood loss,preoperative SII,preoperative NLR,preoperative PLR,and preoperative neutrophil count and percentage are the independent risk factors for PPC.