In normal and obese young males [90--120% and > 160% of ideal body weight (IBW); IBW = 100%], plasma concentrations of testosterone, androstenedione, estrone, and estradiol were measured. Metabolic clearance and production rates of androstenedione and the conversion ratios of androstenedione to testosterone, estrone, and estradiol were determined using the constant infusion technique. In the obese subjects, IBW was inversely correlated (P < 0.001) with plasma concentrations of androstenedione (r = 0.81) and testosterone (r = 0.87), while the levels of estrone (r = 0.92) and estradiol (r = 0.95) increased with IBW (P < 0.001). Thus, when normal and obese subjects were compared as groups, plasma androstenedione decreased form 1.24 +/- 0.13 to 0.93 +/- 0.15 ng/ml (mean +/- SD) and plasma testosterone decreased from 5.89 +/- 0.82 to 3.29 +/- 0.92 ng/ml (P < 0.001), while estrone increased from 28.2 +/- 3.4 to 60.0 +/- 9.4 pg/ml, and estradiol increased from 21.7 +/- 3.5 to 43.9 +/- 5.3 pg/ml. The testosterone to androstenedione and the estradiol to estrone ratios were not different in obesity, but changes in IBW were positively correlated (P < 0.001) with differences in the estrone to androstenedione (r = 0.93) and estradiol to testosterone ratios (r = 0.93), indicating that fat tissue may aromatize androgens, whereas reduction of 17-oxo-steroid appears to be of minor importance. As the MCR of androstenedione increased with IBW (from 2156 to 2636 liters/day P < 0.05) while plasma levels decreased, the apparent production rate of androstenedione was not influenced by the degree of obesity. The conversion of androstenedione to estrone (r = 0.89) and of androstenedione to estradiol (r = 0.82) was enhanced in obese subjects (P < 0.001). We suggest that enhanced aromatization of androstenedione due to an increased adipose tissue mass may account for the high plasma estrogen levels observed in obese men.

In obese male subjects with 160 - 200% of ideal body weight (IBW = 100%) the decrease in total plasma testosterone is biologically ineffective since SHBG is concomitantly decreased from 30.0 +/- 3.6 to 20.0 +/- 3.4 nM/l. Conversely, in massively obese males with greater than 250% of IBW, the decrease in SHBG (to 10.6 +/- 1.8 nM/l) is too small to compensate for total testosterone decrease (from 6.04 +/- 0.57 to 1.72 +/- 0.32 ng/Ml). Therefore, free testosterone is markedly less in the massively obese patients (55 +/- 8 vs. 127 +/- 15 pg/ml in the controls). Despite this significant difference in free testosterone concentrations (p less than 0.01), plasma LH is even lower in the obese (6.8 +/- 0.8 mU/ml) than in the controls (10.0 +/- 1.0 mU/ml). This may be an effect of free estradiol, which rises from 0.48 to 1.52 pg/ml in the massively obese subjects. These alterations are clearly demonstrated by the imbalance of the estradiol/testosterone ratios, which increase 10-fold and 7-fold for the total and for the free sex hormones, respectively. We conclude that the decrease in SHBG, which prevents obese males from developing hypogonadism, is not sufficiently effective in the massively obese patients to compensate the marked decrease in testosterone. This, in connection with the observed increase of free estradiol, may cause hypogonadism and hyperestrogenism in these subjects.