Increased Cortisol Level: A Possible Link Between Climacteric Symptoms and Cardiovascular Risk Factors

Angelo Cagnacci, MD, PhD; Marianna Cannoletta, MD; Simona Caretto, MD; Renata Zanin, MD; Anjeza Xholli, MD; Annibale Volpe, MD


Menopause. 2011;18(3):273-278. 

In This Article


The study was conducted in accordance with the Declaration of Helsinki guidelines for human investigation. Between January and June 2009, women attending the outpatient service for menopause at our University Hospital in Modena were asked to participate in the study. Physicians of the service furnish information and indicate or prescribe measures for the prevention and therapy of perimenopausal to postmenopausal symptoms. Women meeting inclusion and exclusion criteria and willing to participate signed an informed consent form. Inclusion criteria were perimenopause or early postmenopause status (stage 1b or 2 as per Stages of Reproductive Aging Workshop definition).[32] Women in perimenopause with at least 6 months of amenorrhea and follicle-stimulating hormone level higher than 30 IU/L were included. Postmenopause was defined by at least 12 months of amenorrhea in women not receiving hormone therapy and follicle-stimulating hormone levels higher than 40 IU/L. Exclusion criteria were use of hormone therapy or of phytotherapies, neuroactive drugs, substances and medications possibly interfering with cortisol production or cortisol assay. Similarly, women with diseases possibly affecting the hypothalamic-pituitary-adrenal axis function were excluded. All women were white.

For each woman, we collected remote and obstetric-gynecological medical history. Climacteric, anxiety, and depressive symptoms were evaluated by validated Italian translations of self-administered questionnaires.

Vasomotor and psychological symptoms were evaluated by the Greene Climacteric Scale.[33] The Greene Climacteric Scale is composed of 21 items that evaluate vasomotor symptoms (2 items), anxiety (6 items), depression (5 items), somatic symptoms (7 items), and sexuality (1 item). All items have four options: "not at all" (0), "a little" (1), "quite a bit" (2), or "extremely" (3). The sum of the items' scores is used to obtain the Greene Climacteric Scale score as a whole (range, 0–63) or the scores of individual subscales for vasomotor symptoms (range, 0–6), anxiety (range, 0–18), depression (range, 0–15), somatic symptoms (range, 0–21), and sexuality (range, 0–3). Total score and subscale scores were used.

The Y-1 form of the State-Trait Anxiety Inventory (STAI) was used to further evaluate anxiety,[34] and the Zung scale (Self-Rating Depression Scale [SDS]) was used to further evaluate depression.[35] These scales are made up of 20 questions and four possible answers (score, 1–4), with scores ranging from 20 (best) to 80 (worst).

Weight (in kilograms) and height (in meters) were measured while women were wearing light clothes and no shoes. Body mass index (BMI) was calculated as the ratio of weight in kilograms to the square of height in meters. While the women were in standing position, their girths were measured along the horizontal plane at the level of the natural waist (narrowest abdominal circumference) and at the level of the hip (maximum extension of the buttocks). The waist-to-hip ratio (WHR) was calculated. Office blood pressure was measured after at least 10 minutes in the sitting position. Fasting serum values of triglycerides and high-density lipoprotein (HDL) and total cholesterol were evaluated. Low-density lipoprotein cholesterol levels were calculated by the formula of Friedewald: low-density lipoprotein = total cholesterol − (HDL − cholesterol + triglycerides/5). Fasting serum glucose and insulin levels were also evaluated. The homeostatic model assessment of insulin resistance (HOMA-IR; fasting values of glucose × insulin/22.5) was used as an indirect index of insulin resistance.[36] Twenty-four-hour urinary cortisol level was analyzed as an index of the hypothalamus-pituitary-adrenal axis, which eliminates problems due to cortisol fluctuations during the day.[29,31] We chose urinary cortisol, instead of 17-ketosteroids or 17-hydroxysteroids, because it is more representative of free serum cortisol, not being relevantly influenced by obesity and liver and renal function,[37] although possibly influenced by urinary volume.[38] Starting at 7 am, after emptying the bladder of previous nocturnal urine, each woman collected urine in a graduated bottle for the following 24 hours, until 7 am of the following day. Urine samples immediately brought to the laboratory were measured, and a sample of 10 mL was collected.


Serum total cholesterol and triglyceride were measured by enzymatic methods (Olympus AU400; Olympus Diagnostic GmbH, Lismeehan, Ireland). HDL-cholesterol level was determined after precipitation with PEG 6000. The assay for cholesterol has a sensitivity of 5 mg/dL and a coefficient of variation (CV) of 3.0%. The assay for triglycerides has a sensitivity of 1.6 mg/dL and a CV of 2.17%. Glucose was determined by the glucose oxidase method. This assay has a CV of 1.4%.

Insulin was assayed in duplicate by a radioimmunoassay method using available commercial kits (Biodata; Guidonia Montecelio, Rome, Italy). The assay has a sensitivity of 2.1 μU/mL and an intra-assay and interassay CV of 6.2% and 7%, respectively.

Cortisol was assayed in duplicate by means of an immunoassay kit (Bayer Advia Centaur Cor; Bayer, Newbury, UK). The assay has a sensitivity of 0.20 μg/dL and an intra-assay and interassay CV of 3.24% and 3.64%, respectively. Twenty-four-hour cortisol value was calculated by multiplying assay results (given in micrograms per deciliter) × 10 × urine volume (liters in 24 h).

Statistical Analyses

The primary objective was to evaluate whether the Greene Climacteric Scale score is related to the 24-hour urinary cortisol level. The secondary objective was to evaluate whether the 24-hour urinary cortisol level is related to risk factors for cardiovascular disease, such as lipoproteins and insulin resistance.

Factors related to 24-hour urinary cortisol values were evaluated by single regression analysis. Univariate analysis was performed among factors related to 24-hour urinary cortisol level. When colinearity was observed between two related variables (r > 0.7),[39] the one less related to cortisol was not considered in further analysis. Selected variables were entered into stepwise regression analysis. Independent determinants of urinary cortisol level were those exceeding an F value of 4. The resulting independent determinants were entered into a multiple regression analysis to build a linear model for 24-hour urinary cortisol level. Thereafter, urinary cortisol level was related to HOMA-IR and lipids by univariate analysis. Possible other determinants of HOMA and lipids, such as age, months since menopause, age at menopause, BMI, waist, and WHR, were tested by univariate analysis. Collinearity was tested and corrected for. Stepwise regression was used to define independent variables, which were then used to build linear regression models. One-factor analysis of variance followed by the post hoc Fisher exact test was used, as specified. For all the analyses, a P value less than 0.05 was considered as significant. All results are expressed as mean ± SD.

Statistical analyses were performed using the statistical Package Statview 5.0.1 for Apple Macintosh (SAS Institute Inc., Cary, NC).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: