Monitoring of brain tissue oxygenation during aneurysm surgery:: prediction of procedure-related ischemic events

Object. The aim of this study was to evaluate the feasibility of monitoring brain tissue oxygenation (PO2) during aneurysm surgery for the detection of procedure-related ischemia. Methods. Between 1997 and 1998, PO2 was monitored prospectively in a cohort of 40 patients (42 recordings) during aneurysm surgery in the anterior circulation within the vascular territory of the aneurysm-bearing artery. The position of the probe used to measure oxygenation levels was verified on computerized tomography (CT) scanning on the 1 st postoperative day. Because of the mislocation of one probe and the malfunction of another, data from only 38 patients (40 recordings) were suitable for analysis. Relative changes from baseline to absolute nadir values of intraoperative PO2 were correlated with simultaneously recorded somatosensory evoked potentials (SSEPs), and cardiovascular and ventilatory parameters. The frequency of ischemic events was evaluated with the aid of CT on the 1st postoperative day as a substitute parameter for intraoperative ischemia. Clinical outcome was evaluated 30 days post-operatively based on the Glasgow Outcome Scale. Except for three, all patients underwent surgery for treatment of a symptomatic aneurysm. Mean baseline PO2 was 23.9 mm Hg (range 2-67.2 mm Hg) before clip application. A relative decrease in PO2 (20% decrease in value compared with baseline) occur-red in 12 patients and was a sensitive indicator for the risk of ischemia during temporary arterial occlusion, but was less predictive of nonocclusive ischemia (sensitivity 0.5; positive predictive value [PPV] 0.42; p > 0.05). Results of receiver operating characteristic analysis demonstrated a postclipping PO2 nadir of 15 mm Hg as a dichotomizing threshold for the prediction of ischemia. This threshold rendered an improved sensitivity (0.9) and PPV (0.56) for procedure-related ischemia (p = 0.0003). The results of utility analysis revealed this monitoring parameter to be clinically diagnostic. Only PO2 monitoring, and not SSEP at the tibial nerve, was predictive of ischemia within the anterior cerebral artery territory. Conclusions. Using 15 mm Hg as a dichotomizing threshold, intraoperative PO2 monitoring enables one to identify patients at risk for procedure-related ischemia during aneurysm surgery and surpasses SSEP monitoring. This newly defined threshold based on intraoperative PO2 monitoring provides a basis for studies on treatments for procedure-related ischemia during aneurysm surgery.