Automated measurement of "pressure times time dose" of intracranial hypertension best predicts outcome after severe traumatic brain injury.

BACKGROUND: Earlier, more accurate assessment of secondary brain injury is essential in management of patients with traumatic brain injury (TBI). We assessed the accuracy and utility of high-resolution automated intracranial pressure (ICP) and cerebral perfusion pressure (CPP) recording and their analysis in patients with severe TBI.

METHODS: ICP and CPP data for 30 severe TBI patients were collected automatically at 6-second intervals. The degree and duration of ICP and CPP above and below treatment thresholds were calculated as "pressure times time dose" (PTD; mm Hg . h) using automated recordings (PTDa) or manual recordings (PTDm) for early stage (trauma resuscitation unit [TRU]) and total monitoring time (TRU and intensive care unit).

RESULTS: Bland-Altman plots showed lack of agreement between PTDa and PTDm. For ICP >20 mm Hg and CPP <60 mm Hg, PTDa, but not PTDm, was significantly higher in patients with unfavorable outcome (Extended Glasgow Outcome Scale score 4). Total PTDa for ICP >20 mm Hg and CPP <60 mm Hg had high predictive power for functional outcome (area under the receiver operating characteristics curve: 0.92 +/- 0.05 and 0.82 +/- 0.08, respectively) and inhospital mortality (0.76 +/- 0.15 and 0.79 +/- 0.14, respectively) and were strongly correlated with length of intensive care unit stay (p = 0.009 and 0.007), length of hospital stay (p = 0.009 and 0.005), and discharge Glasgow Coma Scale scores (p = 0.008 and p = 0.038). PTDa of CPP >100 mm Hg during TRU monitoring and during the first 24 hours showed highest predictive power for mortality (area under the receiver operating characteristics curve: 0.72 +/- 0.18 and 0.85 +/- 0.13, respectively). PTDa was better than PTDm and the duration of episodes alone in predicting outcome.

CONCLUSIONS: PTD calculation of high resolution ICP and CPP recording is a reliable and feasible way of monitoring severe TBI patients.