海军军医大学学报  2022, Vol. 43 Issue (7): 778-783   PDF    
引用本文
刘智龙, 杜晓辉. 中性粒细胞/淋巴细胞比值联合C反应蛋白/白蛋白比值对直肠癌术后吻合口瘘的预测价值[J]. 海军军医大学学报, 2022, 43(7): 778-783  
LIU Zhi-long, DU Xiao-hui. Role of neutrophil to lymphocyte ratio combined with C reactive protein to albumin ratio in predicting postoperative anastomotic leakage in patients with rectal cancer[J]. Academic Journal of Naval Medical University, 2022, 43(7): 778-783 (in Chinese with English abstract)  
中性粒细胞/淋巴细胞比值联合C反应蛋白/白蛋白比值对直肠癌术后吻合口瘘的预测价值
刘智龙1,2, 杜晓辉3     
1. 中国人民解放军医学院研究生院，北京 100853;
2. 中国人民解放军总医院海南医院普通外科，三亚 572000;
3. 中国人民解放军总医院第一医学中心普通外科，北京 100853
收稿日期: 2021-11-08    接受日期: 2022-03-07
基金项目: 国家自然科学基金（81871317）.
作者简介: 刘智龙，硕士，住院医师.E-mail: liuzhilong301@sina.cn.
通信作者(Corresponding author): 杜晓辉, Tel: 010-66937598, E-mail: duxiaohui301@sina.com.
摘要: 目的 探讨中性粒细胞/淋巴细胞比值（NLR）联合C反应蛋白/白蛋白比值（CAR）对直肠癌术后吻合口瘘的预测价值。方法 回顾性分析2016年8月至2020年8月就诊于中国人民解放军总医院第一医学中心普通外科的187例直肠癌患者的临床资料，根据吻合口瘘发生与否分为吻合口瘘组（27例）和非吻合口瘘组（160例），比较两组在不同时间点的NLR、CAR差异。采用ROC曲线评价NLR、CAR及两者联合对直肠癌患者术后发生吻合口瘘的预测价值。采用多因素logistic回归分析直肠癌患者术后发生吻合口瘘的独立危险因素。结果 吻合口瘘组直肠癌患者的NLR在术后第3、5天均高于非吻合口瘘组（P＜0.05、P＜0.001），CAR在术后第5天高于非吻合口瘘组（P＜0.001）。ROC曲线分析显示，术后第5天的NLR联合CAR预测直肠癌患者术后发生吻合口瘘的AUC值为0.814，术后第3、5天NLR和术后第5天CAR单独预测吻合口瘘的AUC值分别为0.652、0.746和0.789，其中术后第5天NLR联合CAR的AUC值与术后第3天NLR的AUC值差异有统计学意义（P＝0.03）。多因素logistic回归分析显示，饮酒史、术后第3天NLR≥5.610、术后第5天NLR≥11.259和术后第5天CAR≥2.447是直肠癌患者术后发生吻合口瘘的独立危险因素（P均＜0.05）。结论 术后第5天NLR联合CAR、术后第3天NLR、术后第5天NLR和术后第5天CAR均对直肠癌术后吻合口瘘有预测价值，其中术后第5天NLR联合CAR预测效果最佳。
关键词: 中性粒细胞/淋巴细胞比值    C反应蛋白/白蛋白比值    直肠肿瘤    直肠切除术    吻合口瘘    
Role of neutrophil to lymphocyte ratio combined with C reactive protein to albumin ratio in predicting postoperative anastomotic leakage in patients with rectal cancer
LIU Zhi-long1,2, DU Xiao-hui3     
1. Graduate School, Chinese PLA Medical School, Beijing 100853, China;
2. Department of General Surgery, Hainan Hospital, Chinese PLA General Hospital, Sanya 572000, Hainan, China;
3. Department of General Surgery, the First Medical Center, Chinese PLA General Hospital, Beijing 100853, China
Supported by National Natural Science Foundation of China (81871317).
Abstract: Objective To investigate the role of neutrophil to lymphocyte ratio (NLR) combined with C reactive protein to albumin ratio (CAR) in predicting postoperative anastomotic leakage (AL) in patients with rectal cancer. Methods The clinical data of 187 rectal cancer patients who underwent surgery in Department of General Surgery, the First Medical Center, Chinese PLA General Hospital from Aug. 2016 to Aug. 2020 were retrospectively analyzed. Patients were divided into AL group (27 cases) and non-AL group (160 cases), and NLR and CAR were compared between the 2 groups at different time points. Receiver operating characteristic (ROC) curve was used to evaluate the prediction value of NLR, CAR, and their combination for AL. Multivariate logistic regression analysis was used to explore the independent risk factors of postoperative AL in patients with rectal cancer. Results The NLR of rectal cancer patients in the AL group was significantly higher than that in the non-AL group on the 3rd day and 5th day after surgery (P < 0.05 and P < 0.001), and the CAR was significantly higher than that in the non-AL group on the 5th day after surgery (P < 0.001). ROC curve analysis showed that the area under curve (AUC) value of NLR combined with CAR on the 5th day after surgery for predicting AL in rectal cancer patients was 0.814, and the AUC values of NLR on the 3rd day and 5th day after surgery and CAR on the 5th day after surgery were 0.652, 0.746 and 0.789, respectively. The AUC value of NLR combined with CAR on the 5th day after surgery was significantly higher than that of NLR on the 3rd day after surgery (P＝0.03). Multivariate logistic regression analysis showed that drinking history, NLR≥5.610 on the 3rd day after surgery, NLR≥11.259 on the 5th day after surgery, and CAR≥2.447 on the 5th day after surgery were independent risk factors for AL in rectal cancer patients (all P < 0.05). Conclusion NLR combined with CAR on the 5th day after surgery, NLR on the 3rd day after surgery, NLR on the 5th day after surgery and CAR on the 5th day after surgery have predictive value for postoperative AL in rectal cancer patients, and NLR combined with CAR on the 5th day after surgery has the best predictive effect.
Key words: neutrophil to lymphocyte ratio    C reactive protein to albumin ratio    rectal neoplasms    proctectomy    anastomotic leakage    

直肠癌是全世界最常见的消化道恶性肿瘤之一，其发病率和死亡率均居恶性肿瘤的第5位[1]。目前，对直肠癌的治疗仍以手术为主。尽管手术技术和医疗器械不断发展，临床上仍无法避免术后吻合口瘘（anastomotic leakage）的发生。在国外吻合口瘘的发生率和死亡率分别为3%~23%、6%~30%[2-3]；在国内其发生率为2.4%~15.9%，而死亡率最高可达16%[4]；吻合口瘘不仅增加了患者的经济负担，而且降低了患者的生活质量，甚至威胁患者生命。因此，探索吻合口瘘的有效预测指标有重要的临床意义。吻合口瘘的发生可能是炎症、免疫系统和营养状态等因素相互作用的结果[5-6]。而中性粒细胞/淋巴细胞比值（neutrophil to lymphocyte ratio，NLR）和CRP/白蛋白比值（C reactive protein to albumin ratio，CAR）是反映全身炎症、免疫水平和营养状态的重要指标[7-8]。本研究通过回顾性分析直肠癌患者的临床资料，探讨NLR、CAR及两者联合对直肠癌患者术后发生吻合口瘘的预测价值。

1 资料和方法 1.1 病例资料

回顾性分析2016年8月至2020年8月就诊于中国人民解放军总医院第一医学中心普通外科的187例直肠癌患者的临床资料。纳入标准：（1）年龄≥18周岁；（2）病理证实为直肠癌，并行直肠癌根治术；（3）术前外周血中性粒细胞和淋巴细胞计数及术后第1、3、5天的外周血CRP水平、白蛋白水平、中性粒细胞计数和淋巴细胞计数等实验室检查资料结果记录完整。排除标准：（1）患者围手术期输注人血白蛋白；（2）患有肝肾疾病、感染性疾病和血液系统疾病等影响外周血细胞和血清白蛋白的疾病；（3）术后7 d内因非吻合口瘘原因进行二次手术。本研究通过中国人民解放军总医院医学伦理委员会审批。

1.2 吻合口瘘的诊断标准

（1）消化道造影可见造影剂通过吻合口流出；（2）引流管引流出粪渣样引流液；（3）盆腔CT检查提示吻合口周围存在脓肿[4]。观察期为术后1周，期间出现上述任何1项即诊断为吻合口瘘。

1.3 研究方法

根据是否发生吻合口瘘将患者分为吻合口瘘组（27例）和非吻合口瘘组（160例）。收集患者的性别、年龄、糖尿病史、高血压病史、吸烟史、饮酒史、BMI、TNM分期，术前外周血中性粒细胞计数和淋巴细胞计数，术后第1、3、5天外周血CRP水平、白蛋白水平、中性粒细胞计数和淋巴细胞计数等资料，并计算术前和术后第1、3、5天的NLR[中性粒细胞计数（×109/L）/淋巴细胞计数（×109/L）]及术后第1、3、5天的CAR[CRP（mg/L）/白蛋白（g/L）]。其中，肿瘤TNM分期参照美国癌症联合委员会制定的癌症分期手册（第8版）。比较两组上述指标的差异，评价NLR、CAR及两者联合对直肠癌患者术后发生吻合口瘘的预测价值，并分析吻合口瘘的独立危险因素。

1.4 统计学处理

应用SPSS 26.0软件对数据进行统计学分析。计量资料均呈偏态分布，以中位数（下四分位数，上四分位数）表示，两组间比较采用Mann-Whitney U检验。计数资料以例数和百分数表示，两组间比较使用χ2检验。采用ROC曲线评价NLR、CAR及两者联合对直肠癌患者术后发生吻合口瘘的预测价值，并使用DeLong检验对ROC曲线的AUC值进行差异性比较。采用多因素logistic回归分析直肠癌患者术后发生吻合口瘘的独立危险因素。检验水准（α）为0.05。

2 结果 2.1 两组患者NLR和CAR的比较

吻合口瘘组和非吻合口瘘组直肠癌患者术前NLR及术后第1天CAR的差异均无统计学意义（P均＞0.05）。吻合口瘘组直肠癌患者NLR在术后第3、5天均高于非吻合口瘘组（P＜0.05、P＜0.001），CAR在术后第5天高于非吻合口瘘组（P＜0.001）。见表 1。

表 1 两组直肠癌患者术前及术后不同时间点NLR和CAR的比较 Tab 1 Comparison of NLR and CAR of rectal cancer patients before and at different time points after surgery between 2 groups  

2.2 ROC曲线分析

对术后第3、5天的NLR和术后第5天的CAR预测吻合口瘘的效能进行ROC曲线分析，结果如图 1所示。直肠癌患者术后第5天的NLR联合CAR预测吻合口瘘的AUC值为0.814（95% CI 0.722~0.906，P＜0.001），约登指数为0.560，灵敏度为74.1%，特异度为81.9%。术后第3、5天的NLR和术后第5天的CAR预测吻合口瘘的AUC值分别为0.652（95% CI 0.542~0.761，P＝0.012）、0.746（95% CI 0.646~0.846，P＜0.001）和0.789（95% CI 0.692~0.886，P＜0.001），最佳临界值分别为5.610、11.259和2.447，约登指数分别为0.320、0.388和0.517，灵敏度分别为88.9%、44.1%和66.7%，特异度分别为43.1%、98.1%和85%。DeLong检验结果显示，术后第5天NLR联合CAR的AUC值与术后第3天NLR的AUC值差异有统计学意义（Z＝2.18，P＝0.03），其余AUC值比较差异均无统计学意义（P均＞0.05）。

图 1 NLR和CAR预测直肠癌患者术后发生吻合口瘘的ROC曲线 Fig 1 ROC curves of NLR and CAR in predicting anastomotic leakage after surgery in rectal cancer patients NLR: Neutrophil to lymphocyte ratio; CAR: C reactive protein to albumin ratio; ROC: Receiver operating characteristic.

2.3 吻合口瘘影响因素的单因素分析

在男性、有饮酒史、术后第3天NLR≥5.610、术后第5天NLR≥11.259和术后第5天CAR≥2.447的患者中吻合口瘘的发生率较高，差异均有统计学意义（P均＜0.05），表明性别、饮酒史、术后第3天和5天NLR、术后第5天CAR可能与吻合口瘘的发生有关。见表 2。

表 2 直肠癌患者术后发生AL影响因素的单因素分析 Tab 2 Univariate analysis of influencing factors of AL after surgery in rectal cancer patients  

2.4 吻合口瘘影响因素的多因素logistic回归分析

以发生吻合口瘘与否为因变量（否＝0，是＝1），单因素分析中P＜0.05的变量（性别、饮酒史、术后第3天和第5天的NLR、术后第5天的CAR）为自变量，采用全部进入法，进行logistic回归分析。结果显示，饮酒史（OR＝13.511，95% CI 2.772~65.861，P＝0.001）、术后第3天NLR≥5.610（OR＝4.924，95% CI 1.219~19.888，P＝0.025）、术后第5天NLR≥11.259（OR＝36.092，95% CI 4.639~280.797，P＝0.001）和术后第5天CAR≥2.447（OR＝5.219，95% CI 1.683~16.184，P＝0.004）是直肠癌患者术后发生吻合口瘘的独立危险因素。

3 讨论

在目前术后促进康复（enhanced recovery after surgery）的治疗策略中，着重强调患者早期恢复肠道功能、快速出院的同时，存在着因早期出院而未能发现吻合口瘘的风险。吻合口瘘一旦发生，若不能及时识别并采取相应的治疗措施，患者可能发生发热、腹腔脓肿、腹膜炎，严重者可出现代谢紊乱、感染性休克及多器官功能衰竭甚至死亡[9]。目前对吻合口瘘的诊断主要依据临床症状及影像学检查[10]，而确诊的吻合口瘘已经对患者造成了无法挽回的后果。尽管当前建立了对吻合口瘘的预测诊断模型[11-13]，临床外科医师可以通过术中、术后相关指标预测吻合口瘘发生的可能性，如结直肠瘘评分（colon leakage score，CLS）和DULK（Dutch leakage）评分等评价系统，但这些评价系统的预测效能在各学者间观点不一，预测价值未被广泛认可，临床应用较少[14-17]。因此，寻找一种新型、有效的吻合口瘘预测指标意义重大。

目前，吻合口瘘的发生机制尚未完全明了，可能与遗传、肠道微生物、全身炎症、免疫系统、营养状态及吻合口局部血供等因素有关[5-6, 18]。在吻合口愈合的炎症期，大量淋巴细胞、中性粒细胞和巨噬细胞在吻合口处聚集，通过吞噬外来微粒和细菌、分解吻合口处的坏死组织促使吻合口愈合，而持续存在的中性粒细胞可通过释放蛋白酶和具有毒性的氧自由基导致细胞外基质损伤，不利于吻合口愈合[19-21]。另外，淋巴细胞减少会导致细胞外基质中胶原蛋白合成障碍和组织愈合能力下降[22-23]。吻合口愈合离不开患者对营养和能量的摄入，当患者营养不良时吻合口的愈合能力也会下降[24]。而NLR和CAR作为综合反映患者全身炎症、免疫和营养状况的指标，可能与吻合口的愈合和吻合口瘘的发生有关[7-8]。

本研究通过回顾性分析187例直肠癌术后患者的临床资料，探讨围手术期不同时间点NLR、CAR和两者联合对吻合口瘘的预测价值。结果表明，术后第3天和第5天的NLR、术后第5天的CAR及术后第5天的NLR联合CAR对直肠癌患者术后是否发生吻合口瘘均有一定的预测价值。对上述各指标AUC值的差异性进行分析显示，术后第5天NLR联合CAR的AUC值与术后第3天NLR的AUC值差异有统计学意义，说明术后第5天NLR联合CAR对吻合口瘘的预测效果优于术后第3天NLR单一指标的预测效果。但临床实践中仍需重视术后第3天NLR预测吻合口瘘的价值，其时间窗较早，能更早地识别吻合口瘘。临床上，NLR与CAR的监测和计算较为方便、快捷，若发现术后第3天NLR≥5.610、术后第5天NLR≥11.259、术后第5天CAR≥2.447，应及时推迟进食时间、加强抗炎和营养支持，并尽早对患者进行影像学检查以明确诊断，一旦确诊则应及时采取腹腔冲洗引流、内镜下覆膜支架置入等治疗措施，避免二次手术，从而最大程度地减轻吻合口瘘给患者带来的不良影响。

本研究中多因素分析结果还表明，饮酒史是直肠癌患者术后发生吻合口瘘的独立危险因素。乙醇的摄入可以导致骨髓损伤，使纤维蛋白原、因子Ⅶ和血管因子降低，进一步影响机体的止血功能并对伤口的愈合造成不良影响[25-26]。既往也有研究证实饮酒可增加直肠癌术后吻合口瘘及其他术后并发症的发生风险[27-29]。因此，对于有饮酒史的直肠癌术后患者，在围手术期也要警惕发生吻合口瘘的可能性。本研究因部分患者的资料记录不全，故入组病例有限，样本量不大，样本的代表性不强，导致目前已被大多学者证实的吻合口瘘影响因素如性别、糖尿病和BMI等，在本研究中未能发现其与吻合口瘘相关。本研究结果仍需前瞻性、大样本试验进一步证实。

参考文献
[1]
MANCEAU G, BROUQUET A, CHAIBI P, PASSOT G, BOUCHÉ O, MATHONNET M, et al. Multicenter phase Ⅲ randomized trial comparing laparoscopy and laparotomy for colon cancer surgery in patients older than 75 years: the CELL study, a Fédération de Recherche en Chirurgie (FRENCH) trial[J/OL]. BMC Cancer, 2019, 19: 1185. DOI: 10.1186/s12885-019-6376-8.
[2]
PARK J S, CHOI G S, KIM S H, KIM H R, KIM N K, LEE K Y, et al. Multicenter analysis of risk factors for anastomotic leakage after laparoscopic rectal cancer excision: the Korean Laparoscopic Colorectal Surgery Study Group[J]. Ann Surg, 2013, 257: 665-671. DOI:10.1097/SLA.0b013e31827b8ed9
[3]
RAHBARI N N, WEITZ J, HOHENBERGER W, HEALD R J, MORAN B, ULRICH A, et al. Definition and grading of anastomotic leakage following anterior resection of the rectum: a proposal by the International Study Group of Rectal Cancer[J]. Surgery, 2010, 147: 339-351. DOI:10.1016/j.surg.2009.10.012
[4]
中华医学会外科学分会结直肠外科学组. 中国直肠癌手术吻合口漏诊断、预防及处理专家共识(2019版)[J]. 中华胃肠外科杂志, 2019, 22: 201-206.
[5]
FOPPA C, NG S C, MONTORSI M, SPINELLI A. Anastomotic leak in colorectal cancer patients: new insights and perspectives[J]. Eur J Surg Oncol, 2020, 46: 943-954. DOI:10.1016/j.ejso.2020.02.027
[6]
KANG C Y, HALABI W J, CHAUDHRY O O, NGUYEN V, PIGAZZI A, CARMICHAEL J C, et al. Risk factors for anastomotic leakage after anterior resection for rectal cancer[J]. JAMA Surg, 2013, 148: 65-71. DOI:10.1001/2013.jamasurg.2
[7]
ZAHOREC R. Neutrophil-to-lymphocyte ratio, past, present and future perspectives[J]. Bratis1 Lek Listy, 2021, 122: 474-488.
[8]
CAO X, CUI J, LI Z, ZHAO G. Preoperative C-reactive protein/albumin ratio is a prognostic indicator for survival in surgically treated gastrointestinal stromal tumors: a retrospective cohort study[J]. Cancer Manag Res, 2021, 13: 4155-4167. DOI:10.2147/CMAR.S307873
[9]
PEEL A L, TAYLOR E W. Proposed definitions for the audit of postoperative infection: a discussion paper. Surgical Infection Study Group[J]. Ann Royal Coll Surg Engl, 1991, 73: 385-388.
[10]
李子禹, 吴舟桥, 季加孚. 中国胃肠肿瘤外科术后并发症诊断登记规范专家共识(2018版)[J]. 中国实用外科杂志, 2018, 38: 589-595.
[11]
KARLICZEK A, HARLAAR N J, ZEEBREGTS C J, WIGGERS T, BAAS P C, DAM G M V. Surgeons lack predictive accuracy for anastomotic leakage in gastrointestinal surgery[J]. Int J Colorectal Dis, 2009, 24: 569-576. DOI:10.1007/s00384-009-0658-6
[12]
MARKUS P M, MARTELL J, LEISTER I, HORSTMANN O, BRINKER J, BECKER H. Predicting postoperative morbidity by clinical assessment[J]. Br J Surg, 2005, 92: 101-106. DOI:10.1002/bjs.4608
[13]
DEN DULK M, NOTER S L, HENDRIKS E R, BROUWERS M A M, VAN DER VLIES C H, OOSTENBROEK R J, et al. Improved diagnosis and treatment of anastomotic leakage after colorectal surgery[J]. Eur J Surg Oncol, 2009, 35: 420-426. DOI:10.1016/j.ejso.2008.04.009
[14]
YANG S U, PARK E J, BAIK S H, LEE K Y, KANG J. Modified colon leakage score to predict anastomotic leakage in patients who underwent left-sided colorectal surgery[J/OL]. J Clin Med, 2019, 8: 1450. DOI: 10.3390/jcm8091450.
[15]
DEN DULK M, WITVLIET M J, KORTRAM K, NEIJENHUIS P A, DE HINGH I H, ENGEL A F, et al. The DULK (Dutch leakage) and modified DULK score compared: actively seek the leak[J/OL]. Colorectal Dis, 2013, 15: e528-e533. DOI: 10.1111/codi.12379.
[16]
YU X Q, ZHAO B, ZHOU W P, HAN L Z, CAI G H, FANG Z W, et al. Utility of colon leakage score in left-sided colorectal surgery[J]. J Surg Res, 2016, 202: 398-402. DOI:10.1016/j.jss.2015.12.046
[17]
郑鹏, 许剑民. 结直肠吻合术后吻合口瘘的防治[J]. 中华胃肠外科杂志, 2016, 19: 379-382. DOI:10.3760/cma.j.issn.1671-0274.2016.04.005
[18]
ZARNESCU E C, ZARNESCU N O, COSTEA R. Updates of risk factors for anastomotic leakage after colorectal surgery[J/OL]. Diagnostics (Basel), 2021, 11: 2382. DOI: 10.3390/diagnostics11122382.
[19]
LUNDY J B. A primer on wound healing in colorectal surgery in the age of bioprosthetic materials[J]. Clin Colon Rectal Surg, 2014, 27: 125-133. DOI:10.1055/s-0034-1394086
[20]
BAXTER C R. Immunologic reactions in chronic wounds[J]. Am J Surg, 1994, 167(1A): 12S-14S.
[21]
GRAY M, MARLAND J R K, MURRAY A F, ARGYLE D J, POTTER M A. Predictive and diagnostic biomarkers of anastomotic leakage: a precision medicine approach for colorectal cancer patients[J/OL]. J Pers Med, 2021, 11: 471. DOI: 10.3390/jpm11060471.
[22]
RADULESCU D, BALEANU V D, PADUREANU V, RADULESCU P M, BORDU S, PATRASCU S, et al. Neutrophil/lymphocyte ratio as predictor of anastomotic leak after gastric cancer surgery[J/OL]. Diagnostics (Basel), 2020, 10: 799. DOI: 10.3390/diagnostics10100799.
[23]
WANG W T, LEE S S, WANG Y C, LAI Y W, KUO Y R, TANG CHEN Y B, et al. Impaired cutaneous T-cell attracting chemokine elevation and adipose-derived stromal cell migration in a high-glucose environment cause poor diabetic wound healing[J]. Kaohsiung J Med Sci, 2018, 34: 539-546. DOI:10.1016/j.kjms.2018.05.002
[24]
THOMPSON S K, CHANG E Y, JOBE B A. Clinical review: healing in gastrointestinal anastomoses, part Ⅰ[J]. Microsurgery, 2006, 26: 131-136. DOI:10.1002/micr.20197
[25]
KENNEDY N D, WINTER D C. Impact of alcohol & smoking on the surgical management of gastrointestinal patients[J]. Best Pract Res Clin Gastroenterol, 2017, 31: 589-595. DOI:10.1016/j.bpg.2017.10.005
[26]
GLÓRIA L, CRAVO M, CAMILO M E, RESENDE M, CARDOSO J N, OLIVEIRA A G, et al. Nutritional deficiencies in chronic alcoholics: relation to dietary intake and alcohol consumption[J]. Am J Gastroenterol, 1997, 92: 485-489.
[27]
VAN ROOIJEN S, CARLI F, DALTON S O, JOHANSEN C, DIELEMAN J, ROUMEN R D, et al. Preoperative modifiable risk factors in colorectal surgery: an observational cohort study identifying the possible value of prehabilitation[J]. Acta Oncol, 2017, 56: 329-334. DOI:10.1080/0284186X.2016.1267872
[28]
RULLIER E, LAURENT C, GARRELON J L, MICHEL P, SARIC J, PARNEIX M. Risk factors for anastomotic leakage after resection of rectal cancer[J]. Br J Surg, 2003, 85: 355-358. DOI:10.1046/j.1365-2168.1998.00615.x
[29]
TURRENTINE F E, DENLINGER C E, SIMPSON V B, GARWOOD R A, GUERLAIN S, AGRAWAL A, et al. Morbidity, mortality, cost, and survival estimates of gastrointestinal anastomotic leaks[J]. J Am Coll Surg, 2015, 220: 195-206. DOI:10.1016/j.jamcollsurg.2014.11.002