Published works on intelligence quotient (IQ) and development following the use of profound hypothermia and circulatory arrest (TCA) to repair congenital heart defects in infants and young children suggest that little or no psychomotor impairment results. IQ scores derived from cognitive, memory, perceptual, quantitative, and verbal tests (McCarthy scale of the children's abilities, mean score 100, SD 16) were measured in 31 patients 5 years following operations performed with TCA between 1972 and 1976. These patients were compared with three control groups: (1) 19 patients with similar defects but operated upon using moderate hypothermia and continuous cardiopulmonary bypass (CPB); (2) 16 children who were the siblings of the TCA patients; and (3) 14 children who were the siblings of the CPB patients. The hypothermic temperatures reached were closely clustered around 15 degrees C in the TCA group and 28 degrees C in the CPB group. TCA time ranged from 22 to 71 minutes. Statistical analysis, which included, t test, chi square test of association, and Wilcoxon test, showed that the only baseline characteristic which differed between the two patient groups in respect to age at operation, age at testing, and preoperative physiological variables (level of cyanosis, weight, oxygen saturation, and hemoglobin concentration) was weight (p = 0.03). The mean score of the TCA group (91 +/- 4.0, SE) was significantly lower (p = 0.002) than that of their siblings (106 +/- 4.1, SE). The score for the CPB patients (102 +/- 5.2, SE) was not demonstrably different from that of their siblings (96 +/- 5.9, SE). The sibling and patient (TCA) IQ differences were associated with duration of arrest in verbal (p = 0.06), quantitative (p = 0.07), and general cognitive (p = 0.003) scores. A decrease of 0.53 point per minute of arrest time was estimated for the entire group of 31 patients; that is, in the 19 patients with siblings, for each minute increase in circulatory arrest time, the patients dropped 0.69 IQ point below their siblings. These results and analysis of other published data do not support the generally accepted view that TCA can be used entirely without penalty. We question the accepted "safe" limit of circulatory arrest of 60 minutes.