The Effects of Early Bispectral Index to Predict Poor Neurological Function in Cardiac Arrest Patients: A Systematic Review and Meta-Analysis

Abstract

The diagnostic performance of the bispectral index (BIS) to early predict neurological outcomes in patients achieving return of spontaneous circulation (ROSC) after cardiac arrest (CA) remained unclear. We searched PubMed, EMBASE, Scopus and CENTRAL for relevant studies through October 2019. Methodologic quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. Meta-analysis was performed using a linear mixed-effects model to the log-transformed data with a logistic distribution assumption. Bivariate meta-regression was performed to explore heterogeneity. In total, 13 studies with 999 CA adult patients were included. At the optimal threshold of 32, BIS obtained within 72 h of ROSC elicits a pooled sensitivity of 84.9% (95% confidence interval (CI), 71.1% to 92.7%), a pooled specificity of 85.9% (95% CI, 71.2% to 93.8%) and an area under the curve of 0.92. Moreover, a BIS cutoff < 12 yielded a pooled specificity of 95.0% (95% CI, 77.8% to 99.0%). In bivariate meta-regression, the timing of neurological outcome assessment, the adoption of targeted temperature management, and the administration of sedative agents or neuromuscular blocking agents (NMBA) were not identified as the potential source of heterogeneity. BIS retains good diagnostic performance during targeted temperature management (TTM) and in the presence of administrated sedative agents and NMBA. In conclusion, BIS can predict poor neurological outcomes early in patients with ROSC after CA with good diagnostic performance and should be incorporated into the neuroprognostication strategy algorithm.