Relation between neurocognitive impairment, embolic load, and cerebrovascular reactivity following on- and off-pump coronary artery bypass grafting.

OBJECTIVES To evaluate the effect of on-pump and off-pump coronary artery bypass grafting (CABG) on postoperative cognitive impairment and cerebrovascular reactivity, with attention for the perioperative high-intensity transient signals (HITS). DESIGN A prospective comparative study. SETTING Urban university hospital. PATIENTS Candidates for cardiac surgery. METHODS Measurement of HITS as a reflection of embolic load was performed in 50 patients (on-pump CABG, n = 32; off-pump CABG, n = 18). To measure cognitively induced cerebrovascular reactivity, cerebral blood flow velocity (BFV) was measured preoperatively in 66 patients, early postoperatively (after 6 days) in 63 patients, and late postoperatively (after 6 months) in 44 patients during five cognitive tasks. In the same session, seven standardized neuropsychological tests were administered. RESULTS A higher embolic load was found in the on-pump group (p < 0.01). In the on-pump group, aortic cannulation was the most important HITS-prone surgical maneuver. Repeated-measures multivariate analysis of variance (using surgical technique as between-subjects factor and significant differences between both groups as covariates) on the group data revealed no significant differences in neuropsychological performance and BFV immediately after surgery or at 6 months after surgery, compared with preoperative performance. No main effect of surgery was found for neuropsychological performance and BFV. No significant correlations were found between the number of HITS and the degree of postoperative neuropsychological impairment. Individual comparisons revealed that 60% (59.4% in the on-pump group; 61.1% in the off-pump group) of the patients undergoing CABG showed evidence of cognitive impairment soon after surgery. In 24.2%, the cognitive sequelae persisted at 6 months follow-up (31.8% in the on-pump group; 9.1% in the off-pump group). The cognitive impairment index (sum of impaired neuropsychological tests) showed a significant difference after 6 months between both surgery groups with fewer neurocognitive tests that remained impaired in the off-pump group. CONCLUSIONS In off-pump surgery, significantly fewer HITS were observed. On an individual level, more favorable results in neuropsychological test performance were demonstrated in the off-pump group after 6 months. The number of HITS showed no correlation with degrees of early and late postoperative neuropsychological impairment.