结肠镜检查治疗发生结直肠穿孔的影响因素分析及处理策略

Influencing factors analysis and management strategies of colonoscopy-associated colorectal perforation

目的探讨结肠镜检查治疗发生结直肠穿孔的影响因素及处理策略。方法采用回顾性病例对照研究方法。收集2011年1月至2021年3月遵义医科大学附属医院收治的358例行结肠镜检查治疗患者的临床资料;男216例,女142例;年龄为(59±14)岁。患者均行结肠镜检查或治疗。观察指标:(1)结肠镜检查治疗发生结直肠穿孔情况。(2)结肠镜检查治疗发生结直肠穿孔的影响因素分析。(3)结肠镜检查治疗发生结直肠穿孔预测模型的构建。(4)结肠镜检查治疗发生结直肠穿孔的处理。正态分布的计量资料以x±s表示。偏态分布的计量资料以M(Q1,Q3)表示,组间比较采用Wilcoxon秩和检验。计数资料以绝对数表示,组间比较采用χ^(2)检验。多因素分析采用二元Logistic回归模型。采用受试者工作特征(ROC)曲线评价预测模型的灵敏度和特异度。结果(1)结肠镜检查治疗发生结直肠穿孔情况。358例患者中,18例发生结直肠穿孔,其中男6例,女12例;年龄为61(49,69)岁。18例结直肠穿孔患者中,结肠穿孔12例,其中乙状结肠及直肠乙状结肠交界处穿孔10例、横结肠穿孔1例、降结肠穿孔1例;直肠穿孔6例。诊断性穿孔11例,治疗性穿孔7例。(2)结肠镜检查治疗发生结直肠穿孔的影响因素分析。单因素分析结果显示:性别、年龄、结直肠溃疡、结直肠憩室、结直肠肿瘤、腹腔手术史、结肠镜类别、操作医师职称为影响结肠镜检查治疗发生结直肠穿孔的相关因素(χ^(2)=5.77,Z=-3.24,χ^(2)=37.99、97.34、37.99、10.31、8.07、6.73,P<0.05)。多因素分析结果显示:结直肠憩室、腹腔手术史是影响结肠镜检查治疗发生结直肠穿孔的独立危险因素(优势比=287.79,6.74,95%可信区间为23.14~3579.11,1.19~38.27,P<0.05)。治疗性结肠镜是影响结肠镜检查治疗发生结直肠穿孔的独立保护因素(优势比=0.11,95%可信区间为0.23~0.52,P<0.05)。(3)结肠镜检查治疗发生结直肠穿孔预测模型的构建。以结肠镜检查治疗发生结直肠穿孔为因变量,结直肠憩室、腹腔手术史、治疗性结肠镜为自变量构建结肠镜检查治疗发生结直肠穿孔的Logistic回归预测模型。ROC结果显示:该预测模型的灵敏度为0.56,特异度为1.00,曲线下面积为0.78(95%可信区间为0.63~0.92,P<0.05)。(4)结肠镜检查治疗发生结直肠穿孔的处理。18例结肠镜检查治疗发生结直肠穿孔患者中,15例立即行腹腔镜穿孔修补术,2例行内镜下缝合术,1例行保守治疗;均治愈出院,无因结直肠穿孔死亡患者。结论结肠镜检查治疗发生结直肠穿孔部位易发于乙状结肠及直肠乙状结肠交界处。结直肠憩室、腹腔手术史是影响结肠镜检查治疗发生结直肠穿孔的独立危险因素。治疗性结肠镜是影响结肠镜检查治疗发生结直肠穿孔的独立保护因素。结直肠穿孔患者行腹腔镜结直肠穿孔修补术疗效较好。

Objective To investigate the influencing factors and management strategies of colonoscopy-associated colorectal perforation.Methods The retrospective case-control study was conducted.The clinical data of 358 patients who underwent colonoscopy in the Affiliated Hospital of Zunyi Medical University from January 2011 to March 2021 were collected.There were 216 males and 142 females,aged(59±14)years.Patients underwent colonoscopy for diagnosis or treatment.Observation indicators:(1)situations of colonoscopy-associated colorectal perforation;(2)analysis of influencing factors of colonoscopy-associated colorectal perforation;(3)construction of prediction model of colonoscop-associated colorectal perforation;(4)management of colonoscopy-associated colorectal perforation.Measurement data with normal distribution were represented as Mean±SD.Measurement data with skewed distribution were represented as M(Q1,Q3),and comparison between groups was analyzed using the Wilcoxom rank sum test.Count data were discribed as absolute numbers,and comparison between groups was analyzed using the chi-square test.Multivariate ana-lysis was conducted using the binary Logistic regression model.The sensitivity and specificity of the prediction model were evaluated by the receiver operating characteristic(ROC)curve.Results(1)Situations of colonoscopy-associated colorectal perforation.Of the 358 patients,18 cases developed colorectal perforation,including 6 males and 12 females,with an age of 61(49,69)years.Of the 18 patients,there were 12 cases with colon perforation,including 10 cases of sigmoid colon perfora-tion or rectosigmoid junction perforation,1 case of transverse colon perforation and 1 case of descending colon perforation,6 cases with rectal perforation.There were 11 cases with diagnostic perforation and 7 cases with therapeutic perforation.(2)Analysis of influencing factors of colonoscopy-associated colorectal perforation.Results of univariate analysis showed that gender,age,colorectal ulcer,colorectal diverticulum,colorectal tumor,history of abdominal surgery,type of colonoscopy and the experience of operating physician were related factors for colonoscopy-associated colorectal perforation(χ^(2)=5.77,Z=‒3.24,χ^(2)=37.99,97.34,37.99,10.31,8.07,6.73,P<0.05).Results of multi-variate analysis showed that colorectal diverticulum and abdominal surgery history were indepen-dent risk factors for colonoscopy-associated colorectal perforation(odds ratios=287.79,6.74,95%confidence intervals as 23.14‒3579.11,1.19‒38.27,P<0.05).Therapeutic colonoscopy was an independent protective factor for colonoscopy-associated colorectal perforation(odds ratio=0.11,95%confidence interval as 0.23‒0.52,P<0.05).(3)Construction of prediction model of colonoscopy-associated colorectal perforation.With the colonoscopy-associated colorectal perforation as depen-dent variable,colorectal diverticulum,abdominal surgery history and therapeutic colonoscopy as independent variables,a prediction model of colonoscopy-associated colorectal perforation was constructed.The ROC of model showed that the sensitivity was 0.56,the specificity was 1.00,and the area under curve was 0.78(95%confidence interval as 0.63‒0.92,P<0.05).(4)Management of colonoscopy-associated colorectal perforation.Of the 18 cases with colonoscopy-associated colorectal perforation,15 cases underwent laparoscopic perforation repair surgery immediately,2 cases under-went endoscopic suture,and 1 case received conservative treatment.All the patients with perfora-tion were cured and discharged from hospital,without death due to colonoscopy-associated colorectal perforation.Conclusions Colonoscopy-associated colorectal perforation is easy to occur at sigmoid colon or rectosigmoid junction.Colorectal diverticulum and abdominal surgery history are indepen-dent risk factors for colonoscopy-associated colorectal perforation.Therapeutic enteroscopy is an independent protective factor for colonoscopy-associated colorectal perforation.Laparoscopic repair of colon perforation has good effects for patients with colorectal perforation.