Skip Navigation


American Journal of Epidemiology Advance Access originally published online on June 13, 2006
American Journal of Epidemiology 2006 164(5):470-477; doi:10.1093/aje/kwj218
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/5/470    most recent
kwj218v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Bodnar, L. M.
Right arrow Articles by Roberts, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bodnar, L. M.
Right arrow Articles by Roberts, J. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

Original Contribution

Periconceptional Multivitamin Use Reduces the Risk of Preeclampsia

Lisa M. Bodnar1,2,3,4, Gong Tang5, Roberta B. Ness1,2,3, Gail Harger1,2 and James M. Roberts1,2,3

1 Department of Epidemiology, University of Pittsburgh Graduate School of Public Health, Pittsburgh, PA
2 Magee-Womens Research Institute, Pittsburgh, PA
3 Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh School of Medicine, Pittsburgh, PA
4 Department of Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, PA
5 Department of Biostatistics, University of Pittsburgh Graduate School of Public Health, Pittsburgh, PA

Correspondence to Dr. Lisa Bodnar, Department of Epidemiology, University of Pittsburgh Graduate School of Public Health, A742 Crabtree Hall, 130 DeSoto Street, Pittsburgh, PA 15261 (e-mail: bodnar{at}edc.pitt.edu).

Received for publication November 15, 2005. Accepted for publication March 3, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The objective was to assess the independent effect of regular periconceptional multivitamin use on the risk of preeclampsia. Pregnant women (n = 1,835) enrolled in the Pregnancy Exposures and Preeclampsia Prevention Study (Pittsburgh, Pennsylvania, 1997–2001) at less than 16 weeks' gestation were asked whether they regularly used multivitamins or prenatal vitamins in the past 6 months. Women were classified as users or nonusers. The unadjusted prevalence of preeclampsia was 4.4% in nonusers and 3.8% in users. After adjustment for race/ethnicity, marital status, parity, prepregnancy physical activity, and income in a multiple logistic regression model, regular use of multivitamins was associated with a 45% reduction in preeclampsia risk compared with nonuse (odds ratio (OR) = 0.55, 95% confidence interval (CI): 0.32, 0.95). Prepregnancy overweight modified this effect. After confounder adjustment, lean multivitamin users had a 71% reduction in preeclampsia risk compared with lean nonusers (OR = 0.29, 95% CI: 0.12, 0.65). In contrast, there was no relation between multivitamin use and preeclampsia among overweight women (OR = 1.08, 95% CI: 0.52, 2.25). A sensitivity analysis for unmeasured confounding by fruit and vegetable intake supported these conclusions. If confirmed by others, these results suggest that regular use of a multivitamin supplement in the periconceptional period may help to prevent preeclampsia, particularly among lean women.

body mass index; dietary supplements; obesity; pre-eclampsia; pregnancy; vitamins; women

Abbreviations: CI, confidence interval; OR, odds ratio


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Preeclampsia is a pregnancy-specific disorder with global public health significance. In developing countries, preeclampsia accounts for 20–80 percent of maternal mortality (1Go). In developed countries, perinatal mortality of infants of preeclamptic mothers is increased fivefold, and nearly 15 percent of preterm births are indicated premature deliveries for preeclampsia (2Go). Preeclampsia affects roughly 7 percent of first pregnancies (1Go).

Currently, preeclampsia is hypothesized to be a two-stage disorder with maternal-fetal interactions necessary to link the two stages (3Go). Reduced placental perfusion, secondary to abnormal implantation and subsequent reduced placental vascularization (4Go), is thought of as the first stage. The second stage, the maternal syndrome, develops in a subgroup of women with certain genetic, environmental, and/or behavioral factors (3Go) as a response to agents produced by the poorly perfused placenta (5Go). In a suitable maternal environment, oxidative stress and subsequent endothelial activation and injury result (5Go). This endothelial dysfunction initiates the coagulation cascade and ensuing multisystem sequelae (6Go), stage 2 of preeclampsia.

Nutrition has long been hypothesized to have a role in the etiology of preeclampsia (7Go), but only in the last decade has our understanding of the disorder's pathogenesis risen to a level adequate to provide testable hypotheses. It is now well understood that, while preeclampsia is clinically evident late in pregnancy, the causal exposure(s) and many of the pathophysiologic changes are present months earlier. Periconceptional exposures may be particularly relevant, as they may affect implantation and/or decidual vascular remodeling (stage 1). Nonetheless, we are aware of only one previous preeclampsia study that assessed diet around the time of conception. In this case-control study of 109 women with preeclampsia and 259 controls, diet in the past year was assessed at delivery (8Go). These investigators found that intakes of vitamin C, fruits, and vegetables below recommended values were associated with an increased risk of preeclampsia. However, because periconceptional and prenatal intakes were not assessed separately in this study, it is difficult to discern during which time period low vitamin C, fruit, and vegetable intakes were most relevant for the development of preeclampsia.

Given the paucity of data on periconceptional nutrition and preeclampsia risk, we sought to explore this association by assessing the independent effect of regular multivitamin use in the periconceptional period on the risk of preeclampsia. Multivitamins contain a number of micronutrients that have been hypothesized to be relevant for preeclampsia prevention (9Go). Further, certain nutrients found in multivitamins, such as vitamin C and vitamin E, have been given in supplement form starting in mid pregnancy and have been reported to reduce the risk of preeclampsia (10Go).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
The Pregnancy Exposures and Preeclampsia Prevention Study is a prospective cohort study designed to examine factors that predispose women to preeclampsia (11Go). Women were enrolled at less than 16 weeks' gestation from outpatient clinics at Magee-Womens Hospital in Pittsburgh, Pennsylvania (66 percent of the sample), and affiliated private practices (33 percent of the sample) from 1997 to 2001. Women aged 14–44 years, carrying singleton infants, and planning to deliver at Magee-Womens Hospital were eligible. The response rate was 72 percent. After providing informed, written consent, all subjects completed an interviewer-administered questionnaire at enrollment to collect data on sociodemographics, medical history, and health behaviors, including periconceptional dietary supplement use, physical activity, television watching, and smoking. Following delivery, medical records were abstracted to ascertain prepregnancy weight and height, blood pressures and urinary protein measurements throughout gestation, use of hypertensive medications, antepartum and delivery events, and neonatal outcomes. The study was approved by the institutional review board.

Of the 2,891 women enrolled in the study, 722 were excluded from the analysis because they had a spontaneous abortion (n = 194), had a pregnancy termination (n = 63), delivered at another hospital (n = 173), rescinded consent (n = 94), were determined after enrollment to be ineligible (n = 65), had an ectopic pregnancy or other adverse event (n = 11), or were lost to follow-up (n = 122). We further restricted the sample by excluding women with a preexisting medical condition (e.g., pregestational diabetes, chronic hypertension) (n = 56) or a toxicology screen positive for cocaine, opiates, hallucinogens, amphetamines, or barbiturates (n = 46) or women who were missing data on preeclampsia (n = 119) or covariates in the final model (n = 27). Women with a positive toxicology screen were eliminated because drug use may affect diagnostic findings of preeclampsia (e.g., raise blood pressure) and also contribute to adverse pregnancy outcomes. The choice of completing a toxicology screen was at the discretion of the clinician. Indications were relatively broad, including history of prior drug use, very poor compliance with prenatal care, and physical findings related to drug use. We also eliminated 86 observations that represented a second or third pregnancy from the same woman. There were 307 women who reported household size but failed to report household income. To reclaim these subjects in the analysis, we used a multiple imputation procedure (discussed below) (12Go). The final analytical sample was 1,835.

Definition of study variables
Primary exposure variable: periconceptional multivitamin use.
At enrollment at less than 16 weeks' gestation, women were asked, "In the past 6 months, have you taken multivitamins or prenatal vitamins regularly, at least once per week?" Interviewers were instructed not to include supplements that subjects began using during pregnancy. Women who responded that they took a multivitamin regularly in the past 6 months were asked to categorize their usual use as 1–3 times/week, 4–6 times/week, or daily. Data on dose or brand of supplement were not collected. Because 89 percent of regular supplement users reported daily use (762 of 860 total users), we classified women as users or nonusers of periconceptional multivitamin supplements. The study did not assess supplement use at any other time before the clinical onset of preeclampsia.

Primary outcome variable: preeclampsia.
"Preeclampsia" was defined as gestational hypertension and proteinuria and return of all abnormalities to normal by 12 weeks postpartum (13Go). "Gestational hypertension" was defined as systolic blood pressure persistently greater than or equal to 140 mmHg and/or diastolic blood pressure persistently greater than or equal to 90 mmHg for the first time after 20 weeks of gestation. We determined blood pressure as the average of the last five pressures obtained after hospital admission for delivery before medications or clinical perturbations that would alter blood pressure. "Proteinuria" was defined as the excretion of greater than 300 mg of protein in 24 hours, a random sample of greater than 2 g of protein or a catheterized sample of greater than 1 g of protein, or a protein:creatinine ratio of greater than 0.3.

Covariates.
Data on sociodemographic and behavioral variables were obtained from interview-based self-report. Maternal parity (nulliparous, multiparous), marital status (married, unmarried), race/ethnicity (non-Hispanic White, non-Hispanic Black, other), education (less than high school, high school or equivalent, some college), private insurance status (yes, no), household size, and income in the year before the index pregnancy were available. The "poverty index ratio" was defined as the total household income divided by the year-specific poverty threshold (14Go, 15Go). We classified women as less than or equal to 130 percent, 131–299 percent, or greater than or equal to 300 percent of the poverty index. Smoking status in the year before the index pregnancy was classified as 0, 1–10, or greater than or equal to 11 cigarettes/day. Prepregnancy body mass index (weight (kg)/height (m)2) was based on measured height and maternal self-report of prepregnancy weight at the initial visit. We classified women as "lean" or "overweight" if they had a body mass index of less than 25 or of 25 or more, respectively. Women were asked to categorize their usual amount of time spent watching television in the year before the index pregnancy as 0–10, 11–20, 21–30, or greater than 30 hours/week. Women were also asked if they engaged in any leisure-time physical activity in the year before the index pregnancy and, if so, to rate the usual intensity of this activity as low, medium, or high.

Analyses
Multivariable logistic regression was used to estimate the independent effect of regular multivitamin use on risk of preeclampsia. To determine which covariates should be entered into the full multivariable model, we used directed acyclic graphs (16Go, 17Go), theory-based causal diagrams that rely on the investigators' a priori subject-matter knowledge of the causal relations of variables to one another. We fit a parsimonious regression model by specifying a full model with potential effect modifiers and confounding variables (maternal age, race/ethnicity, parity, education, marital status, poverty index ratio, insurance status, prepregnancy smoking, prepregnancy body mass index, prepregnancy physical activity, and prepregnancy television watching). Effect modifications by race/ethnicity, smoking status, parity, and overweight were tested separately using a likelihood ratio test ({alpha} = 0.10). Potential confounders were considered to not be influential and were removed from the model if their inclusion did not satisfy our a priori change-in-estimate criterion (a change in the coefficient of >8 percent). Race/ethnicity, marital status, parity, physical activity, and the poverty index ratio met our definition of confounding and were included in the final model. Alternate specifications of covariates in the models did not alter the findings. Odds ratios were used to approximate risk ratios because preeclampsia was rare in our population (18Go).

In the multiple imputation analysis, the household income of those patients with missing household income but observed household size (307 of 1,835 total women) was imputed 10 times according to the dependence of household income on household size, marital status, maternal age, race/ethnicity, marital status, private insurance status, education, and preeclampsia status (12Go). The imputation was programmed in S-PLUS software (Insightful Corporation, Seattle, Washington), and each imputed data set was analyzed by SAS PROC LOGISTIC (19Go). The results from these 10 imputed data sets were summarized using the multiple imputation technique (12Go). The results obtained after multiple imputation were compared with those from the complete-case analysis (n = 1,528).

Because our study was observational and multivitamin use was not assigned by randomization, unmeasured or unknown confounders may have biased our effect estimate. We performed a sensitivity analysis for unmeasured confounding by fruit and vegetable intake (an unmeasured variable in our data set) that was adapted from the work of Lash and Fink (20Go). Intake of five or more servings of fruits and vegetables per day is more common among supplement users than nonusers (21Go–24Go) and is associated with a reduced risk of preeclampsia (8Go). To quantify the degree of unmeasured confounding by fruit and vegetable intake, we parameterized the relative risk due to confounding using a trapezoidal distribution. The limit of the relative risk due to confounding was calculated according to the method of Flanders and Khoury (25Go). We compared the odds ratio and 95 percent confidence interval from the conventional logistic regression model with two other estimates: the estimates obtained from the sensitivity analysis iterations, which reflected systematic error only, and the estimates obtained after a bootstrapping procedure, which reflected total error (systematic and random) (20Go). Akin to confidence intervals, a sensitivity analysis simulation interval and a bootstrapped sensitivity analysis interval were reported to represent variability in these respective point estimates (20Go).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
About 47 percent of women reported regular periconceptional use of multivitamins. Regular multivitamin users were more likely than nonusers to be at least 25 years of age, non-Hispanic White, married, better educated, nonsmokers, and of normal weight before pregnancy (table 1). Private health insurance and a household income greater than or equal to 300 percent of the poverty index were more common in users than nonusers. Compared with nonusers, regular multivitamin users were more likely to engage in physical activity before pregnancy and watched less television.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Maternal characteristics by regular periconceptional multivitamin use status, Pregnancy Exposures and Preeclampsia Prevention Study, Pittsburgh, Pennsylvania, 1997–2001

 
The overall incidence of preeclampsia was 4.1 percent. Preeclampsia was less common among women reporting regular periconceptional multivitamin use than among nonusers (table 2). After adjustment for race/ethnicity, marital status, parity, prepregnancy physical activity, and the poverty index ratio, regular use of multivitamins in the periconceptional period was associated with a 45 percent reduction in preeclampsia risk (95 percent confidence interval (CI): 0.32, 0.95). The strength of the association increased after confounder adjustment because of negative confounding by marital status, race/ethnicity, physical activity, and the poverty index ratio (26Go). In the complete-case analysis (n = 1,528), results were similar (adjusted odds ratio (OR) = 0.50, 95 percent CI: 0.27, 0.89). Further adjustment for prepregnancy body mass index, smoking, age, and the other covariates mentioned above did not meaningfully alter the results (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Association between regular periconceptional multivitamin use and the risk of preeclampsia, Pregnancy Exposures and Preeclampsia Prevention Study (n = 1,835), Pittsburgh, Pennsylvania, 1997–2001

 
Prepregnancy overweight significantly modified the effect of periconceptional multivitamin use on the risk of preeclampsia (table 3). After adjustment for confounders, lean multivitamin users had a 72 percent reduction in preeclampsia risk compared with lean nonusers of multivitamins (95 percent CI: 0.12, 0.65). In contrast, there was no relation between periconceptional multivitamin use and preeclampsia among overweight women. The complete-case analysis yielded similar findings in lean (adjusted OR = 0.22, 95 percent CI: 0.09, 0.54) and overweight (adjusted OR = 1.22, 95 percent CI: 0.54, 2.76) women after confounder adjustment. We did not observe further effect modification on the multiplicative scale by race/ethnicity, smoking, or parity.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Association between regular periconceptional multivitamin use and the risk of preeclampsia stratified by prepregnancy body mass index, Pregnancy Exposures and Preeclampsia Prevention Study, Pittsburgh, Pennsylvania, 1997–2001

 
The sensitivity analysis suggested that the conventional modeling results in the total population and in each overweight stratum were biased away from the null by systematic error due to unmeasured confounding by fruit and vegetable intake (table 4). For instance, the adjusted odds ratio of 0.55 obtained from the conventional analysis in the total population was attenuated to 0.63 (95 percent simulation interval: 0.56, 0.72) after accounting for systematic error. The simulation interval then widened after bootstrapping to account for both systematic and random error (95 percent bootstrapped sensitivity analysis interval: 0.36, 1.12). Thus, the sensitivity analysis findings indicated that multivitamin users were 0.63 times as likely as nonusers to develop preeclampsia, and that the true odds ratio likely falls between 0.36 and 1.12. The sensitivity analysis results in the total population and within each stratum of overweight generally supported the conclusions drawn in the conventional analysis.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Summary of estimates yielded by the conventional analysis and the sensitivity analysis of the effect of regular periconceptional multivitamin use on the risk of preeclampsia in the total population and stratified by prepregnancy body mass index, Pregnancy Exposures and Preeclampsia Prevention Study, Pittsburgh, Pennsylvania, 1997–2001

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
We observed that regular use of multivitamins in the periconceptional period was associated with a 45 percent reduction in preeclampsia risk. The sensitivity analysis for unmeasured confounding by fruit and vegetable intake, which suggested a 37 percent reduction in risk, supported these findings. Importantly, we observed that this protective effect of multivitamins may be limited to women who enter pregnancy with a body mass index of less than 25. Preeclampsia was 0.29 times as likely among lean women who used periconceptional multivitamins compared with lean nonusers, whereas there was no relation between multivitamin use and preeclampsia in overweight women.

We are unaware of any published study that has examined the influence of periconceptional dietary supplements on the risk of preeclampsia. Therefore, we can only indirectly compare our results with those of studies that have assessed the incidence of preeclampsia in relation to supplement use during pregnancy. Our results are consistent with those of a large uncontrolled trial conducted in 1938–1939 in London by the People's League of Health, which reported that a dietary supplement containing multiple vitamins, minerals, and halibut liver oil provided at 20 weeks' gestation reduced the odds of preeclampsia by 32 percent (95 percent CI: 11, 47) (27Go). Similarly, a multivitamin taken during pregnancy was effective in reducing the risk of gestational hypertension in a randomized, placebo-controlled trial of human immunodeficiency virus-infected women in Tanzania (28Go). In a small, double-blinded, randomized, placebo-controlled trial, supplementation with 1,000 mg/day of vitamin C and 400 IU/day of vitamin E initiated at 20 weeks' gestation among high-risk women reduced the risk of preeclampsia by 60 percent (OR = 0.4, 95 percent CI: 0.2, 0.9) compared with placebo (10Go). Supplementation with at least 1 g of calcium per day during pregnancy has been shown to reduce the risk of preeclampsia by 32 percent (OR = 0.68, 95 percent CI: 0.57, 0.81) (29Go). It is noteworthy that these results have not been consistently reproduced in all studies (30Go). Further, the antioxidant and calcium trials supplemented with pharmacologic doses of these nutrients, not the low doses typically found in multivitamins and prenatal vitamins.

It is biologically plausible that periconceptional multivitamin use protects against preeclampsia (9Go). Periconceptional exposures may influence implantation—a physiologic process known to be abnormal in preeclampsia (4Go). Implantation is characterized by vascular remodeling, heightened inflammation (31Go), oxidative stress (32Go), and rapid cell division—all of which can be influenced by nutritional status. Inadequate nutrient intake has the potential to compromise the well-regulated inflammatory response, antioxidant defenses, and DNA and protein synthesis, thereby leading to abnormal implantation and reduced placental perfusion (33Go). Thus, many nutrients found in typical prenatal vitamins and multivitamins may be implicated, including vitamin C, vitamin E, vitamin A, folic acid, calcium, vitamin D, iron, zinc, selenium, and copper. As homocysteine concentrations are known to be higher in preeclamptic pregnancies than in normal pregnancies (34Go), the folic acid in multivitamins may be an important component of the effect. Research will be needed to elucidate the specific mechanism(s) and the most relevant nutrients.

One of the most intriguing findings from our analysis was the lack of a protective effect of multivitamins among overweight women. Although this effect modification by body mass index has not been tested in previous nutrition–preeclampsia studies, it has been reported in relation to other pregnancy outcomes. Goldenberg and Tamura (35Go) found that zinc supplementation during pregnancy increased birth weight and head circumference and reduced preterm delivery in mothers with a body mass index of less than 26, but had no effect on mothers whose body mass index was 26 or more. Several observational studies have reported that adequate folate intake protects against neural tube defects among lean but not overweight women (36Go, 37Go). Like investigators of these previous studies, we can only speculate on the mechanisms underlying our findings. One potential explanation may be related to the heightened inflammation, oxidative stress, and endothelial dysfunction characteristic of preeclampsia (1Go) and the overweight pregnant state (38Go). It is possible that typical multivitamins, which contain low nutrient doses (most at the recommended dietary allowance), may not be adequate to overcome these metabolic disturbances. Future studies will be needed to test this hypothesis and others.

Our study was limited by a relatively crude measure of periconceptional multivitamin use, as this variable was not a primary interest in the full study. Supplement use defined as a binary variable, as we did, ignores possible heterogeneity in patterns of vitamin use that may be relevant for understanding the exposure–outcome relation (39Go). Unfortunately, our study did not collect data on supplement brand or dose and did not more finely classify subjects' duration or frequency of use. Even with the three-level frequency categories (1–3 days/week, 4–6 days/week, daily), 89 percent of regular users reported daily use. A larger sample of women reporting less-than-daily use would have allowed us to determine the frequency of use that was associated with the best risk profile. Moreover, as the exposure was based solely on self-report, women may have misreported their multivitamin use as regular when they used multivitamins less frequently or not at all, or they may have incorrectly defined multivitamin use started in pregnancy as periconceptional in nature. We assessed regular multivitamin use in the 6 months before enrollment, which occurred at less than 16 weeks' gestation. A smaller window of exposure assessment would have allowed us to determine which time period was the most relevant for preeclampsia prevention.

Unmeasured confounding is a particular concern in studies of dietary supplement use (39Go), as supplement use is highly associated with other healthy behaviors that may reduce the risk of disease (21Go, 40Go–42Go). We attempted to address this problem with a sensitivity analysis of unmeasured confounding by fruit and vegetable intake. Nonetheless, our effect estimate may have been further biased by other unmeasured or unknown confounders and/or measurement error in the covariate data we collected. Indeed, many of our measured covariates such as smoking status, physical activity, television watching, and income were from interview-based self-report, and misclassification was possible. Not all subjects in our study received a urine toxicology screen (260 had a negative screen of 1,835 women retained in the analysis). Although exclusion of those who tested positive for drug use (n = 46) could have caused selection bias, this was unlikely given that the retention of these women in our analysis did not alter our findings (data not shown). Because the majority of our preeclamptic cases had late-onset, mild disease without growth-restricted infants, we were unable to determine whether the protective effect of multivitamins holds for more severe cases, which carry the burden of maternal and infant morbidity and mortality associated with preeclampsia (43Go). Nonetheless, our study had several notable strengths, including its large sample size, prospective design, and adjudicated preeclampsia outcome. Unlike most investigations, our study assessed multivitamin use around the time of conception, which provides novel data in relation to preeclampsia.

Our results confirm previous findings that dietary supplements may be relevant for preeclampsia prevention, and they extend them to suggest that regular use of a multivitamin in the periconceptional period may reduce the risk of preeclampsia, particularly among lean women. If our findings are confirmed by others, they highlight a modifiable risk factor for preeclampsia for which there is a relatively inexpensive, safe, and straightforward intervention available. Our results also emphasize the importance of future investigations into the mechanisms explaining the differential effect of multivitamins by overweight status. Moreover, further understanding of the role of periconceptional dietary intake in the pathophysiology of preeclampsia will allow us to establish which nutrients may be most relevant for preventing this disorder.


   ACKNOWLEDGMENTS
 
This work was partially supported by grants PPG 2PO1 HD30367 and 5MO1 RR00056 from the National Institute of Child Health and Human Development. Dr. Bodnar was supported by grant K12 HD43441 from the same Institute, National Institutes of Health.

The authors are grateful to Dr. Timothy L. Lash and Matthew P. Fox of the Boston University School of Public Health for providing exceptional support in implementing the sensitivity analysis, including supplying SAS code needed to generate the results.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Roberts JM. Pregnancy related hypertension. In: Creasy RK, Resnik R, eds. Maternal fetal medicine. Philadelphia, PA: WB Saunders, 1998:833–72.
  2. Goldenberg RL, Rouse DJ. Prevention of premature birth. N Engl J Med 1998;339:313–20.[Free Full Text]
  3. Roberts JM, Cooper DW. Pathogenesis and genetics of pre-eclampsia. Lancet 2001;357:53–6.[CrossRef][ISI][Medline]
  4. Brosens IA, Robertson WB, Dixon HG. The role of the spiral arteries in the pathogenesis of preeclampsia. Obstet Gynecol Annu 1972;1:177–91.[Medline]
  5. Roberts JM, Hubel CA. Is oxidative stress the link in the two-stage model of pre-eclampsia? Lancet 1999;354:788–9.[ISI][Medline]
  6. Surratt N. Severe preeclampsia: implications for critical-care obstetric nursing. J Obstet Gynecol Neonatal Nurs 1993;22:500–7.[CrossRef][Medline]
  7. Chelsey LC. Hypertensive disorders of pregnancy. New York, NY: Appleton-Century-Crofts, 1978.
  8. Zhang C, Williams MA, King IB, et al. Vitamin C and the risk of preeclampsia—results from dietary questionnaire and plasma assay. Epidemiology 2002;13:409–16.[CrossRef][ISI][Medline]
  9. Roberts JM, Balk JL, Bodnar LM, et al. Nutrient involvement in preeclampsia. J Nutr 2003;133(suppl 2):1684S–92S.[Abstract/Free Full Text]
  10. Chappell LC, Seed PT, Briley AL, et al. Effect of antioxidants on the occurrence of pre-eclampsia in women at increased risk: a randomised trial. Lancet 1999;354:810–16.[ISI][Medline]
  11. Ness RB, Markovic N, Bass D, et al. Family history of hypertension, heart disease, and stroke among women who develop hypertension in pregnancy. Obstet Gynecol 2003;102:1366–71.[Abstract/Free Full Text]
  12. Little RJ, Rubin DB. Statistical analysis with missing data. Hoboken, NJ: Wiley, 2002.
  13. Report of the National High Blood Pressure Education Program Working Group on High Blood Pressure in Pregnancy. Am J Obstet Gynecol 2000;183:S1–22.[CrossRef][ISI][Medline]
  14. US Bureau of the Census. Poverty in the United States: 1990. Washington, DC: US Bureau of the Census, 1991.
  15. US Bureau of the Census. Poverty thresholds. Washington, DC: US Bureau of the Census, 2006. (http://www.census.gov/hhes/www/poverty/threshld.html).
  16. Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology 1990;10:37–48.
  17. Pearl J. Causal diagrams for empirical research. Biometrika 1995;82:669–88.[Abstract/Free Full Text]
  18. Zhang J, Yu KF. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998;280:1690–1.[Abstract/Free Full Text]
  19. SAS Institute, Inc. SAS/STAT software. Cary, NC: SAS Institute, Inc, 2000.
  20. Lash TL, Fink AK. Semi-automated sensitivity analysis to assess systematic errors in observational data. Epidemiology 2003;14:451–8.[ISI][Medline]
  21. Gunther S, Patterson RE, Kristal AR, et al. Demographic and health-related correlates of herbal and specialty supplement use. J Am Diet Assoc 2004;104:27–34.[ISI][Medline]
  22. Looker A, Sempos CT, Johnson C, et al. Vitamin-mineral supplement use: association with dietary intake and iron status of adults. J Am Diet Assoc 1988;88:808–14.[ISI][Medline]
  23. Patterson RE, Neuhouser ML, White E, et al. Cancer-related behavior of vitamin supplement users. Cancer Epidemiol Biomarkers Prev 1998;7:79–81.[Abstract]
  24. Radimer KL, Subar AF, Thompson FE. Nonvitamin, nonmineral dietary supplements: issues and findings from NHANES III. J Am Diet Assoc 2000;100:447–54.[CrossRef][ISI][Medline]
  25. Flanders WD, Khoury MJ. Indirect assessment of confounding: graphic description and limits on effect of adjusting for covariates. Epidemiology 1990;1:239–46.[Medline]
  26. Mehio-Sibai A, Feinleib M, Sibai TA, et al. A positive or a negative confounding variable? A simple teaching aid for clinicians and students. Ann Epidemiol 2005;15:421–3.[CrossRef][ISI][Medline]
  27. Olsen SF, Secher NJ. A possible preventive effect of low-dose fish oil on early delivery and pre-eclampsia: indications from a 50-year-old controlled trial. Br J Nutr 1990;64:599–609.[CrossRef][ISI][Medline]
  28. Merchant AT, Msamanga G, Villamor E, et al. Multivitamin supplementation of HIV-positive women during pregnancy reduces hypertension. J Nutr 2005;135:1776–81.[Abstract/Free Full Text]
  29. Hofmeyr GJ, Atallah AN, Duley L. Calcium supplementation during pregnancy for preventing hypertensive disorders and related problems. Cochrane Database Syst Rev 2002;CD001059. (DOI: 10.1002/14651858.CD001059).
  30. Levine RJ, Hauth JC, Curet LB, et al. Trial of calcium to prevent preeclampsia. N Engl J Med 1997;337:69–76.[Abstract/Free Full Text]
  31. Goodger AM, Rogers PA. Uterine endothelial cell proliferation before and after embryo implantation in rats. J Reprod Fertil 1993;99:451–7.[Abstract]
  32. Halliwell B. Reactive oxygen species in living systems: source, biochemistry, and role in human disease. Am J Med 1991;91:14S–22S.[Medline]
  33. Jauniaux E, Watson AL, Hempstock J, et al. Onset of maternal arterial blood flow and placental oxidative stress: a possible factor in human early pregnancy failure. Am J Pathol 2000;157:2111–22.[Abstract/Free Full Text]
  34. Mignini LE, Latthe PM, Villar J, et al. Mapping the theories of preeclampsia: the role of homocysteine. Obstet Gynecol 2005;105:411–25.[Abstract/Free Full Text]
  35. Goldenberg RL, Tamura T, Neggers Y, et al. The effect of zinc supplementation on pregnancy outcome. JAMA 1995;274:463–8.[Abstract]
  36. Shaw GM, Velie EM, Schaffer D. Risk of neural tube defect-affected pregnancies among obese women. JAMA 1996;275:1093–6.[Abstract]
  37. Werler MM, Louik C, Shapiro S, et al. Prepregnant weight in relation to risk of neural tube defects. JAMA 1996;275:1089–92.[Abstract]
  38. Ramsay JE, Ferrell WR, Crawford L, et al. Maternal obesity is associated with dysregulation of metabolic, vascular, and inflammatory pathways. J Clin Endocrinol Metab 2002;87:4231–7.[Abstract/Free Full Text]
  39. Hoggatt KJ. Commentary: vitamin supplement use and confounding by lifestyle. Int J Epidemiol 2003;32:553–5.[Free Full Text]
  40. Block G, Cox C, Madans J, et al. Vitamin supplement use, by demographic characteristics. Am J Epidemiol 1988;127:297–309.[Abstract/Free Full Text]
  41. Lyle BJ, Mares-Perlman JA, Klein BE, et al. Supplement users differ from nonusers in demographic, lifestyle, dietary and health characteristics. J Nutr 1998;128:2355–62.[Abstract/Free Full Text]
  42. Subar AF, Block G. Use of vitamin and mineral supplements: demographics and amounts of nutrients consumed. The 1987 Health Interview Survey. Am J Epidemiol 1990;132:1091–101.[Abstract/Free Full Text]
  43. Ness RB, Roberts JM. Heterogeneous causes constituting the single syndrome of preeclampsia: a hypothesis and its implications. Am J Obstet Gynecol 1996;175:1365–70.[CrossRef][ISI][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
M. A. Klebanoff and S. R. Cole
Use of Multiple Imputation in the Epidemiologic Literature
Am. J. Epidemiol., August 15, 2008; 168(4): 355 - 357.
[Abstract] [Full Text] [PDF]

Home page
 Am J EpidemiolHome page
R. M. Nilsen, S. E. Vollset, S. A. Rasmussen, P. M. Ueland, and A. K. Daltveit
Folic Acid and Multivitamin Supplement Use and Risk of Placental Abruption: A Population-based Registry Study
Am. J. Epidemiol., April 1, 2008; 167(7): 867 - 874.
[Abstract] [Full Text] [PDF]

Home page
 Reproductive SciencesHome page
J. V. Ilekis, U. M. Reddy, and J. M. Roberts
Review Article: Preeclampsia A Pressing Problem: An Executive Summary of a National Institute of Child Health and Human Development Workshop
Reproductive Sciences, September 1, 2007; 14(6): 508 - 523.
[Abstract] [PDF]

Home page
 Am J EpidemiolHome page
J. M. Catov, L. M. Bodnar, R. B. Ness, N. Markovic, and J. M. Roberts
Association of Periconceptional Multivitamin Use and Risk of Preterm or Small-for-Gestational-Age Births
Am. J. Epidemiol., August 1, 2007; 166(3): 296 - 303.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/5/470    most recent
kwj218v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Bodnar, L. M.
Right arrow Articles by Roberts, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bodnar, L. M.
Right arrow Articles by Roberts, J. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?