腹腔镜全胃切除术中食管-空肠手工吻合与Roux-en-Y吻合的倾向评分匹配近期疗效分析

Short-term efficacy of laparoscopic total gastrectomy with hand-sewn esophagojejunostomy versus Roux-en-Y anastomosis using propensity score matching

目的探讨腹腔镜全胃切除术中食管-空肠手工吻合与Roux-en-Y吻合的近期疗效。方法采用倾向评分匹配及回顾性队列研究方法。收集2014年10月至2021年7月上海交通大学医学院附属仁济医院收治的159例行腹腔镜全胃切除术胃癌患者的临床病理资料;男107例,女52例;年龄为63(28~79)岁。159例患者中,71例行完全腹腔镜全胃切除术中消化道重建采用食管-空肠手工吻合,设为完全腹腔镜组;88例行腹腔镜辅助全胃切除术中消化道重建采用传统Roux-en-Y吻合,设为腹腔镜辅助组。观察指标:(1)倾向评分匹配情况及匹配后两组患者一般资料比较。(2)术中和术后情况。(3)围手术期并发症情况。倾向评分匹配按1∶1最近邻匹配法匹配。正态分布的计量资料以x±s表示,组间比较采用t检验。偏态分布的计量资料以M(范围)表示,组间比较采用Mann-Whitney U检验。计数资料以绝对数或百分比表示,组间比较采用χ^(2)检验或Fisher确切概率法。等级资料比较采用秩和检验。结果(1)倾向评分匹配情况及匹配后两组患者一般资料比较:159例患者中,112例配对成功,完全腹腔镜组和腹腔镜辅助组各56例。倾向评分匹配前完全腹腔镜组患者年龄,肿瘤位置(贲门部、胃体部)分别为61(30~76)岁,26、45例;腹腔镜辅助组患者上述指标分别为65(28~79)岁,50、38例;两组患者上述指标比较,差异均有统计学意义(Z=-2.89,χ^(2)=6.43,P<0.05)。倾向评分匹配后完全腹腔镜组患者性别(男、女),年龄,体质量指数,美国麻醉医师协会分级(Ⅰ级、Ⅱ级、Ⅲ级、Ⅳ级),肿瘤长径,肿瘤位置(贲门部、胃体部),TNM分期(Ⅰ期、Ⅱ期、Ⅲ期)分别为40、16例,(62±9)岁,(22.7±2.8)kg/m^(2),22、26、7、1例,3.5(0.6~17.0)cm,24、32例,22、9、25例;腹腔镜辅助组患者上述指标分别为38、18例,(62±10)岁,(22.7±3.2)kg/m^(2),19、32、5、0例,4.0(0.6~15.0)cm,23、33例,21、7、28例;两组患者上述指标比较,差异均无统计学意义(χ^(2)=0.17,t=-0.09、-0.04,Z=-0.12、-0.82,χ^(2)=0.04,Z=-0.42,P>0.05)。消除年龄、肿瘤位置因素混杂偏倚,具有可比性。(2)术中和术后情况:倾向评分匹配后完全腹腔镜组患者总手术时间,食管-空肠吻合时间,术后24 h疼痛数字评分,术后首次下床活动时间分别为(310±49)min,(37±10)min,(2.3±0.8)分,(2.4±0.7)d;腹腔镜辅助组患者上述指标分别为(344±77)min,(44±12)min,(3.1±1.2)分,(2.9±1.0)d;两组患者上述指标比较,差异均有统计学意义(t=-2.85,-3.05,-4.20,-3.10,P<0.05)。(3)围手术期并发症情况:倾向评分匹配后完全腹腔镜组患者中,6例发生Clavien-Dindo分级2级以上并发症,其中吻合口漏2例,吻合口狭窄、胸腔积液、腹腔感染、肠梗阻各1例,Clavien-Dindo分级2级以上并发症发生率为10.7%(6/56)。腹腔镜辅助组患者中,5例发生Clavien-Dindo分级2级以上并发症,其中吻合口漏2例,腹腔感染、肠梗阻、胆管炎各1例,Clavien-Dindo分级2级以上并发症发生率为8.9%(5/56)。两组患者Clavien-Dindo分级2级以上并发症发生率比较,差异无统计学意义(χ^(2)=0.10,P>0.05)。两组吻合口漏患者经穿刺引流、二次手术、保守治疗后好转,其他并发症经对症治疗后好转。结论与腹腔镜全胃切除术中消化道重建采用传统Roux-en-Y吻合比较,采用食管-空肠手工吻合的总手术时间和食管-空肠吻合时间更短,患者术后疼痛减轻,术后恢复更快,两者围手术期安全性均较好。

Objective To investigate the short-term efficacy of laparoscopic total gastrec-tomy with hand-sewn esophagojejunostomy versus Roux-en-Y anastomosis.Methods The propen-sity score matching and retrospective cohort study was conducted.The clinicopathological data of 159 patients who underwent laparoscopic total gastrectomy in Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine from October 2014 to July 2021 were collected.There were 107 males and 52 females,aged 63(range,28‒79)years.Of 159 patients,71 cases undergoing totally laparoscopic total gastrectomy with hand-sewn esophagojejunostomy were allocated into totally laparoscopic group and 88 cases undergoing laparoscopic-assisted total gastrectomy with Roux-en-Y anastomosis were allocated into laparoscopic-assisted group,respectively.Observation indicators:(1)propensity score matching and comparison of general data of patients between the two groups after matching;(2)intraoperative and postoperative conditions;(3)perioperative complications.Propensity score matching was done by the 1:1 nearest neighbor matching method.Measurement data with normal distribution were expressed as Mean±SD,and t test was used for comparison between groups.Measurement data with skewed distribution were expressed as M(range),and the Mann-Whitney U test was used for comparison between groups.Count data were expressed as absolute numbers or percentages,and comparison between groups was analyzed using the chi-square test or Fisher's exact probability method.The rank sum test was used for comparison of ordinal data.Results(1)Propensity score matching and comparison of general data of patients between the two groups after matching.Of 159 patients,112 cases were successfully matched,including 56 cases in the totally laparoscopic group and 56 cases in the laparoscopic-assisted group.Before propensity score matching,age,cases with tumor located in cardia or gastric body in the totally laparoscopic group were 61(range,30‒76)years,26,45,respectively.The above indicators in the laparoscopic-assisted group were 65(range,28‒79)years,50,38,respectively.There were significant differences in the above indicators between the two groups(Z=‒2.89,χ^(2)=6.43,P<0.05).After propensity score matching,the males and females,age,body mass index,cases of American Society of Anesthesiologists classificationⅠ,Ⅱ,ⅢandⅣ,tumor diameter,cases with tumor located in cardia or gastric body,cases in TNM stageⅠ,ⅡandⅢof patients in the totally laparoscopic group were 40,16,(62±9)years,(22.7±2.8)kg/m^(2),22,26,7,1,3.5(range,0.6‒17.0)cm,24,32,22,9,25.The above indicators of patients in the laparoscopic-assisted group were 38,18,(62±10)years,(22.7±3.2)kg/m^(2),19,32,5,0,4.0(range,0.6‒15.0)cm,23,33,21,7,28,respectively.There was no significant difference in the above indicators between the two groups(χ^(2)=0.17,t=‒0.09,‒0.04,Z=‒0.12,‒0.82,χ^(2)=0.04,Z=‒0.42,P>0.05).The elimination of age and tumor location confounding bias ensured comparability between the two groups.(2)Intraoperative and postoperative conditions:after propensity score matching,the total operation time,time of esophagojejunostomy,postopera-tive 24-hour pain numerical score and time to first out-off bed activities were(310±49)minutes,(37±10)minutes,2.3±0.8 and(2.4±0.7)days for patients in the totally laparoscopic group,versus(344±77)minutes,(44±12)minutes,3.1±1.2 and(2.9±1.0)days in the laparoscopic-assisted group,showing significant differences between the two groups(t=‒2.85,‒3.05,‒4.20,‒3.10,P<0.05).(3)Perioperative complications:after propensity score matching,6 cases of the patients in the totally laparoscopic group had Clavien-Dindo grade 2 or higher complications,including 2 cases of anas-tomotic leak,1 case of anastomotic stenosis,1 case of pleural effusion,1 case of abdominal infection and 1 case of intestinal obstruction.The incidence of Clavien-Dindo grade 2 or higher complications was 10.7%(6/56).In the laparoscopic-assisted group,5 patients had Clavien-Dindo grade 2 or higher complications,including 2 cases of anastomotic leak,1 case of abdominal infection,1 case of intestinal obstruction and 1 case of cholangitis.The incidence of Clavien-Dindo grade 2 or higher complications was 8.9%(5/56).There was no significant difference in the incidence of Clavien-Dindo grade 2 or higher complications between the two groups(χ^(2)=0.10,P>0.05).Patients with anas-tomotic leak were improved after puncture and drainage,secondary surgery and conservative treat-ment,and other complications were improved after symptomatic treatment.Conclusions Com-pared with Roux-en-Y anastomosis in laparoscopic total gastrectomy,the time of hand-sewn esophagojejunostomy and esophago-jejunal anastomosis are shorter,patients have less postopera-tive pain and faster postoperative recovery.Both methods have good peri-operative safety.