Influence of various modes of androgen substitution on serum lipids and lipoproteins in hypogonadal men

We investigated whether the androgen type or application mode or testostero ne (T) serum levels influence serum lipids and lipoprotein levels different ially in 55 hypogonadal men randomly assigned to the following treatment gr oups: mesterolone 100 mg orally daily ([MES] n = 12), testosterone undecano ate 160 mg orally daily ([TU] n = 13), testosterone enanthate 250 mg intram uscularly every 21 days ([TE] n = 15), or a single subcutaneous implantatio n of crystalline T 1,200 mg ([TPEL] n = 15). The dosages were based on stan dard treatment regimens. Previous androgen substitution was suspended for a t least 3 months. Only metabolically healthy men with serum T less than 3.6 nmol/L and total cholesterol (TC) and triglyceride (TG) less than 200 mg/d L were included. After a screening period of 2 weeks, the study medication was taken from days 0 to 189, with follow-up visits on days 246 and 300. Be fore substitution, all men were clearly hypogonadal, with mean serum T less than 3 nmol/L in all groups. Androgen substitution led to no significant i ncrease of serum T in the MES group, subnormal T in the TU group (5.7 +/- 0 .3 nmol/L), normal T in the TE group (13.5 +/- 0.7 nmol/L), and high-normal T in the TPEL group (23.2 +/- 1.1 nmol/L). 5 alpha-Dihydrotestosterone sig nificantly increased in all treatment groups compared with baseline. Compar ed with presubstitution levels, a significant increase of TC was observed i n all treatment groups (TU, 14.4% +/- 3.0%; MES, 18.8% +/- 2.5%; TE, 20.4% +/- 3.0%; TPEL, 20.2% +/- 2.6%). Low-density lipoprotein cholesterol (LDL-C ) also increased significantly by 34.3% +/- 5.5% (TU), 46.4% +/- 4.1% (MES) , 65.2% +/- 5.7% (TE), and 47.5% +/- 4.3% (TPEL). High-density lipoprotein cholesterol (HDL-C) showed a significant decrease by -30.9% +/- 2.8% (TU), -34.9% +/- 2.5% (MES), -35.7% +/- 2.6% (TE), and -32.5% +/- 3.5% (TPEL). Se rum TG significantly increased by 37.3% 11.3% (TU), 46.4% +/- 10.3% (MES), 29.4% +/- 6.5% (TE), and 22.9% +/- 6.7% (TPEL). TU caused a smaller increas e of TC than TE and TPEL, whereas the parenteral treatment modes showed a l ower increase of TG. There was no correlation between serum T and lipid con centrations. Despite the return of serum T to pretreatment levels, serum li pid and lipoprotein levels did not return to baseline during follow-up eval uation. In summary, androgen substitution in hypogonadal men increases TC, LDL-C, and TG and decreases HDL-C independently of the androgen type and ap plication made and the serum androgen levels achieved. Due to the extended washout period for previous androgen medication and the exclusion of men wi th preexisting hyperlipidemia, this investigation demonstrates more clearly than previous studies the impact of androgen effects on serum lipids and l ipoproteins. It is concluded that preexisting low serum androgens induce a "male-type" serum lipid profile, and increasing serum androgens further wit hin the male normal range does not exert any additional effects. The thresh old appears to be above the normal female androgen serum levels and far bel ow the lower limit of normal serum T levels in adult men. These findings ma y have considerable implications for the use of androgens as a male contrac eptive and for androgen therapy in elderly men. Copyright (C) 1999 by W.B. Saunders Company.