妇科恶性肿瘤患者术后胃肠功能紊乱风险预测模型的构建及验证  

Construction and Validation of Risk Prediction Model for Postoperative Gastrointestinal Dysfunction in Patients with Gynecological Malignant Tumor

作　　者：赵晓蕊 龙云[3] 陈思齐 白洁[2] 肖霄 朱社宁[2] ZHAO Xiao-rui;LONG Yun;CHEN Si-qi;BAI Jie;XIAO Xiao;ZHU She-ning(Shanxi University of Traditional Chinese Medicine,Jinzhong 030619,China;Dept.of Nursing Administration,Shenzhen Maternity&Child Healthcare Hospital,Shenzhen 518028,China;Dept.of Gynecology,Shenzhen Maternity&Child Healthcare Hospital,Shenzhen 518028,China;Maternal and Child Medical Institute,Shenzhen Maternity&Child Healthcare Hospital,Shenzhen 518028,China)

机构地区：[1]山西中医药大学,山西晋中030619 [2]深圳市妇幼保健院护理部,广东深圳518028 [3]深圳市妇幼保健院妇科,广东深圳518028 [4]深圳市妇幼保健院妇幼医学研究所,广东深圳518028

出　　处：《护理学报》2022年第3期72-78,共7页Journal of Nursing（China）

摘　　要：目的构建妇科恶性肿瘤患者术后发生胃肠功能紊乱的风险预测模型,并验证该模型的预测效果。方法选择2015年1月1日—2020年12月31日在深圳市妇幼保健院妇科行恶性肿瘤根治术的281例患者作为模型建立对象,将其分为发生术后胃肠功能紊乱(n=109)和未发生术后胃肠功能紊乱(n=172)。通过LASSO模型进行变量的筛选,将LASSO模型筛选出的变量纳入Logistic回归模型,构建妇科恶性肿瘤术后胃肠功能紊乱的风险预测模型,并采用十折交叉验证,计算模型工作曲线下面积(AUC)检验模型的预测效果。结果本研究中术后胃肠功能紊乱的发生率为38.8%;其影响因素分别为:BMI(OR=0.594,95%CI:0.381~0.907)、血清钠水平(OR=0.915,95%CI:0.822~1.015)、大网膜手术切除(OR=8.388,95%CI:1.741~6.569)、盆腔淋巴结清扫切除(OR=4.148,95%CI:1.368~1.665)、麻醉时间(OR=1.528,95%CI:0.910~2.612)和尿pH(OR=0.773,95%CI:0.579~1.015)。受试者操作特征曲线下面积AUC为0.70(95%CI:0.63~0.75,P<0.001),灵敏度为69%,特异度为60%。结论本模型在预测妇科恶性肿瘤患者术后发生胃肠功能紊乱风险具有较为可靠的预测价值,有利于医务人员筛查高风险患者和制定针对性的护理干预措施。Objective To establish a risk prediction model for postoperative gastrointestinal dysfunction in patients with gynecological malignant tumor,and to verify the prediction effect of the model.Methods A total of 281 patients with malignant tumor who underwent radical surgery in Shenzhen Maternity and Child Healthcare Hospital from January 1,2015 to December 31,2020 were selected as the object and they were divided into two groups:patients with postoperative gastrointestinal dysfunction(n=109)and patients without postoperative gastrointestinal dysfunction(n=172).Variables were screened by the Lasso model,and the variables screened out by the Lasso model were incorporated into the logistics regression model to construct the risk prediction model of gastrointestinal dysfunction after gynecological malignant tumor surgery.The area under the working curve(ROC)of the model was calculated to test the differentiation of the model by using the ten-fold cross-validation method.The goodness of fit and prediction effect of the models were tested respectively.Results The incidence of postoperative gastrointestinal dysfunction was 38.8%.The influencing factors were BMI(OR=0.594,95%CI:0.381~0.907),serum sodium level(OR=0.915,95%CI:0.822~1.015),omentum greater surgical resection(OR=8.388,95%CI:1.741~6.569),pelvic lymph node dissection(OR=4.148,95%CI:1.368~1.665),anesthesia time(OR=1.528,95%CI:0.910~2.612)and uric acid alkalinity(OR=0.773,95%CI:0.579~1.015).The AUC under the receiver operating characteristic curve was 0.70(95%CI:0.63~0.75,P<0.001);the sensitivity 69%,and the specificity 60%.Conclusion The model established in the study is relatively reliable in predicting the risk of postoperative gastrointestinal dysfunction in patients with gynecological malignant tumor,and it is helpful for medical personnel to screen high-risk patients and develop targeted nursing interventions.

关 键 词：妇科 恶性肿瘤 危险因素 胃肠功能紊乱 预测模型 

分 类 号：R473.71[医药卫生—护理学]