Objective: Reported associations between endogenous steroid hormone levels and cardiovascular disease in the older population have been contradictory. We evaluated plasma steroid concentrations in terms of the dimensions of the common carotid artery wall layers as a measure of the extent of atherosclerosis.
Methods: A subgroup of 70-year-old participants (32 women and 50 men) from the Prospective Investigation of the Vasculature in Uppsala Seniors study was investigated. All participants had assessments of common carotid artery wall layer parameters (intima thickness, media thickness, and intima-media thickness [IMT] ratio; measured by high-frequency ultrasound at 22 MHz) and endogenous steroid hormone concentrations (measured by liquid chromatography–tandem mass spectrometry).
Results: Low androgen levels, high aromatase enzyme activity (estrone [E1]/androstenedione and estradiol [E2]/testosterone), high E2/E1 ratio, and high estrogen levels (E1, E2, estriol, and E2/sex hormone–binding globulin) were consistently associated (often significantly) with a more unhealthy artery wall (thick intima, thin media, and high IMT ratio) in both sexes. Consistently strong associations were found between the aromatase index E2/testosterone and intima, media, and the IMT ratio. For IMT ratio, in both men (rs = 0.52) and women (rs = 0.58), P was <0.001 for both and remained significant after adjustment for cardiovascular disease risk factors and the Framingham risk score (both P < 0.01).
Conclusions: Low androgens, high aromatase enzyme activity, and high estrogen levels are often significantly associated with an unhealthy artery wall on ultrasound. We suggest that the steroid hormone profile of older individuals with higher estrogens most probably reflects an endogenous response to developing atherosclerosis, rather than a cause-and-effect relationship. However, the reverse causality cannot be excluded.
From the 1Department of Women’s and Children’s Health, Section for Obstetrics and Gynecology, 2Analytical Chemistry/Department of Physical and Analytical Chemistry, and 3Department of Medical Sciences, Internal Medicine, Uppsala University, Uppsala, Sweden; and 4Associated Regional and University Pathologists, Inc. (ARUP) Institute for Clinical and Experimental Pathology, Salt Lake City, UT.
Received January 30, 2012; revised and accepted April 5, 2012.
Funding/support: Support was provided by the following: The Selanders foundation, the Swedish Heart and Lung Association, the Thuréus foundation, the Goljes foundation, the Gamla Tjänarinnor foundation, the Family Planning Fund (Uppsala, Sweden), the Janne Elgquist foundation, the Swedish Research Council (621-2008-3562 to J.B.), and the ARUP Institute for Clinical and Experimental Pathology (Salt Lake City, UT).
Financial disclosures/conflict of interest: The liquid chromatography–tandem mass spectrometry analyses were provided by the ARUP Institute for Clinical and Experimental Pathology, Salt Lake City, UT. T.N. is a member of Novo Nordisk scientific board.
Address correspondence to: Tord Naessen, MD, PhD, Department of Women’s and Children’s Health, Section for Obstetrics and Gynecology, Uppsala University, Uppsala, SE-751 85 Sweden. E-mail: Tord.Naesssen@kbh.uu.se