fewslatfeTv

Methylenetetrahydrofolate Reductase 677 C 3 T Polymorphism,Plasma Folate, Vitamin B12 Concentrations, and Risk of

Preeclampsia among Black African Women from Zimbabwe

Aleksandar Rajkovic,*,1 Kassam Mahomed,† Rima Rozen,‡ M. Rene Malinow,§Irena B. King,\ and Michelle A. Williams\ ,¶

*Departments of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas 77030; †Department of Obstetrics andGynecology, University of Zimbabwe, School of Medicine, Harare, Zimbabwe; ‡Department of Human Genetics, Pediatrics and Biology,McGill University, Montreal, Quebec, Canada; §Division of Pathobiology and Immunology, Oregon Regional Primate Research Center,

Molecular Genetics and Metabolism 69, 33–39 (2000)doi:10.1006/mgme.1999.2952, available online at http://www.idealibrary.com on

Beaverton, Oregon; \Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, Washington; and the¶ ol of P

in rev

Department of Epidemiology, University of Washington, Scho

Received October 25, 1999, and

We conducted a case control study at Harare Ma-ternity Hospital, Zimbabwe. We genotyped a total of171 cases with preeclampsia or eclampsia and 185normotensive control subjects for the methylenetet-rahydrofolate reductase (MTHFR) 677 C 3 T geno-type. The wild-type allele frequency among casesand controls was 91.2 and 91.3%, respectively. Onlyone subject (0.3%) was homozygous for the 677 C 3T MTHFR genotype and this subject had preeclamp-sia. After adjustment for confounding factors, therewas statistically no significant association betweenmaternal MTHFR genotype and risk of preeclamp-sia (adjusted odds ratio 5 1.0; 95% CI, 0.5–1.9). Inaddition, plasma homocyst(e)ine, vitamin B12, andolate concentrations were not statistically differ-nt between normotensive control subjects withild-type genotype as compared with normotensive

ubjects who were heterozygous for the mutant al-ele. Conversely, there was a strong graded associ-tion between maternal plasma folate concentra-ion and risk of preeclampsia. Women with plasmaolate concentrations less than 5.7 nmol/L experi-nced a 10.4-fold increase in risk of preeclampsia.here was no clear pattern of preeclampsia risk anditamin B12 concentrations. © 2000 Academic Press

1 To whom correspondence should be addressed at Departmentof Obstetrics and Gynecology, Baylor College of Medicine, 6550Fannin St., Smith Tower, Suite 901A, Houston, TX 77030. Fax:(713)798-5833. E-mail: rajkovic@bcm.tmc.edu.

33

ublic Health and Community Medicine, Seattle, Washington

ised form November 30, 1999

Key Words: preeclampsia; homocyst(e)ine; folate;vitamin B12; methylenetetrahydrofolate reductase.

Preeclampsia is a major cause of maternal andfetal morbidity and mortality (1). Preeclampsia is agenetically heterogenous disorder with poorly un-derstood maternal and fetal genotype interactionsleading to maternal hypertension, proteinuria, andfetal growth restriction. Finding genes that contrib-ute to the cause of preeclampsia may help us iden-tify women at increased risk for developing the dis-order.

We and others have shown that women whosepregnancies are complicated by preeclampsia havesignificantly higher levels of homocyst(e)ine as com-pared to normotensive controls (2–6). Hyperhomo-cyst(e)inemia is associated with an increased risk ofstroke, myocardial infarction, and thrombosis (7).The mechanism of homocyst(e)ine action is poorlyunderstood. Homocyst(e)ine may injure the vascularendothelium by generating hydrogen peroxide andby impairing basal nitric oxide production (8). Nitricoxide has been implicated in the etiology of pre-eclampsia (9,10) and its interaction with homocys-t(e)ine may modulate vascular response. Nutritionaland genetic factors are thought to regulate homocys-

t(e)ine levels. Deficiency of folic acid, vitamin B6,and B12 can increase plasma homocyst(e)ine concen-trations. A missense mutation 677 C 3 T in the

1096-7192/00 $35.00Copyright © 2000 by Academic Press

All rights of reproduction in any form reserved.

pfa

RjMHtlStaHbdaulaa

cdbdis2

1cnos1s

Pba

mhhtmo

lt

mdwaibtvanap

3 VIC E

MTHFR gene confers thermolability to the MTHFRenzyme and decreased enzyme activity may lead toelevated homocyst(e)ine levels (11). Caucasian andAsian individuals homozygous for the 677 C 3 TMTHFR mutation may be at increased risk for car-diovascular disease and thrombosis (12–14).

We recently reported significantly higher concen-trations of homocyst(e)ine among a ethnically homo-geneous group of black African women from Zimba-bwe (3). To further understand the role of MTHFR677 C 3 T genotype and nutrition in the etiology of

reeclampsia, we genotyped all cases and controlsor the MTHFR 677 C3 T polymorphism. Addition-lly, we measured plasma folate and vitamin B12

concentrations.

PATIENTS AND METHODS

This case-control study, approved by the Medicalesearch Council of Zimbabwe and the Human Sub-

ects Committee of the University of Washingtonedical Center, was conducted at Harare Maternityospital, from June 1995 through April 1996. Po-

ential cases were identified by daily surveillance ofabor and delivery logbooks and medical records.tudy subjects were recruited during their postpar-um stay in the hospital. Preeclampsia was defineds a sustained 15 mm Hg diastolic rise or a 30 mmg rise in systolic blood pressure. If first trimesterlood pressures were unknown, preeclampsia wasefined as persistent (6 h or more) blood pressure oft least 140/90 mm Hg. Proteinuria was defined asrine protein concentration of 0.1 g/L or more in at

east two random specimens collected at least 4 hpart. All 200 eligible cases who were approachedgreed to participate in the study.Controls were women with pregnancies uncompli-

ated by pregnancy-induced hypertension. Each dayuring the enrollment period, controls were num-ered in the order in which they were admitted andelivered within 2 h of a case, and were approachedn the order in which they were identified by re-earch personnel. Of the 201 controls approached,00 (99%) agreed to participate in the study.Blood samples were available for 191 cases and

91 control subjects. After excluding 20 cases and 6ontrol subjects with chronic hypertension diag-osed prior to pregnancy or during the first 20 weeksf the index pregnancy, 171 preeclampsia or eclamp-

4 RAJKO

ia cases (33 eclamptics and 138 preeclamptics) and85 normotensive control subjects remained fortudy.

mtc

A structured interview questionnaire, adminis-tered during participants’ postpartum hospital stay,was used to collect information on maternal socio-demographic, medical, reproductive and life-stylecharacteristics during in-person interviews. Mater-nal and infant records were reviewed to collect de-tailed information concerning antepartum, labor,and delivery characteristics, and conditions of thenewborn.

Blood samples, collected in EDTA 10 ml Vacu-tainer tubes 12–72 h postpartum, were immediatelytransported, in a cooler with ice, to the ReproductiveBiology Laboratory, Parirenyatwa Hospital, Univer-sity of Zimbabwe Medical School. Upon arrival atthe laboratory, two 1.0-ml aliquots of whole bloodwere placed in cryovials and stored at 270°C.

lasma from the remaining sample was separatedy centrifugation and divided into 1.0- to 2.0-mlliquots, placed in cryovials, and stored at 270°C.

Specimens were shipped in liquid nitrogen to theUnited States for biochemical analyses.

Genomic DNA was isolated from whole blood sam-ples and MTHFR 677 C 3 T polymorphism deter-

ined as described previously (11). Plasma totalomocyst(e)ine concentrations were measured byigh-performance liquid chromatography and elec-rochemical detection as described previously withinor modifications (15). The interassay coefficient

f variation was 7.2%. Plasma folate and vitamin B12

concentrations were measured using Bio-Rad’sQuantaphase II radioassay kit (Bio-Rad Laborato-ries, Inc., Hercules, CA). All samples were run induplicates. The interassay coefficients for folate andvitamin B12 were ,6.4 and ,5.2%, respectively. Allaboratory analyses were blinded as to case or con-rol status.

The frequency distributions of maternal sociode-ographic characteristics and medical and repro-

uctive histories according to case control statusere examined. The distribution of continuous vari-bles (e.g., plasma folate concentrations, body massndex, and maternal age), was checked and found toe approximately normal, hence parametric statis-ical analytic procedures were used. Continuousariables are presented as means 6 standard devi-tion (SD). Unadjusted mean differences in mater-al plasma folate concentrations, for instance, weressessed using Student’s t test statistics. Pearson’sroduct-moment correlation coefficient was used to

T AL.

easure the closeness of a linear relationship be-ween maternal plasma homocyst(e)ine and folateoncentrations. Comparison of categorical variables

gmfsedsutv

ieslaquCwt

PRE

was made between case and controls subjects usingx2 or Fisher’s exact tests. To estimate the relativeassociation between varying concentrations of folateand vitamin B12 and risk of preeclampsia, we cate-orized each subject according to quartiles deter-ined by the distribution of plasma concentrations

or each of the analytes among normotensive controlubjects (16). Using the lowest quartile as the refer-nt group, odds ratios and their attendant confi-ence intervals were estimated. The Mantel exten-ion test for linear trend in proportions was used innivariate analyses to test for a linear component ofrend in risk between preeclampsia and folate anditamin B12 concentrations. Logistic regression pro-

cedures were used to calculate maximum likelihoodestimates for the coefficients and their standard er-rors were used to calculate odds ratios and confi-dence intervals (CI), adjusted for confounders (17).In multiple logistic regression models, significancefor monotonic trends was assessed by treating thefour quartiles as a continuous variable after assign-ing a score as its value (17). To assess confounding,we entered variables into a logistic regression modelone at a time and then compared the adjusted andunadjusted odds ratios. Final logistic regressionmodels included covariates that altered unadjustedodds ratios by at least 10%. Results of analysesrestricted to preeclamptics (i.e., excluding the rela-tively small number of eclamptics) were not materi-ally different; hence, we report results for pre-eclampsia and eclampsia cases combined.Hereinafter, we use the terms preeclampsia or pre-eclamptics when referring to the combined group ofcases. All reported P values are two tailed, andconfidence intervals were calculated at the 95%level.

RESULTS

Maternal sociodemographic, medical, and othercharacteristics are presented in Table 1. The fre-quency of the homozygous 677 C3 T (T/T) genotypen this population of black Zimbabwean women isxceedingly low with only one subject (a preeclamp-ia case) possessing the genotype. The wild-type al-ele frequency among cases and controls was 91.2nd 91.3%, respectively. The mutant allele fre-uency was 8.8% for cases and 8.7% for controls. Thenadjusted odds ratio of preeclampsia was 0.9 (95%

MTHFR AND

I, 0.5–1.7) for women with genotype C/T comparedith those women with C/C genotype (Table 2). Af-

er adjustment for confounding by maternal age,

parity, body mass index, gestational age at delivery,and reported use of prenatal vitamin, there re-mained no apparent association between maternalMTHFR genotype and risk of preeclampsia (adjust-ed odds ratio 5 1.0; 95% CI, 0.5–1.9).

Plasma homocyst(e)ine, folate, and vitamin B12

concentrations were not different for normotensivecontrol subjects who were homozygous C/C as com-pared with those normotensive subjects who wereheterozygous for the mutant allele (C/T) (Table 3).Among subjects with the C/C genotype, meanplasma homocyst(e)ine concentrations were 27%higher among preeclamptics as compared with con-trols. Also among these subjects, plasma folate con-centrations were 37% lower on average among casesas compared with controls. Plasma vitamin B12 con-centrations were similar for cases and controls.Among subjects with the C/T genotype, maternalmean plasma homocyst(e)ine concentrations werestatistically significantly higher among cases ascompared with controls. Plasma folate concentra-

TABLE 1Distribution of Preeclampsia Cases and Normo-

tensive Control Subjects According to SelectedCharacteristics, Harare, Zimbabwe 1995–1996

Characteristics

Preeclampsiacases

(n 5 171)

Controlsubjects

(n 5 185)

n % n %

Maternal age (years),19 17 9.9 28 15.1a

19–34 132 76.7 146 78.9$35 23 13.4 11 6.0

Unmarried 21 12.2 26 14.1Nulliparous 92 53.5 71 38.4a

Previous miscarriage 19 11.4 16 8.7Previous stillbirth 16 9.3 8 4.3Worked during pregnancy 51 37.0 61 33.2No use of multivitamins in

pregnancy 117 68.0 93 50.3a

Preterm delivery (,37weeks) 77 44.8 30 16.2a

Body mass indexb,c 27.1 6 4.4 25.3 6 3.8a

Mid-arm circumference (cm)c 26.8 6 3.3 25.2 6 2.5a

Gestational age (weeks)c 36.3 6 2.9 38.5 6 3.2a

a P values ,0.05.b Body mass index at delivery 5 weight (kg)/height (m)2.c Mean 6 SD.

35ECLAMPSIA

tions were 36% lower among cases as compared withcontrols, while vitamin B12 concentrations were sim-ilar for the two groups. Among control subjects,

s

B

age, p

plasma homocyst(e)ine concentrations were nega-tively correlated with plasma folate concentrations(r 5 20.299, P , 0.0001), and weakly but notignificantly correlated with vitamin B12 levels (r 5

20.119, P 5 0.108). Among preeclampsia cases,homocyst(e)ine concentrations were not correlatedwith folate (r 5 20.04, P 5 0.551), or with vitamin

12 (r 5 0.065, P 5 0.395).There was a strong, graded association between

maternal plasma folate concentrations and risk ofpreeclampsia (unadjusted P for linear trend in riskacross quartiles, ,0.0001). After adjusting for po-tential confounding by maternal age, parity, gesta-tional age at delivery, body mass index, and reporteduse of prenatal vitamins, we noted that women inthe lowest quartile experienced a 6.7-fold increasedrisk of preeclampsia as compared with women in thehighest quartile (OR 5 6.7; 95% CI, 3.0–15.1). Com-pared with women in the highest quartile of plasmafolate, subjects in the lowest decile (40 preeclampsiacases and 18 controls with plasma folate concentra-tions ,5.7 nmol/L) experienced a 10.4-fold increase

TAMean Maternal Plasma Homocyst(e)ine, Folate, an

C 3 T Genotype and Preeclampsia Case-

C/C g

Casesn 5 142

Plasma homocyst(e)ine (mmol/L) 12.5 6 5.4Plasma folate (nmol/L) 9.9 6 5.7

TAFrequency of MTHFR 677 C 3 T Genotypes and

Genotype

Preeclampsia(n 5 171) Cont

n % n

C/C 142 83.0 151C/T 28 16.4 32T/T 1 0.6 0

Note. DNA not available for 2 controls.a Unadjusted odds ratio and 95% confidence interval.b Odds ratio and 95% confidence interval adjusted for maternal

vitamin use.

36 RAJKO

Plasma B12 (pmol/L) 264 6 159

a P , 0.05 for case vs control comparison within genotype groupings

in risk of preeclampsia (adjusted odds ratio, 10.4;95% CI, 3.8–28.3). Each 1 nmol/L increase in folatecorresponded to a 10% reduction in risk of pre-eclampsia (adjusted OR, 0.9; 95% CI, 0.8–0.9, Pvalue ,0.0001). As can be seen in Table 4, there wasno clear pattern of preeclampsia risk according tovarying plasma vitamin B12 concentrations.

In Table 5 we summarize the results from analy-ses of the combined effect of the C/T genotype andlow plasma folate (the lowest quartile) on the occur-rence of preeclampsia, respectively. Compared towomen with normal plasma folate and the C/C ge-notype, women with low folate and the C/T genotypeexperienced a 2.1-fold increased risk of preeclampsia(adjusted odds ratio, 2.1; 95% CI, 0.8 to 5.5) al-though this association did not reach statistical sig-nificance. Results from these analyses suggest thatmaternal folate concentration is a more importantdeterminant of preeclampsia risk than MTHFR ge-notype in this population. Women with low plasmafolate concentrations and the C/C genotype experi-enced an almost 4-fold increase in risk of preeclamp-

3min B12 Concentrations According to MTHFR 677ols Status, Harare, Zimbabwe 1995–1996

e C/T genotype

Controlsn 5 151

Casesn 5 28

Controlsn 5 32

9.9 6 3.8a 14.0 6 5.2 9.9 6 3.2a

13.6 6 7.4a 8.4 6 3.4 11.4 6 6.5a

2k of Preeclampsia among Zimbabwean Women

5 183)

ORa (95% CI) ORb (95% CI)%

82.5 1.0 referent 1.0 referent17.5 0.9 (0.5–1.7) 1.0 (0.5–1.9)0.0 — —

arity, body mass index, gestational age at delivery, and prenatal

T AL.

BLEd VitaContr

enotyp

BLERis

rols (n

VIC E

258 6 204 261 6 224 242 6 158

.

hTaol(ptpwtpTm

pHftp

6vwc

age, p

al age

sia as compared with the referent group (adjustedodds ratio, 3.9; 95% CI, 2.1 to 7.2).

DISCUSSION

Our case-control study shows that the common677 C 3 T polymorphism is not associated withpreeclampsia among black African women fromZimbabwe. In our population the frequency of theT677 allele was very low at 8.8%. Only 0.3% of thepopulation under study was homozygous for the 677C 3 T MTHFR polymorphism while ;17% were

eterozygous. Previous studies have reported high677 allele frequencies of 24–40% among Europe-ns while low frequencies in sub-Saharan Africansf 6.6–7.3% (18,19). In U.S. blacks from South Caro-ina the frequency of the T677 allele was also low11%) (20). The lack of 677 C3 T MTHFR polymor-hism association with preeclampsia event argueshat this particular variant may not be useful inredicting the risk of preeclampsia among blackomen of African origin. Online Mendelian Inheri-

ance (OMIM) lists the 677 C3 T MTHFR polymor-

TAOdds Ratios (OR) and 95% Confidence Intervals (C

Plasma Folate and Vitamin B12 Conc

Preeclampsia(n 5 171)

n %

Folate (nmol/L)Quartile 4 [.16.5] 16 9.3Quartile 3 [11.8–16.5] 24 14.0Quartile 2 [7.7–11.7] 59 34.3Quartile 1 [,7.7] 72 42.4Bottom 10% (,5.7) 40 23.4

LinVitamin B12 (pmol/L)

Quartile 4 [.281.9] 57 33.1Quartile 3 [205.0–281.9] 28 16.3Quartile 2 [164.3–204.9] 30 17.4Quartile 1 [,164.3] 57 33.1Bottom 10% (,126.6) 29 17.0

Lin

a Unadjusted odds ratio and 95% confidence interval.b Odds ratio and 95% confidence interval adjusted for maternal

vitamin use.c Odds ratio and 95% confidence interval adjusted for gestation

MTHFR AND

hism as a potential risk factor for preeclampsia.he association between 677 C 3 T MTHFR poly-orphism and preeclampsia has been previously re-

orted among Japanese and Italian women (21,22).owever two other studies on Caucasian population

rom the United States showed no association be-ween 677 C3 T MTHFR polymorphism and risk ofreeclampsia (23).We hypothesized that women heterozygous for

77 C 3 T MTHFR polymorphism may have ele-ated homocyst(e)ine levels as compared to womenith wild-type genotype. In addition, nutrient defi-

iency in folate and vitamin B12 may cause greaterelevation in homocyst(e)ine levels in women het-erozygous for the 677 C 3 T polymorphism. Con-versely, our results indicate that women heterozy-gous for the 677 C 3 T MTHFR polymorphism didnot have significantly different concentrations of ho-mocyst(e)ine, folate, or vitamin B12 when comparedto wild-type controls. Although the odds ratio ofdeveloping preeclampsia in women with C/T geno-type increased from 1.4 to 2.1 with low folate,women with wild-type genotype had an increasedrisk from 1.0 to 3.93 in the presence of low folate.Women heterozygous for the 677 C 3 T MTHFR

4Preeclampsia According to Quartile of Maternaltions, Harare, Zimbabwe, 1995–1996

Controls5 185)

ORa (95% CI) ORb (95% CI)%

24.9 1.0 referent 1.0 referent24.9 1.5 (0.7–3.4) 1.2 (0.5–3.0)24.9 3.7 (1.8–7.8) 3.8 (1.7–8.6)25.4 4.4 (2.1–9.2) 6.7 (3.0–15.1)9.7 6.4 (3.0–13.8) 10.4 (3.8–28.3)c

end P value: ,0.0001 ,0.0001

24.9 1.0 referent 1.0 referent24.9 0.5 (0.3–0.9) 0.7 (0.3–1.3)24.9 0.5 (0.3–1.0) 0.7 (0.4–1.4)25.4 1.0 (0.6–1.8) 1.1 (0.6–2.0)9.7 1.3 (0.6–2.6) 1.5 (0.7–3.5)

end P value: 0.9205 ,0.9514

arity, body mass index, gestational age at delivery, and prenatal

at delivery and prenatal vitamin use.

37ECLAMPSIA

BLEI) of

entra

(n

n

4646464718

ear tr

4646464718

ear tr

PRE

polymorphism in this population do not have ele-vated homocyst(e)ine levels or higher risk of pre-eclampsia.

aCStwStptdwqbhrfla

betmpWhttc

as TT.

age, p

We previously reported in a small group of pre-eclamptic patients from Cleveland, Ohio, that folatelevels did not significantly differ between cases andcontrols. Powers et al. (23) also found no associationmong preeclampsia and folate levels in a group ofaucasian women from Pittsburgh, Pennsylvania.ince low folate levels may cause elevated homocys-(e)ine, thereby increasing the risk of preeclampsia,e examined folate levels and risk of preeclampsia.ince reference values for postpartum folate concen-rations are limited, we calculated odds ratios forreeclampsia for different quartiles of folate concen-rations. The risk of preeclampsia increased withecreasing folate concentrations in a linear fashion,ith odds ratios ranging from 1.2 in the seconduartile to 6.7 in the lowest quartile. Women in theottom 10% for folate concentration experienced theighest risk of preeclampsia, with an adjusted oddsatio of 10.4. It is possible that diet deficient inolate-rich foods in this group of women accounts forow folate levels not usually observed in nutrition-lly richer diets in North America and Europe.Although preeclampsia is characterized clinically

y maternal high blood pressure, proteinuria, anddema, women with the disorder are also more likelyhan normotensive pregnant women to experienceetabolic disturbances similar to those seen in non-

regnant patients with coronary heart disease.omen with preeclampsia may also experience

igher risk of ischemic heart disease as compared to

TAEvaluation of the Joint Association between Ma

Folate in Relation to the Risk of Pre

Exposed groups

Preeclampsia(n 5 171)

n %

Normal folatea and CC genotype 83 48.5Normal folate and CTb genotype 15 8.8Low folate and CC genotype 59 34.5Low folate and CT genotype 14 8.2

Note. DNA not available for 2 controls.a Low folate is defined as maternal plasma concentrations ,7.b CT genotype includes 1 case subject (the only subject) who wc Unadjusted odds ratio and 95% confidence interval.d Odds ratio and 95% confidence interval adjusted for maternal

vitamin use.

38 RAJKO

he general population (24). It is therefore likelyhat genetic risk factors associated with cardiovas-ular disease and thrombosis may also have rele-

vance to preeclampsia. Although low folate as wellas elevated homocyst(e)ine levels are risk factors forpreeclampsia among Zimbabwean women, homocys-t(e)ine concentrations did not correlate with folatelevels in this study. Since a multitude of genetic andnutritional factors control homocyst(e)ine levels, itis possible that genetic heterogeneity of preeclamp-sia and hyperhomocyst(e)inemia accounts for thelack of an expected association between homocys-t(e)ine and folate.

Hyperhomocyst(e)inemia may also represent amarker for interrelated and disrupted metabolicpathways which may have no effect on folate con-centrations. It is possible that other, yet unknown,genetic polymorphism(s) in MTHFR and other genesare specific to the African population and contributeto elevated homocyst(e)ine levels. Sequence analysisof the MTHFR and other genes encoding enzymesimportant in the folate pathway may yield clinicallyrelevant polymorphisms specific to black popula-tions with cardiovascular disease and preeclampsia.

ACKNOWLEDGMENTS

This research was supported in part by awards from the Na-tional Institutes of Health [FIC T37-TW00049 (M.A.W.), HD R01-32562 (M.A.W.), HD R01-34888 (M.A.W.), and P51-RR-00163(M.R.M.)]. The authors thank Sarah Mudzamiri, Simon Mad-zime, and Hong Tang for their expert technical assistance.

5l MTHFR 677 C 3 T Genotype and Low Plasmapsia, Harare, Zimbabwe, 1995–1996

Control(n 5 183)

Unadjusted ORcAdjusted ORd

& 95% CI%

63.4 1.01 1.02

10.4 1.1 1.4 (0.6–3.4)19.1 2.4 3.9 (2.1–7.2)7.1 1.5 2.1 (0.8–5.5)

/L (lowest quartile of the control distribution).

arity, body mass index, gestational age at delivery, and prenatal

T AL.

BLEternaeclam

n

116193513

7 nmol

VIC E

REFERENCES

1. Roberts JM, Redman CWG. Pre-eclampsia: More than preg-nancy-induced hypertension. Lancet 341:1447–1451, 1993.

PRE

2. Rajkovic A, Catalano PM, Malinow MR. Elevated homocys-t(e)ine levels with preeclampsia. Obstet Gynecol 90:168–171, 1997.

3. Rajkovic A, Mahomed K, Malinow MR, Sorenson TK, WoelkGB, Williams MA. Plasma homocyst(e)ine concentrations ineclamptic and preeclamptic African women postpartum. Ob-stet Gynecol 94:355–360, 1999.

4. Sorensen TK, Malinow MR, Williams MA, King IB, LuthyDA. Elevated second-trimester serum homocyst(e)ine levelsand subsequent risk of preeclampsia. Gynecol Obstet Invest48:98–103, 1999.

5. Powers RW, Evans RW, Majors AK, Ojimba JI, Ness RB,Crombleholme WR, Roberts JM. Plasma homocysteine con-centration is increased in preeclampsia and is associatedwith evidence of endothelial activation. Am J Obstet Gynecol179:1605–1611, 1998.

6. Dekker GA, de Vries JIP, Doelitzsch PM, Huijgens PC, vonBlomberg BME, Jakobs C, van Geijn HP. Underlying disor-ders associated with severe early-onset preeclampsia. Am JObstet Gynecol 173:1042–1048, 1995.

7. Malinow MR. Hyperhomocyst(e)inemia. A common and eas-ily reversible risk factor for occlusive atherosclerosis. Circu-lation 81:2004–2006, 1990.

8. Stamler JS, Osborne JA, Jaraki O, Rabbani LE, Mullins M,Singel D, Loscalzo J. Adverse vascular effects of homocys-teine are modulated by endothelium-derived relaxing factorand related oxides of nitrogen. J Clin Invest 91:308–318,1993.

9. Yallampalli C, Garfield RE. Inhibition of nitric oxide syn-thesis in rats during pregnancy produces signs similar tothose of preeclampsia. Am J Obstet Gynecol 169:1316–1320,1993.

10. Arngrimsson R, Hayward C, Nadaud S, Baldursdottir A,Walker JJ, Liston WA, Bjarnadottir RI, Brock DJ, GeirssonRT, Connor JM, Soubrier F. Evidence for a familial preg-nancy-induced hypertension locus in the eNOS-gene region.Am J Hum Genet 61:354–362, 1997.

11. Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Mat-thews RG, Boers GJ, den Heijer M, Kluijtmans LA, van denHeuvel LP, et al. A candidate genetic risk factor for vasculardisease: A common mutation in methylenetetrahydrofolatereductase. Nature Genet 10:111–113, 1995.

12. Morita H, Taguchi J, Kurihara H, Kitaoka M, Kaneda H,Kurihara Y, Maemura K, Shindo T, Minamino T, Ohno M,Yamaoki K, Ogasawara K, Aizawa T, Suzuki S, Yazaki Y.

MTHFR AND

Genetic polymorphism of 5,10-methylenetetrahydrofolatereductase (MTHFR) as a risk factor for coronary arterydisease. Circulation 95:2032–2036, 1997.

13. Kluijtmans LA, van den Heuvel LP, Boers GH, Frosst P,

Stevens EM, van Oost BA, den Heijer M, Trijbels FJ, RozenR, Blom HJ. Molecular genetic analysis in mild hyperhomo-cysteinemia: A common mutation in the methylenetetrahy-drofolate reductase gene is a genetic risk factor for cardio-vascular disease [see comments]. Am J Hum Genet 58:35–41, 1996.

14. Gallagher PM, Meleady R, Shields DC, Tan KS, McMasterD, Rozen R, Evans A, Graham IM, Whitehead AS. Homo-cysteine and risk of premature coronary heart disease. Ev-idence for a common gene mutation. Circulation 94:2154–2158, 1996.

15. Malinow RM, Sexton G, Averbuch M, Grossman M, WilsonD, Upson B. Homocyst(e)inemia in daily practice: Levels incoronary artery disease. Coronary Artery Dis 1:215–220,1990.

16. Mantel N. Chi-square tests with one degree of freedom:Extensions of the Mantel-Haenszel procedure. J Am StatAssoc 48:690–700, 1963.

17. Rothman K. Modern Epidemiology. Boston: Little Brown,1986.

18. Pepe G, Camacho Vanegas O, Giusti B, Brunelli T, MarcucciR, Attanasio M, Rickards O, De Stefano GF, Prisco D,Gensini GF, Abbate R. Heterogeneity in world distributionof the thermolabile C677T mutation in 5,10-methylenetet-rahydrofolate reductase. Am J Hum Genet 63:917–920,1998.

19. Schneider JA, Rees DC, Liu YT, Clegg JB. Worldwide dis-tribution of a common methylenetetrahydrofolate reductasemutation. Am J Hum Genet 62:1258–1260, 1998.

20. Stevenson RE, Schwartz CE, Du YZ, Adams MJ Jr. Differ-ences in methylenetetrahydrofolate reductase genotype fre-quencies, between Whites and Blacks. Am J Hum Genet60:229–230, 1997.

21. Sohda S, Arinami T, Hamada H, Yamada N, Hamaguchi H,Kubo T. Methylenetetrahydrofolate reductase polymor-phism and pre-eclampsia. J Med Genet 34:525–526, 1997.

22. Grandone E, Margaglione M, Colaizzo D, Cappucci G, Pala-dini D, Martinelli P, Montanaro S, Pavone G, Di Minno G.Factor V Leiden, C . T MTHFR polymorphism and geneticsusceptibility to preeclampsia. Thromb Haemostasis77:1052–1054, 1997.

23. Powers RW, Minich LA, Lykins DL, Ness RB, CrombleholmeWR, Roberts JM. Methylenetetrahydrofolate reductase poly-morphism, folate, and susceptibility to preeclampsia. J SocGynecol Invest 6:74–79, 1999.

39ECLAMPSIA

24. Jonsdottir LS, Arngrimsson R, Geirsson RT, Sigvaldason H,Sigfusson N. Death rates from ischemic heart disease inwomen with a history of hypertension in pregnancy. ActaObstet Gynecol Scand 74:772–776, 1995.