Postoperative cognitive deficit after cardiopulmonary bypass with preserved cerebral oxygenation: a prospective observational pilot study

cSO2 was measured by near infrared spectroscopy in 35 patients during cardiopulmonary bypass. cSO2 was kept above 80% of baseline and above 55% during anesthesia including cardiopulmonary bypass. POCD was tested by trail making test, digit symbol substitution test, Ray's auditorial verbal learning test, digit span test and verbal fluency test the day before and 5 days after surgery. POCD was defined as a decline in test performance that exceeded - 20% from baseline in two tests or more. Correlation of POCD with lowest cSO2 and cSO2 - threshold were determined explorative.POCD was observed in 43% of patients. Lowest cSO2 during cardiopulmonary bypass was significantly correlated with POCD (p = 0.015, r2 = 0.44, without Bonferroni correction). A threshold of 65% for cSO2 was able to predict POCD with a sensitivity of 86.7% and a specificity of 65.0% (p = 0.03, without Bonferroni correction).Despite a relevant decrease of cerebral oxygen saturation was avoided in our pilot study during cardiopulmonary bypass, incidence of POCD was comparable to that reported in patients without monitoring. A higher threshold for cSO2 may be needed to reduce the incidence of POCD.Neurologic deficits after cardiac surgery are common complications with clinical manifestations ranging from stroke to subtle neurocognitive deficits [1].Postoperative cognitive deficit (POCD) is a frequent complication of cardiac surgery with and without cardiopulmonary bypass (CPB) [2]. In a study by Newman and colleagues a significant cognitive decline, defined as a 20% reduction from baseline, occurred in 53% of patients at discharge, 36% at 6 weeks, 24% at 6 months, and 42% at 5 years [3]. Other authors report an incidence ranging from 8% to 40% [4].Cognitive impairment is associated with lower general health after cardiac surgery with important implications for the care of patients undergoing cardiac surgery [5]. A retrospective examination of the influence of multimodal neuromonitoring on the incidence o