Background: High parity is associated with greater cardiovascular disease (CVD) among mid-life and older women. Prospective studies of arterial change throughout pregnancy are needed to provide insight into potential mechanisms. This study assessed vascular adaptation across pregnancy in healthy first-time pregnant women. Methods: The Maternal Vascular Adaptation to Healthy Pregnancy Study (Pittsburgh, PA, 2010–2015) assessed 37 primigravid women each trimester, 6–8 weeks after delivery and 1–5 years postpartum, with B-mode ultrasound imaging of common carotid artery (CCA) intima-media thickness (IMT) and inter-adventitial diameter (IAD) to assess associations with physical and cardiometabolic measures. Results: Thirty-seven women (age 28.2 ± 4.5 years, pre-pregnant BMI 24.4 ± 3.2 kg/m ) experienced uncomplicated pregnancies. After adjustment for age and pre-pregnancy BMI, mean (SE) IAD (mm) increased each trimester, from 6.38 (0.08) in the 1st trimester to 6.92 (0.09) in the 3rd trimester, and then returned to 1st trimester levels postpartum (6.35 [0.07], P < 0.001). In contrast, mean (SE) CCA IMT (mm) increased from the 2nd trimester (i.e., 0.546 [0.01]) onward, and remained higher at an average of 2.7 years postpartum (0.581 [0.02], P = 0.03). Weight partially explained changes in IAD. Conclusions: In uncomplicated first pregnancies, IAD increased and returned to 1st trimester levels postpartum. In contrast, CCA IMT remained increased 2 years postpartum. Maternal weight explained vascular changes better than did metabolic changes. Increased postpartum CCA IMT may persist and contribute to long-term CVD risk. Keywords: Common carotid artery intima-media thickness, Inter-adventitial diameter, Pregnancy, Cardiovascular disease, Vascular remodeling Background risk [1–4]. For example, either weight gain or the athero- High parity is associated with greater cardiovascular dis- genic metabolic changes of pregnancy may instigate per- ease (CVD) risk in women . Although some of this sistent unhealthy vascular changes [5, 6]. However, studies risk may be due to socio-economic status and lifestyle that could illuminate these relationships have been limited factors associated with greater parity, acute physiologic by 1) sample sizes inadequate to detect significant differ- changes during pregnancy also may contribute to CVD ences in vessel measures [7, 8], 2) failure to collect serial arterial measures , 3) use of non-standard techniques to * Correspondence: email@example.com assess the vasculature [5, 9], 4) short follow-up [7, 8], and Department of Epidemiology, Graduate School of Public Health, University 5) lack of biomarker collection across the pregnancy cycle of Pittsburgh, 130 De Soto Street, Pittsburgh, PA 15261, USA 2 [5–10]. Department of Health Promotion and Development, School of Nursing, University of Pittsburgh, 3500 Victoria Street, 440 Victoria Building, Pittsburgh, Structural arterial changes during pregnancy can be PA 15261, USA assessed using B-mode ultrasonography of the carotid Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 2 of 9 artery, a well-established, non-invasive, reproducible normal weight newborns in the analytic sample for our technique . Abnormal values of two measures of ar- analysis. terial structure—greater intima-media thickness (IMT) These participants were invited to return for a and inter-adventitial diameter (IAD) of the common follow-up visit 1–5 years after their first postpartum carotid artery (CCA)—are associated with greater CVD visit. Fourteen had moved out of the region and were risk factor burden [12–14], arterial aging , and higher unable to participate. Participants (i.e., five women) were incidence of CVD [13, 16, 17]. The normal changes that excluded if they were pregnant or if they had given birth occur in the CCA IMT and IAD during and after a healthy within the previous 4 months, which generated seven- pregnancy have not been well established. teen potential participants. Of these seventeen, fourteen The primary objective of our Maternal Vascular experienced uncomplicated first pregnancies and were, Adaptation to Healthy Pregnancy (MVP) study was to therefore, included in our analysis. These follow-up visits assess vascular changes in normal first pregnancies, occurred between 2014 and 2015. Participants signed an using an adequate sample size, serial measures, a stan- informed consent document approved by the University dardized technique to assess vasculature, and including of Pittsburgh, Human Research Protection Office. collection of biomarkers. We hypothesized that the vas- culature would transiently adapt to the increased blood Carotid artery measures volume and metabolic requirements of healthy preg- Carotid ultrasounds were performed by a trained re- nancy, and that these adaptations would be associated search vascular sonographer from the University of with pregnancy weight gain and changes in levels of car- Pittsburgh, Ultrasound Research Laboratory (URL). Par- diometabolic factors. ticipants were placed supine, with a right hip wedge for comfort if necessary, and the common carotid artery Methods was scanned bilaterally with high-resolution B-mode Study design and population ultrasound (ACUSON Cypress System, Malvern, PA.) The MVP study prospectively assessed common carotid Digitized images of the common carotid artery were ob- artery measures in a cohort of healthy primigravid tained at end diastole, 1 cm proximal to the carotid bulb, women. Eligible participants recruited from the commu- and IMT was measured as the distance from the nity were healthy, non-smoking primigravid women, media-adventitial interface to the intima-lumen interface aged ≤40 years, at less than 38 weeks of gestational age. of both the near and far wall of the artery. Approxi- Exclusion criteria were the following: 1) vasoactive medi- mately 140 measurements of thickness were made for cation use; 2) infertility history—defined as either experi- each 1-cm segment, and the mean of each segment was encing a period of at least 12 months marked by the calculated. IMT reported represents the mean value for inability to achieve pregnancy or using fertility medica- near and far wall bilaterally. IAD was measured as dis- tions to achieve pregnancy; 3) family history of prema- tance from the adventitial-medial interface of the near ture coronary artery disease; 4) previous abortion; 5) arterial wall to the media-adventitial interface of the multiple gestation. arterial wall using the same CCA segment. Images Study visits were scheduled at 12–14, 24–26, and 36– were read by one reader, using a computerized, 38 weeks of pregnancy, and 6–8 weeks postpartum; all semi-automated reading program system . Repro- visits were conducted between in 2010 to 2013. After ducibility of carotid measures at the URL was excel- telephone screening for eligibility, women began the lent during thetimeperiodof the study, with an study at any one of the pregnancy visits. Each visit in- intraclass correlation coefficient within reader of over volved physical measures (e.g., height and weight) and 0.91 for CCA IMT and over 0.99 for IAD. ultrasound measures of the carotid artery. We calculated that 31 women were needed as participants to generate Demographic, pregnancy history, physical, and laboratory 80% power to detect a 0.5 SD difference for change in measures CCA IMT and IAD given an assumed 0.5 correlation At the initial visit, participants completed a among the repeated observations. Because we estimated self-administered demographic form. Research staff 1) that 1) 10–20% of women develop a pregnancy compli- measured the height of participants using a stadiometer cation and 2) our study would experience 25% attrition, and 2) weighed the participants on a standard balance we targeted recruitment of 46. The study enrolled 44 scale. The mean value of two readings for each measure women, of whom 43 had multiple visits, and six devel- was recorded. Pre-pregnancy weight was identified pref- oped pregnancy complications (one preeclampsia; 3 ges- erentially as the pre-pregnancy weight documented in tational hypertension; 2 preterm births, one of which the prenatal record or, if not available, as a documented had a placental abruption), which left 37 participants weight in the medical record in the 3 months prior to with uncomplicated pregnancies and full term births of the last menstrual period. Pre-pregnancy BMI was Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 3 of 9 calculated as pre-pregnancy weight in kilograms di- placed into the best models together, and significant pre- vided by height in meters, squared. Weight change dictors were retained. A sensitivity analysis was per- was calculated as the difference between current and formed to eliminate three extreme outlier values for pre-pregnancy weight. hsCRP (i.e. ≥ 60 mg/L). P values of 0.05 or less were Pulse and blood pressure were measured, according to considered statistically significant for the analysis. As a a standardized protocol. Three measurements of each sensitivity analysis, the analysis was repeated using only were taken, and the mean of the last two measurements data from women who completed all four initial visits. was recorded and used for our analysis. Data resulting Associations between physical and carotid measures from both demographic and physical measures and re- were not assessed for the second postpartum visit be- cords reviews were collected and managed using REDCap cause 1) associations may differ during pregnancy as a electronic data capture tools hosted at the University of result of dramatic hematologic and hormonal changes Pittsburgh . and 2) the sample size was smaller (i.e., 14) for this visit. Laboratory assays of fasting serum samples collected Statistical analyses were performed using SAS statistical at each visit were performed at the Heinz Nutrition La- software releases 9.3 and 9.4 (SAS Institute, Cary, NC). boratory at the University of Pittsburgh, Graduate School of Public Health, and the following parameters Results were determined using standard laboratory procedures: The mean number of initial study visits was 3.3 (range total cholesterol, high density lipoprotein (HDL-c), low 2–4), and 15 participants (41%) completed all 4 visits. density lipoprotein (LDL-c) , triglycerides , and The average participant age was 28.4 ± 4.6 years, and the glucose . Insulin was measured using a standard average participant pre-pregnancy BMI was 24.3 ± 3.3. radio-immune assay (Linco Research, St. Charles, MO). Participants were predominantly white (91.9%), married HOMA-IR, a measure of insulin resistance, was calcu- or living as married (89.2%), well-educated (89.1% col- lated as (glucose (mg/dl) x insulin (μU/ml))/405 . lege graduate or greater), and employed (64.9% full-time; High-sensitivity C-reactive protein (hsCRP) was mea- 24.3% part-time). Mean birth weight was 3427.2 ± sured with an enzyme-linked immunoassay (Alpha Diag- 224.5 g and mean gestational age at birth was 39.7 ± nostics International Inc., San Antonio, TX). 1.3 weeks. Route of birth was vaginal for 91.2% of Prenatal and birth records were reviewed after the first women, and no newborns had apgar scores less than 7 postpartum study visit to exclude women with complica- at 1 or 5 min of life. At the 6–8-week postpartum visit, tions, which included gestational hypertension, pre- 88% of participants were breastfeeding their infants ex- eclampsia, and preterm birth. Participants completed an clusively. Fourteen participants completed the second interval reproductive and health history form at the sec- postpartum visit 1–5 years (mean 2.7 years) after their ond postpartum visit. first birth, and seven of these participants had experi- enced subsequent pregnancies (i.e., five participants re- Statistical analysis ported having one additional birth, one participant Measures with normal distributions were evaluated as reported having two additional births, and one partici- means ± standard deviations. Measures with pant having a spontaneous abortion). non-normal distributions (i.e., hsCRP and HOMA-IR) Among the participants, IAD increased throughout were analyzed as medians with interquartile range and pregnancy from a mean (SE) of 6.47 (.12) mm in the 1st log-transformed for our analysis. Categorical variables trimester to 6.89 (.10) mm in the 3rd trimester (all P (e.g., employment) were presented as percentages. Linear < 0.05). IAD then returned to early pregnancy values mixed models featuring random intercepts and Toeplitz (i.e., 6.36 [.07] mm, P = 0.76) by the first postpartum variance and covariance structure were used to estimate visit, and we observed no further decrease at the second means for CCA IMT and IAD. postpartum visit (6.42 [0.11] mm) (Table 1). Adjustment Baseline maternal age and pre-pregnancy BMI were for maternal age and pre-pregnancy BMI minimally af- included a priori in all models. Separate models were fected these estimates (Fig. 1). CCA IMT remained constructed for systolic blood pressure, weight, and stable between the 1st and 2nd trimesters and then in- weight change. For CCA IMT, models were also con- creased in the 3rd trimester and through the postpartum structed including IAD, since over time increases in IAD period (i.e., 1st trimester mean [SE] 0.547 [.02] mm, first can cause increases in CCA IMT. Predictors with a sig- postpartum 0.565 [.01] mm, second postpartum 0.581 nificance level of P ≤ 0.2 were then placed into models [0.02] mm) (Table 1). These values changed minimally together, and predictors with a significance level of P ≤ when adjusted for maternal age and pre-pregnancy BMI 0.1 were retained. Next, biomarkers were tested indi- (Fig. 2). vidually in the final models identified for each outcome. Changes in weight, blood pressure, heart rate, lipid, Biomarkers with a significance level of P ≤ 0.1 were then glucose, and hsCRP concentrations followed expected Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 4 of 9 Table 1 Unadjusted values for vascular measures and biomarkers by trimester and postpartum Measure 1st trimester 2nd trimester 3rd trimester 1st postpartum 2nd postpartum Overall n =17 n =32 n =37 n =35 n =14 P-value Inter-adventitial diameter (mm) 6.47 (0.12) 6.79 (0.08) 6.89 (0.10) 6.36 (0.07) 6.42 (0.11) < 0.0001 CCA intima-media thickness (mm) 0.547 (0.02) 0.546 (0.01) 0.553 (0.01) 0.565 (0.01) 0.581 (0.02) 0.03 Weight (kg) 68.7 (2.2) 73.1 (1.6) 79.5 (1.8) 69.2 (1.5) 70.2 (1.9) < 0.0001 Weight change (kg) 0.55 (0.46) 7.27 (0.62) 14.4 (0.91) 4.3 (0.75) 3.7 (1.1) < 0.0001 Systolic blood pressure (mm Hg) 103.7 (2.1) 106.0 (1.7) 110.4 (1.4) 106.2 (1.7) 102.9 (2.5) < 0.0001 Heart rate (bpm) 78.0 (2.4) 79.8 (1.6) 82.0 (1.5) 68.1 (1.5) 63.6 (2.5) < 0.0001 Total cholesterol (mg/dl) 201.7 (8.9) 257.3 (7.0) 273.1 (7.2) 191.4 (5.5) 194.8 (8.4) < 0.0001 LDL-c (mg/dl) 111.2 (7.0) 148.0 (6.3) 155.5 (6.6) 114.7 (4.8) 121.9 (7.0) < 0.0001 Triglycerides (mg/dl) 108.3 (10.5) 176.5 (10.2) 250.7 (13.6) 77.3 (7.0) 87.1 (9.6) < 0.0001 HDL-c (mg/dl) 68.8 (2.2) 74.0 (3.1) 66.8 (2.3) 61.2 (1.8) 55.6 (2.9) < 0.0001 Glucose (mg/dl) 79.3 (1.5) 77.2 (1.1) 77.0 (1.2) 82.6 (1.2) 88.5 (1.9) < 0.0001 Insulin (μU/ml) 8.84 (0.78) 11.25 (0.98) 11.95 (0.81) 8.59 (0.50) 10.76 (0.82) 0.0008 HOMA-IR 1.64 [1.32, 2.09] 2.16 [1.56, 2.56] 2.28 [1.67, 2.51] 1.72 [1.34, 2.13] 2.20 [1.76, 2.87] 0.03 hsCRP (mg/L) 3.58 [2.16, 5.57] 3.36 [2.31, 5.49] 3.29 [2.24, 7.01] 1.20 [.77, 2.44] 0.96 [0.37, 1.50] < 0.0001 Normally distributed values presented as mean (SE) and P value from mixed models. Skewed values presented as median [IQR] and P-value from Wilcoxon rank- sum test. CCA is common carotid artery. Weight change is change from pre-pregnancy weight patterns for healthy pregnancies  (Table 1). Greater Higher SBP was associated with greater CCA IMT; weight was associated marginally with greater IAD, and nonetheless, accounting for SBP did not attenuate the attenuated the increase in IAD that occurred throughout postpartum increase in CCA IMT (Table 4, Model 2). pregnancy (Table 2, Model 3). When metabolic factors Greater weight gain was marginally associated with thin- were considered, higher triglyceride concentrations were ner CCA IMT (Table 4, Models 5, 6, 7), and greater IAD associated (P < 0.0001) with lower IAD, but higher was associated with thicker CCA IMT (Table 4, Model hsCRP was associated (P = 0.0002) with greater IAD 6). In addition, when metabolic factors were considered, (Table 2, Model 5, and Table 3). greater HOMA-IR was associated with lower CCA IMT Fig. 1 Changes in inter-adventitial diameter across pregnancy, adjusted for maternal age and pre-pregnancy BMI. All pairwise comparisons significant at P < .0001 except: 1st Trimester vs. 1st Postpartum P = .0.99, 1st Trimester vs. 2nd Postpartum P = 0.80, 2nd Trimester vs. 3rd Trimester P = .03, 1st postpartum vs. 2nd postpartum P = 0.73. Adjusted for age and pre-pregnancy body mass index. The diamond represents the mean and the horizontal line represents the median Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 5 of 9 Fig. 2 Changes in CCA IMT across pregnancy, adjusted for maternal age and pre-pregnancy BMI. Statistically significant differences are as follows: 1st Trimester vs. 1st Postpartum P = 0.03, 1st Trimester vs. 2nd Postpartum P = 0.01, 2nd Trimester vs. 1st Postpartum P = 0.01, 2nd Trimester vs. 2nd Postpartum P = 0.01. The diamond represents the mean and the horizontal line represents the median Table 2 Associations between inter-adventitial diameter, physical predictors, and significant metabolic predictors b b Predictor Unadjusted Model 1 Model 2 ß (SE) P-value ß (SE) P-value ß (SE) P-value Trimester 1 Ref Ref Ref c c c Trimester 2 0.361 (.07) 0.0001 0.361 (.07) < 0.0001 0.389 (.07) < 0.001 cd cd c Trimester 3 0.498 (.07) 0.0001 0.499 (.07) < 0.0001 0.511 (.07) < 0.001 Postpartum −0.015 (.05) 0.74 −0.014 (.05) 0.76 0.010 (.04) 0.81 Age (years) −0.004 (.02) 0.81 −0.003 (.02) 0.83 Pre-pregnancy BMI (kg/m ) 0.046 (.02) 0.06 0.042 (.02) 0.09 SBP (mmHG) 0.004 (.00) 0.29 b b b Predictor Model 3 Model 4 Model 5 ß (SE) P-value ß (SE) P-value ß (SE) P-value Trimester 1 Ref Ref 0.84 Ref c c c Trimester 2 0.294 (.09) 0.001 0.321 (.09) < 0.001 0.456 (.07) < 0.0001 c c cd Trimester 3 0.338 (.13) 0.009 0.392 (.14) 0.008 0.683 (.12) < 0.0001 Postpartum −0.032 (.05) 0.49 −0.020 (.05) 0.69 − 0.029 (.04) 0.44 Age (years) 0.004 (.02) 0.82 − 0.001 (.02) 0.93 −0.006 (.02) 0.73 Pre-pregnancy BMI (kg/m ) 0.013 (.03) 0.67 0.047 (.02) 0.06 0.006 (.03) 0.85 Weight (kg) 0.015 (.01) 0.08 0.011 (.01) 0.12 Weight change (kg) 0.010 (.01) 0.29 Triglycerides (mg/dl) −0.002 (.00) < 0.0001 Log hsCRP (mg/L) 0.070 (.02) 0.0002 Linear mixed models Model 1: Adjusted for age & pre-pregnancy BMI. Model 2: Model 1 plus SBP. Model 3: Model 1 plus weight. Model 4: Model 1 plus weight change. Model 5: Model 3 plus triglycerides and Log hsCRP c d Different from postpartum at p < .01. Different from second trimester at p < .05 Weight change is from pre-pregnancy weight. β represents change in millimeters Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 6 of 9 Table 3 Associations of individual biomarkers with inter-adventitial diameter and common carotid artery intima-media thickness b b Biomarker Inter-adventitial Diameter Common Carotid Artery Intima-Media Thickness β (SE) P-value β (SE) P-value Total Cholesterol (mg/dl) −0.001 (0.1) 0.18 −0.000 (.00) 0.95 HDL-c (mg/dl) 0.002 (.00) 0.52 −0.000 (.00) 0.39 Triglycerides (mg/dl) −0.001 (.00) 0.01 0.000 (.00) 0.45 LDL-c (mg/dl) −0.001 (.00) 0.38 −0.000 (.00) 0.80 hsCRP (mg/L) 0.004 (.00) 0.03 0.000 (.00) 0.54 Fasting insulin (μU/ml) 0.004 (.01) 0.55 −0.002 (.00) 0.13 Fasting glucose (mg/dl) −0.005 (.00) 0.13 −0.001 (.00) 0.09 Log HOMA-IR −0.013 (.07) 0.86 −0.029 (.01) 0.02 Linear mixed models Models include time point in pregnancy cycle (trimester or postpartum), age, pre-pregnancy BMI, systolic blood pressure, and weight change from pre-pregnancy baseline β represents change in millimeters values (Table 4, Model 7). Accounting for HOMA-IR outliers, were consistent with those from the primary did not affect the increased CCA IMT observed postpar- analyses (Additional file 1: Table S1 and Add- tum (Table 4, Model 7). itional file 2: Table S2). Moreover, for the second Results of sensitivity analyses limited to women who postpartum visit, no reproductive factors (e.g., number completed all four initial visits and that excluded hsCRP of interval pregnancies or breastfeeding status) were Table 4 Associations between common carotid artery intima-media thickness, physical predictors, and significant metabolic predictors b b b Predictor Unadjusted Model 1 Model 2 Model 3 ß (SE) P-value ß (SE) P-value ß (SE) P-value ß (SE) P-value Trimester 1 Ref Ref Ref Ref Trimester 2 0.001 (.01) 0.89 0.001 (.01) 0.89 0.002 (.01) 0.85 0.007 (.01) 0.56 Trimester 3 0.013 (.01) 0.24 0.013 (.01) 0.24 0.009 (.01) 0.43 0.022 (.02) 0.22 c c cd c Postpartum 0.027 (.01) 0.02 0.027 (.01) 0.02 0.026 (.01) 0.03 0.031 (.01) 0.01 Age (yr) 0.004 (.00) 0.03 0.004 (.00) 0.02 0.003 (.00) 0.08 Pre-pregnancy BMI (kg/m ) −0.001 (.00) 0.78 −0.002 (.00) 0.45 0.000 (.00) 0.93 SBP (mm Hg) 0.001 (.00) 0.08 Weight (kg) −0.000 (.00) 0.66 b b b b Predictor Model 4 Model 5 Model 6 Model 7 ß (SE) P-value ß (SE) P-value ß (SE) P-value ß (SE) P-value Trimester 1 Ref Ref Ref Ref Trimester 2 0.016 (.01) 0.19 0.018 (.01) 0.17 0.007 (.01) 0.59 0.009 (.01) 0.51 c c c Trimester 3 0.042 (.02) 0.04 0.041 (.02) 0.046 0.027 (.02) 0.19 0.033 (.02) 0.13 d d c d Postpartum 0.036 (.01) 0.003 0.035 (.01) 0.005 0.034 (.01) 0.006 0.027 (.01) 0.03 Age (years) 0.002 (.00) 0.11 0.003 (.00) 0.07 0.003 (.002) 0.05 0.003 (.00) 0.12 Pre-pregnancy BMI (kg/m ) −0.001 (.00) 0.70 −0.002 (.00) 0.35 −0.004 (.00) 0.15 −0.002 (.00) 0.55 Weight change (kg) −0.002 (.00) 0.13 −0.002 (.00) 0.07 −0.002 (.00) 0.06 −0.002 (.00) 0.08 SBP (mmHg) 0.001 (.00) 0.04 0.001 (.00) 0.04 0.001 (.00) 0.03 Inter-adventitial diameter 0.026 (.01) 0.02 0.017 (.01) 0.17 Log HOMA-IR −0.028 (.01) 0.03 Linear mixed models Model 1: Adjusted for age & pre-pregnancy BMI. Model 2: Model 1 plus SBP. Model 3: Model 1 plus weight. Model 4: Model 1 plus weight change. Model 5: Model 1 plus SBP and weight change. Model 6: Model 5 plus inter-adventitial diameter. Model 7: Model 6 plus HOMA-IR c d Different from second trimester at p < .05. Different from third trimester at p < .05 BMI is body mass index. SBP is systolic blood pressure. Weight change is from pre-pregnancy weight. β represents change in millimeters Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 7 of 9 statistically significantly associated with either carotid increases as arterial diameter increases, which causes ar- measure (data not shown). terial walls to thicken [30, 34]. CCA IMT would thicken during pregnancy as IAD increases, to normalize arterial Discussion wall stresses, as our results confirm . Among our participants with normal first pregnancies, In contrast to the effects of body weight and change in CCA IMT thickened late in pregnancy and remained IAD, the metabolic changes during pregnancy that may thickened at 2.7 years postpartum; IAD, however, in- be considered atherogenic in non-pregnant adults (i.e., creased throughout pregnancy and returned to early increased total cholesterol, LDL-c, triglycerides, pregnancy levels, postpartum. Although our results mir- HOMA-IR, and hsCRP) do not explain the increased ror those described in two classic studies [7, 8], our IAD and CCA IMT that we observed. As expected, we study is the first to follow women for more than 1 year observed an association between higher hsCRP and postpartum. With more participants (i.e., 43) than those greater IAD. Without pregnancy, higher hsCRP concen- studies [7, 8] combined, our study establishes statistically trations are associated with greater carotid IMT [35–37], significant changes in CCA IMT and IAD. While a re- which is associated with greater IAD. However, in our cent study did not demonstrate that CCA IMT was in- study, hsCRP concentrations did not explain the ob- creased in the 3rd trimester, it assessed women earlier in served changes in IAD. Our finding that higher triglycer- the trimester than we did . Our results demonstrate ide concentrations were associated with smaller IAD that unhealthy change in CCA IMT is partially explained  was unexpected, because this relationship differs by changes in IAD and weight—not atherogenic meta- from that observed in non-pregnant adult women. bolic changes. Triglyceride concentrations increase dramatically during An increase in CCA IMT beginning late in pregnancy healthy pregnancy to support fetal growth, and no accepted and persisting postpartum beyond 2 years, in addition to threshold value exists for what constitutes high triglyceride lifestyle changes involved with parenthood and concentrations in pregnancy . However, triglyceride socio-economic profile of women with large families, concentrations can be excessive in pregnancy, as triglycer- could help explain the greater CVD risk that occurs for ide concentrations in the upper percentiles have been asso- women of high parity [1, 3]. Greater IMT is a risk factor ciated with preeclampsia and preterm birth [38–40]. Both for CVD because thickened arteries are 1) less capable high triglyceride concentrations and smaller IAD indeed of responding to changes in blood pressure and 2) could be associated with less healthy pregnancies. Our re- more prone to atherosclerosis . Although studies sults suggest that paradigms of CVD prediction may not be have identified greater CCA IMT in women of higher applicable to the wellness state of pregnancy. parity [6, 27–29], the cause remains unknown. However, Our study benefited from the use of a highly valid and we observed thicker CCA IMT among our participants reproducible measure of carotid structure (i.e., B-mode more than 2 years after childbirth, which suggests that ultrasonography), and high participant retention (i.e., 98%) the acute negative effect of pregnancy on CCA IMT may in the initial study. We also collected serial vascular and persist and could serve as a risk factor for CVD. biomarker measures during and after pregnancy, which The observed changes in CCA IMT and IAD are con- strengthens this study, but the lack of pre-pregnancy mea- sistent with the literature concerning hemodynamic sures poses a limitation. Limitations of the study are changes in pregnancy and the effect of hemodynamic largely due to the rapidly changing hormonal and changes on arteries [30–34]. Importantly, we provide hemodynamic milieus of pregnancy and the postpartum serial measures in pregnancy to characterize this vascu- period. Because the hemodynamic changes of pregnancy lar remodeling and evaluate concomitant metabolic begin as early as 5 weeks of gestation , our 1st trimes- markers. Vascular remodeling is largely due to ter values may not represent a true pre-pregnancy base- hemodynamic factors. Arterial walls adapt to maintain line. For example, thinning of the CCA IMT may have homeostasis between the two main stresses of blood occurred before we could assess it. Similarly, because most flow: shear and tensile stress. First, shear stress is the participants (94%) were breastfeeding at the first postpar- frictional force of blood flowing along the arterial wall. tum visit, their hormonal and cardiovascular status had Increased shear stress causes blood vessels to increase in not attained new postpartum “normal” status. CCA IMT diameter [30–32]. Cardiac output increases early in the may regress after weaning. Our results also might not re- 1st trimester of pregnancy  and peaks at 30–60% flect those for women who formula-feed. Additionally, at above the non-pregnant level in the late 2nd or early 3rd the second postpartum visit, participants exhibited a vary- trimester . Increased cardiac output should increase ing number of subsequent pregnancies, which makes in- IAD resulting from increased shear stress, as our results terpretation difficult. However, our results are consistent demonstrate. Second, tensile stress is the force of blood with those of the Cardiovascular Risk in Young Finns perpendicular to the arterial wall, and this force study, which found that young women who gave birth Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 8 of 9 over a 6-year period had greater progression of CCA IMT adventitial diameter; LDL-c : Low density lipoprotein concentration; MVP : Maternal Vascular Adaptations to Healthy Pregnancy; SBP : Systolic blood than those who had not , and with epidemiologic stud- pressure; SE : Standard error; URL : Ultrasound Research Laboratory ies showing greater CCA IMT in midlife women of higher parity [27–29]. Moreover, although our largely white, Acknowledgments The authors gratefully acknowledge William B. Greene, EdD, Scientific Editor well-educated participants do not represent all first-time and Writer at the University of Pittsburgh, School of Nursing, for his pregnant women in the United States, our study provides assistance with the preparation of this manuscript, and Alyssa Oakes, SN, for valuable baseline data against which arterial remodeling in her assistance with development of the tables. other demographic groups can be assessed. Funding Future work should follow a life course approach, and NAN was supported by NICHD grant T32HD0055162–04 and NHBLI grant seek to enroll women during the preconception period T32HL083825 to the University of Pittsburgh. The project described was supported by the National Institutes of Health through Grant Numbers UL1 to obtain a true baseline and then follow them through RR024153 and UL1TR000005. The funding bodies had no role in design of at least a several month period after weaning. Retention the study; collection, analysis, and interpretation of data; or in writing the for the postpartum visits is critical. Additional studies manuscript. should explore vascular adaptation to pregnancy in Availability of data and materials women in subsequent pregnancies, from different racial The datasets used and analyzed during the current study are available from and ethnic groups, and with higher BMI. Collection of the corresponding author on reasonable request. serum folate levels might provide valuable insights into Authors’ contributions the role folate deficiency during pregnancy plays in dif- NAN performed the original study visits; analyzed the data; and wrote the ferences in vascular adaptation. first, second, and third drafts of the manuscript. MB provided statistical support, contributed to interpretation of results, and read and approved the final manuscript. JC contributed to interpretation of results, and read and Conclusions approved the final manuscript. MD designed the follow up study, performed We found that IAD increased throughout a healthy first study visits, and read and approved the final manuscript. CKM designed the pregnancy and decreased by 8 weeks postpartum. In original study, contributed to interpretation of results, and read and approved the final manuscript. JMR contributed to interpretation of results, contrast, postpartum CCA IMT thickening persisted for and read and approved the final manuscript. AS contributed to more than 2 years. These adaptations can be explained— interpretation of results and read and approved the final manuscript. PGT partially—by pregnancy-related changes in weight and provided statistical support, contributed to interpretation of results, and read and approved the final manuscript. EBM contributed to interpretation of IAD; moreover, they are not substantially explained by results, and read and approved the final manuscript. changes in metabolic measures. Therefore, our results suggest that pregnancy represents a unique setting of Ethics approval and consent to participate This study was approved by the University of Pittsburgh, Human Research rapid physiologic changes that maintain homeostasis Protection Office, ID # PRO09050089 and ID# PRO14060316. This body during a period of acute stress. approved an informed consent document that was signed by all study Understanding normal vascular adaptation to preg- participants. nancy can not only engender an improved understand- Competing interests ing of the physiology of pregnancy complications, but The authors declare that they have no competing interests. also better identify women at risk for complications early in pregnancy. If it persists, the greater CCA IMT de- Publisher’sNote tected postpartum may help explain the higher CVD risk Springer Nature remains neutral with regard to jurisdictional claims in in women of higher parity. published maps and institutional affiliations. Author details Additional files 1 Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, 130 De Soto Street, Pittsburgh, PA 15261, USA. Department of Health Promotion and Development, School of Nursing, University of Additional file 1: Table S1 Associations between inter-adventitial diam- Pittsburgh, 3500 Victoria Street, 440 Victoria Building, Pittsburgh, PA 15261, eter, physical predictors, and significant metabolic predictors for the 15 USA. Department of Obstetrics and Gynecology, School of Medicine, women who completed all 4 study visits. These are data about carotid University of Pittsburgh, 3550 Terrace Street, Pittsburgh, PA 15213, USA. measures, physical predictors, and significant metabolic predictors for the Department of Clinical and Translational Research, School of Medicine, 15 women who completed all 4 initial study visits. (DOCX 15 kb) University of Pittsburgh, 3550 Terrace Street, Pittsburgh, PA 15213, USA. Additional file 2: Table S2 Associations between common carotid Magee-Womens Research Institute, Magee-Womens Hospital of University artery intima-media thickness, physical predictors, and significant meta- of Pittsburgh Medical Center (UPMC), 204 Craft Avenue, Pittsburgh, PA bolic predictors for the 15 women who completed all 4 study visits. 15213, USA. These are data about carotid measures, physical predictors, and signifi- cant metabolic predictors for the 15 women who completed all 4 initial Received: 9 February 2018 Accepted: 22 May 2018 study visits. (DOCX 15 kb) Abbreviations References BMI: Body mass index; CCA IMT : Common carotid artery intima-media thick- 1. Parikh NI, Cnattingius S, Dickman PW, Mittleman MA, Ludvigsson JF, ness; CVD : Cardiovascular disease; HDL-c : High density lipoprotein Ingelsson E. Parity and risk of later-life maternal cardiovascular disease. Am concentration; hsCRP : High-sensitivity C-reactive protein; IAD : Inter- Heart J. 2010;159(2):215–21. e216 Niemczyk et al. BMC Pregnancy and Childbirth (2018) 18:195 Page 9 of 9 2. Bertuccio P, Tavani A, Gallus S, Negri E, La Vecchia C. Menstrual and 23. Matthews D, Hosker J, Rudenski A, Naylor B, Treacher D, Turner R. reproductive factors and risk of non-fatal acute myocardial infarction in Italy. Homeostasis model assessment: insulin resistance and beta-cell function Eur J Obstet Gynecol Reprod Biol. 2007;134(1):67–72. from fasting plasma glucose and insulin concentrations in man. 3. Ness RB, Harris T, Cobb J, Flegal KM, Kelsey JL, Balanger A, Stunkard AJ, Diabetologia. 1985;28(7):412–9. D'Agostino RB. Number of pregnancies and the subsequent risk of 24. Liu LX, Arany Z. Maternal cardiac metabolism in pregnancy. Cardiovasc Res. cardiovascular disease. NEnglJMed. 1993;328(21):1528–33. 2014;101(4):545–53. 25. Carallo C, Irace C, Pujia A, De Franceschi MS, Crescenzo A, Motti C, Cortese 4. Ness RB, Schotland HM, Flegal KM, Shofer FS. Reproductive history and C, Mattioli PL, Gnasso A. Evaluation of common carotid hemodynamic coronary heart disease risk in women. EpidemiolRev. 1994;16(2):298–314. forces. Relations with wall thickening. Hypertension. 1999;34(2):217–21. 5. Akhter T, Larsson A, Larsson M, Wikstrom AK, Naessen T. Artery wall layer 26. Lakatta EG. Arterial and cardiac aging: major shareholders in cardiovascular dimensions during normal pregnancy: a longitudinal study using disease enterprises: part III: cellular and molecular clues to heart and arterial noninvasive high-frequency ultrasound. Am J Physiol Heart Circ Physiol. aging. Circulation. 2003;107(3):490–7. 2013;304(2):H229–34. 27. Humphries KH, Westendorp IC, Bots ML, Spinelli JJ, Carere RG, Hofman A, 6. Skilton MR, Bonnet F, Begg LM, Juonala M, Kahonen M, Lehtimaki T, Viikari Witteman JC. Parity and carotid artery atherosclerosis in elderly women: the JS, Raitakari OT. Childbearing, child-rearing, cardiovascular risk factors, and Rotterdam study. Stroke. 2001;32(10):2259–64. progression of carotid intima-media thickness: the cardiovascular risk in 28. Wolff B, Volzke H, Robinson D, Schwahn C, Ludemann J, Kessler C, John U, Felix young Finns study. Stroke. 2010;41(7):1332–7. SB. Relation of parity with common carotid intima-media thickness among 7. Mersich B, Rigo J Jr, Besenyei C, Lenard Z, Studinger P, Kollai M. Opposite women of the study of health in Pomerania. Stroke. 2005;36(5):938–43. changes in carotid versus aortic stiffness during healthy human pregnancy. 29. Skilton MR, Serusclat A, Begg LM, Moulin P, Bonnet F. Parity and carotid ClinSci(Lond). 2005;109(1):103–7. atherosclerosis in men and women: insights into the roles of childbearing 8. Visontai Z, Lenard Z, Studinger P, Rigo J Jr, Kollai M. Impaired baroreflex and child-rearing. Stroke. 2009;40(4):1152–7. function during pregnancy is associated with stiffening of the carotid artery. 30. Pries AR, Reglin B, Secomb TW. Remodeling of blood vessels: responses of Ultrasound ObstetGynecol. 2002;20(4):364–9. diameter and wall thickness to hemodynamic and metabolic stimuli. 9. Sator MO, Joura EA, Gruber DM, Obruca A, Zeisler H, Egarter C, Huber JC. Non- Hypertension. 2005;46(4):725–31. invasive detection of alterations of the carotid artery in pregnant women with 31. Herity NA, Ward MR, Lo S, Yeung AC. Review: clinical aspects of vascular high-frequency ultrasound. Ultrasound ObstetGynecol. 1999;13(4):260–2. remodeling. J Cardiovasc Electrophysiol. 1999;10(7):1016–24. 10. Iacobaeus C, Andolf E, Thorsell M, Bremme K, Jorneskog G, Ostlund E, Kahan 32. Kiechl S, Willeit J. The natural course of atherosclerosis. Part II: vascular T. Longitudinal study of vascular structure and function during normal remodeling. Bruneck Study Group. Arterioscler Thromb Vasc Biol. 1999; pregnancy. Ultrasound Obstet Gynecol. 2017;49(1):46–53. 19(6):1491–8. 11. Stein JH, Korcarz CE, Hurst RT, Lonn E, Kendall CB, Mohler ER, Najjar SS, 33. Abbas AE, Lester SJ, Connolly H. Pregnancy and the cardiovascular system. Rembold CM, Post WS. Use of carotid ultrasound to identify subclinical IntJCardiol. 2005;98(2):179–89. vascular disease and evaluate cardiovascular disease risk: a consensus 34. Bokov P, Chironi G, Orobinskaia L, Flaud P, Simon A. Carotid circumferential statement from the American Society of Echocardiography carotid intima- wall stress homeostasis in early remodeling: theoretical approach and media thickness task force. Endorsed by the Society for Vascular Medicine. J clinical application. J Clin Ultrasound. 2012;40(8):486–94. Am Soc Echocardiogr. 2008;21(2):93–111. quiz 189-190 35. Toprak A, Kandavar R, Toprak D, Chen W, Srinivasan S, Xu JH, Anwar A, 12. Bonithon-Kopp C, Touboul PJ, Berr C, Magne C, Ducimetiere P. Factors of Berenson GS. C-reactive protein is an independent predictor for carotid carotid arterial enlargement in a population aged 59 to 71 years: the EVA artery intima-media thickness progression in asymptomatic younger adults study. Stroke. 1996;27(4):654–60. (from the Bogalusa heart study). BMC Cardiovasc Disord. 2011;11:78. 13. Polak JF, Wong Q, Johnson WC, Bluemke DA, Harrington A, O'Leary DH, 36. Ciccone MM, Scicchitano P, Zito A, Cortese F, Boninfante B, Falcone VA, Yanez ND. Associations of cardiovascular risk factors, carotid intima-media Quaranta VN, Ventura VA, Zucano A, Di Serio F, et al. Correlation between thickness and left ventricular mass with inter-adventitial diameters of the inflammatory markers of atherosclerosis and carotid intima-media thickness common carotid artery: the multi-ethnic study of atherosclerosis (MESA). in obstructive sleep apnea. Molecules. 2014;19(2):1651–62. Atherosclerosis. 2011; 37. Ock SY, Cho KI, Kim HJ, Lee NY, Kim EJ, Kim NK, Lee WH, Yeo GE, Heo JJ, 14. Joensuu T, Salonen R, Winblad I, Korpela H, Salonen JT. Determinants of Han YJ, et al. The impacts of C-reactive protein and atrial fibrillation on femoral and carotid artery atherosclerosis. J Intern Med. 1994;236(1):79–84. carotid atherosclerosis and ischemic stroke in patients with suspected 15. Lakatta EG, Levy D. Arterial and cardiac aging: major shareholders in ischemic cerebrovascular disease: a single-center retrospective observational cardiovascular disease enterprises: part I: aging arteries: a "set up" for cohort study. Korean Circ J. 2013;43(12):796–803. vascular disease. Circulation. 2003;107(1):139–46. 38. Gallos ID, Sivakumar K, Kilby MD, Coomarasamy A, Thangaratinam S, Vatish 16. Eigenbrodt ML, Sukhija R, Rose KM, Tracy RE, Couper DJ, Evans GW, Bursac M. Pre-eclampsia is associated with, and preceded by, hypertriglyceridaemia: Z, Mehta JL. Common carotid artery wall thickness and external diameter as a meta-analysis. BJOG. 2013;120(11):1321–32. predictors of prevalent and incident cardiac events in a large population 39. Mudd LM, Holzman CB, Catov JM, Senagore PK, Evans RW. Maternal lipids at study. CardiovascUltrasound. 2007;5:11. mid-pregnancy and the risk of preterm delivery. Acta Obstet Gynecol Scand. 17. Kozakova M, Morizzo C, La Carrubba S, Fabiani I, Della Latta D, Jamagidze J, 2012;91(6):726–35. Chiappino D, Di Bello V, Palombo C. Associations between common carotid 40. Catov JM, Bodnar LM, Kip KE, Hubel C, Ness RB, Harger G, Roberts JM. Early artery diameter, Framingham risk score and cardiovascular events. Nutr pregnancy lipid concentrations and spontaneous preterm birth. Am J Metab Cardiovasc Dis. 2017; 27(4):329–34. Obstet Gynecol. 2007;197(6):610 e611–7. 18. Wendelhag I, Gustavsson T, Suurkula M, Berglund G, Wikstrand J. Ultrasound measurement of wall thickness in the carotid artery: fundamental principles and description of a computerized analysing system. Clin Physiol. 1991;11(6):565–77. 19. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81. 20. Friedewald W, Levy R, Fredrickson D. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499–502. 21. Bucolo G, David H. Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem. 1973;19(5):476–82. 22. Bondar RJ, Mead DC. Evaluation of glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides in the hexokinase method for determining glucose in serum. Clin Chem. 1974;20(5):586–90.
BMC Pregnancy and Childbirth – Springer Journals
Published: May 31, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera