Utility of neuron-specific enolase in traumatic brain injury; relations to S100B levels, outcome, and extracranial injury severity

Table 2 Univariate outcome prediction

	GOS 1–5 (proportional odds analysis)
	p value	Nagelkerke’s pseudo-R² (coefficient)
Gender (female)	0.760	0.000 (-)
Age	<0.001	0.151 (-, higher age = lower GOS)
Pupil unresponsiveness	<0.001	0.074 (-, if present = lower GOS)
Glasgow Coma Score (GCS)	<0.001	0.070 (+, higher GCS = higher GOS)
Multitrauma	0.867	0.000 (+)
Hypoxemia	0.040	0.015 (-)
Hypotension	0.494	0.002 (-)
Glucose	0.010	0.022 (-)
Hemoglobin	<0.001	0.044 (+)
Head Abbreviated Injury Score (AIS)	0.002	0.029 (-)
Marshall	0.190	0.004 (-)
Rotterdam	<0.001	0.048 (-)
Stockholm	<0.001	0.132 (-)
Stockholm subarachnoid hemorrhage (SAH)	<0.001	0.079 (-)
S100B admission (log)	<0.001	0.051 (-)
Neuron-specific enolase (NSE) admission (log)	0.025	0.013 (-)
S100B area under the curve (AUC) 48 h (log)	<0.001	0.132 (-)
NSE AUC 48 h (log)	0.001	0.038 (-)
	GOS 1–3 vs 4–5 (bivariate regression analysis)
S100B admission (log)	<0.001	0.061 (-)
NSE admission (log)	0.096	0.009 (-)
S100B AUC 48 h (log)	<0.001	0.127 (-)
NSE AUC 48 h (log)	0.002	0.032 (-)
	GOS 1 vs 2–5 (bivariate regression analysis)
S100B admission (log)	0.001	0.054 (-)
NSE admission (log)	0.008	0.027 (-)
S100B AUC 48 h (log)	<0.001	0.179 (-)
NSE AUC 48 h (log)	<0.001	0.077 (-)

Table illustrating different un-imputed parameters versus different outcome dichotomizations. Nagelkerke’s pseudo-R² and regression coefficients are shown to facilitate interpretation. A negative coefficient means that a higher level of the parameter correlated to a lower Glasgow Outcome Score (GOS) (e.g., age) and vice versa (e.g., GCS). Bold indicates significance (p <0.05)

ISSN: 1364-8535