Effects of atorvastatin monotherapy and simvastatin plus cholestyramine on arterial endothelial function in patients with severe primary hypercholesterolaemia

Atherosclerosis 137 (1998) 197 – 203

Effects of atorvastatin monotherapy and simvastatin plus cholestyramine on arterial endothelial function in patients with severe primary hypercholesterolaemia L.A. Simons a,*, D. Sullivan b, J. Simons a, D.S. Celermajer c a

Uni6ersity of New South Wales Lipid Research Department, St Vincent’s Hospital, Darlinghurst NSW 2010, Australia b Department of Biochemistry, Royal Prince Alfred Hospital, Camperdown NSW 2050, Australia c Department of Cardiology, Royal Prince Alfred Hospital, Camperdown NSW 2050, Australia Received 20 May 1997; received in revised form 2 September 1997; accepted 15 October 1997

Abstract Endothelial dysfunction is an important early event in atherogenesis. Changes in arterial endothelial physiology were studied in patients with severe primary hypercholesterolaemia participating in an ongoing clinical trial evaluating atorvastatin and simvastatin. Endothelial function was assessed non-invasively using brachial ultrasound and the primary outcome measure was flow-mediated endothelium-dependent dilatation (FMD) in response to reactive hyperaemia. Patients were studied upon entry while still using simvastatin 40 mg daily and again after a 10-week washout (baseline). Over the next 30 weeks, 20 patients received atorvastatin titrated up to 80 mg daily and 12 patients received simvastatin titrated up to 40 mg daily (plus cholestyramine 4 g daily in 10/12), followed by a final ultrasound study. During simvastatin washout, total and low density lipoprotein (LDL) cholesterol rose by a median 23–29% and 30–34%, respectively. During atorvastatin therapy, total and LDL cholesterol fell by a median of 41 and 46%, respectively, triglycerides fell by 45% and high density lipoprotein (HDL) cholesterol rose by 10%. During simvastatin plus cholestyramine therapy, the respective median changes were − 32, − 39, − 44 and + 11%. Patients at baseline showed evidence of impaired FMD and this improved significantly on either treatment, from a median + 2.2 to +5.5% on atorvastatin and from +1.8 to +4.5% on simvastatin plus cholestyramine (P B 0.01 for both treatments). Typical response in healthy subjects would be from +8 to +9%. FMD at baseline was correlated with HDL cholesterol (r =0.49, P B0.01). Change in FMD was inversely correlated with baseline FMD (r= − 0.54, PB 0.001). Endothelial dysfunction in primary hypercholesterolaemia was improved by treatment with atorvastatin or simvastatin plus cholestyramine and this effect may result in the prevention of future coronary events. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Endothelial dysfunction; Vascular ultrasound; Flow-mediated dilatation; Hypercholesterolaemia; Atorvastatin; Simvastatin.

1. Introduction Endothelial dysfunction is an important early event in atherogenesis [1,2] and is also important in established coronary artery disease where loss of endothelium-dependent relaxation may cause dynamic narrowing at lesion sites [3]. A major functional conse* Corresponding author. Tel.: + 61 2 93612301; fax: + 61 2 93612234; e-mail: [email protected]

quence of endothelial damage is reduced availability of endothelium-derived relaxing factor (nitric oxide), a local vasodilator [2]. This factor also inhibits platelet aggregation, smooth muscle cell proliferation and interactions between endothelial cells and blood leukocytes [2]. Endothelial dysfunction has been demonstrated in the presence of coronary risk factors such as hypercholesterolaemia, cigarette smoking and hypertension [4–6]. It has been shown to improve in the presence of various and often unrelated treatments, including lipid-

0021-9150/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 0 2 1 - 9 1 5 0 ( 9 7 ) 0 0 2 5 2 - 9

198

L.A. Simons et al. / Atherosclerosis 137 (1998) 197–203

lowering drugs [7–11], anti-oxidants [5,12], oestrogen replacement [13] and angiotensin-converting enzyme inhibitors [14]. Investigations of endothelial function have generally relied on invasive techniques, including cardiac catheterisation and intra-coronary injection of acetylcholine. High-resolution ultrasound imaging of the brachial or other superficial arteries has provided an opportunity to study arteries in a non-invasive manner [15]. The observed changes in flow-mediated dilatation (FMD) in a peripheral artery are thought to be representative of changes in endothelial function in the coronary circulation, as judged by coronary responses to acetylcholine [16,17] and by the finding that brachial artery dilatation can be blocked by an antagonist to nitric oxide production [17]. Although trials with clinical outcomes are needed to provide definitive evidence that any intervention is genuinely beneficial, examination of endothelial function by ultrasound provides a simple surrogate for study of the potential effects of any new intervention, such as a cholesterol-lowering drug. Changes in brachial artery FMD observed have been reported during conduct of a controlled trial employing the drug atorvastatin, a recently developed inhibitor of the enzyme hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase [18 – 20].

2. Methods

2.1. Study design A multi-centre study commenced in 1996 which aimed to assess safety, efficacy and cost-effectiveness of atorvastatin in comparison to simvastatin (9 supplementary cholestyramine) in 144 patients with severe primary hypercholesterolaemia. This study remains in progress. The present report relates to a sub-study of patients attending the two participating Sydney centres. In brief, eligible patients needed to have primary hypercholesterolaemia, to be currently managed on a fat – reduced diet and simvastatin 40 mg daily and to have low density lipoprotein (LDL) cholesterol ] 5.0 mmol/l and triglycerides B4.0 mmol/l. After completion of biochemical and ultrasound observations at this entry point (‘simvastatin observations’), drug therapy was discontinued for a period of approximately 10 weeks, but standard dietary advice was continued for the remainder of the study. Biochemistry and ultrasound observations were then repeated (‘baseline observations’). Patients were then randomised in an open, parallel design to receive either 10 – 80 mg daily of atorvastatin (two thirds of patients) or 10 – 40 mg daily of simvastatin (with or without 4 g/day of cholestyramine) (one third of patients). The initial dose of each

drug was 10 mg and this was uptitrated every 6 weeks if patients had not reached target LDL cholesterol B 3.5 mmol/l. Ultrasound examination was repeated after 30 weeks treatment, when lipid levels were well stabilised and with constant treatment over the final 12 weeks (‘treatment observations’). For the purposes of this present report, a single set of biochemical readings obtained at the time of each ultrasound examination was used. The sub-study protocol was completed in 20 patients randomised to atorvastatin (18/20 ultimately studied on 80 mg in a single evening dose) and in 12 patients randomised to simvastatin (all ultimately studied on 40 mg in a single evening dose plus 10/12 on cholestyramine 4 g/day). The numbers completing the study are not strictly in the proportion 2:1, because some did not meet the entry criteria or left the study prematurely (see below). In order that all ultrasound data be utilised a ‘per protocol’ analysis was conducted rather than ‘intention-to-treat’. This decision was taken before results were examined. The protocol was approved by Institutional Ethics Committees and all patients gave informed, written consent.

2.2. Clinical and laboratory procedures Dietary compliance was supervised by the attending physician. Drug compliance was assessed by tablet counts at each titration step and study conclusion. All patients consumed more than 85% of the required intake. Standard clinical observations were made at each visit. Plasma cholesterol and triglycerides [21,22] and high density lipoprotein (HDL) cholesterol levels [23] were measured after a 10–12 h fast. There was no cross-standardisation between the respective clinic laboratories, but each met the performance standards of the Australian Lipid Standardisation Program and used standards traceable to the Centers for Disease Control, Atlanta. LDL cholesterol was calculated using the Friedewald equation [24]. (In-trial triglyceride levels occasionally exceeded 4.5 mmol/l and this prevented LDL calculation in two subjects.) Routine blood biochemistry, liver function tests, creatine kinase measurements and haematology were performed at regular intervals.

2.3. Brachial artery ultrasound The ultrasound method for measuring endotheliumdependent and endothelium-independent arterial dilatation has been described previously [25]. Briefly, the brachial artery diameter was measured on B-mode ultrasound images, with the use of a 7.0-MHz linear-array transducer and a standard Acuson 128XP/10 system (Mountain View, CA). The brachial artery was scanned in longitudinal section 2–15 cm above the

L.A. Simons et al. / Atherosclerosis 137 (1998) 197–203

elbow and the centre of the artery was identified when the clearest picture of the anterior and posterior intimal layers was obtained. The transmit (focus) zone was set to the depth of the near wall. 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 obtained and the velocity of arterial flow was measured with a 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 the inflation of pneumatic tourniquet placed around the forearm (distal to the scanned segment) to a pressure of 250 mmHg for 4.5 min, followed by release. A second scan was performed continuously for 30 s before and 90 s after deflation of the cuff, including a repeat recording of flow velocity for the first 15 s after the cuff was released. Thereafter, 10 – 15 min was allowed for recovery of the vessel, after which an additional resting scan was performed. Sublingual glyceryl trinitrate spray (400 mg) was then administered and 3 – 4 min later the last scan was performed. The scan operator was unaware of the treatment allocation. The diameter of the vessel was measured by two independent observers who were also unaware of the treatment allocation or stage of the protocol. FMD and glyceryl trinitrate-induced dilatation were calculated by each observer as the percentage change from resting vessel size and the average result recorded. It has been previously shown that this method is accurate and reproducible for measuring small changes in arterial diameter [25]. Volume flow was calculated by multiplying the velocity-time integral of the Doppler flow signal (corrected for angle) by the heart rate and the cross-sectional area of the vessel. Reactive hyperaemia was calculated as the maximal flow recorded in the first 15 s after cuff deflation divided by the flow during the first resting (baseline) scan.

2.4. Sample size calculation and statistical analysis The primary outcome of this study was the change in FMD between the baseline measurement and trial completion. Based on a single ultrasound examination in each phase, the sample sizes required were previously explored in order to detect a given improvement in FMD [25]. A sample of 15 subjects is sufficient to detect a 1.9% improvement in FMD, a sample of 20 subjects is sufficient to detect a 1.7% improvement in FMD, with power of 80% and a 0.05. The present study was therefore capable of detecting important changes in FMD in those on atorvastatin or on simvastatin plus cholestyramine, but was not adequately powered to contrast differences between the outcomes on atorvastatin and simvastatin plus cholestyramine. The lipid and lipoprotein data were analysed as percent change from the baseline treatment, other data

199

was analysed as absolute values. Since the same subjects were studied on three occasions, the respective values were evaluated by analysis of variance for repeated measures. Statistical analysis was conducted using SPSS Release 7 for Windows 95.

3. Results

3.1. Effects on lipids and lipoproteins A summary of entry characteristics for all those completing the protocol is presented in Table 1. Although there were no statistically significant differences between the atorvastatin and simvastatin groups, the former group included fewer men and more smokers. The majority of patients had a past history of coronary heart disease, but none were using nitrates at the time of study. A similar proportion were using aspirin or anti-hypertensive drugs. These drugs could conceivably have influenced vascular responses and were continued unchanged throughout the study. A summary of lipid and ultrasound findings is presented in Table 2. Although the time sequence of the study was ‘simvastatin treatment’, baseline and then atorvastatin or simvastatin (9cholestyramine), the results in Table 2 begin with baseline findings for all variables. These are contrasted by percentage change with the respective treatments for lipids and by actual values for ultrasound results. The patients exhibited a severe degree of hypercholesterolaemia due to excess of LDL cholesterol. Upon withdrawal of simvastatin, atorvastatin and simTable 1 Entry characteristicsa Treatment group in trial Number Age (y) Sex (% M) Body mass index (kg/m2) Smoker Current (%) Former (%) Previous CHD (%) Family history CHD (%) Anti-hypertensive drugs (%) Aspirin (%) Blood pressure (mmHg) Vessel diameter (mm)

Atorvastatin

Simvastatin

20 49 9 9 60 28.1 9 3.1

12 49 9 13 83 27.2 9 2.2

10 50 80 65 15 65 128/81918/12 3.84 90.58

10 42 67 67 25 58 130/779 19/11 4.23 9 0.57

CHD, coronary heart disease. a Entry data for all patients while using simvastatin 40 mg/day. Mean 9S.D. for continuous variables. The atorvastatin and simvastatin treatment groups were contrasted by Mann-Whitney test (continuous variables) and x 2 (categorical variables): no significant differences were found. 10/12 patients in the simvastatin group also received supplementary cholestyramine 4 g daily.

L.A. Simons et al. / Atherosclerosis 137 (1998) 197–203

200 Table 2 Summary of lipid and ultrasound results Treatment group

Lipids and lipoproteins Cholesterol LDL cholesterol Triglycerides HDL cholesterol

ATORV SIMV ATORV SIMV ATORV SIMV ATORV SIMV

Ultrasound studies Flow-mediated dilatation (%)

ATORV SIMV

Baseline (mmol/l) Previous SIMV Rx (% change)

10.9 (9.3, 12.7) 10.7 (9.6, 12.6) 8.9 (7.0, 10.3) 8.3 (7.3, 10.4) 2.2 (1.4, 4.3) 2.0 (1.8, 3.3) 0.89 (0.77, 1.22) 0.96 (0.79, 1.34)

−29 −23 −34 −30 −10 −32 +10

(−36, −18) (−39, −14) (−44, −22) (−43, −14) (−37, 0) (−38, +10) (−1, +26)

+7 (+1, +18)

2.2 (0.4, 6.3) 1.8 (1.4, 5.3)

Current trial (% change)a

−41 −32 −46 −39 −45 −44 +10

Analysis of variance (P)

Within treatment

Between treatmentb

B0.001

\0.4

B0.001

\0.5

B0.001

\0.5

B0.01

\0.3

5.5 (3.6, 7.2) B0.01 4.5 (1.6,

\0.5

B0.10

\0.8

(−48, −32) (−44, −19) (−56, −35) (−47, −29) (−55, −23) (−52, −11) (+2, +23)

+11(−1, +20)

4.3 (1.7, 7.1) 2.5 (0.8, 6.3) 7.4)

Glyceryl trinitrate-induced dilatation (%) Hyperaemia (%) Flow (ml/min)

ATORV SIMV ATORV SIMV ATORV SMIV

14 (12, 20) 14 380 490 45 66

(11, 20) (320, 428) (408, 695) (34, 67) (23, 82)

15 (9,18) 12 385 445 43 46

(8, 16) (310,359) (403, 613) (31, 74) (40, 66)

13 (9, 16) 15 385 340 60 66

(11, 18) (263, 488) \0.30 (268, 453) (27, 84) B0.05 (36, 98)

\0.1 \0.4

a

Atorvastatin and simvastatin groups in this column contrasted by Mann-Whitney test, there were no significant differences in any variable. 10/12 patients in the simvastatin group also received supplementary cholestyramine 4 g daily. Values are median (25th, 75th percentiles). b Refers to a comparison of atorvastatin versus simvastatin groups.

vastatin groups exhibited median 23 – 29% and 30–34% rises in total and LDL cholesterol, respectively. The groups exhibited similar falls in HDL cholesterol upon simvastatin withdrawal, but the ‘simvastatin’ group showed a larger rise in triglycerides. There was 30 weeks of titrated atorvastatin treatment which was associated with a median 41% reduction in total cholesterol, 46% reduction in LDL cholesterol, 45% reduction in triglycerides and a 10% increase in HDL cholesterol (all changes statistically significant). The respective median changes on simvastatin plus cholestyramine were −32, −39, − 44 and + 11%. Differences between atorvastatin and simvastatin groups were not statistically significant.

3.2. Vascular ultrasound results The degree of reactive hyperaemia produced by cuff inflation and release was similar across all treatment classes and all treatment groups (Table 2). Flow-mediated dilatation fell after simvastatin withdrawal, but increased by a median 3.3% on atorvastatin and 2.7% on simvastatin over the next 30 weeks. These changes over time were statistically significant (P B 0.01 for both groups). Glyceryl trinitrate-induced dilatation, a

measure of endothelium-independent relaxation, was unchanged across the study. Since the broad changes in FMD were similar in both treatment groups, further analyses were conducted in the overall group of 32 patients. FMD at baseline was significantly correlated with HDL cholesterol (r= 0.49, PB 0.01) (see Fig. 1). HDL cholesterol was a significant positive predictor of FMD at baseline in a linear regression model (P B 0.03, adjusted R2, 0.28) which also controlled for total cholesterol, gender, vessel diameter, age, previous coronary disease, cigarette smoking and hypertension (none of which were significant predictors). The observed changes in FMD after 30 weeks treatment were not significantly correlated with changes in total cholesterol (r= − 0.18, P\ 0.4, n=15) after exclusion of current and former smokers. (It was anticipated that smoking might confound this relationship and these subgroup analyses were pre-specified in the protocol.) Change in FMD was inversely correlated with FMD at baseline (r= − 0.54, PB 0.001), but not with HDL cholesterol. The change in FMD was examined in a linear regression model which controlled for change in total cholesterol, HDL cholesterol and FMD at baseline, gender, age, vessel diameter, previous coro-

L.A. Simons et al. / Atherosclerosis 137 (1998) 197–203

201

Fig. 1. The relationship between HDL cholesterol and flow-mediated dilatation at baseline (all 32 subjects pooled, r= 0.49, PB 0.01).

nary disease, cigarette smoking and hypertension. There were no significant predictors.

3.3. Tolerance and safety of medication In general, medications were well tolerated and only two patients required permanent withdrawal of medication. They were two patients on simvastatin 20 mg daily who developed a rash, one of them also complaining of joint pains. One patient on atorvastatin 10 mg daily developed joint pains and the drug was temporarily suspended (this same patient reported a sleep disturbance). Other adverse events were occasionally noted, but they were not necessarily caused by drug therapy nor did they require cessation of treatment. Full details of all adverse events, clinical and biochemical, will be published later as part of the total trial experience.

4. Discussion HMG-CoA reductase inhibitors of cholesterol synthesis are now established drugs in the management of hypercholesterolaemia where dietary treatment alone is judged insufficient. Clinical trial data demonstrate that simvastatin and pravastatin reduce the future risk of coronary disease and improve survival in patients with or without known coronary disease [26,27]. As new members of this class of drugs become available, it has been relatively easy to demonstrate safe and effective LDL cholesterol reduction. Recent research with ator-

vastatin demonstrates effective LDL reduction and greater efficacy than its predecessors [18,20]. This drug is also effective in reducing triglyceride concentrations and raising HDL cholesterol [19]. The present results, though not based on a large number of patients, indicate a 46% reduction in LDL cholesterol with atorvastatin. This very favourable outcome is a smaller effect than that noted in a previous report [18], but patients in the earlier study had a milder degree of hypercholesterolaemia. Simvastatin plus cholestyramine produced a 39% reduction in LDL cholesterol. Prior to starting cholestyramine this group had experienced a 35% reduction in LDL cholesterol. The additional efficacy of combining simvastatin with resin had previously been demonstrated [28]. This data, plus the smaller observed changes in LDL cholesterol following simvastatin withdrawal, serve to confirm that a titrated dose of atorvastatin will achieve a considerably greater reduction in LDL cholesterol than simvastatin alone. The extent of reduction in triglycerides with atorvastatin was as expected [19], although the effect with simvastatin was larger than expected [26,28] and exceeded that observed after simvastatin withdrawal for either arm of the study (Table 2). It is noted that the entry results on simvastatin were not obtained during the controlled circumstances of the study. As we have little information on the effect of atorvastatin on clinical event rates, the use of surrogate markers for arterial disease may be informative. Changes in FMD in a peripheral artery are representative of changes in the coronary circulation [16,17] and a recent

202

L.A. Simons et al. / Atherosclerosis 137 (1998) 197–203

report highlights an association between brachial endothelial function and the extent of coronary disease [29]. The patients in the present investigation had severe hypercholesterolaemia and many had other vascular risk factors such as smoking; therefore the subjects as a group generally manifested significant endothelial dysfunction. For example, the average FMD in the baseline state was around 3%, less than half the dilatation found in healthy young adults [30]. This finding is consistent with impairment due to increasing age and severe hypercholesterolaemia [4,31,32]. However, one might speculate that HDL protects endothelial function against some adverse effects of LDL (see Fig. 1). The capacity for endothelium-independent relaxation was relatively unimpaired in the patients under study, averaging around 14% (healthy young adults average 20%) [30]. A previous study using simvastatin in healthy young adults without hypercholesterolaemia showed FMD rising from 5 to 16% over 12 weeks [33]. What is uncertain is the degree of reversibility in endothelial dysfunction that may be achievable when a severe defect has been present for a long time. In our relatively small study, there were no significant predictors of change in FMD on multivariate analysis. On univariate testing, change in FMD was inversely correlated with FMD at baseline, suggesting that vessels with greater cholesterol-related endothelial dysfunction, in the absence of therapy, might have greater capacity for treatment-related functional recovery. After controlling for cigarette smoking, there was no significant relationship between statin-induced changes in cholesterol levels and improvement in FMD. It is unlikely that aggressive management of hypercholesterolaemia would necessarily reverse dysfunction induced by another factor such as smoking. While the use of atorvastatin or simvastatin plus cholestyramine has been associated with a significant 2.7 – 3.3% increase in FMD, final average values on treatment were around 5% and these are still well below healthy responses. A similar conclusion may be drawn from our findings following cessation of simvastatin (Table 2). We know that simvastatin and pravastatin improve FMD [8,11,33] and that both drugs improve clinical outcomes in patients with hypercholesterolaemia [26,27]. Atorvastatin has a similarly favourable effect to other drugs on endothelial function and to the extent that an improvement in endothelial function may be relevant to an improved clinical outcome, atorvastatin may be having a similar, clinically beneficial effect. Although cholesterol-lowering therapy appears to have only minor effects on atherosclerotic plaque size [34], recent studies have indicated a more profound and rapid effect on arterial endothelial function [35,36]. LDL apheresis may improve endothelium-dependent dilatation within min [36] and such an effect is sustainable for up to 6 months [9,10]. With cessation of

therapy, microvascular endothelial responses may deteriorate within 2 weeks [35], emphasising the need for continuing therapy to maximise therapeutic benefit. In the current study, we have extended these findings to show that cessation of therapy with an HMG-CoA reductase inhibitor results in significant deterioration of conduit artery endothelial function within 10 weeks and that benefits of cholesterol-lowering on endothelium-dependant dilatation may be observed with two different but highly effective HMG-CoA reductase inhibitor regimens over 30 weeks. Improved endothelial function has been demonstrated to occur with a spectrum of potentially antiatherogenic treatments. Flow-mediated dilatation of the brachial artery is endothelium-dependent and predominantly mediated by the release of nitric oxide [37]. Endothelial dysfunction is a well documented complication of hypercholesterolaemia [31] and is related to either decreased local release of NO by the endothelium and/or to excess production of superoxide anions, with consequent degradation of NO before it reaches the underlying smooth muscle [38]. The statin-related improvement in FMD in this study is likely due to enhanced endothelial production of NO and/or decreased local formation of oxygen-derived free radicals. In regard to lipid-lowering drugs, a common link is an important reduction in LDL cholesterol. As improved endothelial function has been linked recently to a significant decrease in episodic myocardial ischaemia [39] and possibly to reduced coronary risk, therapeutic strategies that reverse arterial endothelial dysfunction may have important prognostic benefits.

Acknowledgements The study was supported by a grant from ParkeDavis Pharmaceutical Research. DSC is supported by the Medical Foundation of the University of Sydney. We acknowledge the expert assistance of Jacqui Robinson and Robyn McCredie in performance of ultrasound examinations, Margaret von Konigsmark in data preparation, Sandra Fitness and Mary Threlfall in patient management and Amanda Patrick in preparation of the manuscript.

References [1] Henderson AH. Endothelium in control. Br Heart J 1991;65:116 – 25. [2] Boger RH, Bode-Boger SM, Frolich JC. The L-arginine-nitric oxide pathway: role in atherosclerosis and therapeutic implications. Atherosclerosis 1996;127:1 – 11. [3] Ludmer PL, Selwyn AP, Shook TL, et al. Paradoxical vasoconstriction induced by acetylcholine in atherosclerotic coronary arteries. New Engl J Med 1986;315:1046 – 51.

L.A. Simons et al. / Atherosclerosis 137 (1998) 197–203 [4] Vita JA, Treasure CB, Nabel EG, et al. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation 1990;81:491–7. [5] Celermajer DS, Sorensen KE, Georgakopoulos D, et al. Cigarette smoking is associated with dose-related and potentially reversible impairment of endothelium-dependent dilatation in healthy young adults. Circulation 1993;88:2149–55. [6] Panza JA, Quyyumi AA, Callahan TS, Epstein SE. Effect of anti-hypertensive treatment on endothelium-dependent vascular relaxation in patients with essential hypertension. J Am Coll Cardiol 1993;21:1145–51. [7] Leung WH, Lau CP, Wong CK. Beneficial effect of cholesterollowering therapy on coronary endothelium-dependent relaxation in hypercholesterolaemic patients. Lancet 1993;341:1496– 500. [8] Egashira K, Hirooka Y, Kai H. Reduction in serum cholesterol with pravastatin improves endothelium-dependent coronary vasomotion in patients with hypercholesterolemia. Circulation 1994;89:2519 – 24. [9] Anderson TJ, Meredith IT, Yeung AC. The effect of cholesterollowering and anti-oxidant therapy on endothelium-dependent coronary vasomotion. New Engl J Med 1995;332:488–93. [10] Treasure CB, Klein JL, Weintraub WS, et al. Beneficial effects of cholesterol-lowering therapy on the coronary endothelium in patients with coronary artery disease. New Engl J Med 1995;332:481 – 7. [11] O’Driscoll G, Green D, Taylor RR. Simvastatin, an HMGCoenzyme A reductase inhibitor, improves endothelial function in 1 month. Circulation 1997;95:1126–31. [12] Levine GN, Frei B, Koulouris SN, et al. Ascorbic acid reverses endothelial vasomotor dysfunction in patients with coronary artery disease. Circulation 1996;93:1107–13. [13] McCrohon JA, Adams MR, McCredie RJ, et al. Hormone replacement therapy is associated with improved arterial physiology in healthy post-menopausal women. Clin Endocrinol 1996;45:435 – 41. [14] Mancini GBJ, Henry GC, Macaya C, et al. Angiotensin-converting enzyme inhibition with quinapril improves endothelial vasomotor dysfunction in patients with coronary artery disease. Circulation 1996;94:258–65. [15] Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 1992;340:1111–5. [16] Anderson TJ, Uehata A, Gerhard MD, et al. Close relation of endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol 1995;26:1235–41. [17] Lieberman EH, Gerhard MD, Uehata A, et al. Flow-induced dilatation of the human brachial artery is impaired in patients B 40 years of age with coronary artery disease. Am J Cardiol 1996;78:1210 – 4. [18] Nawrocki JW, Weiss SR, Davidson MH, et al. Reduction of LDL cholesterol by 25–60% in patients with primary hypercholeserolemia by atorvastatin, a new HMG-CoA reductase inhibitor. Arterioscler Thromb Vasc Biol 1995;15:678–82. [19] Bakker-Arkema RG, Davidson MH, Goldstein RJ, et al. Efficacy and safety of a new HMG-CoA reductase inhibitor, atorvastatin, in patients with hypertriglyceridemia. J Am Med Assoc 1996;275:128 – 33. [20] Bertolini S, Bon GBB, Campbell LM, et al. Efficacy and safety of atorvastatin compared to pravastatin in patients with hypercholesterolemia. Atherosclerosis 1997;130:191–7.

.

203

[21] Allain CC, Poon LC, Chan CSG, et al. Enzymatic determination of total serum cholesterol. Clin Chem 1974;20:470 – 5. [22] Bucolo G, David H. Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem 1973;19:476–82. [23] Burstein M, Scholnick HR, Morfin R. Rapid methods for the isolation of lipoproteins from human serum by precipitation with polyanions. J Lipid Res 1970;11:583 – 95. [24] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without the use of the preparative ultracentrifuge. Clin Chem 1972;18:499 – 502. [25] Sorensen KE, Celermajer DS, Spiegelhalter DJ, et al. Non-invasive measurement of human endothelium-dependent arterial responses: accuracy and reproducibility. Br Heart J 1995;74:247 – 53. [26] Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383– 9. [27] Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. New Engl J Med 1995;333:1301 – 6. [28] Simons LA, Simons J, Parfitt A. Successful management of primary hypercholesterolaemia with simvastatin and low-dose colestipol. Med J Aust 1992;157:455 – 9. [29] Neunteufl T, Katzenschlager R, Hassan A, et al. Systemic endothelial dysfunction is related to the extent and severity of coronary artery disease. Atherosclerosis 1997;129:111 – 8. [30] Celermajer DS, Adams MR, Clarkson P, et al. Passive smoking and impaired endothelium-dependent arterial dilatation in healthy young adults. New Engl J Med 1996;334:150 –4. [31] Sorensen KE, Celermajer DS, Georgakopoulos D, et al. Impairment of endothelium-dependent dilatation is an early event in children with familial hypercholesterolemia, and is related to the Lp(a) level. J Clin Invest 1994;93:50 – 5. [32] Celermajer DS, Sorensen KE, Spiegelhalter DJ, et al. Ageing is associated with endothelial dysfunction in healthy men years before the age-related decline in women. J Am Coll Cardiol 1994;24:471 – 6. [33] Vogel RA, Corretti MC, Plotnick GD. Changes in flow-mediated brachial artery vasoactivity with lowering of desirable cholesterol levels in healthy middle-aged men. Am J Cardiol 1996;77:37 – 40. [34] Blankenhorn DH, Azen SP, Kramsch DM, et al. Coronary angiographic changes with lovastatin therapy. Ann Intern Med 1993;119:969 – 76. [35] Stroes ESG, Koomans HA, de Bruin TWA, Rabelink TJ. Vascular function in the forearm of hypercholesterolaemic patients off and on lipid-lowering medication. Lancet 1995;346:467–71. [36] Tamai O, Matsuoka H, Itabe H, et al. Single LDL apheresis improves endothelium-dependant vasodilatation in hypercholesterolemic humans. Circulation 1997;95:76 – 82. [37] Joannides R, Haefeli WE, Linder L, et al. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation 1995;91:1314 – 9. [38] Ohara Y, Peterson TE, Harrison DG. Hypercholesterolemia increases endothelial superoxide anion production. J Clin Invest 1993;91:2546 – 51. [39] Andrews TC, Raby K, Barry J, et al. Effect of cholesterol reduction on myocardial ischemia in patients with coronary disease. Circulation 1997;95:324 – 8.