贲门癌切除术后围术期死亡原因分析及术后并发症logistic回归分析及风险预测模型的建立

发布时间：2018-05-18 18:33

本文选题：贲门癌 + 危险因素　；参考：《广东医学》2017年15期

【摘要】：目的探讨并分析贲门癌围术期死亡原因及术后并发症危险因素,建立logistics回归模型。方法收集305例贲门癌切除术患者的临床资料,根据围术期及住院期间术后有无发生并发症分为两组,将33个可能对贲门癌术后发生并发症有相关影响的有代表性的因素进行回顾性分析,通过计算机用logistic回归模型分析术后并发症相关危险因素,进行单因素、多因素、相关性及共线性检验分析,并建立风险预测模型,总结实际意义。结果305例贲门癌中手术死亡7例,手术死亡率2.3%。死亡原因:循环系统并发症2例(包括心源性休克并急性冠脉综合征1例,心源性猝死1例,占死亡组的28.60%),呼吸系统并发症2例(包括肺部感染、重症肺炎、肺不张等引起的呼吸衰竭2例,占死亡组的28.60%),呼吸系统与循环系统并发症并存3例(包括呼吸循环衰竭2例,呼吸衰竭并心肌梗死1例,占死亡组的42.80%),吻合口瘘1例(14.30%),乳糜胸1例(14.30%)。单因素logistic回归分析显示,在所分析的33个因素中,有9个因素与贲门癌切除术后发生并发症有关,分别为术前合并心脏病、病变部位、手术时长、手术切除范围、手术年代、术中输血、术中出血量、肠内营养时间、重症监护室(ICU)治疗时间,差异有统计学意义(P0.05)。多因素logistic回归分析提示:术前合并心脏病、手术切除范围、术中输血、肠内营养时间等指标有统计学意义(P0.05),术前合并心脏病为独立危险因素,术中输血为保护因素,并进入logistic回归方程,获得预测模型P=1/(1+e~((18.256-1.079X33+0.963X19-0.788X26+0.725X30)))。结论合并心脏病的贲门癌患者手术评估需严格掌握,欲防止并发症发生,围术期需注意心功能的动态变化,保证心功能能够维持机体正常生理功能;手术切除范围愈大,术后并发症发生风险愈高,故术中应根据病变部位和范围在切除病变的前提下尽可能减少创伤,有助于减少术后并发症的发生;贲门癌手术创伤较大,术中予输血有助于术后减少术后并发症发生;肠内营养时间的合理性与科学性能影响围术期并发症发生率,故需结合具体病情开始肠内营养时间并需动态观察营养前后机体的变化。
[Abstract]:Objective to investigate the causes of perioperative death and risk factors of postoperative complications of cardiac carcinoma and to establish a logistics regression model. Methods the clinical data of 305 patients with cardiac cancer resection were collected and divided into two groups according to the complications during perioperative period and hospitalization. A retrospective analysis of 33 representative factors related to postoperative complications of cardiac cancer was carried out. The related risk factors of postoperative complications were analyzed by logistic regression model. Correlation and collinearity analysis and risk prediction model are established to summarize the practical significance. Results among 305 cases of cardiac carcinoma, 7 cases died, and the operative death rate was 2.3%. The causes of death included 2 cases of circulatory complications (including 1 case of cardiogenic shock and acute coronary syndrome, 1 case of sudden cardiac death, 28.60% of death group), 2 cases of respiratory complications (including pulmonary infection, severe pneumonia). There were 2 cases of respiratory failure caused by atelectasis, 28.60% of death group, 3 cases of respiratory and circulatory complications (including 2 cases of respiratory and circulatory failure, 1 case of respiratory failure and myocardial infarction). In the death group, 42.80 cases had anastomotic fistula, 1 case had anastomotic fistula and 1 case had chylothorax. Univariate logistic regression analysis showed that 9 of the 33 factors were associated with postoperative complications of cardiac cancer, including preoperative heart disease, location of lesion, length of operation, resection range, and age of operation. There were significant differences in blood transfusion, intraoperative blood loss, enteral nutrition time and ICU treatment time in intensive care unit (ICU). Multivariate logistic regression analysis showed that preoperative heart disease, surgical resection range, intraoperative blood transfusion, time of enteral nutrition were statistically significant (P 0.05), preoperative heart disease was an independent risk factor, and intraoperative blood transfusion was a protective factor. After entering the logistic regression equation, the prediction model P ~ (1 / 1) e~((18.256-1.079X33 0.963X19-0.788X26 0.725 X _ (30) was obtained. Conclusion the surgical evaluation of cardiac cancer patients with heart disease should be strictly grasped. In order to prevent complications, attention should be paid to the dynamic changes of cardiac function in perioperative period to ensure that cardiac function can maintain normal physiological function. The higher the risk of postoperative complications is, the more trauma should be minimized under the premise of resection according to the location and scope of the lesion, which is helpful to reduce the incidence of postoperative complications. Intraoperative transfusion was helpful to reduce postoperative complications, and the rationality of enteral nutrition time and scientific performance affected the incidence of perioperative complications. Therefore, it is necessary to start enteral nutrition time combined with specific condition and observe the changes of body before and after nutrition.
【作者单位】：新疆医科大学第一附属医院胸外科;
【分类号】：R735

【相似文献】