Assessment of calcium homeostasis in the critically ill surgical patient. The diagnostic pitfalls of the McLean-Hastings nomogram.

Hypocalcemia is a common problem in critically ill surgical patients. We prospectively evaluated whether measurement of the total serum calcium (Ca) concentration or calculation of the serum ionized Ca level (by the McLean-Hastings nomogram) accurately reflects the measured serum ionized Ca level. Although 71% and 58% of 156 predominantly surgical intensive care unit (ICU) patients were hypocalcemic by the total serum Ca or calculated ionized Ca level, respectively, only 12% were hypocalcemic by directly measured serum ionized Ca measurement. The total serum Ca and calculated ionized Ca concentrations were sensitive (95% and 89%, respectively) but lacked specificity (32% and 46%, respectively) in predicting ionized hypocalcemia. Analyses of Ca binding to albumin in the serum of surgical ICU patients and normal subjects suggested that there is a circulating factor in critically ill patients that increases the binding of Ca to albumin. These observations may explain why the McLean-Hastings nomogram underestimates the protein-induced changes in serum Ca in critically ill surgical subjects. We conclude that: total serum Ca and calculated ionized Ca concentrations are poor indicators of the true serum ionized Ca status in critically ill surgical patients, and we recommend direct measurement of serum ionized Ca levels in these patients; and variability in the affinity of Ca for binding proteins in critical illness may explain the poor correlation between serum total and ionized Ca measurements.