Predictive value of percent change in brain natriuretic peptide for endotracheal tube extubation in invasive mechanically ventilated patients during spontaneous breathing trial: a meta-analysis

Abstract

Objective To assess the predictive value of percent change in brain natriuretic peptide (BNP%) for endotracheal tube extubation in invasive mechanically ventilated patients during spontaneous breathing trial (SBT). Methods Searches were conducted in the PubMed, Embase, Medline, Ovid, Cochrane Central Register of Controlled Trails and Information Sciences Institute Web of Science for studies on the predictive value of BNP% for endotracheal tube extubation in invasive mechanically ventilated patients published up to March 31, 2018. The Quality Assessment of Diagnostic Accuracy Studies (Quasas-2) was used for quality assessment. The MetaDiSc was used for heterogeneity analysis, and the pooled diagnostic odds ratio (DOR), pooled sensitivity, pooled specificity, pooled positive likelihood ratio (PLR), pooled negative likelihood ratio (NLR), and the area under the summary receiver operating characteristic (SROC) curve were calculated. Results A total of 213 patients were recruited from 5 trials. Meta-analysis showed that the pooled DOR was 22.33 [95% confidence interval (CI) (9.25, 53.91)], the pooled sensitivity was 85% [95%CI (0.78, 0.91)], the pooled specificity was 80% [95%CI (0.68, 0.90)], the pooled PLR was 3.72 [95%CI (2.11, 6.55)], and the pooled NLR was 0.18 [95%CI (0.12, 0.28)]. The pooled DOR, sensitivity, specificity, PLR, and NLR showed no significant heterogeneity (all P > 0.1, all I2 < 50%), and the area under the SROC curve was 0.913. Conclusion BNP% can elevate the predictive value for endotracheal tube extubation in invasive mechanically ventilated patients during SBT. Key words: Brain natriuretic peptide; Ventilators, mechanical; Spontaneous breathing trial