Metabolic alkalosis in children undergoing cardiac surgery

To define the frequency of metabolic alkalosis and its pathogenesis in children after open-heart surgery.

Retrospective chart review.

Multidisciplinary, tertiary, pediatric intensive care unit.

Fifty-six consecutive children undergoing open-heart surgery.

Metabolic alkalosis occurred in 29 (52%) of 56 patients. Seventy-two percent of patients < 12 months of age developed metabolic alkalosis as compared with 30% of patients > 12 months of age (p < .01 by chi-square). Patients developing metabolic alkalosis were younger, received more furosemide, had lower serum chloride concentrations, and underwent longer cardiopulmonary bypass times than nonmetabolic alkalosis patients. By stepwise multiple linear regression analysis, only age (p < .05) and serum chloride concentrations (p < .001) had independent correlations with the development of metabolic alkalosis; both variables had inverse correlations with arterial pH (r2 = .42). Patients with metabolic alkalosis also developed significantly (p < .01 by two tailed Student's t-test) lower serum ionized calcium concentrations (4.2 +/- 0.5 mg/dL [1.05 mmol/L]) as compared with nonmetabolic alkalosis patients (4.6 +/- 0.4 mg/dL [1.15 mmol/L]).

Postoperative metabolic alkalosis occurs frequently in children undergoing open-heart surgery. Chloride depletion seems to be the predominant factor in the pathogenesis of metabolic alkalosis. Younger age can serve as a positive predictor for the development of metabolic alkalosis in this subset of patients.