DEFINING GROWTH-HORMONE DEFICIENCY IN ADULTS

The absence of a distinct clinical syndrome calls for a strategy to re liably identify patients with hyposomatotropism. However, there is no consensus as to the most appropriate method of defining growth hormone (GH) deficiency in adults. Since GH secretion falls with senescence a nd is also reduced by obesity, both of these factors must be controlle d for in such an evaluation. We have investigated the relative diagnos tic merits of measuring (1) peak GH response to insulin-induced hypogl ycemia (ITT), (2) mean 24-hour GH concentration derived from 20-minute sampling (IGHC), (3) serum IGF-I levels, and (4) serum insulin-like g rowth factor (IGF)-binding protein-3 (IGFBP-3) levels. These tests wer e undertaken in 23 patients considered GH-deficient from extensive org anic pituitary disease and in 35 sex-matched normal subjects of simila r age and body mass index. The ITT was the only test capable of distin guishing patients with organic GH deficiency from matched normal subje cts. The sensitivity of the GH radioimmunoassay (0.2 ng/mL) limited th e utility of IGHC measurements, since many subjects from both groups h ad undetectable values. Using a GH assay with a 100-fold greater sensi tivity, we found a better but still incomplete separation of values be tween the two groups. There was a significant overlap of IGF-I and IGF BP-3 values, with only a third of GH-deficient subjects having low IGF -I values. The limitation of IGF-I has been confirmed by others, altho ugh its sensitivity as a diagnostic test is greater in young adults. W e conclude that organic GH deficiency in adults can be reliably diagno sed by the ITT. IGF-I and IGFBP-3 measurements are unreliable, because these levels may be normal in GH deficiency. Copyright (C) 1995 by W. B. Saunders Company.