Exercise during pregnancy is associated with maternal health benefits including a 40% reduction in the risk of common pregnancy complications including gestational diabetes mellitus, gestational hypertension and preeclampsia, without increasing the risk of adverse fetal outcomes (1–5). This benefit may be conferred via positive cardiovascular adaptations associated with exercise. One of the most prominent hemodynamic changes during pregnancy is the reduction in systemic vascular resistance secondary to a peripheral vasodilation (6). Enhanced vasodilation during pregnancy has been attributed to an increase in flow-mediated dilation pathways (nitric oxide [NO] (7), as well as endothelium-derived hyperpolarization [EDH] (8)).
Flow-mediated dilation (FMD) is an index of endothelium-dependent vasodilation (9) and is widely used to predict cardiovascular morbidity and mortality (10). Pregnancy has been associated with both an improvement (11–13) or no change (14,15) in FMD responses compared with nonpregnant women. In nonpregnant populations, exercise has been associated with an overall enhancement of endothelial function by increasing NO- and EDH-mediated vasodilation (16,17) and improving reactive oxygen species modulation of vasodilation (18,19). However, the impact of moderate-to-vigorous physical activity (MVPA) on FMD during pregnancy has been minimally investigated (15,20), especially within the context of the current guidelines for physical activity during pregnancy (3,4). Although a study evaluating self-reported physical activity suggested that exercise is not associated with changes in FMD during pregnancy (15), a clinical trial in pregnant women found that a moderate-to-vigorous intensity aerobic exercise training intervention during pregnancy was associated with an improvement in FMD responses compared with pregnant women who undertook their usual physical activity (20). However, MVPA or “exercise” defines only one aspect of physical activity. At the opposite end of the physical activity spectrum, sedentary behavior is defined as “any waking behavior characterized by an energy expenditure ≤1.5 METs while in a sitting, lying, or reclining posture (21).” In nonpregnant populations, 6 to 8 h·d−1 of sedentary behavior has been associated with an increased risk of cardiovascular mortality and incidence of type 2 diabetes independent of physical activity (22). The role of sedentary behavior on FMD responses during pregnancy has not been examined. Understanding the potential influence of sedentary behavior on maternal cardiovascular health is very relevant and important because (a) only 15% of pregnant women adhere to the current guidelines for physical activity during pregnancy (23,24), which recommends women engage in at least 150 min·wk−1 of moderate-intensity exercise (3,4); (b) pregnant women are more sedentary and less active than nonpregnant populations, spending approximately 70% of waking hours sedentary (25); and (c) it has been shown that there is an increase in sedentary behavior with increasing gestational age (26).
It is now recognized that shear stress (the frictional force generated by blood flow) and shear pattern (anterograde, retrograde) are important mechanisms influencing endothelial function. Anterograde shear stress is associated with increased NO production (27), whereas retrograde shear stress resulted in an acute decreased endothelial function in males (28). Increased cardiac output and volume expansion can influence shear stress (29). Thus, one could speculate that pregnancy is associated with changes in shear patterns because of the cardiovascular adaptations that women undergo during pregnancy include increased cardiac output and blood expansion. However, the role of MVPA and sedentary behavior on anterograde and retrograde blood flow, velocity, and shear rate in pregnant populations has not been assessed.
Pregnancy is also associated with alterations in glucose metabolism leading to insulin resistance to accommodate the metabolic needs of the growing fetus (30). Women with impaired glucose metabolism and gestational diabetes have an impaired FMD response compared with normoglycemic pregnant women (31) which persists into the postpartum period (32). Physical activity has been suggested to improve vascular function by a combination of mechanisms (glucose metabolism improvement, increase in NO bioavailability (16)); however, this has not been examined during pregnancy.
With this as a background, the purpose of the present study was to determine the role of MVPA and sedentary behavior on FMD and glucose metabolism. We hypothesized that late pregnant women with objectively measured higher levels of MVPA and lower sedentary behavior would have an enhanced FMD response compared with pregnant women with lower MVPA activity levels and higher sedentary behavior. We further hypothesized that pregnant active women would have better glucose metabolism regulation than pregnant inactive women.
Seventy normotensive, euglycemic pregnant women in their third trimester were assessed. Participants arrived at the laboratory at 8:00 am after a 12-h fast. Caffeine and strenuous exercise were also avoided 12 h before testing. All procedures contributing to this work comply with the ethical standards set by the latest revision of the Declaration of Helsinki and had been approved by the Health Research Ethics Board at the University of Alberta (Pro00041144). Written informed consent was obtained from all participants.
Blood Sampling and Hemodynamic Monitoring
Blood samples were drawn from the left antecubital vein and participants were fed a light standardized meal consisting of a whole wheat bagel, jam (no sugar added), and a juice box. Blood samples were centrifuged, separated, and stored at −80°C until analysis. Glucose (hexokinase, Seimens Advia 1800), insulin (chemiluminescence microparticle immunoassay, Abbott Architect i2000), estradiol (electrochemiluminescence, Roche Cobas), progesterone (chemiluminescence competitive immunoassay, Siemens Centaur), and testosterone (two-site sandwich chemiluminescence, Siemens Centaur) were assessed. Homeostatic model assessment of insulin resistance (HOMA-IR) was calculated (fasting serum glucose × fasting serum insulin/22.5 (33).
Heart rate and blood pressure were measured continuously using a standard three-lead electrocardiogram based on lead II, and finger photoplethysmography (Finometer Pro; Finapres Medical Systems, Amsterdam, The Netherlands), respectively. Cardiac output was assessed using the Modelflow algorithm in the Finometer. Participants were seated in a semi-recumbent position in a dentist-style chair in a semi-darkened laboratory.
Flow-mediated Dilation Protocol
Our flow-mediated dilation protocol and analysis adhere with current guidelines for FMD (34). The right arm was positioned on a table at heart level. A blood pressure cuff attached to a manual inflation device was placed on the forearm distal to the elbow. Brachial artery diameter and velocity were recorded using Doppler Ultrasound (General Electric Health Care, USA) with a 60° insonation angle during 3 min of baseline, followed by a manual cuff inflation (5 min, 50 mm Hg above their systolic blood pressure) and release of the cuff (5 min).
Brachial artery blood flow velocity
Mean blood velocity was processed using qDAT (General Electric Health Care, USA), and recorded in labChart (ADInstruments, USA). Arterial blood velocity during baseline and after cuff release was averaged into 3-s bins. Mean blood velocity was used to calculate blood flow (πr2 × blood velocity × 60 [mL·min−1]; where r represents brachial artery radius) during baseline and after cuff release. Baseline brachial artery resistance was calculated as (mean arterial blood pressure/cardiac output) and baseline brachial artery conductance was calculated as (cardiac output/mean arterial blood pressure).
Brachial artery diameter
Brachial artery diameter (video saved as. AVI) was measured using an edge-detection software (Brachial Analyzer; Medical Imaging Applications, USA). Baseline, occlusion, and postocclusion diameters were measured frame by frame and then averaged into 3-s bins. Peak diameter and the time required to reach peak diameter were calculated.
As blood viscosity was not measured, we calculated shear rate as follows. Mean shear rate was calculated as 4 × mean blood velocity [cm·s−1]/diameter [mm]. The shear rate stimulus was calculated as the area under the curve (AUC) or the sum of shear rates from cuff release to peak dilation (average shear rate [s−1] × time to peak dilation [s]). In addition, anterograde shear rate (4 × anterograde blood velocity [cm·s−1]/diameter [mm]), retrograde shear rate (4 × retrograde velocity [cm·s−1]/diameter [mm]), and oscillatory shear index (absolute value of retrograde shear rate/absolute value of retrograde shear rate + anterograde shear rate) were calculated during baseline and after cuff release (34).
The FMD response was calculated as absolute change (mm) or percentage change.
This response was normalized by the shear stimulus (AUC), to take into consideration the stimulus magnitude.
Physical Activity Patterns
Participants were instructed to wear an ActiGraph accelerometer (model wGT3X-BT; ActiGraph LLC, Pensacola, FL) on their right hip during waking hours for seven consecutive days. In addition, participants documented their nonwear times in an activity log. Data were only included in the analysis if the total wear time was over 600 min·d−1 on at least 4 d per standard protocols (25). Accelerometers recorded accelerations over 60-s time intervals (epoch). Time spent sedentary, time spent engaging in light activity, and time spent engaging in MVPA was calculated using Freedson accelerometer count ranges (<100 counts per minute, 100–1951 counts per minute, and ≥1952 counts per minute, respectively). In addition, the time accumulated in bouts lasting more than 10 min in duration was also calculated. Data were analyzed using Actilife software (ActiGraph LLC, Pensacola, FL).
The Shapiro–Wilk test was used to assess the normality of the continuous data. According to the distribution of the variables, descriptive data were expressed as mean and SD. A multivariable regression analysis (least squares) between FMD responses, age, gestational age, prepregnancy body mass index (BMI), was run to determine which of the above mentioned variables should be considered as a covariate in our analysis (see Figure, Supplemental Digital Content 1, the analysis showed that age and gestational age should be covariates in our analysis, https://links.lww.com/MSS/B822). We used an allometric scaling technique to control for a potential influence of baseline diameter on the FMD result. (35). Briefly, a log-linked generalized linear model was applied where: the absolute change in FMD was the outcome, and the natural log of the baseline diameter was a covariate. A univariate analysis of variance to determine differences between active (or meeting 150 min MVPA) and inactive (those not meeting 150 min MVPA) women in all the parameters measured was carried out; for these analyses, age and gestational age were considered as covariates. Partial correlations controlling for age and gestational age were carried out between FMD responses and metrics of physical activity, glucose, insulin, HOMA-IR, and sex hormones within the groups. Statistical significance was defined as P < 0.05. Data were analyzed using IBM SPSS (IBM Analytics, USA). We have calculated the effect size (G*Power 188.8.131.52; Software, Germany) for the primary outcome as follows: with an effect size (F = 4.169), a total sample size of 70; 2 groups, 2 covariates, and numerator df = 1. Assuming an α of 0.05. The critical F value was 3.986.
Seventy pregnant women (31.6 ± 2.9 yr) in their third trimester [28–39 wk] were recruited. Following the 2019 Canadian Guideline for Physical Activity throughout Pregnancy (3–5), we categorized the women into active (≥150 min·wk−1) or inactive (<150 min·wk−1) according to their objective accelerometry data.
Active pregnant women (>150 min·wk−1 MVPA, n = 32) engaged in 266.7 ± 99.3 min·wk−1 MVPA, whereas inactive pregnant women (<150 min·wk−1 MVPA, n = 38) engaged in 76.1 ± 42.5 min·wk−1 MVPA (P < 0.0001; Table 1). Active pregnant women had higher DBP (F = 3.039, P = 0.03), higher glucose levels (F = 3.382, P = 0.02) and lower progesterone levels (F = 10.443, P < 0.0001) than inactive pregnant women (Table 1).
No differences among the groups were found regarding the brachial artery diameters at baseline, peak or occlusion (Table 1). Time to peak dilation, shear stimulus, absolute, and the percentage change in FMD were not different among the groups. The normalized FMD response, however, was higher in active pregnant women compared with inactive pregnant women (6.5 ± 4.4 vs 3.9 ± 3.5; F = 4.619; P = 0.005). Similarly, a significant difference in FMD between groups was observed when the allometric scaling of the FMD was performed (see Table, Supplemental Digital Content 2, https://links.lww.com/MSS/B823. Briefly, we generated a log-linked generalized linear model where the absolute change in FMD was the outcome, and the natural log of the baseline diameter was a covariate).
Active pregnant women had greater baseline retrograde velocity (F = 3.867) and baseline retrograde blood flow (F = 3.531), whereas their postocclusion mean shear rate (F = 3.254) and postocclusion anterograde shear rate (F = 3.230) were lower (Table 1).
In active pregnant women, the normalized FMD response, light activity or sedentary behavior were not correlated with any other variables (Table 2). The MVPA (min·wk−1) was positively correlated with systemic vascular resistance (r = 0.607, P = 0.02) and inversely correlated with insulin concentrations and HOMA-IR (Fig. 1). The MVPA bouts ≥10 min were inversely correlated with prepregnancy weight (r = −0.545, P = 0.04), prepregnancy BMI (r = −0.620, P = 0.01), current weight (r = −0.553, P = 0.03), current BMI (r = −0.660, P = 0.007), baseline blood vessel diameter (r = −0.525, P = 0.04), insulin concentrations, and HOMA-IR (Fig. 1).
In inactive pregnant women, the normalized FMD response was not correlated with any other variables (Table 3). Sedentary behavior, light activity, or MVPA were not correlated with glucose levels, insulin or HOMA-IR (Table 3). Interestingly, sedentary behavior was correlated with baseline retrograde flow and velocity as well as baseline oscillatory shear index, whereas the opposite was true for light activity (Fig. 2).
Our data demonstrate that women who meet or exceed current recommendations of 150 min·wk−1 MVPA have better endothelial function compared with women who do not engage in 150 min·wk−1 of MVPA. In active pregnant women, MVPA was inversely correlated with insulin concentrations, suggesting that physical activity also enhances glucose metabolism during pregnancy. Interestingly, the correlations between insulin concentrations and HOMA-IR and MVPA were stronger when the bouts of MVPA were ≥10 min, although this is not causative, it may suggest that glucose metabolism is enhanced when longer durations of physical activity are performed. Finally, we found that vascular responsiveness to the FMD test and glucose metabolism was not related to sedentary behavior in either active or inactive women.
Physical activity, FMD, and brachial artery flow
This study sought to address a knowledge gap related to the relationship between physical activity (light activity and MVPA), sedentary behavior, and endothelial function during pregnancy. Endothelial function is influenced by the direction (anterograde, retrograde) and magnitude of shear stress (36). Anterograde shear stress has been associated with an increase in NO production in endothelial cells (27), whereas retrograde shear rate acutely impairs endothelial function (28). Our results demonstrated that active pregnant women had a greater baseline retrograde blood flow and velocity with a higher diastolic blood pressure. Because during the third trimester of pregnancy, blood pressure returns to prepregnancy values, our findings may suggest that active pregnant women have less blood vessel compliance. Nonetheless, these findings need to be interpreted carefully because the magnitude of these differences may not be physiologically relevant.
High levels of sedentary behavior during pregnancy have been associated with excess gestational weight gain, reduced glucose tolerance, elevated cholesterol, and systemic inflammation during pregnancy, all of which would be expected to be detrimental to endothelial function (37,38). Contrary to our hypothesis, sedentary behavior was not found to be associated with detriments in glucose metabolism in our participants. Previous data in women at risk of developing gestational diabetes mellitus have shown that prolonged sedentary time was associated with higher fasting glucose levels (39), whereas in women with gestational diabetes, breaking sedentary behavior was associated with lower fasting and postprandial glucose levels (39). There were no differences between sedentary behavior in our active and inactive pregnant women. Moreover, our participants were otherwise healthy. Thus, given the nature of our cohort of participants (no history of cardiometabolic diseases, cross-sectional assessment during their third trimester), the likelihood of finding relationships between sedentary behavior and FMD and glucose may be lower than anticipated.
Sedentary behavior was not found to be associated with FMD responses in this sample of healthy pregnant women. These findings are aligned with previous findings in older healthy adults, where reducing sedentary time did not improve vascular endothelial function (40). However, in inactive women, greater sedentary time was associated with lower amounts of retrograde shear rate. Putting all the evidence derived from this cohort in context, there were no correlations between anterograde or mean flow and/or shear rate. In addition, no differences between the groups regarding baseline blood vessel diameter, vascular conductance or resistance were found. Hence, given that the order of magnitude of anterograde flow is (5×) higher than the retrograde flow. We believe that this finding is minimally relevant to perfusion.
Finally, physical activity has been associated with enhanced vascular function in nonpregnant populations (16–19); however, this association is less clear in pregnant women. Our data align with results from Ramirez-Velez et al. (20), who showed that MVPA during pregnancy improves endothelial function (increased FMD) in late pregnancy. This was observed after normalizing the data by the shear stimulus; indicating that this is a critical factor to account for when interpreting FMD results. This is particularly relevant in pregnant populations where increases in blood volume and cardiac output, concurrent with a functional reduction in hematocrit may all influence shear stress.
The discrepancies between our findings and Ramirez-Velez et al., with those from Weissgerber et al. (15), could be explained by a) the authors quantified physical activity using an activity survey. b) Because shear rate responses, FMD response, and diameters were not different among the pregnant groups (active vs inactive). c) Authors acknowledge the sample size as a limiting factor in the study.
The current Canadian Guideline for Physical Activity throughout pregnancy recommends women engage in at least 150 min·wk−1 of moderate-intensity exercise (3,4). Women who follow these recommendations have a reduction in the risk of developing hypertensive disorders of pregnancy by approximately 20% (1). One of the potential mechanisms that can explain this would be an improvement of vascular function. Even though we did not follow-up our pregnant women, our cross-sectional data suggest that exercise indeed improves vascular function in this population.
Physical activity, hormones, and cardiometabolic adaptations to pregnancy
Pregnant women undergo several cardiometabolic adaptations to meet the increased metabolic demands of the growing fetus. These adaptations include insulin resistance, and increased glucose utilization, heart rate, stroke volume, and cardiac output while maintaining normal blood pressure (41). Even though we found that fasting glucose level was slightly higher in active women, the difference between the groups was small (0.2 mmol·L−1) and there was no difference in insulin concentrations or HOMA-IR. Nonetheless, MVPA during pregnancy was inversely correlated with fasting insulin concentrations and HOMA-IR only in active pregnant women; suggesting that higher levels of physical activity may influence insulin control in these women. Moreover, the correlations were stronger when MVPA was analyzed as bouts ≥10 min showing that planned physical activity may be better to improve glucose metabolism.
Physical activity patterns appear to have an important influence on sex hormones. Previous data have shown that in healthy, sedentary eumenorrheic women, an aerobic exercise program for 16 wk (MVPA, 150 min·wk−1) decreased progesterone levels (42). This is similar to what we observed between our groups of women. However, there were no associations between vascular function and hormone concentrations. Although there is evidence that progesterone may influence vascular tone (43), the magnitude of the observed difference was small, and the physiological relevance remains unclear.
We found that 46% of our sample met the guidelines for exercise during pregnancy, whereas nationally representative cohorts have suggested that only 9% to 15% of women meet these guidelines (23,24). We recruited our participants through posters located in the university areas and through social media posts. Most of the women recruited in our study come from the university area, therefore, a recruitment bias could explain why our population has double the rate of compliance with the guidelines.
Physical activity and sedentary behavior are challenging to quantify, especially during pregnancy (44). Thus, a strength of our study is the use of objectively measured physical activity and sedentary behavior. Although accelerometers do not measure posture (standing vs sitting); a defining feature of sedentary behavior; our accelerometry derived measure of sedentary behavior was similar to data derived from the only two studies using activPAL devices in pregnant women (39,45). Similar to results from Wagnild et al., (39) women in our study spent approximately 9.4 ± 1.4 h·d−1 being sedentary, whereas women from the Di Fabio et al., (45) study spent 12.9 ± 2.2 h·d−1 in sedentary behavior. However, future studies assessing the relationship between sedentary behavior and FMD should consider the use of an inclinometer.
In active women, exercise enhances vascular function (FMD responses). Moreover, greater levels of activity may also further benefit glucose metabolism. Combined these data demonstrate the cardiovascular benefits of meeting current guidelines for MVPA during pregnancy (3–5).
The authors would like to thank Miss Marina James for her technical assistance. This research has been funded by generous supporters of the Lois Hole Hospital for Women through the Women and Children’s Health Research Institute (WCHRI, RES0018745). L. M. R. is funded by the Molly Towell Perinatal Research Foundation (RES0041143). S. M. F. was supported by a WCHRI Summer Studentship (2414). R. J. S. is funded by the Canadian Institutes for Health Research, WCHRI Doctoral Research Award (GSD-146252) and Alberta Innovates Graduate Studentship (RES042403). S. T. D. is a Canada Research Chair in Maternal and Perinatal Cardiovascular Health. C. D. S. and M. H. D. are supported by a Heart and Stroke Foundation of Canada Grant in Aid (G-16-00014033). M. H. D. is funded by a Heart & Stroke Foundation of Canada (HSFC)/Health Canada Improving Heart Health for Women Award, National and Alberta HSFC New Investigator Award. (HSFC NNIA Davenport), and NSERC discovery grant (RES0043852). C. D. S. is funded by a HSFC Joint National and Alberta New Investigator Award (HSFC NNIA Steinback).
Conflict of interest: The authors report no conflicts of interest and that the funding sources did not play a role in the design, collection, analysis, and interpretation of the data; in the writing of the manuscript and the decision to submit the manuscript for publication. The results of the present study do not constitute endorsement by ACSM. The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.
1. Davenport MH, Ruchat SM, Poitras VJ, et al. Prenatal exercise for the prevention of gestational diabetes mellitus and hypertensive disorders of pregnancy: a systematic review and meta-analysis. Br J Sports Med
2. Davenport MH, Kathol AJ, Mottola MF, et al. Prenatal exercise is not associated with fetal mortality: a systematic review and meta-analysis. Br J Sports Med
3. Mottola MF, Davenport MH, Ruchat SM, et al. No. 367-2019 Canadian guideline for physical activity throughout pregnancy. J Obstet Gynaecol Can
4. Mottola MF, Davenport MH, Ruchat SM, et al. 2019 Canadian guideline for physical activity throughout pregnancy. Br J Sports Med
5. Mottola MF, Davenport MH, Ruchat SM, et al. N° 367-2019 Lignes Directrices Canadiennes Sur L’activité Physique Durant La Grossesse. J Obstet Gynaecol Can
6. Duvekot JJ, Cheriex EC, Pieters FA, Menheere PP, Peeters LH. Early pregnancy changes in hemodynamics and volume homeostasis are consecutive adjustments triggered by a primary fall in systemic vascular tone. Am J Obstet Gynecol
7. Williams DJ, Vallance PJ, Neild GH, Spencer JA, Imms FJ. Nitric oxide-mediated vasodilation in human pregnancy. Am J Physiol
. 1997;272(2 Pt 2):H748–52.
8. Morton JS, Davidge ST. Arterial endothelium-derived hyperpolarization: potential role in pregnancy adaptations and complications. J Cardiovasc Pharmacol
9. Thijssen DH, Black MA, Pyke KE, et al. Assessment of flow-mediated dilation in humans: a methodological and physiological guideline. Am J Physiol Heart Circ Physiol
10. Green DJ, Jones H, Thijssen D, Cable NT, Atkinson G. Flow-mediated dilation and cardiovascular event prediction: does nitric oxide matter? Hypertension
11. Dorup I, Skajaa K, Sorensen KE. Normal pregnancy is associated with enhanced endothelium-dependent flow-mediated vasodilation
. Am J Physiol
. 1999;276(3 Pt 2):H821–5.
12. Sierra-Laguado J, Garcia RG, Lopez-Jaramillo P. Flow-mediated dilatation of the brachial artery in pregnancy. Int J Gynaecol Obstet
13. Savvidou MD, Donald AE, Nicolaides KH. Assessment of endothelial function in normal twin pregnancy. Ultrasound Obstet Gynecol
14. Quinton AE, Cook CM, Peek MJ. A longitudinal study using ultrasound to assess flow-mediated dilatation in normal human pregnancy. Hypertens Pregnancy
15. Weissgerber TL, Davies GA, Tschakovsky ME. Brachial artery flow-mediated dilation is not affected by pregnancy or regular exercise participation. Clin Sci (Lond)
16. Green DJ, Maiorana A, O’Driscoll G, Taylor R. Effect of exercise training on endothelium-derived nitric oxide function in humans. J Physiol
. 2004;561(Pt 1):1–25.
17. Gunduz F, Kocer G, Ulker S, Meiselman HJ, Baskurt OK, Senturk UK. Exercise training enhances flow-mediated dilation in spontaneously hypertensive rats. Physiol Res
18. Rush JW, Turk JR, Laughlin MH. Exercise training regulates SOD-1 and oxidative stress in porcine aortic endothelium. Am J Physiol Heart Circ Physiol
19. Adams V, Linke A, Krankel N, et al. Impact of regular physical activity on the NAD(P)H oxidase and angiotensin receptor system in patients with coronary artery disease. Circulation
20. Ramirez-Velez R, Aguilar de Plata AC, Escudero MM, et al. Influence of regular aerobic exercise on endothelium-dependent vasodilation and cardiorespiratory fitness in pregnant women. J Obstet Gynaecol Res
21. Tremblay MS, Aubert S, Barnes JD, et al. Sedentary behavior research network (SBRN)—terminology consensus project process and outcome. Int J Behav Nutr Phys Act
22. Patterson R, McNamara E, Tainio M, et al. Sedentary behaviour and risk of all-cause, cardiovascular and cancer mortality, and incident type 2 diabetes: a systematic review and dose response meta-analysis. Eur J Epidemiol
23. Evenson KR, Wen F. Prevalence and correlates of objectively measured physical activity and sedentary behavior among US pregnant women. Prev Med
24. Santo EC, Forbes PW, Oken E, Belfort MB. Determinants of physical activity frequency and provider advice during pregnancy. BMC Pregnancy Childbirth
25. Sobierajski FM, Purdy GM, Usselman CW, et al. Maternal physical activity is associated with improved blood pressure regulation during late pregnancy. Can J Cardiol
26. Duncombe D, Wertheim EH, Skouteris H, Paxton SJ, Kelly L. Factors related to exercise over the course of pregnancy including women’s beliefs about the safety of exercise during pregnancy. Midwifery
27. Kuchan MJ, Frangos JA. Role of calcium and calmodulin in flow-induced nitric oxide production in endothelial cells. Am J Physiol
. 1994;266(3 Pt 1):C628–36.
28. Thijssen DH, Dawson EA, Tinken TM, Cable NT, Green DJ. Retrograde flow and shear rate acutely impair endothelial function in humans. Hypertension
29. Califano JP, Reinhart-King CA. Exogenous and endogenous force regulation of endothelial cell behavior. J Biomech
30. Catalano PM, Tyzbir ED, Roman NM, Amini SB, Sims EA. Longitudinal changes in insulin release and insulin resistance in nonobese pregnant women. Am J Obstet Gynecol
. 1991;165(6 Pt 1):1667–72.
31. Paradisi G, Biaggi A, Ferrazzani S, De Carolis S, Caruso A. Abnormal carbohydrate metabolism during pregnancy : association with endothelial dysfunction. Diabetes Care
32. Davenport MH, Goswami R, Shoemaker JK, Mottola MF. Influence of hyperglycemia during and after pregnancy on postpartum vascular function. Am J Physiol Regul Integr Comp Physiol
33. Singh B, Saxena A. Surrogate markers of insulin resistance: a review. World J Diabetes
34. Thijssen DHJ, Bruno RM, van Mil A, et al. Expert consensus and evidence-based recommendations for the assessment of flow-mediated dilation in humans. Eur Heart J
35. Atkinson G, Batterham AM. Allometric scaling of diameter change in the original flow-mediated dilation protocol. Atherosclerosis
36. Cheng JL, Au JS, MacDonald MJ. Peripheral artery endothelial function responses to altered shear stress patterns in humans. Exp Physiol
37. Gollenberg AL, Pekow P, Bertone-Johnson ER, Freedson PS, Markenson G, Chasan-Taber L. Sedentary behaviors and abnormal glucose tolerance among pregnant Latina women. Med Sci Sports Exerc
38. Fazzi C, Saunders DH, Linton K, Norman JE, Reynolds RM. Sedentary behaviours during pregnancy: a systematic review. Int J Behav Nutr Phys Act
39. Wagnild JM, Hinshaw K, Pollard TM. Associations of sedentary time and self-reported television time during pregnancy with incident gestational diabetes and plasma glucose levels in women at risk of gestational diabetes in the UK. BMC Public Health
40. Suboc TB, Knabel D, Strath SJ, et al. Associations of reducing sedentary time with vascular function and insulin sensitivity in older sedentary adults. Am J Hypertens
41. Davenport MH, Skow RJ, Steinback CD. Maternal responses to aerobic exercise in pregnancy. Clin Obstet Gynecol
42. Smith AJ, Phipps WR, Arikawa AY, et al. Effects of aerobic exercise on premenopausal sex hormone levels: results of the WISER study, a randomized clinical trial in healthy, sedentary, Eumenorrheic women. Cancer Epidemiol Biomarkers Prev
43. Dos Santos RL, Da Silva FB, Ribeiro RF, Stefanon I. Sex hormones in the cardiovascular system. Horm Mol Biol Clin Invest
44. Chasan-Taber L, Evenson KR. Next steps for measures of physical activity during pregnancy. Matern Child Health J
45. Di Fabio DR, Blomme CK, Smith KM, Welk GJ, Campbell CG. Adherence to physical activity guidelines in mid-pregnancy does not reduce sedentary time: an observational study. Int J Behav Nutr Phys Act