Defining acidosis in postoperative cardiac patients using Stewart's method of strong ion difference

Objective: To define the true incidence and nature of acidosis in pediatric patients postcardiac surgery, using Stewart's direct method of measuring strong ion difference. We also wished to compare the ability of standard indirect methods (base deficit, lactate, anion gap, and corrected anion gap) to accurately predict tissue acidosis. Design: A single-center prospective observational study. Setting: A pediatric intensive care unit in a tertiary referral center. Patients: Pediatric patients who had undergone cardiac surgery were studied in the immediate postoperative period. Patients who had undergone both open and closed cardiac surgery were included. Interventions: Routine arterial blood gas analysis and laboratory electrolyte measurements were made in patients immediately on admission to the pediatric intensive care unit (PICU) after cardiac surgery and each morning until discharge from the PICU. Measurements and Main Results: Figge's equations were used to calculate strong ion difference and total tissue acids (unmeasured acids and lactate). These direct methods then were compared to indirect measurements: base deficit, lactate anion gap, and anion gap corrected for albumin. We collected 150 samples from 44 patients. Tissue acidosis occurred overall in 60 of 150 samples. This was due to raised unmeasured acids alone in 44 of 60 (73.3%), raised lactate alone in six of 60 (10%), and a combination of the two in ten of 60 (16.6%). Hyperchloremia occurred in 19 of 150 samples overall and 12 of 25 (48%) samples immediately after cardiopulmonary bypass. Measured base deficit showed a poor correlation with true tissue acidosis (r =-.48, p < .001) and the worst discriminatory ability (area under the curve, 0.72; 0.62-0.82). Anion gap corrected for albumin had the best correlation (r = .95, p < .001) and highest area under the curve (0.90; 0.85-0.95). Conclusions: Metabolic acidosis occurs frequently postcardiac surgery and is largely due to raised unmeasured acids and less commonly raised lactate. Hyperchloremia is common, particularly after cardiopulmonary bypass. Base deficit correlates poorly with true tissue acidosis, and corrected anion gap offers the most accurate bedside alternative to Stewart's method of tissue acid calculation.