Zetia – down for the count?
Once again it would appear that Zetia (ezetimibe) is in desperate trouble, barely clinging to respectability in the lipid-lowering armamentarium. In 2008 the ENHANCE trial showed that ezetimibe added to 80 mg of simvastatin produced no additional increment of improvement in carotid intima-media thickness as compared to simvastatin plus placebo, despite the expected additional LDL lowering. This small study produced a remarkable level of unfavorable publicity for this agent. A study published later that year, in which the combination of simvastatin and ezetimibe was used in patients with aortic stenosis, showed no effect on the valve disease, but an unexpected increase in cancer incidence, and ezetimibe took another hit. A companion analytic article in that issue of the New England Journal by Peto and colleagues looked at several studies that used ezetimibe and concluded that it was unlikely that ezetimibe was increasing cancer risk, but the possibility could not be ruled out. Partial salvation arrived in the form of the Stop Atherosclerosis in Native Diabetes Study (SANDS), in which ezetimibe was added if the LDL cholesterol goal was not achieved with lifestyle modification and statin therapy. In this study ezetimibe proved useful in achieving improvement in carotid intima-media thickness, with such improvement being primarily correlated with the attaining of equivalent LDL cholesterol reductions, whether from a statin alone or a statin plus ezetimibe. Ezetimibe was here used as it might be used clinically, being added when first-line therapy had not succeeded in reaching the goal LDL. Finally, of course, there was ARBITER 6-HALTS, presented at the AHA meeting and simultaneously published in the New England Journal. Here again the endpoint is carotid intima-media thickness, the total number of subjects (with coronary disease or a coronary risk equivalent) relatively small (363), but the randomization now comparing extended-release niacin (goal of 2000 mg per day) with ezetimibe, both added to statin therapy. Niacin resulted in significant regression of carotid intima-media thickness whereas ezetimibe showed no such effect. Curiously, greater reductions in LDL in the ezetimibe group were associated with increases in carotid intima-media thickness. Furthermore, despite the relatively small number of subjects, there were significantly fewer clinical events in the niacin group.
Where does this leave us? Statin therapy pushed to full dose is clearly the primary approach to achieving goal LDL levels. Extended release niacin should be the next agent in line, either added to full dose statin therapy or used synergistically in patients with low HDL levels. The data suggesting that niacin is useful goes back to the Coronary Drug Project, in which a 15 year follow-up published in 1986 showed a mortality benefit in subjects in the niacin arm. A study published in 2001 in the New England Journal by Greg Brown and colleagues, very reminiscent of ARBITER 6-HALTS results, showed both angiographic and clinical improvements in subjects treated with both statin and niacin therapy despite a trial design (only 160 subjects) that would have been thought to preclude the attainment of clinical endpoints. But large-scale trials of niacin are only now being carried out. Likewise findings from a large-scale clinical endpoint study of ezetimibe (IMPROVE-IT) are not expected to be available until 2012. But the data on niacin, going back many years, is consistently positive whereas the data on Ezetimibe has to raise concerns.
Ezetimibe was brought to market without clinical endpoint data. It rode on the coattails of numerous studies that suggested that lowering LDL, however accomplished, produced good results. The FDA approved Ezetimibe quite quickly, the thinking being that LDL lowering accomplished by pharmacologic therapy (statins, cholestyramine), diet therapy (Dean Ornish's regression study) or even ileal bypass (the POSCH trial) had all been associated with favorable outcomes. Nonetheless, it was certainly a leap of faith to assume that LDL lowering, no matter how accomplished, would always yield a favorable outcome.
Still, while acknowledging that niacin in its extended-release form is a very useful and synergistic agent which further lowers LDL while raising HDL and lowering triglycerides and Lp(a), it must be recognized that niacin is often not tolerated by many patients because of flushing and GI upset. So with the bar being progressively lowered to an LDL goal of <70 mg/dl and perhaps even lower, there are many patients who do not attain such a goal on statins alone, and may not tolerate niacin. That leaves us with the choice (assuming that diet and weight loss have been taken as far as possible) of ezetimibe or resin therapy (fibrates primarily lowering triglycerides). Resins such as cholestyramine or Welchol do lower LDL, but the effect is relatively modest and these agents cause G.I. symptomatology that is often not tolerated.
Given therefore the still somewhat conflicting data on ezetimibe from small studies using surrogate endpoints, it may be appropriate to continue using this agent in individuals where the alternative is to leave the LDL level higher than is desirable. It is also true that the degree of LDL lowering seen with ezetimibe is quite variable, with some individuals having no response and others having quite striking responses, possibly reflecting the amount of cholesterol in bile and the volume of bile produced by given individuals. Restricting the use of ezetimibe to those individuals who have a good response is more likely to produce a favorable outcome, as was suggested by the SANDS trial. In my clinics this week I saw six individuals on ezetimibe, and after careful consideration I did not stop this medication in any of them. These are patients who were either intolerant of statins or had sufficient disease to justify LDL levels much lower than was achievable without the addition of ezetimibe. Given the available data, such an approach seems reasonable.
Some thought on using niacin
- Extended release niacin is preferable to short-acting niacin. Several studies have shown that it is better tolerated and produces better results.
- The prescription form of extended release niacin, Niaspan, is generally preferable, in part because it has been taken through the FDA process and shown to be effective, and also because it is generally covered by third-party payers. There is one form of over-the-counter niacin – Slo-niacin by Upsher-Smith, which is also effective and of high quality. Other forms should generally be avoided because there is absolutely no quality control over supplements and certain forms of niacin have produced serious liver toxicity (“Goldline” niacin in particular).
- "No-flush” niacin does not produce flushing, but it is also completely worthless (references available on request)
- Niacin should be avoided in patients with poorly controlled reflux, with esophagitis, with high uric acid levels (niacin can precipitate gout), with a history of sensitive skin and rashes, and with poorly controlled diabetes. Niacin can worsen glucose tolerance but can be used in patients with reasonably well controlled diabetes.
Here are some specifics
- Niacin should be started at low dose and the dose built up slowly. Start with 500 mg daily, increasing by 500 mg every few weeks or as tolerated, with the effective range being 1000 – 2000 mg. Below 1000 mg produces only a modest effect and above 2000 mg generally produces only more toxicity.
- Niacin should always be taken with food. The company recommends taking it with a snack before bedtime, the theory being that the patient will sleep through a possible flushing episode. I personally do not like to encourage patients to eat before going to bed and generally tell them to take niacin after the evening meal. There are some patients on higher amounts of niacin who do better splitting the medication into two daily doses.
- Patients on niacin should take a full dose aspirin daily, aspirin blocking much of the prostaglandin-mediated flushing.
- Tolerance to niacin builds up very rapidly and so flushing is generally worst at the very beginning of therapy and quickly improves. The flip side of this is that tolerance is lost equally quickly, and I have a number of patients who tell me if they miss a single day they will flush on the following day. Thus niacin is an "unforgiving" medication and is not an appropriate medication for the casually adherent pill-taker. It is important to emphasize to patients that the drug must be taken exactly as prescribed and not forgotten. Someone on 2000 mg a day who goes on a weekend trip, forgets to take the medication along, and three days later resumes 2000 mg a day is likely to be quite unhappy.
- Excesses of alcohol, spicy foods, or hot environments will all exacerbate flushing. I have one Indian patient who was on high-dose niacin without difficulty, but on a trip to India the combination of heat and spicy food rapidly resulted in his needing to stop taking niacin altogether until his return.