Endothelium-Dependent Dilatation Is Impaired in Young Healthy Subjects With a Family History of Premature Coronary Disease
Background A family history of premature coronary artery disease (CAD) in a first-degree relative is an independent risk factor for coronary disease. Both genetic and environmental influences are likely to be responsible and may interact, but their relative importance is unclear.
Methods and Results We studied endothelial function in 50 first-degree relatives (31 men, 19 women; mean age, 25±8 years) of patients (men ≤45years, women ≤55years) with proven CAD. All subjects were well, lifelong nonsmokers, not diabetic, and not hypertensive and took no medications. Using high-resolution external vascular ultrasound, we measured brachial artery diameter at rest and in response to reactive hyperemia (with increased flow causing an endothelium-dependent vasodilatation) and to sublingual glyceryltrinitrate (GTN, an endothelium-independent dilator). Vascular responses were compared with those of 50 healthy control subjects matched for age and sex. Flow-mediated dilatation (FMD) was impaired in the family history group (4.9±4.6% versus 8.3±3.5% in control subjects, P<.005). In contrast, GTN caused dilatation in all subjects (family history, 17.1±8.8%; control subjects, 19.0±6.3%; P=NS), suggesting that reduced FMD was due to endothelial dysfunction. When the family history subjects were subdivided, those found to have a serum cholesterol >4.2 mmol/L (group A, n=10) had mildly impaired FMD compared with control subjects (5.5±5.1% versus 8.3±3.5%). In others whose affected relative had coronary risk factors (group B, n=24), FMD was also only slightly reduced (6.2±4.8% versus 8.3±3.5%). In contrast, subjects with no risk factors and whose affected relative had a normal cardiovascular risk factor profile (group C, n=16) had markedly impaired FMD (2.9±3.7% versus 8.3±3.5%). Although ANOVA of the three family history subgroups did not reach statistical significance (F=2.55, P=.09), pairwise analysis showed that FMD in group C was significantly impaired compared with group B (P=.026).
Conclusions Healthy young adults with a family history of premature coronary disease may have impaired endothelium-dependent dilatation, even in the absence of other cardiovascular risk factors. Those subjects, who were free of risk factors and whose affected first-degree relative was free of risk factors, had the most impaired endothelial function, suggesting a genetic influence on early arterial physiology that may be relevant to later clinical disease.
A history of premature CAD in a first-degree relative is an independent risk factor for coronary heart disease.1 2 3 4 5 Epidemiological studies suggest that the first-degree relatives of coronary patients have a 2.5- to 7-fold increase in risk of death from coronary disease compared with those without a family history of CAD.6
This inherited vascular risk may be mediated in a number of ways. Several major cardiovascular risk factors are known to be genetically influenced, including diabetes and hyperlipidemia, and levels of other cardiovascular risk factors such as homocysteine and fibrinogen may also have an inherited basis.7 8 9 10 It is also likely that a variety of genes interact to alter the arterial wall and/or its susceptibility to damage from risk factors.11
Damage to the vascular endothelium is an initiating event in experimental studies of atherogenesis.12 An early marker of endothelial dysfunction is the loss of endothelium-dependent dilatation, thought to be related to reduced activity of NO.13 14 Recent work suggests that NO not only regulates vascular tone but also has a key “antiatherogenic” role with regulation of vascular permeability, the inhibition of platelet adhesion/aggregation, leukocyte/vessel wall interaction, and smooth muscle proliferation.15
Endothelial dysfunction has been demonstrated in humans with clinical evidence of atherosclerosis16 and young, asymptomatic subjects who have established cardiovascular risk factors such as smoking, hypercholesterolemia, and diabetes.17 18 19 In this study, we have examined a cohort of young subjects identified by a family history of premature CAD but in whom these potentially confounding risk factors were excluded to examine the influence of family history on endothelial function.
We have found early endothelial dysfunction in the first-degree relatives of young coronary patients in whom family history was the only major cardiovascular risk factor. These subjects are likely to have the strongest genetically determined influence on their risk of subsequent arterial disease.
We identified 122 patients (men ≤45 years, women ≤55 years) with premature CAD (angiographically proven—≥50% stenosis of one or more of the major epicardial coronary arteries, or with myocardial infarction—chest pain, development of Q waves on the resting ECG, and appropriate cardiac enzyme rise) from coronary care and angiography records from three London hospitals, over a 12-month period. These patients were assessed for cardiovascular risk factors with a questionnaire concerning smoking history and family history of CAD, and details of their lipid profile and history of hypertension were obtained from hospital records or their primary care physician. Those with familial hypercholesterolemia (n=14) or diabetes mellitus (n=6) identified in medical records were excluded. The remaining 102 patients were contacted to determine whether they had any first-degree relatives between 15 and 40 years of age who were lifelong nonsmokers and not hypertensive (resting BP<140/90) or diabetic (fasting plasma glucose <5.2 mmol/L) and who were receiving no regular vasoactive medications. Of the 88 who replied, 29 patients had 50 first-degree relatives (31 men, 19 women) who were willing to participate. Details of smoking history, cholesterol levels, BP, and family history were obtained in a cardiovascular risk questionnaire from the patients with premature coronary disease.
The 50 subjects attended for study after a 12-hour overnight fast, and venous blood was withdrawn for measurement of plasma glucose (Kodac autoanalyzer), total cholesterol (cholesterol C-system high performance CHOD-PAP method), HDL cholesterol, afterprecipitation of apoprotein B–containing lipoproteins, and triglycerides (GPO-PAP high-performance enzymatic colorimetric test, Boehringer-Mannheim GmbH, Diagnostica). LDL cholesterol was calculated according to the formula of Friedwald et al.20 Lipoprotein(a) was measured with an ELISA method (Immuno GmbH). Fibrinogen and homocysteine levels were not measured. Supine BP was recorded after 5 minutes of rest, and a noninvasive study of vascular reactivity was performed (see below). Aspirin and nonsteroidal anti-inflammatory drugs were avoided for 5 days before the study.
Vascular responses were compared with those of 50 healthy control subjects matched for age and sex who were free of identifiable cardiovascular risk factors and recruited from hospital staff and their relatives. The study was approved by the local research ethics committee, and each subject gave informed consent.
The 50 individuals with a family history of premature CAD were then subdivided into three prospectively defined groups: group A, those who, on testing, had an elevated serum LDL cholesterol (>4.2 mmol/L); group B, those whose first-degree relative had known coronary risk factors (cigarette smoker >1 pack-year, LDL cholesterol >4.2 mmol/L or total cholesterol >5.5 mmol/L, and BP >140/90 mm Hg); and group C, those with no identifiable cardiovascular risk factors themselves and whose affected relative had no known cardiovascular risk factors. If the risk factor profile of the affected first-degree relative was not clearly defined, the subject was included in group B (those with relatives with risk factors).
Vascular Reactivity Study
Arterial endothelial and smooth muscle function were studied noninvasively by examination of brachial artery responses to endothelium-dependent and -independent stimuli as we have previously described.21 Arterial diameter was measured from B-mode ultrasound images at rest, in response to reactive hyperemia (with increased flow producing endothelium-dependent vasodilatation), again at rest, and after sublingual GTN (an endothelium-independent vasodilator) by use of a standard 7-MHz linear array transducer and Acuson 128XP/10 system.
The subject lay at rest for at least 10 minutes before the first scan and remained supine throughout the study. The brachial artery was scanned in a longitudinal section 2 to 15 cm above the elbow, and the center of the artery was identified when the clearest picture of the anterior and posterior wall layers was obtained. The transmit (focus) zone was set to a depth of the near wall, in view of the greater difficulty in evaluating the near compared with the far wall “m” line (the interface between the media and adventitia). Depth and gain settings were set to optimize images of the lumen/arterial wall interface, images were magnified using a resolution box function (leading to a line width of approximately 0.065 mm), and machine operating parameters were not changed during the study.
When a satisfactory transducer position was found, the skin was marked and the arm remained in the same position throughout the study. A resting scan was then recorded. Increased flow was then induced by inflation of a pneumatic tourniquet placed around the forearm (distal to the segment of artery being scanned) to a pressure of 250 to 300 mm Hg for 4.5 minutes, followed by release. A second scan was taken 30 seconds before release of the cuff and continued for 90 seconds after cuff deflation. The brachial artery dilatation to flow with this technique can be blocked by infusion of L-NMMA,22 a specific antagonist for NO synthase, and responses correlate with invasive tests of coronary endothelial function.23 Thereafter, a period of 10 to 15 minutes was allowed for vessel recovery, after which another resting scan was taken. Sublingual GTN spray (400 μg) was then administered, and the last scan was performed 3 to 4 minutes later. Doppler-derived flow measurements (using a pulsed-wave Doppler signal at a 70° angle to the vessel with the range gate [1.5 mm] in the center of the artery) were obtained during the first resting scan (baseline blood flow) and again during the first 15 seconds of reactive hyperemia (allowing the flow increase to be expressed as a percentage of the baseline flow). The ECG was monitored continuously throughout the study.
All scans were recorded on super VHS videotape for later analysis. Vessel diameters were measured by two “blinded” observers unaware of the clinical details and the stage of the experiment. The mean of the two observations was then taken. We have shown previously that this method is accurate and reproducible for measurement of small changes in arterial diameter, with low interobserver error for measurement of FMD.24 The arterial diameter was measured at a fixed distance from an anatomical marker (such as a fascial plane or vein seen in cross section), using ultrasonic calipers. Measurements were taken from the anterior to the posterior m-line at end diastole. The mean diameter was calculated from four cardiac cycles incident with the R wave on the ECG. For the reactive hyperemia scan, diameter measurements were taken 50 to 60 seconds after cuff deflation. Diameter changes were derived as percentage change relative to the first baseline scan (100%). Volume blood flow was calculated by multiplying the velocity time integral of the Doppler flow signal (corrected for angle) by the heart rate and the vessel cross-sectional area (π×r2). The flow velocity used in our calculation is taken from the center of the artery and therefore gives an overestimate, but relative flow values before and after cuff inflation are accurate. Reactive hyperemia was calculated as the maximum flow measured during the first 15 seconds after cuff deflation divided by the flow during the first resting (baseline) scan.
Descriptive data are expressed as mean± SD. The family history and control groups were compared by use of two sample (independent) Student’s t tests. Corrections for multiple comparisons were made when appropriate. One-way ANOVA was then performed to compare FMD and GTN responses in the three family history subgroups. Pairwise comparisons were then made, with the primary comparison being between those subjects, themselves free of cardiovascular risk factors, whose affected relative had cardiovascular risk factors (group B) and those whose affected relative was free of cardiovascular risk factors (group C). For the whole group, family history subjects and control subjects, we used univariate and multivariate regression analysis to explore the relationship between both FMD and GTN-mediated dilatation and age, resting brachial artery diameter, reactive hyperemia, BP, total cholesterol, and family history of premature CAD. Similar analyses were performed to study determinants of FMD in the family history group alone. Statistical significance was inferred at P<.05.
Characteristics of Premature CAD Patients and Study Subjects
Of the 29 patients (17 men, 12 women) whose offspring participated in the study, 18 were identified from coronary angiography and 11 from coronary care records. The risk factor profile of these patients was as follows. Of these patients, 10 were lifelong nonsmokers, 6 smoked ≤10 pack-years, and 13 smoked >10 pack-years. Eleven patients had a history of hypercholesterolemia (total cholesterol >5.5 mmol/L), 8 patients had a history of hypertension (resting supine BP>140/90 mm Hg), and 21 patients themselves had a first-degree relative with CAD. Nine patients had no identifiable coronary risk factors (except for a family history of CAD).
The characteristics of the 50 offspring of the CAD patients (family history subjects) and their matched control subjects are shown in the Table⇓. Total cholesterol levels were significantly higher in the family history group, but other characteristics, including BP, were similar. Within the family history subjects, 10 had a serum LDL cholesterol >4.2 mmol/L (group A), 24 had a first-degree relative with cardiovascular risk factors (group B), and 16 had an affected relative with no identifiable risk factors (except for family history, group C).
Vascular Reactivity Studies
Resting brachial artery diameter, baseline flow, and the degree of reactive hyperemia were similar in family history subjects and control subjects (the Table⇑). In the 50 family history subjects, FMD was markedly reduced (5.0±4.6% versus 8.2±3.5%, P<.001) compared with the control group; as expected for this heterogeneous population (groups A, B, and C), there was a wide range of responses (Fig 1a⇓). GTN caused dilatation in all subjects (family history subjects, 17.1±8.8%; control subjects, 19.0±6.3%; P=.21), suggesting that reduced FMD in the family history group is due to endothelial dysfunction (Fig 1b⇓).
We examined both FMD and GTN responses in the three subgroups of family history subjects (groups A, B, and C), although these groups are relatively small. In those family history subjects with an LDL cholesterol >4.2 mmol/L (group A), FMD was mildly impaired (5.5±5.1% versus 8.2±3.5% in control subjects), as might be expected from previous work.25 Similarly, FMD was only slightly depressed in the family history subjects whose first-degree relatives had major cardiovascular risk factors (group B) (6.2±4.8% versus 8.2±3.5% in the control subjects). In contrast, the group with no risk factors in either themselves or their affected relatives (group C) had markedly abnormal FMD (2.9±3.7% versus 8.2±3.5% in the control subjects, which is highly significant [P<.0001]). Although ANOVA of the three family history subgroups did not reach statistical significance (F=2.55, P=.09), pairwise analysis showed that FMD in group C was significantly more impaired than in group B (P=.026) (Fig 2⇓). The GTN responses were not significantly different within the three family history groups or compared with the control subjects.
Determinants of FMD and GTN Dilatation
On multivariate regression analysis of the combined group of 100 family history subjects and control subjects, the strongest correlation was between FMD and a history of premature CAD (partial r=−.32, P<.005), but FMD was also related to vessel size (r=−.24, P<.05) as has previously been shown.21 Within the 50 family history subjects, similar multivariate regression analyses were performed, but no correlation was found between FMD and total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, lipoprotein(a), or resting BP. GTN-mediated responses within the whole group of 100 subjects were inversely correlated with vessel size (partial r=−.21, P<.05), but no other relationships were found. There were no significant determinants of GTN response within the family history group alone.
We have shown that young adults with a family history of premature CAD have impaired endothelium-dependent FMD in large arteries of the systemic circulation. This was most obvious in subjects without additional risk factors themselves or in their affected first-degree relative, suggesting a direct inherited influence on arterial function from an early age, which may be relevant to the risk of subsequent development of vascular disease.
Epidemiological studies have examined the relationship between a family history of premature CAD and the development of atherosclerotic disease in populations with other cardiovascular risk factors.2 3 4 5 6 Several studies have shown that a family history of CAD is independently associated with the complications of atherosclerotic disease, and some workers have demonstrated a clustering of cardiovascular risk within families, suggesting a genetic influence.1
Our results provide objective evidence of abnormal arterial physiology, long before the development of clinical CAD, in young subjects with an adverse family history. The abnormality is likely to reflect endothelial dysfunction, a key early event in atherogenesis.
The noninvasive method used enables accurate and reproducible assessment of endothelial function in humans.21 24 The technique compares dilatation in response to increased flow with that to GNT. FMD is known to depend on the ability of endothelium to release NO in response to shear stress and consequently can be substantially attenuated by interarterial administration of L-NMMA, a specific antagonist of NO synthase.22 A close correlation has been established between endothelial function in the brachial artery, assessed with our method, and endothelial function in the coronary arteries, studied invasively by infusion of acetylcholine.23
In previous studies, we have demonstrated the influence of a number of recognized cardiovascular risk factors, including cigarette smoking, hypercholesterolemia, and diabetes, on this measure of early arterial function.17 18 25 We therefore purposefully excluded subjects with identifiable risk factors, ie, smoking, hypertension, diabetes, and familial hypercholesterolemia, from our cohort. Nevertheless, subjects identified on the basis of a family history of CAD form a heterogeneous population, and it is not surprising that a wide range of vascular responses was found.
The influence of family history of CAD on vascular physiology and cardiovascular risk is likely to be multifactorial. Some subjects identified on the basis of a family history of CAD have risk factors such as elevated LDL cholesterol. This was the case in 20% of our subjects (group A), and they had a small reduction in FMD compared with the control subjects, as found in our previous studies.25 The presence of elevated cholesterol may be a chance association with family history or may represent an inherited form of hyperlipidemia, despite the fact that we excluded subjects with familial hypercholesterolemia. Other subjects with a family history of CAD had first-degree relatives who themselves had a clearly abnormal cardiovascular risk factor profile that probably contributed to their cardiovascular disease (group B). As might be expected, FMD in these subjects was not significantly decreased, suggesting a low inherited risk. Any impairment of FMD is likely to be due to shared environmental influences such as environmental tobacco smoke exposure and diet. Most interesting were the subjects who themselves had no identifiable vascular risk factors and whose first-degree relative with premature CAD had a normal cardiovascular risk factor profile. These subjects had the most abnormal endothelial function. Because shared environmental influences are unlikely to have an important effect, an inherited abnormality of the vessel wall or an inherited increased vulnerability to circulating risk factors is likely to be present.26 We cannot, however, exclude the influences of other inherited biochemical and serological factors such as homocysteine and fibrinogen on arterial function. An increased plasma homocysteine level has been shown to be an independent risk factor for CAD in a recent large meta-analysis.27 We have previously found that the heterozygous parents of homozygous children with homocystinuria (who have markedly raised plasma homocysteine levels and impaired FMD) had normal endothelial function,28 and elevated homocysteine levels are unlikely to be the basis for the markedly abnormal endothelial responses seen in a third of our subjects. Although we were unable to control for environmental factors that might have been shared between the affected relative and the study subjects, these cannot explain the marked abnormality of endothelium-dependent relaxation in the group C subjects, who were least likely to have been exposed to factors such as passive smoking. It is well known that reported paternity is unreliable in ≈10% of families. Although we could not correct for this in our study, its influence would operate to weaken the effects we have found.
The inherited basis of coronary disease likely to be represented in the group C subjects remains poorly understood. Single gene abnormalities, such as ACE gene polymorphism, have been linked to cardiovascular disease later in life, but in most cases, the development of atherosclerosis and its complications is likely to have a polygenic basis.11 Our findings suggest that the NO pathway is abnormal from an early stage but cannot distinguish between an abnormality of NO availability or nonspecific manifestations of another genetically determined abnormality.29 Whatever the basis of the impaired endothelial function, decreased bioavailability of NO may nevertheless contribute to the inherited predisposition to the development of arterial disease.
Although long-term follow-up studies investigating clinical outcome in young people identified as having endothelial dysfunction have not yet been conducted, our current understanding of the evolution of the atherosclerosis would suggest a contribution to increased risk of later CAD.12 In this group of young people with a strong family history of premature CAD, impaired FMD may therefore act as a phenotypic marker for those with the inherited tendency to develop CAD. Impaired FMD, used alone or in conjunction with other markers, might facilitate genetic linkage, or sibling pair analyses, helping to identify those genes that influence the development of this disease.
Selected Abbreviations and Acronyms
|CAD||=||coronary artery disease|
|L-NMMA||=||NG-nitro-l-arginine methyl ester|
Dr Clarkson was supported by the British Heart Foundation, Dr Celermajer by the National Heart Foundation of Australia, and Ann Donald by the Coronary Artery Disease Association.
- Received February 10, 1997.
- Revision received June 23, 1997.
- Accepted July 3, 1997.
- Copyright © 1997 by American Heart Association
Grech ED, Ramsdale DR, Bray CL, Faragher EB. Family history as an independent risk factor of coronary artery disease. Eur Heart J. 1992;13:1311-1315.
Wang XL, Tam C, McCredie RM, Wilcken DE. Determinants of severity of coronary artery disease in Australian men and women. Circulation. 1994;89:1974-1981.
Slack J, Evans KA. The increased risk of death from ischaemic heart disease in the first degree relatives of 121 men and 96 women with ischaemic heart disease. J Med Genet. 1966;3:239-257.
Wang PH, Korc M. Searching for the Holy Grail: the cause of diabetes. Lancet. 1995;346(suppl):4.
Dammerman M, Breslow JL. Genetic basis of lipoprotein disorders. Circulation. 1995;91:505-512.
Stampfer MJ, Malinow MR. Can lowering homocysteine levels reduce cardiovascular risk? N Engl J Med. 1995;323:328-329.
Scarabin PY, Bara L, Richard S, Poirier O, Cambou JP, Arveiler D, Luc G, Evans AE, Samama M, Cambien F. Genetic variation at the β-fibrinogen locus in relation to plasma fibrinogen concentration and risk of myocardial infarction: the ECTIM Study. Arterioscler Thromb. 1993;13:886-891.
Lusis AJ, Rotter JI, Sparkes RS, eds. Molecular Genetics of Coronary Artery Disease: Candidate Genes and Process in Atherosclerosis. Basal, Switzerland: Karger; 1992.
Zeiher AM, Drexler H, Wollschlager H, Just H. Modulation of coronary vasomotor tone in humans: progressive endothelial dysfunction with different early stages of coronary atherosclerosis. Circulation. 1991;83:391-401.
Cooke JP, Tsao PS. Is NO an endogenous antiatherogenic molecule? Arterioscler Thromb Vasc Biol. 1994;14:653-655.
Vita JA, Treasure CB, Nabel EG, McLenachan JM, Fish RD, Yeung AC, Vekshtein VI, Selwyn AP, Ganz P. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation. 1990;81:491-497.
Celermajer DS, Sorensen KE, Georgakopoulous D, Bull C, Thomas O, Robinson J, Deanfield JE. Cigarette smoking is associated with a dose dependent and potentially reversible impairment of endothelium-dependent dilatation in healthy young adults. Circulation. 1993;88:2149-2155.
Clarkson P, Celermajer DS, Sampson M, Sorensen KE, Adams MR, Yue DK, Betteridge DJ, Deanfield JE. Endothelial dysfunction in insulin-dependent diabetes mellitus is related to disease duration and LDL-cholesterol levels. J Am Coll Cardiol. 1996;99:99-100.
Sorensen KE, Celermajer DS, Georgakopoulous D, Hatcher G, Betteridge DJ, Deanfield JE. Impairment of endothelium-dependent dilation is an early event in children with familial hypercholesterolemia and is related to the lipoprotein(a) level. J Clin Invest. 1994;93:50-55.
Friedwald WT, Levi RI, Fredrickson DS. Estimation of the concentration of low density lipoprotein cholesterol in the plasma without the use of the preparative ultracentrifuge. Clin Chem. 1958;18:499-502.
Joannides R, Haefeli WE, Linder L, Richard V, Bakkali EH, Thuillez C, Luscher TF. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation. 1995;91:1314-1319.
Sorensen KE, Celermajer DS, Spiegalhalter DS, Georgakopoulous D, Robinson J, Deanfield JE. Non-invasive measurement of endothelium dependent arterial responses in man: accuracy and reproducibility. Br Heart J. 1995;74:247-253.
Kaprio J, Norio R, Pesonen E, Sarna S. Intimal thickening of the coronary arteries in infants in relation to family history of coronary artery disease. Circulation. 1993;87:1960-1968.