Statin therapy with ezetimibe or niacin in high-risk patients

T h e n e w e ng l a n d j o u r na l o f m e dic i n e Unfortunately, the premature termination of cin or ezetimibe and carotid intima–media thickness. N Engl J the ARBITER 6–HALTS trial, the small number Med 2009;361:2113-22.
2. Brown BG, Zhao X-Q, Chait A, et al. Simvastatin and niacin,
of patients studied, and the limited duration of antioxidant vitamins, or the combination for the prevention of follow-up preclude us from conclusively declaring coronary disease. N Engl J Med 2001;345:1583-92.
niacin the adjunctive agent of choice on the basis 3. Brown G, Albers JJ, Fisher LD, et al. Regression of coronary
artery disease as a result of intensive lipid-lowering therapy in of the evidence. A decrease of 0.014 mm in the men with high levels of apolipoprotein B. N Engl J Med carotid intima–media thickness over 14 months 1990;323:1289-98.
does not necessarily merit changes in our lipid- 4. Taylor AJ, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Arte-
rial Biology for the Investigation of the Treatment Effects of lowering guidelines at this time. However, in Reducing cholesterol (ARBITER) 2. Circulation 2004;110:3512-7. conjunction with another, smaller, recently pub- [Errata, Circulation 2004;110:3615, 2005;111(24):e446.] lished study involving magnetic resonance im- 5. Barter PJ, Caulfield M, Eriksson M, et al. Effects of torcetra-
pib in patients at high risk for coronary events. N Engl J Med aging of the carotid artery,14 for now, we would 2007;357:2109-22.
support the use of niacin as the preferred ad- 6. Sharma K, Blaha MJ, Blumenthal RS, Musunuru K. Clinical
junctive agent to be used in combination with and research applications of carotid intima-media thickness. the maximal dose of a potent statin in persons 7. Finn AV, Kolodgie FD, Virmani R. Correlation between ca-
who have low levels of HDL cholesterol and es- rotid intimal/medial thickness and atherosclerosis: a point of tablished cardiovascular disease. However, we view from pathology. Arterioscler Thromb Vasc Biol 2009 Au- await the completion of the definitive clinical tri- 8. Mackinnon AD, Jerrard-Dunne P, Sitzer M, Buehler A, von
Kegler S, Markus HS. Rates and determinants of site-specific In summary, the ARBITER 6–HALTS results progression of carotid artery intima-media thickness: the ca- rotid atherosclerosis progression study. Stroke 2004;35:2150-4.
are provocative and are an important contribu- 9. Kastelein JJ, Akdim F, Stroes ESG, et al. Simvastatin with or
tion to preventive cardiology research. However, without ezetimibe in familial hypercholesterolemia. N Engl J the secondary choices for optimizing cholesterol- Med 2008;358:1431-43. [Erratum, N Engl J Med 2008;358:1977.] 10. Crouse JR III, Raichlen JS, Riley WA, et al. Effect of rosuva-
lowering therapy, constituting part of the “C” statin on progression of carotid intima-media thickness in low- component of the “ABCDEs” of secondary pre- risk individuals with subclinical atherosclerosis: the METEOR vention of cardiovascular disease, should not Trial. JAMA 2007;297:1344-53.
11. Ridker PM, Danielson E, Fonesca FAH, et al. Rosuvastatin to
overshadow the importance of the rest of the prevent vascular events in men and women with elevated C-reac- ABCDEs: assessment of risk, antiplatelet therapy, tive protein. N Engl J Med 2008;359:2195-207.
blood-pressure management, cholesterol modifi- 12. Fleg JL, Mete M, Howard BV, et al. Effect of statins alone
versus statins plus ezetimibe on carotid atherosclerosis in type cation and cigarette-smoking cessation, dietary 2 diabetes. The SANDS (Stop Atherosclerosis in Native Diabetics and weight modification, and exercise habits.15 Study) trial. J Am Coll Cardiol 2008;52:2198-205.
13. Meaney A, Ceballos G, Asbun J, et al. The VYtorin on Ca-
No potential conflict of interest relevant to this article was re- rotid intima-media thickness and overall rigidity (VYCTOR) study. J Clin Pharmacol 2009;49:838-47.
14. Lee M, Robson MD, Yu LM, et al. Effects of high-dose mod-
From the Johns Hopkins Ciccarone Center for the Prevention of ified-release nicotinic acid on atherosclerosis and vascular func- tion: a randomized, placebo-controlled, magnetic resonance imaging study. J Am Coll Cardiol 2009;54:1787-94.
This article (10.1056/NEJMe0908838) was published on No- 15. Blaha MJ, Bansal S, Rouf R, Golden SH, Blumenthal RS,
Defilippis AP. A practical “ABCDE” approach to the metabolic syndrome. Mayo Clin Proc 2008;83:932-41.
1. Taylor AJ, Villines TC, Stanek EJ, et al. Extended-release nia-
Copyright 2009 Massachusetts Medical Society. Statin Therapy with Ezetimibe or Niacin in High-Risk Patients
John J.P. Kastelein, M.D., Ph.D., and Michiel L. Bots, M.D., Ph.D.
The use of statin therapy to reduce levels of low- risk. This residual risk has resulted in the search density lipoprotein (LDL) cholesterol in patients for additional therapeutic approaches. One ap- who are at high risk for vascular events results proach is the further reduction of the LDL cho- in a 30 to 40% reduction in the rate of clinical lesterol level through treatment with higher doses events.1 However, many patients continue to be at of statins or through the addition of the choles- n engl j med 361;22 nejm.org november 26, 2009 Downloaded from nejm.org by JEFFREY SHANES on February 24, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved. terol-absorption inhibitor ezetimibe. However, to obtained in a modest sample of 208 patients, date, only high-dose statin therapy has shown a followed for only 14 months, show a clear supe- clear clinical benefit beyond a moderate reduc- riority of niacin over ezetimibe. In fact, the ad- tion of the LDL cholesterol level.1 Strategies to dition of extended-release niacin to statin therapy lower LDL cholesterol further, through the addi- led to a decrease in the common carotid intima– tion of ezetimibe, are not supported by defini- media thickness, whereas the addition of ezetimibe tive evidence.2-4 Two studies of clinical end merely prevented an increase in this thickness. points involving ezetimibe are ongoing: one of The authors emphasize the additional post hoc patients with renal failure (SHARP [Study of finding that large reductions in the LDL choles- Heart and Renal Protection]; ClinicalTrials.gov terol level, induced by ezetimibe, were associated number, NCT00125593), and the other of patients with an increase in the carotid intima–media with the acute coronary syndrome (IMPROVE-IT thickness.
[Improved Reduction of Outcomes: Vytorin Ef- Given the importance of these findings, several ficacy International Trial]; NCT00202878). The aspects of the study should be put in perspective, large number of hard clinical end points (>5000) including the quality of the surrogate marker required to achieve sufficient statistical power in (carotid intima–media thickness), the statistical IMPROVE-IT makes it uncertain whether the trial approach, and the potential consequences of pre- A second approach to reducing residual risk in The approach to the assessment of the com- patients receiving statin therapy is to increase the mon carotid intima–media thickness was careful level of high-density lipoprotein (HDL) cholester- and well designed; Taylor and colleagues went to ol. Niacin is one of the most effective oral agents great lengths to reduce measurement error in currently available to increase the HDL choles- the assessment. This was achieved through the use of one ultrasonographer for nearly the entire In this issue of the Journal, Taylor et al. report study, the acquisition of images in a blinded fash- on the ARBITER 6–HALTS trial (Arterial Biology ion, the use of the baseline carotid ultrasono- for the Investigation of the Treatment Effects of graphic examinations to localize the site of re- Reducing Cholesterol 6–HDL and LDL Treatment measurement at the 8- and 14-month visits, the Strategies; NCT00397657).5 The trial was designed use of standardized settings on state-of-the-art to address the question of whether the addition ultrasonography equipment, the duplicate mea- of ezetimibe (to further reduce LDL cholesterol surement of the carotid intima–media thickness levels) or the addition of niacin (to improve levels in a blinded manner, and the evaluation of the of multiple lipoproteins) is more effective in de- reproducibility of measurements by the observer creasing the progression of carotid intima–media over time (called “reader drift”). The reproducibili- thickness in patients receiving statin therapy. In ty of measurements of the carotid intima–media a previous study (ARBITER 2) from the same au- thickness is among the best reported worldwide, thors conducted in a similar group of patients, with a very high intraobserver correlation coef- the addition of niacin to statin therapy resulted ficient and extremely small differences in both in the slowing of progression of the carotid the mean and individual measurements taken at intima–media thickness.6 The patients enrolled each time point.
in the ARBITER 6–HALTS study had a high risk of vascular events, and the mean duration of 6–HALTS trial restricted the authors to the analy- standard statin treatment before randomization sis of 208 participants who had finished their 14-month visit by the time of study termination. In ARBITER 6–HALTS, the authors used the Although at the time of termination, ultrasono- between-group difference in the change from graphic examinations were upcoming for partici- baseline in thickness of the common carotid ar- pants who had not yet completed the study, only tery after 14 months as the primary outcome. data for patients who had already undergone the The results of this important and provocative 14-month visit were included in the final analy- open-label trial are intriguing, even though the ses. This was a missed opportunity to enhance trial was prematurely terminated. The findings, precision and power, because the power of the n engl j med 361;22 nejm.org november 26, 2009 Downloaded from nejm.org by JEFFREY SHANES on February 24, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e mixed-effect linear regression models usually used The decision to stop a trial before completion in trials of carotid intima–media thickness lies generally involves a data and safety monitoring in the fact that all thickness measurements can board comprising experienced investigators with be included from all participants who have at clinical as well as statistical expertise. The board least one postrandomization measurement.7 That acts according to predefined criteria, since an approach is preferred over the “complete case” early observed benefit may be an overestimation analysis used in the ARBITER 6–HALTS trial, of the real between-group difference, owing to which may lead to bias; the imputation of data chance, that may not persist over time.10,11 The ARBITER 6–HALTS findings might have been However, it is reassuring that the baseline strengthened had the trial continued to com- characteristics and measurements of the carotid pletion.
intima–media thickness did not differ signifi- In summary, despite the several limitations of cantly between the participants who completed this investigator-initiated trial, the primary results the study and those who dropped out, although are likely to be correct although the magnitude of the sample size of the study, and thus the number the difference between the treatment arms may be who dropped out, was small and may preclude overestimated. Whether this result is due to the the assessment of statistical significance, owing effect of niacin on HDL cholesterol, LDL choles- terol, remnants, Lp(a) lipoprotein, high-sensitivity Taylor and colleagues emphasize the results C-reactive protein, or any combination of these of a post hoc analysis of the relationships between cannot be answered by the current data.
changes in the LDL cholesterol level and in the Together, the results available to date provide carotid intima–media thickness for each of the support for the concept that the use of statins to two treatment groups. In clinical trials, post hoc reduce LDL cholesterol to target levels with the analyses should be performed only after careful subsequent addition of a drug to raise HDL cho- consideration and appropriate statistical model- lesterol levels (niacin), rather than a drug to lower ing.9 In the ARBITER 6–HALTS report, the authors LDL cholesterol levels (ezetimibe), is a more effec- assessed the relationships by using single-variable tive treatment for patients at high cardiovascular correlation coefficients. Such an approach is not risk. Two well-powered studies of clinical end the most rigorous in the current, evidence-based points — AIM-HIGH (Atherothrombosis Interven- era. Post hoc analyses are very sensitive to bias tion in Metabolic Syndrome with Low HDL/High due to confounding. In this trial, multivariable Triglycerides and Impact on Global Health Out- models should have been used. At a minimum, comes; NCT00120289) and HPS2-THRIVE (Heart an analysis for interaction should have been per- Protection Study 2: Treatment of HDL to Reduce formed to give statistical support for the fact that the Incidence of Vascular Events; NCT00461630) the relationship between changes in LDL choles- — will provide additional, pivotal information terol level and carotid intima–media thickness was regarding the current hypothesis that the addition indeed different between the two groups, be- of niacin to statin therapy leads to a further re- yond what would be expected due to chance. duction of the risk of clinical events in our pa- The conclusions about this relationship from tients.
the post hoc analysis in the ARBITER 6–HALTS Dr. Kastelein reports receiving consulting fees, lecture fees, study are at odds with those in the recent report and grant support from Merck and lecture fees from Schering- Plough; and Dr. Bots, lecture fees and grant support from Astra- of the SANDS (Stop Atherosclerosis in Native Zeneca. No other potential conflict of interest relevant to this Diabetics Study) trial (NCT00047424),3 in which article was reported.
the addition of ezetimibe to statin therapy was From the Department of Vascular Medicine, Academic Medical associated with the prevention of a progression Center, Meibergdreef (J.J.P.K.); and the Julius Center for Health of carotid intima–media thickness. However, both Sciences and Primary Care, University Medical Center Utrecht, Utrecht (M.L.B.) — both in the Netherlands.
the ARBITER 6–HALTS and SANDS studies are small and used a surrogate marker; we believe This article (10.1056/NEJMe0908841) was published on No- that firm conclusions about this relationship must await the findings of large studies involving clini- 1. LaRosa JC, He J, Vupputuri S. Effect of statins on risk of
coronary disease: a meta-analysis of randomized controlled tri- n engl j med 361;22 nejm.org november 26, 2009 Downloaded from nejm.org by JEFFREY SHANES on February 24, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved. 2. Kastelein JJ, Akdim F, Stroes ES, et al. Simvastatin with or ins. Circulation 2004;110:3512-7. [Errata, Circulation 2004;110:
without ezetimibe in familial hypercholesterolemia. N Engl J 3615, 2005;111(24):e446.] Med 2008;358:1431-43. [Erratum, N Engl J Med 2008;358: 7. Crouse JR III, Raichlen JS, Riley WA, et al. Effect of rosuva-
statin on progression of carotid intima-media thickness in low- 3. Fleg JL, Mete M, Howard BV, et al. Effect of statins alone risk individuals with subclinical atherosclerosis: the METEOR
versus statins plus ezetimibe on carotid atherosclerosis in type Trial. JAMA 2007;297:1344-53.
2 diabetes: the SANDS (Stop Atherosclerosis in Native Diabetics 8. van der Heijden GJ, Donders AR, Stijnen T, Moons KG. Im-
Study) trial. J Am Coll Cardiol 2008;52:2198-205.
putation of missing values is superior to complete case analysis 4. Rossebø AB, Pedersen TR, Boman K, et al. Intensive lipid and the missing-indicator method in multivariable diagnostic
lowering with simvastatin and ezetimibe in aortic stenosis. research: a clinical example. J Clin Epidemiol 2006;59:1102-9.
9. Barrett-Connor E. Looking for the pony in the HERS data:
5. Taylor AJ, Villines TC, Stanek EJ, et al. Extended-release nia-
Heart and Estrogen/progestin Replacement Study. Circulation cin or ezetimibe and carotid intima–media thickness. N Engl J 2002;105:902-3.
10. Pocock S, White I. Trials stopped early: too good to be true?
6. Taylor AJ, Sullenberger LE, Lee HJ, Lee JK, Grace KA. Arte-
rial Biology for the Investigation of the Treatment Effects of 11. Montori VM, Devereaux PJ, Adhikari NK, et al. Randomized
Reducing Cholesterol (ARBITER) 2: a double-blind, placebo- trials stopped early for benefit: a systematic review. JAMA controlled study of extended-release niacin on atherosclerosis 2005;294:2203-9.
progression in secondary prevention patients treated with stat- Copyright 2009 Massachusetts Medical Society. Deleterious Effects of Right Ventricular Pacing
In this issue of the Journal, Yu et al. report results cardiography to assess left ventricular ejection of the Pacing to Avoid Cardiac Enlargement fraction and dyssynchrony in the majority of pa- (PACE) study, a prospective, double-blind, multi- tients, but they did not comment on the number center trial designed to determine whether bi- of patients who had dyssynchrony at baseline or ventricular pacing is superior to right ventricu- whether there were changes during long-term lar apical pacing for the prevention of adverse follow-up. The study was relatively small in changes in left ventricular function.1 The inclu- scope because it was powered to assess the pri- sion criteria focused on patients with a left ven- mary end points rather than mortality or other tricular ejection fraction of 45% or more who outcome measures. Although the atrioventricu- required pacing because of high-grade atrioven- lar intervals were programmed to achieve 97% tricular block or sinus-node dysfunction. The right ventricular pacing, we do not know how investigators observed significant differences in many patients with sinus-node dysfunction had the primary end points of left ventricular ejec- fusion of ventricular activation by pacing and tion fraction and left ventricular end-systolic vol- normal conduction. More aggressive pharmaco- ume. During a follow-up period that was limited logic therapy might have reduced the changes to 12 months, these end points did not change that were observed with right ventricular pac- in patients who were assigned to biventricular ing. Only a minority of patients were treated pacing, but there was a decline of 6.7 percentage with beta-blockers or angiotensin-converting– points in left ventricular ejection fraction and a enzyme inhibitors because the left ventricular 25% increase in left ventricular end-systolic vol- function was normal at baseline. It remains un- ume in patients who were assigned to right ven- determined whether the deleterious effects of tricular pacing. The left ventricular ejection frac- right ventricular pacing that were observed at 12 tion declined to less than 45% in 9% of the months would progress inexorably over time, patients in the right-ventricular-pacing group. but it is doubtful that this question could be an- With regard to the secondary end points, the swered without jeopardizing the safety of pa- investigators did not detect significant differ- tients with sinus-node dysfunction who did not ences in the distance covered in a 6-minute walk, require right ventricular pacing.
the scores on a quality-of-life assessment, or The observations by Yu et al. add to a growing hospital admissions for heart failure.
body of evidence that right ventricular pacing in- The investigators used three-dimensional echo- creases the risk of left ventricular dysfunction and n engl j med 361;22 nejm.org november 26, 2009 Downloaded from nejm.org by JEFFREY SHANES on February 24, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

Source: http://www.ccmconsultants.com/assets/1/7/Statin_Therapy_with_Ezetimibe_or_Niacin_in_High-Risk_Patients_-_Editorial.pdf