Efficacy of fenofibrate and simvastatin on endothelial function and inflammatory markers in patients with combined hyperlipidemia: relations with baseline lipid profiles

Atherosclerosis 170 (2003) 315–323

Efficacy of fenofibrate and simvastatin on endothelial function and inflammatory markers in patients with combined hyperlipidemia: relations with baseline lipid profiles Tzung-Dau Wang a , Wen-Jone Chen b , Jong-Wei Lin a , Ching-Chih Cheng a , Ming-Fong Chen a , Yuan-Teh Lee a,∗ a b

Division of Cardiology, Department of Internal Medicine, National Taiwan University Hospital and National Taiwan University College of Medicine, 7 Chung-Shan South Road, Taipei, Taiwan, ROC Department of Emergency Medicine, National Taiwan University Hospital and National Taiwan University College of Medicine, 7 Chung-Shan South Road, Taipei, Taiwan, ROC Received 26 February 2003; received in revised form 13 June 2003; accepted 8 July 2003

Abstract Given that combination therapy with statin plus fibrate confers a risk of myopathy, it is worthwhile to determine whether statin or fibrate monotherapy is associated with greater clinical benefit in individuals with combined hyperlipidemia. In this randomized double-blind study, we compared the efficacy of simvastatin and fenofibrate on indexes of endothelial function (flow-mediated dilation (FMD) of the brachial artery) and inflammatory markers (plasma high-sensitivity C-reactive protein (CRP), interleukin-1␤ (IL-1␤), soluble CD40, and soluble CD40 ligand (sCD40L) levels), as surrogate indicators of future coronary heart disease (CHD), in patients with combined hyperlipidemia. A total of 70 patients with plasma triglyceride levels between 200 and 500 mg/dl and total cholesterol levels of >200 mg/dl were randomly assigned to receive either simvastatin (20 mg/day) (n = 35) or micronized fenofibrate (200 mg/day) (n = 35) for 8 weeks. Treatment with simvastatin was associated with significantly greater reduction of total cholesterol and low-density lipoprotein cholesterol (LDL-C), while the decrease in triglycerides was significantly greater in patients receiving fenofibrate. Both fenofibrate and simvastatin markedly reduced plasma levels of high-sensitivity CRP, IL-1␤, and sCD40L, and improved endothelium-dependent FMD without mutual differences. The changes in plasma inflammatory markers did not correlate with baseline clinical characteristics in both groups. However, the improvement in FMD with fenofibrate treatment correlated inversely with baseline high-density lipoprotein cholesterol (HDL-C) levels, whereas the improvement in FMD with simvastatin treatment was positively related to HDL-C levels. Accordingly, in the subgroup with a baseline HDL-C of ≤40 mg/dl, only fenofibrate significantly improved the endothelium-dependent FMD. On the other hand, in the subgroup with HDL-C >40 mg/dl, only treatment with simvastatin achieved significant improvement in FMD. The data here indicate that in patients with combined hyperlipidemia, both fenofibrate and simvastatin have comparative beneficial effects on various inflammatory markers and differential beneficial effects on endothelial function according to baseline HDL-C levels. These findings should be validated by additional prospective studies, in which patients are stratified by baseline HDL-C prior to randomization. © 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Endothelium; Fibrates; Hypercholesterolemia; Inflammation; Lipids; Statins

1. Introduction Combined hyperlipidemia is characterized by elevated levels of total cholesterol, low-density lipoprotein choles-

∗ Corresponding author. Tel.: +886-2-2356-2013; fax: +886-2-2391-3682. E-mail address: [email protected] (Y.-T. Lee).

terol (LDL-C), and triglycerides, and decreased levels of high-density lipoprotein cholesterol (HDL-C) [1,2]. In both observational studies and clinical trials, patients with combined hyperlipidemia had disproportionately higher risk for coronary heart disease (CHD) compared with those with elevated LDL-C levels alone [3–7]. However, it is still uncertain whether optimal lipid-lowering therapy for individuals with combined hyperlipidemia should consist of statin monotherapy, fibrate monotherapy, or a combination therapy

0021-9150/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0021-9150(03)00296-X

316

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

of statin plus fibrate, because no head-to-head comparisons have been made in large-scale studies to examine their relative efficacy on the reduction of CHD events [5–7]. Given that combination therapy with statin plus fibrate confers a risk of myopathy and rhabdomyolysis [8], it is therefore worthwhile to determine whether statin or fibrate monotherapy can provide the benefit for patients with combined hyperlipidemia. Both endothelial dysfunction and vascular inflammation play important roles in the pathogenesis of atherosclerosis [9]. In patients at high risk of CHD, endothelial dysfunction is observed in morphologically intact vessels before the onset of clinically manifested vascular disease [10]. As a consequence, assessment of endothelial function, by measuring the flow-mediated dilation (FMD) of the brachial artery, is currently regarded as a potential tool for prediction of CHD risks [10,11]. Treatment with statins or fibrates alone has been demonstrated to improve endothelium-dependent FMD of the brachial artery in hypercholesterolemic individuals [12,13]. Nevertheless, there have been only two recent studies comparing the effects of both lipid-lowering agents on endothelium-dependent vascular reactivity in combined hyperlipidemia and the results are inconsistent [14,15]. C-reactive protein (CRP), a sensitive marker of vascular inflammation, has been shown to be a powerful independent predictor of future CHD in numerous prospective epidemiological studies [16,17]. Treatment with statins not only reduced LDL-C, but also reduced CRP in an LDL-independent manner [18,19]. In contrast, the effects of fibrates on levels of CRP remain controversial [13,14]. Recent studies have demonstrated that the multipotent immunomodulator CD40 ligand (sCD40L) and its receptor, CD40, play an important role in the various stages of atherogenesis [9,20,21]. In addition to the 39 kDa cell-associated form, CD40 ligand also occurs in a soluble, biologically active form. It has been shown that elevations of circulating soluble CD40 ligand predict future CHD in healthy individuals and patients with acute coronary syndromes, irrespective of plasma lipid and CRP levels [22,23]. There is paucity of data regarding the effects of statins and fibrates on plasma levels of sCD40L and soluble CD40. In this study, the efficacy of statin monotherapy (simvastatin) and fibrate monotherapy (fenofibrate) on both endothelial function and markers of inflammation, as surrogate indicators of future CHD, was compared in patients with combined hyperlipidemia. We further examined the correlations between changes in markers of inflammation and endothelial function and various baseline characteristics to see if there was any differential effect of either statin or fibrate monotherapy in different predisposing conditions. These findings provide a rational basis in considering either statins or fibrates as the first-line lipid-lowering therapy for combined hyperlipidemia.

2. Methods 2.1. Subjects and study design Patients were recruited from the clinics at the National Taiwan University Hospital, without restriction to sex or socioeconomic status. Inclusion criteria were following: age 18–80 years, plasma triglyceride level between 200 and 500 mg/dl, total cholesterol level >200 mg/dl, and total cholesterol/HDL-C ratio >5. The major exclusion criteria were: acute coronary event, stroke, or coronary revascularization within the preceding 3 months; insulin-dependent diabetes mellitus or poorly controlled non-insulin-dependent diabetes mellitus (HbA1c >8%), severe obesity, overt liver disease, chronic renal failure, hypothyroidism, myopathy, alcohol or drug abuse, several other significant diseases, or use of other lipid-lowering therapy, immunosuppressants, erythromycin and/or neomycin, ketoconazole, and hormone-replacement therapy. All subjects gave written informed consent and the study was approved by the local ethics committee. Eligible patients were instructed to adhere to the American Heart Association Step 1 diet throughout the study and underwent an 8-week run-in period during which previous lipid-lowering therapy was discontinued. After the run-in phase, patients were then randomized to receive either simvastatin (20 mg/day) (n = 35) or micronized fenofibrate (200 mg/day) (n = 35) for the 8-week double-blind phase. Patients assigned to the simvastatin group were arranged to take placebo with breakfast and simvastatin at bedtime, whereas those assigned to the fenofibrate group were arranged to take fenofibrate with breakfast and placebo at bedtime. The patients were seen at the screening visit (i.e. before the 8-week run-in), 1 week before randomization (baseline laboratory and vascular studies), at entry (randomization), and at 4 and 8 weeks of treatment. On week 8, physical examinations, laboratory assessments, and vascular studies were repeated. 2.2. Laboratory assays Two fasting blood samples were obtained at baseline, 7 days apart, and at the end of the 8-week drug-therapy phase (weeks 7.5 and 8). Venous blood samples were placed into tubes containing EDTA and were centrifuged within 30 min at 550 × g for 10 min. The plasma was then separated and stored at −70 ◦ C until analysis. Levels of total cholesterol, total triglycerides, LDL-C and HDL-C were assayed by routine laboratory techniques with the use of methodology of the Lipid Research Clinics, as reported previously [3]. If plasma triglycerides were >400 mg/dl, LDL-C was assessed by a direct method [3]. High-sensitivity CRP was assayed by rate nephelometry (Dade Behring, Newark, Del.). Plasma concentrations of interleukin-1␤ (IL-1␤), soluble CD40, and sCD40L were determined in duplicate using commercially available immunosorbent kits (IL-1␤, R&D Systems;

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

soluble CD40 and sCD40L, Bender MedSystems). Dilution curves of the plasma samples were parallel to those of standard. Routine chemical clinical analyses were performed by standard methods subject to strict quality control. The coefficients of variation were <5% for every type of measurement.

317

ious baseline characteristics. Statistical significance was set at P < 0.05.

3. Results 3.1. Baseline data

2.3. Vascular studies Endothelium-dependent flow-mediated vasodilation in response to reactive hyperemia and endothelium-independent nitroglycerin-induced vasodilation were evaluated in the right brachial artery 1 week before randomization and after 8 weeks of active treatment. Ultrasound measurements were performed using a high-resolution ultrasound machine (Hewlett-Packard 5500) equipped with an L11-3 linear array transducer. Arterial diameters were measured at rest, during reactive hyperemia, again at rest (after vessel recovery), and after administration of 0.6 mg sublingual nitroglycerin. The condition of reactive hyperemia was induced by inflation of a pneumatic cuff on the upper arm to suprasystolic pressure, followed by cuff deflation after 4.5 min. The brachial artery was scanned in longitudinal section 2–8 cm above the elbow, and the arterial diameter was measured on B-mode images with the use of ultrasonic calipers. The end-diastolic arterial diameter was measured from one media–adventitia interface to the other at the clearest section three times at baseline, every 20 s after reactive hyperemia, and after administration of nitroglycerin. The maximum vessel diameter was taken as the average of the three consecutive maximum diameter measurements after hyperemia and nitroglycerin, respectively. Vasodilation was then calculated as the percent change in diameter compared with baseline. In the laboratory, the measurements were performed by a single experienced operator, blinded to the medication studied, in a temperature-controlled room (21–24 ◦ C) at the same time of the day on patients fasted overnight. The intraobserver variation was 1.5%. Medications were omitted on the morning of the visit, and nitrates were withheld for 24 h before the studies. 2.4. Statistical analysis The data were analyzed by non-parametric methods to avoid assumptions about the distribution of measured variables. Comparisons between groups were made using the Mann–Whitney U-test. The differences between baseline and post-treatment values were analyzed using the Wilcoxon signed-rank test. Mann–Whitney analysis was used for comparison of the percentage changes between baseline and post-treatment values in patients receiving fenofibrate versus those receiving simvastatin. The association of these measurements with other baseline biochemical parameters was assessed by the Spearman rank correlation test. Multivariate regression analysis was performed to test the independent association between indexes of endothelial function and var-

Baseline characteristics of the 70 patients with combined hyperlipidemia enrolled in the study are listed in Table 1. There were no significant differences between the two treatment groups in demographic characteristics, conventional risk factors, diagnosis of CHD at enrollment, and plasma lipid and blood pressure levels. Only six (17.1%) patients were free of any conventional risk factor and documented CHD in either treatment group. The two groups showed similar use of various classes of drugs at baseline and during the treatment periods (data not shown). There were neither drop-outs from the study nor serious adverse events. Safety screening biochemistry (including creatine phosphokinase and transaminases) did not change significantly during the study. 3.2. Effects of fenofibrate and simvastatin on biochemical parameters and indexes of vascular reactivity Table 2 shows the effect of both medications on plasma lipid concentrations. After 8 weeks of treatment, patients Table 1 Baseline characteristics of participants Characteristics Age (years) Male sex, n (%) Body-mass index (kg/m2 ) Coronary heart disease, n (%) Hypertension, n (%) Diabetes mellitus, n (%) Current smoker, n (%) Family history of premature coronary heart disease Blood pressure (mmHg) Systolic Diastolic Fasting glucose (mg/dl) Total cholesterol (mg/dl) Triglycerides (mg/dl) LDL-C (mg/dl) HDL-C (mg/dl) Non-HDL-C (mg/dl) Total cholesterol/HDL-C LDL-C/HDL-C

Fenofibrate (n = 35)

Simvastatin (n = 35)

P

60.7 ± 8.6 13 (37.1) 26.3 ± 2.6 9 (25.7)

59.1 ± 11.1 14 (40.0) 25.5 ± 3.1 9 (25.7)

0.576 0.651 0.260 1.000

25 3 5 2

23 2 7 2

0.406 0.509 0.348 1.000

(71.4) (9.4) (14.3) (5.7)

137 ± 15 75 ± 10 107.2 234.6 265.3 142.7 41.3 193.3 5.85 3.54

± ± ± ± ± ± ± ±

18.0 30.5 84.7 29.2 8.3 28.1 1.19 0.82

(65.7) (5.7) (20.0) (5.7)

135 ± 21 79 ± 11

0.639 0.270

± ± ± ± ± ± ± ±

0.136 0.360 0.243 0.462 0.417 0.228 0.474 0.422

104.5 230.6 241.1 140.5 43.0 185.4 5.45 3.34

33.5 31.4 66.0 31.0 9.3 25.7 0.83 0.68

Plus–minus values are means ± S.D. HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol. To convert values for cholesterol to mmol/l, multiply by 0.02586; to convert values for triglycerides to mmol/l, multiply by 0.01129.

318

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

Table 2 Changes in lipid values in fenofibrate and simvastatin groups Characteristics

Fenofibrate (n = 35)

Simvastatin (n = 35)

P

Table 3 Changes in inflammatory markers and indexes of endothelial function in fenofibrate and simvastatin groups Characteristics

Fenofibrate (n = 35)

Simvastatin (n = 35)

P

Total cholesterol (mg/dl) Baseline 234.6 ± 30.5 Study end 196.4 ± 34.0 P <0.001

230.6 ± 31.4 179.0 ± 23.8 <0.001

0.360 0.038 0.029

Triglycerides (mg/dl) Baseline 265.3 ± 84.7 Study end 136.6 ± 42.7 P <0.001

High-sensitivity CRP (mg/dl) Baseline 0.17 (0.09–0.23) Study end 0.13 (0.07–0.20) P 0.018

0.18 (0.08–0.38) 0.12 (0.06–0.19) 0.020

0.632 0.522 0.522

241.1 ± 66.0 170.4 ± 42.8 <0.001

0.243 0.038 <0.001

LDL-C (mg/dl) Baseline Study end P

Interleukin-1␤ (pg/ml) Baseline 0.58 (0.50–0.75) Study end 0.32 (0.10–0.58) P 0.006

0.58 (0.37–0.75) 0.32 (0.12–0.38) 0.002

0.496 0.389 0.387

142.7 ± 29.2 126.8 ± 28.3 0.003

140.5 ± 31.0 105.3 ± 22.2 <0.001

0.462 0.004 0.003

HDL-C (mg/dl) Baseline Study end P

Soluble CD40 (ng/ml) Baseline 158 (106–219) Study end 145 (121–185) P 0.696

152 (110–232) 136 (75–155) 0.447

0.567 0.203 0.642

41.3 ± 8.3 44.6 ± 12.8 0.108

43.0 ± 9.3 45.8 ± 11.1 0.138

0.417 0.906 0.282

Non-HDL-C (mg/dl) Baseline 193.3 ± 28.1 Study end 151.8 ± 33.8 P <0.001

Soluble CD40 ligand (ng/ml) Baseline 5.30 (2.90–6.04) Study end 3.73 (1.17–6.67) P 0.015

6.43 (2.56–8.83) 2.56 (0.85–4.12) 0.001

0.318 0.094 0.150

185.4 ± 25.7 134.1 ± 23.6 <0.001

0.228 0.040 0.122

Vessel size (mm) Baseline Study end P

4.2 (3.8–4.4) 4.2 (3.9–4.5) 0.550

0.331 0.443 0.402

4.3 (4.0–4.7) 4.2 (4.1–4.7) 0.367

Total cholesterol/HDL-C Baseline 5.85 ± 1.19 Study end 4.81 ± 1.79 P 0.012

5.45 ± 0.83 4.11 ± 0.90 <0.001

0.474 0.539 0.938

LDL-C/HDL-C Baseline Study end P

Flow-mediated vasodilation (%) Baseline 8.2 (5.6–9.4) Study end 12.0 (9.6–13.7) P <0.001

7.6 (4.8–10.1) 11.0 (9.2–12.9) <0.001

0.668 0.409 0.863

3.34 ± 0.68 2.44 ± 0.71 <0.001

0.422 0.149 0.197

Nitroglycerin-induced vasodilation (%) Baseline 14.2 (12.9–17.7) Study end 15.1 (13.1–17.3) P 0.367

14.9 (8.1–16.9) 15.9 (10.7–20.4) 0.187

0.637 0.710 0.591

3.54 ± 0.82 3.14 ± 1.34 0.071

Plus–minus values are means ± S.D. HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol. To convert values for cholesterol to mmol/l, multiply by 0.02586; to convert values for triglycerides to mmol/l, multiply by 0.01129.

receiving simvastatin achieved a significantly greater reduction of total cholesterol (22% versus 16%, P = 0.029) and LDL-C (25% versus 11%, P = 0.003), while the decrease in triglycerides (49% versus 29%, P < 0.001) was significantly greater in patients receiving fenofibrate. The marginal increases in HDL-C by both drugs (8% with fenofibrate, 6% with simvastatin) were not statistically significant. As a result, patients receiving simvastatin had significantly lower values of total cholesterol, LDL-C, and non-HDL-C, but higher values of triglycerides than patients treated with fenofibrate after 8 weeks of treatment. Even though the decrease in non-HDL-C was greater in patients receiving simvastatin, the differences in the reduction between the two groups were not statistically significant (28% versus 21%, P = 0.12). The effects of both drugs on various inflammatory markers and indexes of endothelial function are shown in Table 3. There were no differences between the two treatment groups in cytokine concentrations and ultrasound parameters at baseline. Compared with baseline values, treatment with fenofibrate and with simvastatin were associated with sig-

Data presented are median (interquartile range). CRP: C-reactive protein.

nificant reductions in CRP, IL-1␤, and sCD40L concentrations, while soluble CD40 levels were not significantly affected. After treatment, the endothelium-dependent FMD of the right brachial artery markedly improved in both groups, whereas the endothelium-independent nitroglycerin-induced vasodilation did not change significantly. The changes in CRP, IL-1␤, and sCD40L concentrations, and in FMD in patients receiving fenofibrate were not significantly different from those in patients treated with simvastatin. Since fenofibrate and simvastatin were administered at different times of the day, there might have been systematic differences in the time from the last dose when assessments of FMD at the study end were made. Therefore, FMD was assessed 12 and 24 h after the last dose of the study drugs administered in 10 patients taking either fenofibrate (200 mg/day) (n = 5) or simvastatin (20 mg/day) (n = 5) for 8 weeks. In patients receiving fenofibrate, there was only 1.4% variation in two consecutive FMDs taken 12 h apart, whereas in patients treated with simvastatin, the variation was 1.6% (data not shown). Hence, the influence of the different time elapsed from last dose between the two groups on FMD was not considered clinically significant.

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

In both treatment groups, the decreases of CRP, IL-1␤, and sCD40L did not correlate with demographic characteristics, conventional risk factors, diagnosis of CHD, blood pressure levels, and plasma lipid levels (Table 4). Likewise, the changes in FMD did not correlate with demographic characteristics, conventional risk factors, diagnosis of CHD, and blood pressure levels in both groups. However, in the fenofibrate treatment group, the changes in FMD correlated negatively with baseline HDL-C levels. In contrast, the changes in FMD correlated positively with HDL-C levels and negatively with total cholesterol/HDL-C and LDL-C/HDL-C ratios for patients receiving simvastatin (Table 4). The negative correlation between baseline HDL-C levels and changes in FMD in the fenofibrate group, as well as the positive correlation between HDL-C levels and changes in FMD in the simvastatin group, remained statistically significant after adjustment for age, sex, body-mass index, conventional risk factors, and plasma LDL-C, CRP, IL-1␤, and sCD40L levels in multivariate analysis models (data not shown). Relations between the individual changes in FMD and baseline HDL-C levels in each group are presented in the Fig. 1. The patients were further stratified in both groups according to their baseline HDL-C levels to see whether the differential effects of fenofibrate and simvastatin on endothelial function could be clearly demonstrated in the subgroups. The cut-off point of HDL-C was 40 mg/dl, as recommended by the ATP III guidelines [8]. The results showed that for patients with HDL-C ≤40 mg/dl, only treatment with fenofibrate was associated with significant improvement in FMD (Table 5). For patients with HDL-C

Post-treatment FMD/Baseline FMD (%)

3.3. Correlations between changes in markers of inflammation and endothelial function and various baseline characteristics

Fenofibrate 400 r = -0.41 P = 0.02

300

200

100

0 20

30

40

50

60

70

HDL-C (mg/dL)

Simvastatin Post-treatment FMD/Baseline FMD (%)

Among the inflammatory markers, baseline sCD40L levels correlated significantly with baseline levels of soluble CD40 (r = 0.46, P = 0.001) and IL-1␤ (r = 0.44, P = 0.004). There were no significant correlations between CRP and levels of IL-1␤, soluble CD40 and sCD40L (data not shown).

319

400 r = 0.39 P = 0.02

300

200

100

0 20

30

40

50

60

70

HDL-C (mg/dL)

Fig. 1. Correlation between the effects of fenofibrate (upper panel) and simvastatin (lower panel) on endothelium-dependent flow-mediated vasodilation and baseline high-density lipoprotein cholesterol levels. The regression lines are drawn. FMD: flow-mediated dilation; HDL-C: high-density lipoprotein cholesterol.

>40 mg/dl, the changes in FMD were significant only for those receiving simvastatin. There were no significant changes in nitroglycerin-induced vasodilation in all subgroups. It is of interest that in the fenofibrate treatment group, patients with HDL-C ≤40 mg/dl had a significantly

Table 4 Relationships of changes in inflammatory markers and indexes of endothelial function with baseline lipid values in fenofibrate and simvastatin groups Variable

Total cholesterol Triglycerides LDL-C HDL-C Non-HDL-C Total cholesterol/HDL-C LDL-C/HDL-C

Fenofibrate

Simvastatin

CRP

IL-1␤

sCD40L

FMD

CRP

IL-1␤

sCD40L

FMD

0.21 0.04 0.07 0.18 0.06 −0.28 −0.29

−0.14 −0.22 0.06 0.09 −0.04 −0.16 −0.14

0.14 −0.12 0.21 −0.34 −0.22 0.32 0.31

−0.26 0.10 −0.16 −0.41∗ −0.14 0.19 −0.01

−0.05 0.32 −0.18 −0.19 0.11 0.19 −0.10

−0.37 −0.30 0.07 0.18 −0.34 −0.33 −0.13

0.24 0.02 0.22 −0.08 0.17 0.06 0.08

0.17 0.01 0.01 0.39∗ 0.02 −0.41∗ −0.43∗

CRP: C-reactive protein; FMD: flow-mediated dilation; HDL-C: high-density lipoprotein cholesterol; IL-1␤: interleukin-1␤; LDL-C: low-density lipoprotein cholesterol; sCD40L: soluble CD40 ligand. ∗ P < 0.05.

320

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

Table 5 Changes in vascular reactivity in fenofibrate and simvastatin groups stratified by baseline HDL-C levels Variable

Fenofibrate

Simvastatin

HDL-C ≤40 mg/dl (n = 16)

HDL-C >40 mg/dl (n = 19)

HDL-C ≤40 mg/dl (n = 14)

HDL-C >40 mg/dl (n = 21)

4.1 (3.8–4.6) 4.2 (4.0–4.9) 0.255

4.4 (4.2–4.6) 4.3 (4.1–4.6) 0.258

4.2 (3.8–4.4) 4.2 (4.0–4.4) 0.683

4.2 (4.2–4.7) 4.2 (3.9–4.6) 0.840

Flow-mediated vasodilation (%) Baseline 7.5 (5.2–9.3) Study end 12.7 (10.0–14.2) P 0.001

8.2 (7.2–12.1) 10.1 (5.5–13.8) 0.463

8.4 (6.5–11.4) 9.2 (7.6–13.1) 0.271

7.4 (4.6–9.4) 11.2 (9.4–11.8) 0.001

Nitroglycerin-induced vasodilation (%) Baseline 15.6 (14.1–17.9) Study end 14.2 (11.6–19.3) P 0.437

13.9 (11.6–16.4) 15.4 (14.1–17.3) 0.139

14.5 (11.8–18.2) 14.9 (9.4–19.7) 0.925

15.0 (7.7–16.4) 16.4 (11.6–20.9) 0.188

Vessel (mm) Baseline Study end P

Data presented are median (interquartile range). HDL-C: high-density lipoprotein cholesterol.

greater reduction of triglycerides than those with HDL-C >40 mg/dl (52% versus 43%, P = 0.043). In the simvastatin treatment group, patients with HDL-C >40 mg/dl had markedly greater reduction of LDL-C than those with HDL-C ≤40 mg/dl (32% versus 10%, P = 0.026) (data not shown). However, in subgroups stratified by HDL-C, the correlations between lipoprotein changes and improvement in FMD did not reach statistical significance for patients receiving either fenofibrate or simvastatin (data not shown). This might be due to the small number of cases in subgroups. There were no differential effects of both drugs on endothelial function in subgroups stratified by other baseline characteristics except HDL-C levels.

4. Discussion In this study, it was demonstrated that first, in patients with combined hyperlipidemia, treatment with simvastatin was associated with a significantly greater reduction of total cholesterol and LDL-C, while the decrease in triglycerides was significantly greater in patients receiving micronized fenofibrate. Second, both fenofibrate and simvastatin markedly reduced plasma levels of high-sensitivity CRP, IL-1␤, and sCD40L, and improved endothelium-dependent vascular reactivity without mutual differences. Third, the beneficial effects of both drugs on plasma inflammatory markers did not correlate with baseline clinical characteristics. However, the improvement of vascular reactivity with fenofibrate treatment correlated inversely with baseline plasma HDL-C levels, whereas the improvement of vascular reactivity with simvastatin treatment was positively related to baseline HDL-C levels. Accordingly, in the subgroup with baseline HDL-C ≤40 mg/dl, fenofibrate was superior to simvastatin in improving endothelium-dependent vascular reactivity. On the other hand, treatment with simvastatin achieved significantly greater improvement in vascular re-

activity than treatment with fenofibrate in the subgroup with HDL-C >40 mg/dl. As expected, different effects of fenofibrate and simvastatin on plasma lipids and lipoproteins were found in patients with combined hyperlipidemia [24]. Accordingly, if the therapeutic efficacy of both drugs was assessed by their ability to reduce plasma levels of LDL-C and non-HDL-C as suggested by the recent ATP III guidelines [8], simvastatin would be considered superior to fenofibrate in this regard. However, several studies have consistently demonstrated that fibrate therapy, rather than statins, reduces plasma levels of atherogenic remnant lipoproteins and promotes a shift in LDL size toward larger, less atherogenic particles [25,26], which in combination could not be fully evaluated by the mere values of LDL-C and non-HDL-C. Moreover, the extent of CHD event reduction with statin or fibrate monotherapy was similar in subgroups with combined hyperlipidemia in large-scale lipid-intervention trials [5–7]. It is therefore difficult to judge the superiority of either agent by their lipid-lowering effects alone. In this study, the baseline CRP levels (median, 0.17 mg/dl) were relatively low compared to data obtained from other large-scale studies [17–19,27]. This finding indicates that the activity of vascular inflammation was very low in these patients with combined hyperlipidemia. However, after the 8-week treatment of either simvastatin or fenofibrate, both lipid-lowering agents still significantly reduced CRP levels to a similar extent. Statins have been proven to reduce CRP levels in several large-scale studies [18,19]. In contrast, fenofibrate, as a peroxisome proliferator-activated receptor-␣ (PPAR-␣) agonist, has only been shown to be effective in reducing CRP levels in small-sized studies [13,28]. For the first time, it has been demonstrated that both simvastatin and fenofibrate had comparable efficacy in reducing CRP levels in individuals with combined hyperlipidemia. This observation is of importance in light of the notion that purposeful lowering of CRP levels

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

might have a major impact on preventing CHD events [29]. Instead of CRP, recent studies have supported the pivotal role of CD40–CD40L interactions in atherosclerosis, thrombosis, and inflammation [9]. One previous study has shown that statin treatment for 8 weeks decreased sCD40L [30], while another study demonstrated that statin treatment for 3 weeks did not significantly affect sCD40L levels in hypercholesterolemic individuals [31]. The results here confirmed previous observation that among individuals with combined hyperlipidemia, statin treatment for 8 weeks significantly decreased sCD40L. This study further demonstrated that an 8-week fibrate treatment reduced sCD40L to a similar extent as well. Moreover, both statin and fibrate monotherapy significantly decreased plasma levels of IL-1␤, a downstream cytokine to CD40/CD40L dyad [32], without mutual differences. It is of interest to note that sCD40L level did not correlate with CRP levels in this and other studies [23]. However, it is still uncertain whether sCD40L levels provide additional or superior benefits in CHD risk stratification than CRP levels. In this study, both fenofibrate and simvastatin treatment for 8 weeks markedly improved the endothelium-dependent FMD, rather than endothelium-independent nitroglycerininduced vasodilation, in the brachial artery of individuals with combined hyperlipidemia without mutual differences. More interestingly, it was demonstrated for the first time that the effects of simvastatin on FMD correlated positively with baseline HDL-C levels, whereas the effects of fenofibrate on FMD correlated inversely. Furthermore, in subjects with baseline HDL-C ≤40 mg/dl (the recommended cut-off point in ATP III), fenofibrate was superior to simvastatin in improving FMD. Conversely, treatment with simvastatin achieved significantly greater improvement in FMD than treatment with fenofibrate in subjects with HDL-C >40 mg/dl. There are several possible explanations for this intriguing finding. First, in the simvastatin treatment group, it was found that individuals with HDL-C >40 mg/dl had higher total cholesterol and LDL-C levels and significantly greater reduction of LDL-C than those with HDL-C ≤40 mg/dl. This finding is consistent with one previous study, which demonstrated that the LDL-C-lowering effects of statins were positively related to baseline LDL-C levels [33]. Given that the lipid-lowering effect of statins correlates with the improvement in FMD in several studies [34], it may partly explain why individuals with HDL-C >40 mg/dl had markedly greater improvement in FMD than those with HDL-C ≤40 mg/dl in the simvastatin group. The non-lipid effects of statins, including the down-regulation of endothelin-1 expression and the increase in endothelial nitric oxide synthase activity could also contribute to the beneficial effects of statins on the improvement in FMD [14]. However, it is uncertain whether these non-lipid effects of statins varied in individuals with different baseline lipid and lipoprotein levels. Second, in the fenofibrate

321

treatment group, individuals with HDL-C ≤40 mg/dl had significantly greater reduction of triglycerides than those with HDL-C >40 mg/dl, which is consistent with previous studies that demonstrate that the magnitude of triglycerides reduction by fenofibrate is inversely related to baseline HDL-C levels [26]. Moreover, studies have shown that the triglycerides-lowering effect of fibrates correlates with the improvement in FMD [12,15]. Hence, this explains why individuals with HDL-C ≤40 mg/dl had markedly greater improvement in FMD than those with HDL-C >40 mg/dl in the fenofibrate group. Furthermore, the pleiotropic metabolic action of fenofibrate mediating through PPAR-␣ might improve FMD in individuals with features of insulin resistance, of which low HDL-C is a key component [2,14]. Whether the observed differential effects of fenofibrate and simvastatin on endothelial function in patients in different HDL-C strata could translate into clinical benefits deserves further validation in large-scale studies. In contrast to fibrate therapy, which has been consistently demonstrated to be more efficacious for patients with low HDL-C in several large-scale studies [5,6], evidence supporting a differential benefit of statin therapy in patients with low HDL-C is scarce [7]. Despite the recently published post-hoc analysis of the Scandinavian Simvastatin Survival Study, which showed that patients with combined hyperlipidemia and low HDL-C levels had significantly greater reduction of CHD events than those with high levels of both LDL-C and HDL-C, it is still not known whether there is any difference between the subgroups with low and average HDL-C levels as in the present study [35]. There are several limitations of this study. First, the proportion of participants with other coronary risk factors, instead of combined hyperlipidemia, was relatively high in this study. Albeit these coronary risk factors and related medical therapy might confound the therapeutic efficacy of both lipid-lowering drugs, there were no significant differences in baseline characteristics and medication use between the two treatment groups, thus excluding this possibility. Second, because the stratification according to baseline HDL-C occurred post-hoc, the finding of a differential response to fenofibrate or simvastatin based on pre-treatment HDL-C levels is inherently “hypothesis generating” rather than “hypothesis proving”. The present observations should be validated by additional prospective studies, in which patients are stratified by baseline HDL-C prior to randomization.

5. Conclusions This study demonstrated that instead of having different effects on lipoprotein profiles, both fenofibrate and simvastatin had comparative beneficial effects on various plasma inflammatory markers and endothelial function, which were all independent predictors of future CHD. We further showed that for individuals with HDL-C levels ≤40 mg/dl, fenofibrate, rather than simvastatin, had a significantly

322

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323

beneficial effect on endothelial function. On the other hand, for individuals with HDL-C >40 mg/dl, only simvastatin improved endothelial function. The data here suggest that it might be more appropriate to prescribe fenofibrate monotherapy for individuals with combined hyperlipidemia and low HDL-C, a feature of insulin resistance. Simvastatin monotherapy is more efficacious for those whose lipid profile abnormality, despite belonging to the same combined hyperlipidemia category, is primarily LDL-C elevation and average HDL-C. These findings should be further validated in larger-scale clinical studies using hard CHD events as the end-points.

Acknowledgements This work was supported in part by grants from the National Taiwan University Hospital (NTUH90-N008, NTUH91-N017, NTUH 92-N001), the National Science Council (NSC90-2314-B-002-218), and the Academia Sinica (IBMS-CRC90-T05), Republic of China.

References [1] Vicari RM, Wan GJ, Aura AM, Alexander CM, Markson LE, Teutsch SM. Use of simvastatin treatment in patients with combined hyperlipidemia in clinical practice. Arch Fam Med 2000;9:898– 905. [2] Grundy SM. Hypertriglyceridemia, atherogenic dyslipidemia, and the metabolic syndrome. Am J Cardiol 1998;81(Suppl 4A):18B–25B. [3] Wang TD, Chen WJ, Chien KL, et al. Efficacy of cholesterol levels and ratios in predicting future coronary heart disease in a Chinese population. Am J Cardiol 2001;88:737–43. [4] Stein DT, Devaraj S, Balis D, Adams-Huet B, Jialal I. Effect of statin therapy on remnant lipoprotein cholesterol levels in patients with combined hyperlipidemia. Arterioscler Thromb Vasc Biol 2001;21:2026–31. [5] Manninen V, Tenkanen L, Koskinen P, et al. Joint effects of serum triglyceride and LDL cholesterol and HDL cholesterol concentrations on coronary heart disease risk in the Helsinki Heart Study: implications for treatment. Circulation 1992;85:37–45. [6] The BIP Study Group. Secondary prevention by raising HDL cholesterol and reducing triglycerides in patients with coronary artery disease: the Bezafibrate Infarctioin Prevention (BIP) Study. Circulation 2000;102:21–7. [7] Ballantyne CM, Olsson AG, Cook TJ, Mercuri MF, Pedersen TR, Kjekshus J. Influence of low high-density lipoprotein cholesterol and elevated triglyceride on coronary heart disease events and response to simvastatin therapy in 4S. Circulation 2001;104:3046–51. [8] Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;285:2486–97. [9] Glass CK, Witztum JL. Atherosclerosis: the road ahead. Cell 2001;104:503–16. [10] Suwaidi JA, Hamasaki S, Higano ST, Nishimura RA, Holmes DR, Lerman A. Long-term follow-up in patients with mild coronary artery disease and endothelial dysfunction. Circulation 2000;101:948–54. [11] Neunteufl T, Heher S, Katzenschlager R, et al. Late prognostic value of flow-mediated dilation in the brachial artery of patients with chest pain. Am J Cardiol 2000;86:207–10.

[12] Capell WH, DeSouza CA, Poirier P, et al. Short-term triglyceride lowering with fenofibrate improves vasodilator function in subjects with hypertriglyceridemia. Arterioscler Thromb Vasc Biol 2003;23:307–13. [13] Alonso R, Mata P, De Andres R, Villacastin BP, Martinez-Gonzalez J, Badimon L. Sustained long-term improvement of arterial endothelial function in heterozygous familial hypercholesterolemia patients treated with simvastatin. Atherosclerosis 2001;157:423–9. [14] Malik J, Melenovsky V, Wichterle D, et al. Both fenofibrate and atorvastatin improve vascular reactivity in combined hyperlipidemia (fenofibrate versus atorvastatin trial-FAT). Cardiovasc Res 2001;52:290–8. [15] Sebestjen M, Zegura B, Keber I. Both cerivastatin and fenofibrate improve arterial vasoreactivity in patients with combined hyperlipidemia. J Int Med 2002;251:77–85. [16] Tracy RP. Inflammation markers and coronary heart disease. Curr Opin Lipidol 1999;10:435–41. [17] Ridker PM, Hennekens CH, Buring JE, et al. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000;432:836–43. [18] Ridker PM, Rifai N, Pfeffer MA, et al. Long-term effects of pravastatin on plasma concentration of C-reactive protein: the cholesterol and recurrent events (CARE) investigators. Circulation 1999;100:230–5. [19] Albert MA, Danielson E, Rifai N, et al. Effect of statin therapy on C-reactive protein levels: the pravastatin inflammation/CRP evaluation (PRINCE): a randomized trial and cohort study. JAMA 2001;286:64–70. [20] Phipps RP. Atherosclerosis: the emerging role of inflammation and the CD40–CD40 ligand system. Proc Natl Acad Sci USA 2000;97:6930–2. [21] Lutgens E, Gorelik L, Daemen MJ, et al. Requirement for CD154 in the progression of atherosclerosis. Nat Med 1999;5:1313–6. [22] Heeschen C, Dimmeler S, Hamm CW, et al. Soluble CD40 ligand in acute coronary syndromes. N Engl J Med 2003;348:1104–11. [23] Schonbeck U, Varo N, Libby P, Buring J, Ridker PM. Soluble CD40L and cardiovascular risk in women. Circulation 2001;104:2266–8. [24] Farnier M, Bonnefous F, Debbas N, Irvine A. Comparative efficacy and safety of micronized fenofibrate and simvastatin in patients with primary type IIa or IIb hyperlipidemia. Arch Int Med 1994;154:441– 9. [25] Frost RJA, Otto C, Geiss HC, Schwandt P, Parhofer KG. Effects of atorvastatin versus fenofibrate on lipoprotein profiles, low-density lipoprotein subfraction distribution, and hemorheologic parameters in type 2 diabetes mellitus with mixed hyperlipoproteinemia. Am J Cardiol 2001;87:44–8. [26] Despres JP, Lemieux I, Salomon H, Delaval D. Effects of micronized fenofibrate versus atorvastatin in the treatment of dyslipidemic patients with low plasma HDL-cholesterol levels: a 12-week randomized trial. J Int Med 2002;251:490–9. [27] Kuller LH, Tracy RP, Shaten J, et al. Relation of C-reactive protein and coronary heart disease in the MRFIT nested case-control study. Multiple risk factor intervention trial. Am J Epidemiol 1996;144:537– 47. [28] Staels B, Koenig W, Habib A, et al. Activation of human aortic smooth-muscle cells is inhibited by PPAR alpha but not by PPAR gamma activators. Nature 1998;393:790–3. [29] Ridker PM, Rifai N, Clearfield M, et al. Measurement of C-reactive protein for the targeting of statin therapy in the primary prevention of acute coronary events. N Engl J Med 2001;344:1959–65. [30] Cipollone F, Mezzetti A, Porreca E, et al. Association between enhanced soluble CD40L and prothrombotic state in hypercholesterolemia: effects of statin therapy. Circulation 2002;106:399–402. [31] Garlichs CD, John S, Schmeiber A, et al. Upregulation of CD40 and CD40 ligand (CD154) in patients with moderate hypercholesterolemia. Circulation 2001;104:2395–400.

T.-D. Wang et al. / Atherosclerosis 170 (2003) 315–323 [32] Wang CY, Mazer SP, Minamoto K, et al. Suppression of murine cardiac allograft arteriopathy by long-term blockade of CD40–CD154 interactions. Circulation 2002;105:1609–14. [33] de Sauvage Nolting PRW, Buirma RJA, Hutten BA, Kastelein JJP. Baseline lipid values partly determine the response to high-dose simvastatin in patients with familial hypercholesterolemia. Atherosclerosis 2002;164:347–54.

323

[34] Treasure CB, Klein JL, Weintraub WS, et al. Beneficial effects of cholesterol-lowering therapy on the coronary endothelium in patients with coronary artery disease. N Engl J Med 1995;332:481– 7. [35] Robins S. Low high-density lipoprotein cholesterol and response to simvastatin therapy in Scandinavian Simvastatin Survival Study (4S). Circulation 2002;106:e8 [Comment].