Hyperhomocysteinemia as a risk factor for deep-vein thrombosis

HYPERHOMOCYSTEINEMIA AS A RISK FACTOR FOR DEEP-VEIN THROMBOSIS

MARTIN DEN HEIJER, M.D., Tun KOSTER, M.D., HENKJ. BLOM, PH.D., GERARD M.J. Bös, M.D.,

ERNEST BRIET, M.D., PIETER H. REITSMA, PH.D., JAN P. VANDENBROUCKE, M.D.,

AND FRITS R. ROSENÜAAI., M.D.

Abstract Background. Previous studies have suggest-ed that hyperhomocysteinemia may be a risk factor for venous thrombosis. Tb assess the risk of venous throm-bosis associated with hyperhomocysteinemia, we stud-ied plasma homocysteine levels in patients with a first episode of deep-vein thrombosis and in normal control subjects.

Methode. We measured plasma homocysteine levels

in 269 patients with a first, objectively diagnosed epi-sode of deep-vein thrombosis and in 269 healthy con-trols matched to the patients according to age and sex. Hyperhomocysteinemia was defined äs a plasma ho-mocysteine level above the 95th percentile in the control group (18.5 μηιοί per liter).

Results. Of the 269 patients, 28 (10 percent) had

plas-ma homocysteine levels above the 95th percentile for the controls, äs compared with 13 of the controls (matched odds ratio, 2.5; 95 percent confidence interval, 1.2 to 5.2). The association between elevated homocysteine levels and venous thrombosis was stronger among women than among men and increased with age. The exclusion of subjects with other established risk factors for thrombosis (e.g., a deficiency of protein C, protein S, or antithrombin; resistance to activated protein C; pregnancy or recent childbirth; or oral-contraceptive use) did not materially af-fect the risk estimates.

Conclusions. High plasma homocysteine levels are a

risk factor for deep-vein thrombosis in the general popu-lation. (N Engl J Med 1996;334:759-62.)

©1996, Massachusetts Medical Society.

M

·-In ciassic homocystinuria, half the vascular complica-tions are of venous origin,:ibut until reccntly it has bcen

unclear whether mild hyperhomocysteinemia is also a risk factor for venous thrombosis.2·1'5 In a casc—control

stucly, Falcon et al. found that hyperhomocysteinemia was Ά risk faclor for thrombosis in people younger than 40 years of agc.(> They rcportecl that the diffcrcncc in homocysteine levels between casc patients and control subjects was particularly evident after mcthionine load-ing (sincc mcthionine is a prccursor of homocysteine). Recently, wc found hyperhomocysteinemia to bc a risk factor for recurrcnt venous thrombosis in patients bc-twecn 20 and 70 years of age, äs compared with controls from the gcncral population.7 Although the results of

thcse studies support the hypothesis that mild hyperho-mocysteinemia is a risk factor for venous thrombosis, the studies werc not designed to cstimate the risk in the general population.

Wc measured homocysteine conccntrations in paticnls and matched control subjccls participating in the Leiden Thrombophilia Study.8"" This is a population-based

casc-control study designed to measure the effect ofsev-eral acquired and genetic risk factors for thrombosis in the general population. Bccause of the data available on the study subjects, we were ablc to investigatc wheth-er the effect of hypwheth-erhomocysteinemia was independent of other well-established risk factors for thrombosis, such äs a deficiency of protein G, protein S, or anti-thrombin; use of oral contraccptivcs; and pregnancy or From tho Dcparlmenl of Hematology, Municipal Hospital Leyenburg, The Hague (M.H., G.M.J.B.); the Departmcnts of Clinical Epidcmiology (T.K., J.P.V., F.R.R.) and Hematology (E.B., P.H.R., F.R.R.), University Hospital, Leiden; and the Laboratory of Pcdialrics and Neurology. Department of Pediatrics, Univer-sity Hospital, Nijmcgen (H.J.B.) — all in the Netherlands. Address reprint re-ciucsls to Dr. den Heijcr at the Department of Hematology, Municipal Hospital Leyenburg, P.O. Box 40551, 2504 LN The Hague, the Netherlands.

Stipported by grants from the Prevention Fund of the Netherlands (28-2263-1) and the Netherlands Hearl Foundation (89.063).

recent childbirth. Recently, resistance lo activated pro-lein G caused by a single point mutation in the factor V gcnc (factor V Leiden) has been rcported to bc the most common hercditary cause of venous thrombosis.12

Since hyperhomocysteinemia also appears to bc com-mon, we cxamincd the risk of thrombosis in pcrsons with both abnormaliücs.

METHODS

The methods by which blood samples were obtained and intervic\\' data werc colleeted have been described elsewhcre.^" The stndy pro-tocol was approved by the local elhics committce, and all participants gave thcir informed conscnt. Brietty, consccnlivc patients youngcr ihan 70 years of age wbo had a first episode of deep-vein thi ombosis, objcclivcly confirmcd (by impedancc plcthysmography, Dopplcr ul-trasonography, compression ulul-trasonography, or conlrasl venogra-phy), between 1988 and 1993 and who had no known cancer were sc-lected from the tiles of three anticoagulation clinies in the Netherlands (in Leiden, Amsterdam, and Rotterdarn). These chnics monitor the anticoagulant trcatmcnt of virtually all patients in well-dcfincd geo-graphic areas. Each patient was askcd to find bis or her own healthy control subject of the samc sex and age (within fwe years) by asking neighbors or friends. Wc restricted the prcscnl analysis to casc pa-tients and controls who were seen at the Leiden Anticoagulation Clin-ic and whose blood samples werc processed and frozcn on site with minimal dclay. (Blood samples from participants in Amsterdam and Rotterdam were also processed in Leiden, which caused delays of sev-eral hours, and homocysteine measuremcnls were therefore less accu-ratc than those measured in samples from subjects in Leiden.15)

The total homocysteine conccntration was measured in citrated plasma by automatcd high-performancc liquid chromatography with reverse-phase Separation and fluoresccnt detection (witb a Gilson 232-401 sample processor, Spcctra-Physics 8800 solvent-delivery system, and Spcctra-Physics LC 304 fluorometer). Wc used the method de-scribed by Fiskcrstrand et al.Li with somc modifications.'' If not other-wisc stated, hyperhomocysteinemia was defined äs a homocysteine lev-cl above the 95th percentile in the control group (18.5μιτκ>1 per liter).

Wc calculated matched odds ratios äs estimates of the relative risk

of thrombosis for homocysteine values above a given point, with the matching factor takcn into account. The univariate matched odds ra-tio is the rara-tio of the number of pairs of case patients and controls in which the homocysteine valne for the casc patient was above the spec-ificd level and the value for the control was below that level to the number of pairs in which the homocysteine value for the conlrol was above the spccified level and the value for the case patient was bclow that level. The 95 percent confidence intervals werc calculated from

760 THE NEW ENGLAND JOURNAL OF MEDICINE March 21, 1996

a conditional logislic-regression algorithm by thc maximum-likcli-hood method, with Egret Software. We also invcstigated a possiblc dosc-rcsponse rclation by calculating odds ratios for several ranges of homocyslcinc concentrations in a conditional logistic model. In aclcli-tion, we calculated odds ratios for men and women separately and for several age groups in order to study possiblc diffcrenccs in risk among thcsc subgroups.

Wc further explored the differences in risk between men and wom-cn by taking risk factors specific to women into account — specifical-ly, the use of oral contraccptivcs, prcgnancy, and rccent childbirth. We analyzed the risk of thrombosis among womcn less lhan 50 ycars old, both with and without thc inclusion of womcn with thesc risk fac-tors, by calculating unmatchcd odds ratios. The use of unmalchccl odds ratios was necessary because in thc restricted groups many matchcd pairs would not havc been complete. Sincc the matchcd and unmatched odds ratios diel not diffcr substantially in any oi our anal-yscs, we considered this approach justiiicd.

Wc also assessed whcthcr thc increased risk associatcd with hyper-homocysteinenna in both sexes was confoundcd by othcr risk factors, such äs a dcficicncy of protein C, protein S, or antithrombin. Wc rc-peatcd the analysis alter cxcluding subjccls with abnormally low lev-els of thesc protcins (mcasurcd, äs prcviously reported, with a single lest") and estimatcd thc risk associalcd will) hypcrhomocystcincmia in pcrsons with normal protein C, protein S, and antilhrombin Icvcls.

Finally, wc looked at the possibility of an intcraction bctwccn hy-pcrhomocysteincmia and hetcrozygosity (carrier Status) for factor V Leiden, a rathcr common dcfcct that causcs resistance to activated protein C. We analyzed this intcraction by calculating univariatc odds ratios for thrombosis m persons with both or cithcr öl thrsc risk fac-tors, äs compared with persons with neithcr risk faclor.

RESULTS

Thc ralio of male lo fcmale subjects among both thc casc paticnts and the controls was 1:1.3, and the mcan age was 44 years (ränge, 16 to 70 for the case patients and 16 to 71 for thc controls); both thcse variables were uscd in matching thc case patients and the controls.

Thc median plasma homocystcinc Icvcl in thc pa-tients was 12.9 μηιοί per litcr (ränge, 4.8 to 60.2), and that in the controls was 12.3 /xmol per liter (ränge, 6.4 to 37.5). The homocystcine concentrations of individual case paticnts and controls are shown in Figurc 1.

70 i ω 60· E 50· ω

l

8 3o^

l

Controls Case Patients

Figure 1. Plasma Homocysteine Levels in 269 Patients with Deep-Vein Thrombosis and 269 Controls.

Values shown have been rounded.

Thc 95th perccntilc of thc homocystcinc Icvcls in the control group was 18.5 μιτιοί per litcr. Of thc 269 pa-tients, 28 (10 pcrcent) exceeded this cutoff, äs com-pared with 13 (5 percent, by definition) in the control group. The matched odds ratio for decp-vcin thrombo-sis in subjects with a homocysteinc concentration abovc the 95th perccntilc, äs compared with those whose ho-mocysteinc Icvcls were at or below that value, was 2.5 (95 percent confidcnce intcrval, 1.2 to 5.2). Whcn thc cutoff was set at thc 90th pcrcentilc, thc matchcd odds ratio was 1.9 (95 percent confidence interval, 1.1 to 3.3); it was 4.0 (95 percent confidence interval, 1.4 to 12.0) whcn the cutoff was the 97.5th pcrccntile (Tablc 1).

In order to evaluate the possibility of a dose—rcsponsc relation, we stratificd thc paticnts and controls accord-ing to their homocysteine concentrations and calculated odds ratios for thrombosis in thc paticnts at thc higher levels äs compared with thosc at the lowest level. As Figurc 2 shows, the risk of thrombosis did not increase among subjects with homoeysteinc levels up to 18 μιτιοί per liter; the risk was greatly increased abovc 22 μηιοί per liter, howevcr, inclicating a thrcshold effcct rathcr than a continuous dose—rcsponsc rclation.

Odds ratios for several age groups and for men and women separately arc shown in Table 2. For both scxcs, there was a sharp increase in thc risk of thrombosis as-sociatcd with hypcrhomocysteincmia at inereasing ag-es. The overall odds ratio for thrombosis associatcd with hyperhomocysteinemia in women was 7.0 (95 percent confidence intcrval, 1.6 to 30.8), and in men it was 1.4 (95 percent confidence intcrval, 0.6 to 3.4), with the cutoff sct at thc 95th perccntilc of the homocysteine levels in thc control group (P = 0.067 for thc compari-son between the sexcs). Whcn wc calculated thc 95th pcrccntile of thc distribulion of homocystcinc levels for men and womcn separately, wc found a 95th percentilc of 17.1 μπιοί per liter among women and 20.0 μιτιοί per liter among men in thc control group. Using thcsc cut-offs for hypcrhomocystcincmia, we found an odds ratio for thrombosis of 3.8 (95 percent confidence intcrval, 1.4 to 10.2) for womcn and 1.8 (95 percent confidence intcrval, 0.6 to 5.4) for men.

Thc higher rate of hyperhomocysteinemia in wom-en than in mwom-en was prcscnl al all agcs, making it un-likcly thal thc diffcrence was cluc to risk factors spe-cific to women, such äs thc use of oral contraccptivcs, rccenf childbirth, or prcgnancy. Indecd, whcn we cx-cluclcd womcn wilh thcse risk factors, the unmatchcd odds ratio for thrombosis that was associated wilh hy-pcrhomocystcincmia (with the 95th pcrccntile ihr hol h sexes — 18.5 μηιοί per litcr — äs the cutoff for hyper-homocysleincmia) among women under thc age of 50 was 11.3 (95 perccnl confidence intcrval, 2.7 to 46.0), whcrcas it was 2.8 (95 percent confidenee interval, 0.9 to 8.7) for all women, both those wilh and l hose with-out thcsc risk factors, under llic age of 50.

Vol 334 No 12 HYPERHOMOGYSTEINEMIA AS A RISK FACTOR FOR DEEP-VEIN THROMBOSIS 761

Table 1. Pairwise Distribution of Plasma Ho-mocysteine Values in 269 Gase Patients and Their Matched Controls, According to Vari-ous Definitions of Hyperhomocysteinemia.*

CUTOTT 90τπ PCRCCNIILE (]6 6 μηιοί per liter)

CASC PATIENTS CON1RO1S

Abovc cutoff Below cutoff Above cutoff 6 pairs

20 pan s

38 pairs 205 paus

Below cutoff

Matched odds lalio foi thrombosis, I 9 (95% CI, l 1-33)

C u i o n 9 5 i i ] Pi RC f N I I I ϊ ( 1 8 5 μηιοί pci l i t e r ) C A S I P A I I F N I S CONTRO1 S

Above cutoff Below cutofi Above culoff

Bciow cutofi

Matched odds latio for thrombosis, 2 5 (95% CI, l 2-5 2)

Cu KM i 97 * i i i l Pl RU Ν Π Ι r (21 Ι μ m o l per l i t c i ) CASL i ' A I I I N l S CON1ROLS

Above cutoff Below cutoff Above cutotf Below cutoff 3 paus 1 0 pairs 25 pairs 23 1 pairs 2 patrs 4 pairs 16 pairs 247 pairs Matched odds latio foi thiombosis.

40(95%CI, 3 4-120)

^Füi cach cutolf point, subjccts classihcd äs ha\mg hypciho-mocystcmemia weie thosc with plasma homocyslcmc 3c\els above the culoll value, and subjects elassified äs not having hypcr homocystememia wcie thosc with levcls at 01 bclow thc cutoff valuc ('below culoll') Tbe pcieentiles used äs culofls were toi (he disttibution of bomocysteine valucs in the contiof gioup Odds ratlos wcrc calculated äs the nsk of thiombosis in the stibjects with hypethoinocystememia äs Lompaied with that m the subjects withotit hypci homocystcincnua CI dcnotcs conhdcncc intcnal

cighl had protcin S dcficicncy, and cight had anlithrom-bin deficicncy. After cxcluding thesc subjccts, we found a matchcd odds ratio for decp-vein ihrombosis of 2.6 (95 percent confidcnce interval, 1.2 to 5.9), äs comparcd with 2.5 (95 pcrccnt confidcnce interval, 1.2 to 5.2) \vhcn thosc subjccts werc included; this result shows that the cffect of homocystcine is largely indepenclent of thesc deficicnries in clotting-factor Inhibitors.

With respect to thc combination of factor V Leiden and hypcrhomocyslcincmia, wc calculated odds ratios for thrombosis in subjccts with both risk factors or either onc in rclation to subjects with ncithcr. A total of 47 of the pa-ticnts carricd the factor V Leiden rautalion, äs comparcd with 7 of thc controls. Thc small numbcr with both dc-fccts madc thc results statistically unstable and some-what sensitive to the cutoff chosen for clevalcd homocys-teine levels. When thc 90th pcrccntilc was uscd äs thc cutolf, thc odcLs ralio for thrombosis associated with thc presence of bolh risk factors (factor V Leiden and hypcr-homoeysteinernia) was 3.5 (95 percent confidcnce intci-val, 0.7 lo 16.9); the odds ratios foi thrombosis associated with factor V Leiden alone and hyperhomocjsleinemia alonc, talculated sepaialely, were 9.5 and 2.2, rcspcctive-ly. With the c)5th peicentile used äs thc cutofi, the odds latio foi the combination of risk factois was 2.0 (95 pei-ccnt tonddence inlcual, 0.4 to 10.9), \\heieas the odds ratios for cach i isk factor separately remamed virtualh

unchanged. The statistical uncertainty of results based on these data is reflected in the wide confidence inter-vals, which do not exclude a relative risk äs high äs 16.9.

DlSCUSSION

Our study shows that hyperhomocysteinemia is a risk factor for deep-vein ihrombosis in the gencral popula-tion. Moreovcr, our results suggest that thc association between mild hyperhomocysteinemia and venous throm-bosis is similar in degree to that reported for hyperhomo-cysteinemia and arterial vascular discase.'1'16 An unex-pected finding was the substantial increase in the risk of thrombosis at the highest plasma homocysteine levels. Our data suggest that there may be a thrcshold levcl above which homocysteine has a thrombogcnic cffect.

Falcon et al. reported that hyperhomocysteinemia was a risk factor for juvenile thrombosis.6 Our data imply that hyperhomocysteinemia is a risk factor for thrombo-sis in adult subjects äs well, since we found an increasing odds ratio with increasing age.

When we analyzed men and women separately, we found a differcnce in the risk of thrombosis associated with hyperhomocysteinemia. Even when we used diffcr-cnt cutoff points for hyperhomocysteinemia in men and women by calculating the 95th percentiles of their ho-mocysteine distributions in the control group separate-ly, wc found that thc odds ratio was roughly twice äs high for women äs for men. This suggests that women may be more susceptible to thc pathologic effects of el-cvatccl homocysteine levels, cven though their homocys-teine levels arc in gencral lowcr than those of men.1 This cffect cannot bc explained by risk factors spccific to women (such äs pregnancy, rccent childbirth, and oral-contraccpüve usc); an cffect of thesc risk factors was unlikcly in any casc bccausc thc differcnce betwccn men and women who did not havc such risk factors was even more pronouneed.

Hyperhomocysteinemia remained a risk factor for

4-o ro 3-CC (0 -o § ^ 1-:12 12-14 >14-16 >16-18 >18-22 Plasma Homocysteine (^mol/liter)

>22

Figure 2 Odds Ratio for Thrombosis According to Plasma Ho-mocysteine Level.

762 TIIL NEW ENGLAND JOURNAL ΟΓ MLDICINE iVLuch 21, 1996

Table 2. Odds Ratlos for Thrombosis Associated with Hyperhomocystememia, According to Age

and Sex.·1

conlributcs to the prcvcnlion οί ι

c-<30 30-50 »50 All agcs 0 5 ( 0 1 - 5 5 ) 1 0 ( 0 1 - 1 6 0 ) 0 7 ( 0 1 - 4 0 ) l 3 (0 3-4 6) 7 0 (0 9-56 9) 2 4 (0 8-6 8) 2 5 ( 0 5 - 1 2 9 ) o» | 5 5 ( 1 2 - 2 4 8 ) 14(06-34) 70(16-308) 2 5 ( 1 2 - 5 2 )

ΛΊΐΙι the nsk in those withoul hypcihomocystememia Cl dcnotcs conh icnce interval

l1 he odds ratio was 12 0 (95 pcrccnt confidcncc jnlcrval ! 6 to 92 3) λΊιεη Ihe cutoll used was thc 90th peicendlc in the contiol gioup

dccp-vcin thrombosis aftcr wc excludcd subjects with other wcll-established risk factors; that is, the associa-tion with thrombosis was not cxplaincd by thc prcscncc of other hercditary risk factors for thrombosis, such äs a dcficicncy of protem G, protem S, or antithrombin. The same was truc of the most common hereclitary risk fac-tor for deep vcnous thrombosis, resistancc to activalcd protcin G, sincc hypcrhomocysteinemia also increased thc risk of thrombosis in those without this abnormal-ity. We investigated a possiblc interaction between re-sistancc to activatcd protcin G (factor V Leiden) and hy-pcrhomocysteinemia. Although we found that the risk of thrombosis may be higher m carners of the mutation who havc hyperhomocysteinemia than in noncarriers with hyperhomocysteinemia, the combined effect in our subjects seemed smaller than for factor V Leiden alonc. Becausc of the small numbers mvolvcd, thc only rca-sonablc conclusion is that thc two factors do not poten-tiatc each other.

Many hypotheses have been proposed to cxplain how hyperhomocysteinemia may lead to vcnous thrombosis and atheiosclerosis. One hypothesis is that homocys-teine has a toxic effect on thc vascular cndothelium and on the clotting cascadc.1·2 Scveral in vitro studies seem

to support this view.17·18 However, virtually all these

stud-ies used amounts of homocysteine that produced high-er-than-physiologic concentrations. Alternatively, hy-perhomocysteinemia may reflect abnormal methiomne metabolism that affects the methylation of DNA and cell membranes.19

Elevated homocysteine levels may result from low lev-els of fohc acid, vitamm B6, or vitamm B u. Moreover,

sev-eral genetic alterations in enzymes involved in homocys-teine metabolism have been described.20 22 It remains

unclear whether hyperhomocysteinemia of different caus-es entails the same risk of thrombosis. Ncverthelcaus-ess, it is well knowri that vitamin supplementation lowers ho-mocysteine concentrations in almost all subjects with hy-perhomocysteinemia, rcgardlcss of thc undcrlying causc. We conclude that mild hyperhomocysteinemia is a risk factor for deep-vein thrombosis in the general pop-ulation. The ncxt question to be answercd is whether homocystcine-lowering therapy — folic acid, Vitamin

B6, or vitamin B

currcnt vcnous thrombosis.2

Wc arc tndcbtcd to Mi s T Vissci, Mi s A. \ an Bc< k, Mi s M T WB Ic Poik-l-Otholl, and Mis A de G i a a l - I k s s lui ihon ( \ c r l l i n t as-sistanco

REFERENCES

1 Ueland PM, Refsum II, Btatlstiom L Plasma homocyslcme and caidiovasetl-lai discasc In Franus RB Ji cd AlhciosLlcioüc caidiovasuicaidiovasetl-lai cltscasc hc-moslasis, and endothchal tunclion NewYoik Maitcl Dckkei, 1993 183-236 2 Rees MM, Rodgeis GM Homoeyslememid assouaüon öl a mclabolic

disoidct with vasculai discasc and thtombosis Thromb Rcs 199ΐ 71 337

59

3 Mudd SH, Skovby F, Levy HL, et al Thc natinal histoiy öl homocjslmuua

duc to cystathioninc ß-synlhase dehcicncy Am J Hum Gcnct 1985,37 l 3l

4 Bicnvcnu T, Ankn A, Chadclaux B, Kamoun P Dosage de l homocysteine plasmatiquc dans l'cxploialion des thtomboscs du sujeljatinc Picsse Mcd 1991,20985 8

5 Biallstrom L, Tengboin L, Lagcislcdt C, Isiaelsson B, Hullbcig B Plas-ma homocysteine m vcnous thtombocmbolism Hacmostasis 1991,21 51 7

6 Falcon CR, Callanco M, Pan/cn D, Maitinclli I, Mannucci PM High picv-alcnce of hypeihomocyst(c)memia in patients with juvemle vcnous llnom bosis Aitenosclei Thiomb 1994,14 1080-3

7 den Heijcr M, Blom HJ Gcints WBT, et al Is hypcthomocystcinacmia a risk factoi toi lecunent venous lluombosis' Lancet 1995,345 882 5 8 Koslei T, Roscndaal FR, Bnel E, et al Piotein C deficiency in a conliolled

scncs öl unsclecled outpalicnls an mlicqucnt but clcai nsk lactoi ioi vcnous thiombosis (Leiden Thrombophilia Study) ßlood 1995,85 2756 6l 9 Kostei T, Rosendaal FR, de Ronde II, ßnet E, Vandenbiouckc JP, Beitina

RM Venous thiombosis clue to pooi anticoagulant icsponsc to aclivatcd prolem C Leiden Thrombophiha Study Lancet 1993,342 1503 6 10 Kostei T, Blann AD, Briet E, Vandenbroucke JP, Rosendaal FR Rolc öl

clolting lactoi VIII in cllcct öl von Willcbiand lacloi on occtincncc öl deep-vein thrombosis Lancet 1995,345 152-5

1 1 Vandenbroucke JP, Koster T, Briet E, Reilsma PH, Bcitma RM, Rosendaal FR Increased risk of venous thrombosis m oial-contiaccptivc useis who aie caineis öl factoi V Leiden Imitation Lancel 1994,344 1453-7 12 Bertina RM, Koclcman BPC, Kostei T, et al Mutation in blood coagulation

factoi V associaled with resistance lo activatcd piotein C Natuie 1994 369 64-7

13 Fiskerstiand T, Refsum H, Kvalheim G, Ueland PM Homocysteine and olh ei thiols in plasma and unne aulomaled deteimmation and sample stabihly C h n C h e m 1993,39263-71

14 te Poele-Pothoff MTWB, van den ßeig M, Franken DG, et al Thiee diflei-ent melhods foi thc determination of total homocysteine in plasma Ann Clm Biochem 1995,32 218 20

15 Seihub J, Jacques PF, Boslom AG, et al Association between plasma ho-mocysteine conccntrations and cxtiaciamal caiotid-aitciy Stenosis N Engl J M e d 1995,332286 91

16 Stampfet MJ, Malmow MR, Willett WC, et al A prospcctive study öl plas-ma homocyst(e)me and nsk of myocaidial infaiction m US physicians JAMA 1992,268 877-81

17 Rodgers GM, Kane WH, Pitas RE Formation of factoi Va by atheiosde-rolic rabbit aorta mediales facloi Xa-catalyzed protluombin activation J Clm luvest 1988,81 1911-9

18 Rodgeis GM, Gönn MT Homocysteine, an atheiogenic Stimulus, leduces protem C activation by aitenal and venous endothehal cclls Blood 1990, 75 895-901

19 Blom HJ, van der Molen EF Pathobiochemical nnphcations oi hypcihomo-cystememia Fibimolysis 1994,8 SuppI 2 86-7

20 Boers GHJ, Smals AGH, Tnjbels FJM, et al Hetcrozygosity ioi homocys-tinuna m piematuie penpheral and ceiebral occlusivc aitenal disease N Engl J Med 1985,313 709- 1 5

21 Engbersen AMT, Franken DG, Boeis GHJ, Stevens EMB, Tnjbels FJM, Blom HJ Theimolabile 5,10-melhyIenetetrahydiololalc icductasc äs a cause of mild hyperhomocysteinemia Am T Hum Gcnet 1995,56 142-50

22 Frosst P, Blom HJ, Goyette P, et al A candtdate genelic nsk tactoi foi vas-cular disease a common mutation in methylenclelrahydiofolate icductasc NatGenet 1995,10 111-13

23 den Heijei M, Bös GMJ, Geirits WBJ, Blom HI Will a decrcase öl blood homocysteine by vitamin supplementation icduce the risk foi vasculai dis-ease'' Fibimolysis 1994,8 Suppl 2 91-2

24 Stampfer MJ, Willett WC Homocysteine and margmal vitamm dchciency the tmportance of adequate vitamin mlake JAMA 1993,270 2726-7 25 Stampfer MJ, Malmow MR Can lowenng homocysleme levels leduce

car-diovascular nsk'' N Engl J Med 1995,332 328-9