Objectives: To prospectively document the occurrence of ionized hypocalcemia in infants and children treated with extracorporeal membrane oxygenation (ECMO), to determine if the type of calcium salt (calcium chloride or gluconate) used in priming the ECMO circuit affected ionized calcium, to determine if ionized calcium concentrations correlate with total calcium, protein, albumin, or total magnesium values, and to determine if the hypotension usually observed after ECMO initiation correlates with low circulating ionized calcium concentrations.
Design: Prospective study.
Setting: Pediatric ICU and neonatal ICU.
Patients: Sixteen neonatal and three pediatric patients who were started on ECMO for cardiopulmonary support.
Interventions: The ECMO circuit was primed in a standardized manner, 100 mg of calcium gluconate was added in group 1 patients and 100 mg of calcium chloride was added in group 2 patients.
Measurements: Ionized calcium was measured from the circuit before initiation of ECMO and from the patient before, and then 5, 10, 15, 30, 60, 120, and 240 mins after initiation of ECMO. Total calcium and ionized calcium concentrations were measured simultaneously every 6 hrs. Serum total protein, albumin, magnesium, and ionized calcium values were measured from blood samples collected simultaneously twice daily.
Results: A significant decrease in the mean serum ionized calcium value occurred 5 mins after the initiation of ECMO in both groups, p less than .001. The ionized calcium value remained significantly decreased until 30 mins after the initiation of ECMO. There were no differences between the ionized calcium concentrations obtained during priming with calcium gluconate vs. those concentrations obtained with calcium chloride priming (p = .79). Throughout the course of ECMO, the serum ionized calcium concentrations ranged from 0.60 to 1.86 mmol/L. Poor correlations existed between circulating ionized calcium values and total calcium (r2 = .30), total protein (r2 = .20), albumin (r2 = .20), and magnesium concentrations (r2 = .10). There was a good correlation between the patients' BP and ionized calcium concentrations after bypass was initiated (r2 = .87).
Conclusion: Our data demonstrate that ionized hypocalcemia is a frequent occurrence after the initiation of ECMO. Since there is a poor correlation between ionized calcium and total calcium, ionized calcium concentrations should be measured directly in these patients.