OBJECTIVE: Barbiturate-induced coma can be used in patients to treat intractable intracranial hypertension when other therapies, such as osmotic therapy and sedation, have failed. Despite control of intracranial pressure, cerebral infarction may still occur in some patients, and the effect of barbiturates on outcome remains uncertain. In this study, we examined the relationship between barbiturate infusion and brain tissue oxygen (PbtO(2)). METHODS: Ten volume-resuscitated brain-injured patients who were treated with pentobarbital infusion for intracranial hypertension and underwent PbtO(2) monitoring were studied in a neurosurgical intensive care unit at a university-based Level I trauma center. PbtO(2), intracranial pressure (ICP), mean arterial pressure, cerebral perfusion pressure (CPP), and brain temperature were continuously monitored and compared in settings in which barbiturates were or were not administered. RESULTS: Data were available from 1595 hours of PbtO(2) monitoring. When pentobarbital administration began, the mean ICP, CPP, and PbtO(2) were 18 +/- 10, 72 +/- 18, and 28 +/- 12 mm Hg, respectively. During the 3 hours before barbiturate infusion, the maximum ICP was 24 13 mm Hg and the minimum CPP was 65 20 mm Hg. In the majority of patients (70%), we observed an increase in PbtO(2) associated with pentobarbital infusion. Within this group, logistic regression analysis demonstrated that a higher likelihood of compromised brain oxygen (PbtO(2) < 20 mm Hg) was associated with a decrease in pentobarbital dose after controlling for ICP and other physiological parameters (P < 0.001). In the remaining 3 patients, pentobarbital was associated with lower PbtO(2) levels. These patients had higher ICP, lower CPP, and later initiation of barbiturates compared with patients whose PbtO(2) increased. CONCLUSION: Our preliminary findings suggest that pentobarbital administered for intractable intracranial hypertension is associated with a significant and independent increase in PbtO(2) in the majority of patients. However, in some patients with more compromised brain physiology, pentobarbital may have a negative effect on PbtO(2), particularly if administered late. Larger studies are needed to examine the relationship between barbiturates and cerebral oxygenation in brain-injured patients with refractory intracranial hypertension and to determine whether PbtO(2) responses can help guide therapy.