IGF-I is a marker of activity of the growth axis. It is accepted that the interpretation of IGF-I values requires age- and gender-specific reference ranges (RR).
We have studied IGF-I assay performance by circulating 2 blood samples with different clinical scenarios to 23 centers [IDS-OCTEIA (n=2), DPC-Immulite (3), DPC-Immulite 2000 (4), in-house RIA (1), Nichols-Advantage (10) and Nichols-IRMA (3)] participating in the UK-NEQAS IGF-I scheme. Participants were asked to a) measure IGF-I, b) interpret the result against their RR, and c) provide information on the source of their RR.
Scenario 1: 6 year old boy with short stature. All method median IGF-I was 4.0nmol/L (range 3.0-7.3nmol/L, CV=23.0%). Nine laboratories reported the IGF-I as within their RR. Scenario 2: 38 year old woman with increased sweating, diabetes and a diagnosis of possible acromegaly. All method median IGF-I was 50.8nmol/L (24.3-60.9nmol/L, CV=16.3%). Seven laboratories interpreted their value as being within the RR. Median IGF-I was 15% (-46 to +76%) above the upper limit of their RR. All participants indicated they used age-related RR (15/23 manufacturer-derived, 5/23 in combination with their own ranges), however, there was almost no concordance in the ranges quoted by labs using the IDS-OCTEIA, Nichols-IRMA and Nichols-Advantage assays, this despite the recent availability of well-developed ranges (Horm. Res. 2003:60;53) for the latter. 40% and 60% of laboratories, respectively, do not quote gender-related ranges for children and adults.
Variability in IGF-I results, coupled with use of inappropriate normative data, undermines IGF-I's contribution to clinical practice.
01 - 03 Nov 2004
Society for Endocrinology