Methods

This was a 12-week multicenter, randomized, double-blind, parallel-arm study conducted from February 2007 to October 2008 at 115 sites: 75 in the United States, 18 in Russia, 8 in the Ukraine, 5 in Canada, 5 in Romania, and 4 in Poland. The investigational review boards reviewed and approved the protocol and amendments, and all patients provided written informed consent before any study procedures ( ClinicalTrials.gov identifier NCT00418834 ). The patients included men and women ≥65 years old at high risk for coronary heart disease with or without atherosclerotic vascular disease (AVD) who had not reached a LDL cholesterol level of <70 or <100 mg/dl, respectively, with atorvastatin 10 mg/day. Patients were also potentially eligible if they were taking atorvastatin 10 or 20 mg, taking a stable daily dose of a statin of equal or lesser potency than atorvastatin 20 mg with good compliance (>80% of daily doses for 6 weeks before visit 1), or naive to lipid-lowering therapy. These patients were switched to atorvastatin 10 mg during the run-in phase of the study and were required to meet the risk and LDL cholesterol level eligibility criteria. All patients were instructed to consume a cholesterol-lowering diet throughout the study period.

The patients were eligible for enrollment if they met the following criteria: established coronary heart disease and other AVD and LDL cholesterol ≥70 but ≤160 mg/dl; no AVD but diabetes mellitus (type 1 or 2) or multiple risk factors and a 10-year risk of coronary heart disease of >20% (as determined by the Framingham calculation) and LDL cholesterol ≥100 but ≤190 mg/dl; triglycerides ≤350 mg/dl, alanine aminotransferase and aspartate aminotransferase ≤1.5 times the upper limit of normal with no active liver disease, creatine kinase ≤2 times the upper limit of normal, thyroid-stimulating hormone ≥0.3 or ≤5.0 μIU/ml, and hemoglobin A1c <8.5%. Patients were excluded from the study if they had uncontrolled hypertension (systolic blood pressure >160 mm Hg or diastolic blood pressure >100 mm Hg) or impaired renal function (creatinine ≥2.0 mg/day or a history of nephrotic range proteinuria), were taking lipid-lowering agents (except for simvastatin 10, 20, or 40 mg; atorvastatin 20 mg; pravastatin 10, 20, or 40 mg; fluvastatin 20, 40, or 80 mg; ezetimibe 10 mg; lovastatin 10, 20, or 40 mg; or rosuvastatin 5 mg) within 6 weeks or fibrates within 8 weeks of screening, or were taking prescription and/or over-the-counter drugs with potential drug interactions with statins within 6 weeks of the study start. Patients receiving maintenance therapy with psyllium or other over-the-counter lipid-lowering therapies for ≥6 weeks before study entry were allowed into the study if they agreed to maintain the same treatment regimen throughout the study period.

Computerized allocation schedules with block sizes of 4 were created for each of the following baseline LDL cholesterol strata and the presence or absence of AVD: (1) LDL cholesterol ≥70 but <100 mg/dl with AVD, (2) LDL cholesterol ≥100 but <130 mg/dl with AVD, (3) LDL cholesterol ≥100 but <130 mg/dl without AVD, (4) LDL cholesterol ≥130 but ≤160 mg/dl with AVD, (5) LDL cholesterol ≥130 but <160 mg/dl without AVD, and (6) LDL cholesterol ≥160 but ≤190 mg/dl without AVD. Patients were centrally randomized in a 1:1 ratio using an interactive voice response system. All study personnel, including the investigators, study site personnel, patients, monitors, and central laboratory personnel, were unaware of the treatment allocation throughout the study period. During the 4-week run-in period (5 weeks for switch and naive patients), all patients received single-blind atorvastatin 10 mg/day. After the run-in period, the patients were randomized to ezetimibe 10 mg plus atorvastatin 10 mg daily or atorvastatin 20 mg daily for 6 weeks. After the first 6 weeks, patients in the ezetimibe 10 mg plus atorvastatin 10 mg group continued the same treatment. The patients in the atorvastatin 20-mg group were titrated to atorvastatin 40 mg/day (atorvastatin 20/40 mg) for an additional 6 weeks, regardless of their LDL cholesterol level. All patients were supplied as follows: during the run-in phase, bottles of single-blind atorvastatin 10 mg, placebo to ezetimibe 10 mg, and placebo to atorvastatin 20 mg; at randomization, bottles of blinded ezetimibe 10 mg or placebo and blinded atorvastatin 10 mg or placebo and blinded atorvastatin 20 mg or placebo; at week 6, bottles of blinded ezetimibe 10 mg or placebo and blinded atorvastatin 10 mg or placebo and blinded atorvastatin 40 mg or placebo. Patients took 1 tablet/day from each bottle (labeled A, B, or C).

The primary efficacy end point was the percentage of change in LDL cholesterol level after 6 weeks of treatment. The secondary end points included the general safety and tolerability of atorvastatin 10 mg plus ezetimibe after 12 weeks of treatment. In addition, the following secondary end points were assessed after 6 weeks: (1) the percentage of patients achieving LDL cholesterol <70 mg/dl and (2) the percentage of patients achieving LDL cholesterol <100 mg/dl for high-risk patients without AVD and <70 mg/dl for high-risk patients with AVD. After 12 weeks, the percentage of change from baseline in LDL cholesterol levels, the percentage of patients achieving LDL cholesterol <70 mg/dl, and the percentage of patients achieving LDL cholesterol <100 mg/dl for high-risk patients without AVD and <70 mg/dl for high-risk patients with AVD were assessed. Additional efficacy end points included an evaluation of high-density lipoprotein (HDL) cholesterol, non-HDL cholesterol, total cholesterol, triglycerides, apo B, apo A-I, total cholesterol/HDL cholesterol ratio, apo B/apo A-I ratio, LDL cholesterol/HDL cholesterol ratio, non-HDL cholesterol/HDL cholesterol ratio, and high-sensitivity C-reactive protein after 6 and 12 weeks. The definition of metabolic syndrome was determined from the American Heart Association/National Heart, Lung, and Blood Institute scientific statement on the metabolic syndrome. At screening, vital signs (eg, heart rate, blood pressure, weight, height, and waist circumference) were measured. All lipid determinations used were obtained through central laboratories (PPD, Highland Heights, Kentucky; and PPD, Zaventem, Belgium). For patients with triglycerides ≤400 mg/dl, the LDL cholesterol measurements were calculated using the Friedewald equation. For patients with triglycerides >400 mg/dl, the LDL cholesterol measurements were obtained directly using the β-quantification method.

The focus of the safety and tolerability end points was the 12-week, double-blind treatment period. To determine the change from baseline in vital signs or laboratory measurements, the patients were required to have a baseline measurement and at least one “on-treatment” measurement to be included in these analyses. Adverse events (AEs) were summarized by system organ class and specific AE term. Prespecified safety parameters included consecutive elevations of alanine aminotransferase and/or aspartate aminotransferase ≥3 times the upper limit of normal, creatine kinase elevations ≥10 times the upper limit of normal, creatine kinase elevations ≥10 times the upper limit of normal with muscle symptoms (within ±7 days of the laboratory result), creatine kinase elevations ≥10 times the upper limit of normal with muscle symptoms (within ±7 days of the laboratory result) considered by the investigators to be related to the study drug, hepatitis-related AEs, gallbladder-related AEs, gastrointestinal-related AEs, allergic reaction or rash AEs, and potential Hy’s law condition.

The target enrollment was 450 evaluable patients per group. With this number of patients, assuming a within-group standard deviation of 17%, the study had >99% power to detect a 10% difference in the percentage of change from baseline in LDL cholesterol (α = 0.05, 2-sided) and 94% power to detect a 4% difference between treatment groups for the primary and secondary end points. For the primary efficacy end point, the between-group differences were analyzed using analysis of covariance with terms for treatment, baseline LDL cholesterol, and AVD status. The least squares mean for each treatment, between-treatment difference, and 95% confidence intervals were estimated from the analysis of covariance model. A logistic regression model was used to compare treatment groups with respect to patients achieving prespecified LDL cholesterol levels at weeks 6 and 12. The terms for treatment, baseline LDL cholesterol, and AVD status were included. The Wald chi-square statistic was used to test the differences between the treatment groups. Other end points (with the exception of triglycerides and high-sensitivity C-reactive protein) were evaluated using analysis of covariance with terms for treatment, baseline covariate, and AVD status. The analysis of triglycerides was done using a nonparametric method and analysis of covariance applied to Tukey’s normalized ranks. The between-treatment group difference in the median was estimated using the Hodges-Lehmann location shift, and the distribution-free 95% confidence interval for the between-treatment difference was determined using Wilcoxon’s rank sum test statistic. A longitudinal data analysis method proposed by Liang and Zeger was used for high-sensitivity C-reactive protein. The repeated measures model included terms for treatment, time, and interaction of time by treatment, and the data were transformed by the natural logarithm. The treatment difference in geometric mean percentage of change from baseline was calculated according to the difference in the back-transformed least squares mean.

All randomized patients who took at least one dose of study medication were included in the safety analyses. Prespecified AEs were subject to inferential testing for statistical significance with p values and 95% confidence intervals provided for between-group comparisons using the Miettinen and Nurminen method. For AEs occurring in ≥1% of patients within one of the treatment groups, the 95% confidence intervals for between-group differences were calculated. Summary statistics were tabulated by treatment group for AEs occurring in <1% of patients within at least one of the treatment groups and select laboratory tests.

Results

The flow of participants through the study is shown in Figure 1 , and the baseline demographics for each treatment group are summarized in Table 1 . The risk factors ( Table 1 ) and lipid values ( Table 2 ) were similar between the treatment groups.

Flow of patients through study. A = atorvastatin; EZ = ezetimibe 10 mg.

Table 2

Percentage of change from treated baseline at weeks 6 and 12 in lipid parameters, lipid ratios, and high-sensitivity C-reactive protein

Parameter	Week 6			Week 12
	A10 + E (n = 515)	A20 (n = 515)	Treatment Difference ⁎	A10 + E (n = 516)	A20/40 (n = 509)	Treatment Difference †
Low-density lipoprotein cholesterol (mg/dl)
Baseline	103 ± 28	101 ± 21		103 ± 28	102 ± 21
Least squares mean % change	−27	−13	−14	−23	−18	−5
95% confidence interval	−29 to −25	−15 to −11	−16 to −12	−25 to −20	−21 to −15	−7 to −2
p Value			<0.001			0.001
High-density lipoprotein cholesterol (mg/dl)
Baseline	55 ± 14	55 ± 13		55 ± 14	54 ± 12
Least squares mean % change	3	1	2	2	−1	3
95% confidence interval	1 to 4	−1 to 2	0.3 to 4	1 to 4	−2 to 1	2 to 5
p Value			0.021			<0.001
Total cholesterol (mg/dl) ‡
Baseline	183 ± 32	182 ± 26		183 ± 32	182 ± 26
Least squares mean % change	−16	−8	−8	−14	−12	−2
95% confidence interval	−17 to −15	−9 to −7	−9 to −7	−15 to −12	−13 to −10	−4 to −0.2
p Value			<0.001			0.029
Triglycerides (mg/dl) ‡ §
Baseline	113 ± 54	116 ± 62		113 ± 54	117 ± 62
Median % change	−13	−6	−6	−12	−9	−2
95% confidence interval	−14 to −10	−8 to −4	−9 to −3	−14 to −9	−11 to −7	−5 to 1
p Value			<0.001			0.150
Non–high-density lipoprotein cholesterol (mg/dl)
Baseline	128 ± 31	127 ± 25		128 ± 31	127 ± 25
Least squares mean % change	−24	−11	−12	−20	−16	−4
95% confidence interval	−25 to −22	−13 to −10	−14 to −10	−22 to −18	−18 to −14	−7 to −2
p Value			<0.001			0.001
Apo B (mg/dl) § ¶
Baseline	103 ± 23	102 ± 21		104 ± 23	102 ± 21
Least squares mean % change	−17	−8	−9	−14	−11	−3
95% confidence interval	−19 to −15	−9 to −6	−11 to −7	−16 to −12	−13 to −9	−6 to −1
p Value			<0.001			0.007
Apo A-I (mg/dl) § ¶
Baseline	165 ± 29	164 ± 27		165 ± 29	164 ± 27
Least squares mean % change	−1	−2	1	1	−2	3
95% confidence interval	−2 to 0.3	−3 to −0.3	−1 to 2	−1 to 2	−3 to −1	1 to 4
p Value			0.401			<0.001
Total cholesterol/high-density lipoprotein cholesterol ratio
Baseline	4 ± 1	4 ± 1		4 ± 1	4 ± 1
Least squares mean % change	−17	−8	−9	−14	−10	−5
95% confidence interval	−18 to −15	−9 to −6	−11 to −7	−16 to −13	−12 to −8	−7 to −2
p Value			<0.001			<0.001
Low-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio
Baseline	2 ± 1	2 ± 1		2 ± 1	2 ± 1
Least squares mean % change	−27	−13	−15	−23	−16	−7
95% confidence interval	−30 to −25	−15 to −10	−17 to −12	−26 to −20	−19 to −13	−10 to −4
p Value			<0.001			<0.001
Apo B/apo A-I ratio § ¶
Baseline	1 ± 0.2	1 ± 0.2		1 ± 0.2	1 ± 0.2
Least squares mean % change	−15	−5	−10	−14	−8	−5
95% confidence interval	−17 to −13	−7 to −3	−12 to −8	−16 to −11	−10 to −6	−8 to −3
p Value			<0.001			<0.001
Non–high-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio
Baseline	3 ± 1	3 ± 1		3 ± 1	3 ± 1
Least squares mean % change	−24	−11	−13	−20	−14	−7
95% confidence interval	−26 to −22	−13 to −9	−15 to −11	−23 to −18	−17 to −11	−10 to −4
p Value			<0.001			<0.001
High-sensitivity C-reactive protein ∥
Baseline #	2	2		2	2
Least squares mean % change	−14	−11	−3	−22	−15	−7
95% confidence interval	−20 to −7	−17 to −4	−12 to 6	−27 to −16	−21 to −8	−16 to 2
p Value			0.534			0.090

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Comparative Analysis of Cardiac Magnetic Resonance Viability Indexes to Predict Functional Recovery After Successful Percutaneous Coronary Intervention in Acute Myocardial Infarction
2. N-Terminal Pro–Brain Natriuretic Peptide and Outcomes in Patients Undergoing Surgical Ventricular Restoration
3. The RecordAF Study: Design, Baseline Data, and Profile of Patients According to Chosen Treatment Strategy for Atrial Fibrillation
4. Erratum for Shah et al. “Usefulness of Various Plasma Biomarkers for Diagnosis of Heart Failure in Children With Single Ventricle Physiology” Am J Cardiol 2009;104:1280–1284

Stay updated, free articles. Join our Telegram channel

Join