Determination of glucose in serum and whole blood: statistical relationships between values obtained by different methods.

Several different methods are available for measuring blood glucose in the clinical laboratory on physician’s office. As a result, confusion often arises in the clinical interpretation of glucose values that have been derived by different methocis. There are, of course, many variables to be considered when comparing methods, but the end result is that each method for glucose in whole blood or serum has a somewhat different normal range. Unless clearly understood, this difference may lead to misinterpretation of reported values. Equations for interconverting results obtained by various methods would help alleviate this problem. Our purpose is to present equations for estimating or predicting relationships between glucose values for whole blood or serum, as obtained by different methods. With their use, the sometimes significant difference between reported values obtained by two different methods may be interpreted in their proper perspective. However, clinicians should recognize that many variablessuch as specimen collection, handling, and testing-