脑出血患者术后延迟脱机的危险因素分析及预测模型建立

Analysis of risk factors for postoperative delayed weaning in patients with intracerebral hemorrhage and establishment of a predictive model

目的 分析脑出血患者术后延迟脱机的危险因素并建立预测模型。方法 选择2014年10月-2021年10月北京大学人民医院重症医学科(ICU)收治的因脑出血行外科手术的207例患者的临床资料进行回顾性分析。根据2007年欧洲呼吸学会共识,将207例患者分为延迟脱机组(n=66)与非延迟脱机组(n=141)。比较两组患者基本情况及手术相关状况;采用多因素logistic回归分析患者延迟脱机的危险因素,并构建延迟脱机风险评估列线图并验证。结果 与非延迟脱机组比较,延迟脱机组患者合并心脑血管疾病比例较高、入院格拉斯哥昏迷评分法(GCS)评分较低、术前合并脑疝比例较高,急诊手术、接受骨瓣减压比例较高,且术后急性生理与慢性健康(APACH)Ⅱ评分、GCS评分、发生肺不张比例、脱机失败率、死亡发生率均较高,住ICU时间及住院时间均较长(P<0.05)。多因素logistic回归分析提示,急诊手术、术后氧合指数低、术后至脱机前最高GCS评分低及发生肺不张是导致术后脱机延迟的独立危险因素(P<0.05),联合此4项指标构建Nomogram风险预测模型,其受试者工作特征(ROC)曲线下面积(AUC)为0.855(95%CI 0.804~0.907),Hosmer-Lemeshow检验表明模型拟合度较好(P=0.659)。结论 延迟脱机可导致患者机械通气时间、住ICU时间及住院时间延长,脱机失败率及院内死亡风险增加。Nomogram模型对预测延迟脱机具有一定价值,可帮助临床医师早期发现高风险患者,并及时调整通气策略。

Objective To identify the risk factors for postoperative delayed weaning in patients with intracerebral hemorrhage and to establish a predictive model.Methods A retrospective study was conducted on 207 patients who underwent surgery for intracerebral hemorrhage at the intensive care unit(ICU)of Peking University People's Hospital from October 2014 to October 2021.Utilizing the 2007 European Respiratory Society consensus criteria,patients were divided into delayed weaning group(n=66)and non-delayed weaning group(n=141).The demographic and operation-related conditions of the two groups were compared.Multiple logistic regression analysis was employed to identify the risk factors for delayed weaning,and a risk assessment nomogram was constructed and validated.Results Compared with the non-delayed weaning group,the delayed weaning group exhibited a significantly higher proportion of comorbid cardiovascular and cerebrovascular diseases,a lower Glasgow coma scale(GCS)score on admission,a greater incidence of preoperative brain herniation,a higher proportion of patients undergoing emergency surgeries and decompression craniotomies,and higher postoperative APACHEⅡscore,GCS score,incidence of atelectasis,weaning failure rate and mortality rate,and longer ICU stay and hospital stay(P<0.05).Multiple logistic regression analysis indicated that emergency surgery,low postoperative oxygenation index,low postoperative score of highest GCS before weaning,and incidence of atelectasis were independent risk factors for postoperative delayed weaning(P<0.05).A nomogram predictive model was established using these four predictors,with an area under the receiver operating characteristic(ROC)curve(AUC)of 0.855(95%CI 0.804-0.907),and the Hosmer-Lemeshow test showed good model fit(P=0.659).Conclusions Postoperative delayed weaning is associated with extended durations of mechanical ventilation,ICU stay,and hospital stay,as well as an increased risk of weaning failure and in-hospital mortality rate.The nomogram model provides valuable insights for the early recognition of patients at high risk for postoperative delayed weaning,thereby facilitating timely adjustment in ventilation management strategies.