Objective: A prospective observational study was conducted to investigate whether episodes of ischemia are detected by continuous cerebral monitoring and whether such episodes are related to clinical outcome.
Methods: Forty patients (35 after subarachnoid hemorrhage and 5 after complex aneurysm surgery) were monitored for a total of 174 days (mean, 4 d; range, 1-12 d). Brain tissue partial pressures of oxygen and carbon dioxide, pH, and temperature were measured continuously using Neurotrend sensors (Codman, Bracknell, England). Bedside analysis of extracellular chemistry was performed hourly using microdialysis. Glasgow Outcome Scale score was assessed at 3 to 6 months.
Results: Patients with poor World Federation of Neurosurgical Societies grades (4 and 5) or an unfavorable outcome (severe disability or death) had, on average, higher lactate and lactate/pyruvate ratio but lower glucose/lactate ratio (P < or = 0.05). Brain tissue partial pressure of oxygen decreased to below 1.1 kPa in 78% of the patients for 18% (95% confidence interval, 12-24%) of time monitored. There were 197 episodes in which brain tissue partial pressures of oxygen decreased to below 1.1 kPa for at least 30 minutes. Unfavorable outcome was associated with more of these episodes (8.8 episodes; 95% confidence interval, 4.4-13.2 episodes) than favorable outcome (2.2 episodes; 95% confidence interval, 1.1-3.3 episodes), as well as an episode of glutamate levels of more than 10 micromol/L or lactate/pyruvate ratio more than 40 (P < 0.05, chi(2) test).
Conclusion: Intraparenchymal oximetry and microdialysis can detect but fail to predict the development of delayed cerebral ischemia. There were associations between episodes of low brain oxygen, abnormal microdialysis, and unfavorable outcome, but these associations were weak.