Pregnancy has been described as a stress test that can identify women at risk of future chronic disease [
The development of endothelial dysfunction is an early pathologic event in the natural history of CVD. Endothelial function can be assessed noninvasively using the ultrasound-based technique of flow-mediated dilatation (FMD) [
This study was conducted in the setting of an ongoing research program, in which pregnant women are recruited at the time of GDM screening and then followed longitudinally in the years thereafter. For this study, women were recruited in late second or early third trimester of pregnancy and all underwent a 3-hour, 100 g oral glucose tolerance test (OGTT) for ascertainment of gestational glucose tolerance status. This OGTT classified gestational glucose tolerance status as either (i) GDM (as defined by 2 or more glucose values on the OGTT meeting National Diabetes Data Group (NDDG) criteria [
Subject preparation for brachial artery FMD studies followed established guidelines [
On the morning of the vascular study, participants completed an interviewer-administered questionnaire that addressed personal and family medical history, use of medications (including vitamins and contraceptives), smoking history, physical activity, and menstrual patterns. Height and weight were measured by medical scale. Anatomical waist, iliac waist, and trochanteric hip were each measured twice with mean values calculated for analyses.
All FMD studies were conducted between 07:00–11:00 AM in a quiet, dimly lit, temperature-controlled room (24–26°C). Sitting blood pressure (BP) was measured from the left arm (Baumanometer wall unit 33, W.A. Baum Co Inc., Copiague, NY) at the beginning of the test. Subjects then lay supine and the left arm BP was again recorded with pulse rate. After >10 minutes of rest, ECG electrodes (Medi-Trace 530 Series, Ludlow Technical Products Ltd., Gananoque, ON, Canada) were attached. A Versa form Pillow (
Setup of FMD studies showing proximal manual probe holding technique and distal cuff position.
All brachial artery imaging were performed using a non-ECG gated Vivid Q GE ultrasound machine (GE Healthcare, Mississauga, ON, Canada), with manual probe placement by an experienced ultrasonographer who had performed >100 brachial ultrasound scans prior to this study. The ultrasonographer was blinded to gestational and current glucose tolerance status of the participants. All scans were continuous, employing an insonation angle of <60° and captured both brachial artery diameter and velocity measurements. Three separate scans were recorded per subject. A resting scan captured 1 minute of baseline brachial artery diameter and velocity data. Then, the cuff was inflated (Hokanson DS400 manual cuff inflator, Bellevue, Seattle, WA) to systolic BP >200 mmHg for 5 minutes. Scanning was then resumed in the final 10–20 seconds prior to cuff deflation and continued for 200 seconds. The final recovery scan captured the last 2 minutes of data. All data were stored using B-mode ultrasound on the echocardiography machine, resulting in a 15-second interruption in data acquisition during storage. After all images were obtained, supine and sitting BP were again recorded, and the average of pre- and poststudy data measurements was determined.
The following five measures of vascular function were assessed. FMD of the brachial artery was the primary vascular function outcome of interest. FMD in response to the hyperemia that occurs following cuff deflation reflects endothelium-dependent vasodilation. Specifically, hyperemia provides a mechanical stimulus (shear stress) for increased endothelial production of nitric oxide, which in turn diffuses into the underlying vascular smooth muscle cells and stimulates their production of cyclic guanosine monophosphate, which leads to vasodilation. Historically, FMD has been measured between 45 and 60 seconds after cuff deflation, and an impaired FMD at this time, that is, Peak shear rate is a surrogate of peak shear stress [ Reactive hyperemia measures the downstream dilatation of the microvasculature in response to ischemia and generates the shear stress stimulus for FMD [ Brachial artery diameter is a determinant of FMD [
Following image acquisition and storage, analysis of the brachial artery diameter was performed using near- and far-wall detection by customized edge detection software by Medical Imaging Applications (MI.A Vascular Tools 5, Brachial Analyzer, Coralville, Iowa) (Figure
(a) Image of a longitudinal view of a brachial artery analyzed by customized edge detection software producing diameter measurements. (b) Velocity data for the brachial artery at (i) rest and (ii) hyperemia on cuff deflation.
FMD60 was calculated using the formula:
where
The velocity measurements were calculated manually from the time-velocity integral software of the Vivid Q machine. For the rest scans, the average of 4 traced velocity envelopes was used to obtain the time-averaged mean (TAmeanrest) velocity. For the hyperemic phase, the first envelope after cuff deflation was omitted, and the next 15 envelopes were traced. The average of these values was calculated and used as the TAmeanhyperemia.
The blood flow (BF) at rest and hyperemia was calculated using the following formula:
Reactive hyperemia (RH) was calculated by the following formula:
Shear rate was calculated according to the following formula:
In our laboratory, the interobserver coefficient of variation for FMD was 20.3% and the intraobserver coefficient of variation was 11.2%. These were calculated as the difference between the paired values divided by the mean and divided by
All analyses were conducted using SAS version 9.1 (SAS Institute, Cary, NC). Continuous variables were tested for normality of distribution, and transformations of skewed variables were used, where necessary, in subsequent analyses. The study participants were stratified into 2 groups based on their glucose tolerance status in pregnancy: (i) those with normal glucose tolerance in pregnancy and (ii) those with GDM or GIGT. Univariate differences between the 2 groups were assessed by Wilcoxon two-sample test for continuous variables and
Brachial artery FMD studies were performed on 158 women between May 2011 and June 2012 (Figure
Schematic showing disposition of the study population.
The mean time since the index pregnancy was ~6 years for both groups (Table
Characteristics of study population stratified by glucose tolerance status in pregnancy.
Characteristic | Gestational glucose tolerance status |
| |
---|---|---|---|
NGT ( |
GIGT/GDM ( |
||
Age (years) |
|
|
0.91 |
Caucasian/noncaucasian | 47/12 | 44/14 | 0.62 |
Time from delivery to FMD study (years) |
|
|
0.59 |
Time from delivery to OGTT (years) |
|
|
0.81 |
Body mass index (kg/m2) |
|
|
0.12 |
Waist: hip ratio |
|
|
0.37 |
Oral glucose tolerance test: | |||
Fasting glucose (mmol/L) |
|
|
0.13 |
2-hour glucose (mmol/L) |
|
|
0.05 |
AUCgluc |
|
|
0.01 |
Prediabetes or diabetes ( |
14 | 19 | 0.28 |
Mean seated systolic BP (mmHg) |
|
|
0.50 |
Mean supine systolic BP (mmHg) |
|
|
0.15 |
Mean seated diastolic BP (mmHg) |
|
|
0.03 |
Mean supine diastolic BP (mmHg) |
|
|
0.04 |
Heart rate (beats per minute) |
|
|
0.31 |
Menstrual state (Follicular/Luteal) | 39/18 | 37/16 | 0.87 |
Oral contraception (N/Y) | 53/6 | 48/10 | 0.27 |
Smoker (N/Y) | 57/2 | 54/4 | 0.39 |
Flow-mediated dilatation % |
|
|
0.61 |
Flow-mediated dilatation60 % |
|
|
0.33 |
Baseline diameter (mm) |
|
|
0.66 |
Peak shear rate (s−1) |
|
|
0.32 |
Reactive hyperemia % |
|
|
0.07 |
Data are presented as mean ± standard deviation. N: no, Y: yes.
On unadjusted comparison (Table
Mean adjusted levels of vascular measures from gestational glucose tolerance group after adjustment for age, years since delivery, ethnicity, current BMI, mean supine DBP, current glucose intolerance, smoking history, and menstrual status.
Vascular measure | Gestational glucose tolerance status |
| |
---|---|---|---|
NGT ( |
GIGT/GDM ( |
||
Flow-mediated dilatation (%) |
|
|
0.19 |
Flow-mediated dilatation60 (%) |
|
|
0.09 |
Baseline diameter (mm) |
|
|
0.60 |
Peak shear rate (s−1) |
|
|
0.80 |
Reactive hyperemia (%) |
|
|
0.14 |
Adjusted data are presented as mean ± standard error.
The unadjusted and adjusted comparisons of all 5 vascular measures were repeated after further stratifying the GDM/GIGT group into the 38 women with GDM and the 20 women with GIGT. Again, with this approach, there were no significant differences between the NGT, GIGT, and GDM groups in FMD, FMD60, baseline diameter, peak shear rate, and reactive hyperemia (data not shown).
Finally, multiple linear regression analyses were performed to identify independent determinants of the vascular function measures amongst the following covariates: age, gestational glucose intolerance, years since delivery, ethnicity, current BMI, supine DBP, current glucose intolerance, smoking, and menstrual cycle status. On these analyses, Caucasian ethnicity emerged as a negative predictor of baseline diameter (
In light of the absence of differences in FMD between the NGT and GIGT/GDM groups, we performed post hoc power calculations based on the sample size. These calculations showed that a sample size of 117 participants would provide 82% power to declare no difference in FMD between study groups at a 5% significance level, under the assumption of 2.1% as the largest clinically significant difference in FMD between groups. Given the observed difference of FMD between the two groups of 0.8 and standard deviation of 3.8 with the sample size of
In this study, we demonstrate that, when assessed at 6 years postpartum, women with previous gestational dysglycemia exhibit similar FMD to that seen in women who maintained normal glucose tolerance in pregnancy. Furthermore, these two groups of women did not differ with respect to other vascular function measures, including FMD60, baseline arterial diameter, peak shear rate, and reactive hyperemia. It thus appears that, despite their long-term cardiovascular risk, women with glucose intolerance in pregnancy do not show evidence of endothelial dysfunction at 6 years postpartum.
Previous studies of FMD in women with a history of gestational dysglycemia have yielded inconsistent findings. In a study of 40 women assessed at 2 months postpartum, Davenport et al. found that brachial artery FMD was lower in women with previous GDM (
In this context, the current study was specifically designed to address this unresolved question of endothelial function in women with previous gestational dysglycemia, while accounting for these limitations. First, the sample size of this study is more than 2-fold greater than that of the previous reports. Second, participants underwent detailed clinical characterization, including prospective ascertainment of glucose tolerance status both during pregnancy and again when undergoing the vascular studies ~6 years later. Moreover, this phenotypic detail enabled application of strict exclusion criteria that were designed to limit the influence of confounding factors (e.g., as shown in Figure
Supported by these strengths in design, the current study demonstrates that women with and without previous glucose intolerance in pregnancy have similar FMD, FMD60, baseline arterial diameter, peak shear rate, and reactive hyperemia at 6 years postpartum. This FMD result supports the findings of Hannemann and colleagues at 5 years postpartum, but differs from the two other reports at 2–6 months after delivery. We believe that it is unlikely that time since delivery is a direct determinant of endothelial function, particularly since this covariate was not independently associated with any of the vascular function outcomes on multiple linear regression analyses. However, it remains possible that a factor specific to the early postpartum could result in abnormalities in FMD in patients with GDM. One such consideration would be breastfeeding, which has been associated with reduced long-term cardiovascular risk [
In addition, the clinical implications of these data are that FMD measurement may not be a sensitive or useful method for identifying those women with previous glucose intolerance who are at the highest risk of future CVD. Unlike in men [
A limitation of this study is that the assessment of vascular function at a single point in time (i.e., 6 years postpartum) does not provide insight on the potential for differential longitudinal changes in endothelial function that may emerge over time in the study groups. Similarly, the effect over time of incident prediabetes/diabetes on the relationship between gestational dysglycemia and vascular function also remains to be established. Indeed, on multiple linear regression analyses, current glucose intolerance nearly reached statistical significance as an independent negative predictor of FMD60 (
In summary, at 6 years postpartum, women with previous glucose intolerance in pregnancy do not exhibit impairment of FMD as compared to their peers. Furthermore, these two groups of women also do not differ with respect to other measures of vascular function, including FMD60, baseline arterial diameter, peak shear rate, and reactive hyperemia. Thus, despite their long-term cardiovascular risk, women with glucose intolerance in pregnancy do not show evidence of endothelial dysfunction 6 years later. Future investigation should be directed at identification and validation of alternate biomarkers of subsequent vascular risk that may be evident in such women before the development of diabetes, metabolic syndrome, or cardiovascular events.
The authors declare that they have no conflict of interests.
This work was supported by operating grants from the Canadian Institutes of Health Research (CIHR) (Grant no. MOP 84206) the Heart and Stroke Foundation of Ontario (HSFO) (Grant no. NA 6747) and by a philanthropic donation in memory of Bram and Bluma Appel. S. Brewster was supported by Mount Sinai Hospital, Department of Medicine Postdoctoral Fellowship. J. S. Floras holds the Canada Research Chair in Integrative Cardiovascular Biology. B. Zinman holds the Sam and Judy Pencer Family Chair in Diabetes Research at Mount Sinai Hospital and University of Toronto. R. Retnakaran holds an Ontario Ministry of Research and Innovation Early Researcher Award.