The impact of middle cerebral artery blood flow velocity on delayed neurocognitive recovery in patients undergoing carotid endarterectomy

Abstract

Background. A significant numbers of patients undergo Carotid Endarterectomy (CEA) experience cognitive chan- ges in the early postoperative period. The middle cerebral artery (MCA) is susceptible to fluctuations in blood flow velocity (BFV) changes during carotid cross-clamping. BFV can be measured in real time using Transcranial Dopp- ler (TCD) ultrasonography. Postoperative hypoperfusion syndrome is a proven risk factor for delayed neurocognitive recovery (dNCR), intraoperative reductions in cerebral blood flow velocity may contribute to postoperative cogni- tive decline. Aim. The aim of this study was to evaluate the association between MCA BFV fluctuations and delayed neurocogni- tive recovery and identify risk factors for dNCR. Methods. 27 patients without prior neurocognitive disor- ders undergoing CEA in LUHS Hospital, Kaunas Clinics, were enrolled in the study. Adenbrooke cognitive testing was performed prior to surgery and 1 day after surgery. TCD ultrasonography was performed 1 day before surgery and during CEA. Results. 21 patients were examined; 15 (71.4%) male. Ac- cording to the ACE-III test results, patients were divided into two groups: with dNCR (dNCR group) and without dNCR (non-dNCR group). dNCR developed for seven pa- tients (33.3%). ACM BFV was consistently higher in the postoperative dNCR group, with a statistically significant difference observed before carotid cross-clamping (median [min–max]: 43.1 [11.2–75.1] vs 26.4 [11.9–40.0] cm/s; p = 0.043). No statistically significant differences were ob- served at the remaining time points (p < 0.05). Bias-re- duced logistic regression showed that preoperative cogni- tive impairment was the dominant predictor of postoperative dNCR (penalised â » +3.8). In models exclud- ing preoperative cognitive impairment, preoperative demen- tia (â » +1–4), higher pre-clamping ACM BFV (standard- ized â » +0.6), and lower end-of-surgery rSO? (â » –0.04 to –0.17 per 1% increase) showed consistent directional asso- ciations. No predictor reached statistical significance due to limited sample size. Conclusions. Postoperative dNCR occurred in one-third of patients. Higher pre-clamping MCA BFV was associated with postoperative dNCR, suggesting that altered baseline cerebrovascular haemodynamics may play a role in cogni- tive decline. Elevated pre clamping MCA BFV in carotid stenosis reflects a compensatory hyperdynamic flow from collateral circulation. dNCR was most strongly predicted by preoperative cognitive impairment. When preoperative dNCR was excluded from analysis, dementia, higher pre clamping MCA blood flow velocity, and lower end of sur- gery rScO? showed consistent but non-significant associa- tions, likely due to the limited sample size.