Harmonization of growth hormone measurements with different immunoassays by data adjustment.
Müller A, Scholz M, Blankenstein O, Binder G, Pfäffle R, Körner A, Kiess W, Heider A, Bidlingmaier M, Thiery J, Kratzsch J
Purpose
The purpose of this paper was to determine the effects of using different commercially available immunoassays for the measurement of human growth hormone (hGH) in children and adolescents.
Conclusion
When 8 different immunoassays were used to measure hGH concentrations in the same 312 specimens, large differences were observed between assays. Average differences in measurements of the same specimen by different assays ranged between 0.36 ng/mL (DxI versus ELISA) and 2.71 ng/mL (IMMULITE versus BC-IRMA) but individual differences were as high as 11.4 ng/mL. With an hGH concentration cut-off value of 6 ng/mL for the diagnosis of growth hormone deficiency (GHD), 29% of the patients in this study would have a diagnosis dependent on the immunoassay used.
Studies (1)
Purpose
The purpose of this study was to determine the effects of using different commercially available immunoassays for the measurement of hGH in children and adolescents. Both individual specimens and pooled specimens were analyzed. All specimens were kept at <-25 degrees C prior to analysis.
Summary of Findings
When 8 different immunoassays were used to measure hGH concentrations in the same 312 specimens, large differences were observed between assays. The assay with the highest hGH measurements was the IMMUNLITE 2000 (median: 5.44 ng/mL) and the assay giving the lowest hGH measurements was the BC-IRMA (median: 2.67 ng/mL), while the overall median for all 8 immunoassays was 4.10 ng/mL. Average differences in measurements of the same specimen by different assays ranged from 0.36 ng/mL (DxI versus ELISA) to 2.71 ng/mL (IMMULITE versus BC-IRMA) but individual differences were as high as 11.4 ng/mL. Correlation coefficients between pairs of assays ranged from 0.864 (iSYS and in house radioimmunoassay) to 0.992 (AutoDELFIA and Liaison). With an hGH concentration cut-off value of 6 ng/mL for the diagnosis of GHD, 29% of the patients in this study would have a diagnosis dependent on the immunoassay used. However, adjustment of hGH measurements using linear regression or conversion factors would reduce the amount of immunoassay dependent diagnoses to less than 8%. Further, quantile normalization would reduce immunoassay dependent diagnoses to less than 1%.
Biospecimens
FluidSerumPreservation Types
FrozenDiagnoses
Other diagnosesPlatforms
ProteinImmunoassayProteinRadioimmunoassayProteinELISAPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Preaquisition Diagnosis/ patient condition Suspected growth hormone deficiencyImmunoassay Specific Technology platform AutoDELFIABCR-IRMAELISAIMMULITE 2000In house radioimmunoassayLiaisonUniCel DxI 800 AccessiSYS