Abstract.We previously demonstrated that individual subjects have fairly constant ratios of serum concentrations of 20 kDa-(20K) and 22 kDa-GH (22K). The aim of this study is to demonstrate the possibility of utilizing the changes in the ratio of 20K/22K for detecting the exogenous administration of 22K. A male patient with idiopathic dilated cardiomyopathy (age 51) received 22K (4 U, s.c.) every other day. The concentrations of 20K and 22K in serum and urine were measured using enzyme-linked immunosorbent assays before and after administration.The administration of 22K increased total GH concentration, and markedly decreased the ratio of 20K/22K in serum, especially 2-10 h after the administration.From calculations, it became clear that the concentration of exogenous 22K reached a peak between 2-4 h after the administration and decreased to a negligible level after 24 h. The ratio of 20K/22K in the 0-24 h urine was 5 times lower than that in the 24-48 h urine. These data suggest that, by monitoring the ratio of 20K/22K in serum or urine, it is possible to determine whether or not GH has been externally administered and to calculate the serum GH that has been administered. GROWTH hormone (GH) therapy has been indicated for the treatment of GH deficient patients, and for some other indications and purposes [1, 2]. In addition, GH has been a subject to be tested for doping as one of prohibited drugs in sports [3, 4]. Two different kinds of GHs, 22 kDa-(22K) and 20 kDa-GH (20K), are known to be produced through alternative splicing of mRNA [5]. The 20K is a naturally occurring isoform lacking residues 32-46 of 22K [6] and comprises approximately 5-20% of circulating GH [7][8][9][10][11]. Recently, we have constructed two enzyme-linked immunosorbent assays (ELISAs), which specifically detect 20K or 22K in blood [9,10]. The level of circulating 20K was highly correlated to that of 22K in both normal subjects and patients, and the proportion of 20K in each individual subject was fairly constant even after pharmacological and physiological stimuli. Currently, we conducted a pharmacokinetic study of recombinant 20K in human subjects [11], and found that the 24-h profile of serum 20K level in the placebo group is parallel to that of 22K. In contrast, administration of 20K resulted not only in remarkable decrease in the ratio of serum concentrations of 20K and 22K, but also apparent suppression of GH secretion. These results strongly suggest that, by monitoring the ratio of 20K/22K either in serum or urine, it may be possible to determine whether or not GH has been externally administered and to calculate the concentration of serum GH that has been administered. The present