摘要目的 探讨动脉先行路径联合预先门静脉-肠系膜上静脉分流重建胰十二指肠切除术治疗复杂胰头部肿瘤的临床疗效.方法 采用回顾性队列研究方法.收集2011年2月至2016年12月海军军医大学附属长海医院收治的91例胰头部肿瘤患者的临床病理资料,患者均行联合血管切除重建胰十二指肠切除术,其中27例复杂胰头部肿瘤行动脉先行路径联合预先门静脉-肠系膜上静脉分流重建胰十二指肠切除术,设为PD-PVR组;23例复杂胰头部肿瘤行常规路径胰十二指肠切除术,设为PD-CVR组;41例胰头部肿瘤行胰十二指肠联合血管切除术,设为PD-SVR组.观察指标:(1)3组患者术中及术后情况比较.(2)3组患者术后病理学检查情况比较.(3)3组患者术后生存情况比较.采用门诊和电话方式定期进行随访,随访患者术后生存情况.随访时间截至2017年12月.计数资料组间比较用x2检验或Fisher确切概率法.正态分布的计量资料以x±s表示,多组间比较采用方差分析,两两比较采用LSD法;偏态分布的计量资料以M(Q)表示,组间比较采用Kurskal-Wallis秩和检验,两两比较采用校正p值的成对比较.等级资料比较采用非参数检验.采用Kaplan-Meier法绘制生存曲线和计算生存率,Log-rank检验进行生存分析.结果 (1)3组患者术中及术后情况比较:PD-PVR组患者手术时间,肝门阻断时间,术中出血量,术中输血例数,门静脉或肠系膜上静脉切除长度,使用人工血管,术后并发症Clavien-Dindo分级(1、2、3、4、5级),术后住院时间分别为(274±36) min,(22±7)min,1 661 mL(110 mL,3 800 mL),20例,(5.6±1.4) cm,6例,11、1、1、1、1例,(20±7)d;PD-CVR组分别为(281±41)min,(27±5) min,1 888 mL(176 mL,4 162 mL),18例,(5.4±1.5)cm,3例,1、8、2、0、0例,(21±7) d;PD-SVR组分别为(201±36) min,(16±6) min,1 052 mL(74 mL,3 926 mL),17例,(3.2±2.0)cm,5例,15、2、3、1、1例,(13±6)d;3组患者上述指标比较,差异均有统计学意义(F=37.060,34.530,x2=13.771,14.015,F=32.260,x2=39.309,F=19.880,P＜0.05).术后并发症均经对症支持治疗后有所好转.(2)3组患者术后病理学检查情况比较:PD-PVR组、PD-CVR组、PD-SVR组患者肿瘤最大径分别为(3.6±1.3)cm、(4.0±1.3)cm、(2.6± 1.3) cm,3组比较,差异有统计学意义(F=7.845,P＜0.05).PD-PVR组患者肿瘤分化程度(高分化、中分化、低分化),肿瘤分期(ⅡA期、ⅡB期),神经侵犯,阳性淋巴结,胰颈部切缘阳性,胆管切缘阳性,肠系膜上静脉切缘阳性例数分别为0、13、14例,17、10例,17例,21例,0,0,0;PD-CVR组分别为1、12、10例,10、13例,15例,19例,1例,0,0;PD-SVR组分别为1、29、11例,17、24例,30例,29例,2例,1例,1例;3组患者上述指标比较,差异均无统计学意义(x2=4.122,3.306,0.902,1.214,P＞0.05).(3)3组患者术后生存情况比较:91例患者中,52例获得术后随访(PD-PVR组16例、PD-CVR组14例、PD-SVR组22例),随访率为57.1％(52/91),随访时间为3.0～69.3个月,中位随访时间为18.0个月.52例患者中位生存时间为16.6个月,1、2、3年总体生存率分别为63.5％、35.7％、26.8％.PD-PVR组、PD-CVR组、PD-SVR组患者生存时间分别为12.3个月(3.9～69.3个月)、15.0个月(3.0～63.3个月)、20.0个月(6.0～65.2个月),3组生存情况比较,差异有统计学意义(x2 =6.201,P＜0.05);PD-PVR组与PD-CVR组比较,差异无统计学意义(x2=0.001,P＞0.05),PD-PVR组与PD-SVR组比较,差异有统计学意义(x2=4.412,P＜0.05).结论 动脉先行路径联合预先门静脉-肠系膜上静脉分流重建的胰十二指肠切除术缩短了肝门血流阻断时间,有助于复杂胰腺肿瘤术中的风险控制,提高手术安全性,改善患者预后.

abstractsObjective To investigate the clinical efficacy of pancreaticoduodenectomy using artery first approach and prior portal vein (PV)-superior mesenteric vein (SMV) shunting for complicated pancreatic head tumors.Methods The retrospective cohort study was conducted.The clinicopathological data of 91 patients with pancreatic head cancer who were admitted to the Changhai Hospital of Navy Medical University from February 2011 to December 2016 were collected.Among 91 patients undergoing pancreaticoduodenectomy combined with vascular resection and reconstruction,27 using artery first approach and PV-SMV shunting and reconstruction were allocated into the PD-PVR group,23 using conventional approach were allocated into the PD-CVR group,and 41 receiving pancreaticoduodenectomy with vascular resection were allocated into the PD-SVR group.Observation indicators:(1) comparison of intra-and post-operative situations among groups;(2) comparison of postoperative pathological results among groups;(3) comparison of postoperative survival among groups.Follow-up using outpatient examination and telephone interview was performed to detect postoperative survival up to December 2017.Comparisons among groups of count data were done by the chi-square test or Fisher exact probability.Measurement data with normal distribution were represented as x±s,comparisons among groups were analyzed using the ANOVA,and pairwise comparison was done using the LSD method.Measurement data with skewed distribution were described as M(Q),comparison among groups was analyzed using the Kurskal-Wallis rank sum test,and pairwise comparison was done using the paired comparison with adjusted P value.Ordinal data were compared using the nonparametric test.The survival rate and curve were respectively calculated and drawn by the Kaplan-Meier method,and Log-rank test was used for survival analysis.Results (1) Comparison of intra-and post-operative situations among groups:operation time,time of hepatic inflow occlusion,volume of intraoperative blood loss,cases with intraoperative blood transfusion,length of PV or SMV removal,cases with vascular grafts,grading 1,2,3,4 and 5 of Clavien-Dindo classification of postoperative complications and duration of hospital stay were respectively (274±36)minutes,(22±7)minutes,1 661 mL (110 mL,3 800 mL),20,(5.6±1.4)cm,6,11,1,1,1,1,(20±7)days in the PD-PVR group and (281±41)minutes,(27±5)minutes,1 888 mL (176 mL,4 162 mL),18,(5.4±1.5) cm,3,1,8,2,0,0,(21±7)days in the PD-CVR group and (201± 36)minutes,(16±6)minutes,1052 mL (74 mL,3 926 mL),17,(3.2±2.0) cm,5,15,2,3,1,1,(13± 6)days in the PD-SVR group,with statistically significant differences among groups (F=37.060,34.530,x2=13.771,14.015,F=32.260,x2 =39.309,F =19.880,P＜0.05).Patients with postoperative complications were improved by symptomatic and supporting treatment.(2) Comparison of postoperative pathological results among groups:mnaximum tumour dimension was respectively (3.6± 1.3) cm,(4.0± 1.3) cm and (2.6± 1.3) cm in the PD-PVR,PD-CVR and PD-SVR groups,with a statistically significant difference among groups (F =7.845,P＜0.05).Cases with high-differentiated,moderate-differentiated and low-differentiated tumors of tumor differentiation,staging Ⅱ A and Ⅱ B of tumor staging,nerve invasion,positive lymph node,positive resection margins in neck of pancreas,bile duct and SMV were respectively 0,13,14,17,10,17,21,0,0,0 in the PD-PVR group and 1,12,10,10,13,15,19,1,0,0 in the PD-CVR group and 1,29,11,17,24,30,29,2,1,1 in the PD-SVR group,with no statistically significant difference among groups (x2 =4.122,3.306,0.902,1.214,P＞0.05).(3) Comparisons of postoperative survival among groups:of 91 patients,52 were followed up for 3.0-69.3 months,with a median time of 18.0 months and follow-up rate of 57.1％ (52/91),including 16 in the PD-PVR group and 14 in the PD-CVR group and 22 in the PD-SVR group.The median survival time,1-,2-and 3-year overall survival rates in 52 patients were respectively 16.6 months,63.5％,35.7％ and 26.8％.The survival time in the PD-PVR,PD-CVR and PD-SVR groups were respectively 12.3 months (3.9-69.3 months),15.0 months (3.0-63.3 months) and 20.0 months (6.0-65.2 months),with a statistically significant difference in survival among groups (x2=6.201,P＜0.05),and between PD-PVR and PD-SVR groups (x2 =4.412,P＜0.05).There was no statistically significant difference in survival between PD-PVR and PD-CVR groups (x2 =0.001,P＞0.05).Conclusion Pancreaticoduodenectomy using artery first approach and PV-SMV shunting and reconstruction for complicated pancreatic head tumors can reduce the time of hepatic inflow occlusion,it also contributes to the risk control of surgery for complicated pancreatic tumors,increases surgical safety and improves patients' prognosis.