Serum versus Heparinized Plasma for Eighteen Common Chemistry Tests: Is Serum the Appropriate Specimen?

Abstract

The results of 18 chemistry tests performed on serum and heparinized plasma from 206 unselected patients admitted to Barnes Hospital are reported. In 138 cases, platelet counts, leukocyte counts, and erythrocyte counts were also determined. For 11 of the 18 tests, statistically significant differences between serum and plasma values were observed. The differences for calcium, glucose, inorganic phosphorus, potassium, and total protein were felt to be of sufficient magnitude to alter clinical interpretation in certain instances. The magnitude of differences between serum and plasma potassium correlated positively with the platelet counts and leukocyte counts but could not be predicted by these values. A relatively high correlation (r = .459) was found for differences between serum and plasma calcium and the differences between serum and plasma total protein. The use of combinations of heparin and fluoride or heparin and iodoacetate as anticoagulant-antiglycolytic agents caused inaccurate potassium results. Iodoacetate appears to be more effective than fluoride in stabilizing values for reducing sugars. The data suggest that plasma specimens are preferable to serum specimens for routine clinical chemistry analysis and are essential for accurate potassium measurement.