全麻病人术后恶心呕吐的危险因素及预模型的构建

目的 筛选全麻病人术后恶心呕吐(PONV)的危险因素,并构建预测模型.方法 选择本院2008年8月至2008年9月全麻下行口腔、耳鼻喉、食管、肺部、骨科、普通外科、妇科手术的病人,年龄＞15岁、ASA Ⅰ～Ⅲ级,收集病人和手术等有关因素,记录术后24 h内恶心呕吐的发生情况.采用Logistic回归分析的方法筛选PONV的危险因素.根据Logistic回归分析得出的危险因素数量和危险因素及其相关系数(多元回归)分别构建两种PONV预测模型,绘制ROC曲线,计算曲线下面积(AUC)以评估预测模型的效能.结果 共纳入350例全麻病人,术后恶心和呕吐的发生率分别为36.6%和22.3%.Logistic回归分析结果显示,女性、PONV史、晕动史、偏头痛史和妇科或腹腔镜手术是PONV的独立危险因素.以危险因素数量建立的预测模型的AUC为0.749±0.027,多元回归预测模型的AUC为0.755±0.027.结论 女性、PONV史、晕动史、偏头痛史、妇科手术或腹腔镜手术是全麻病人PONV的危险因素,根据危险因素的数量可初步筛选PONV高危人群.

Risk factors for postoperative nausea and vomiting after general anesthesia and establishment of predictive model

Objective To investigate the incidence of and risk factors for postoperative nausea and vomiting(PONV)after general anesthesia and pick out patients at high risk of PONV.Methods Patients undergoing stomatological,otorhinolaryngological,esophageal,pulmonary,orthopedic,general and gynecological surgery under general anesthesia from August 2008 to September 2008 in our hospital were included in this study.The clinical data about the patients who developed PONV were collected.Univariate and multivariate analyses were performed with chi-spuare test and logistic stepwise regression analysis to pick out the potential risk factors for PONV.Predictive models were established and their efficacy was assessed with receptor operating curve.Results A total of 350 patients were retrospectively studied.The incidences of postoperative nausea and vomiting were36.6% and 22.3% respectively.Statistical analysis showed that the risk factors for PONV included female,history of PONV,motion sickness and migraine,and gynecological or laparoscopic operation.Predictive models were established based on the number of risk factors and coefficients in logistic regression analysis and the areas under curve were 0.749±0.027 and 0.755±0.027 respectively.Conclusion The risk factors for PONV include female,history of PONV,motion sickness and migraine,and gynecological or laparoscopic operation.The incidence of PONV is fairly high after general surgery and could be predicted from the risk factors.