Measurement of inaccuracy and imprecision of HCG methods using dilutions of the WHO 4th IS-HCG standard and a pregnant patient's serum.

INTRODUCTION: Differences in human chorionic gonadotropin (hCG) results obtained by seven different methods were documented by analyzing dilutions of the WHO 4th International Standard (IS) and a pregnant patient's serum. MATERIALS AND METHODS: Biases of +30.9 to -37.5% and +36.8 to -36.1% from the target concentration were found for the WHO 4th IS and patient sample dilutions, respectively. RESULTS: Imprecision was calculated from replicate measurements of hCG on the different sample dilutions. Imprecision ranged from 1.0% to 18.9% and 1.1% to 5.3% for the WHO 4th IS and patient sample dilutions, respectively. Using a maximum allowable error of 12.5% for hCG measurements, we found that two instruments were so biased that their hCG measurements could not be interchanged with hCG values produced by any of the other systems. DISCUSSION: It is ideal to use only one hCG methodology for the serial monitoring of hCG; otherwise, hCG methods should be carefully chosen to minimize inter-method bias.

Unique approach to derivation of random error in laboratory assays: application to glycohemoglobin testing demonstrates poor clinical performance for immunochemistry assay.

The measurement of glycohemoglobin is the best measure of mean glucose within a 3-4 month range. As it is used for patient education, counseling, feedback control, and ultimately for patient motivation, its measurement should be optimally accurate and precise. Duplicate hemoglobin A1c readings were used to determine physiological (changes over time between measurements) and analytic variation of two widely used laboratory assays: Bio-Rad Variant II's high-performance liquid chromatography (HPLC) system and Roche's immunoassay. The average variation of grouped duplicates was calculated and graphed against corresponding time intervals. Regression to the y-intercept (0 day separation between readings) was used to determine the analytic variation. Analytic coefficients of variation (CVs) for the HPLC and immunoassay were determined as 2.6% and 5.1%, respectively. The CV of the immunoassay method exceeds physiologically established limits of 2-3% and those of the National Glycohemoglobin Standardization Program (3-4%). The Bio-Rad HPLC system produces a CV within these limits.