Thursday, February 19, 2009

(Brief) Reaction to Article: "Percutaneous Coronary Intervention versus Coronary-Artery Bypass Grafting for Severe Coronary Artery Disease"

"Percutaneous Coronary Intervention versus Coronary-Artery Bypass Grafting for Severe Coronary Artery Disease" by Patrick W. Serruys, M.D., et. al. (New England Journal of Medicine (NEJM), March 5, 2009)

I was drawn tonight to the above NEJM article by an article posted on the New York Times (NYT) website, "Heart Stents Found as Effective as Bypass for Many Patients" by Roni Caryn Rabin. This NEJM article has yet to be published in print but was published in advance online.

My first two reactions to the study (and there were many others but these I feel are the most important) :

(i) The follow up of patients was only for one year. I suspect the majority of us are hoping to live much longer than one year after our procedure... Follow up at, let's say, 5 years would be much more valuable, and I think much more revealing. There are usually two periods of divergence between patient groups that we see when evaluating vascular surgeries - such as evaluating the benefits of placing stents versus performing surgery with carotid stenosis (blockages of the arteries that supply blood to the brain and increase people's risk of strokes): early on and long term. Surgeries typically have higher risks early on but lower long term, and with stents the reverse is typically true. Thus, a further divergence between patient groups after one year would not be a surprise to the medical community.

(ii) There are
significant differences between the medicines the two groups received after their procedure (even simple medicines like aspirin and cholesterol lowering medicines like statins - drugs that we know reduce strokes, heart attacks, etc.) I encourage you to look at Table 2 of the article and compare the two columns, PCI (patients receiving stents) and CABG (patients receiving surgical bypasses). The patients who had bypass surgery consistently, in regard to each of the medicines listed, were less like to receive the medical standard of care. Those differences make the comparison between the groups of patients much more difficult. Rather than comparing apples and apples, the study is forced to compare more apples and oranges. One wonders at how the two groups would have differed if they had both received the general standard of medical care after their procedures. Perhaps the surgical bypass patients would have done even better. One would expect so.

By the way for the women out there, 78% of the study participants were men - a flaw that continues to plague studies, even today.

If you are interested, I encourage you to attempt to read the article - you will understand more than you think and be able to make your own opinions. If not, I encourage you to read the editorial accompanying the article, Coronary Revascularization in Context by Richard A. Lange, M.D., and L. David Hillis, M.D. (NEJM, March 5, 2009).

Consumers should be aware of the limitations of what they read and are told. For those interested, previous entries on related subjects: The almighty cardiac stent and Reaction to Article: "Communicating Medical News -- Pitfalls of Health Care Journalism".

Sunday, February 8, 2009

Placebos, doctors, and fibromyalgia (Addendum)

As discussed in the last entry, one of the problems with surveying doctors' prescribing habits in treating fibromyalgia is that it disregards the controversy surrounding the diagnosis. Therefore, does doctors' use of placebos in treating fibromyalgia really represent their general use of placebos? Published online today is an article that elucidates further the skepticism that surrounds this diagnosis.

For those interested: "
Drugmakers' push boosts 'murky' ailment" written by Matthew Perrone, AP Business Writer.

Sunday, February 1, 2009

Placebos, doctors, and fibromyalgia

A placebo, according to the American Heritage Dictionary, is “a substance containing no medication and prescribed or given to reinforce a patient's expectation to get well.”

When we think of placebos, we generally think of sugar pills; however, the concept of placebos has evolved to include pills with active ingredients and more broadly any “treatment whose benefits (in the opinion of the clinician) derive from positive patient expectations and not from the physiological mechanism of the treatment itself.” (BMJ 2008;337:a1938; Prescribing "placebo treatments": results of national survey of US internists and rheumatologists).

Using this broad definition, a group of researchers set forth to study the current practices of U.S. doctors in prescribing placebos. Its study results, published in the British Medical Journal (BMJ) in October 2008, were reported in all the major newspapers, including an article in the New York Times.

The study, as it was reported, found that approximately half of the American internists and rheumatologists surveyed regularly prescribe placebos.

While the general public likely was surprised by these results, as some doctors might have been, anyone reading the actual study likely was not. Despite the suggestion that the study surveyed the general attitudes and behaviors of physicians in prescribing placebos, the study actually asked the questions specifically in the framework of fibromyalgia.


Fibromyalgia is a poorly understood condition of chronic pain over multiple muscle and soft tissue areas of the body often accompanied by nonspecific symptoms, such as fatigue, headache, and sensation of joint or tissue swelling. The symptoms of fibromyalgia occur with varying severity and can be aggravated by various factors, such as stress or poor sleep.

While a physical exam reveals tender points located symmetrically on both sides of the body, the patients look healthy and have no other real findings on exam, labs, or x-rays. Once other potential diagnoses are excluded, the diagnosis of fibromyalgia is made.

What causes fibromyalgia? We don't know. How best to treat it? We don't know.

Some studies suggest that prolonged sleep deprivation can cause symptoms similar to fibromyalgia, but many patients with fibromyalgia do not have sleep disturbances. An association of fibromyalgia with mood disorders has led some doctors to attribute the symptoms to psychiatric causes, but again the majority of patients with fibromyalgia do not have psychiatric disorders. In part because co-existing illnesses can confound the diagnosis, a clear explanation for fibromyalgia does not yet exist.

The current explanation for fibromyalgia is that the brains of patients with fibromyalgia are overly sensitive to the nerve signals emanating from the patients' muscle and soft tissues and, as well, over-interpret these inputs as pain. (There are parallels here with irritable bowel syndrome.)

Despite this knowledge and the knowledge that patients with fibromyalgia suffer from real pain, doctors in general, without any objective findings in their patient, have a hard time understanding the disorder and treating it. They tend to see fibromyalgia as in their patients' heads and can become as frustrated as the patients in trying to treat the chronic pain.

Treatments for fibromyalgia

The available treatments for fibromyalgia include patient education, aerobic exercise, muscle strengthening, cognitive behavioral therapy, and, of course, medications. The medications are of varying but limited benefit and include some tricyclic antidepressants, serotonin reuptake inhibitors (SSRIs), anticonvulsants, and muscle relaxants in combination with acetaminophen (Tylenol). Consistent with a lack of evidence for muscle or tissue inflammation in fibromyalgia, there is generally no benefit from anti-inflammatory agents, such as NSAIDs (a group of pain relievers that include over the counter medications such as ibuprofen and naproxen) or prednisone. The prescribing of narcotics other than tramadol are usually avoided due to a lack of evidence and a concern for long-term abuse.

A look back now at the BMJ study

The study surveyed doctors to explore current practices in prescribing placebos. Despite how the study was reported in the lay press, the questions were asked in the framework of treating fibromyalgia. Given what you now know about fibromyalgia (a condition of chronic pain with an unknown cause that is difficult to treat) are you as surprised by the study results? Clearly, if the study asked the questions in the framework of other diagnoses, such as rheumatoid arthritis or heart disease, you would not expect half of doctors surveyed to report regularly using placebos.

The study revealed the use of the following placebos: over the counter pain relievers (41%), vitamins (38%), sedatives (13%; usually implies benzodiazepines or 'valium-like' medicines - drugs that soothe or calm and can induce sleep), antibiotics (13%), saline (3%, salt water), and sugar pills (2%). Saline and sugars pills, in not containing active ingredients, were the only two placebos in the traditional sense of the word and made up 5% of cases in which placebos were used (not 5% of doctor-patient encounters for fibromyalgia).

Risk-benefit analysis

As observed in the study, the placebos of today are generally not sugar pills but pills with active ingredients. In fact, obtaining sugar pills from pharmacies is now extremely difficult as very few pharmacies will, and only on special request, make them – it requires time and know-how. I suspect doctors' fear of being sued also plays a role in the prescribing of more active than inactive placebos, as a medicine with an active ingredient appears more easy to justify.

The beauty of sugar pills, of course, is that they carry little risk to the patient. If ultimately the patient and doctor want the patient to feel better and if a pill, whether a placebo or not, can be effective with little risk to the patient, then the risk-benefit analysis would seem to favor giving it.

There are obvious risks, however, in prescribing medicines with active ingredients: side effects, allergic reactions, interactions with other medicines the person is taking, etc. The use of antibiotics in the study as a purported placebo in treating fibromyalgia is, therefore, of a bit more concern. The antibiotics were presumably prescribed to allay patients' fears of Lyme disease or other chronic bacterial infections, reflecting unfortunately already known inappropriate use of antibiotics (see past entry: Antibiotics and the common cold). Without evidence of a clear bacterial infection, a risk-benefit analysis, particularly in fibromyalgia, would of course generally not favor antibiotics as they may lead to real, potentially serious medical problems.

A placebo effect for doctors?

Placebos are not only for patients, though. They can be for patients' families and even the doctors themselves. While doctors in the study may have justified the use of antibiotics as placebos for their patients, they very well may have been treating themselves - perhaps their own concerns about missing a diagnosis behind persistent, nonspecific complaints or a desire to discover a fortunate, effective and sufficient treatment for the patients' problems.

Less discussed is this idea that sometimes in prescribing specific treatments to patients, doctors are actually, in essence, treating themselves or treating themselves in addition to the patient – when, to rework the earlier definition of placebo, positive doctor expectations may outstrip the known benefits from the physiological mechanism of the treatment itself. The doctor's desire to help the patient get better or be more comfortable may at times be too great.

Of course, not for naught, positive doctor expectations may lead to positive patient expectations and treatment benefits via the placebo effect.

A few more thoughts from the study

The authors of the BMJ study on placebos bring up two additional interesting and unanswered questions for discussion: the ethical considerations of prescribing placebos and whether placebos work if patients know they are taking placebos.