smoking and hypercholesterolaemia, before

Transcription

1 Saturday 7 November 1992 No 8828 ORIGINAL ARTICLES Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis Endothelial dysfunction is an early event in experimental studies of atherogenesis, preceding formation of plaques. We have devised a noninvasive method for testing endothelial function, to find out whether abnormalities are present in symptom-free children and young adults at high risk of atherosclerosis. With high-resolution ultrasound, we measured the diameter of the superficial femoral and brachial arteries at rest, during reactive hyperaemia (with increased flow causing endothelium-dependent dilatation), and after sublingual glyceryl trinitrate (GTN; causing endothelium-independent dilatation) in 100 subjects 50 controls without vascular risk factors (aged 8-57 years), 20 cigarette smokers (aged years), 10 children with familial hypercholesterolaemia (FH; aged 8-16 years), and 20 patients with established coronary artery disease (CAD). Adequate scans were obtained in all but 6 cases. Flow-mediated dilatation was observed in arteries from all control subjects. Dilatation was inversely related to baseline vessel diameter (r= -0 81, p<0 0001); in arteries of 6 0 mm or less, mean dilatation was 10 (SE 2)%. In smokers, FH children, and adults with CAD, flow-mediated dilatation was much reduced or for comparison with each relevant absent (p<0 001 control group). Dilatation in response to GTN was present in all groups. Endothelial dysfunction is present in children and adults with risk factors for atherosclerosis, such as smoking and hypercholesterolaemia, before anatomical evidence of plaque formation in the arteries studied. This may be an important early event in atherogenesis. Lancet 1992; 340: Introduction The disease process of atherosclerosis begins in childhood. The first histological signs are lipid deposits in the intima of systemic arteries.! Fatty streaks can be found in the aortas of children older than 3 years and in the coronary arteries by adolescence.2,3 Endothelial dysfunction is an early physiological event in atherogenesis.4 Studies in vitro have shown that the endothelium is abnormal in the earliest stages, before plaques exist and certainly before clinical detection of disease, and that endothelial injury predisposes to thrombosis, leucocyte adhesion, and proliferation of smooth muscle cells in the arterial wall.s An important functional consequence of endothelial dysfunction is the inability to release endothelium-derived relaxing factor (EDRF).6 In vivo, coronary endothelial dysfunction has been shown in response to various pharmacological and physiological stimuli, largely in adults with symptoms of established coronary atherosclerosis.7,11 These studies used invasive angiography, which is clearly not suitable for the ADDRESSES: Cardiothoracic and Medical Units, Hospital for Sick Children, Great Ormond Street, London (D. S. Celermajer, FRACP, K. E. Sorensen, MD, V. M. Gooch, BSc, O. I. Miller, BMed, I. D. Sullivan, FRACP, Prof J. K. Lloyd, MD, J E. Deanfield, MRCP) and Medical Research Council Biostatistics Unit, Cambridge, UK (D. J. Spiegelhalter, PhD). Correspondence to Dr David S. Celermajer, Cardiothoracic Unit, Hospital for Sick Children, Great Ormond Street, London WC1N 3JH, UK.

2 1112 investigation of either the early development of vascular damage in young symptom-free subjects, or for serial studies of progression or reversibility. We have developed a non-invasive means of studying early changes in vascular physiology in systemic arteries. High-resolution ultrasound is used to follow changes in vessel diameter in response to increased flow and to glyceryl trinitrate (GTN). In arteries lined by healthy endothelium, increased flow causes dilatation of the vessel, 9,10 via release of EDRF.111 This mechanism fails in endothelial dysfunction.12,13 By contrast, GTN causes vasodilatation by direct action on the smooth muscle; its effect is therefore independent of the endothelium. We have used this method to find out whether endothelial dysfunction is present in subjects at high risk of atherosclerosis before clinical evidence of vascular disease. Patients and methods 100 subjects aged 8-67 years were studied; the superficial femoral artery was used in 60 and the brachial artery in were studied on two occasions, and 12 were studied three times, all within 2 months of their first study. 50 control subjects (hospital staff and their families), aged 8-57 years, were lifelong non-smokers, normotensive, and non-diabetic, had no family history of premature vascular disease, and had plasma total cholesterol of 5-5 mmoljl or less (range mmol/1); of these, 16 were children (aged 8-14 years). 30 subjects were clinically well but had at least one risk factor for vascular disease; 20 adults aged years had smoked 20 or more cigarettes per day for at least 5 years, and 10 children aged 8-16 years had familial hypercholesterolaemia (FH). The remaining 20 subjects were aged years and had established coronary artery disease (CAD); all had had either coronary artery surgery or angioplasty. 6 patients were excluded because of inadequate scan quality; scans from the remaining subjects in each risk group were compared with those of appropriate controls. All studies were approved by the local ethics committee. The diameter of the target artery was measured from twodimensional ultrasound images, with a 70 MHz linear array transducer and a standard 128XP/10 system (Acuson, Mountain View, California, USA). In each study, scans were taken at rest, during reactive hyperaemia, again at rest, and after sublingual GTN. The target artery (superficial femoral just distal to the bifurcation of the common femoral artery, or the brachial 2-15 cm above the elbow) was scanned in longitudinal section. The transmit "T)-. Fig 2-Flow-mediated dilatation in control subjects, FH children, smokers, and subjects with CAD. Horizontal lines=group means; 0 =subjects with homozygous FH, BA= brachial artery, SFA=superficial femoral artery. (focus) zone was set to the depth of the near wall, in view of the greater difficulty of evaluating the near than the far wall "m" line (the interface between media and adventitia).12 Depth and gain settings were set to optimise images of the lumen/arterial wall interface, images were magnified by a resolution box function (leading to a television line width of about mm), and machine operating parameters were not changed during any study. The subject lay at rest for at least 10 min before a first resting scan was recorded, and arterial flow velocity was measured, by pulsed doppler signal at a 70 angle to the vessel, with the range gate (1-5 mm) in the centre of the artery. Increased flow was then induced by inflation of a pneumatic tourniquet to a pressure of 300 means of a mm Hg for 45 min. A second scan was taken for 30 s before and 90 s after cuff deflation, including a repeat flow velocity recording for the first 15 s after cuff release. 15 min was allowed for vessel recovery, and then a further resting scan was taken. Sublingual GTN spray (400 ug) was then administered, and 3-4 min later the last scan was done. The electrocardiogram was monitored continuously, and in the first 40 subjects studied the blood pressure was recorded in the left arm every 2 min. Vessel diameter was measured by two observers, unaware of clinical details and the stage of the experiment. The arterial diameter was measured at a fixed distance from an anatomical marker, such as a bifurcation, with ultrasonic calipers. Measurements were taken from the anterior to the posterior "m" line at end diastole, incident with the R-wave on the electrocardiogram. For the reactive hyperaemia scan, diameter measurements were taken s after cuff deflation. Four cardiac cycles were analysed for each scan and the measurements averaged. Vessel diameters in scans after reactive hyperaemia, 15 min rest, and GTN were expressed as percentages of the first control scan. Flow was calculated from doppler flow velocity and vessel diameter; since velocity was taken from the centre of the artery absolute values may be overestimated but relative values before and after cuff inflation are accurate.13 Reactive hyperaemia was calculated as the maximum TABLE I-STUDIES OF SUPERFICIAL FEMORAL ARTERY IN CHILDREN Fig 1-Relation between resting arterial diameter and induced dilatation in systemic arteries of control subjects. A =superficial femoral artery in adults; C7 =superficial femoral artery in children, 0 = brachial artery m adults.

3 1113 TABLE 11-SUPERFICIAL FEMORAL ARTERY STUDIES IN ADULTS p,= for comparison of controls and smokers, P2 = for comparison of controls and CAD subjects flow recorded in the first 15 s after cuff deflation divided by the flow during the resting (baseline) scan. The mean vessel diameter and percent dilatation for each patient were obtained by averaging the measurements taken over all occasions on which that patient was studied. Patients were grouped according to age (child 16 years, adult 17), artery studied, and whether the patient was a control, a smoker, had FH, or had CAD. For control subjects, the relation between percent flow-mediated dilatation and vessel diameter was assessed by linear regression, for vessel size alone and then with adjustment for age, cholesterol, and type of artery. This procedure was repeated for percent GTNinduced dilatation. Two-sample t tests were used to compare control and other groups. Descriptive statistics are expressed as mean and standard error. Statistical significance was taken as p<005. Sources of variability were studied for baseline measurement of vessel diameter, percent flow-mediated dilatation, and percent dilatation to GTN. Interobserver variability (reproducibility = variation between measurements made by different observers on each of the 127 examinations) was summarised by the range across observers of the measurements on each occasion. The variability between occasions (repeatability), for the 21 patients who had been studied more than once, was summarised by calculating the range across occasions of the mean measurements at each visit. Finally, a joint analysis of all sources of variation (between patients, between-occasions within-patients, and between-observers within-occasions) was carried out by a nested analysis of variance, which allowed estimates of standard deviations and coefficients of variation. Results In 6 subjects, scans were of insufficient quality for assessment of vessel diameter (superficial femoral artery in 2 smokers and 3 subjects with CAD, and brachial artery in 1 normal subject). 3 control subjects (2 children, 1 adult) declined sublingual GTN. All subjects tolerated the studies well. In no subject did systolic or diastolic blood pressure change by more than 10 mm Hg during the study. No control subject, hypercholesterolaemic child, or adult smoker had clinical evidence of vascular disease, nor any ultrasound evidence of arterial narrowing in the vessels studied. L0/7f0/.? In 49 control subjects reactive hyperaemia was 440 (24)% in the superficial femoral artery (n 30) and 628 (44) % in = the brachial artery (n = 19). The arteries dilated 1-18% in response to this increase in flow. Dilatation was correlated inversely with resting vessel diameter (r 0-81, = - p < 0001; fig 1). In multivariate analysis, flow-mediated dilatation was still closely correlated with resting vessel diameter (p < ), but not with subject age, cholesterol concentration, or target artery. Vessels of 6-0 mm or less dilated 10% (2, range 2-18%). Vessel dilatation to GTN was also inversely correlated with resting vessel diameter (r=-0-80, p<00001), and findings in multivariate analysis were similar to those for flow-mediated dilatation. Hypercholesterolaemic children We studied 10 children with FH (2 homozygous, 8 heterozogyous). Treatment was simvastatin and cholestyramine in the 2 homozygotes, diet and cholestyramine in 4 heterozygotes, and diet alone in the remaining 4. There was significantly smaller flow-mediated dilatation in the FH children than normal children (fig 2, table I). Flow-mediated dilatation was reduced or absent in all 10 FH children; only 1 showed more than 2% dilatation. By contrast, dilatation to GTN did not differ from that in control subjects. Resting flow and flow increase during reactive hyperaemia were similar in the two groups (table I). Adult smokers In superficial femoral artery studies in 8 adult smokers and 14 adult controls flow-mediated dilatation was much lower in the smokers (table II, fig 2). These groups had similar vessel diameter and cholesterol concentrations. Dilatation to GTN and flow increase during reactive hyperaemia were similar in both groups. In brachial artery studies, 10 adult smokers had significantly less flowmediated dilatation than controls (table III, fig 3); there was also a significant difference in GTN-induced dilatation TABLE!tt BRACH!AL ARTERY STUDIES IN ADULTS _. p, = for companson of controls and smokers, P2 = for comparison of controls and CAD subjects

4 1114 was 2.8% (0-10%). From the nested analysis of variance, the estimated coefficient of variation between visits was 2-3%. Fig 3-Scans from brachial artery of a control adult male (top panels) and from an adult male smoker (lower panels), showing the artery at rest (left) and reactive hyperaemia (right). In control subject artery dilates from 4-1 to 4-6 mm; in smoker artery measures 46 mm in both scans. Scale markers on left of each frame are 5 mm apart. (table III). Resting flow and flow increase during reactive hyperaemia were similar in the two groups. CAD patients Vessel responses were analysed in 17 subjects with CAD (7 superficial femoral artery, 10 brachial artery). Plaques were seen in the superficial femoral arteries of 4 subjects, causing 20-50% stenosis; no plaques were seen in brachial arteries. The CAD patients were significantly older than the healthy subjects. Flow-mediated dilatation was reduced or absent in all cases. Dilatation to GTN was significantly less than that in control subjects (tables II and III). Reproducibility and repeatability The ratio of scan 3 to scan 1 (2 resting scans) from each study (284 data points: 127 subject visits, two observers for each visit, a third observer in 30 cases) was (0016; range ). The mean range of interobserver difference for the measurement of percent flow-mediated dilatation for 97 studies with two observers and 30 studies with three observers was 1-7% (0-7%). From the full nested analysis of variance, the estimated coefficient of variation was 1-4% (table IV). The study of repeatability was based on 21 subjects (12 controls, 6 smokers, 3 subjects with coronary disease) who underwent repeat scans (12 had a third scan). In each case, there was a directionally similar response to increased flow and to GTN; normal subjects showed dilatation in response to an increase in flow on each occasion, although the extent of dilatation varied, and abnormal subjects had reproducible failure to dilate. The mean range across visits of observed percent flow-mediated dilatation TABLE IV-REPRODUCIBILITY (INTER-OBSERVER VARIABILITY) AND REPEATABILITY (BETWEEN-OCCASION WITHIN-PATIENT VARIABILITY) Discussion Flow-mediated dilatation is impaired in young symptomfree subjects with risk factors for vascular disease, before anatomical evidence of atherosclerosis appears. We found that endothelial dysfunction can occur in systemic arteries of hypercholesterolaemic children as young as 8 years and in young adult smokers, which suggests that physiologically important changes may start as early as the first decade of life. Additional follow-up over many years would be required to prove that endothelial dysfunction in our young patients will be associated with subsequent atherosclerosis; however, both spatial and temporal correlation between early endothelial dysfunction and later development of atherosclerosis have lately been documented Coronary bifurcations show maximum endothelial dysfunction and are the commonest sites of plaque formation,l4 and endothelial dysfunction precedes occlusive vascular disease both in primates 15 and in human heart-transplant recipients.16 Impairment of endothelial function in early life may result in abnormal reactions between the vessel wall and platelets, neutrophils, and macrophages, and thus may contribute to the initial stages of atherogenesis.5 External ultrasound imaging provides detailed pictures of arterial wall anatomy.12,17 Small changes in vessel diameter can be reliably detected by means of high-frequency linear array transducers. The theoretical limit of axial resolution of 7-0 MHz ultrasound in the near field is about mm (depending on the number of cycles within each ultrasound pulse),12 and ultrasonic calipers are accurate to 0-1 mm. Beam plot and phantom experiments have shown that changes in axial distance from the transducer of 0-1 mm or more can be visualised and measured accurately (Acuson, unpublished). With the 7-0 MHz linear array transducer, we and othersl have found a low coefficient of variation for measurements of arterial wall anatomy and vessel diameter (1-3% variation between observers) and a high correlation between consecutive control measurements made within a study. In our study, the measured error (1-3%) is significantly less than the difference between normal and abnormal groups of subjects (7-10%). All measurements were made at end-diastole to avoid possible errors resulting from variable arterial compliance.l8 Measurements were made from anterior to posterior "m" lines, rather than between the intimal layers, since it was difficult to image the intimal cell layer consistently. We may therefore have underestimated flow-mediated dilatation in normal arteries by overestimating resting vessel diameter. For arteries of diameter less than 6-0 mm, flow-mediated dilatation is about 10% in control subjects; however, for arteries of more than 6-0 mm in diameter, flow-mediated dilatation is small even in healthy subjects. Our data from studies in adult femoral arteries must therefore be interpreted with caution. For this reason, our method is best applied to the study of smaller arteries in adults (such as the brachial or internal carotid) and larger arteries in children (such as the superficial femoral). Invasive studies on adult coronary arteries show that flow-mediated dilatation is 10-12% for vessels of about 3 mm diameter,8 which is consistent with our observations (fig 1). The availability of a well tolerated and widely applicable non-invasive method for testing endothelial function in vivo enables study of the relation between risk factors and

5 1115 vascular damage in young apparently healthy subjects, including children. It is currently suitable only for superficial arteries, and cannot be used to study coronary vessels. However, atherosclerosis is a diffuse process, and often follows a parallel course in the coronary and proximal femoral arteries.19 Others have used indirect methods to measure arterial diameter non-invasively, with dual-crystal pulsed doppler systems calculating the width of the column of moving blood, rather than directly measuring diameter by visualisation of the vessel walls.20 That technique, however, has potential limitations of low resolution and of overestimation of flow-mediated dilatation as a result of changes in the flow velocity profile during reactive hyperaemia. Hypercholesterolaemia predisposes to atherosclerosis in clinical and experimental studies.121 In animals with experimental diet-induced hypercholesterolaemia there is impairment of endothelial function before histological changes are visible,zz and this impairment can be reversed by L-arginine, which suggests relative deficiency of EDRF in the vascular endothelium. 23 Vita et a124 have shown in adults that the vasomotor response of angiographically smooth coronary arteries to acetylcholine is related to serum cholesterol concentration. Our study shows how early this interaction may begin, with abnormal endothelial function present in children aged 8-16 years. Smoking has long been recognised as one of the major risk factors for vascular disease. It is associated with severe aortic atherosclerosis in patients over 25 years o1d,25 and with early lesions in those under 25 years.1 In our study most of the smokers, some as young as 17 years, had impaired endothelial function in a systemic artery. The results of such a test might provide motivation for the individual to stop smoking when population-based statistics do not. The progress and regression of human atherosclerosis can be affected by therapy, but the change in luminal narrowing in treated groups has been small.26,27 On the other hand, a decrease in cardiovascular mortality has been shown in patients who are treated with lipid-lowering agents28 and who stop smoking,29 which may reflect a beneficial effect on endothelial physiology. However, the maximum potential for prevention and reversibility of atherosclerosis would be expected with intervention at an earlier stage of the disease, before atherosclerotic plaques are established.30 Angiographic techniques cannot be used for serial evaluation of the systemic arteries of symptom-free young people. The physiological abnormalities in children and young adults at increased risk of atherosclerosis shown by our non-invasive method, however, provide a marker that may enable objective assessment of the effects of risk factor modification. We are doing serial studies on the effects of cholesterol lowering or cessation of smoking on endothelial function in symptom-free subjects at increased risk of vascular disease. We thank Mary Jane Potter for expert assistance in preparing the paper. D. S. C. is supported by a grant from the Medical Foundation, University of Sydney, Australia, and K. E. S. by a grant from the Danish Heart Foundation. This study was supported by a grant from Corda. REFERENCES 1. McGill HC Jr. Relationship of atherosclerosis in young men to serum lipoprotein cholesterol concentrations and smoking: a preliminary report from the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group. JAMA 1990; 264: Stary HC. Evolution and progression of atherosclerotic lesions in coronary arteries in children and young adults. Arteriosclerosis 1989; 99 (suppl I): I Enos WF, Holmes RH, Beyer J. Coronary disease among United States soldiers killed in action in Korea: preliminary report. JAMA 1953; 152: Healy B. Endothelial cell dysfunction: an emerging endocrinopathy linked to coronary disease. J Am Coll Cardiol 1990; 16: Ross R. The pathogenesis of atherosclerosis an update. N Engl J Med 1986; 8: Furchgott R, Zawadzki D. The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 1980; 288: Ludmer PL, Selwyn AP, Shook TL, et al. Paradoxical vasoconstriction induced by acetylcholine in atherosclerotic coronary arteries. N Engl J Med 1986; 315: Nabel EL, Selwyn AP, Ganz P. Large coronary arteries in humans are responsive to changing blood flow: an endothelium-dependent mechanism that fails in patients with atherosclerosis. J Am Coll Cardiol 1990; 16: Laurent S, Lacolley P, Brunel P, Laloux B, Pannier B, Safar M. Flow-dependent vasodilation of brachial artery in essential hypertension. Am J Physiol 1990; 258: H Rubanyi GM, Romero C, Vanhouette PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol 1986; 250: Pohl U, Holtz J, Busse R, Bassenge E. Crucial role of endothelium in the vasodilator response to increased flow in vivo. Hypertension 1986; 8: Wendelhag I, Gustavsson T, Suurkula M, Berglund G, Wikstrand U. Ultrasound measurement of wall thickness in the carotid artery: fundamental principles and description of computerised analysing system. Clin Physiol 1991; 11: Chaveau M, Levy B, Dessanges JF, Savin E, Bailliart O, Martineaud JP. Quantitative doppler blood flow measurement method and in vivo calibration. Cardiovasc Res 1985; 19: McLenachan JM, Vita J, Fish RD, et al. Early evidence of endothelial vasodilator dysfunction at coronary branch points. Circulation 1990; 82: Harrison DG, Armstrong ML, Freiman PC, Heistad DD. Restoration of endothelium-dependent relaxation by dietary treatment of atherosclerosis. J Clin Invest 1987; 80: Fish RD, Nabel EG, Selwyn AP, et al. Responses of coronary arteries in cardiac transplant patients to acetylcholine. J Clin Invest 1988; 81: Salonen R, Salonen JT. Determinants of carotid intima-media thickness: a population based ultrasonography study in Eastern Finnish men. J Intern Med 1991; 229: Reneman RS, Van Merode T, Hick P, Muytjens MM, Hoeks APG. Age-related changes in carotid wall properties in man. Ultrasound Med Biol 1986; 12: Olsson AG. Regression of femoral atherosclerosis. J Am Coll Cardiol 1991; 83: Anderson EA, Mark AL. Flow-mediated and reflex changes in large peripheral artery tone in humans. Circulation 1989; 79: Anderson KM, Castelli WP, Levy DL. Cholesterol and mortality: 30 years of follow-up from the Framingham study. JAMA 1987; 257: Cohen RA, Zitnay KM, Haudenschild CC, Cunningham LD. Loss of selective endothelial vasoactive functions caused by hypercholesterolemia in pig coronary arteries. Circ Res 1988; 63: Cooke JP, Andon NA, Girerd XJ, Jirsch AT, Creager MA. Arginine restores cholinergic relaxation of hypercholesterolemic rabbit thoracic aorta. Circulation 1991; 83: Vita JA, Treasure CB, Nabel EG, et al. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation 1990; 81: McGill HC Jr. The cardiovascular pathology of smoking. Am Heart J 1988; 115: Arntzenius AC, Kromhout D, Barth JD, et al. Diet, lipoproteins, and the progression of atherosclerosis: the Leiden Intervention trial. N Engl J Med 1985; 312: Brown G, Albers JJ, Fisher LD, et al. Regression of coronary artery disease as a result of intensive lipid lowering therapy in men with high levels of apolipoprotein B. N Engl J Med 1990; 323: Frick MH, Elo O, Haapa K, et al. Helsinki Heart Study: primary prevention trial with gemfibrozil in middle aged men with dyslipidemia. N Engl J Med 1987; 317: Rosenberg L, Kaufman DW, Helmrich SP, Shapiro S. The risk of myocardial infarction after quitting smoking in men under 55 years of age. N Engl J Med 1985; 313: Kwiterovich PO Jr. Diagnosis and management of familial dyslipoproteinemia in children and adolescents: current issues in pediatric and adolescent endocrinology. Pediatr Clin North Am 1990; 37: