Impact of Preeclampsia and Parity on Sex-based Discrepancies in Subclinical Carotid Atherosclerosis in Type 1 Diabetes

Abstract

Context

The excess risk of fatal and nonfatal cardiovascular events is roughly twice as high in women than in men with type 1 diabetes.

Objective

To evaluate the impact of preeclampsia and parity on sex-based discrepancies in preclinical atherosclerosis and on the diagnostic performance of a cardiovascular risk scale.

Design

Cross-sectional study.

Setting

Single tertiary hospital.

Patients

A total of 728 people with type 1 diabetes (48.5% women) without cardiovascular disease and age ≥40 years, nephropathy, and/or ≥10 years of diabetes duration with another risk factor.

Intervention

Standardized carotid ultrasonography.

Main Outcome Measures

Carotid plaque determined by ultrasonography and cardiovascular risk estimated according to the Steno T1 Risk Engine (Steno-Risk).

Results

Nulliparous women and parous women without previous preeclampsia had a lower risk for carotid plaque than men (adjusted odds ratio: .48, 95% confidence interval [.28-.82]; adjusted odds ratio: .51 [.33-.79], respectively), without differences in the preeclampsia group. The prevalence of carotid plaque increased as the estimated cardiovascular risk increased in all subgroups except for preeclampsia group. The area under the curve of the Steno-Risk for identifying ≥2 carotid plaques was lower in the preeclampsia group (men: .7886; nulliparous women: .9026; women without preeclampsia: .8230; preeclampsia group: .7841; P between groups = .042). Neither the addition of parity nor preeclampsia in the Steno-Risk led to a statistically significant increase in the area under the curve.

Conclusion

The risk for carotid plaque in women compared with men decreased as exposure to obstetric factors diminished. However, the addition of these factors did not improve the prediction of the Steno-Risk.

Despite improvements in the treatment and management of type 1 diabetes, this population still experiences higher rates of cardiovascular disease (CVD) than the general population (1, 2), with sex discrepancies. Although a higher number of cardiovascular events have been reported in men than in women with type 1 diabetes, the excess risk of fatal and nonfatal cardiovascular events is roughly twice as high in women than in men (compared with their counterparts without diabetes) (3, 4).

Unlike men, women are exposed to specific cardiovascular risk factors (CVRF), such as those related to the reproductive area. Indeed, international guidelines on the prevention of CVD recommend considering preeclampsia as an additional individual risk-enhancing clinical factor in the general population (5, 6). Several studies have reported the increased risk of CVD in women who developed preeclampsia during pregnancy (7-9). In fact, our group showed that a history of preeclampsia was associated with more advanced preclinical carotid atherosclerosis not only in the general population but also in women with type 1 diabetes (10). In addition to adverse pregnancy outcomes, parity has also been described as a CVRF, showing a J-shaped association between parity and incident CVD (11, 12). However, data in the type 1 diabetes population are scarce and are mainly focused on the relationship between parity and microvascular complications and CVRF (13, 14)

Given the importance of CVD in people with type 1 diabetes, tools to predict cardiovascular events are essential to tailoring cardiovascular risk management according to the risk of each individual. Specific type 1 diabetes scales that take into account specific CVRF of this entity (such as diabetes duration or diabetes-related complications) have been developed and show better performance than the classical scales. In this setting, the Steno T1 Risk Engine (Steno-Risk) has shown to have an independent association with the identification of several markers of subclinical atherosclerosis and CVD events in several type 1 diabetes populations (15-17).

As mentioned previously, there is growing evidence of the role of sex-specific reproductive factors in the risk of CVD in women with type 1 diabetes. However, in contrast to the general population, there is no evidence about whether these factors could improve the discrimination of specific type 1 diabetes scales, such as Steno-Risk (18). Considering this background, this study aimed to evaluate the risk of preclinical atherosclerosis according to sex-specific reproductive factors, such as preeclampsia and parity. Additionally, the diagnostic performance of adding these factors to the Steno-Risk equation for the identification of preclinical atherosclerosis plaque was studied.

Materials and Methods

Study Design and Participants

This was a cross-sectional investigation of a specific prospective protocol for CVD risk assessment in subjects with type 1 diabetes (which included a structured evaluation of CVRF and a standardized carotid ultrasonography to assess subclinical atherosclerosis) (15, 19). Briefly, this protocol included patients diagnosed with type 1 diabetes and no previous personal history of CVD (coronary artery disease, ischemic stroke, peripheral vascular disease, or heart failure), but with conditions associated with high risk according to the standards proposed by the main CVD prevention guidelines (20-22): (A) age ≥40 years, (B) individuals with any stage of diabetic nephropathy regardless of age, and (C) individuals of any age, with at least 10 years of disease duration and 1 additional CVRF (active smoking habit, diabetic retinopathy, hypertension, triglycerides >150 mg/dL, low high-density lipoprotein [HDL] cholesterol [<40 mg/dL in men, <45 mg/dL in women], family history of premature CVD in first-degree relatives [<55 years of age in men and <65 years in women], severe hypoglycemia [defined as an episode of confirmed hypoglycemia requiring external assistance for recovery], or hypoglycemia unawareness [defined as a score >3 in the Clarke test], with the validated Spanish version (23), and history of preeclampsia/eclampsia in at least 1 previous pregnancy in women).

The study protocol was conducted according to the Declaration of Helsinki. All patients provided informed consent, and the study was approved by the Ethical Committee of the Hospital Clínic, Barcelona, Spain (approval number HCB/2017/0977).

Clinical and Laboratory Measures

Both demographic and clinical data including duration of type 1 diabetes, family history of premature CVD in first-degree relatives, history of microvascular diabetes complications, and medical treatment (multiple-dose insulin, continuous subcutaneous insulin infusion, lipid-lowering agents, and antihypertensive and antiplatelet drugs) were obtained from medical records.

Diabetic nephropathy was assessed according to the albumin-to-creatinine ratio, with <30 mg/g considered as normal and ≥30 mg/g as diabetic nephropathy (confirmed in at least 2 of 3 consecutive determinations). The use of angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers, without a history of hypertension or CVD, was also considered as diabetic nephropathy. Fundus oculi was used for the diagnosis of diabetic retinopathy, which was always confirmed by an ophthalmologist.

Anthropometric measurements (weight, height, and waist circumference) were also obtained. Patients were weighed wearing light clothing and being barefoot, with the use of a calibrated electronic scale. The body mass index was calculated as weight in kilograms divided by the square height in meters. The midpoint between the lowest rib and the iliac crest was used to measure waist circumference.

Laboratory parameters were measured in fasting blood and first-morning urine spot samples. Lipid profile (including total cholesterol, triglycerides, and HDL cholesterol), glucose, creatinine, and albumin-to-creatinine ratio were assessed with standardized assays. Low-density lipoprotein (LDL) cholesterol was determined with the Friedewald formula. Non-HDL cholesterol was calculated as the deduction of HDL cholesterol from total cholesterol. The estimated glomerular filtration rate (eGFR) was assessed with the Chronic Kidney Disease Epidemiology Collaboration equation. Glycated hemoglobin (HbA1c) values (Tosoh G8 Automated HPLC Analyzer; Tosoh Bioscience, South San Francisco, CA) (Diabetes Control and Complications Trial aligned, normal range 4-6% [20-42 mmol/mol]) were also recorded.

Sex-specific Reproductive Factors

All women with type 1 diabetes were asked about their history of previous pregnancies. The diagnosis of preeclampsia included internationally recommended criteria (24) and was generally defined as de novo hypertension (≥140 mm Hg systolic and/or ≥90 mm Hg diastolic) after 20 weeks of gestation and proteinuria (300 mg/24 hours or ≥1+ in the dipstick test). All the gestational information (even in women without pregnancy complications) was confirmed by verifying medical records.

Cardiovascular Risk Estimation

Cardiovascular risk was assessed with the Steno-Risk equation (17). Briefly, Steno-Risk estimates the 10-year risk of fatal or nonfatal CVD (ischemic heart disease, ischemic stroke, heart failure, and peripheral artery disease) based on 10 variables (age, sex, diabetes duration, HbA1c, systolic blood pressure, LDL cholesterol, albuminuria, eGFR, smoking habit, and regular exercise [≥3.5 hours/week]). Individuals were classified as having low (<10%), moderate (10%-19.9%), or high (≥20%) risk, accordingly.

Carotid B-mode Ultrasound Imaging

Carotid plaques were evaluated with high-resolution B-mode ultrasound (ACUSON X700 [Siemens Healthineers] or Aplio a450 [Canon]) with the same electric linear array 5- to 10-MHz transducer. Predefined and standardized imaging protocols to measure carotid intima-media thickness (IMT) were used, and the presence of plaque was determined as previously described (15, 25). Carotid images were visualized with B-mode and color Doppler in longitudinal and transverse planes to appraise circumferential asymmetry. Carotid plaques were defined according to focal echo structures intruding into the arterial lumen by at least 50% of the surrounding IMT value or when IMT was at least 1.5 mm as measured from the media-adventitia interface to the intima-lumen surface (26). Two experienced endocrinologists performed all the procedures, and IMT and plaque ascertainment and measurements were made by the same researcher (A.J.A.) using semiautomatic software.

Statistical Analysis

Data are presented as median [25th and 75th percentiles] or number (percentage) unless otherwise indicated. The Mann-Whitney U test, Kruskal-Wallis, Pearson χ² test, and ANOVA were performed, as appropriate, for comparisons between groups. The Bonferroni test was used as a post hoc analysis to make pairwise comparisons, correcting for multiple analyses.

Logistic regression modeling was used to estimate the association between sex and carotid plaque burden (presence of at least 1 carotid plaque or presence of ≥2 carotid plaque). Sex was included in the model as a binary variable (men/women) or categorical variable (men, nulliparous women, and parous women with/without history of preeclampsia), with men being the reference category. Two adjusted models were performed. Model 1 included the same variables of Steno-Risk: age, diabetes duration, HbA1c, systolic blood pressure, LDL cholesterol, albuminuria, eGFR, smoking habit, and regular exercise. Model 2 included model 1 plus statin use. Interaction analysis was performed between sex and age (≤ or >50 years).

Furthermore, receiver operating characteristic curves were used to assess the optimal model for identifying individuals with carotid plaque. Area under the curve (AUC) comparisons were performed between the Steno-Risk equations according to sex category. Additionally, among women with type 1 diabetes, the diagnostic performance of adding preeclampsia and parity (number of live births) to the Steno-Risk equation for the identification of preclinical atherosclerosis plaque was evaluated. All analyses were performed using STATA version 14.0 (Stata Corp., College Station, TX, USA). A 2-sided P < .05 was considered statistically significant.

Results

Subject Characteristics

A total of 728 subjects were included in the study, with a median age of 47.0 [40.9-55.1] years (60.4% aged ≤50 years), a median duration of diabetes of 25.9 [20.1-33.7] years, a median HbA1c of 7.4% [6.9%-8%], and 38.6% were smokers (Supplemental Table 1 (27)). Among women (n = 353, 48.5% of the sample), 125 (35.4%) were nulliparous, 175 (49.6%) were parous without a history of preeclampsia, and 51 (15%) had a history of preeclampsia. Parous women reported having 1 (48.2%), 2 (47.4%), or 3+ (4.5%) live births, without significant differences in the history of preeclampsia between groups (Supplemental Table 2 (27)). Nulliparous women presented a better cardiovascular profile, with lower rates of hypertension, lower systolic and diastolic blood pressure, higher HDL cholesterol, and lower use of statins compared with men, without significant differences in rate of premature CVD in first-degree relatives, smoking habit, HbA1c, diabetes duration, or diabetic nephropathy (Table 1). Conversely, the rates of retinopathy were higher in nulliparous women compared with men (46.5% vs 35.7%, P < .05). Despite women with previous preeclampsia having higher HDL cholesterol concentrations than men, these levels were lower compared with their counterparts without preeclampsia or without children (Table 1). The remaining characteristics are shown in Table 1.

Prevalence of Preclinical Carotid Atherosclerosis

In the whole cohort, 39.4% of subjects had at least 1 carotid plaque (22.8% with 2 or more plaques) with significant between-sex differences (men: 43.5% vs women: 35.2%; P = .023). Indeed, the crude and adjusted logistic regression model showed that women had a lower risk for carotid plaque than men (crude odds ratio [OR]: .68 [95% CI, .50-.92]; adjusted OR [aOR]: .54 [95% CI, .38-.78]).

When dividing the whole cohort according to parity and previous preeclampsia, nulliparous women had a lower prevalence of carotid plaque (26%) than their counterparts (vs 38.8% in parous women without preeclampsia, P = .025; vs 41.2% in parous women with preeclampsia, P = .046) and men (vs 43.5%, P < .001) (Table 2). However, when adjusting for well-known CVRF (age, diabetes duration, HbA1c, systolic blood pressure, LDL cholesterol, albuminuria, eGFR, smoking habit, and regular exercise), both parity and preeclampsia were associated with carotid plaque. Not having children or no history of preeclampsia was associated with a lower risk of carotid plaque compared with men (aOR: .48 [.28-.82]; aOR: .51 [.33-.79], respectively). Conversely, the risk of carotid plaque in parous women with a history of preeclampsia was not significantly different than men (aOR: .87 [.44-1.74]). The inclusion of statin use did not modify the results (Table 2).

An interaction analysis between age and sex was statistically significant (P for interaction = .037); thus, the logistic models were performed stratifying for age (≤ or >50 years). The association observed between the presence of carotid plaque and sex-specific reproductive factors was maintained among subjects aged younger than 50 years (nulliparous women: aOR: .41 [95% CI, .18-.89]; parous women without preeclampsia: aOR .43 [95% CI, .21-.86]). However, the associations were blunted among those aged >50 years (Table 2).

Last, the same analysis was performed for the presence of 2 or more carotid plaques, showing similar results (Table 2).

Relationships Between Estimated Risk and Preclinical Carotid Atherosclerosis

The use of the Steno-Risk equation classified low, moderate, and high cardiovascular risk in 25.1%, 42.4%, and 32.5% of men, respectively, and 35.3%, 36.9%, and 27.8% of women, respectively (P = .010). When taking into account pregnancy factors, the most prevalent Steno-Risk category was moderate in all groups except for nulliparous women, with 51.2% of the latter being classified as low risk (P < .001) (Fig. 1).

In the whole cohort, the prevalence of carotid plaque increased as the estimated cardiovascular risk increased (low risk: 11.7%; moderate risk: 38.4%; high risk: 68.2%; P for trend <.001). As shown in Fig. 1, similar results were found when the sample was stratified for sex and pregnancy factors, except for parous women with preeclampsia. In this group, no significant association was observed between Steno-Risk categories and the presence of carotid plaque (low risk: 25%; moderate risk: 45%; high risk: 53.3%; P = .251). We also investigated whether Steno-Risk predicted the presence of carotid plaque by sex. Steno-Risk showed good performance in identifying 1 or more carotid plaques regardless of sex and sex-specific reproductive factors (men: AUC .7651; nulliparous women: AUC .7988; women without preeclampsia: AUC .8353; women with preeclampsia: AUC .6206; P = .051) (Fig. 2). However, the AUC was significantly lower in the preeclampsia group compared with the other 3 groups for the prediction of the presence of 2 or more carotid plaques (men: AUC .7886; nulliparous women: AUC .9026; women without preeclampsia: AUC .8230; women with preeclampsia: AUC .7841; P = .042) (Fig. 2).

Finally, we evaluated the diagnostic performance of the Steno-Risk equation alone and with the addition of pregnancy factors to the model for the prediction of carotid plaque among women with type 1 diabetes. Neither the inclusion of parity (number of live births) nor preeclampsia in the Steno-Risk model led to a statistically significant increase in the AUC (Steno-Risk equation alone: AUC .7921 vs plus parity: AUC .7920, P = .969; Steno-Risk equation alone: AUC .7808 vs plus preeclampsia: .7887, P = .208) (Supplemental Table 2 (27)). There were also no significant differences in the diagnostic performance when stratifying for age or when evaluating the identification of 2 or more carotid plaques (Supplemental Table 3, Supplemental Table 4 (27)).

Discussion

In the present study, the inclusion of specific reproductive factors modified the association between preclinical carotid atherosclerosis and sex. Although the lowest risk of carotid plaque was observed in nulliparous women, there was no significant difference in carotid plaque burden between men and women with a history of preeclampsia. In fact, the diagnostic performance of the Steno-Risk scale for the identification of carotid plaque was lower in women with previous preeclampsia. However, the inclusion of either parity or preeclampsia in the Steno-Risk model did not improve the prediction of cardiovascular risk. To the best of our knowledge, no previous study has evaluated the role of parity and preeclampsia in the risk of carotid atherosclerosis in a type 1 diabetes population.

CVD in women has been widely underrecognized and underestimated (28). Indeed, women were less likely to be prescribed cardiovascular medication such as statins than men in primary prevention, both in the general population (29) and, as our results reinforced, in women with type 1 diabetes. In part, this could be explained by the fact that women were more often classified as having low/moderate cardiovascular risk according to international guidelines on CVD prevention (30). Thus, additional individual risk-enhancing clinical factors, such as sex-specific variables, should be taken into account (5, 22). In this context, the deleterious impact of preeclampsia on cardiovascular health years after pregnancy has been described (8, 9). Our group has previously shown a higher prevalence of carotid atherosclerosis burden when a history of preeclampsia was present among the type 1 diabetes population (10). Data from the present study, including almost 350 women with type 1 diabetes, confirmed the aforementioned results. The prevalence of carotid plaques observed was twice as high in the preeclampsia group compared with that of nulliparous women. Interestingly, the lowest odds for carotid plaque in women with type 1 diabetes compared with men disappeared when preeclampsia was taken into account. Although the higher absolute risk of CVD in men with type 1 diabetes compared with women has been largely described (31-33), no previous study has included obstetrics factors in the evaluation of sex disparities in CVD among the type 1 diabetes population. Overall, our results highlight that women with type 1 diabetes who developed preeclampsia during pregnancy should receive patient counseling about their future cardiovascular health and start preventive and therapeutic measures to reduce cardiovascular risk.

In the present study, nulliparous women with type 1 diabetes showed a better cardiovascular profile (ie, lower rates of hypertension and higher HDL cholesterol) but had higher rates of retinopathy compared with men. In addition to hypertension, poor glycemic control is the main modifiable risk factor for diabetic retinopathy (34). In our cross-sectional study, there was no significant difference in HbA1c levels. However, these data could not fully capture the glycemic control throughout diabetes duration. Indeed, parous women were more likely to use advanced diabetes technology such as continuous subcutaneous insulin infusion therapy (Table 1 of our study (35)). On the other hand, although it is known that pregnancy is associated with progression of diabetic retinopathy (36), follow-up studies comparing women with or without children found that the prevalence and progression of diabetic retinopathy occur less frequently in parous women compared with nulliparous women (37, 38). Taken together, it is possible that the intensive self-management behaviors adopted in pregnancy and the increased uptake of diabetes technologies could help women who have conceived to improve their glucose control compared with women who have never conceived. In the same line as CVRF, nulliparous women showed an approximately 50% decreased risk of carotid plaque presence compared with men. In the general population, multiparity has been associated with a higher risk of CVD (11, 39). Indeed, a recent Mendelian randomization study highlighted that the correlation between the number of live births and CVD is likely to be a direct causal link, beyond sociodemographic and clinical factors (12). Nonetheless, few studies have focused on women with type 1 diabetes (13, 40, 41). A cohort from the T1D Exchange Network and Clinic Registry, including 497 women with type 1 diabetes over a 5-year follow-up period, did not show differences between nulliparous (n = 469) and parous women (n = 28) in relation to either in microvascular complications or cardiovascular events (13). Compared with our study, those women were younger (aged < 30 years) with a shorter diabetes duration (median < 18 years), and assessment of subclinical carotid atherosclerosis was not carried out. On the other hand, our findings show that the “protective” effect of nulliparity varied across age groups. The lower risk of carotid plaque in nulliparous women compared with men was no longer observed among those aged >50 years. Previous large nationwide cohort studies have described a reduction in the sex difference in the absolute number of cardiovascular events after age 50 years (3, 33). In this context, our findings suggest that factors such as age, the onset of menopause (42), and other established risk factors such as glycemic control (33) may exert a greater influence on the risk of subclinical atherosclerosis at that age. This influence could potentially diminish the protective effect of parity.

As mentioned previously, the risk of carotid plaque compared with men decreased as exposure to obstetric factors diminished (parity and preeclampsia). However, adding this information to a validated calculator for cardiovascular risk, such as Steno-Risk, did not improve discrimination for the identification of subclinical atherosclerosis among women with type 1 diabetes. These results were in accordance with the previous large studies in the general population (18, 43). A hypothesis was that the incremental information provided by obstetric factors may have been partly captured by any subsequent increases in classical risk factors. However, despite its limitations, a recent Scientific Statement from the American Heart Association emphasized the importance of recognizing adverse pregnancy outcomes when CVD risk is evaluated in women (44). In this regard, we also observed that with the Steno-Risk scale the discrimination of women with previous preeclampsia was weaker than for the other 3 groups (men, nulliparous women, and parous women without preeclampsia). Indeed, the rate of carotid plaque was approximately 3 times higher in the low-risk group among women with previous preeclampsia in comparison to the other 3 groups. Overall, these results highlight that the cardiovascular risk classification obtained by the Steno-Risk scale among women with previous preeclampsia should be considered with caution, and other tools (such as carotid ultrasonography) should be applied for more accurate risk reclassification, especially in the low-risk category.

Our study has several strengths. This is the first study that assesses the presence of subclinical atherosclerosis by taking into account both parity and a history of preeclampsia in a cohort of subjects with type 1 diabetes. Second, we selected a large sample using standardized procedures to reduce variability, and thereby provide more accurate results. Last, the revision of all the medical records regarding gestational information avoided recall bias, a flaw frequently observed in previous retrospective studies (45). However, our study also has limitations. First, the study was carried out in a single tertiary hospital and only prespecified high-risk patients without CVD were included. Thus, extrapolation to other patients with a different setting or cardiovascular profile should be made with caution. Second, data of some reproductive factors were missing, such as age at menarche, age at first birth, or menopausal status. Despite this limitation, the inclusion of an easily collected (having children yes/no) variable into routine clinical practice could overcome the well-known recall bias of adverse obstetric outcomes in the cardiovascular risk reclassification (46). Third, there were no data about mild hypertensive disorders of pregnancy such as gestational hypertension. This could lead to the inclusion of women with higher cardiovascular risk in the control group (parous without preeclampsia). Despite this limitation, we found significant differences between groups highlighting the importance of the identification of preeclampsia in the cardiovascular risk reclassification in women with type 1 diabetes. Finally, the cross-sectional design does not allow assessment of the appearance of atherosclerotic disease or cardiovascular events. However, the presence of carotid plaque has been associated with incident CVD in large follow-up studies in the general population (47, 48). Furthermore, the predictive capacity for future cardiovascular events of carotid plaque has also been demonstrated in subjects with a coronary artery calcium score (the gold standard for CVD reclassification) of 0 (5, 22, 49). This is particularly interesting because women are more likely to have coronary artery calcium scores of 0 than men (50).

In conclusion, the present study highlights the importance of specific reproductive factors in the presence of preclinical carotid atherosclerosis, especially in younger women with type 1 diabetes. Furthermore, the presence of a history of preeclampsia identifies a subset of women in whom the diagnostic performance of a specific cardiovascular risk scale, such as Steno-Risk, was poor. Overall, both parity and preeclampsia are easily identifiable factors during routine clinical practice and thereby provide a window of opportunity for tailoring better strategies to prevent CVD in this high-risk population.

Funding

V.P. received a research grant from Fundació Docència i Recerca MútuaTerrassa, “Beca FMT d’Intensificació per a professionals de la Salud MT 2021.”

Disclosures

None of the funding sources played a role in the design, collection, analysis, or interpretation of the data or in the decision to submit the manuscript for publication.

Data Availability

Some or all datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

References

Abbreviations

 
• aOR

  adjusted odds ratio

 
• AUC

  area under the curve

 
• CVD

  cardiovascular disease

 
• CVRF

  cardiovascular risk factor

 
• eGFR

  estimated glomerular filtration rate

 
• HbA1c

  glycated hemoglobin

 
• HDL

  high-density lipoprotein

 
• IMT

  intima-media thickness

 
• LDL

  low-density lipoprotein

 
• OR

  odds ratio

 
• Steno-Risk

  Steno T1 Risk Engine