The lower limit of reactivity as a potential individualised cerebral perfusion pressure target in traumatic brain injury: a CENTER-TBI high-resolution sub-study analysis

Table 3 Multivariate logistic regression analysis for mortality

Model	AUC (95% CI)	AIC	p value	AdjustedR²
IMPACT core	0.84 (0.78–0.91)	135.06	< 0.001	0.49
IMPACT core + Dose ICP > 20	0.88 (0.82–0.94)	125.12	< 0.001	0.56
IMPACT core + Dose CPP < LLR	0.86 (0.79–0.93)	128.69	< 0.001	0.54
IMPACT core + %Time CPP < LLR	0.85 (0.78–0.92)	130.46	< 0.001	0.53
IMPACT core + Dose CPP < LLR if ICP > 20	0.88 (0.81–0.94)	123.15	< 0.001	0.57
IMPACT core + Dose ICP > 20 if CPP < LLR	0.87 (0.81–0.94)	125.42	< 0.001	0.56

The table presents the multivariate logistic regression analysis for mortality prediction for variables that describe the relationship between CPP and LLR and between CPP, LLR and ICP, with IMPACT core covariates. The table shows only models where the tested variable added significant contribution in predicting mortality, when considered together with the IMPACT core variables. IMPACT: International Mission for Prognosis and Analysis of Clinical Trials
ICP intracranial pressure, CPP cerebral perfusion pressure, LLR lower limit of reactivity. AIC Akaike Information Criterion, AUC Area Under the Curve

ISSN: 1364-8535