The estrogen controversy

I was planning to address this issue once again today because of the NY Times article – Hormone Studies: What Went Wrong? I couldn limit my ranting, because Medpundit has already provided an excellent rant.

However, I cannot resist adding my own reflections. I present a paradigm for medical knowledge. I hope this discussion adds texture to our understanding.

Medical research has not failed here. Nor have physicians failed in their actions. Nor has the public health system failed.

At any time in medical history, we (physicians) must make health care recommendations base upon the best available evidence at the time. Researchers, and most practicing physicians, know that randomized controlled trials trump epidemiologic studies. The physicians who designed the Women’s Health Initiative understood the importance of a prospective randomized study. Until we have such data, we cannot know the results of any intervention.

However, while we await randomized controlled trials, we must make decisions. Some have described medicine as the art and science of making decisions under conditions of uncertainty. The more general ones practice, the more uncertainty one must accept.

When a patient enters your office, or you enter the patient’s hospital room, the physician must make many decisions. What questions should I ask? Which physical exam maneuvers should I perform? What tests should I order? What prevention should I recommend?

While these decisions are somewhat data driven, the data are often not definitive. We (physicians) must play the odds. We must take the available data, using those data to make the best possible decision for our patient.

I have found that physicians generally act with the patient’s best interest as the first and only concern. (I do admit that it pains me to have to add the adverb generally to that statement.) Given the uncertainty in the data, we (the medical community) will make mistakes. For many years, we prescribed antiarrhhthmics to post-MI patients with more that 6 PVCs per minute. Then a well done randomized controlled trial demonstrated that the anti-arrhythmic drugs were really proarrhythmic and lead to an increased death rate. The common practice prior to the study release made sense. Almost all cardiologists had embraced that approach. It took a randomized controlled trial to provide common practice wrong.

This situation is similar. The evidence for post-menopausal estrogens was strong. Several analyses of the epidemiologic data (and not just the Nurses’ Health Study) supported the role of post-menopausal estrogens.

I view this story as a positive one. This story demonstrates the importance of scientific inquiry. We can only approximate the truth, but we can always strive to get close to that elusive golden ring.

We now know more than we did prior to the Women’s Health Initiative. As usual, each study raises as many questions as it answers. Researchers will continue to provide information that will help us understand the varied effects of post-menopausal estrogens. In the meantime, we will continue to make decisions based upon the best available data. Some of those decisions will actually be helping patients; some of those decisions may be hurting patients. We can only hope that the helping outweighs the hurting.

On carbohydrates and weight loss

The Carbo War, Cont’d This article summarizes the controversy over low carb diets. It includes a common sense approach to dieting. I recommend this article strongly.

hsCRP – a conversation

As our knowledge of coronary artery disease evolves, so do we change our thoughts on both primary and secondary prevention. Highly sensitive C reactive protein (hsCRP) has received much attention over the past few years (including several rants here). New Test for Hearts at Risk: What It Can and Can’t Do

A panel of experts has recommended that tens of thousands of people be tested for a new indicator of heart disease. The indicator, C-reactive protein, is produced in the liver in response to injury or inflammation, and it can help identify people at higher risk for heart attack and stroke, including those with normal cholesterol.

But the panel, convened by the Centers for Disease Control and Prevention and the American Heart Association, stopped short of recommending the test for every adult and said that the test should not replace assessments of other risk factors, like cholesterol, high blood pressure and diabetes.

People with a low risk of heart disease and those with a high risk will most likely not benefit from the test, called high sensitivity C-reactive protein, or hs-CRP, the panel says.

In an interview, Dr. Thomas A. Pearson of the University of Rochester School of Medicine and Dentistry, co-chairman of the panel, helped clarify the recommendations.

I recommend this question and answer session. Some physicians may want to print this to hand out to inquiring patients (caution NY Times links “disappear” in ~ one month). Therefore I will include this snippet.

Q. According to your panel, who should be tested for hs-CRP?

A. Typically it’s people who are borderline high risk. First, a doctor would measure a patient’s major risk factors, like cholesterol, high blood pressure, smoking, diabetes, sex and age. Then he’d calculate the patient’s 10-year risk of developing a heart attack or dying of heart disease. We have simple formulas we use for that. If the risk falls between 10 percent and 20 percent, you’re in a situation in which you don’t know whether to treat this person or not, so doing this test would help inform your decision.

Q. Can people figure out their own 10-year risk?

A. They can go to the American Heart Association’s Web site, americanheart .com, which has a risk assessment tool.

Q. Can you give a sense of someone who has a 10 percent to 20 percent chance of developing a heart attack or dying of heart disease in the next 10 years?

A. Smoking gets you there very fast; or having two or three elevated risk factors, like high bad cholesterol, high blood pressure and being an older male, which is a substantial number of Americans. If you’re obese, you probably have several risk factors, and that will get you into that category.

Q. Why do you suggest not testing people for the heart disease indictor if their risk of having a heart attack in the next 10 years is less than 10 percent? Or those whose risk is greater than 20 percent?

A. We feel very strongly you shouldn’t test someone in whom you wouldn’t do anything different on the basis of the test. You wouldn’t do it in someone who has coronary disease, a previous heart attack or stroke or diabetes, as we’re going to treat these people aggressively no matter what the results of hs-CRP are.

Similarly if you have a young person with few risk factors, you wouldn’t test him either, because you’re not going to do anything about it. If a young person had a 2 percent risk factor, and had a high hs-CRP, it would only bring his risk up to 4 percent, which is a level we still wouldn’t treat.

I believe the panel has acted cautiously in their recommendations. We are starting to order hsCRP for consideration of secondary prevention – i.e., known atherosclerotic disease and normal cholesterol levels. If hsCRP comes back high, we will use that as an excuse to prescribe statins.