Brain temperature measured by using proton MR spectroscopy predicts cerebral hyperperfusion after carotid endarterectomy.

PURPOSE To determine whether brain temperature measured by using preoperative proton magnetic resonance (MR) spectroscopy could help identify patients at risk for cerebral hyperperfusion after carotid endarterectomy (CEA). MATERIALS AND METHODS Institutional review board approval and informed consent were obtained. Acquisition of proton MR spectroscopic data by using point-resolved spectroscopy without water suppression was performed before CEA in the bilateral cerebral hemispheres of 84 patients with unilateral internal carotid artery stenosis (> or =70%) and without contralateral internal carotid artery steno-occlusive disease. Brain temperature was calculated from the chemical shift difference between water and N-acetylaspartate signals at proton MR spectroscopy. Cerebral blood flow (CBF) was also measured by using single photon emission computed tomography and N-isopropyl-p-[(123)I]-iodoamphetamine before and immediately after CEA and on the 3rd postoperative day. The relationship between each variable and the development of post-CEA hyperperfusion (CBF increase > or = 100% compared with preoperative values) was evaluated with univariate statistical analysis followed by multivariate analysis. RESULTS A linear correlation was observed between preoperative brain temperature difference (the value in the affected hemisphere minus the value in the contralateral hemisphere) and increases in CBF immediately after CEA (r = 0.763 and P < .001) when the preoperative brain temperature difference was greater than 0. Cerebral hyperperfusion immediately after CEA was observed in nine patients (11%). Elevated preoperative brain temperature difference was the only significant independent predictor of post-CEA hyperperfusion. When elevated brain temperature difference was defined as a marker of hemodynamic impairment in the affected cerebral hemisphere, use of preoperative brain temperature difference resulted in 100% sensitivity and 87% specificity, with a 47% positive predictive value and a 100% negative predictive value for the prediction of post-CEA hyperperfusion. Hyperperfusion syndrome developed on the 3rd and 4th postoperative days in two of the nine patients who exhibited hyperperfusion immediately after CEA. CONCLUSION Brain temperature measured by using preoperative proton MR spectroscopy may help identify patients at risk for post-CEA cerebral hyperperfusion.