N端脑钠素前体对机械通气撤机结局的预测意义

目的 探讨达到撤机条件的机械通气患者撤机前N端脑钠素前体(NT-proBNP)水平与撤机结局间的关系.方法 选择2008年8月至2009年12月重症监护病房(ICU)机械通气患者126例,统计其病因构成.在患者达到撤机条件时取血测定NT-proBNP水平,随后依次进行自主呼吸试验(SBT)并序贯撤机拔管.根据48 h内的撤机结局,比较撤机成功组与失败组NT-proBNP水平的差异,并绘制受试者工作特征曲线(ROC曲线),寻找预测撤机结局的最佳NT-proBNP水平.结果 撤机前患者的上机病因组成仍以肺部感染(占33.3%)和外科手术后(占30.2%)为主,以心力衰竭(心衰)为上机原因的仅占11.9%.撤机前患者的血浆NT-proBNP水平与撤机结局存在关联:失败组(38例)lg NT-proBNP水平较成功组(88例)明显增高(3.97±0.48比2.99±0.67,P<0.05);NT-proBNP的ROC曲线下面积为0.875±0.043,95%可信区间(95%CI)为0.792～0.959,其预测撤机失败的截点值为3 914.5 ng/L,在此截点值下的敏感性为78.3%、特异性为91.1%.结论 不论机械通气的病因如何,心功能因素在撤机时都必须考虑;3 914.5 ng/L血浆NT-proBNP水平可以预测撤机结局,达到该截点值应对患者进行改善心功能的治疗,从而提高撤机成功率.

Prognostic implication of plasma N-terminal-pro-brain natriuretic peptide in weaning from mechanical ventilation

Objective To explore the relationship between plasma N-terminal-pro-brain natriuretic peptide (NT-proBNP) levels of patients before weaning and outcome of weaning from mechanical ventilation (MV) in patients. Methods A total of 126 intensive care unit (ICU) patients with MV were enrolled from August 2008 to December 2009, and the cause composition was recorded. Plasma NT-proBNP levels were measured in patients with MV, in whom the clinical data had fulfilled the criteria for weaning from MV, spontaneous breathing trial, weaning and extubation were performed successively. The enrolled patients were divided into two groups namely success group and failure group according to weaning outcome within 48-hour. The plasma NT-proBNP levels in two groups were compared, and receiver operating characteristic (ROC) curve for predicted weaning outcome was plotted to find the cut-off point value of NT-proBNP. Results The major causes of MV were pulmonary infection (33.3%) and surgical operations (30.2%), and heart failure accounted for only 11.9%. The plasma NT-proBNP levels before weaning were negatively correlated with the consequences of weaning. The plasma NT-proBNP levels in failure group (n=38) were significantly higher than those in success group (n=88, lg NT-proBNP: 3.97±0.48 vs. 2.99±0.67, P<0.05). The NT-proBNP area under ROC curve was 0.875±0.043 [95% confidence interval (95%CI) was 0.7920.959]. The cut-off point value which could be used to predict the outcome of weaning was 3 914.5 ng/L. The sensitivity and specificity of the cut-off point value were 78.3% and 91.1%, respectively. Conclusion Irrespective of the causes of institution of MV, the cardiac function must be considered as an important factor in affecting the outcome of weaning. The plasma NT-proBNP level of 3 914.5 ng/L can be used to predict weaning outcome. The cardiac function should be improved to a point within the range of cut-off point value in order to improve the success rate of weaning.