Plasma homocysteine and its association with carotid
intimal-medial wall thickness and prevalent coronary heart disease:
NHLBI Family Heart Study
Michael Y. Tsai
a,*, Donna K. Arnett
b, John H. Eckfeldt
a, Roger R. Williams
1,
R. Curtis Ellison
caDepartment of Laboratory Medicine and Pathology,Uni6ersity of Minnesota Medical School,420 Delaware Street SE,Box 609,Mayo,
Minneapolis,MN 55455-0392, USA
bDepartment of Epidemiology,Uni6ersity of Minnesota Medical School,Minneapolis,MN, USA cBoston Uni6ersity School of Medicine,Boston,MA, USA
Received 29 April 1999; received in revised form 23 August 1999; accepted 28 September 1999
Abstract
Mildly elevated plasma total homocysteine (tHcy) levels have been associated with increased risk of coronary heart disease (CHD). Carotid artery intimal-medial wall thickening is a predictor of cardiovascular disease and has been previously shown to be positively associated with plasma tHcy in studies of asymptomatic subjects. In the current study we examined 1467 subjects with regard to their fasting plasma tHcy levels and intimal-medial wall thickness as measured by B-mode ultrasound and early onset CHD. The results showed that there is a significant positive association between plasma tHcy levels and carotid-artery wall thickness in participants 55 years or older even after the tHcy levels are adjusted for age, smoking and anti-hypertensive medication. The direction and magnitude of the relationship is similar although the result was not statistically significant in younger participants (B55 years). Early onset CHD at any age was not significantly different across the tHcy quintiles. The lack of an association of tHcy and CHD in the presence of a positive association with intimal-medial wall thickening may be a reflection of increased statistical power of quantitative versus qualitative traits. We conclude that the present finding of a positive association between tHcy and intimal-medial wall thickness strengthens the in vitro finding of the stimulating effect of homocysteine on vascular smooth muscle cell growth. Vascular smooth muscle cell proliferation may be an important mechanism through which mildly elevated plasma tHcy promotes atherosclerosis. © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords:Artery; Cardiovascular disease; Carotid; Epidemiology; Homocysteine
www.elsevier.com/locate/atherosclerosis
1. Introduction
Over the past decade, a large number of case-control and cross-sectional studies [1 – 3] and several prospec-tive studies [4 – 6] have all demonstrated an association between mildly elevated plasma total homocysteine (tHcy) and coronary and peripheral vascular disease. A meta-analysis showed that each 5 mmol/l increment in
tHcy is associated with a 60% increase in men and 80% increase in women in coronary heart disease (CHD) [7]. Also, a recent report indicates that plasma tHcy con-centration is a strong predictor of mortality in patients with angiographically confirmed CHD [8]. Thus, hyper-homocysteinemia is now becoming increasingly ac-cepted as an important risk factor or risk marker for atherosclerotic occlusive vascular disease.
Despite the compelling evidence, some controversies remain. For example, several recent prospective studies did not show a statistically significant association be-tween plasma tHcy and CHD incidence [9 – 11]. More-over, the mechanism by which mildly elevated plasma tHcy causes increased atherogenicity is not clearly * Corresponding author. Tel.:+1-612-626-3629; fax:+
1-612-273-6994.
E-mail address:[email protected] (M.Y. Tsai)
1Deceased, formerly of the Cardiovascular Genetics, University of
Utah, Salt Lake City, UT, USA.
understood. The NHLBI Family Heart Study (FHS) was established to identify and evaluate non-genetic and genetic determinants of CHD, preclinical athero-sclerosis, and cardiovascular disease risk factors in in-dividuals and families [12]. In the current report, we describe the association between fasting plasma tHcy and carotid intimal-medial wall thickness and early onset CHD in selected participants of the FHS.
2. Methods
2.1. Study population
FHS recruited individuals and families from three existing population-based studies, the Framingham Heart Study, the Utah Family Tree Study, and the Atherosclerosis Risk in Communities (ARIC) Study (Minneapolis, Forsyth County cohorts). Participants were recruited and examined in two phases. In phase I, conducted in 1993 – 1995, randomly selected partici-pants were sent questionnaires regarding the family history of CHD and related conditions on their par-ents, children, and siblings. Of 4679 individuals con-tacted, responses were obtained from 3150 (67%); their other family members were then contacted, and self-re-ported health data were obtained from a total of 22 908 individuals (86% of those contacted). Based on the validation of the CHD events reported in phase I, random families (n=541) and high risk families (de-termined by a calculated family risk score greater than expected and at least two validated CHD events in the family, n=610) participated in phase II at the four sites combined. All family members over 25 years of age were invited to participate in phase II, including parents, siblings, and children; a total of 5348 partici-pants were examined. The phase II examination in-cluded multiple interviews and physiologic measure-ments, as previously described [12].
For the purpose of efficiency, a subset of the total sample was selected for special assays and genotyping. The first series included sibships in which at least two of the siblings were CHD affected full sibs (n=243 sibs). The second series was defined on the basis of the carotid artery intimal-medial thickness (described in detail below), in which at least one sibling was above the 90th percentile of the distribution of intimal-me-dial thickness in the phase II population (n=231 sibs); the 90th percentile was selected based on prior experience in the ARIC study. The third series was derived from a computed risk score of CHD (i.e. the Individual Risk Score, IRS) based on measured car-diovascular risk factors and sibships in which at least one of the siblings was above the age – sex specific 80th percentile of the IRS (n=1090 sibs). The IRS was estimated for each phase II individual using
sex-spe-cific proportional hazards models to predict age-of-on-set of CHD; the variables included carotid artery intimal-medial thickness, LDL and HDL cholesterol, Lp(a), body mass index, blood pressure, hypertension, diabetes, the residual Family Risk Score (i.e. an esti-mate of familial risk not accounted for by any particu-lar risk factor domain). The final series included 200 randomly selected, unrelated subjects. These randomly selected subjects were recruited for case-control com-parison. In comparisons involving continuous vari-ables such as the intimal-medial wall thickness, inclusion of these individuals provides a full range of the distribution. Since several participants were reflected in more than one of the above groups, the final sample size was 1467.
2.2. Physiologic measurements
Sitting brachial artery blood pressure measurements were taken after a 5-min rest with a Hawksley random zero sphygmomanometer by trained and certified tech-nicians; the average of the last two of three blood pressure measurements was used in analysis. Partici-pants wore scrub suits while height and weight were measured; body mass index was calculated as weight (kg) over height (cm)2. Smoking status was based on
self-report. The FHS Lipid Laboratory, which is cer-tified by the Center for Disease Control, measured plasma cholesterol and triglyceride, and HDL choles-terol using enzymatic methods. LDL cholescholes-terol was calculated using the Friedewald equation [13].
2.3. Assay of plasma tHcy
2.4. Ultrasound measurements of carotid artery intima-media thickness
Carotid B-mode real-time ultrasound examinations were performed by trained and certified technicians with a Biosound 2000IIsa using a common scanning protocol at each of the field centers. Readings were performed at a central ultrasound reading center by trained and certified readers. Measurements of intimal-medial wall thickness were derived in the far wall of three segments of the right and left extracranial carotid arteries: (1) the common carotid artery (1 cm proximal to the dilatation of the carotid bulb); (2) the bifurcation (the 1 cm segment proximal to the flow divider); and (3) the internal carotid artery (the 1 cm segment in the internal branch distal to the flow divider). Eleven con-tiguous measurements (1 mm each) were attempted in each of the six segments. The overall average value of the mean wall thickness at the six segments combined was calculated and used in analysis.
2.5. Early onset CHD
All CHD events were validated using medical records and/or death certificates. A CHD event was defined by myocardial infarction or revascularization procedure (angioplasty or coronary artery bypass grafting). Early onset CHD was defined to be before age 55 years in men or before 65 years in women; the total CHD definition ignored age-of-onset.
2.6. Statistical methods
Descriptive statistics were calculated for all study variables. To evaluate the association of plasma tHcy
with intima-media thickness, an analysis of covariance was performed. Logistic regression was used to evaluate the relationship between tHcy and early onset CHD. Additional models were subsequently developed to fur-ther adjust for factors associated with IMT/CHD and possibly to tHcy (age, smoking, systolic blood pressure, hypertension medication, and current smoking). Signifi-cant interactions between age and tHcy with intima-me-dia thickness and tHcy with early onset CHD were found; descriptive characteristics, the intima-media models, and the CHD models were calculated sepa-rately for B55 years and ]55 years of age (55 years was selected based on the sample mean age). No other interactions between tHcy and conventional risk factors were detected. All analyses were conducted using the Statistical Analysis System (SAS) version 6.6 [16].
3. Results
Table 1 gives the descriptive characteristics for the study population. The association between plasma tHcy and demographic, anthropomorphic, biochemical, and physiologic variables is described in Table 2(a) for participants B55 years of age and in Table 2(b) for those ]55 years of age. In the younger group, tHcy levels were significantly and positively associated with LDL cholesterol, systolic blood pressure, male sex, and current cigarette smoking; age, fibrinogen, and antihy-pertensive medication use were not associated with tHcy levels. In contrast, in the older age group, age, fibrinogen, antihypertensive medication, male sex and current cigarette smoking were positively and signifi-cantly associated with tHcy; while LDL cholesterol and systolic blood pressure were not. HDL cholesterol was inversely related to tHcy in both age groups; however, neither diabetes nor body mass index demonstrated a significant association with tHcy.
Carotid artery intimal-medial wall thickness and S.D. according to plasma tHcy quintiles and age groups are listed in Table 3. In the younger age group, there was a positive but non-significant association between inti-mal-medial wall thickness and tHcy. In participants ]55 years of age, intimal-medial wall thickness was significantly greater in the upper two quintiles relative to the three lower quintiles. The older age group also had greater intimal-medial wall thicknesses relative to their younger counterparts.
Table 4 shows the logistic odds ratios for early CHD and total CHD according to quintiles of plasma tHcy concentration. An age by tHcy interaction was not detected for early onset or total CHD. Neither early onset CHD nor total CHD was associated with plasma tHcy levels.
Table 1
Descriptive characteristics of participants (n=1467)
Mean S.D.
Fibrinogen (mg/l) 78.9
LDL cholesterol (mg/dl) 129.8 36.0 49.1
HDL cholesterol (mg/dl) 15.6
120.5
Systolic blood pressure (mmHg) 19.1 Body mass index (kg/cm2) 28.6 5.7
HBP medication (%)a 0
12.6
Early-onset CHD (%) 0
21.0 0
Total CHDb
aAntihypertensive medication.
bIncludes myocardial infarction and/or revascularization
Table 2
Mean and S.D.aor frequency of demographic, anthropomorphic, biochemical and physiologic characteristics according to quintiles of plasma
tHcy concentration for participantsB55 years of age (a) and]55 years of age (b)
P-6alue
tHcy (mmol/l)
6.71–8.05 8.06–9.36
56.70 9.37–11.08 ]11.09
(a)
128 128 115
160
No. of subjects –
44.0 (7.6)
Age (years) 42.9 (7.9) 45.3 (7.3) 45.8 (7.2) 45.8 (7.7) B0.01
6
Early-onset CHD (%) 2 6 17 11 B0.001
305.2 (686) 312.3 (734) 302.1 (614)
307.3 (642) 304.7 (629)
Fibrinogen (mg/l) \0.25
123.11 (35.8) 130.3 (37.4) 127.5 (34.2) 133.7 (31.5)
LDL cholesterol (mg/dl) 121.3 (32.9) 0.03
45.4 (13.3) 47.2 (14.6) 44.8 (13.8)
50.8 (16.2) 44.0 (14.8)
HDL cholesterol (mmol/l) B0.001
111.3 (15.6)
Systolic blood pressure (mmHg) 112.2 (14.0) 114.9 (15.8) 114.5 (14.0) 117.7 (14.8) B0.01 29.0 (7.1)
Current smoker (%) 11 35 B0.001
8 6 8
No. of subjects –
63.0 (5.2) 64.0 (5.4) 64.5 (5.6) 65.2 (5.9) B0.001
Age (years) 62.4 (5.3)
329.5 (725) 337.2 (953) 335.1 (797)
330.2 (725) 358.6 (934)
Fibrinogen (mg/l) 0.006
128.99 (37.8)
LDL cholesterol (mg/dl) 132.9 (40.2) 129.8 (35.1) 133.7 (34.0) 135.0 (36.7) \0.25 56.6 (15.1)
HDL cholesterol (mg/dl) 49.5 (15.8) 49.8 (15.1) 51.2 (15.5) 48.1 (17.0) B0.001 123.9 (18.8) 128.1 (22.4) 123.3 (19.2)
123.8 (18.1) 127.7 (21.4)
Systolic blood pressure (mmHg) 0.06
27.7 (4.0) 28.4 (5.7) 28.3 (5.4)
HBP Medication (%)b 0.001
16 10 20 22
Early-onset CHD (%) 12 0.01
aS.D. is presented in parentheses. bAntihypertension medication.
4. Discussion
The present study showed that concentration of plasma tHcy is positively associated with intimal-medial wall thickness of the carotid artery, as measured by ultrasound in older participants in the NHLBI FHS. Plasma tHcy level is also positively associated with age, sex, smoking and antihypertensive medication. Adjust-ment for these risk factors, however, did not affect the observed positive finding between intimal-medial wall thickness and plasma tHcy.
A case control study of 287 pairs of participants in the Atherosclerosis Risk in Communities Study (ARIC) also demonstrated a positive association between carotid artery wall thickness and plasma tHcy in asymptomatic adults [3]. In the ARIC study, 57% of the pairs were men and 67% were]55 years old. Recently, several epidemiological and clinical studies have confi-rmed the positive association between elevated tHcy and increased carotid artery wall thickness [17 – 21]. For example, Bots et al. [17] found in the Rotterdam Study that elevated tHcy is associated with increased intimal-medial wall thickness in subjects aged 55 – 74 [17].
Simi-larly, McQuillan et al. [18] confirmed this association in 1111 subjects with a mean age of 52 years. Voutilainen et al. [19], in their study of 513 men and women aged 45 – 69, found a positive association between elevated tHcy and increased carotid artery wall thickness in men but not in women. Carotid intimal-medial wall thick-ness increases with age as well as with several major risk factors for atherosclerosis, such as serum choles-terol, cigarette smoking and blood pressure [22], and is generally considered to be a reliable measure of
subclin-Table 3
Mean carotid intimal-medial wall thickness and S.D. according to quintiles of plasma tHcy concentrationa
tHcy (mmol/l) Wall thickness Wall thickness
B55 years (mm) ]55 years (mm)
]11.09 0.74890.238 0.95290.340 9.37–11.08 0.70790.231 0.92090.376 0.88690.291 8.06–9.36 0.70090.189
6.71–8.05 0.69390.198 0.87390.316
56.70 0.69090.179 0.80690.219
aP\0.25 in ageB55 years,PB0.02 in age]55 years for
Table 4
Logistic odds ratios for early CHD and total CHD according to quintiles of plasma tHcy concentrationa
tHcy (mmol/l) AgeB55 years Age]55 years
Early CHD Total CHD Early CHD Total CHD
P-value Odds ratio
Odds ratio P-value Odds ratio P-value Odds ratio P-value
0.31 0.66 0.47 1.31
0.55 0.43
]11.09 1.30 0.41
1.92
9.37–11.08 0.21 2.03 0.17 1.25 0.52 1.38 0.33
0.41
8.06–9.36 0.14 0.42 0.15 0.65 0.26 0.80 0.51
0.05 0.19 0.05 1.22
0.19 0.57
6.71–8.05 1.14 0.69
– 1.0 – 1.0 0
56.70 1.0 1.0 –
aAdjusted for age, systolic blood pressure, antihypertensive medication, current smoking and sex.
ical atherosclerosis. Thus the current study involving almost 1467 individuals showing a positive association between intimal-medial wall thickness and plasma tHcy, combined with the large number of individuals studied in the previous publications [17 – 19] lend strong support to the hypothesis that elevated tHcy is an independent risk factor for atherosclerosis.
In the current study plasma tHcy concentration is not associated with either early onset CHD or total CHD in the overall sample despite a positive associa-tion with carotid intimal-medial thickness. This may be related to the design of the study, which includes both CHD cases and non-CHD individuals with elevated cardiovascular risk factors. The lack of an association of tHcy and CHD in the presence of a positive associa-tion with intimal-medial wall thickening may reflect enhanced power for the latter as a quantitative trait. Currently, there is controversy regarding whether homocysteine is a causative risk factor for CHD. While numerous case/control studies consistently showed that patients with myocardial infarction, stroke, and periph-eral vascular disease have higher plasma tHcy levels than controls, these results were not confirmed by some of the more recent prospective studies, such as ARIC [11], and Multiple Risk Factor Intervention Trial (MR-FIT) [10]. Our finding of a positive association between tHcy and intimal-medial wall thickness confirmed pre-vious reports [1,3] and supported the hypothesis that homocysteine may be a causative risk factor for CHD. A previous in vitro study had also demonstrated that homocysteine promotes the growth of smooth muscle cells while inhibiting endothelial cell growth [23]. Thus, the results of these studies, when taken together, may be interpreted as supporting the in vitro finding that homocysteine promotes the growth of vascular wall smooth muscle cells, resulting in increased intimal-me-dial wall thickness. However, the inconsistent findings between plasma tHcy and CHD has led to suggestions that homocysteine is merely a marker of atherosclerotic burden, and increased homocysteine may be the result of the inflammatory process associated with
atheroscle-rosis [24]. Thus, there is now an urgent need to distin-guish whether moderately elevated plasma tHcy is a marker of the process of atherosclerosis or is an active participant in enhancing the progress of atherosclerosis and/or thrombosis. Experimental animal model studies and longitudinal human studies where homocysteine is measured serially along with non-invasive means of assessing atherosclerosis such as ultrasound and/or elec-tron beam computerized tomography are needed. Since plasma tHcy has been shown to be a strong predictor of mortality in patients with angiographically confirmed CAD [8], secondary prevention trials, in which vitamin therapies are used to lower tHcy levels, are also needed. If vitamin therapies reduce the risk of CHD and/or mortality, then the hypothesis that a moderately ele-vated plasma tHcy level is as an independent risk factor for CHD will be strengthened.
Acknowledgements
This project was funded in part with federal funds from the NHLBI Family Heart Study N01-HC-25106, N01-HC-25107, N01-HC-25108, and N01-HC-25109.
References
[1] Selhub J, Jacques PF, Bostom AG, et al. Association between plasma homocysteine concentrations and extracranial carotid artery stenosis. New Engl J Med 1995;332:286 – 91.
[2] Graham IM, Daly LE, Refsum HM, et al. Plasma homocysteine as a risk factor for vascular disease. The European concerted action plan. J Am Med Assoc 1997;277:1775 – 81.
[3] Malinow MR, Nieto FJ, Szklo M, Chambless LE, Bond G. Carotid artery intimal-medial wall thickening and plasma homo-cyst(e)ine in asymptomatic adults. The Atherosclerosis Risk in Communities Study. Circulation 1993;87:1107 – 13.
[4] Stampfer MJ, Malinow MR, Willet WC, et al. A prospective study of plasma homocyst(e)ine and risk of myocardial infarc-tion in US physicians. J Am Med Assoc 1992;268:877 – 81. [5] Arnesen E, Refsum H, Bonaa KH, Ueland PM, Forde OH,
[6] Perry II, Retsum H, Morris RW, Ebrahim SB, Ueland PM, Shaper AG. Prospective study of serum homocysteine concentra-tion and risk of stroke in middle-aged British men. Lancet 1995;346:1395 – 8.
[7] Boushey CJ, Beresford SAA, Omenn GS, Motulsky AG. A quantitative assessment of plasma homocysteine as a risk factor for vascular disease: probable benefits of increasing folic acid intakes. J Am Med Assoc 1995;274:1049 – 57.
[8] Nygard O, Nordrehaug JE, Refsum H, Ueland PM, Farstad M, Vollset SE. Plasma homocysteine levels and mortality in patients with coronary artery disease. New Engl J Med 1997;337:230 – 6. [9] Alfthan G, Pekkanen J, Jauhiainen M, et al. Relation of serum homocysteine and lipoprotein (a) concentrations to atheroscle-rotic disease in a prospective Finnish population based study. Atherosclerosis 1994;106:9 – 19.
[10] Evans RW, Shaten J, Hempel JD, Cutler JA, Kuller LH. Homo-cyst(e)ine and Risk of Cardiovascular Disease in the Multiple Risk Factor Intervention Trial. Arterioscler Thromb Vasc Biol 1997;17:1947 – 53.
[11] Folsom AR, Nieto FJ, McGovern PG, et al. Prospective study of coronary heart disease incidence in relation to fasting total homocysteine, related genetic polymorphisms, and B vitamins. The atherosclerosis risk in communities (ARIC) study. Circula-tion 1998;98:204 – 10.
[12] Higgins M, Province M, Heiss G, et al. NHLBI Family Heart Study: objectives and design. Am J Epidemiol 1996;143:1219 – 28. [13] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol without the use of the preparative ultracentrifuge. Clin Chem 1972;18:499 – 502.
[14] Ubbink JB, Vermaak H, Bissbort S. Rapid high-performance liquid chromatographic assay for total homocysteine levels in human serum. J Chromatogr 1991;565:441 – 6.
[15] Garg UC, Zheng Z, Folsom AR, et al. Short-term and long-term variability of plasma homocysteine measurement. Clin Chem 1977;43:141 – 5.
[16] SAS/STAT User’s Guide, Version 6, 4th edn, SAS Institute Inc., Cary, NC, 1990.
[17] Bots ML, Launer LJ, Lindemans J, Hofman A, Grobbee DE. Homocysteine, atherosclerosis and prevalent cardiovascular dis-ease in the elderly: the Rotterdam Study. J Intern Med 1997;242:339 – 47.
[18] McQuillan BM, Beilby JP, Nidorf M, Thompson PL, Hung J. Hyperhomocysteinemia but not the C677T mutation of methylenetetrahydrofolate reductase is an independent risk de-terminant of carotid wall thickening. The Perth Carotid Ultra-sound Disease Assessment Study (CUDAS). Circulation 1999;99:2383 – 8.
[19] Voutilainen S, Alfthan G, Nyysso¨nen K, Salonen R, Salonen JT. Association between elevated plasma total homocysteine and increased common carotid artery wall thickness. Ann Med 1998;30:300 – 6.
[20] Demuth K, Moatti N, Hanon O, Benoit MO, Safar M, Girerd X. Opposite effects of plasma homocysteine and the methylenetetrahydrofolate reductase C677T mutation on carotid artery geometry in asymptomatic adults. Thromb Vasc Biol 1998;18:1838 – 43.
[21] Tonstad S, Joakimsen O, Stensland-Bugge E, et al. Risk factors related to carotid intima-media thickness and plaque in children with familial hypercholesterolemia and control subjects. Arte-rioscler Thromb Vasc Biol 1996;16:984 – 91.
[22] ARIC investigators, Heiss G, Sharrett AR, Barnes R, Chambless LE, Szklo M, Alzola C. Carotid atherosclerosis measured by B-mode ultrasound in populations: associations with cardiovas-cular risk factor in the ARIC study. Am J Epidemiol 1991;134:250 – 6.
[23] Tsai JC, Perrella MA, Yoshizumi M, et al. Promotion of vascu-lar smooth muscle cell growth by homocysteine: a link to atherosclerosis. Proc Natl Acad Sci USA 1994;91:6369 – 73. [24] Kuller LH, Evans RW. Homocysteine, vitamins, and
cardiovas-cular disease. Circulation 1998;98:196 – 9.
