Hyperhomocysteinaemia: A Risk Factor for Extracranial Carotid Artery Atherosclerosis R. Clarke I, D. Fitzgerald 2, C. O'Brien 2, C. O'Farrell 2, G. Roche 3, R. A. Parker 4, I. Graham I
Department of Cardiology, The Adelaide Hospital, Dublin1; The Department of Vascular Medicine, St. Mary's Hospital, Dublin2; The National Screening Centre for Metabolic Diseases, The Children's Hospital, Temple Street, Dublin ~and the Department of Preventive Medicine, Vanderbilt University, Nashville, Tennessee, U.S.A 4. Abstract Hyperhomocysteinemia arising from impaired methionine metabolism, and usually due to a deficiency of cystathionine B-synthase is a significant and independent risk factor for symptomatic vascular disease. It is not known if hyperhomocysteinemia in apparently healthy asymptomatic subjects is associated with atherosclerosis and whether such a relationship is independent of conventional risk factors. The prevalence of asymptomatic extracranial carotid artery atherosclerosis was determined by duplex ultrasound examination in 25 obligate heterozygotes with respect for cystathionine Bsynthase deficiency (whose children were known to be homozygous for this genetic defect) and in 21 controls. Hyperhomocysteinemia was determined by a standard methionine-loading test and conventional risk factors were also recorded. Twelve of 25 obligate heterozygotes and 8 of 21 normal controls had evidence of extracranial carotid artery atherosclerosis. Hyperhomocysteinemia as a genetic trait was not a significant risk marker, but the actual homocysteine level was associated with an increased risk of carotid disease. After adjustment for the effects of other significant risk factors, the odds ratio ofhyperhomocysteinemia for carotid disease was 1.038 per unit increase in homocysteine level (P=0.03). Hyperhomocysteinemia is a weak risk factor for asymptomatic extracranial carotid atherosclerosis and the relative risk a~ssociated with this genetic trait is less than that observed in a study of patients pres-enting with clinical manifestations of vascular disease. between hyperhomocysteinemia and premature cerebrovascular5-s,peripheral vascular5,6,9and coronary vascular disease ~,1°-15. We have shown that hyperhomocysteinemia is a significant risk marker for premature cardiovascular disease including coronary heart disease and that this effect is independent of that of conventional risk factors5. In several of these studies5,6, hyperhomocysteinemia was usually associated with reduced cystathionine 13-synthase activity suggesting heterozygous inheritance of this genetic defect. Although vascular disease subjects are more likely than normal to have hyperhomocysteinemia, it is not known whether apparently healthy heterozygotes for this condition are at increased risk of vascular disease. Mudd et al conducted an international survey based on questionnaire responses of 394 obligate heterozygotes and found that only 5% of obligate heterozygotes were likely to have fatal vascular event before the age of 501~. The design and conclusions of this study have been criticized xT. In contrast, Rubba et al reported a higher prevalence of arteriosclerosis as assessed by noninvasive ultrasound methods in both homozygous and heterozygous subjects from 8 families with cystathionine 13-synthase deficiency is. The aim of this study was to ascertain the relationship between hyperhomocysteinemia and asymptomatic atherosclerosis as assessed by duplex ultrasound of the extracranial arteries. High resolution duplex ultrasound is sensitive in detecting minor atherosclerotic disease and can be used to make direct measurements of arterial wall thickness ~9,2°. This technique was used to determine the prevalence of extracranial carotid artery atherosclerosis in obligate heterozygotes (whose children were known to be homozygous for homocys-
Introduction Cardiovascular diseases are the leading cause of death in western countries 1. Several studies have shown that the lesions of atherosclerosis begin years or even decades before the onset of clinically observable signs and symptoms2'3. The occurrence of hypercholesterolemia, hypertension and cigarette smoking are established risk factors that have been used to identify asymptomatic people who are likely to develop cardiovascular disease2~. Homocystinuria, an autosomal recessive condition, is usually due to a deficiency of cystathionine B-synthase, an enzyme which converts methionine to cystathionine4. Persons who are homozygous for this condition have markedly elevated blood homocysteine levels and have a significantly increased risk of premature vascular disease4. In addition to classical cystathionine B-synthase deficiency, several rarer genetic defects and deficiency of nutritional factors (such as folate and vitamin B~2) can affect homocysteine levels4. Heterozygotes for homocystinuria show abnormal elevations of homocysteine (hyperhomocysteinemia) in response to oral methionine-loading. Recently hyperhomocysteinemia has been implicated as a risk factor for vascular disease. Several studies reported an association Supported in part by a grant from the Health Research Board of Ireland. Address reprint requests to: Prof. Ian Graham, Department of Cardiology The Adelaide Hospital, Peter Street, Dublin 8, Ireland. Tel : 758971'; Fax : 781392. 61
62 Clarke et al. tinuria from this genetic defec0 and compared this with a control population which was of a similar age and gender. We recorded the peak-methionine load homocysteine level and conventional risk factors in all subjects. We compared the relationship between hyperhomocysteinemia and asymptomarie vascular disease with that of other risk factors, such as hypertension, cigarette smoking and hypercholesterolemia.
IJ.M.S.
March, 1992
Twenty-five obligate heterozygotes for cyastathionine Bsynthase deficiency (12 men and 13 women) less than 55 years old (mean age 42) whose children were known to be homozygous for this condition, were compared with 21 controls. The control subjects, who had been randomly selected from an itidustrial health screening program had a similar age and sex as the cases, but were not individually matched to the cases. All subjects gave informed consent to a protocol which had been approved by the hospital's Ethics committee.
nique at several points along the arteries. The complete examination was recorded on videotape. All scans were carried out by a single experienced ultrasonographer who was blinded to the category of the subject. The presence of carotid disease was assessed by an independent vascular physician who was also blinded to the category of the subject. A further assessment of possible carotid disease was made by obtaining actual measurements of the intimal-medial thickness of the far wall of the common carotid arteries at 1 cm proximal to the bifurcation of the common carotid using a modification of a previously reported method~9. The intimal-medial wall thickness of the far wall of the common carotid arteries was measured as the distance from the leading edge of the ftrst echogenic line to the leading edge of the second echogenic line by a second experienced ultrasonographer who was again blinded to the category of the patient (Figure 1). The mean of the right and left carotid artery intimal-medial wall thickness for each subject was calculated.
Detection of hyperhomocysteinemia
Statistical methods
All obligate heterozygotes and normal subjects underwent a methionine-loading test in which L-methionine was given by mouth in a dose of 100 mg per kilogram of body weight after an overnight fasP. Dietary methionine was restricted to a maximum of 14 mg for the eight hour study period. Venous blood samples were obtained immediately before and four, six and eight hours after methionine administration for measurement of serum non-protein bound homocysteine. The samples were deproteinized with solid sulfosalicylic acid and stored at -20°C until analyzed. Methionine, homocysteine and homocysteine-cysteine mixed disulfides were measured in deproteiuized serum samples by ion-exchange chromatography with a Locarte amino acid analyzer. Norleucine was used as the internal standard, and mixed disulfide standard was prepared by the method, described by Frimpter21. Ninhydrin was used as the detection agent. The total concentration of non-protein-bound homocysteine was calculated as twice the concentration of homocystine plus the concentration of homocysteine-cysteine mixed disulfides.
Descriptive analyses used either Wilcoxon Rank Sum test (continuous data) or Fisher's exact test (binary data). Odds ratios were calculated fxom the parameters of a univariate logistic regression. The 95 percent confidence interval for the odds ratios were calculated using the tog odds ratio + 1.96 times the asymptotic standard error of the log odds ratio, and
Subjects and methods
Risk factors
Hypertension was considered present if the subject was receiving antihypertensive therapy or had a supine diastolic blood pressure greater than 90 mm Hg on 3 measurements made in hospital. Hypercholesterolemia was indicated by a serum cholesterol concentration of 6.1 mmoi per litre (235 mg per decilitre) or more. Cigarette smokers were classified as current smokers or non-smokers. The latter included former smokers who had quit smoking for at least 6 months before the study. Assessment of vascular disease
The ultrasonic examinationoftheextracranialcarotid system was performed with the subject lying in the supine position using an ATL Mark 5 Duplex system with a 7,5 mHz probe. The arteries were examined in the longitudinal and transverse planes from the base of the common carotid through to the most distal accessible part of the internal and external carotid arteries. Blood flow was measured using the doppler tech-
Fig. 1 - Measurement of intimal-medial wall thickness of the far wall of the carotid artery.
rot.
Hyperhomocysteinemia 63
161
No. 3
TAB LE I Obligate Heterozygotes Carotid Disease Normal Carotids
n(%)
n age (years) males cholesterol (mMol/1) Diastolic BP (ram Hg) Smokers Homocysteine (btMol/1)
12 45.67_+,5.76" 7 (58%) 6.065:1.0 81+15 3 (25%) 55+34
13 39.23+4.66 5 (38%) 5.4:t0.8 75+10 6 (46%) 45+22
P-Values OH vs Controls
> 0.05 > 0.25 > 0.5 > 0.5 > 0.5 > 0.5 < 0.0001
Normal Subjects Carotid Disease Normal Carotids 8 45.5+3.16 7 (88%) 6.38+1.63 78:1_'9 1 (13%) 16+6
13 43.38+7.33 5 (38%) 5.79!-_1.27 78+11 4 (31%) 15+4
Entries are the mean + SD or the number positive (%). Statistical significance of disease status within a population is denoted by *(p
Department of Cardiology, The Adelaide Hospital, Dublin1; The Department of Vascular Medicine, St. Mary's Hospital, Dublin2; The National Screening Centre for Metabolic Diseases, The Children's Hospital, Temple Street, Dublin ~and the Department of Preventive Medicine, Vanderbilt University, Nashville, Tennessee, U.S.A 4. Abstract Hyperhomocysteinemia arising from impaired methionine metabolism, and usually due to a deficiency of cystathionine B-synthase is a significant and independent risk factor for symptomatic vascular disease. It is not known if hyperhomocysteinemia in apparently healthy asymptomatic subjects is associated with atherosclerosis and whether such a relationship is independent of conventional risk factors. The prevalence of asymptomatic extracranial carotid artery atherosclerosis was determined by duplex ultrasound examination in 25 obligate heterozygotes with respect for cystathionine Bsynthase deficiency (whose children were known to be homozygous for this genetic defect) and in 21 controls. Hyperhomocysteinemia was determined by a standard methionine-loading test and conventional risk factors were also recorded. Twelve of 25 obligate heterozygotes and 8 of 21 normal controls had evidence of extracranial carotid artery atherosclerosis. Hyperhomocysteinemia as a genetic trait was not a significant risk marker, but the actual homocysteine level was associated with an increased risk of carotid disease. After adjustment for the effects of other significant risk factors, the odds ratio ofhyperhomocysteinemia for carotid disease was 1.038 per unit increase in homocysteine level (P=0.03). Hyperhomocysteinemia is a weak risk factor for asymptomatic extracranial carotid atherosclerosis and the relative risk a~ssociated with this genetic trait is less than that observed in a study of patients pres-enting with clinical manifestations of vascular disease. between hyperhomocysteinemia and premature cerebrovascular5-s,peripheral vascular5,6,9and coronary vascular disease ~,1°-15. We have shown that hyperhomocysteinemia is a significant risk marker for premature cardiovascular disease including coronary heart disease and that this effect is independent of that of conventional risk factors5. In several of these studies5,6, hyperhomocysteinemia was usually associated with reduced cystathionine 13-synthase activity suggesting heterozygous inheritance of this genetic defect. Although vascular disease subjects are more likely than normal to have hyperhomocysteinemia, it is not known whether apparently healthy heterozygotes for this condition are at increased risk of vascular disease. Mudd et al conducted an international survey based on questionnaire responses of 394 obligate heterozygotes and found that only 5% of obligate heterozygotes were likely to have fatal vascular event before the age of 501~. The design and conclusions of this study have been criticized xT. In contrast, Rubba et al reported a higher prevalence of arteriosclerosis as assessed by noninvasive ultrasound methods in both homozygous and heterozygous subjects from 8 families with cystathionine 13-synthase deficiency is. The aim of this study was to ascertain the relationship between hyperhomocysteinemia and asymptomatic atherosclerosis as assessed by duplex ultrasound of the extracranial arteries. High resolution duplex ultrasound is sensitive in detecting minor atherosclerotic disease and can be used to make direct measurements of arterial wall thickness ~9,2°. This technique was used to determine the prevalence of extracranial carotid artery atherosclerosis in obligate heterozygotes (whose children were known to be homozygous for homocys-
Introduction Cardiovascular diseases are the leading cause of death in western countries 1. Several studies have shown that the lesions of atherosclerosis begin years or even decades before the onset of clinically observable signs and symptoms2'3. The occurrence of hypercholesterolemia, hypertension and cigarette smoking are established risk factors that have been used to identify asymptomatic people who are likely to develop cardiovascular disease2~. Homocystinuria, an autosomal recessive condition, is usually due to a deficiency of cystathionine B-synthase, an enzyme which converts methionine to cystathionine4. Persons who are homozygous for this condition have markedly elevated blood homocysteine levels and have a significantly increased risk of premature vascular disease4. In addition to classical cystathionine B-synthase deficiency, several rarer genetic defects and deficiency of nutritional factors (such as folate and vitamin B~2) can affect homocysteine levels4. Heterozygotes for homocystinuria show abnormal elevations of homocysteine (hyperhomocysteinemia) in response to oral methionine-loading. Recently hyperhomocysteinemia has been implicated as a risk factor for vascular disease. Several studies reported an association Supported in part by a grant from the Health Research Board of Ireland. Address reprint requests to: Prof. Ian Graham, Department of Cardiology The Adelaide Hospital, Peter Street, Dublin 8, Ireland. Tel : 758971'; Fax : 781392. 61
62 Clarke et al. tinuria from this genetic defec0 and compared this with a control population which was of a similar age and gender. We recorded the peak-methionine load homocysteine level and conventional risk factors in all subjects. We compared the relationship between hyperhomocysteinemia and asymptomarie vascular disease with that of other risk factors, such as hypertension, cigarette smoking and hypercholesterolemia.
IJ.M.S.
March, 1992
Twenty-five obligate heterozygotes for cyastathionine Bsynthase deficiency (12 men and 13 women) less than 55 years old (mean age 42) whose children were known to be homozygous for this condition, were compared with 21 controls. The control subjects, who had been randomly selected from an itidustrial health screening program had a similar age and sex as the cases, but were not individually matched to the cases. All subjects gave informed consent to a protocol which had been approved by the hospital's Ethics committee.
nique at several points along the arteries. The complete examination was recorded on videotape. All scans were carried out by a single experienced ultrasonographer who was blinded to the category of the subject. The presence of carotid disease was assessed by an independent vascular physician who was also blinded to the category of the subject. A further assessment of possible carotid disease was made by obtaining actual measurements of the intimal-medial thickness of the far wall of the common carotid arteries at 1 cm proximal to the bifurcation of the common carotid using a modification of a previously reported method~9. The intimal-medial wall thickness of the far wall of the common carotid arteries was measured as the distance from the leading edge of the ftrst echogenic line to the leading edge of the second echogenic line by a second experienced ultrasonographer who was again blinded to the category of the patient (Figure 1). The mean of the right and left carotid artery intimal-medial wall thickness for each subject was calculated.
Detection of hyperhomocysteinemia
Statistical methods
All obligate heterozygotes and normal subjects underwent a methionine-loading test in which L-methionine was given by mouth in a dose of 100 mg per kilogram of body weight after an overnight fasP. Dietary methionine was restricted to a maximum of 14 mg for the eight hour study period. Venous blood samples were obtained immediately before and four, six and eight hours after methionine administration for measurement of serum non-protein bound homocysteine. The samples were deproteinized with solid sulfosalicylic acid and stored at -20°C until analyzed. Methionine, homocysteine and homocysteine-cysteine mixed disulfides were measured in deproteiuized serum samples by ion-exchange chromatography with a Locarte amino acid analyzer. Norleucine was used as the internal standard, and mixed disulfide standard was prepared by the method, described by Frimpter21. Ninhydrin was used as the detection agent. The total concentration of non-protein-bound homocysteine was calculated as twice the concentration of homocystine plus the concentration of homocysteine-cysteine mixed disulfides.
Descriptive analyses used either Wilcoxon Rank Sum test (continuous data) or Fisher's exact test (binary data). Odds ratios were calculated fxom the parameters of a univariate logistic regression. The 95 percent confidence interval for the odds ratios were calculated using the tog odds ratio + 1.96 times the asymptotic standard error of the log odds ratio, and
Subjects and methods
Risk factors
Hypertension was considered present if the subject was receiving antihypertensive therapy or had a supine diastolic blood pressure greater than 90 mm Hg on 3 measurements made in hospital. Hypercholesterolemia was indicated by a serum cholesterol concentration of 6.1 mmoi per litre (235 mg per decilitre) or more. Cigarette smokers were classified as current smokers or non-smokers. The latter included former smokers who had quit smoking for at least 6 months before the study. Assessment of vascular disease
The ultrasonic examinationoftheextracranialcarotid system was performed with the subject lying in the supine position using an ATL Mark 5 Duplex system with a 7,5 mHz probe. The arteries were examined in the longitudinal and transverse planes from the base of the common carotid through to the most distal accessible part of the internal and external carotid arteries. Blood flow was measured using the doppler tech-
Fig. 1 - Measurement of intimal-medial wall thickness of the far wall of the carotid artery.
rot.
Hyperhomocysteinemia 63
161
No. 3
TAB LE I Obligate Heterozygotes Carotid Disease Normal Carotids
n(%)
n age (years) males cholesterol (mMol/1) Diastolic BP (ram Hg) Smokers Homocysteine (btMol/1)
12 45.67_+,5.76" 7 (58%) 6.065:1.0 81+15 3 (25%) 55+34
13 39.23+4.66 5 (38%) 5.4:t0.8 75+10 6 (46%) 45+22
P-Values OH vs Controls
> 0.05 > 0.25 > 0.5 > 0.5 > 0.5 > 0.5 < 0.0001
Normal Subjects Carotid Disease Normal Carotids 8 45.5+3.16 7 (88%) 6.38+1.63 78:1_'9 1 (13%) 16+6
13 43.38+7.33 5 (38%) 5.79!-_1.27 78+11 4 (31%) 15+4
Entries are the mean + SD or the number positive (%). Statistical significance of disease status within a population is denoted by *(p
