A neuropsychologic (NP) study was carried out on 49 patients who had had cardiac valvular replacement to investigate the determinants of postoperative outcome. The results were assessed together with operative data and with neurologic and quantitative EEG (QEEG) findings. Of the operative variables, perfusion time and unexpected intraoperative events were the most important factors leading to postoperative NP impairment. Clinical neurologic outcome correlated positively with NP changes. Differences between various cardiologic and age groups were found in the postoperative NP findings. Changes in frequency in QEEG reflected the NP changes more reliably than did amplitude. Significantly one of the tests, a modification of the Stroop color test, was found to be prognostically important, suggesting that it might be possible to construct test batteries of value in predicting high operative risk. The results show that multiparameter investigation using NP testing together with clinical and QEEG follow-up is useful in the assessment of cerebral disorders attributable to open-heart surgery.