Meta-analysis of Glasgow coma scale and simplified motor score in predicting traumatic brain injury outcomes

Abstract

Objective: To perform a systematic review and meta-analysis to compare the simplified motor score (SMS) and Glasgow Coma Scale (GCS) in predicting outcomes in patients with traumatic brain injury (TBI).

Data sources and study selection: Ovid EMBASE, Ovid Medline, Ovid PsycInfo, evidence-based medicine reviews and Scopus and related conference proceedings were searched through 28 February 2012 for studies comparing SMS and GCS in predicting the outcomes [emergency tracheal intubation (ETI), clinically significant brain injuries (CSBI), neurosurgical intervention (NSI) and mortality] in patients with TBI. A random-effects model was used for meta-analysis.

Data synthesis: Five retrospective studies were eligible, enrolling a total of 102 132 subjects with TBI (63.4% males), with 14 670 (14.4%) ETI, 16 201 (15.9%) CSBI, 4730 (4.6%) NSI and 6725 (6.6%) mortality. Pooled AUC of the GCS and SMS were as follows: CSBI 0.79 and 0.75 (p = 0.16), NSI 0.83 and 0.81 (p = 0.34), ETI 0.85 and 0.82 (p = 0.31) and mortality 0.90 and 0.87 (p = 0.01). The difference in AUC for mortality was 0.03. Large heterogeneity between the studies was observed in all analyses (I(2 )> 50%).

Conclusion: In patients with TBI, SMS predicts different outcomes with similar accuracy as GCS except mortality. However, due to heterogeneity and limited numbers of studies, further prospective studies are required.