Association between Preoperative Cortical Magnetic Susceptibility and Postoperative Changes in the Cerebral Blood Flow on Cognitive Improvement following Carotid Endarterectomy

Introduction: While patients who experience improved cognition following carotid endarterectomy (CEA) typically demonstrate restored brain perfusion after the procedure, it is worth noting that less than 50% of patients in whom postoperative cerebral blood flow (CBF) restoration is achieved actually show improved cognition after postoperatively. This suggests that factors beyond the mere restoration of CBF may play a role in postoperative cognitive improvement. Increased iron deposition in the cerebral cortex may cause neural damage, and quantitative susceptibility mapping (QSM) obtained using magnetic resonance imaging (MRI) quantifies magnetic susceptibility in the cerebral cortex, allowing for the assessment of iron deposition in vivo. The purpose of the present study was to determine whether preoperative cortical magnetic susceptibility as well as postoperative changes in CBF are associated with cognitive improvement after CEA. Methods: Brain MRI with a three-dimensional gradient echo sequence was preoperatively performed in 53 patients undergoing CEA for ipsilateral internal carotid artery stenosis (>= 70%), and QSM with brain surface correction and vein removal was obtained. Cortical magnetic susceptibility was measured in the cerebral hemisphere ipsilateral to surgery on QSM. Preoperatively and at 2 months after the surgery, brain perfusion single-photon emission computed tomography and neuropsychological assessments were conducted. Using these collected data, we evaluated alterations in CBF within the affected hemisphere and assessed cognitive improvements following the operation. Results: A logistic regression analysis showed that a postoperative greater increase in CBF (95% confidence interval [CI], 1.06-1.90; p = 0.0186) and preoperative lower cortical magnetic susceptibility (95% CI, 0.03-0.74; p = 0.0201) were significantly associated with postoperatively improved cognition. Although sensitivity, specificity, and positive and negative predictive values with the cutoff value lying closest to the upper left corner of a receiver operating characteristic curve for the prediction of postoperatively improved cognition did not differ between postoperative changes in CBF and preoperative cortical magnetic susceptibility, the specificity and the positive predictive value were significantly greater for the combination of postoperative changes in CBF and preoperative cortical magnetic susceptibility (specificity, 95% CI, 93-100%; positive predictive value 95% CI, 68-100%) than for the former parameter alone (specificity, 95% CI, 63-88%; positive predictive value 95% CI, 20-64%). Conclusion: Preoperative cortical magnetic susceptibility as well as postoperative changes in CBF are associated with cognitive improvement after CEA.