| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Letter to the Editor |
Department of Medicine, Columbia University College of Physicians and Surgeons, St. Luke’s-Roosevelt Hospital Center, New York, New York 10019
Address correspondence to: Dr. Gerald B. Phillips, M.D., St. Luke’s-Roosevelt Hospital Center, 1000 Tenth Avenue, New York, New York 10019. E-mail: gbpl{at}columbia.edu.
To the editor:
In the study of Muller et al. (1), the testosterone level was negatively associated with the insulin level and had a significant negative trend with the number of components of the "metabolic syndrome" (2), whereas the estradiol level had a significant positive trend with the number of components of the "metabolic syndrome." In addition, the testosterone level showed a significant negative association with waist circumference and triglyceride, and the estradiol level showed a significant positive association. These relationships suggest that the ratio of estradiol-to-testosterone (E/T) could have related more strongly to these variables than either hormone alone.
In 1977, glucose and insulin were observed to correlate with sex hormones (3). In that study in nonobese, nondiabetic men, the glucose area, insulin area, and insulin area/glucose area (insulin resistance) correlated negatively with testosterone but more strongly and positively with E/T even though estradiol did not correlate significantly with the glucose or insulin areas. E/T also correlated with the cholesterol and ß-lipoprotein levels (3). In the four subsequent studies from this laboratory that tested the sex hormone-insulin relationship, controlled for age and BMI, all found that E/T correlated with fasting insulin more strongly than did testosterone, even though estradiol did not correlate with insulin in any of the studies (4). In two of these studies, neither estradiol nor testosterone correlated significantly with insulin, whereas E/T correlated strongly (P < 0.001) (4). In our most recent study, in which visceral adipose tissue (VAT) was measured, the E/T-insulin, but not the testosterone-insulin, correlation survived control for VAT; and in a multiple regression analysis, estradiol, as well as testosterone and VAT, were associated with insulin (4). Consistent with an estradiol-insulin relationship was the finding of an elevated estradiol level (P < 0.01) and E/T (P < 0.005) in nonobese men with noninsulin-dependent diabetes on no drugs (5). Additional evidence of a relationship of estradiol to insulin sensitivity and to components of the "metabolic syndrome" is apparent in studies of men and their knockout mouse models with deficient estrogen action resulting from mutations of either of two genes (6).
Most studies on the relationship of testosterone to insulin and/or risk factors for myocardial infarction (MI) in men have not measured estradiol, possibly owing to the difficulty in measuring reliably the minuscule levels found in men (7).
However, estradiol may be importantly related to insulin and risk factors for MI. The evidence that estradiol and testosterone appear to relate in opposite directions to insulin and to major risk factors for MI [though not, for example, to HDL (4)] in men suggests that E/T may define certain of these sex hormone relationships better than either hormone alone. I believe that further investigation of E/T, as well as estradiol, in these relationships is indicated.
Received June 6, 2005.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
