Growth hormone assays and dynamic tests in the diagnosis of acromegaly and GH deficiency in adults

The measurement of serum growth hormone (GH) is the cornerstone of the diagnosis and management of GH-related disorders, acromegaly, and GH deficiency (GHD). GH secretion from the anterior pituitary gland is pulsatile and influenced by several factors (e.g. exercise, stress, glucose). Therefore, the GH result of a single blood sample is not diagnostic. Thus, the diagnosis is based on the suppression and stimulation tests of pituitary GH secretion. GH measurement is challenging due to the heterogenous structure and binding proteins of GH. The first assays for GH were based on radioimmunoassay (RIA) using polyclonal antibodies. The old International Standard (IS) reference preparations for GH assay calibration were of pituitary origin and contained a pooled cadaver-derived mixture of GH isoforms. During the 1990s, a significant methodological change in serum GH assays occurred, when RIAs were gradually replaced by more specific sandwich assays with monoclonal antibodies and highly sensitive labels. The use of monoclonal antibodies and IS preparations containing different GH isoforms caused significant between-assay and between-laboratory variation of GH levels. During the 1990s, the standard of recombinant 22-kDa GH became available. At the beginning of the 2000s, the WHO IS 98/574 (22 kDa) was established. During the last 10 years, GH immunoassay manufacturers have adopted this new standard for calibration of GH assays. Together with monoclonal antibodies, this has resulted in a significant reduction of GH concentrations obtained by recent GH assays. The aim of this study was to evaluate the interlaboratory variation of different GH assays used in Finland between 1998 to 2003 on the basis of an external quality assessment round of GH. Moreover, we wanted to compare the clinical use of two automated GH assays, AutoDELFIA and Immulite 2000 assays (calibrated against the same standard), with old polyclonal RIA with older standardization. The utility of consensus cut-offs of the oral glucose tolerance test (OGTT) and insulin tolerance test (ITT) in clinical practice was also studied by using those two automated assays. At the beginning of this study, the clinical use of a new GHRH+ARG test was assessed in control subjects and GHD patients by using the AutoDELFIA GH assay. Later, the cut-offs for the GHRH+ARG test were evaluated by using the Immulite 2000 XPi GH assay, which is calibrated against the current WHO IS 98/574. Our results showed that the international cut-off values for GH in consensus statements and guidelines cannot be directly adopted in diagnostic use without a clinical validation of the particular GH assay in use. We also showed that male control subjects in the OGTT and GHRH+ARG test had significantly lower GH values than women, suggesting that gender-specific cut-offs are needed. In our study, the ITT was an unreliable test for GHD. Almost half of apparently healthy adults have peak GH values