OBJECTIVE: To prospectively examine whether blood pressure changes persist after pregnancy among women of reproductive age.
METHODS: This was a prospective, population-based, observational cohort of 2,304 (1,167 black, 1,137 white) women (aged 18–30 years) who were free of hypertension at baseline (1985–1986) and reexamined up to six times at 2, 5, 7, 10, or 20 years later (2005–2006). We obtained standardized blood pressure measurements before and after pregnancies and categorized women into time-dependent groups by the cumulative number of births since baseline within each time interval (zero births [referent]; one interim birth and two or more interim births; nonhypertensive pregnancies). The study assessed differences in systolic and diastolic blood pressures among interim birth groups using multivariable, repeated measures linear regression models stratified by baseline parity (nulliparous and parous), adjusted for time, age, race, baseline covariates (blood pressure, body mass index, education, and oral contraceptive use), and follow-up covariates (smoking, antihypertensive medications, oral contraceptive use, and weight gain).
RESULTS: Among nulliparas at baseline, mean (95% confidence interval) fully adjusted systolic and diastolic blood pressures (mm Hg), respectively, were lower by –2.06 (–2.72 to –1.41) and –1.50 (–2.08 to –0.92) after one interim birth, and lower by –1.89 (–2.63 to –1.15) and –1.29 (–1.96 to –0.63) after two or more interim births compared with no births (all P<.001). Among women already parous at baseline, adjusted mean blood pressure changes did not differ by number of subsequent births.
CONCLUSION: A first birth is accompanied by persistent lowering of blood pressure from preconception to years after delivery. Although the biologic mechanism is unclear, pregnancy may create enduring alterations in vascular endothelial function.
LEVEL OF EVIDENCE: II
A first birth is accompanied by lower systolic and diastolic blood pressures that persist years after delivery, independent of weight gain, secular trends, and behaviors.
From the 1Division of Research, Epidemiology and Prevention, Kaiser Permanente, Oakland, California; 2Division of Preventive Medicine, University of Alabama at Birmingham, Birmingham, Alabama; 3Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania; 4Department of Obstetrics and Gynecology, Magee-Womens Hospital, University of Pittsburgh, Pittsburgh, Pennsylvania; and 5Division of Prevention and Population Sciences, National Heart, Lung, and Blood Institute, Bethesda, Maryland.
Supported by contracts # N01-HC-48047, N01-HC-48048, N01-HC-48049, N01-HC-48050, and N01-HC-95095, from the National Heart, Lung, and Blood Institute and Career Development Award, Grant number K01 DK059944 from the National Institute of Diabetes, Digestive, and Kidney Diseases.
Corresponding author: Erica P. Gunderson, PhD, Epidemiologist, Research Scientist II, Division of Research, Epidemiology, and Prevention Section, Kaiser Permanente, 2000 Broadway, Oakland, CA 94612; e-mail: email@example.com.
Financial Disclosure The authors have no potential conflicts of interest to disclose.
Pregnancy produces marked alterations in vascular physiology; if these persist, pregnancy might provide insights into mechanisms of adult blood pressure regulation.1 Hemodynamic adaptations during pregnancy include increases in cardiac output and blood volume and decreases in perfusion pressure and total systemic vascular resistance, together elevating uteroplacental blood flow without compromising maternal circulation. Mean arterial pressure declines in the first trimester of pregnancy, likely due to a drop in systemic vascular reactivity to norepinephrine and nitric oxide-induced blunting of vascular reactivity to angiotensin II.2–4 Angiogenic factors (vascular endothelial growth factor, VEGF, and placental growth factor, PlGF) produced by placental trophoblast, signal the growth of placental and uterine spiral arteries.5 This acutely contributes to reductions in maternal peripheral vascular resistance. But do these physiologic adaptations leave a lasting imprint on the postpartum mother?
Blood pressure is somewhat lower in parous as compared with nulliparous women according to several, but not all, previous epidemiologic studies.6 The causality of this association, however, must be viewed with skepticism on several grounds. First, it is not clear whether women with previous births have lower blood pressure or whether women without any births have higher blood pressure. Second, studies of parity and blood pressure have generally been cross-sectional or retrospective, leaving temporality unclear. Third, not all results have been consistent. Finally, not all parity is equivalent with respect to cardiovascular effects; women with hypertensive disorders of pregnancy are at elevated risk for future hypertension.7 Whether nonhypertensive births reduce blood pressure remote from pregnancy, therefore, remains unclear. A study of 30 women that measured preconception blood pressures reported a nonsignificant 2 mm Hg lower mean arterial blood pressure at 1 year postpartum.8
We sought to prospectively examine changes in blood pressure from the time before to after pregnancy relative to women who did not give birth within a large, population-based cohort of women of reproductive age from the Coronary Artery Risk Development in Young Adults (CARDIA) Study. We hypothesized that delivery of a pregnancy uncomplicated by hypertension persistently lowers levels of both systolic and diastolic blood pressure compared with no births.
MATERIALS AND METHODS
The CARDIA Study is a multicenter, longitudinal, observational study designed to describe the development of risk factors for coronary heart disease in young black and white men and women. In 1985–1986, 5,115 participants (2,787 women; 52% black), aged 18 to 30 years, were recruited from four geographic areas: Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota; and Oakland, California.9,10 Retention rates were 91%, 86%, 81%, 79%, 74%, and 72% of the surviving cohort 2, 5, 7, 10, 15, and 20 years (2005–2006) after baseline.11,12 Institutional review boards at each participating study center approved the study. Written, informed consent was obtained from subjects for all study procedures.
Of 2,787 women, we excluded women reporting a hysterectomy or removal of both ovaries (n=25), history of hypertension (n=249) or current antihypertensive medication use (n=5), elevated systolic (140 mm Hg or more) and/or diastolic (90 mm Hg or more) blood pressure (n=14), past or current heart medication (n=34), or current asthma medication (n=58) at baseline. We also excluded women missing self-reported hypertension status (n=10) or nonpregnant systolic and diastolic blood pressures (n=88) at baseline and/or at all follow-up exams. Our analytic sample included 2,304 (1,167 black, 1,137 white) women, of whom 1,531 were nulliparous and 773 parous at baseline. Women excluded tended to be of black race, smokers, older, less educated, and have higher body mass index and larger waist girth at baseline than the analytic sample.
After an initial 5-minute rest, blood pressure was measured three times at 1-minute intervals using the Hawksley (Lancing, Sussex, UK) random-zero sphygmomanometer through year 15 and the Omron (Omron Corp., Schaumburg, IL) HEM907XL oscillometer at year 20; the first and fifth phase Korotkoff sounds were recorded. The appropriate cuff size (small, medium, large, extra large) was based on the upper arm circumference, which was measured by the blood pressure technician at the midpoint between the acromion and the olecranon. The second and third measurements were averaged.
Among 200 CARDIA participants across 5-year age groups, we established calibration equations for transforming Omron levels to random-zero levels. Blood pressure was measured three times using the Omron and random-zero devices together simultaneously according to CARDIA protocol for random-zero blood pressure measurements and the protocol developed by Dr. Ronald Prineas. The average of the second and third readings from the random-zero (RZ) device was regressed on Omron measurements for both systolic (SBP) and diastolic (DBP) blood pressures for the year 20:
Calibrated RZ SBP=3.74+0.96*OMRON average SBP, R-Square=.94.
Calibrated RZ DBP=1.30+0.97*OMRON average DBP, R-Square=.91.
Changes in blood pressure (systolic and diastolic) were computed for six time intervals each starting from baseline (0 to 2, 0 to 5, 0 to 7, 0 to 10, 0 to 15, and 0 to 20 years) by subtracting the baseline (year 0) from each follow-up measurement.
Antihypertensive medication use was self-reported at baseline and in years 2, 5, 7, 10, 15, and 20. We dichotomized responses for all exams as one fixed variable; ever or never using antihypertensive medication during follow-up.
At each examination we asked each woman whether she was currently pregnant or breast-feeding; number of pregnancies since the last examination, including abortions, miscarriages, and live births or stillbirths, length of gestation, and delivery dates. We classified pregnancies ending in miscarriages, abortions, and/or less than 20 weeks gestation as pregnancy losses and 20 weeks or longer as births. Women were asked at each examination whether they had developed “hypertension without toxemia” or “toxemia” in a pregnancy. Hypertension without toxemia was classified as gestational hypertension and toxemia was preeclampsia. Nulliparity at baseline was defined as no live births before baseline. Births after baseline are called “interim” births.
We formed time-dependent interim birth groups based on the cumulative number of interim births and pregnancy hypertension status. We assigned women to one of five groups for each interval: 0 births (referent), one birth, and two or more (2+) births; nonhypertensive pregnancies, and one or more (1+) births with gestational hypertension, and one or more (1+) births with preeclampsia. Women remained in these categories for all subsequent intervals until the end of the follow-up regardless of future pregnancies.
The validation study of pregnancy hypertension included 290 (217 multiparas, 83 primiparas) CARDIA women whose medical records were obtained for 366 pregnancies in 1986–1996. We abstracted antepartum and intrapartum blood pressures, urinary protein levels, medication use, and International Classification of Diseases, 9th Revision (ICD-9) codes. Preeclampsia was defined as elevated blood pressure (systolic 140 mm Hg or more, diastolic 90 mm Hg or more) and 1+ proteinuria on two or more occasions after 20 weeks of gestation, and/or a discharge diagnosis ICD-9 codes 642.4–642.5, and/or medications specifically used to treat preeclampsia before or during labor. Gestational hypertension was defined as elevated blood pressure and absence of 1+ proteinuria on two or more occasions after 20 weeks of gestation and/or a discharge diagnosis ICD-9 code 642.3. The sensitivity and specificity, respectively, for self-report of preeclampsia was 36% and 93%, and for gestational hypertension it was 8% and 98%. Negative predictive values were 98% and 91% for preeclampsia and gestational hypertension, respectively. Because of substantial misclassification (ie, overreporting pregnancy hypertension), results for gestational hypertension and preeclampsia pregnancies are not presented.
Weight and height and waist circumference were measured by certified technicians according to a standardized protocol described previously.13 Body weight was measured to the nearest 0.1 kg using a calibrated balance beam scale in participants wearing light clothing. Height (without shoes) was measured to the nearest 0.5 cm using a vertical ruler. Waist circumference was measured to the nearest 0.5 cm at the minimal abdominal girth.14 Body mass index (BMI) was computed as weight in kilograms divided by squared height in meters. Weight gain was calculated by subtracting baseline weight from weights at each examination.
Sociodemographic and behavioral data were collected at each examination using self- and interviewer-administered questionnaires. Trained interviewers assessed dietary intake during the previous month using the CARDIA Dietary History at baseline, and daily physical activity scores using the CARDIA Physical Activity History at each examination.15 Baseline categorical variables were smoking (never, former, or current), years of education (12 or less, 13–15, 16 or more), marital status (never married, widowed, divorced or separated, or married), oral contraceptive (OC) use (never, past, current), alcohol intake quartiles, and physical activity scores (race-specific quartiles). Daily dietary nutrient estimates of total fat, protein, carbohydrate, and energy intake (kJ) were used to obtain the percentages of kilocalories from fat, protein, carbohydrate, and crude fiber. Physical activity quartiles have been correlated positively with symptom-limited graded treadmill exercise test duration.16 Time-dependent smoking and OC use groups were categorized as current, past, or never for each examination.
Baseline characteristics by interim birth groups at the end of follow-up were assessed, including frequency distributions, means, standard deviations, medians, and interquartile range. Multiple linear regression methods (analysis of variance) were used to assess baseline differences in continuous variables and weight gain during follow-up among interim birth groups. We used χ2 statistics to examine differences among interim birth groups by race, baseline categorical variables (marital status, oral contraceptive use, education, smoking), ever taking antihypertensive medication, oral contraceptive use, and smoking during follow-up. Wilcoxon rank sum and Kruskal-Wallis one-way tests were used to assess differences in alcohol intake and physical activity (median, interquartile range) due to skewedness in the distributions. P values were obtained from two-tailed tests (significance <.05).
Systolic and diastolic blood pressure measurements from years 0, 2, 5, 7, 10, 15, and 20 were assembled, along with fixed variables, race, and age, and time-dependent interim births groups over the six intervals. Repeated measures linear regression methods (SAS 9.1, PROC MIXED; SAS Institute Inc., Cary, NC) were used to estimate unadjusted and adjusted differences in mean changes in systolic and diastolic blood pressures among interim births (0 births as referent).
We assessed effect modification by time, race, baseline parity, and BMI in the association of interim birth groups with blood pressure changes. Two-way interaction terms for race, baseline parity, and BMI were examined simultaneously in the models to assess potential heterogeneity by including appropriate cross-product terms. We found evidence of significant two-way interaction terms for baseline parity (P<.05) and race (P<.001) with interim birth groups, but not for BMI (P>.10) for systolic blood pressure. Unadjusted and multivariable adjusted means for systolic and diastolic blood pressure changes were also contrasted among interim birth groups separately for each race. A three-way interaction term for baseline parity, race, and time-dependent interim births showed no evidence for effect modification by race and baseline parity (P>.10). We also examined heterogeneity in interim birth group effects over time by inclusion of appropriate cross-product terms for time of examinations and time-dependent interim birth groups and found no evidence of heterogeneity (all P>.10). Both baseline and time-dependent covariates included in models as confounders were selected based on their association with outcome measures independent of association with interim birth groups. Model fit was assessed by comparison of Akaike’s Information Criteria, which is a measure of the goodness of fit of an estimated statistical model, assessing both precision and model complexity.17 Results are stratified by baseline parity (nulliparous and parous) for races combined.
Unadjusted models included indicator variables for time (examination years). Next, Model 1 was adjusted for baseline covariates (relevant blood pressure measure, BMI, education, race, smoking, OC use, age) and time. Time-dependent pregnancy losses and asthma medication and baseline covariates (dietary intake, physical activity) did not act as confounders. Next, we added antihypertensive medication use during follow-up and time-dependent variables (OC use, smoking) by stepwise addition. We examined study center as a fixed variable in all adjusted models, but it had little effect on the estimates. Last, we examined time-dependent weight gain as a potential mediator. Overall significant (P<.05) differences in mean blood pressure changes among time-dependent interim birth groups and pair-wise comparisons relative to 0 births were tested within baseline parity models.
Among 2,304 women (51% black, 49% white), mean (±standard deviation) age was 25 (±3.7) years, and 66% were nulliparous at baseline (60% white, 40% black). Of these, 635 nulliparas and 298 primiparas or multiparas gave birth at least once after baseline and did not report hypertension during pregnancy (Table 1). Among 474 women who delivered two or more interim births, only 102 had three and 33 had four births. Among nulliparas, those with interim birth(s) (Table 1) tended to be younger and married; have smaller waist girth and lower BMI and were less likely to use oral contraceptives at baseline. Parous women with interim birth(s) were younger, more educated, had lower BMI and smaller waist girth, higher percent kilocalories as carbohydrates and lower percent kilocalories as saturated fat than those without births. Women with interim births (nonhypertensive pregnancies) were less likely to have taken antihypertensive medications or oral contraceptives or to have smoked, but gained more weight (Table 2).
Among nulliparas at baseline, unadjusted systolic and diastolic blood pressures were 2 mm Hg lower for one interim birth and two or more interim births compared with none (Table 3, Fig. 1). Differences became slightly weaker with multivariate adjustment for both baseline and time-dependent covariates. Fully adjusted mean (95% confidence interval) systolic and diastolic blood pressures (mm Hg), respectively, were lower by –2.06 (–2.72 to –1.41) and –1.50 (–2.08 to –0.92) for one interim birth and by –1.89 (–2.63 to –1.15) and –1.29 (–1.96 to –0.63) for two or more interim births compared with no births (all P<.001).
Among women parous at baseline, unadjusted systolic blood pressures were –1 mm Hg lower for one interim birth and –3 mm Hg lower for two or more interim births compared with none, and diastolic blood pressures were –1.60 mm Hg lower for two or more interim births compared with none (Table 4, Fig. 1). In fully adjusted models, systolic or diastolic blood pressures did not differ among interim birth groups for women parous at baseline.
Race-specific blood pressure changes showed that among white women, mean changes (95% confidence interval) in systolic blood pressure (mm Hg) for primiparas and multiparas, respectively, compared with nulliparas were more pronounced, –2.26 (–3.01 to –1.52) and –2.51 (–3.31 to –1.72), P<.001, and more modest among black women, –1.27 (–2.45 to –0.08), P<.05, and –0.45 (–1.09 to 1.98). Moreover, diastolic blood pressures were also significantly lower for white primiparas, –1.79 (–2.47 to –1.10), and multiparas, –1.67 (–2.40 to –0.94), compared with nulliparas, all P<.001. In black women, diastolic blood pressures did not differ significantly for primiparas, –0.68 (–1.70 to 0.33), and multiparas, +0.33 (–0.99 to 1.65), compared with nulliparas.
Several unique strengths of our study enhance its validity. First, blood pressure was measured using research methodology both before and after pregnancy, allowing us to examine prospective changes in blood pressure as a result of pregnancy. Standardized blood pressure measurements were available before conception for 100% of women in our analysis, and 89% had at least four measurements after baseline (2- to 5-year intervals) up to 20 years, and 92% of births occurred before year 2001. Other strengths include the 72% retention rate at 20 years of follow-up and the internal comparison group of nonparous women.
The second, unique, strength was our ability to segregate out pregnancies with hypertension. Blood pressure and vascular resistance are higher before, during, and well after pregnancy among women with preeclampsia or gestational hypertension. Our study benefited from being able to examine such pregnancies separately. Our focus was on how nonhypertensive pregnancies related to future blood pressure. A high negative predictive value (above 90%) for self-report of hypertension in pregnancy allowed us to precisely identify women without such conditions. Moreover, by excluding women with preexisting hypertension and controlling for treatment of hypertension during follow-up, we further enhanced the focus on women with nonhypertensive pregnancies in relation to blood pressure. Our validation study found a very low true-positive rate for self-report of gestational hypertension and preeclampsia, suggesting that self-report alone does not accurately identify those diagnosed with these specific complications. Women tended to overreport pregnancy hypertension. Whenever possible, physicians should try to confirm patient self-reports of blood pressure conditions with medical records.
Limitations of our study include variable timing of blood pressure measurements before conception and after delivery and lack of blood pressure measurements during pregnancy. Self-reported hypertensive medication use outside of pregnancy is also a limitation, although we adjusted for this factor in our models. Self-reports are not entirely accurate measures of actual hypertensive medication use. Nonetheless, nondifferential misclassification would have biased our results toward the null. Moreover, hypertensive medications were more common among women with no interim births, and their exclusion likely further biased our result toward the null.
To our knowledge, only one previous study examined blood pressure changes from before to after pregnancy8 and found a nonsignificant, 2 mm Hg lower mean arterial pressure of 2 mm Hg at 1 year postpartum among 30 primiparas and multiparas combined.8 An analysis of the Medical Research Council National Survey of Health that collected reproductive history data before later life blood pressure assessments18 found lower adjusted systolic and diastolic blood pressures of 2.9 and 1.5 mm Hg among women aged 36 years with any compared with no children. The magnitude of these midlife associations is similar to ours, and, like our finding, additional children beyond one did not further lower blood pressure. Another study measured blood pressure before an index pregnancy, but all measurements were during pregnancy. Mean arterial pressure was significantly lower in the index as compared with the first (preceding) pregnancy.19 This suggests either that blood pressure declined after the first pregnancy or that first compared with other pregnancies have differential effect on blood pressure.
Several cross-sectional studies examined the relationship between reproductive history and blood pressure or hypertension, with conflicting results.6 Some showed no significant association or a clinically small negative relationship,20–22 and another23 found lower blood pressure by 0.47 mm Hg with each additional pregnancy among premenopausal women and modestly lower, by 0.39 mm Hg, among postmenopausal women. These data suggest that parity may be inversely associated with blood pressure at mid life, but this may resolve with age.
Another important issue raised by our findings is the clinical significance of a 2-mm Hg reduction in mean blood pressure for women’s long-term health. Based on intervention trials, a reduction in the mean blood pressure by 2 mm Hg would result in a reduction in stroke mortality of 6%, coronary heart disease mortality of 4%, and a reduction in total mortality of 3% for the population. Therefore, our findings provide evidence for a clinically significant effect on health among the population of women who have pregnancies uncomplicated by hypertension.24
What might mediate the decline in blood pressure after a first pregnancy? There may be no direct effect of pregnancy. Instead, passage through the stress of pregnancy without hypertensive complications may simply define a group of women who have more healthy vascular function. On the other hand, pregnancy may have a direct, lasting effect on blood pressure. Vascular resistance falls substantially early in pregnancy, accompanied by reductions in blood pressure.
Mechanisms likely involved in the pregnancy-related decline in vascular resistance include resistance to pressors such as angiotensin II, angiogenesis, and weight change. Angiotensin II resistance is demonstrable as early as the tenth week of pregnancy.25 It is more pronounced among women with pregnancies uncomplicated by hypertension than in women with pregnancy-induced hypertension. Second, angiogenesis may help regulate blood pressure during pregnancy.26 VEGF inhibitors, used to treat cancer patients, can induce hypertension.27 During pregnancy, proangiogenic proteins, including VEGF and PlGF, are markedly upregulated.5 Indeed, preeclampsia has been proposed to result from lack of angiogenic endothelial signaling. Preeclamptic pregnancies have high concentrations of soluble fms-like tyrosine kinase 1, which correlate negatively with blood pressure. Last, weight change may affect blood pressure. Gestational weight gain is highly variable, but on average about 1 kg is retained by 12 months postpartum or longer,28 although 15–20% of women retain at least 5 kg.29 Weight loss is associated with blood pressure decline, although this is not linear: large weight reductions after surgery have similar effects to much smaller reductions after lifestyle modification.30 Weight loss alone is an unlikely explanation for our findings, because lower blood pressure remained after adjusting for BMI and weight gain; thus, blood pressure declines prepartum compared with postpartum despite weight gain.
What remains unknown is whether and how physiologic alterations during pregnancy affect physiology after pregnancy. Does vascular reactivity remain suppressed after pregnancy? Does angiogenesis remain up-regulated or do pregnancy-related alterations in endothelial function or vascular bed size remain postpartum? Another important issue that requires further study is whether blood pressure lowering after a first birth may be weaker in black women than white women because of other risk factors.
A first birth may be accompanied by persistent lower levels of blood pressure from preconception to years after delivery, accounting for preconception levels, weight gain, secular trends, and behaviors. Few lifestyle modifications persistently lower blood pressure. Thus, pregnancy is a natural model that may provide insights into the pathophysiology of hypertension. Although the biologic mechanism for blood pressure reduction is unclear, pregnancy may create enduring alterations in endothelial function.
1. McLaughlin MK, Roberts JM. Hemodynamic changes. In: Lindheimer MD, Roberts JM, Cunningham FG, editors. Chesley’s hypertensive disorders in pregnancy. 2nd ed. Stamford (CT): Appleton & Lange; 1999. p. 69–102.
2. Williams DJ, Vallance PJ, Neild GH, Spencer JA, Imms FJ. Nitric oxide-mediated vasodilation in human pregnancy. Am J Physiol 1997;272:H748–52.
3. Anumba D, Ford GA, Boys RG, Robson SC.. The role of nitric oxide in the modulation of vascular tone in normal pregnancy. Br J Obstet Gynaecol 1996;103:1169–70.
4. Nisell H, Hjemdahl P, Linde B. Cardiovascular responses to circulating catecholamines in normal pregnancy and in pregnancy-induced hypertension. Clin Physiol 1985;5:479–93.
5. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350:672–83.
6. Ness RB, Schotland HM, Flegal KM, Shofer FS. Reproductive history and coronary heart disease risk in women. Epidemiol Rev 1994;16:298–314.
7. Ness RB, Hubel CA. Risk for coronary artery disease and morbid preeclampsia: a commentary. Ann Epidemiol 2005;15:726–33.
8. Clapp JF III, Capeless E. Cardiovascular function before, during, and after the first and subsequent pregnancies. Am J Cardiol 1997;80:1469–73.
9. Cutter GR, Burke GL, Dyer AR, Friedman GD, Hilner JE, Hughes GH, et al. Cardiovascular risk factors in young adults. The CARDIA baseline monograph. Control Clin Trials 1991;12 suppl:1–77.
10. Friedman GD, Cutter GR, Donahue RP, Hughes GH, Hulley SB, Jacobs DR Jr, et al. CARDIA: study design, recruitment, and some characteristics of the examined subjects. J Clin Epidemiol 1988;41:1105–16.
11. Lewis CE, Jacobs DR Jr, McCreath H, Kiefe CI, Schreiner PJ, Smith DE, et al. Weight gain continues in the 1990s: 10-year trends in weight and overweight from the CARDIA study. Coronary Artery Risk Development in Young Adults. Am J Epidemiol 2000;151:1172–81.
12. Steffen LM, Kroenke CH, Yu X, Pereira MA, Slattery ML, Van Horn L, et al. Associations of plant food, dairy product, and meat intakes with 15-y incidence of elevated blood pressure in young black and white adults: the Coronary Artery Risk Development in Young Adults (CARDIA) Study. Am J Clin Nutr 2005;82:1169–77.
13. Lewis CE, Smith DE, Wallace DD, Williams OD, Bild DE, Jacobs DR Jr. Seven-year trends in body weight and associations with lifestyle and behavioral characteristics in black and white young adults: the CARDIA study. Am J Public Health 1997;87: 635–42.
14. Smith DE, Lewis CE, Caveny JL, Perkins LL, Burke GL, Bild DE. Longitudinal changes in adiposity associated with pregnancy. The CARDIA Study. Coronary Artery Risk Development in Young Adults Study. JAMA 1994;271:1747–51.
15. Anderssen N, Jacobs DR Jr, Sidney S, Bild DE, Sternfeld B, Slattery ML, et al. Change and secular trends in physical activity patterns in young adults: a seven-year longitudinal follow-up in the Coronary Artery Risk Development in Young Adults Study (CARDIA). Am J Epidemiol 1996;143:351–62.
16. Sidney S, Haskell WL, Crow R, Sternfeld B, Oberman A, Armstrong MA, et al. Symptom-limited graded treadmill exercise testing in young adults in the CARDIA study. Med Sci Sports Exerc 1992;24:177–83.
17. Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr 1974;19:716–23.
18. Hardy R, Lawlor DA, Black S, Wadsworth ME, Kuh D. Number of children and coronary heart disease risk factors in men and women from a British birth cohort. BJOG 2007;114:721–30.
19. Bernstein IM, Thibault A, Mongeon JA, Badger GJ. The influence of pregnancy on arterial compliance. Obstet Gynecol 2005;105:621–5.
20. Lee-Feldstein A, Harburg E, Hauenstein L. Parity and blood pressure among four race-stress groups of females in Detroit. Am J Epidemiol 1980;111:356–66.
21. Kritz-Silverstein D, Wingard DL, Barrett-Connor E. The relation of reproductive history and parenthood to subsequent hypertension. Am J Epidemiol 1989;130:399–403.
22. Lao XQ, Thomas GN, Jiang CQ, Zhang WS, Yin P, Schooling M, et al. Parity and the metabolic syndrome in older Chinese women: the Guangzhou Biobank Cohort Study. Clin Endocrinol (Oxf) 2006;65:460–9.
23. Ness RB, Kramer RA, Flegal KM. Gravidity, blood pressure, and hypertension among white women in the Second National Health and Nutrition Examination Survey. Epidemiology 1993;4:303–9.
24. Whelton PK, He J, Appel LJ, Cutler JA, Havas S, Kotchen TA, et al. Primary prevention of hypertension: clinical and public health advisory from The National High Blood Pressure Education Program. JAMA 2002;288:1882–8.
25. Gant NF, Daley GL, Chand S, Whalley PJ, MacDonald PC. A study of angiotensin II pressor response throughout primigravid pregnancy. J Clin Invest 1973;52:2682–9.
26. Wolf M, Shah A, Lam C, Martinez A, Smirnakis KV, Epstein FH, et al. Circulating levels of the antiangiogenic marker sFLT-1 are increased in first versus second pregnancies. Am J Obstet Gynecol 2005;193:16–22.
27. Yang JC, Haworth L, Sherry RM, Hwu P, Schwartzentruber DJ, Topalian SL, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003;349:427–34.
28. Gunderson EP, Murtaugh MA, Lewis CE, Quesenberry CP, West DS, Sidney S. Excess gains in weight and waist circumference associated with childbearing: The Coronary Artery Risk Development in Young Adults Study (CARDIA). Int J Obes Relat Metab Disord 2004;28:525–35.
29. Gunderson EP, Abrams B. Epidemiology of gestational weight gain and body weight changes after pregnancy. Epidemiol Rev 2000;22:261–74.
© 2008 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
30. Aucott L, Poobalan A, Smith WC, Avenell A, Jung R, Broom J. Effects of weight loss in overweight/obese individuals and long-term hypertension outcomes: a systematic review. Hypertension 2005;45:1035–41.