摘要
目的了解结肠癌伴肠梗阻患者术后切口感染的病原菌分布,分析结肠癌伴肠梗阻患者术后切口感染的危险因素,并构建预测结肠癌伴肠梗阻患者术后切口感染的列线图模型。方法选择2016年1月-2018年12月在本院行外科手术治疗的结肠癌伴肠梗阻患者423例作为训练集;另选取2019年1月-2021年12月在本院行外科手术治疗的结肠癌伴肠梗阻患者410例作为验证集,均依据术后切口感染情况分成感染组(n=69)与未感染组(n=354),用二元Logistic回归分析影响结肠癌伴肠梗阻患者术后切口感染的危险因素;构建预测结肠癌伴肠梗阻患者术后切口感染的列线图模型,用受试者工作特征(ROC)与校准曲线验证预测结肠癌伴肠梗阻患者术后切口感染列线图模型的区分度和一致性。结果423例结肠癌伴肠梗阻患者中,69例发生术后切口感染(构成比16.31%),共培养出96株细菌,包括36株革兰阳性菌(占37.50%)、60株革兰阴性菌(占62.50%)。感染组与未感染组在年龄、是否合并糖尿病、是否合并高血压、手术方式、是否术前化疗、术前是否低蛋白、手术时间、术后留置导尿管时间方面比较差异有统计学意义(P<0.05)。Logistic回归分析结果显示:手术时间≥120 min、合并糖尿病、术后留置导尿管时间≥3 d、年龄≥60岁、术前化疗、手术方式为开腹均是影响结肠癌伴肠梗阻患者术后切口感染的危险因素(P<0.05)。训练集ROC曲线下面积为0.892(95%CI:0.850~0.934);校准曲线斜率接近1,Hosmer-Lemeshow拟合优度检验χ2=8.892,P=0.352。外部验证结果显示ROC曲线下面积为0.786(95%CI:0.729~0.843),校准曲线斜率接近1。结论手术时间≥120 min、合并糖尿病、术后留置导尿管时间≥3 d、年龄≥60岁、术前化疗、手术方式为开腹均是影响结肠癌伴肠梗阻患者术后切口感染的危险因素,据此构建的列线图预测模型,能较准确预测结肠癌伴肠梗阻患者术后切口感染风险。
Objective To investigate the distribution of pathogens of postoperative incision infection in patients with colon cancer and intestinal obstruction,analyze the influencing factors of postoperative incision infection in patients with colon cancer and intestinal obstruction,and to construct a nomogram model for predicting postoperative incision infection in patients with colon cancer and intestinal obstruction.Methods A total of 423 patients with colon cancer and intestinal obstruction who underwent surgical treatment in our hospital from January 2016 to December 2018 were gathered as the modeling group;in addition,410 patients with colon cancer with intestinal obstruction who underwent surgical treatment in our hospital from January 2019 to December 2021 were selected as the validation group.According to postoperative wound infection,the patients were separated into infected group(n=69)and uninfected group(n=354).Logistic regression was performed to analyze the risk factors of postoperative incision infection in patients with colon cancer and intestinal obstruction;a nomogram model was constructed for predicting postoperative incision infection in patients with colon cancer and intestinal obstruction,the discrimination and consistency of the nomogram model for predicting postoperative incision infection in patients with colon cancer and intestinal obstruction were validated by receiver operating characteristic(ROC)and calibration curve.Results Among the 423 patients with colon cancer and intestinal obstruction,69(16.31%)had postoperative wound infection,a total of 96 bacterial strains were cultured,including 36 Gram-positive bacteria(37.50%)and 60 Gram-negative bacteria(62.50%).There were statistically significant differences between the infected group and the uninfected group in terms of age[78.26%(54/69).48.31%(171/354)],diabetes mellitus[60.87%(42/69),25.42%(90/354)],hypertension[52.17%(36/69),25.42%(90/354)],surgical method[78.26%(54/69),45.76%(162/354)],preoperative chemotherapy[39.13%(27/69),8.47%(30/354)],preoperative low protein[78.26%(54/69),43.22%(153/354)],operation time[39.13%(27/69),17.80%(63/354)],and postoperative indwelling catheter time[43.48%(30/69),13.56%(48/354)](P<0.05).Logistic regression analysis showed that operation time≥120 min(OR=5.017,95%CI=2.332-10.796,P=0.000),complicated with diabetes mellitus(OR=7.001,95%CI=3.399-14.419,P=0.000),postoperative indwelling catheter time≥3 days(OR=4.844,95%CI=2.271-10.334,P=0.000),age≥60 years(OR=3.761,95%CI=1.672-8.461,P=0.001),preoperative chemotherapy(OR=11.513,95%CI=4.743-27.946,P=0.000),and laparotomy(OR=4.606,95%CI=2.012-10.545,P=0.000)were all risk factors for postoperative incision infection in patients with colon cancer and intestinal obstruction(P<0.05).In the modeling group,the area under the ROC curve was 0.892(95%CI:0.850-0.934);the slope of the calibration curve was close to 1,Hosmer-Lemeshow goodness-of-fit test X2=8.892,P=0.352.The external verification results show that the slope of the calibration curve is close to 1,and the area under the ROC curve is 0.786(95%CI=0.729-0.843).Conclusion Operation time≥120 min,diabetes mellitus,postoperative indwelling catheter time≥23 days,age≥60 years,preoperative chemotherapy,and laparotomy are all risk factors for postoperative incision infection in patients with colon cancer and intestinal obstruction,the nomogram prediction model constructed on this basis can more accurately predict the risk of postoperative incision infection in patients with colon cancer and intestinal obstruction.
作者
苗菲菲
朱晓翠
刘兆玮
张红
李宁
张长庚
MIAO Fei-fei;ZHU Xiao-cui;LIU Zhao-wei;ZHANG Hong;LI Ning;ZHANG Chang-geng(Laboratory Department of Hengshui People's Hospital,Hebei Province,Hengshui 053000,Hebei,China)
出处
《中国病原生物学杂志》
CSCD
北大核心
2023年第3期336-341,共6页
Journal of Pathogen Biology
基金
衡水市科技计划项目(No.2020014064Z)
关键词
结肠癌
肠梗阻
切口感染
影响因素
列线图预测模型
colon cancer
intestinal obstruction
incision infection
influencing factors
nomogram prediction model
