METHODS: Forty-six patients, including 34 women and 12 men (Glasgow Coma Scale Score < or = 8 and median age 58.5 years) who underwent
PbtO(2) monitoring were studied prospectively during a 2-year period in a neurosurgical intensive care unit at a University Level I Trauma Center. Brain
oxygen tension, intracranial pressure (ICP), mean arterial pressure, cerebral perfusion pressure (
CPP), and brain temperature were continuously monitored, and treatment was directed toward ICP,
CPP, and
PbtO(2) targets. The relationship between
PbtO(2) and 1-month survival was examined.
RESULTS: Data were available from 5424 hours of
PbtO(2) monitoring. For the entire cohort the mean ICP,
CPP, and
PbtO(2) were 13.85 +/- 2.40, 84.05 +/- 3.41, and 30.79 +/- 1.91 mm Hg, respectively. Twenty-five patients died (54%). The mean daily
PbtO(2) was higher in survivors than nonsurvivors (33.94 +/- 2.74 vs 28.14 +/- 2.59 mm Hg; p = 0.05). In addition, survivors had significantly shorter episodes of compromised
PbtO(2) (defined as 15-25 mm Hg) than nonsurvivors (125.85 +/- 15.44 vs 271.14 +/- 55.23 minutes; p < 0.01). Intracranial pressure was similar in survivors and nonsurvivors. In contrast, the average
CPP was significantly lower in nonsurvivors than survivors (76.96 +/- 5.50 vs 92.49 +/- 2.75 mm Hg; p = 0.01). When
PbtO(2) was stratified according to
CPP level, survivors had higher
PbtO(2) levels. Following logistic regression, the number of episodes of compromised
PbtO(2) (odds ratio 1.1, 95% confidence interval 1.003-1.2) and number of episodes of
cerebral hypoxia (< 15 mm Hg; odds ratio 1.3, 95% confidence interval 1.0-1.7) were more frequent in those who died.
CONCLUSIONS: Patient deaths after SAH may be associated with a lower mean
PbtO(2) and longer periods of compromised cerebral oxygenation than in survivors. This knowledge may be used to help direct
therapy.