Relation of pH strategy and developmental outcome after hypothermic circulatory arrest

To examine whether pH management during core cooling is a risk factor for adverse developmental outcome, we studied 16 children with transposition of the great arteries and intact ventricular septum who underwent a Senning procedure in infancy (median age 32 days, range 2 to 154 days) between 1983 and 1988. Information was collected retrospectively on many aspects of perfusion, including lowest carbon dioxide tension during core cooling, duration of core cooling, and duration of circulatory arrest. The pH strategy changed from pH-stat to alpha-stat in 1985, resulting in a wide range of pH values and carbon dioxide tension (34 to 76 mm Hg) during the study period. All children had rapid core cooling to a rectal temperature of 19.8 degrees +/- 2.7 degrees C (mean +/- standard deviation) and a tympanic temperature of 16.6 degrees +/- 3.0 degrees C. Development was assessed at median age 48.0 (11 to 79) months with the Bayley Scales (n = 4, children younger than 30 months) or the McCarthy Scales (n = 12, children older than 30 months). The mean core-cooling duration was 14.5 +/- 6.2 minutes, circulatory arrest time was 43.4 +/- 6.6 minutes, and total bypass plus circulatory arrest time was 89.7 +/- 12.7 minutes. Lower carbon dioxide tension (alpha-stat) before onset of circulatory arrest was associated with worse developmental outcome (r = 0.71, p = 0.002). This relationship remained highly significant when we controlled for sociodemographic and intraoperative variables. including core-cooling time, circulatory arrest time, and total elapsed time. Duration of circulatory arrest was not associated with developmental outcome. We conclude that when relatively rapid core cooling is used to achieve hypothermia before circulatory arrest in young infants, a more alkaline pH strategy such as alpha-stat may result in less effective cerebral protection.