Diabetes abrogates sex differences and aggravates cardiometabolic risk in postmenopausal women

Cardiovascular disease (CVD) is a major public health problem worldwide and the leading cause of death in Portugal and in most industrialized and developing countries [1, 2]. This mortality of cardiovascular (CV) cause appears to be increasing in countries where type 2 diabetes mellitus (T2DM) is prevalent; therefore, diabetic patients are associated with a risk 2 to 4 times higher of CVD development, accounting for 50-80% of deaths as well as for the increased morbidity and loss of quality of life in these patients [3, 4]. Apart from being per se a risk factor for CVD, T2DM is often associated with a higher prevalence of other important risk factors, including hypertension, obesity, insulin resistance, microalbuminuria and dyslipidaemia [5, 6], a multifactorial condition, often referred as metabolic syndrome, which is responsible for the high CV morbidity and mortality in those patients [7]. Pharmacological treatment is crucial for delaying the progression of diabetes, yet it remains inadequate in preventing the increased risk of CVD in patients with T2DM, especially women [6].

The CV risk seems to be different for men and women. The apparent cardioprotective effects of endogenous estrogens seem to prevent CVD in premenopausal women, when compared with age-matched men; however, following menopause and the consequent loss of hormonal effects, gender-based differences in CVD are reduced [8, 9]. The increased CVD risk after menopause seems to be associated with the emergence of the features of metabolic syndrome [10], but the precise causes remain to be fully elucidated. Obesity is an independent risk factor for macrovascular disease across sexes [11]; however, despite higher incidence of obesity in premenopausal women, rates of macrovascular disease are lower than in men. Interestingly, this sex difference, which normally vanishes after menopause, is rapidly lost in premenopausal T2DM patients, with CVD reaching 2- to 5-fold higher rates than in non-diabetic women [12]. In fact, women with T2DM, compared with age-matched non-diabetic women, exhibit several-fold higher rates of death related to coronary artery disease (CAD), with event rates nearly identical to those observed in T2DM men [13]. Traditional cardiovascular risk factors (CVRFs) cannot completely account for these sex differences in cardiovascular mortality [14]. So, it seems clear that more studies are needed to understand the precise influence of gender and menopause in the risk for CVD, especially in diabetic patients, in order to achieve effective preventative and disease management strategies to reduce the CVD risk associated with the disease, particularly in postmenopausal women.

As the leading cause of death in T2DM patients, the CVD is a complex phenomenon, which involves metabolic, lipidic, oxidative, inflammatory, as well as genetic factors. Beyond hyperglycemia, a number of other common risk factors may contribute to CVD in diabetic women. Low plasma levels of high-density lipoprotein cholesterol (HDL-c) have been largely recognized as a risk factor for coronary heart disease (CHD) [15, 16] and they are a common feature of the dyslipidaemia linked to diabetes and insulin resistance [17]. Recent findings suggest that monitoring the type of HDL particles (carry distinct and specific proteins or lipids and differentiated by their density and size – large, intermediate and small), rather than their total quantity, is a more reasonable way of determining the CV risk, suggesting that different subpopulations may have a different role in reverse cholesterol transport and CVD risk protection [18]. In fact, some recent studies have been reporting that large HDL levels are reduced in patients with CAD compared to healthy subjects and inversely related to both disease severity and progression of coronary lesions [19]. Paraoxonases (PONs) are enzymes from the HDL, and have been indicated as one of the best candidates for the protective activity of HDL against CVD development, namely protection against LDL oxidation [20]. In fact, oxidized LDL (Ox-LDL) and HDL are indeed antagonists in the development of CVD [21]. Several studies have shown a strong positive correlation between the lower levels of HDL and the development of atherosclerosis, but their role in the determination of CVR in gender and menopause in a diabetic population remains to be clarified.

Chronic inflammation is currently viewed as a key factor in the development of atherosclerosis, contributing to raise the overall CV risk, namely in diabetic patients. An inflammatory imbalance, as manifested by increased pro-inflammatory cytokines, such as the tumor necrosis factor alpha (TNF-α), increased C-reactive protein (hsCRP), and/or reduced levels of anti-inflammatory and anti-atherogenic mediators, such as adiponectin, have been considered a key factor for the increased CVR in some pathologies [22, 23], deserving more attention in respect to the gender and menopause, mainly in diabetes. Similar importance is now attributed to the phenomenon of angiogenesis, which has the vascular endothelial growth factor (VEGF) as the key biomarker, which has been suggested as a novel promising target for T2DM treatment [24]. Another new marker that deserves our attention is the uric acid; although uric acid can act as an antioxidant, excess serum accumulation is often associated with several conditions, and has been suggested as an independent risk factor for carotid atherosclerosis in patients with type 2 diabetes [25]. Furthermore, a prospective follow-up study showed that hyperuricemia is associated with higher risk of T2DM, independent of obesity, dyslipidemia and hypertension [26]. Thus, it would be important to understand its influence concerning gender and menopause modulation of CV risk, mainly in diabetes.

This study aimed to evaluate the influence of gender and menopause on CV risk in a diabetic population, using both traditional and new non-traditional markers.

The risk for coronary artery disease (CAD), the main cause of death in women, increases after menopause [9]. Premenopausal women are at lower risk for CAD than postmenopausal and than men [8, 9]. Although previous studies have spotlighted the effects of estrogens, no conclusive evidence has proven their role in reducing the incidence of CVD [27]. Indeed, hormone replacement therapy for the menopausal women does not confer cardiovascular protection according to the Women’s Health Initiative trial [28]. Therefore, estrogen deficiency may indirectly contribute to the increased risk of CVD in postmenopausal women. Some studies suggest that the cardiovascular effects usually attributed to menopause are merely a consequence of the older age of menopausal women [29]. In contrast, others demonstrated that menopause is associated with a modest increase in total fatness and an accelerated accumulation of central body fat that exceeds changes normally attributed to the aging process [30]. Indeed, the transition from premenopausal to postmenopausal status is associated with the emergence of various risk factors for metabolic syndrome and the rising incidence of CAD during menopause occurs in parallel with an increase in the incidence of T2DM [10]. The presence of diabetes increases the risk for CAD in both premenopausal and postmenopausal women and probably counteracts the protective effect of estrogens on the vasculature [31, 32], so that premenopausal diabetic women show the same risk for CAD as men and 2- to 5-fold higher rates than in non-diabetic women [12, 13]. In addition, CAD is considered as one of the most important complications of DM in both sexes. Hypertension and dyslipidemia are risk factors for CAD among diabetic patients and it is well established that patients with diabetes have more extensive and more rapidly progressive CAD than non-diabetic subjects [12, 13]. Several studies showed a 2 to 4-fold higher prevalence of atherosclerotic disease in diabetic compared to non-diabetic individuals [33, 34]. Diabetic women have greater mortality risk from CAD than non-diabetic men and women [35]. Since the traditional CVRFs cannot completely account for these sex differences in cardiovascular mortality, more research is pivotal to understand the precise influence of gender and menopause in the risk for CVD, especially in diabetic patients.

This study has compared the effects of gender and menopause on cardiovascular parameters/markers in a diabetic population under antidiabetic, antidyslipidemic and antihypertensive medication, compared with matched controls. Male and female subpopulations, from both groups, presented identical age, percentage of males/females and BMI. Female waist circumference normal range is distinct between males and females cannot be directly compared. As expected, diabetic patients (from both genders) presented higher values of glycemia, BMI and WC, when compared with the age and gender-matched controls. However, systolic and diastolic blood pressure were significantly reduced in the diabetic patients (male and female), confirmative of the antihypertensive medication taken. The data of the control subjects shows slight higher values of blood pressure that might be related with the age of these individuals, which were chosen to be age-matched with the diabetic ones in order to minimize the influence of this factor in the analysis. Additionally, lower values of total-c, LDL-c, Ox-LDL-c and non-HDL-c were found in the diabetic patients when compared with the control subjects. However, increased contents of serum TGs and reduced of HDL-c were found for the diabetic subjects. This classical lipid profile is in agreement with what could be expected with the type of antidyslipidemic therapy practiced, since statins, the main class of drugs used, have less impact on TGs and HDL-c than on total-cholesterol and LDL-c. Despite the pharmacological and behavioral interventions directed to control risk factors for CVD in diabetic patients, the incidence of CVD remains alarmingly high. Dyslipidaemia, which is associated with increased CVD mortality in diabetic individuals [36], is among the most important modifiable risk factors. There is now a wide variety of antidyslipidemic drugs [37, 38]; however, with the increasing combination of risk factors commonly found in several CV diseases, including in T2DM, the control of dyslipidaemia is not enough, requiring a more effective modulation of HDL-c, which remains only slightly modifiable with the current pharmacological arsenal [39, 40]. Our study in diabetic patients under antidyslipidemic therapy reinforces the need of better intervention on HDL-c.

Concerning other markers of cardiovascular/metabolic risk, the male diabetic patients presented increased hsCRP, uric acid and reduced adiponectin concentrations, while the female diabetic subjects also presented increased hsCRP and uric acid, as well as TNF-α and VEGF, with reduced adiponectin levels. Thus, both the male and female diabetic patients showed impaired markers of cardiometabolic risk, which is accompanied by increased waist circumference. This pro-inflammatory profile, a key factor in the development of atherosclerosis, is in agreement with other studies which have reported increased pro-inflammatory mediators, such as TNF-α and hsCRP, and reduced levels of anti-inflammatory and anti-atherogenic mediators, such as adiponectin [21, 41]. Furthermore, adiponectin levels of diabetic woman were negatively associated with hsCRP, in agreement with a recent study [42]. Additionally, increase in hsCRP levels was greater in women than men, in agreement with previous data [43], and might be due to the significantly increased concentration in the premenopausal diabetic women. Hence, even though the reduced blood pressure and some of the traditional marker of lipid profile (total-c, LDL-c and non-HDL-c), which is in agreement with the medication taken, the diabetic patients presented obesity and visceral adiposity, accompanied by markers of low-grade inflammation, and uncorrected TGs and HDL-c contents, which are less modifiable with the most used antidyslipidemic agents (mainly statins). Collectively, the impaired parameters might be viewed as predictors/markers of an increased cardiometabolic risk in this diabetic population.

Regarding the differences among men and women, in the control population females presented lower glycemia, unchanged total-c, TGs, LDL-c and non-HDL-c, together with reduced Ox-LDL, TNF-α, adiponectin and uric acid. However, lower PON1 activity was found in women, which might be explained by the men’s compensatory increment of PON1 activity against the pro-oxidative profile (viewed by the significantly higher Ox-LDL contents). In agreement, despite identical total HDL-c concentrations in both subgroups, the quality of HDL was better in female, as they presented higher contents of large HDL-c and reduced of small HDL-c. Therefore, collectively, this data is indicative of a better cardiometabolic profile and lower CV risk in non-diabetic females when compared with males. However, when the subpopulations of diabetic patients are compared, the differences between male and female are significantly reduced. Indeed, almost all the parameter of reduced risk found in the control females (vs control males) are no long different, and women even presented higher values of HbA1c, VEGF and hsCRP, despite the higher adiponectin concentration, which might be viewed as the exception of this clearly worse cardiometabolic profile. If compared with the female control population, this indication is even clearer. The differences of adiponectin encountered might be due to differences on fat distribution (which is known to differ between genders, with men having more visceral and less subcutaneous fat) as well as due to the effect of sex hormones, that are involved in the metabolism of adipose tissue and fat distribution. Our results may be due to increased visceral fat in female control population (proportionately larger than the male). However, when comparing the diabetic female population with control female we found that adiponectin levels are decreased in diabetic women, in agreement with previous data [44], which is consistent with the other results of our study that show a worse cardiometabolic profile of women in the presence of T2DM. The significant decrease of adiponectin levels in diabetic men, when compared with control men, is consistent with other studies which have previously suggested that low adiponectin levels are associated with fatty liver disease in women and low testosterone levels in men with type 2 diabetes [45].

Diabetic women presented not only the expected increased glycemia and obesity (higher BMI and WC) when compared with the female control subjects, but also increased hsCRP, TNF-α, uric acid and VEGF, accompanied by reduced adiponectin. Additionally, despite unchanged values of classical lipid profile (total-c, LDL-c, non-HDL-c), due to medication, the diabetic females showed increased TGs and reduced HDL-c quality, confirmed by the reduced amount of large HDL-c and increased of small HDL-c, regardless of normal total HDL-c quantity. Low levels of HDL-c are a major CHD risk factor in type 2 diabetic subjects [46]. In T2DM patients, TGs are usually elevated and HDL metabolism is perturbed with evidence of both qualitative and quantitative alterations [17]. Despite the growing body of evidence indicating that determination of HDL subpopulations may add an important information on CHD risk prediction [15, 18], data on HDL subpopulation distribution and its modulating factors in women when compared with male are limited, including in diabetic populations. Our data is in agreement with the study of Russo et al. (2010) that showed a lipid and HDL subpopulation profiles more atherogenic in the diabetic women [47].

Some of the markers of cardiometabolic risk analyzed showed important correlations in the female diabetic population, in opposition with the data obtained in the male one, reinforcing the previous data and indications. Indeed, the waist circumference in the diabetic women is positively and significantly correlated with the concentrations of hsCRP, TNF-α and VEGF, which was not encountered in the male diabetic individuals, for whom the increased uric acid is indeed the best association with waist circumference. Additionally, TNF-α contents are positively and significantly correlated with VEGF levels and hsCRP concentrations are inversely and significantly correlated with adiponectin values. Collectively, this data suggests that obesity, and especially abdominal adiposity, in female diabetic subjects, is metabolically more active and, consequently, deleterious than in male. This data is in agreement with the fact that women with T2DM, compared with age-matched non-diabetic women, exhibit several-fold higher rates of death related to CAD, with event rates nearly identical to those observed in T2DM men [13], as is also in line with other study in type 1 diabetes [48]. Our data suggested that diabetes abrogate the protective effect of gender encountered in women without diabetes, which agrees with the suggestion that CHD risk is higher in men and that the difference is reduced in diabetic populations [49]. Nevertheless, the traditional CVRFs fail to completely explain these sex differences, and the new “non-classical” factors seem to be able to improve knowledge and clarify this discrepancies, in particular the more atherogenic lipid sketch and pro-inflammatory and pro-angiogenic profile, viewed mainly by the contents of HDL-c subpopulations and the serum values of TNF-α, hsCRP and VEGF.

Regarding the influence of menopause, glycemia was significantly higher after menopause, both for control and diabetic females. However, postmenopausal diabetic woman presented significantly increased BMI and waist circumference when compared with the control postmenopausal subjects, while no changes were observed between groups at premenopausal stage. Systolic blood pressure was significantly increased after menopause, for both control and diabetic groups. However, diabetic patients presented a trend to reduced SBP and significantly lower values of DBP, when compared with the control population, both for pre and postmenopause comparisons, which might be due to antihypertensive medication. In relation to the classical lipid profile, in both the 2control and diabetic groups, no changes were encountered between pre and postmenopausal women for total-c, TGs, LDL-c and non-HDL-c. Additionally, both pre and postmenopausal diabetic patients presented unchanged or even decrease (LDL-c) values for all of those parameters. This profile is in agreement, again, with the type of antidyslipidemic therapy practiced by those patients, mainly statins, which decrease total-c and LDL-c contents, but have less impact on TGs and HDL-c levels. Indeed, TGs concentrations were significantly higher, but only in the postmenopausal diabetic women when compared with the control postmenopausal subjects. In addition, although total HDL-c concentration was unchanged in pre and postmenopausal diabetic women, when compared with the control matched subpopulations, in the postmenopausal stage the HDL-c quality is worse, viewed by the significantly reduced proportion of large HDL-c and increased of small HDL-c. Although further studies are needed to clarify the complex role of the different HDL particles in the development of CHD, several lines of evidence indicate that evaluating HDL subpopulations profile may provide some adjunctive information on CHD risk definition, independently from total HDL-c measurements [15, 18]. Our data clearly reinforces this recommendation, in particular in the postmenopausal diabetic women, in whom the quality of HDL is poor.

Concerning other markers of cardiovascular/metabolic risk, before menopause the diabetic women presented significantly increased hsCRP, uric acid and VEGF, with a trend to reduced adiponectin contents, when compared with the non-diabetic females. These changes were maintained after the menopause, accompanied by increased TNF-α concentration. In this population, an important association was found between some of the markers of cardiometabolic risk. Indeed, the waist circumference in the postmenopausal diabetic women is positively and significantly correlated with the concentrations of hsCRP, TNF-α and VEGF. Additionally, TNF-α contents are positively and significantly correlated with VEGF levels, hsCRP concentrations are inversely and significantly correlated with adiponectin values and, finally, uric acid was inversely and significantly correlated with large HDL-c. Thus, the postmenopausal diabetic females have an increased obesity which seems to be metabolically more deleterious than in premenopausal diabetic women, namely due to obesity-induced low-degree chronic inflammation, through enhanced adipose tissue-derived cytokine expression, in agreement with previous recent reports [50]. The transition from pre to postmenopause may be associated with features of the metabolic syndrome, including an increased central body fat, a shift toward a more atherogenic lipid profile, as well as other risk factors. Obesity is an independent risk factor for macrovascular disease across sexes, but despite higher incidence of obesity in premenopausal women, rates of macrovascular disease are lower in premenopausal women than in men. Interestingly, this sex difference, which normally vanishes after menopause, is rapidly lost in premenopausal DM patients, with CVD reaching 2- to 5-fold higher rates than in non-diabetic women [12]. In fact, women with T2DM, compared with age-matched non-diabetic women, exhibit significantly higher rates of death related to CAD, similar to those observed in T2DM men [13]. The emergence of the CVD risk factors in the postmenopausal women may be a direct result of ovarian failure, or an indirect result of the metabolic consequences of body fat centralization with estrogen deficiency. Previous studies have also demonstrated that menopause is associated with a modest increase in total fatness and an accelerated accumulation of central body fat that exceeds changes normally attributed to the aging process [30, 51]. Similarly in our studies, postmenopausal women, when compared to premenopausal women, had a higher WC, TGs level and other non classical markers, suggesting an increased CV risk. Previous reports have indicated that the diabetic postmenopausal women have more severe CAD and CV risk compared to non-diabetic women [52].

Although the small number of premenopausal diabetic women could be viewed as a study limitation, that deserves further strengthening, postmenopausal diabetic women have a clearly poor cardiometabolic profile since several parameters that are unchanged between premenopausal diabetic subjects when compared with the premenopausal controls, are aggravated in the postmenopausal diabetic ones, versus the corresponding postmenopausal controls, including significantly increased TGs, small HDL, WC, BMI and TNF-α values, aggravated contents of VEGF and uric acid, significantly lower of large HDL and a trend to reduced adiponectin concentration. Collectively, our data reinforces the suggestion that the multi-targeted treatment of all risk factors is even more justified in postmenopausal women, which is mainly suggested by non-traditional markers in this diabetic population medicated for hypertension and dyslipidemia. The apparently more deleterious visceral obesity, the more atherogenic lipid sketch and the pro-inflammatory profile in diabetic patients in general, but in the postmenopausal women in particular, urges precise attention and proper multi-therapeutic intervention.

Our study suggests that diabetes abrogate the protective effect of female gender on non-diabetic subjects when compared with male, and that postmenopausal diabetic females presented worsen cardiometabolic profile, including a more atherogenic lipid sketch and a pro-inflammatory and pro-angiogenic profile. The traditional CVRFs fail to completely explain these differences, which are better clarified using “non-classical” markers, such as contents of HDL-c subpopulations, rather than total content, and mediators of inflammation and angiogenesis, namely TNF-α, hsCRP, uric acid and VEGF. Multi-therapeutic intervention, directed to obesity, atherogenic lipid particles and inflammatory mediators, is advisory in order to efficiently prevent the serious cardiovascular complications of diabetes in this higher-risk population.