Comparison of pre-existing cognitive impairment, amnesic mild cognitive impairment, and multiple domain mild cognitive impairment in men scheduled for coronary artery surgery.

Context: Preoperative cognitive impairment (PreCI), amnesic mild cognitive impairment (aMCI; a risk factor for Alzheimer's disease) and multiple domain MCI with amnesia (mdMCI+a) identify preoperative cognitive abnormalities in patients scheduled for coronary artery bypass graft surgery (CABG).

Objectives: Preoperative cognitive function is important in determining the incidence and severity of postoperative cognitive dysfunction, but the methods used to measure preoperative cognition are diverse. The incidence of preoperative cognitive abnormalities in cardiac surgical patients is unknown.

Design: Observational study.

Setting: Veteran Administration Medical Center in the United States of America.

Participants: Elective CABG surgery patients and nonsurgical controls.

Main Outcome Measures: Recent verbal and nonverbal memory and executive functions were assessed using psychometric tests. Associations were determined using multivariate regression analysis. We quantified the incidences of PreCI, aMCI and mdMCI+a in patients before elective CABG surgery.

Results: Patients (n = 100) scheduled for surgery had lower cognitive scores compared with nonsurgical controls (n = 100) on eight of 10 psychometric tests. Differences between patients and controls were 'large' for the Immediate and Delayed Story Recall tests (Cohen's d; wherein d is defined as a difference between two means divided by the pooled SD for those two means), 'medium' for the Immediate Word List Recall test and 'small' for five other tests. Twenty-five, 20, and 21 patients satisfied the criteria for PreCI, aMCI and mdMCI+a, respectively. Fifteen patients satisfied the criteria for both PreCI and mdMCI+a, whereas only five did so for both PreCI and aMCI. PreCI was not significantly associated with aMCI (P > 0.05), but was significantly associated with mdMCI+a (P < 0.00001).

Conclusion: PreCI, aMCI and mdMCI+a identified preoperative cognitive abnormalities in different groups of patients scheduled for CABG surgery. The findings emphasise that identification of patients with preoperative cognitive dysfunction is methodology-dependent. The optimal approach to subsequently identify patients who may be at greatest risk of progressive cognitive deterioration after CABG surgery remains to be determined.