胃癌根治术后非计划再手术对临床疗效的影响及其发生的危险因素分析（附4124例报告）

Effect of unplanned reoperation on clinical efficacy and its risk factor after radical resection of gastric cancer （A report of 4 124 cases）

目的：探讨胃癌根治术后非计划再手术对临床疗效的影响，及其发生的原因和危险因素。方法：采用回顾性病例对照研究方法。收集2005年1月至2014年12月福建医科大学附属协和医院收治的4 124例行胃癌根治术患者的临床病理资料。患者首次手术方式均采用开腹或腹腔镜胃癌根治术。观察指标：（1）首次手术治疗情况与病理学检查结果及随访情况。（2）术后恢复情况。（3）胃癌根治术后非计划再手术发生的原因及时间间隔。（4）影响胃癌根治术后非计划再手术的单因素分析。（5）影响胃癌根治术后非计划再手术的多因素分析。采用门诊、电话和微信等方式专人进行随访，了解患者术后30 d内的恢复情况。正态分布的计量资料以±s表示，组间比较采用独立样本t检验。计数资料和单因素分析采用x2检验或Fisher确切概率法检验。多因素分析采用logistic回归模型。结果：（1）首次手术治疗情况与病理学检查结果及随访情况：①4 124例患者首次手术治疗情况，行腹腔镜手术2 608例，行开腹手术1 516例；行全胃切除术2 259例，行远端胃切除术1 865例。②4 124例患者术后病理学检查结果：T1期883例、 T2期468例、T3期959例、T4a期1 814例；N0期1 414例、N1期571例、N2期683例、N3期1 456例；TNM分期Ⅰ期1 073例、Ⅱ期825例、Ⅲ期2 226例。③4 124例患者首次手术后30 d内全部获得随访，随访率为100.000%（4 124/4 124），其中术后行非计划再手术52例，术后未行非计划再手术4 072例，术后早期非计划再手术率为1.261%（52/4 124）。（2）术后恢复情况：4 072例术后未行非计划再手术患者中，575例发生并发症，并发症发生率为14.121%（575/4 072）；术后死亡17例，病死率为0.417%（17/4 072），术后住院时间为（14.0±9.0）d。52例术后行非计划再手术患者中，二次手术后23例发生并发症，并发症发生率为44.231%（23/52）；死亡6例，病死率为11.538%（6/52），术后住院时间为（28.0±13.0）d。两组患者上述指标比较，差异均有统计学意义（x2=37.550，t=10.900，P〈0.05）。（3）胃癌根治术后非计划再手术发生的原因及时间间隔：52例患者术后行非计划再手术距首次手术总时间间隔为（6.9±6.7）d。52例患者术后行非计划再手术的原因分别为：腹腔内出血23例（2例死亡）、吻合口出血7例、吻合口漏6例（2例死亡）、腹腔感染5例、小肠梗阻5例（1例死亡）、腹部切口裂开3例、小肠穿孔2例、胰液漏1例（患者死亡），其距首次手术时间间隔分别为（3.9±3.8）d、（0.9±0.5）d、（7.9±4.7）d、（14.9±4.6）d、（16.4±9.9）d、（10.0±6.0）d、 （6.7±5.2）d、12.0 d。（4）影响胃癌根治术后非计划再手术的单因素分析结果显示：患者年龄、体质量指数（BMI）和术中出血量是影响非计划再手术的危险因素（x2=5.468，7.589，5.041，P〈0.05）。（5）影响胃癌根治术后非计划再手术的多因素分析分析结果显示：患者年龄〉70岁、BMI〉25 kg/m^2及术中出血量〉100 mL是影响非计划再手术发生的独立危险因素（比值比=1.950，2.288，1.867；95%可信区间：1.074-3.538，1.230-4.257，1.067-3.267，P〈0.05）。结论：患者胃癌根治术后行非计划再手术后并发症发生率高、病死率高、住院时间长。胃癌根治术后腹腔内出血、吻合口出血和吻合口漏是非计划再手术发生的主要原因，患者年龄〉70岁、BMI〉25 kg/m2及术中出血量〉100 mL是影响胃癌根治术后非计划再手术发生的独立危险因素。

Objective：To explore the effect of unplanned reoperation （URO） on clinical efficacy after radical resection of gastric cancer （GC）, and its causes and risk factors analysis affecting URO. Methods：The retrospective casecontrol study was conducted. The clinicopathological data of 4 124 patients who underwent radical resection of GC in the Union Hospital of Fujian Medical University between January 2005 and December 2014 were collected. The initial operation was open or laparoscopic radical resection of GC. Observation indicators： （1） initial operation situations, results of pathologic examination and followup situations; （2） postoperative recovery situations; （3） causes and time interval of URO after radical resection of GC; （4） univariate analysis affecting URO after radical resection of GC; （5） multivariate analysis affecting URO after radical resection of GC. Followup using outpatient examination, telephone interview and Wechat was performed to detect postoperative 30day recovery of patients. Measurement data with normal distribution were represented as ±s, and comparisons between groups were done using the independentsample t test. Count data and univariate analysis were done using the chisquare test or Fisher exact probability. Multivariate analysis was done using the logistic regression model. Results：（1） Initial operation situations, results of pathologic examination and followup situations： ① Initial operation situations of 4 124 patients, 2 608 and 1 516 underwent respectively laparoscopic surgery and open surgery; 2 259 and 1 865 underwent respectively total gastrectomy and distal gastrectomy. ② Results of pathological examination of 4 124 patients： 883, 468, 959 and 1 814 were respectively in T1, T2, T3 and T4a stages; 1 414, 571, 683 and 1 456 were in N0, N1, N2 and N3 stages; 1 073, 825 and 2 226 were inⅠ,Ⅱ and Ⅲ stages. ③ All the 4 124 patients were followed up within 30 days after initial operation, with a followup rate of 100.000%（4 124/4 124）, including 52 with URO and 4 072 without URO, with a early URO rate of 1.261%（52/4 124）. （2） Postoperative recovery situations： of 4 072 patients without URO, 575 had postoperative complications, with an incidence of 14.121%（575/4 072）; 17 died after operation, with a mortality of 0.417%（17/4 072）, and duration of postoperative hospital stay was （14.0±9.0）days. Of 52 patients with URO, 23 had complications after reoperation, with an incidence of 44.231%（23/52）; 6 died after reoperation, with a mortality of 11.538%（6/52）, and duration of postoperative hospital stay was （28.0±13.0）days. There were statistically significant differences in above indicators between groups （x2=37.550, t=10.900, P〈0.05）. （3） Causes and time interval of URO after radical resection of GC： total time interval between initial operation and URO of 52 patients was （6.9±6.7）days. Causes of URO of 52 patients： 23 （2 deaths）, 7, 6 （2 deaths）, 5, 5 （1 death）, 3, 2 and 1 （death） patients were respectively due to intraperitoneal hemorrhage, anastomotic bleeding, anastomotic leakage, intraabdominal infection, small bowel obstruction, dehiscence of abdominal incisions, enteric perforation and pancreatic fistula, and time intervals between initial operation and URO of them were respectively （3.9±3.8）days, （0.9±0.5）days, （7.9±4.7）days, （14.9±4.6）days, （16.4±9.9）days, （10.0±6.0）days, （6.7±5.2）days and 12.0 days. （4） Univariate analysis affecting URO after radical resection of GC： results showed that age, body mass index （BMI） and volume of intraoperative blood loss were risk factors affecting URO after radical resection of GC （x2=5.468, 7.589, 5.041, P〈0.05）. （5） Multivariate analysis affecting URO after radical resection of GC： results showed that age 〉 70 years old, BMI 〉 25 kg/m2 and volume of intraoperative blood loss 〉 100 mL were independent risk factors affecting occurrence of URO after radical resection of GC （odds ratio=1.950, 2.288, 1.867; 95% confidence interval： 1.074-3.538, 1.230-4.257, 1.067-3.267, P〈0.05）. Conclusions：URO can increase postoperative complications and mortality, and extend duration of hospital stay after radical resection of GC. Intraabdominal bleeding, anastomotic bleeding and anastomotic leakage are the main causes affecting occurrence of URO after radical resection of GC, and age 〉 70 years old, BMI 〉 25 kg/m2 and volume of intraoperative blood loss 〉 100 mL are independent risk factors affecting occurrence of URO after radical resection of GC.