Monday, January 26, 2009

Which Vitamin Should I Take?

A staff member in my office asked me which vitamin(s) I would recommend for her to take. She stated her energy was down and wanted to know the best vitamins. This is also a common question I get asked by patients. The staff member seemed a little surprised when I told her I didn't really recommend any vitamins.

There is little proof that vitamins (and other dietary supplements for that matter) can actually help you. On the contrary, if anything the evidence shows that vitamins might even harm you! I have blogged in the past about this, mentioning a review of 68 trials in over 200,000 patients published in February 2007 in JAMA showed that treatment with beta carotene, vitamin A, and vitamin E may increase your risk of death!

Benefits of vitamins like E and C were linked to their antioxidant properties, and have been suggested to prevent colds, improve memory and extend life. However, I have blogged previously about a study on Vitamin E and C that was published in the Journal of the American Geriatrics Society that studied thousands of seniors showing that this did not prevent dementia. In addition, an extensive review of the literature found that vitamin C did not prevent the common cold.

The list of studies goes on and on. There is just no convincing data that these things are helpful. Part of the reason may be that supplements work when you have a diet lacking vitamins and minerals. In the US, our diets are not healthy, but clearly we are getting some nutrition given that the majority of us are overweight or obese. There are a few exceptions where vitamins and supplements may actually help:

1. Folic acid for women of child bearing age. This prevents neural tube defects in babies .
2. Calcium. Calcium is recommend for most adult women to prevent osteoporosis. However, even here there is no data to support that calcium prevents fractures. Yet, it is an important part of bone metabolism, so it is recommended.
3. Vitamin D. We are learning more and more about Vitamin D. I have blogged before about a study published in the Annals of Internal Medicine showing that women with low vitamin D levels have a much greater risk of fracture than women with normal or high vitamin D levels, and it just today it was reported that a low level of vitamin D in older patients is associated with a higher risk of cognitive impairment.

Why Does it Matter?
So what? Even if there are no studies to prove that vitamins and supplements work, they are probably safe, and may help. This is for the most part true. In addition, one can argue that research is lacking because there is no financial incentive (unlike the drug companies) to do the research needed. However, I don't like to recommend things for patients which they will need to pay out of pocket for unless I know it will help them. Whereas some pharmacies are giving away generic antibiotics, and charging only $4 for some generic medications, some women's multivitamins sell for over $34 a bottle (and that's just at Target).

Bottom Line
It's not clear whether or not vitamins really help. If you are a women of child bearing age, you should probably take calcium and folic acid. If you are a senior, you should probably take calcium and vitamin D. If you want to take a multi-vitamin, a generic, inexpensive, once daily mutli-vitamin should be fine. I would not take extra doses of other vitamins or supplements, especially vitamins A, C, or E.

Tuesday, January 13, 2009

FDA, Thanks for the Warning!

As reported in the Associated Press, the FDA just announced that it had completed its analysis of Singulair for suicide, and found absolutely no link.

"Regulators said that only one in nearly 10,000 patients taking Singulair in clinical trials reported suicidal tendencies. No patients committed suicide."

I blogged about this when the FDA announced it was going to look into the matter back in March. I am no fan of Singulair (see here ), but was extremely skeptical of this link given the relatively clean safety profile of singulair, the lack of biologic plausibility, and the fact since millions of patients with allergies and asthma take Singulair (which is unfortunate, because it doesn't work all that well and extremely expensive), if there was a really link it would've shown up by now.

I have no problem with the FDA investigating concerns with medications. This is what they are supposed to do. The problem I have is that they report the fact that they are investigating a problem, without any data or information to be informative to doctors or patients. I raised these concerns when the FDA initially proposed this strategy. There is a reason that in the United States a person is presumed innocent until found guilty. Because of our 24/7 media cycle and their desire to publish scary stories about medicine, when the FDA announces it is looking into a medication, this automatically makes headlines. Patient and physicians, being provided limited to no data as well as no clinical context, assume the worst. Many patients will simply stop taking a medication, without even informing their doctor.
Please recall that though the Vioxx scandal was one of the first events to put the spotlight on the FDA, the Avandia scare magnified this. As I have mentioned multiple times, well done studies on Avandia (see here ) have not led to the same conclusion as Dr. Nissen's poorly done meta-analysis. The ramifications of that dreaded publication by the New England Journal of Medicine have been numerous and include decreasing the availability of newer drugs for diabetes, guidelines which push patients onto insulin sooner, and the FDA needlessly worrying the public about safe (though ineffective) medications with these early warnings of investigations.

Thursday, January 8, 2009

Vytorin: To Switch or Not to Switch? That is the Question

The FDA has just released their review of the ENHANCE trial involving Vytorin. I have blogged about Vytorin, a combination of a statin medication (simvastatin) and another cholesterol lowering medication that blocks cholesterol absorption from the gut called ezetimibe (Zetia), and the ENHANCE trial many times (here, here, here, here, here, and here)

The FDA's analysis concluded that after 2 years of treatment with either Vytorin or simvastatin alone, carotid artery thickness (a marker of risk for cardiovascular disease) increased by 0.011 mm in the Vytorin group and by 0.006 mm in the simvastatin group, which was not statistically significantly different. However, the levels of LDL (bad) cholesterol decreased by 56% in the Vytorin group and by 39% in the simvastatin group, which was statistically significantly different.

Here's what the FDA concluded:

The results from ENHANCE do not change FDA’s position that an elevated LDL cholesterol is a risk factor for cardiovascular disease and that lowering LDL cholesterol reduces the risk for cardiovascular disease. Based on current available data, patients should not stop taking Vytorin or other cholesterol lowering medications and should talk to their doctor if they have any questions about Vytorin, Zetia, or the ENHANCE trial.

First, the FDA is stating that despite the fact that LDL lowering did not correlate with improvement in atherosclerosis, LDL is still a reliable surrogate marker for cardiovascular disease and thus a good target of therapy.

As I have mentioned previously, when you take a drug to prevent a heart attack or stroke, you want to know whether that drug actually decreases heart attacks or strokes (outcomes). The problem is that doing outcomes studies requires a large number of patients to be studied over a long period of time, which is difficult and costly. Therefore, many of the studies we use to make medical decisions rely on markers or proxies for disease called "surrogates." Cholesterol (specifically the bad cholesterol or LDL) is a known surrogate marker for heart attack and stroke, as is carotid artery thickness, the endpoint in ENHANCE. Though both surrogates correlate well with the outcome of heart attacks and stroke, in the ENHANCE study lowering the LDL did not cause a decrease in carotid artery thickness. Despite this, the FDA still believes that LDL is a good surrogate.

Second, the FDA states that because LDL is still a good surrogate and because Vytorin lowered LDL, depsite that fact that it failed to do any good in the ENHANCE study, the FDA recommends that patients NOT stop Vytorin. This one is more problematic.

I agree that LDL is a good surrogate, but we have been burned by surrogates before. High levels of homocysteine are linked to heart attacks, and folic acid lowers homocysteine, but when we looked at whether folic acid reduced heart attacks (outcomes), it actually caused MORE heart attacks, and is no longer recommended.

We know the lowering the LDL with a statin reduces heart attacks, and this relationship is in fact linear (the lower the LDL with a statin the lower the risk of heart attacks-see figure from 2004 NIH guideline update). The questions is whether lowering the LDL with something other than a statin, i.e. ezetimibe (Zetia) will reduce the risk of heart attacks and strokes. It may very well, but we have no evidence to date. Though ENHANCE may not have been the perfect trial to look at this, it is the only study we have using Vytorin/Zetia looking at something other than LDL, and this study was negative. Finally, it may be more about the statin than the LDL. The JUPITER study (which I have also blogged about) showed that lowering the CRP with a statin (Crestor)decreased heart attacks. LDL's were also lowered, but where CRP was elevated at the beginning of the study, LDL was relatively normal and would not have normally required a statin.
Thus, though the FDA suggests to not to stop taking Vytorin because it does lower LDL (and also does not appear to have a major safety issue), the question is whether or not you should be on Vytorin (statin plus zetia) or just a statin? Based on the figure above, by using a combination (Vytorin) instead of a statin alone to achieve LDL lowering, are you denying yourself the maximum amount of statin that you need to acheive outcomes? We don't have the studies to answer this question, but in my opinion it would be better to lower your LDL with a stronger statin like Lipitor and Crestor, then to use a weaker statin (like simvastatin) plus Zetia, i.e. Vytorin. The highest dose of Vytorin lowers cholesterol just a little bit more than the highest dose of Lipitor and Crestor, but not by much. In addition, adding Zetia to any statin will lower your LDL an additional 25%. Thus, based on the available data, it would be better to use the dose of a statin needed to lower your LDL. If the maxium dose of a statin failed to achieve the LDL to goal, or one could not tolerate the maxium dose of a statin, the adding Zetia seems a reasonable option.
Bottom Line:
1. LDL is still a good surrogate market and target for therapy.
2. The best way to lower LDL is with a statin, and lower appears to be better.
3. If you are on Vytorin, I wouldn't stop it, but would discuss with your doctor about of switching to something like Crestor or Lipitor.
4. If you are unable to reach your LDL goal on the highest tolerable dose of Crestor or Lipitor, than adding Zetia seems reasonable.

Sunday, January 4, 2009

The Problem With Insulin

Before folks like Amy Tenderich at her excellent blog Diabetes Mine gets upset with me, insulin is a miracle of modern science. Frederick Banting and J.J.R. Macleod's Nobel Prize wining discovery of insulin has saved millions of lives and helped to prevent countless complications of diabetes such as blindness and kidney failure. While patients with Type 1 diabetes, who produce no insulin on their own, rely on insulin therapy to survive, the use of insulin in Type 2 diabetes is more complicated, since patients with early stage Type 2 diabetes initially have high levels of insulin and some type 2 diabetics with prolonged beta-cell failure eventually stop making useful insulin. Though where and exactly how to use insulin in a Type 2 diabetic is debated, newer recommendations from the American Diabetes Association (ADA) are promoting earlier use of insulin in Type 2 diabetics, which may not be entirely evidence based or in the best interest of the patient. This has caused me great concern.

The Problems With Insulin

Insulin is not just another treatment for Type 2 diabetes. The main complication of insulin use is hypoglycemia, or low blood sugar. Symptoms range to feeling nervous to complete coma. Hypoglycemia from insulin is a frequent cause of ER admissions, hospitalizations, and can even cause death. For both for sugar management and prevention of hypoglycemia, patients who use of insulin require frequent blood sugar monitoring. Though these monitors have become easier to use for patients, this is still a burden as well as an added cost. In addition, for the best blood sugar control, many endocrinologists recommend a basal/bolus approach to insulin therapy which is one injection of a 24 hour long acting insulin (like Lantus) and mealtime injections of short acting inuslins (like Humalog). Thus, the basal/bolus approach requires four injections a day and several sugar checks.

Though insulin is the most effective agent for lowering blood sugar, this may not be the best thing for all type 2 diabetic patients. Three studies presented this summer at the ADA (VADT, ACCORD and ADVANCE which I have blogged about before-here and here) not only failed to show cardiovascular benefit with aggressive diabetes control, but there was substantial hypoglycemia in the intervention arms (which used more insulin) and in ACCORD, the study was stopped early because of increased deaths.

In the January issue of Diabetes Care (the ADA's journal which each year publishes updates to their guidelines), a paper explaining how to interpret these studies come to similar conclusions of my previous posts, mainly that good diabetes control (A1c less than 7%) is still very important. They also mention " it is biologically plausible that severe hypoglycemia could increase the risk of cardiovascular death in participants with high underlying CVD risk. This might be further confounded by the development of hypoglycemia unawareness, particularly in patients with coexisting cardiovascular autonomic neuropathy (a strong risk factor for sudden death). Death from a hypoglycemic event may be mistakenly ascribed to coronary artery disease, since there may not have been a blood glucose measurement and since there are no anatomical features of hypoglycemia detected postmortem."

In other words, it may have been hypoglycemia, which is caused primarily by insulin, that killed the aggressively treated diabetics.

In fact, a recent commentary in JAMA similarly mentions that "the paradox that blood glucose levels are strongly associated with the risk of cardiovascular disease but that glycemic control does not prevent cardiovascular disease may perhaps be explained by insulin resistance or hyperinsulinemia." and "If insulin levels are toxic to the cardiovascular system, then treatments designed to reduce insulin levels rather than glucose levels might be associated with a reduced risk of cardiovascular events in patients with type 2 diabetes."

In other words, glycemic control is important, but other factors such as insulin levels and insulin resistance might be more important, so driving down the sugar with insulin in type 2 diabetics may not have been the way to go.

The Problem With the Older Diabetes Pills

The two most commonly prescribed (and guideline recommended) pills for diabetes are metformin and sufonlyureas. They are generic, inexpensive, have a long track record, and will lower blood sugars in type 2 diabetics. The problem with these medications (beside side effects, particularly hypoglycemia with sulfonylureas), is that they only work for a short period of time.

Analysis have shown that though both metformin and sufonlyurea lower blood sugar; and, compared to usual care, those whose goal A1c was less than 7% had lower microvascular (eye and kidney problems) complications; about half of all patients taking either of these medications lost diabetes control by three years, with the vast majority losing control at 9 years. Though a recent report on the UKPDS showed that just being in the intervention group reduced heart attacks and strokes 10 years after the follow up period, imagine how many heart attacks and strokes might have been prevented had patients maintained glycemic control over a much longer period of time!

It should be no surprise that patients fail on metformin and sufonlyurea because they primarily work by lowering sugar, and don't seem (especially with sufonlyurea) to address the underlying defects of type 2 diabetes which are insulin resistance, beta cell dysfunction and glucose regulation. The good news is that newer agents for diabetes do address this. Published on December 7th, 2006, in one of the largest diabetes studies ever done, the ADOPT trial showed that Avandia, a thiazolidinedione or TZD, substantially reduced loss of diabetes control compared to both metformin (Avandia 32% better) and sufonlyurea(Avandia 63% better). TZD's work by reducing insulin resistance, and also can improve beta cell dysfunction and both effects were seen in ADOPT. After 5 years of treatment, only 15% of patients failed on Avandia monotherapy. This is far better than the failures from UKPDS.

The Problem With the Diabetes "Experts"

You would think that with all the problems with insulin and the older diabetes pills that diabetes experts and researchers would welcome newer therapies for Type 2 diabetes, but the trend is just the opposite.

In the editorial to the ADOPT study, Dr. David Nathan, author of the current American Diabetes guidelines stated that "given the modest glycemic benefit of rosiglitazone and higher cost metformin remains the logical choice when initiating pharmacotherapy for type 2 diabetes." Yet, at this point in time, even he seemed convinced that Avandia was a better second choice then sulfonylurea, stating "the benefits of rosiglitazone over glyburide, which is often not recommended for older patients because of its increased risk of hypoglycemia, are more convincing. "

However, only a few months later (2/1/2007), in response to the recent approval of Januvia, Dr. Nathan published this Perspective piece in the New England Journal of medicine , in which he railed against new diabetes medicines stating that newer diabetes drugs like Avandia and Januvia , "are generally no more potent, and often less effective in lowering glycemia, than the three oldest classes (insulin, the sulfonylureas, and the biguanides), all of which are more than 50 years old. Moreover, the newer classes are uniformly more expensive and are associated with adverse effects — some that are shared by the older drugs, but others that are new." In essence, Dr. Nathan is suggesting (despite the problems with insulin and the failure of the older drugs that I mentioned above), that we just stick with the status quo, and use caution before approving newer diabetes medicine. See my Letter to the Editor in the NEJM criticising this position.

The current guidelines for the medical management of diabetes (which are taken directly from Dr. Nathan's previous consensus statement) state the patients with Type 2 diabetes should be started on metformin, and when that fails, add either sulfonlyurea, insulin or TZD. The language of that paper states that TZD's are an option, but are expensive and cause side effects. In the more recently published consensus statement (also authored by Dr. Nathan) that will likely be adopted in the 2009 guidelines, Avandia and Januvia are not recommended. The recommended choices of treatment for type 2 diabetes are metformin, sulfonylurea and insulin. Less preferred are Actos and Byetta. Thus, given that we know the older medications fail, the newer guidelines will likely push more and more type 2 patients to use insulin. I have previously commented that this trend is already happening.

I recently blogged about conflicts of interest that go beyond the drug companies, and used the example of differing prostate cancer screening guidelines. The government's guideline (which wants to save money) does not recommend this practice, where as the urologist's guidelines (who benefit from aggressive screening) do recommend prostate cancer screening. Both guidelines use the same studies, and neither guideline is right or wrong, but must be interpreted from the perspective of the authors.

So, given the problems with insulin and the older diabetes pills as elicited above, why would the "experts" of the new guidelines recommend these therapies over newer therapies with potential benefits (and leave out some newer therapies altogether)?

One answer (in my opinion) is that endocrinologists make recommendations that will lead more patients toward insulin is because they have an incentive to do so. Most patients with type 2 diabetes are cared for not by specialists but by primary care physicians. Though many primary care physicians are quite comfortable using insulin, many referrals to endocrinologist come from primary care because a patient has failed pills and now needs to be started on insulin. Insulin therapy requires substantial time, education and close patient follow up. This is time that many PCP's simply don't have, whereas most endocrinologists work with a certified diabetes educator who is specifically trained just for this purpose.

One can look at the failure of Exubera, Pfizer's inhaled insulin, as evidence of how many endocrinologists cling to injections. Pfizer did not pull Exubera because it didn't work (worked great) or that patients didn't like it (my patients and I were extremely upset when this was pulled), but because endocrinologists didn't prescribe it. Whereas endocrinologist see starting insulin therapy as common and easy, my perspective as a PCP is that patients do not want to take insulin and see it as an absolute last resort. (Based on the evidence above, I would tend to agree).

The Problem With the New ADA Guidelines

To be fair, not all endocrinologists agree with the ADA or Dr. Nathan, similar to the way not all Republicans agree with George Bush. However, guidelines, especially ones from such as prestigious group as the ADA, have a tremendous influence on the way that primary care physicians practice medicine. Thus, these guidelines will likely have a tremendous impact on the way future diabetic care is delivered.

The problem with the new guidelines is that they been extremely affected by two factors: one poorly done, controversial study on Avandia (see here ) and one (or a few) highly influential endocrinologist, like Dr. Nathan, who sees no need for new diabetes meds, and would prefer that once patients lose control on the two older medicines most commonly used (which most will based on the data), that they be started on insulin. Furthermore, these recommendations which will encourage more insulin use are coming from specialist who might have an incentive to make such recommendations. Thus, these recommendations are not entirely evidence based and seem not to be in the best interest of patients to whom insulin is a burden and potential danger.