| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Department of Laboratory Medicine, Collegium Medicum, Nicholas Copernicus University, Bydgoszcz, Poland;
aaddress correspondence to this author at: Department of Laboratory Medicine, Collegium Medicum, Nicholas Copernicus University, Sklodowskiej-Curie 9, 85-094 Bydgoszcz, Poland; fax 048-052-585-3603, e-mail kizdiagn{at}cm.umk.pl)
Cessation of ovarian function and sex hormone deficiency are associated with metabolic disorders that increase the risk of cardiovascular disease in women after menopause (1)(2). Changes in sex steroids may influence inflammatory processes and lipid metabolism during the menopausal transition. The status of estrogen in women at early perimenopause is similar to that of premenopause, whereas decreased estradiol is more likely in late perimenopause (3–11 months of amenorrhea) (3). Several studies have demonstrated that lipid concentrations, body weight, blood pressure, and insulin resistance increase after menopause and that endothelial function is impaired (4)(5)(6)(7). Recent data indicate that chronic inflammation may lead to atherosclerosis in healthy populations. It has been suggested that systemic markers of inflammation such as C-reactive protein (CRP) and interleukin-6 (IL-6) are important predictors of cardiovascular risk, and measurement of their concentrations may increase the predictive value of traditional lipid screening (8)(9). CRP assayed by high-sensitivity methods enables the evaluation of low-grade inflammation in early atherosclerosis. Rifai et al. (10)(11) recently proposed algorithms for the prediction of cardiovascular disease risk using CRP and total cholesterol/HDL-cholesterol (TC/HDL-C) or LDL-cholesterol (LDL-C) values.
We investigated the relationship between hormonal status and cardiovascular risk assessed by inflammatory markers and traditional laboratory variables in women during menopausal transition. Our study included 80 early perimenopausal [PERI; mean (SD) age, 47 (3) years] and 78 postmenopausal [POST; mean age (SD), 53 (4) years] healthy nonsmoking women. The PERI women had irregular menses during the previous 12 months and at least one climacteric symptom (hot flashes, night sweats, vaginal dryness, sleeping difficulties, or mood swings). The POST women were divided into 2 groups: 40 early-POST (1–5 years without menses) and 38 late-POST (6–10 years without menses). None of the women had hypertension, thyroid or liver disease, diabetes mellitus, cardiovascular disease, or chronic inflammation or was taking lipid-lowering or antiinflammatory agents, hormonal replacement therapy, or oral contraceptives before the study.
The study was approved by the Collegium Medicum Ethics Committee. All participants gave written informed consent.
Fasting blood samples were collected (on the third day of the menstrual cycle in PERI women) and stored at –70 °C until analysis. Serum was assayed for TC, HDL-C, triglycerides (TGs), glucose, apolipoproteins AI and B, 17ß-estradiol, and follicle-stimulating hormone (FSH; Roche Diagnostics), and plasma was assayed for fibrinogen (Dade Behring). The serum CRP concentration was measured by a high-sensitivity method (Dade Behring); soluble intercellular adhesion molecule (sICAM-1), soluble vascular cell adhesion molecule (sVCAM-1), IL-6, and insulin were measured by ELISA (Bender MedSystems, DRG MedTek).
A body mass index (BMI) of 25–29.9 kg/m2 was defined as overweight; a BMI >30 kg/m2 or a waist-to-hip ratio >0.8 was used as an index of obesity or abdominal obesity. Cardiovascular risk was calculated by algorithms proposed by Rifai and Ridker (10)(11) that use the CRP concentration and the TC/HDL-C ratio or CRP and LDL-C values.
Data were expressed as means (SD) or medians (25th and 75th percentiles). Variables with nongaussian distribution were compared by Mann–Whitney U-test or Kruskal–Wallis ANOVA; others were compared by one-way ANOVA. Pearson or Spearman rank correlation tests were used. In multiple linear regression analysis, age, BMI, waist-to-hip ratio, LDL-C, HDL-C, TG, estradiol, FSH, and insulin/glucose ratio were assessed as the independent variables. The results were presented as the regression coefficient (ß), and the multiple coefficient of determination (R2) was expressed as a percentage. P <0.05 was considered statistically significant.
The characteristics of the women in the study are presented in Table 1
. Markedly higher values for lipids, apolipoproteins, and BMI were found in POST women compared with the PERI group, whereas no differences were observed between the 2 POST groups. Hypercholesterolemia [TC >5.18 mmol/L (200 mg/dL)] was observed in 87% of POST women, and the frequency of overweight and obesity was higher after menopause (53%) than at perimenopause (38%).
|
Among the inflammatory markers, only the sICAM-1 concentration in POST women significantly exceeded that in perimenopausal women (Table 1
). CRP concentrations were similar in all groups (Table 1
). Markedly increased CRP values (>3 mg/L) were found in 25% of late-POST and 15% of PERI women (difference not significant). Fibrinogen concentrations were slightly higher after menopause (Table 1
). A high frequency of increased fibrinogen concentrations >3.5 g/L occurred in 37% of PERI and 50% of POST women. IL-6 and sVCAM-1 concentrations, which were similar in the study groups, were within the reference ranges (Table 1
).
CRP, IL-6, and fibrinogen concentrations were associated with insulin (P = 0.001, P = 0.01, and P = 0.001, respectively), insulin/glucose ratio (P = 0.001, P = 0.028, and P = 0.004, respectively), and BMI (P = 0.001, P = 0.01, and P = 0.001, respectively). The sICAM-1 concentration was also positively related to BMI values (P = 0.01).
We found weak but significant negative correlations between HDL-C and CRP (P = 0.02), IL-6 (P = 0.04), and sICAM-1 (P = 0.04). Fibrinogen correlated with LDL-C concentration (P = 0.002). sICAM-1 was the only variable that was inversely related to the estradiol concentration (P = 0.049), whereas sVCAM-1 did not correlate with any measured variables.
In multiple regression analysis, CRP, sICAM-1, and fibrinogen independently correlated with BMI in the whole group. BMI explained 18% of CRP variance, 15% of sICAM-1 variance, and 24% of fibrinogen variance. IL-6 was independently associated with indices of insulin resistance (14%). Other independent variables forced into the final models had a weaker influence on inflammation markers.
Multiple regression analysis performed in the PERI or POST group revealed that BMI had a stronger influence on CRP values in the latter group (ß = 0.29, P = 0.03, 18% vs ß = 0.55, P = 0.0017, 37%, respectively). CRP concentrations were related to TG values (ß = 0.38, P = 0.009, 19%) and HDL-C (ß = –0.16, P = 0.33, 5.9%) only in the POST group.
Insulin resistance had a similar effect on CRP in both POST and PERI women, whereas the influence on IL-6 values was markedly stronger in POST than in PERI women (ß = 0.58, P = 0.01, 26% vs ß = 0.22, P = 0.18, 7%, respectively). The relationship between sICAM-1 and fibrinogen was similar in PERI and POST women.
Women after menopause had significantly higher relative cardiovascular risk only if estimated by algorithm using CRP and LDL-C values [median 3.0 (1.7–4.9) in POST vs 1.7 (1.7–2.9) in PERI women; P = 0.009]. When CRP and TC/HDL-C values were used for the calculation, the risk was similar in PERI and POST women [1.7 (1.2–2.35) vs 1.85 (1.4–3.5)]; however, POST women were more than 2 times more likely to have cardiovascular risk in the highest quartile than PERI women (30% vs 14%).
Both inflammatory processes and a disturbed lipid profile may mediate the development and progression of atherosclerosis. Several circulating inflammation factors such as cellular adhesion molecules and cytokines are associated with cardiovascular disease (8)(12)(13). Among systemic inflammatory markers, CRP has been regarded as a strong independent predictor of cardiovascular events in a healthy population (10)(11)(14). The median CRP concentration in healthy European women >45 years of age was 1.2–1.7 mg/L (15). In our study, healthy PERI (>40 years of age) and POST women had a lower median CRP concentration (1.05 mg/L), and 25% of them had values >3 mg/L. Higher fibrinogen concentrations observed after menopause have also been reported previously (16).
Slight increases in CRP and fibrinogen concentrations in women across the menopausal transition were not associated with sex hormone changes. Similarly, the IL-6 concentration was not related to FSH or estradiol concentrations (1)(17).
Unlike other measured inflammatory markers, sICAM-1 was only weakly inversely related to estradiol concentration. This univariate relationship did not remain significant in multivariate analysis.
In this study, sICAM-1 values, substantially increased after menopause, were strongly and independently associated with BMI. Similar to the findings of Oger et al. (18), we did not observe higher sVCAM-1 concentrations after menopause. sVCAM-1 values were not related to any variables that change after menopause.
Finally, we estimated relative cardiovascular risk according to Rifai and Ridker (10)(11). We have shown that the highest calculated cardiovascular risk was 2 times more likely among healthy POST than PERI women and associated more with hypercholesterolemia than with CRP concentrations.
The impact of menopause on increased cardiovascular risk seems to be related mainly to BMI, insulin resistance, and increased TC and sICAM-1 concentrations. Changes in the concentrations of sex hormones at the menopausal transition did not directly influence the inflammation state in clinically healthy women.
Acknowledgments
This work was supported by a grant (SD 29/02) from the Collegium Medicum, Nicholas Copernicus University, Bydgoszcz, Poland.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
