The aim of this study was to compare blood flow velocity changes in the middle cerebral artery before, during and after heart surgery with cardiopulmonary bypass for patients with and without postoperative cognitive dysfunction.

A total of 100 patients, undergoing elective coronary artery bypass grafting or/and valve surgery enrolled in the study. A neurocognitive test evaluation included Adenbrooke, Mini-Mental State Examination and Trial Making test before and 7-10 days after surgery. Middle cerebral artery mean blood flow velocity was evaluated 1 day before the surgery, after anaesthesia induction, before cardiopulmonary bypass, at the beginning, ending and after cardiopulmonary bypass, and post surgery in intensive care unit. Blood samples for glial fibrillary acidic protein were measured after anaesthesia induction, 24 hours and 48 hours after surgery. According to neurocognitive tests results patients were divided in to two groups: patients with and without postoperative cognitive dysfunction.

Of the 100, 86 patients completed investigation. After induction, blood flow velocity of the middle cerebral artery was lower in postoperative cognitive dysfunction group (41.2; min 27.91, max 49.47) than in the H group (41.2, min 21.9, max 84.3) p = 0.034, and during cardiopulmonary bypass, blood flow velocity of the middle cerebral artery was lower in the postoperative cognitive dysfunction group (37.35, min 26.6, max 44.02) than the H group (42.3, min 20.1, max 86.5), p = 0.001. After the surgery, blood flow velocity of the middle cerebral artery was lower in the postoperative cognitive dysfunction group (40.7, min 29.7, max 50.4) than in the H group (45.3, min 34.12, max 59.88), p = 0.05. Results of cognitive tests had weak correlation (rho, 0.391) with middle cerebral artery’s blood flow velocity after anaesthesia induction (p = 0.001) and during bypass (p = 0.018). The receiver operating characteristic analysis showed that the blood flow velocity of the middle cerebral artery during bypass (area under the curve = 0.735) was a fair predictor for postoperative cognitive dysfunction (p = 0.001). No significant correlations were found among glial fibrillary acidic protein, middle cerebral artery blood flow velocity, and cognitive tests results.