The preoperative cognitive test performance of 77 candidates for elective open-heart surgery that did not suffer from carotid artery stenosis, was compared with the performance of 37 normal controls. Multivariate analysis of variance, with state anxiety score and depression score as covariates, showed a significantly reduced cognitive performance in the surgical candidates. Post hoc analyses using univariate F tests revealed a significantly impaired word fluency, manual dexterity, verbal learning, and psychomotor speed. A stepwise multiple regression analysis revealed that, besides the usual demographic factors, preoperative variables indicating a more precarious heart function also contributed significantly to the estimation of the proportion of variance of the results on several impaired cognitive tests. Future research should focus on the etiology of the cognitive impairment in cardiovascular disease. (JINS, 1997, 3, 480–484.)