- Email: [email protected]
A Randomized Trial of the Effect of Statin and Fibrate Therapy on Arterial Function in CKD
A Randomized Trial of the Effect of Statin and Fibrate Therapy on Arterial Function in CKD Gursharan Dogra, MBBCh, FRACP, PhD,1,2 Ashley Irish, MBBS, FRACP,2 Dick Chan, PhD,1 and Gerald Watts, MD, PhD, FRCP, FRACP1 Background: Although patients with chronic kidney disease (CKD) are at increased risk of cardiovascular disease (CVD), the roles of lipid-modifying therapies in decreasing CVD risk are unclear. Our aim is to compare the effects of statin and fibrate therapy on arterial function as a risk marker of CVD. Study Design: Double-blind, randomized, placebo-controlled, parallel-group study. Setting & Participants: Ambulatory patients with stages 3 to 5 CKD. Intervention: 6 weeks of atorvastatin, 40 mg/d, or gemfibrozil, 600 mg twice daily, with placebo. Outcomes & Measurements: Primary outcome was arterial function assessed by means of endothelial-dependent flow-mediated dilatation (FMD) and small-artery compliance (C2). Secondary outcomes included endothelial-independent glyceryl trinitrate–mediated dilatation (GTNMD), largeartery compliance (C1), and levels of lipids, lipoproteins, and oxidized low-density lipoprotein, as well as markers of insulin resistance and inflammation. Results: Compared with placebo, atorvastatin significantly decreased low-density lipoprotein (⫺52%), triglyceride (⫺30%), and oxidized low-density lipoprotein levels (⫺41%; P ⬍ 0.0001). Gemfibrozil significantly decreased triglyceride levels (⫺40%) and increased high-density lipoprotein levels (⫹20%; P ⬍ 0.0001). Neither atorvastatin nor gemfibrozil had a significant effect on markers of insulin resistance or inflammation. There was no significant change in FMD, GTNMD, or C1 with either atorvastatin or gemfibrozil. There was improvement in C2 with atorvastatin (⫹1.1 mL/mm Hg ⫻ 100) compared with placebo (P ⫽ 0.024), but not with gemfibrozil compared with placebo. Limitations: Small sample size leading to inadequate power, short duration of therapy, and use of a heterogeneous group of patients with CKD and dialysis patients. Conclusion: In patients with advanced CKD, atorvastatin is associated with improvement in dyslipidemia and small-artery stiffness, but not endothelial function. Gemfibrozil improves dyslipidemia, but has no effect on arterial function. Am J Kidney Dis 49:776-785. © 2007 by the National Kidney Foundation, Inc. INDEX WORDS: Chronic kidney disease; endothelial function; arterial compliance; dyslipidemia; statin; fibrate.
lthough the markedly increased risk of cardiovascular disease (CVD) in patients with chronic kidney disease (CKD) is well established,1 the role of uremic dyslipidemia relative to other novel CVD risk factors remains uncertain. Some studies showed an inverse relationship between total cholesterol level and risk of death in patients with CKD,2 whereas more recent studies showed that in the absence of inflammation, there is a more conventional relationship
A
between cholesterol level and CVD mortality.3 The benefits of lipid-lowering therapy also are unclear in patients with CKD, with results of randomized controlled trials pending. Although post hoc analyses from large trials of lipidlowering with pravastatin and gemfibrozil showed significant cardiovascular benefit in patients with early-stage CKD,4,5 these findings were not replicated in 1 study of statin therapy in patients with diabetes with CKD on dialysis therapy6 or
From the 1School of Medicine and Pharmacology, University of Western Australia and Western Australian Heart Research Institute; and 2Department of Nephrology, Royal Perth Hospital, Perth, Western Australia. Received October 23, 2006; accepted in revised form March 6, 2007. Originally published online as doi:10.1053/j.ajkd.2007.03.003 on April 25, 2007. Support: This study was supported by Medical Research Grants from Pfizer Cardiovascular Lipid (CVL) grants, the Faculty of Medicine and Dentistry at the University of
Western Australia, and the Royal Perth Hospital Medical Research Foundation. Potential conflicts of interest: None. Trial registration: www.cochrane-renal.org; study number: CRG080500011. Address correspondence to Gursharan Dogra, MBBCh, FRACP, PhD, School of Medicine and Pharmacology, Level 4, MRF Bldg, Royal Perth Hospital, Box X2213 GPO Perth, Western Australia 6847. E-mail: [email protected] © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/4906-0008$32.00/0 doi:10.1053/j.ajkd.2007.03.003
776
American Journal of Kidney Diseases, Vol 49, No 6 (June), 2007: pp 776-785
Lipid-Lowering in Chronic Kidney Disease
in smaller studies of patients with stages 4 to 5 CKD.7,8 The negative findings may be related to the presence of advanced atherosclerosis with arterial intima-medial calcification as a primary feature6,9 or the inability of the chosen lipidlowering agents to significantly modify uremic dyslipidemia and such concomitant CVD risk factors as increased acute-phase response, hyperoxidative stress, and hypercoagulability.10,11 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors, or statins, inhibit cholesterologenesis, leading to upregulation of low-density lipoprotein (LDL) receptors and decreased veryLDL synthesis. This results in decreased LDL cholesterol and triglyceride levels and a small increase in high-density lipoprotein (HDL) cholesterol levels. Statins were shown to improve endothelial function in at-risk groups12,13 and have anticoagulant, anti-inflammatory, and antioxidant effects that may contribute to their ability to decrease cardiovascular risk.14 Fibrates decrease the synthesis of very-LDL triglycerides and increase lipolytic activity, resulting in triglyceride lowering and an increase in HDL cholesterol levels, thus specifically targeting the type of dyslipidemia seen in patients with CKD. As with statins, fibrates also have significant non–lipidlowering effects, including antifibrinolytic and antithrombogenesis properties that may specifically benefit uremic patients.15,16 Thus, the cardiovascular benefits of statins and fibrates and the lipid and nonlipid mechanisms involved warrant ongoing examination in patients with earlyand advanced-stage CKD with and without diabetes. Endothelial dysfunction is an early phase of atherosclerosis, associated with an abnormality in physiological characteristics of nitric oxide17 that is predictive of coronary events.18-20 It can be detected noninvasively by using high-resolution ultrasonography to measure postischemic flow-mediated dilatation (FMD) of peripheral conduit arteries.21 Arterial stiffness or impaired arterial compliance is a consequence of structural changes related to alterations in physical properties of the arterial media and functional changes associated with an imbalance in the release of such vasoactive substances as nitric oxide.22 Impaired small-artery compliance (C2) was shown to be an independent risk marker for cardiovascular events in nonrenal patients,23 and
777
arterial stiffness is highly predictive of increased cardiovascular mortality in patients with CKD.24-27 Our primary aim is to compare the independent effects of statin and fibrate therapies during a 6-week period on endothelial function and C2 in patients with advanced stages 3 to 5 CKD. The short duration of therapy was chosen based on studies by our group and others that showed significant changes in both endothelial function and arterial compliance during treatment periods ranging from 4 to 6 weeks.13,28-30 Secondarily, we examined the effect of these therapies on glyceryl trinitrate–mediated vasodilatation, largeartery compliance (C1), and levels of lipids, lipoproteins, oxidized LDL, and markers of inflammation and insulin resistance to explore the potential pleiotropic effects of statins and fibrates. METHODS Subjects and Study Design Patients with CKD stages 3 to 5 (Cockcroft-Gault estimated glomerular filtration rate ⬍ 60 mL/min [⬍1 mL/s]) and aged 18 to 80 years were recruited from Perth Renal Units (Western Australia). Dialysis patients were established on renal replacement therapy for 6 months or longer with acceptable indices of dialysis adequacy. Patients with CKD with diabetes and current smokers were included, as well as those using aspirin, angiotensin-converting enzyme inhibitors, and/or other antihypertensive agents. Exclusion criteria were nephrotic-range proteinuria (protein ⬎ 3 g/d), bilateral arteriovenous fistulas, abnormal liver function test results (alanine aminotransferase) or muscle enzyme (creatine kinase) levels greater than 2 times the upper limit of normal, alcohol consumption greater than 30 g/d, active uppergastrointestinal dyspepsia, a clinical cardiovascular event including chest pain or angina within the preceding 6 months, use of anticoagulant or immunosuppressive therapy, and previous intolerance of statins or fibrates. Patients with CKD on lipid-modifying therapy underwent a 6-week washout period, and those on antioxidant vitamin or fish oil therapy underwent a 2-week washout period before study participation. The study received approval from the Royal Perth Hospital Ethics Committee, and all volunteers gave informed written consent. A double-blind, randomized, placebo-controlled, parallelstudy design was used to examine the independent effects of atorvastatin and gemfibrozil compared with placebo on brachial artery endothelial function and systemic arterial compliance. After a 4-week run-in while following a low-fat diet, patients were randomly allocated to receive 6 weeks of either atorvastatin, 40 mg/d; gemfibrozil, 600 mg twice daily; or placebo. Block randomization using random number tables was performed by a clinical trials pharmacist, thereby ensuring allocation concealment. Randomization also was substratified for the status of the patient (predialy-
778 sis, hemodialysis, or peritoneal dialysis), as well as the presence or absence of diabetes mellitus. All investigators, patients, renal staff, and the vascular function technician were blinded to treatment allocation. Clinical, laboratory, and vascular function tests described next were performed at baseline and after 6 weeks of study medication. Safety of therapy was assessed by means of liver and muscle enzyme level determinations (alanine aminotransferase and creatine kinase). Pill counting was used to assess compliance.
Clinical and Laboratory Methods All subjects gave a detailed medical history and underwent medical examination and electrocardiography. Clinical, laboratory, and vascular function tests on hemodialysis patients were always performed on the day after a dialysis session. Blood pressure and heart rate were measured using a Dinamap 1846 SX/P monitor (Critikon Ltd, Tampa, FL) after resting subjects for 10 minutes in the supine position. Weight, height, and waist and hip circumference were measured, and body mass index (weight/height2 [kilograms per square meter]) and waist-hip ratio were calculated. Venous blood samples were obtained after a 12-hour fast, with minimal venous stasis and with subjects in the supine position. Variables were measured by using standard laboratory techniques unless stated otherwise: High-sensitivity C-reactive protein was assayed using a high-sensitivity immunonephelometric method (Dade Behring Marburg GmbH, Marburg, Germany). High-sensitivity interleukin 6 was measured using a high-sensitivity quantitative enzyme immunoassay (Quantikine HS; R&D Systems Inc, Minneapolis, MN). Oxidized LDL was measured using an immunosorbent assay kit (Mercodia AB, Uppsala, Sweden). Insulin sensitivity was calculated using the Homeostasis Model Assessment (HOMA) score (HOMA score ⫽ fasting insulin [U/mL] ⫻ serum glucose [mmol/L]/22.5),31 in which a higher score signifies increasing insulin resistance. Calcium-phosphate product was derived from serum calcium ⫻ phosphate. The interassay coefficient of variation of all analytical assays was less than 7%, except for high-sensitivity interleukin 6, which was 9.9% to 16.5%.
Vascular Function Tests Brachial artery ultrasonography (Acuson Aspen 128 ultrasound device; Acuson Corp, Mountainview, CA) was used to measure endothelium-dependent postischemic FMD and endothelium-independent glyceryl trinitrate–mediated dilatation (GTNMD) using methods described elsewhere.32 Maximal FMD and GTNMD responses were calculated as percentage of change in brachial artery diameter from baseline. Systemic arterial compliance was measured using the CR2000 (Hypertension Diagnostics Inc/Pulse Wave CR2000; Research Cardiovascular Profiling System, Minneapolis, MN). Radial artery pulse contour analysis is used to derive C1 and C2 using a modified version of the Windkessel model, as described elsewhere.22
Statistical Methods and Analysis The desired effect size of a 2% change in FMD, as well as a 2-mL/mm Hg change in C2, was estimated to allow for a mean difference in FMD in repeated within-subject measure-
Dogra et al ments of 1.6% in our hands and take into consideration studies showing that a 1% decrease in FMD is associated with an odds ratio of a cardiovascular event of 1.35,19 and a 2-unit decrease in C2 is associated with an odds ratio of a cardiovascular event of 1.5.23 Comparing treatment arm with placebo (atorvastatin versus placebo or gemfibrozil versus placebo) with n ⫽ 35 per group (total n ⫽ 105), the study would have had greater than 80% power to detect a 2% ⫾ 2.5% (SD) FMD change and 2 ⫾ 2.6-mL/mm Hg C2 change with an ␣ error of 0.05. Analyses were performed using SPSS, version 11.5 (Statistical Package for Social Sciences; SPSS Inc, Chicago, IL). Results are expressed as mean ⫾ SD except for skewed variables, which are presented as geometric mean and 95% confidence interval and were log transformed before statistical analysis. Comparisons were performed between the atorvastatin and placebo groups and the gemfibrozil and placebo groups by using independent t-tests at baseline and general linear modeling adjusting for the corresponding baseline variable for results after 6 weeks of therapy. Posttreatment results also are represented as an estimate of the mean difference and 95% confidence interval. Categorical variables were compared using chi-square and Fisher exact tests. Statistical significance is defined at the 5% level for primary outcome variables (FMD and C2), and using Bonferroni correction for multiple testing, statistical significance is defined at the 0.4% level for secondary outcome variables.
RESULTS
Figure 1 shows in detail the flow of patients from recruitment through randomization, allocation, follow-up, and analysis. Of 119 eligible patients randomly assigned to atorvastatin, gemfibrozil, or placebo, 105 patients completed the baseline study visit and started study medications (Fig 1). Baseline characteristics of these 105 patients as a total group were described in detail elsewhere.33 As shown in Fig 1, after the start of trial medication, an additional 15 patients withdrew because of adverse events, leaving 90 patients who completed the study and were available for analysis (31 patients, atorvastatin; 27 patients, gemfibrozil; and 32 patients, placebo). Baseline characteristics of these patients within their individual allocated treatment groups are listed in Table 1. As listed in Table 2, serum creatinine, hemoglobin, serum albumin, calcium, phosphate, and calcium-phosphate product values were not significantly different among the 3 groups. Estimated glomerular filtration rate was not significantly different among the predialysis subgroups (Table 2). As listed in Table 3, lipid, lipoprotein, and oxidized LDL values were not significantly different between the gemfibrozil and placebo
Lipid-Lowering in Chronic Kidney Disease
779
Figure 1. Recruitment and patient randomization.
groups at baseline. However, total cholesterol and LDL cholesterol levels were significantly greater in the atorvastatin group compared with the placebo group at baseline. Values for markers of insulin resistance (glucose, insulin, and HOMA score), as well as markers of inflammation (Creactive protein and interleukin 6) also were not significantly different among the 3 groups at baseline (Table 3). Finally, all measures of vascular function (FMD, GTNMD, C1, and C2) were not significantly different between the gemfibrozil and placebo groups or between the atorvastatin and placebo groups at baseline (Table 4). Tables 5 and 6 list effects of atorvastatin and gemfibrozil therapy compared with placebo on
lipid, lipoprotein, and oxidized LDL levels, as well as insulin resistance and inflammation. Compared with placebo, atorvastatin therapy was associated with significant decreases in total cholesterol, LDL cholesterol, triglyceride, and oxidized LDL levels, but no change in HDL cholesterol levels. Gemfibrozil therapy was associated with a significant increase in HDL cholesterol levels and decrease in triglyceride levels, but no significant changes in total cholesterol, LDL cholesterol, and oxidized LDL levels. There was no significant effect of atorvastatin or gemfibrozil on markers of insulin resistance and inflammation. Tables 7 and 8 list effects of lipid-modifying therapy compared with placebo on brachial ar-
780
Dogra et al Table 1. Baseline Patient Characteristics in the Placebo, Atorvastatin, and Gemfibrozil Groups Placebo (n ⫽ 32)
Age (y) Predialysis Hemodialysis Peritoneal dialysis Renal diagnosis Chronic glomerulonephritis Diabetic nephropathy Unknown Other Diabetes mellitus Men Smokers Whites Previous cardiovascular disease Use of antihypertensives Use of erythropoietin n Body mass index (kg/m2) Waist-hip ratio Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Pulse pressure (mm Hg)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
62 ⫾ 11 18 9 5
65 ⫾ 14 18 9 4
58 ⫾ 13 17 8 2
11 4 4 13 7 20 5 27 7 24 17 27.9 ⫾ 6.4 0.94 ⫾ 0.12 134.9 ⫾ 24.2 72.4 ⫾ 11.9 62.5 ⫾ 17.4
11 5 4 11 6 19 6 25 6 21 16 26.5 ⫾ 5.2 0.92 ⫾ 0.09 142.6 ⫾ 23.5 77.4 ⫾ 11.6 65.2 ⫾ 19.1
6 3 8 10 6 17 7 18 5 17 10 26.2 ⫾ 4.7 0.91 ⫾ 0.08 134.1 ⫾ 25.9 74.9 ⫾ 12.2 59.3 ⫾ 21.0
Note: Results expressed as mean ⫾ SD or number of patients. P is nonsignificant (⬎0.05) for all comparisons between the placebo and atorvastatin groups and the placebo and gemfibrozil groups.
tery FMD and GTNMD. Compared with placebo, neither atorvastatin nor gemfibrozil therapy was associated with a significant improvement in FMD or GTNMD. As also listed in Tables 7 and 8, compared with placebo, lipid-modifying therapy (atorvastatin or gemfibrozil) was not associated with improvement in C1. As listed in Table 7, there was a small, but significant, improvement in C2 with atorvastatin (⫹1.1 mL/mm Hg ⫻ 100) compared with placebo (P ⫽ 0.024), but not with gemfibrozil compared with placebo, as listed in Table 8.
Compliance with gemfibrozil, but not atorvastatin, was lower than with placebo (placebo, 96.5%; atorvastatin, 91.2%; gemfibrozil, 89.1%). There was a greater incidence of self-reported gastrointestinal side effects (nausea, bloating, and diarrhea) with gemfibrozil, but not with atorvastatin, compared with placebo (placebo, 0 of 32; atorvastatin, 1 of 31; gemfibrozil, 11 of 27). There also was weight loss with gemfibrozil (median, ⫺1.1 kg; range, ⫺3.8 to ⫹0.8 kg) compared with placebo (median, ⫹0.4 kg; range, ⫺1.6 to ⫹2.7 kg). There were no significant
Table 2. Baseline Biochemical Variables in the Placebo, Atorvastatin, and Gemfibrozil Groups
Estimated glomerular filtration rate for predialysis subgroup (mL/min) Hemoglobin (g/dL) Serum albumin (g/dL) Serum calcium (mg/dL) Serum phosphate (mg/dL) Calcium-phosphate product
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
31 (33) ⫾ 13 (n ⫽ 18) 12.2 ⫾ 1.4 4.2 ⫾ 0.3 9.7 (9.5-10.0) 4.2 (3.7-4.8) 40.7 (35.1-48.0)
28 (25) ⫾ 12 (n ⫽ 18) 12.6 ⫾ 1.4 4.1 ⫾ 2.9 9.5 (9.3-9.7) 4.6 (4.1-5.2) 43.7 (38.1-50.4)
34 (29) ⫾ 16 (n ⫽ 17) 12.5 ⫾ 1.5 4.2 ⫾ 0.4 9.5 (9.3-9.8) 3.9 (3.5-4.4) 37.1 (32.6-43.1)
Note: Results expressed as mean ⫾ SD, mean (median) ⫾ SD, and geometric mean (95% confidence interval). P is nonsignificant (⬎0.05) for all comparisons between the placebo and atorvastatin groups and the placebo and gemfibrozil groups. To convert estimated glomerular filtration rate in mL/min to mL/s, multiply by 0.01667; hemoglobin in g/dL to g/L, multiply by 10; albumin in g/dL, multiply by 10; calcium in mg/dL to mmol/L, multiply by 0.2495; phosphate in mg/dL to mmol/L, multiply by 0.3229.
Lipid-Lowering in Chronic Kidney Disease
781
Table 3. Baseline Values for Lipids, Lipoproteins, Insulin Resistance, and Markers of Inflammation in the Placebo, Atorvastatin, and Gemfibrozil Groups
Cholesterol (mg/dL)* LDL (mg/dL)* Triglycerides (mg/dL)† HDL (mg/dL)† Oxidized LDL (U/L)† Serum glucose (mg/dL)† Serum insulin (U/mL)† HOMA score† C-Reactive protein (mg/L)† Interleukin 6 (ng/L)†
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
197 ⫾ 49 112 ⫾ 43 148 (124-186) 46.0 (41.8-51.0) 77.8 ⫾ 26.7 92 (83-104) 19.6 (15.5-24.6) 4.5 (3.4-5.9) 5.3 (3.2-8.7) 4.3 (3.3-5.7)
228 ⫾ 49 139 ⫾ 44 177 (142-213) 47.5 (42.2-53.4) 89.1 ⫾ 31.0 86 (77-99) 14.3 (10.7-19.1) 3.1 (2.2-4.3) 3.2 (2.2-4.7) 3.1 (2.6-3.8)
217 ⫾ 35 131 ⫾ 34 153 (124-195) 49.9 (44.1-56.1) 80.1 ⫾ 20.5 86 (79-95) 14.7 (10.8-19.8) 3.2 (2.3-4.4) 3.1 (1.9-5.0) 3.0 (2.4-3.8)
Note: Results expressed as mean ⫾ SD and geometric mean (95% confidence interval). To convert cholesterol, HDL cholesterol and LDL cholesterol in mg/dL to mmol/L, multiply by 0.02586; triglycerides in mg/dL to mmol/L, multiply by 0.01129; glucose in mg/dL to mmol/L, multiply by 0.05551; insulin in U/mL to pmol/L, multiply by 7.175. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, Homeostasis Model Assessment. *P ⬍ 0.05 for atorvastatin group compared with placebo group, but not significant for gemfibrozil group compared with placebo group. †P is nonsignificant (P ⬎ 0.05) for comparisons between the placebo and atorvastatin groups and placebo and gemfibrozil groups.
episodes of myalgia, myositis, or creatine kinase or alanine aminotransferase level increase with atorvastatin or gemfibrozil, and within the predialysis group, there was no significant increase in serum creatinine level. DISCUSSION
This study shows that lipid-lowering therapy during 6 weeks has no effect on endothelial function and a minimal effect on arterial compliance in patients with advanced stages 3 to 5 CKD. Although atorvastatin therapy was associated with a significant decrease in total cholesterol, LDL cholesterol, triglyceride, and oxidized LDL levels, there was no improvement in FMD and only a small improvement in C2. Gemfibrozil therapy was associated with a decrease in
triglyceride levels and increase in HDL cholesterol levels, but had no significant effects on FMD or C2 and was poorly tolerated by patients. This suggests that correction of dyslipidemia alone, by either targeting intensive LDL cholesterol lowering or correcting high triglyceride and low HDL cholesterol levels, is unlikely to significantly improve arterial function in the short term. A similar lack of improvement in endothelial function and other surrogate CVD risk markers was shown previously by studies incorporating lipid-lowering with a statin as part of a multiple risk factor–lowering approach in patients with advanced CKD.7,34 The improvement in arterial compliance with atorvastatin in our study, although of a small degree, is consistent with a previous study showing improvement in arterial
Table 4. Baseline Brachial Artery Vascular Function and Systemic Arterial Compliance in the Placebo, Atorvastatin and Gemfibrozil Groups
Resting brachial artery diameter (mm) FMD (%) GTNMD (%) C1 (mL/mm Hg ⫻ 10) C2 (mL/mm Hg ⫻ 100)
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
3.7 ⫾ 0.7 3.5 ⫾ 2.3 16.0 ⫾ 9.2 12.3 (10.7-14.1) 3.8 (3.1-4.7)
3.5 ⫾ 0.6 4.2 ⫾ 4.6 16.1 ⫾ 6.3 12.4 (10.9-14.2) 3.9 (3.1-4.8)
3.7 ⫾ 0.6 4.1 ⫾ 3.5 15.4 ⫾ 8.5 13.2 (11.1-15.7) 4.2 (3.3-5.3)
Note: Results expressed as mean ⫾ SD or geometric mean (95% confidence interval). P is nonsignificant (⬎0.05) for all comparisons between the placebo and atorvastatin groups and placebo and gemfibrozil groups. Abbreviations: FMD, flow-mediated dilatation; GTNMD, glyceryl trinitrate–mediated dilatation; C1, large-artery compliance; C2, small-artery compliance.
782
Dogra et al Table 5. Lipids, Lipoproteins, Insulin Resistance, and Markers of Inflammation After Treatment With Placebo and Atorvastatin
Cholesterol (mg/dL) LDL (mg/dL) Triglycerides* (mg/dL) HDL* (mg/dL) Oxidized LDL (U/L) Serum glucose (mg/dL) Serum insulin (U/mL) HOMA score C-Reactive protein (mg/L) Interleukin 6 (ng/L)
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Mean Difference (95% confidence interval)
P
194 ⫾ 50 112 ⫾ 41 151 (126-182)* 46.1 (41.7-51.0)* 76.1 ⫾ 25.1 87 (79-96)* 18.3 (14.0-24.0)* 3.9 (3.0-5.2)* 4.7 (3.0-7.6)* 4.2 (3.2-5.5)*
139 ⫾ 24 64 ⫾ 16 124 (106-144)* 45.8 (40.8-51.5)* 50.1 ⫾ 17.6 88 (77-100)* 15.1 (11.6-19.8)* 3.3 (2.4-4.5)* 5.2 (2.9-8.7)* 4.3 (3.4-5.5)*
⫺75.9 (⫺89.1 to ⫺62.7) ⫺62.9 (⫺73.6 to ⫺52.2) ⫺50.1 (⫺77.8 to ⫺22.5)† ⫺1.7 (⫺4.9 to ⫹1.5)† ⫺31.7 (⫺40.1 to ⫺23.3) ⫹9.0 (⫺0.9 to ⫹19.0)† ⫹0.3 (⫺2.1 to ⫹2.9)† ⫹0.9 (⫺0.4 to ⫹1.7)† ⫹12.2 (⫺6.5 to ⫹30.9)† ⫹0.4 (⫺1.7 to ⫹2.5)†
⬍0.0001 ⬍0.0001 ⬍0.0001 0.298 ⬍0.0001 0.084 0.171 0.064 0.209 0.277
Note: Results in atorvastatin and placebo columns expressed as mean ⫾ SD unless noted otherwise. To convert cholesterol, HDL cholesterol and LDL cholesterol in mg/dL to mmol/L, multiply by 0.02586; triglycerides in mg/dL to mmol/L, multiply by 0.01129; glucose in mg/dL to mmol/L, multiply by 0.05551; insulin in U/mL to pmol/L, multiply by 7.175. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, Homeostasis Model Assessment. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
stiffness with fluvastatin in dialysis patients.35 This possible divergent benefit of statin therapy may be a consequence of differential effects on resistance versus conduit arteries and on endothelial vasodilators other than nitric oxide. In addition, the small benefit of atorvastatin on arterial compliance may be related to the significantly greater LDL cholesterol lowering (50%) achieved in our study than previous studies,7 as well as decrease in oxidized LDL levels. This minimal effect of atorvastatin on arterial compliance may be consistent with the significant decrease in some secondary cardiac end points, but not in the
primary end points, in Assessment of Lescol in Renal Transplantation and the recent German Diabetes and Dialysis Study.6,36 The clinical significance of increased degrees of LDL cholesterol lowering will be tested further in the Study of Heart and Renal Protection, which combines ezetimibe with simvastatin, the results of which are still pending.37 Our study also shows that gemfibrozil is poorly tolerated in patients with advanced CKD, and although it corrects uremic dyslipidemia, it has no effect on endothelial function or arterial compliance. To our knowledge, this is the first study
Table 6. Lipids, Lipoproteins, Insulin Resistance, and Markers of Inflammation After Treatment With Placebo and Gemfibrozil
Cholesterol (mg/dL) LDL (mg/dL) Triglycerides (mg/dL) HDL (mg/dL) Oxidized LDL (U/L) Serum glucose (mg/dL) Serum insulin (U/mL) HOMA score C-Reactive protein (mg/L) Interleukin 6 (ng/L)
Placebo (n ⫽ 32)
Gemfibrozil (n ⫽ 27)
Mean Difference (95% confidence interval)
P
194 ⫾ 50 112 ⫾ 41 151 (126-182)* 46.1 (41.7-51.0)* 76.1 ⫾ 25.1 87 (79-96)* 18.3 (14.0-24.0)* 3.9 (3.0-5.2)* 4.7 (3.0-7.6)* 4.2 (3.2-5.5)*
208 ⫾ 41 114 ⫾ 39 96 (74-124)* 58.7 (52.1-66.3)* 73.9 ⫾ 21.5 89 (76-103)* 13.9 (10.4-18.5)* 3.0 (2.2-4.2)* 3.5 (2.1-5.9)* 4.0 (3.1-5.2)*
⫺5.3 (⫺16.1 to ⫹5.4) ⫺3.0 (⫺13.3 to ⫹7.4) ⫺56.0 (⫺84.9 to ⫺27.1) ⫹9.6 (⫹5.4 to ⫹13.8) ⫺4.1 (⫺11.7 to ⫹3.6) ⫹13.8 (⫺10.0 to ⫹37.7)† ⫺1.1 (⫺3.7 to ⫹1.4)† ⫹0.2 (⫺0.4 to ⫹0.9)† ⫺0.4 (⫺5.1 to ⫹4.2)† ⫹0.2 (⫺1.8 to ⫹2.2)†
0.328 0.564 ⬍0.0001 ⬍0.0001 0.292 0.269 0.972 0.483 0.671 0.231
Note: Results in the gemfibrozil and placebo columns expressed as mean ⫾ SD unless noted otherwise. To convert cholesterol, HDL cholesterol and LDL cholesterol in mg/dL to mmol/L, multiply by 0.02586; triglycerides in mg/dL to mmol/L, multiply by 0.01129; glucose in mg/dL to mmol/L, multiply by 0.05551; insulin in U/mL to pmol/L, multiply by 7.175. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, Homeostasis Model Assessment. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
Lipid-Lowering in Chronic Kidney Disease
783
Table 7. Brachial Artery Vascular Function and Systemic Arterial Compliance After Treatment With Placebo and Atorvastatin
FMD (%) GTNMD (%) C1 (mL/mm Hg ⫻ 10) C2 (mL/mm Hg ⫻ 100)
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Mean Difference (95% confidence interval)
P
4.0 ⫾ 2.6 15.3 ⫾ 7.6 12.5 (10.9-14.5)* 3.9 (3.2-4.7)*
3.5 ⫾ 4.1 16.7 ⫾ 7.9 13.3 (11.6-15.3)* 4.8 (3.9-5.9)*
⫺1.0 (⫺2.1 to ⫹0.2) ⫹1.5 (⫺1.1 to ⫹4.1) ⫹0.7 (⫺0.9 to ⫹2.2)† ⫹1.1 (⫹0.1 to ⫹2.2)†
0.099 0.254 0.408 0.024
Note: Results in atorvastatin and placebo columns expressed as mean ⫾ SD unless noted otherwise. Abbreviations: FMD, flow-mediated dilatation; GTNMD, glyceryl trinitrate–mediated dilatation; C1, large-artery compliance; C2, small-artery compliance. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
to examine the effects of fibrate therapy on surrogate CVD markers. However, although there are no clinical trials of fibrates and CVD in patients with advanced CKD, post hoc analysis of the Veteran’s Affairs-High Density Lipoprotein Cholesterol Intervention Trial showed a significant benefit of gemfibrozil in patients with mild CKD, with minimal side effects.5 Patients in our study had moderate to severe CKD and cannot be compared with the Veteran’s AffairsHigh Density Lipoprotein Cholesterol Intervention Trial patients. A notable finding of our study is the high incidence of gastrointestinal side effects from gemfibrozil therapy, resulting in poor compliance with pills. Although this raises the possibility that gemfibrozil is less likely to have a role in clinical trials in this advanced CKD population at the dose of 600 mg twice daily, it is possible that a lower dose of 600 mg/d may have equal lipid-lowering efficacy with reduced side effects. Although our findings suggest that dyslipidemia is not a primary contributor to arterial dysfunction in patients with CKD, with little benefit to be obtained from statin and fibrate therapy,
recent observational studies suggested significant potential cardiovascular benefit from statin therapy.38,39 The negative findings in our studies and others could be explained by the presence of advanced and irreversible vascular disease driven primarily by calcific arteriopathy, rather than atherosclerosis, contributing to the extremely high rate of cardiac events that characterizes patients with advanced CKD. Thus, starting statin therapy earlier in the course of CKD may be beneficial, as suggested by the Assessment of Lescol in Renal Transplantation study40 and post hoc analyses showing that statin and fibrate use in patients with early renal disease is of considerable cardiovascular benefit.4,5,41 Furthermore, in patients with advanced CKD, other nonconventional CVD risk factors, such as calcium-phosphate product, inflammation, oxidative stress, and insulin resistance, are likely to have a significant contributory role33,42,43 and warrant separate examination. In our study, despite the known pleiotropic effects of statins and fibrates, levels of oxidized LDL, but none of the inflammatory markers and markers of insulin resistance, were decreased by statin therapy, and neither was
Table 8. Brachial Artery Vascular Function and Systemic Arterial Compliance After Treatment With Placebo and Gemfibrozil
FMD (%) GTNMD (%) C1 (mL/mm Hg ⫻ 10) C2 (mL/mm Hg ⫻ 100)
Placebo (n ⫽ 32)
Gemfibrozil (n ⫽ 27)
Mean Difference (95% confidence interval)
P
4.0 ⫾ 2.6 15.3 ⫾ 7.6 12.5 (10.9-14.5)* 3.9 (3.2-4.7)*
3.7 ⫾ 3.2 16.1 ⫾ 8.0 13.6 (11.5-16.1)* 4.2 (3.4-5.3)*
⫺0.7 (⫺1.8 to ⫹0.4) 1.2 (⫺1.1 to ⫹3.4) 0.2 (⫺1.5 to ⫹1.9)† 0.1 (⫺0.8 to ⫹1.0)†
0.214 0.299 0.780 0.854
Note: Results in the gemfibrozil and placebo columns expressed as mean ⫾ SD unless noted otherwise. Abbreviations: FMD, flow-mediated dilatation; GTNMD, glyceryl trinitrate–mediated dilatation; C1, large-artery compliance; C2, small-artery compliance. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
784
Dogra et al
decreased by fibrate therapy, possibly explaining a negative effect on arterial function. Additionally, the negative findings of this study could be attributed to a number of limitations, including the small sample size leading to inadequate power, short duration of therapy, and use of a heterogeneous group of patients with CKD and dialysis patients. A longer duration of therapy warrants examination because the improvement in C2 on atorvastatin therapy may be more significant and extend to endothelial function as well, or, conversely, the improvements may not be sustained. An additional limitation to consider is that this and other studies did not measure specific apolipoprotein-defined lipoprotein subclasses that might better explain the lack of efficacy of these drugs.44 In conclusion, despite improvement in dyslipidemia in patients with stages 3 to 5 CKD, short-term atorvastatin and gemfibrozil therapy failed to improve endothelial function, and atorvastatin, but not gemfibrozil, was associated with a small improvement in C2. ACKNOWLEDGMENT The authors acknowledge financial support from Pfizer Cardiovascular Lipid (CVL) grants, Faculty of Medicine and Dentistry at the University of Western Australia, and the Royal Perth Hospital Medical Research Foundation and thank the Perth Renal Units, which referred patients for the study, and Lisa Rich, for technical assistance with ultrasonography.
REFERENCES 1. Baigent C, Burbury K, Wheeler D: Premature cardiovascular disease in chronic renal failure. Lancet 356:147152, 2000 2. Lowrie EG, Lew NL: Death risk in hemodialysis patients: The predictive value of commonly measured variables and an evaluation of death rate differences between facilities. Am J Kidney Dis 15:458-482, 1990 3. Liu Y, Coresh J, Eustace JA, et al: Association between cholesterol level and mortality in dialysis patients: Role of inflammation and malnutrition. JAMA 291:451-459, 2004 4. Tonelli M, Isles C, Curhan GC, et al: Effect of pravastatin on cardiovascular events in people with chronic kidney disease. Circulation 110:1557-1563, 2004 5. Tonelli M, Collins D, Robins S, et al: Effect of gemfibrozil on change in renal function in men with moderate chronic renal insufficiency and coronary disease. Am J Kidney Dis 44:832-839, 2004 6. Wanner C, Krane V, Marz W, et al: Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med 353:238-248, 2005 7. Isbel NM, Haluska B, Johnson DW, et al: Increased targeting of cardiovascular risk factors in patients with
chronic kidney disease does not improve atheroma burden or cardiovascular function. Am Heart J 151:745-753, 2006 8. Rakhit DJ, Marwick TH, Armstrong KA, et al: Effect of aggressive risk factor modification on cardiac events and myocardial ischaemia in patients with chronic kidney disease. Heart 92:1402-1408, 2006 9. Blacher J, Guerin AP, Pannier B, et al: Arterial calcifications, arterial stiffness, and cardiovascular risk in endstage renal disease. Hypertension 38:938-942, 2001 10. Yeun JY, Kaysen GA: C-Reactive protein, oxidative stress, homocysteine, and troponin as inflammatory and metabolic predictors of atherosclerosis in ESRD. Curr Opin Nephrol Hypertens 9:621-630, 2000 11. Irish AB, Green FR: Factor VII coagulant activity (VIIc) and hypercoagulability in chronic renal disease and dialysis: Relationship with dyslipidaemia, inflammation, and factor VII genotype. Nephrol Dial Transplant 13:679-684, 1998 12. 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 332:481-487, 1995 13. Dogra GK, Watts GF, Chan DC, et al: Statin therapy improves brachial artery vasodilator function in patients with type 1 diabetes and microalbuminuria. Diabet Med 22:239-242, 2005 14. Vaughan CJ, Gotto AM Jr, Basson CT: The evolving role of statins in the management of atherosclerosis. J Am Coll Cardiol 35:1-10, 2000 15. Irish AB, Thompson CH: The effects of gemfibrozil upon the hypercoagulable state in dyslipidaemic patients with chronic renal failure. Nephrol Dial Transplant 11:22232228, 1996 16. Watts GF, Dimmitt SB: Fibrates, dyslipoproteinaemia and cardiovascular disease. Curr Opin Lipidol 10:561-574, 1999 17. Rubanyi GM: The role of endothelium in cardiovascular homeostasis and diseases. J Cardiovasc Pharmacol 22:S1S14, 1993 (suppl 4) 18. Suwaidi JA, Hamasaki S, Higano ST, et al: Longterm follow-up of patients with mild coronary artery disease and endothelial dysfunction. Circulation 101:948-954, 2000 19. 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 86:207-210, 2000 20. Schachinger V, Britten MB, Zeiher AM: Prognostic impact of coronary vasodilator dysfunction on adverse longterm outcome of coronary heart disease. Circulation 101: 1899-1906, 2000 21. Celermajer DS, Sorensen KE, Gooch VM, et al: Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 340:11111115, 1992 22. Woodman RJ, Watts GF: Measurement and application of arterial stiffness in clinical research: Focus on new methodologies and diabetes mellitus. Med Sci Monit 9:RA81RA89, 2003 23. Grey E, Bratteli C, Glasser SP, et al: Reduced small artery but not large artery elasticity is an independent risk
Lipid-Lowering in Chronic Kidney Disease marker for cardiovascular events. Am J Hypertens 16:265-269, 2003 24. London GM, Guerin AP, Marchais SJ, et al: Cardiac and arterial interactions in end-stage renal disease. Kidney Int 50:600-608, 1996 25. London GM: Arterial function in renal failure. Nephrol Dial Transplant 13:S12-S15, 1998 (suppl 4) 26. Blacher J, Safar ME, Guerin AP, et al: Aortic pulse wave velocity index and mortality in end-stage renal disease. Kidney Int 63:1852-1860, 2003 27. London GM, Blacher J, Pannier B, et al: Arterial wave reflections and survival in end-stage renal failure. Hypertension 38:434-438, 2001 28. Tsunekawa T, Hayashi T, Kano H, et al: Cerivastatin, a hydroxymethylglutaryl coenzyme A reductase inhibitor, improves endothelial function in elderly diabetic patients within 3 days. Circulation 104:376-379, 2001 29. McVeigh GE, Brennan GM, Cohn JN, et al: Fish oil improves arterial compliance in non-insulin-dependent diabetes mellitus. Arterioscler Thromb 14:1425-1429, 1994 30. Nestel PJ, Pomeroy SE, Sasahara T, et al: Arterial compliance in obese subjects is improved with dietary plant n-3 fatty acid from flaxseed oil despite increased LDL oxidizability. Arterioscler Thromb Vasc Biol 17:1163-1170, 1997 31. Matthews DR, Hosker JP, Rudenski AS, et al: Homeostasis Model Assessment: Insulin resistance and betacell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28:412-419, 1985 32. Woodman RJ, Playford DA, Watts GF, et al: Improved analysis of brachial artery ultrasound using a novel edge-detection software system. J Appl Physiol 91:929-937, 2001 33. Dogra G, Irish A, Chan D, et al: Insulin resistance, inflammation, and blood pressure determine vascular dysfunction in CKD. Am J Kidney Dis 48:926-934, 2006 34. Fathi R, Isbel N, Short L, et al: The effect of longterm aggressive lipid lowering on ischemic and atheroscle-
785 rotic burden in patients with chronic kidney disease. Am J Kidney Dis 43:45-52, 2004 35. Ichihara A, Hayashi M, Ryuzaki M, et al: Fluvastatin prevents development of arterial stiffness in haemodialysis patients with type 2 diabetes mellitus. Nephrol Dial Transplant 17:1513-1517, 2002 36. Holdaas H, Fellstrom B, Jardine AG, et al: Effect of fluvastatin on cardiac outcomes in renal transplant recipients: A multicentre, randomised, placebo-controlled trial. Lancet 361:2024-2031, 2003 37. Baigent C, Landry M: Study of Heart and Renal Protection (SHARP). Kidney Int Suppl 84:S207-S210, 2003 38. Seliger SL, Weiss NS, Gillen DL, et al: HMG-CoA reductase inhibitors are associated with reduced mortality in ESRD patients. Kidney Int 61:297-304, 2002 39. Mason NA, Bailie GR, Satayathum S, et al: HMGcoenzyme A reductase inhibitor use is associated with mortality reduction in hemodialysis patients. Am J Kidney Dis 45:119-126, 2005 40. Jardine AG, Holdaas H, Fellstrom B, et al: Fluvastatin prevents cardiac death and myocardial infarction in renal transplant recipients: Post-hoc subgroup analyses of the ALERT Study. Am J Transplant 4:988-995, 2004 41. Heart Protection Study Collaborative Group: MRC/ BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: A randomised placebo-controlled trial. Lancet 360:7-22, 2002 42. Shinohara K, Shoji T, Emoto M, et al: Insulin resistance as an independent predictor of cardiovascular mortality in patients with end-stage renal disease. J Am Soc Nephrol 13:1894-1900, 2002 43. London GM, Guerin AP, Marchais SJ, et al: Arterial media calcification in end-stage renal disease: Impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant 18:1731-1740, 2003 44. Alaupovic P, Attman PO, Knight-Gibson C, et al: Effect of fluvastatin on apolipoprotein-defined lipoprotein subclasses in patients with chronic renal insufficiency. Kidney Int 69:1865-1871, 2006
lthough the markedly increased risk of cardiovascular disease (CVD) in patients with chronic kidney disease (CKD) is well established,1 the role of uremic dyslipidemia relative to other novel CVD risk factors remains uncertain. Some studies showed an inverse relationship between total cholesterol level and risk of death in patients with CKD,2 whereas more recent studies showed that in the absence of inflammation, there is a more conventional relationship
A
between cholesterol level and CVD mortality.3 The benefits of lipid-lowering therapy also are unclear in patients with CKD, with results of randomized controlled trials pending. Although post hoc analyses from large trials of lipidlowering with pravastatin and gemfibrozil showed significant cardiovascular benefit in patients with early-stage CKD,4,5 these findings were not replicated in 1 study of statin therapy in patients with diabetes with CKD on dialysis therapy6 or
From the 1School of Medicine and Pharmacology, University of Western Australia and Western Australian Heart Research Institute; and 2Department of Nephrology, Royal Perth Hospital, Perth, Western Australia. Received October 23, 2006; accepted in revised form March 6, 2007. Originally published online as doi:10.1053/j.ajkd.2007.03.003 on April 25, 2007. Support: This study was supported by Medical Research Grants from Pfizer Cardiovascular Lipid (CVL) grants, the Faculty of Medicine and Dentistry at the University of
Western Australia, and the Royal Perth Hospital Medical Research Foundation. Potential conflicts of interest: None. Trial registration: www.cochrane-renal.org; study number: CRG080500011. Address correspondence to Gursharan Dogra, MBBCh, FRACP, PhD, School of Medicine and Pharmacology, Level 4, MRF Bldg, Royal Perth Hospital, Box X2213 GPO Perth, Western Australia 6847. E-mail: [email protected] © 2007 by the National Kidney Foundation, Inc. 0272-6386/07/4906-0008$32.00/0 doi:10.1053/j.ajkd.2007.03.003
776
American Journal of Kidney Diseases, Vol 49, No 6 (June), 2007: pp 776-785
Lipid-Lowering in Chronic Kidney Disease
in smaller studies of patients with stages 4 to 5 CKD.7,8 The negative findings may be related to the presence of advanced atherosclerosis with arterial intima-medial calcification as a primary feature6,9 or the inability of the chosen lipidlowering agents to significantly modify uremic dyslipidemia and such concomitant CVD risk factors as increased acute-phase response, hyperoxidative stress, and hypercoagulability.10,11 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors, or statins, inhibit cholesterologenesis, leading to upregulation of low-density lipoprotein (LDL) receptors and decreased veryLDL synthesis. This results in decreased LDL cholesterol and triglyceride levels and a small increase in high-density lipoprotein (HDL) cholesterol levels. Statins were shown to improve endothelial function in at-risk groups12,13 and have anticoagulant, anti-inflammatory, and antioxidant effects that may contribute to their ability to decrease cardiovascular risk.14 Fibrates decrease the synthesis of very-LDL triglycerides and increase lipolytic activity, resulting in triglyceride lowering and an increase in HDL cholesterol levels, thus specifically targeting the type of dyslipidemia seen in patients with CKD. As with statins, fibrates also have significant non–lipidlowering effects, including antifibrinolytic and antithrombogenesis properties that may specifically benefit uremic patients.15,16 Thus, the cardiovascular benefits of statins and fibrates and the lipid and nonlipid mechanisms involved warrant ongoing examination in patients with earlyand advanced-stage CKD with and without diabetes. Endothelial dysfunction is an early phase of atherosclerosis, associated with an abnormality in physiological characteristics of nitric oxide17 that is predictive of coronary events.18-20 It can be detected noninvasively by using high-resolution ultrasonography to measure postischemic flow-mediated dilatation (FMD) of peripheral conduit arteries.21 Arterial stiffness or impaired arterial compliance is a consequence of structural changes related to alterations in physical properties of the arterial media and functional changes associated with an imbalance in the release of such vasoactive substances as nitric oxide.22 Impaired small-artery compliance (C2) was shown to be an independent risk marker for cardiovascular events in nonrenal patients,23 and
777
arterial stiffness is highly predictive of increased cardiovascular mortality in patients with CKD.24-27 Our primary aim is to compare the independent effects of statin and fibrate therapies during a 6-week period on endothelial function and C2 in patients with advanced stages 3 to 5 CKD. The short duration of therapy was chosen based on studies by our group and others that showed significant changes in both endothelial function and arterial compliance during treatment periods ranging from 4 to 6 weeks.13,28-30 Secondarily, we examined the effect of these therapies on glyceryl trinitrate–mediated vasodilatation, largeartery compliance (C1), and levels of lipids, lipoproteins, oxidized LDL, and markers of inflammation and insulin resistance to explore the potential pleiotropic effects of statins and fibrates. METHODS Subjects and Study Design Patients with CKD stages 3 to 5 (Cockcroft-Gault estimated glomerular filtration rate ⬍ 60 mL/min [⬍1 mL/s]) and aged 18 to 80 years were recruited from Perth Renal Units (Western Australia). Dialysis patients were established on renal replacement therapy for 6 months or longer with acceptable indices of dialysis adequacy. Patients with CKD with diabetes and current smokers were included, as well as those using aspirin, angiotensin-converting enzyme inhibitors, and/or other antihypertensive agents. Exclusion criteria were nephrotic-range proteinuria (protein ⬎ 3 g/d), bilateral arteriovenous fistulas, abnormal liver function test results (alanine aminotransferase) or muscle enzyme (creatine kinase) levels greater than 2 times the upper limit of normal, alcohol consumption greater than 30 g/d, active uppergastrointestinal dyspepsia, a clinical cardiovascular event including chest pain or angina within the preceding 6 months, use of anticoagulant or immunosuppressive therapy, and previous intolerance of statins or fibrates. Patients with CKD on lipid-modifying therapy underwent a 6-week washout period, and those on antioxidant vitamin or fish oil therapy underwent a 2-week washout period before study participation. The study received approval from the Royal Perth Hospital Ethics Committee, and all volunteers gave informed written consent. A double-blind, randomized, placebo-controlled, parallelstudy design was used to examine the independent effects of atorvastatin and gemfibrozil compared with placebo on brachial artery endothelial function and systemic arterial compliance. After a 4-week run-in while following a low-fat diet, patients were randomly allocated to receive 6 weeks of either atorvastatin, 40 mg/d; gemfibrozil, 600 mg twice daily; or placebo. Block randomization using random number tables was performed by a clinical trials pharmacist, thereby ensuring allocation concealment. Randomization also was substratified for the status of the patient (predialy-
778 sis, hemodialysis, or peritoneal dialysis), as well as the presence or absence of diabetes mellitus. All investigators, patients, renal staff, and the vascular function technician were blinded to treatment allocation. Clinical, laboratory, and vascular function tests described next were performed at baseline and after 6 weeks of study medication. Safety of therapy was assessed by means of liver and muscle enzyme level determinations (alanine aminotransferase and creatine kinase). Pill counting was used to assess compliance.
Clinical and Laboratory Methods All subjects gave a detailed medical history and underwent medical examination and electrocardiography. Clinical, laboratory, and vascular function tests on hemodialysis patients were always performed on the day after a dialysis session. Blood pressure and heart rate were measured using a Dinamap 1846 SX/P monitor (Critikon Ltd, Tampa, FL) after resting subjects for 10 minutes in the supine position. Weight, height, and waist and hip circumference were measured, and body mass index (weight/height2 [kilograms per square meter]) and waist-hip ratio were calculated. Venous blood samples were obtained after a 12-hour fast, with minimal venous stasis and with subjects in the supine position. Variables were measured by using standard laboratory techniques unless stated otherwise: High-sensitivity C-reactive protein was assayed using a high-sensitivity immunonephelometric method (Dade Behring Marburg GmbH, Marburg, Germany). High-sensitivity interleukin 6 was measured using a high-sensitivity quantitative enzyme immunoassay (Quantikine HS; R&D Systems Inc, Minneapolis, MN). Oxidized LDL was measured using an immunosorbent assay kit (Mercodia AB, Uppsala, Sweden). Insulin sensitivity was calculated using the Homeostasis Model Assessment (HOMA) score (HOMA score ⫽ fasting insulin [U/mL] ⫻ serum glucose [mmol/L]/22.5),31 in which a higher score signifies increasing insulin resistance. Calcium-phosphate product was derived from serum calcium ⫻ phosphate. The interassay coefficient of variation of all analytical assays was less than 7%, except for high-sensitivity interleukin 6, which was 9.9% to 16.5%.
Vascular Function Tests Brachial artery ultrasonography (Acuson Aspen 128 ultrasound device; Acuson Corp, Mountainview, CA) was used to measure endothelium-dependent postischemic FMD and endothelium-independent glyceryl trinitrate–mediated dilatation (GTNMD) using methods described elsewhere.32 Maximal FMD and GTNMD responses were calculated as percentage of change in brachial artery diameter from baseline. Systemic arterial compliance was measured using the CR2000 (Hypertension Diagnostics Inc/Pulse Wave CR2000; Research Cardiovascular Profiling System, Minneapolis, MN). Radial artery pulse contour analysis is used to derive C1 and C2 using a modified version of the Windkessel model, as described elsewhere.22
Statistical Methods and Analysis The desired effect size of a 2% change in FMD, as well as a 2-mL/mm Hg change in C2, was estimated to allow for a mean difference in FMD in repeated within-subject measure-
Dogra et al ments of 1.6% in our hands and take into consideration studies showing that a 1% decrease in FMD is associated with an odds ratio of a cardiovascular event of 1.35,19 and a 2-unit decrease in C2 is associated with an odds ratio of a cardiovascular event of 1.5.23 Comparing treatment arm with placebo (atorvastatin versus placebo or gemfibrozil versus placebo) with n ⫽ 35 per group (total n ⫽ 105), the study would have had greater than 80% power to detect a 2% ⫾ 2.5% (SD) FMD change and 2 ⫾ 2.6-mL/mm Hg C2 change with an ␣ error of 0.05. Analyses were performed using SPSS, version 11.5 (Statistical Package for Social Sciences; SPSS Inc, Chicago, IL). Results are expressed as mean ⫾ SD except for skewed variables, which are presented as geometric mean and 95% confidence interval and were log transformed before statistical analysis. Comparisons were performed between the atorvastatin and placebo groups and the gemfibrozil and placebo groups by using independent t-tests at baseline and general linear modeling adjusting for the corresponding baseline variable for results after 6 weeks of therapy. Posttreatment results also are represented as an estimate of the mean difference and 95% confidence interval. Categorical variables were compared using chi-square and Fisher exact tests. Statistical significance is defined at the 5% level for primary outcome variables (FMD and C2), and using Bonferroni correction for multiple testing, statistical significance is defined at the 0.4% level for secondary outcome variables.
RESULTS
Figure 1 shows in detail the flow of patients from recruitment through randomization, allocation, follow-up, and analysis. Of 119 eligible patients randomly assigned to atorvastatin, gemfibrozil, or placebo, 105 patients completed the baseline study visit and started study medications (Fig 1). Baseline characteristics of these 105 patients as a total group were described in detail elsewhere.33 As shown in Fig 1, after the start of trial medication, an additional 15 patients withdrew because of adverse events, leaving 90 patients who completed the study and were available for analysis (31 patients, atorvastatin; 27 patients, gemfibrozil; and 32 patients, placebo). Baseline characteristics of these patients within their individual allocated treatment groups are listed in Table 1. As listed in Table 2, serum creatinine, hemoglobin, serum albumin, calcium, phosphate, and calcium-phosphate product values were not significantly different among the 3 groups. Estimated glomerular filtration rate was not significantly different among the predialysis subgroups (Table 2). As listed in Table 3, lipid, lipoprotein, and oxidized LDL values were not significantly different between the gemfibrozil and placebo
Lipid-Lowering in Chronic Kidney Disease
779
Figure 1. Recruitment and patient randomization.
groups at baseline. However, total cholesterol and LDL cholesterol levels were significantly greater in the atorvastatin group compared with the placebo group at baseline. Values for markers of insulin resistance (glucose, insulin, and HOMA score), as well as markers of inflammation (Creactive protein and interleukin 6) also were not significantly different among the 3 groups at baseline (Table 3). Finally, all measures of vascular function (FMD, GTNMD, C1, and C2) were not significantly different between the gemfibrozil and placebo groups or between the atorvastatin and placebo groups at baseline (Table 4). Tables 5 and 6 list effects of atorvastatin and gemfibrozil therapy compared with placebo on
lipid, lipoprotein, and oxidized LDL levels, as well as insulin resistance and inflammation. Compared with placebo, atorvastatin therapy was associated with significant decreases in total cholesterol, LDL cholesterol, triglyceride, and oxidized LDL levels, but no change in HDL cholesterol levels. Gemfibrozil therapy was associated with a significant increase in HDL cholesterol levels and decrease in triglyceride levels, but no significant changes in total cholesterol, LDL cholesterol, and oxidized LDL levels. There was no significant effect of atorvastatin or gemfibrozil on markers of insulin resistance and inflammation. Tables 7 and 8 list effects of lipid-modifying therapy compared with placebo on brachial ar-
780
Dogra et al Table 1. Baseline Patient Characteristics in the Placebo, Atorvastatin, and Gemfibrozil Groups Placebo (n ⫽ 32)
Age (y) Predialysis Hemodialysis Peritoneal dialysis Renal diagnosis Chronic glomerulonephritis Diabetic nephropathy Unknown Other Diabetes mellitus Men Smokers Whites Previous cardiovascular disease Use of antihypertensives Use of erythropoietin n Body mass index (kg/m2) Waist-hip ratio Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Pulse pressure (mm Hg)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
62 ⫾ 11 18 9 5
65 ⫾ 14 18 9 4
58 ⫾ 13 17 8 2
11 4 4 13 7 20 5 27 7 24 17 27.9 ⫾ 6.4 0.94 ⫾ 0.12 134.9 ⫾ 24.2 72.4 ⫾ 11.9 62.5 ⫾ 17.4
11 5 4 11 6 19 6 25 6 21 16 26.5 ⫾ 5.2 0.92 ⫾ 0.09 142.6 ⫾ 23.5 77.4 ⫾ 11.6 65.2 ⫾ 19.1
6 3 8 10 6 17 7 18 5 17 10 26.2 ⫾ 4.7 0.91 ⫾ 0.08 134.1 ⫾ 25.9 74.9 ⫾ 12.2 59.3 ⫾ 21.0
Note: Results expressed as mean ⫾ SD or number of patients. P is nonsignificant (⬎0.05) for all comparisons between the placebo and atorvastatin groups and the placebo and gemfibrozil groups.
tery FMD and GTNMD. Compared with placebo, neither atorvastatin nor gemfibrozil therapy was associated with a significant improvement in FMD or GTNMD. As also listed in Tables 7 and 8, compared with placebo, lipid-modifying therapy (atorvastatin or gemfibrozil) was not associated with improvement in C1. As listed in Table 7, there was a small, but significant, improvement in C2 with atorvastatin (⫹1.1 mL/mm Hg ⫻ 100) compared with placebo (P ⫽ 0.024), but not with gemfibrozil compared with placebo, as listed in Table 8.
Compliance with gemfibrozil, but not atorvastatin, was lower than with placebo (placebo, 96.5%; atorvastatin, 91.2%; gemfibrozil, 89.1%). There was a greater incidence of self-reported gastrointestinal side effects (nausea, bloating, and diarrhea) with gemfibrozil, but not with atorvastatin, compared with placebo (placebo, 0 of 32; atorvastatin, 1 of 31; gemfibrozil, 11 of 27). There also was weight loss with gemfibrozil (median, ⫺1.1 kg; range, ⫺3.8 to ⫹0.8 kg) compared with placebo (median, ⫹0.4 kg; range, ⫺1.6 to ⫹2.7 kg). There were no significant
Table 2. Baseline Biochemical Variables in the Placebo, Atorvastatin, and Gemfibrozil Groups
Estimated glomerular filtration rate for predialysis subgroup (mL/min) Hemoglobin (g/dL) Serum albumin (g/dL) Serum calcium (mg/dL) Serum phosphate (mg/dL) Calcium-phosphate product
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
31 (33) ⫾ 13 (n ⫽ 18) 12.2 ⫾ 1.4 4.2 ⫾ 0.3 9.7 (9.5-10.0) 4.2 (3.7-4.8) 40.7 (35.1-48.0)
28 (25) ⫾ 12 (n ⫽ 18) 12.6 ⫾ 1.4 4.1 ⫾ 2.9 9.5 (9.3-9.7) 4.6 (4.1-5.2) 43.7 (38.1-50.4)
34 (29) ⫾ 16 (n ⫽ 17) 12.5 ⫾ 1.5 4.2 ⫾ 0.4 9.5 (9.3-9.8) 3.9 (3.5-4.4) 37.1 (32.6-43.1)
Note: Results expressed as mean ⫾ SD, mean (median) ⫾ SD, and geometric mean (95% confidence interval). P is nonsignificant (⬎0.05) for all comparisons between the placebo and atorvastatin groups and the placebo and gemfibrozil groups. To convert estimated glomerular filtration rate in mL/min to mL/s, multiply by 0.01667; hemoglobin in g/dL to g/L, multiply by 10; albumin in g/dL, multiply by 10; calcium in mg/dL to mmol/L, multiply by 0.2495; phosphate in mg/dL to mmol/L, multiply by 0.3229.
Lipid-Lowering in Chronic Kidney Disease
781
Table 3. Baseline Values for Lipids, Lipoproteins, Insulin Resistance, and Markers of Inflammation in the Placebo, Atorvastatin, and Gemfibrozil Groups
Cholesterol (mg/dL)* LDL (mg/dL)* Triglycerides (mg/dL)† HDL (mg/dL)† Oxidized LDL (U/L)† Serum glucose (mg/dL)† Serum insulin (U/mL)† HOMA score† C-Reactive protein (mg/L)† Interleukin 6 (ng/L)†
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
197 ⫾ 49 112 ⫾ 43 148 (124-186) 46.0 (41.8-51.0) 77.8 ⫾ 26.7 92 (83-104) 19.6 (15.5-24.6) 4.5 (3.4-5.9) 5.3 (3.2-8.7) 4.3 (3.3-5.7)
228 ⫾ 49 139 ⫾ 44 177 (142-213) 47.5 (42.2-53.4) 89.1 ⫾ 31.0 86 (77-99) 14.3 (10.7-19.1) 3.1 (2.2-4.3) 3.2 (2.2-4.7) 3.1 (2.6-3.8)
217 ⫾ 35 131 ⫾ 34 153 (124-195) 49.9 (44.1-56.1) 80.1 ⫾ 20.5 86 (79-95) 14.7 (10.8-19.8) 3.2 (2.3-4.4) 3.1 (1.9-5.0) 3.0 (2.4-3.8)
Note: Results expressed as mean ⫾ SD and geometric mean (95% confidence interval). To convert cholesterol, HDL cholesterol and LDL cholesterol in mg/dL to mmol/L, multiply by 0.02586; triglycerides in mg/dL to mmol/L, multiply by 0.01129; glucose in mg/dL to mmol/L, multiply by 0.05551; insulin in U/mL to pmol/L, multiply by 7.175. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, Homeostasis Model Assessment. *P ⬍ 0.05 for atorvastatin group compared with placebo group, but not significant for gemfibrozil group compared with placebo group. †P is nonsignificant (P ⬎ 0.05) for comparisons between the placebo and atorvastatin groups and placebo and gemfibrozil groups.
episodes of myalgia, myositis, or creatine kinase or alanine aminotransferase level increase with atorvastatin or gemfibrozil, and within the predialysis group, there was no significant increase in serum creatinine level. DISCUSSION
This study shows that lipid-lowering therapy during 6 weeks has no effect on endothelial function and a minimal effect on arterial compliance in patients with advanced stages 3 to 5 CKD. Although atorvastatin therapy was associated with a significant decrease in total cholesterol, LDL cholesterol, triglyceride, and oxidized LDL levels, there was no improvement in FMD and only a small improvement in C2. Gemfibrozil therapy was associated with a decrease in
triglyceride levels and increase in HDL cholesterol levels, but had no significant effects on FMD or C2 and was poorly tolerated by patients. This suggests that correction of dyslipidemia alone, by either targeting intensive LDL cholesterol lowering or correcting high triglyceride and low HDL cholesterol levels, is unlikely to significantly improve arterial function in the short term. A similar lack of improvement in endothelial function and other surrogate CVD risk markers was shown previously by studies incorporating lipid-lowering with a statin as part of a multiple risk factor–lowering approach in patients with advanced CKD.7,34 The improvement in arterial compliance with atorvastatin in our study, although of a small degree, is consistent with a previous study showing improvement in arterial
Table 4. Baseline Brachial Artery Vascular Function and Systemic Arterial Compliance in the Placebo, Atorvastatin and Gemfibrozil Groups
Resting brachial artery diameter (mm) FMD (%) GTNMD (%) C1 (mL/mm Hg ⫻ 10) C2 (mL/mm Hg ⫻ 100)
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Gemfibrozil (n ⫽ 27)
3.7 ⫾ 0.7 3.5 ⫾ 2.3 16.0 ⫾ 9.2 12.3 (10.7-14.1) 3.8 (3.1-4.7)
3.5 ⫾ 0.6 4.2 ⫾ 4.6 16.1 ⫾ 6.3 12.4 (10.9-14.2) 3.9 (3.1-4.8)
3.7 ⫾ 0.6 4.1 ⫾ 3.5 15.4 ⫾ 8.5 13.2 (11.1-15.7) 4.2 (3.3-5.3)
Note: Results expressed as mean ⫾ SD or geometric mean (95% confidence interval). P is nonsignificant (⬎0.05) for all comparisons between the placebo and atorvastatin groups and placebo and gemfibrozil groups. Abbreviations: FMD, flow-mediated dilatation; GTNMD, glyceryl trinitrate–mediated dilatation; C1, large-artery compliance; C2, small-artery compliance.
782
Dogra et al Table 5. Lipids, Lipoproteins, Insulin Resistance, and Markers of Inflammation After Treatment With Placebo and Atorvastatin
Cholesterol (mg/dL) LDL (mg/dL) Triglycerides* (mg/dL) HDL* (mg/dL) Oxidized LDL (U/L) Serum glucose (mg/dL) Serum insulin (U/mL) HOMA score C-Reactive protein (mg/L) Interleukin 6 (ng/L)
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Mean Difference (95% confidence interval)
P
194 ⫾ 50 112 ⫾ 41 151 (126-182)* 46.1 (41.7-51.0)* 76.1 ⫾ 25.1 87 (79-96)* 18.3 (14.0-24.0)* 3.9 (3.0-5.2)* 4.7 (3.0-7.6)* 4.2 (3.2-5.5)*
139 ⫾ 24 64 ⫾ 16 124 (106-144)* 45.8 (40.8-51.5)* 50.1 ⫾ 17.6 88 (77-100)* 15.1 (11.6-19.8)* 3.3 (2.4-4.5)* 5.2 (2.9-8.7)* 4.3 (3.4-5.5)*
⫺75.9 (⫺89.1 to ⫺62.7) ⫺62.9 (⫺73.6 to ⫺52.2) ⫺50.1 (⫺77.8 to ⫺22.5)† ⫺1.7 (⫺4.9 to ⫹1.5)† ⫺31.7 (⫺40.1 to ⫺23.3) ⫹9.0 (⫺0.9 to ⫹19.0)† ⫹0.3 (⫺2.1 to ⫹2.9)† ⫹0.9 (⫺0.4 to ⫹1.7)† ⫹12.2 (⫺6.5 to ⫹30.9)† ⫹0.4 (⫺1.7 to ⫹2.5)†
⬍0.0001 ⬍0.0001 ⬍0.0001 0.298 ⬍0.0001 0.084 0.171 0.064 0.209 0.277
Note: Results in atorvastatin and placebo columns expressed as mean ⫾ SD unless noted otherwise. To convert cholesterol, HDL cholesterol and LDL cholesterol in mg/dL to mmol/L, multiply by 0.02586; triglycerides in mg/dL to mmol/L, multiply by 0.01129; glucose in mg/dL to mmol/L, multiply by 0.05551; insulin in U/mL to pmol/L, multiply by 7.175. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, Homeostasis Model Assessment. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
stiffness with fluvastatin in dialysis patients.35 This possible divergent benefit of statin therapy may be a consequence of differential effects on resistance versus conduit arteries and on endothelial vasodilators other than nitric oxide. In addition, the small benefit of atorvastatin on arterial compliance may be related to the significantly greater LDL cholesterol lowering (50%) achieved in our study than previous studies,7 as well as decrease in oxidized LDL levels. This minimal effect of atorvastatin on arterial compliance may be consistent with the significant decrease in some secondary cardiac end points, but not in the
primary end points, in Assessment of Lescol in Renal Transplantation and the recent German Diabetes and Dialysis Study.6,36 The clinical significance of increased degrees of LDL cholesterol lowering will be tested further in the Study of Heart and Renal Protection, which combines ezetimibe with simvastatin, the results of which are still pending.37 Our study also shows that gemfibrozil is poorly tolerated in patients with advanced CKD, and although it corrects uremic dyslipidemia, it has no effect on endothelial function or arterial compliance. To our knowledge, this is the first study
Table 6. Lipids, Lipoproteins, Insulin Resistance, and Markers of Inflammation After Treatment With Placebo and Gemfibrozil
Cholesterol (mg/dL) LDL (mg/dL) Triglycerides (mg/dL) HDL (mg/dL) Oxidized LDL (U/L) Serum glucose (mg/dL) Serum insulin (U/mL) HOMA score C-Reactive protein (mg/L) Interleukin 6 (ng/L)
Placebo (n ⫽ 32)
Gemfibrozil (n ⫽ 27)
Mean Difference (95% confidence interval)
P
194 ⫾ 50 112 ⫾ 41 151 (126-182)* 46.1 (41.7-51.0)* 76.1 ⫾ 25.1 87 (79-96)* 18.3 (14.0-24.0)* 3.9 (3.0-5.2)* 4.7 (3.0-7.6)* 4.2 (3.2-5.5)*
208 ⫾ 41 114 ⫾ 39 96 (74-124)* 58.7 (52.1-66.3)* 73.9 ⫾ 21.5 89 (76-103)* 13.9 (10.4-18.5)* 3.0 (2.2-4.2)* 3.5 (2.1-5.9)* 4.0 (3.1-5.2)*
⫺5.3 (⫺16.1 to ⫹5.4) ⫺3.0 (⫺13.3 to ⫹7.4) ⫺56.0 (⫺84.9 to ⫺27.1) ⫹9.6 (⫹5.4 to ⫹13.8) ⫺4.1 (⫺11.7 to ⫹3.6) ⫹13.8 (⫺10.0 to ⫹37.7)† ⫺1.1 (⫺3.7 to ⫹1.4)† ⫹0.2 (⫺0.4 to ⫹0.9)† ⫺0.4 (⫺5.1 to ⫹4.2)† ⫹0.2 (⫺1.8 to ⫹2.2)†
0.328 0.564 ⬍0.0001 ⬍0.0001 0.292 0.269 0.972 0.483 0.671 0.231
Note: Results in the gemfibrozil and placebo columns expressed as mean ⫾ SD unless noted otherwise. To convert cholesterol, HDL cholesterol and LDL cholesterol in mg/dL to mmol/L, multiply by 0.02586; triglycerides in mg/dL to mmol/L, multiply by 0.01129; glucose in mg/dL to mmol/L, multiply by 0.05551; insulin in U/mL to pmol/L, multiply by 7.175. Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; HOMA, Homeostasis Model Assessment. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
Lipid-Lowering in Chronic Kidney Disease
783
Table 7. Brachial Artery Vascular Function and Systemic Arterial Compliance After Treatment With Placebo and Atorvastatin
FMD (%) GTNMD (%) C1 (mL/mm Hg ⫻ 10) C2 (mL/mm Hg ⫻ 100)
Placebo (n ⫽ 32)
Atorvastatin (n ⫽ 31)
Mean Difference (95% confidence interval)
P
4.0 ⫾ 2.6 15.3 ⫾ 7.6 12.5 (10.9-14.5)* 3.9 (3.2-4.7)*
3.5 ⫾ 4.1 16.7 ⫾ 7.9 13.3 (11.6-15.3)* 4.8 (3.9-5.9)*
⫺1.0 (⫺2.1 to ⫹0.2) ⫹1.5 (⫺1.1 to ⫹4.1) ⫹0.7 (⫺0.9 to ⫹2.2)† ⫹1.1 (⫹0.1 to ⫹2.2)†
0.099 0.254 0.408 0.024
Note: Results in atorvastatin and placebo columns expressed as mean ⫾ SD unless noted otherwise. Abbreviations: FMD, flow-mediated dilatation; GTNMD, glyceryl trinitrate–mediated dilatation; C1, large-artery compliance; C2, small-artery compliance. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
to examine the effects of fibrate therapy on surrogate CVD markers. However, although there are no clinical trials of fibrates and CVD in patients with advanced CKD, post hoc analysis of the Veteran’s Affairs-High Density Lipoprotein Cholesterol Intervention Trial showed a significant benefit of gemfibrozil in patients with mild CKD, with minimal side effects.5 Patients in our study had moderate to severe CKD and cannot be compared with the Veteran’s AffairsHigh Density Lipoprotein Cholesterol Intervention Trial patients. A notable finding of our study is the high incidence of gastrointestinal side effects from gemfibrozil therapy, resulting in poor compliance with pills. Although this raises the possibility that gemfibrozil is less likely to have a role in clinical trials in this advanced CKD population at the dose of 600 mg twice daily, it is possible that a lower dose of 600 mg/d may have equal lipid-lowering efficacy with reduced side effects. Although our findings suggest that dyslipidemia is not a primary contributor to arterial dysfunction in patients with CKD, with little benefit to be obtained from statin and fibrate therapy,
recent observational studies suggested significant potential cardiovascular benefit from statin therapy.38,39 The negative findings in our studies and others could be explained by the presence of advanced and irreversible vascular disease driven primarily by calcific arteriopathy, rather than atherosclerosis, contributing to the extremely high rate of cardiac events that characterizes patients with advanced CKD. Thus, starting statin therapy earlier in the course of CKD may be beneficial, as suggested by the Assessment of Lescol in Renal Transplantation study40 and post hoc analyses showing that statin and fibrate use in patients with early renal disease is of considerable cardiovascular benefit.4,5,41 Furthermore, in patients with advanced CKD, other nonconventional CVD risk factors, such as calcium-phosphate product, inflammation, oxidative stress, and insulin resistance, are likely to have a significant contributory role33,42,43 and warrant separate examination. In our study, despite the known pleiotropic effects of statins and fibrates, levels of oxidized LDL, but none of the inflammatory markers and markers of insulin resistance, were decreased by statin therapy, and neither was
Table 8. Brachial Artery Vascular Function and Systemic Arterial Compliance After Treatment With Placebo and Gemfibrozil
FMD (%) GTNMD (%) C1 (mL/mm Hg ⫻ 10) C2 (mL/mm Hg ⫻ 100)
Placebo (n ⫽ 32)
Gemfibrozil (n ⫽ 27)
Mean Difference (95% confidence interval)
P
4.0 ⫾ 2.6 15.3 ⫾ 7.6 12.5 (10.9-14.5)* 3.9 (3.2-4.7)*
3.7 ⫾ 3.2 16.1 ⫾ 8.0 13.6 (11.5-16.1)* 4.2 (3.4-5.3)*
⫺0.7 (⫺1.8 to ⫹0.4) 1.2 (⫺1.1 to ⫹3.4) 0.2 (⫺1.5 to ⫹1.9)† 0.1 (⫺0.8 to ⫹1.0)†
0.214 0.299 0.780 0.854
Note: Results in the gemfibrozil and placebo columns expressed as mean ⫾ SD unless noted otherwise. Abbreviations: FMD, flow-mediated dilatation; GTNMD, glyceryl trinitrate–mediated dilatation; C1, large-artery compliance; C2, small-artery compliance. *Geometric mean (95% confidence interval) and variables log transformed before analysis. †Mean difference (95% confidence interval) calculated before log transformation for skewed variables.
784
Dogra et al
decreased by fibrate therapy, possibly explaining a negative effect on arterial function. Additionally, the negative findings of this study could be attributed to a number of limitations, including the small sample size leading to inadequate power, short duration of therapy, and use of a heterogeneous group of patients with CKD and dialysis patients. A longer duration of therapy warrants examination because the improvement in C2 on atorvastatin therapy may be more significant and extend to endothelial function as well, or, conversely, the improvements may not be sustained. An additional limitation to consider is that this and other studies did not measure specific apolipoprotein-defined lipoprotein subclasses that might better explain the lack of efficacy of these drugs.44 In conclusion, despite improvement in dyslipidemia in patients with stages 3 to 5 CKD, short-term atorvastatin and gemfibrozil therapy failed to improve endothelial function, and atorvastatin, but not gemfibrozil, was associated with a small improvement in C2. ACKNOWLEDGMENT The authors acknowledge financial support from Pfizer Cardiovascular Lipid (CVL) grants, Faculty of Medicine and Dentistry at the University of Western Australia, and the Royal Perth Hospital Medical Research Foundation and thank the Perth Renal Units, which referred patients for the study, and Lisa Rich, for technical assistance with ultrasonography.
REFERENCES 1. Baigent C, Burbury K, Wheeler D: Premature cardiovascular disease in chronic renal failure. Lancet 356:147152, 2000 2. Lowrie EG, Lew NL: Death risk in hemodialysis patients: The predictive value of commonly measured variables and an evaluation of death rate differences between facilities. Am J Kidney Dis 15:458-482, 1990 3. Liu Y, Coresh J, Eustace JA, et al: Association between cholesterol level and mortality in dialysis patients: Role of inflammation and malnutrition. JAMA 291:451-459, 2004 4. Tonelli M, Isles C, Curhan GC, et al: Effect of pravastatin on cardiovascular events in people with chronic kidney disease. Circulation 110:1557-1563, 2004 5. Tonelli M, Collins D, Robins S, et al: Effect of gemfibrozil on change in renal function in men with moderate chronic renal insufficiency and coronary disease. Am J Kidney Dis 44:832-839, 2004 6. Wanner C, Krane V, Marz W, et al: Atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med 353:238-248, 2005 7. Isbel NM, Haluska B, Johnson DW, et al: Increased targeting of cardiovascular risk factors in patients with
chronic kidney disease does not improve atheroma burden or cardiovascular function. Am Heart J 151:745-753, 2006 8. Rakhit DJ, Marwick TH, Armstrong KA, et al: Effect of aggressive risk factor modification on cardiac events and myocardial ischaemia in patients with chronic kidney disease. Heart 92:1402-1408, 2006 9. Blacher J, Guerin AP, Pannier B, et al: Arterial calcifications, arterial stiffness, and cardiovascular risk in endstage renal disease. Hypertension 38:938-942, 2001 10. Yeun JY, Kaysen GA: C-Reactive protein, oxidative stress, homocysteine, and troponin as inflammatory and metabolic predictors of atherosclerosis in ESRD. Curr Opin Nephrol Hypertens 9:621-630, 2000 11. Irish AB, Green FR: Factor VII coagulant activity (VIIc) and hypercoagulability in chronic renal disease and dialysis: Relationship with dyslipidaemia, inflammation, and factor VII genotype. Nephrol Dial Transplant 13:679-684, 1998 12. 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 332:481-487, 1995 13. Dogra GK, Watts GF, Chan DC, et al: Statin therapy improves brachial artery vasodilator function in patients with type 1 diabetes and microalbuminuria. Diabet Med 22:239-242, 2005 14. Vaughan CJ, Gotto AM Jr, Basson CT: The evolving role of statins in the management of atherosclerosis. J Am Coll Cardiol 35:1-10, 2000 15. Irish AB, Thompson CH: The effects of gemfibrozil upon the hypercoagulable state in dyslipidaemic patients with chronic renal failure. Nephrol Dial Transplant 11:22232228, 1996 16. Watts GF, Dimmitt SB: Fibrates, dyslipoproteinaemia and cardiovascular disease. Curr Opin Lipidol 10:561-574, 1999 17. Rubanyi GM: The role of endothelium in cardiovascular homeostasis and diseases. J Cardiovasc Pharmacol 22:S1S14, 1993 (suppl 4) 18. Suwaidi JA, Hamasaki S, Higano ST, et al: Longterm follow-up of patients with mild coronary artery disease and endothelial dysfunction. Circulation 101:948-954, 2000 19. 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 86:207-210, 2000 20. Schachinger V, Britten MB, Zeiher AM: Prognostic impact of coronary vasodilator dysfunction on adverse longterm outcome of coronary heart disease. Circulation 101: 1899-1906, 2000 21. Celermajer DS, Sorensen KE, Gooch VM, et al: Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis. Lancet 340:11111115, 1992 22. Woodman RJ, Watts GF: Measurement and application of arterial stiffness in clinical research: Focus on new methodologies and diabetes mellitus. Med Sci Monit 9:RA81RA89, 2003 23. Grey E, Bratteli C, Glasser SP, et al: Reduced small artery but not large artery elasticity is an independent risk
Lipid-Lowering in Chronic Kidney Disease marker for cardiovascular events. Am J Hypertens 16:265-269, 2003 24. London GM, Guerin AP, Marchais SJ, et al: Cardiac and arterial interactions in end-stage renal disease. Kidney Int 50:600-608, 1996 25. London GM: Arterial function in renal failure. Nephrol Dial Transplant 13:S12-S15, 1998 (suppl 4) 26. Blacher J, Safar ME, Guerin AP, et al: Aortic pulse wave velocity index and mortality in end-stage renal disease. Kidney Int 63:1852-1860, 2003 27. London GM, Blacher J, Pannier B, et al: Arterial wave reflections and survival in end-stage renal failure. Hypertension 38:434-438, 2001 28. Tsunekawa T, Hayashi T, Kano H, et al: Cerivastatin, a hydroxymethylglutaryl coenzyme A reductase inhibitor, improves endothelial function in elderly diabetic patients within 3 days. Circulation 104:376-379, 2001 29. McVeigh GE, Brennan GM, Cohn JN, et al: Fish oil improves arterial compliance in non-insulin-dependent diabetes mellitus. Arterioscler Thromb 14:1425-1429, 1994 30. Nestel PJ, Pomeroy SE, Sasahara T, et al: Arterial compliance in obese subjects is improved with dietary plant n-3 fatty acid from flaxseed oil despite increased LDL oxidizability. Arterioscler Thromb Vasc Biol 17:1163-1170, 1997 31. Matthews DR, Hosker JP, Rudenski AS, et al: Homeostasis Model Assessment: Insulin resistance and betacell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28:412-419, 1985 32. Woodman RJ, Playford DA, Watts GF, et al: Improved analysis of brachial artery ultrasound using a novel edge-detection software system. J Appl Physiol 91:929-937, 2001 33. Dogra G, Irish A, Chan D, et al: Insulin resistance, inflammation, and blood pressure determine vascular dysfunction in CKD. Am J Kidney Dis 48:926-934, 2006 34. Fathi R, Isbel N, Short L, et al: The effect of longterm aggressive lipid lowering on ischemic and atheroscle-
785 rotic burden in patients with chronic kidney disease. Am J Kidney Dis 43:45-52, 2004 35. Ichihara A, Hayashi M, Ryuzaki M, et al: Fluvastatin prevents development of arterial stiffness in haemodialysis patients with type 2 diabetes mellitus. Nephrol Dial Transplant 17:1513-1517, 2002 36. Holdaas H, Fellstrom B, Jardine AG, et al: Effect of fluvastatin on cardiac outcomes in renal transplant recipients: A multicentre, randomised, placebo-controlled trial. Lancet 361:2024-2031, 2003 37. Baigent C, Landry M: Study of Heart and Renal Protection (SHARP). Kidney Int Suppl 84:S207-S210, 2003 38. Seliger SL, Weiss NS, Gillen DL, et al: HMG-CoA reductase inhibitors are associated with reduced mortality in ESRD patients. Kidney Int 61:297-304, 2002 39. Mason NA, Bailie GR, Satayathum S, et al: HMGcoenzyme A reductase inhibitor use is associated with mortality reduction in hemodialysis patients. Am J Kidney Dis 45:119-126, 2005 40. Jardine AG, Holdaas H, Fellstrom B, et al: Fluvastatin prevents cardiac death and myocardial infarction in renal transplant recipients: Post-hoc subgroup analyses of the ALERT Study. Am J Transplant 4:988-995, 2004 41. Heart Protection Study Collaborative Group: MRC/ BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: A randomised placebo-controlled trial. Lancet 360:7-22, 2002 42. Shinohara K, Shoji T, Emoto M, et al: Insulin resistance as an independent predictor of cardiovascular mortality in patients with end-stage renal disease. J Am Soc Nephrol 13:1894-1900, 2002 43. London GM, Guerin AP, Marchais SJ, et al: Arterial media calcification in end-stage renal disease: Impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant 18:1731-1740, 2003 44. Alaupovic P, Attman PO, Knight-Gibson C, et al: Effect of fluvastatin on apolipoprotein-defined lipoprotein subclasses in patients with chronic renal insufficiency. Kidney Int 69:1865-1871, 2006