A. Tchernof et al., RELATION OF STEROID-HORMONES TO GLUCOSE-TOLERANCE AND PLASMA-INSULIN LEVELS IN MEN - IMPORTANCE OF VISCERAL ADIPOSE-TISSUE, Diabetes care, 18(3), 1995, pp. 292-299
Citations number
46
Categorie Soggetti
Endocrynology & Metabolism","Medicine, General & Internal
OBJECTIVE - Low plasma testosterone levels are associated with hyperin
sulinemia and glucose intolerance in men. However, it is unclear wheth
er these abnormalities are related to the concomitant alteration in re
gional adipose tissue (AT) accumulation associated with reduced androg
en levels. RESEARCH DESIGN AND METHODS - we measured plasma steroid le
vels in a sample of 79 men, ranging from lean to obese (aged 29-42 yea
rs), for whom an oral glucose tolerance rest (OGTT), anthropometric an
d computed tomography (CT) measurements of body fatness, and AT distri
bution were performed. Sex hormone binding globulin (SHBG) and the fol
lowing steroids were measured after extraction from plasma and chromat
ography: dehydroepiandrosterone, androstenedione, androst-5-ene-3 beta
,17 beta-diol, testosterone, estrone, and estradiol (E(2)). RESULTS -
Several significant negative correlations were found between adrenal C
-19 steroid precursors, testosterone, SHBG, and fasting insulin levels
, as well as between plasma glucose and insulin concentrations measure
d during the OGTT (-0.25 less than or equal to r less than or equal to
-0.35, 0.05 greater than or equal to P greater than or equal to 0.001
). The best steroid correlate of plasma glucose and insulin homeostasi
s indexes was the E(2):testosterone ratio (0.34 less than or equal to
r less than or equal to 0.42, 0.005 greater than or equal to P greater
than or equal to 0.001). However, after correction of steroid levels
for either fat mass, body mass index (BMI), or visceral AT area, as me
asured by CT, no significant residual associations were noted between
testosterone, adrenal C-19 steroid, SHBG, and estrogen levels and inde
xes of plasma glucose-insulin homeostasis, although the positive assoc
iation between the E(2):testosterone ratio and glucose area remained s
ignificant after adjustment for total body fat mass and BMI. Furthermo
re, 15 pairs of obese subjects, matched for visceral AT area, showing
either low or high levels of the steroids studied, did not differ in f
asting insulin and postglucose plasma insulin levels or in glucose tol
erance. CONCLUSIONS - These results suggest that the previously report
ed relationships between androgen levels and indexes of plasma glucose
-insulin homeostasis are mediated, to a large extent, by concomitant a
lterations in levels of total body fat and visceral AT in men.