Science Based Wellness

In a previous post I stated that I had a philosophical difference with Healthy Skepticism, an organisation dedicated to combating misleading drug promotions. Dr Peter Mansfield, the founder of Healthy Skepticism, posted a comment in which he stated that no philosophical differences existed.

I am afraid that I must respectfully differ with Dr. Mansfield. I do have a philosophical difference or, at any rate, a difference of opinion, with Healthy Skepticism. Rather than argue in the abstract I shall use a “case study” to illustrate our differences. To be specific I shall analyse the much hyped JUPITER study on the use of the cholesterol lowering drug, Rosuvastatin. AstraZeneca markets Rosuvastatin as “Crestor.”

One of the top 10 medical breakthroughs of 2008?

The TIME Magazine website contains a segment called The Top 10 Everything of 2008. The segment lists what TIME’s editors believe are the top 10 stories of 2008 for various news categories.

One of the “top 10″ categories is the top 10 medical breakthroughs of 2008. The alleged medical breakthrough that caught my attention appears second on the list. It is called Inflammation vs. Cholesterol.

Statins, cardiovascular disease and inflammation

Recall that high LDL-C levels often portend a heart attack. However it turns out that only about half of all heart attack victims have elevated LDL-C levels. The rest have “normal” LDL-C levels. Further, many people with elevated LDL-C levels never suffer a heart attack.

What accounts for the difference? Why do some people with apparently healthy LDL-C levels get heart attacks and why do many people with elevated LDL-C levels never suffer any sort of cardio-vascular problem?

The difference may be inflammation. Inflammation plays a role in both the formation of atheromatic plaque and in the rupture of the plaque cap that leads to the formation of blood clots. Could it be that it is the combination of elevated LDL-C levels and inflammation that leads to heart attacks? Could it be that inflammation is the root cause of heart attacks in people who have what appear to be healthy LDL-C levels?

A widely used class of cholesterol lowering drugs known as statins also have anti-inflammatory properties. In addition to reducing LDL-C levels they combat inflammation rather like aspirin and Panadol. Is it possible that it is the combination of LDL-C and inflammation reduction that make statins allegedly so effective in protecting people against heart attacks?

Testing for inflammation

But how do we know whether inflammation is present?

Just as there is a test to determine LDL-C levels there is a test to determine whether inflammation is present in the body. The test has the unfortunate acronym of hs-CRP which stands for high-sensitivity C-Reactive Protein.

The liver produces elevated levels of C-reactive protein when inflammation is present in the body. The hs-CRP test measures the level of C-reactive protein in the blood.

The JUPITER trials

And this brings us to the JUPITER trials.

JUPITER stands for “Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin.” Rosuvastatin is a statin, a cholesterol lowering drug, marketed by AstraZeneca under the name “Rosuvastatin”. JUPITER was intended to test the theory that Rosuvastatin could reduce the incidence of heart attacks and other cardiovascular diseases such as stroke among people with elevated CRP levels and normal LDL-C levels. AstraZeneca, Merck, Novartis and other pharmaceutical companies financed the JUPITER trials. Dr. Paul Ridker, the principal researcher, is joint patent-holder, along with Brigham and Women’s Hospital, of the hs-CRP assay used to determine CRP levels. The results of the JUPITER trial were published in the New England Journal of Medicine in November 2008.

The results created a sensation – especially among scientifically illiterate journalists. As TIME Magazine put it:

“…Ridker’s recent research, published in the New England Journal of Medicine, showed that when people with normal cholesterol and high levels of CRP… took statins, their CRP levels plummeted and their heart attack risk fell 54%.”

Well that’s fantastic! Heart attacks are a leading cause of death and disability in all industrialised countries. Isn’t it wonderful to have a drug that can cut the risk by more than half? No wonder the editors of TIME considered JUPITER one of the top 10 medical breakthroughs of 2008.

Does reality match the hype?

Let’s take a closer look at the JUPITER results and see whether the reality matches the hype.

17,802 people in 26 countries participated in the study. They were assigned randomly to a placebo group and to a group that received 20mg of Rosuvastatin daily. Both groups were monitored for the onset of cardiovascular diseases – called “primary endpoints” in the study. The primary end points of interest were:

• Myocardial infarction - ie heart attack
• stroke
• Arterial revascularization - for example, bypass surgery
• Hospitalisation for unstable angina.
• Death from cardiovascular disease. Cardiovascular disease is any disease involving the heart or blood vessels.

If significantly fewer members of the Rosuvastatin group suffered a “primary end point” as compared to the placebo group then it is reasonable to say that Rosuvastatin has some prophylactic value.

In plain English, did taking Rosuvastatin protect the participants against the onset of a cardiovascular condition such as heart disease or stroke? Did those who took Rosuvastatin have to undergo fewer bypass surgeries? Did fewer of the Rosuvastatin group die as a result of cardiovascular disease?

A cursory reading of the JUPITER trial results appears to bear out the prophylactic value of taking Rosuvastatin. The incidence of “primary end points” in the Rosuvastatin group was just slightly more than half that of the incidence in the placebo group.

In plain language, participants in the Rosuvastatin group were much less likely to be afflicted with cardiovascular disease than those on the placebo group. The actual reduction in the likelihood of getting a cardiovascular condition was a whopping 44%. Who wouldn’t want to cut their risk of getting a cardiovascular condition by 44%?

The results were considered so positive that the trial was cut short because it was felt unethical to deprive the placebo group of the benefits of Rosuvastatin. No wonder the editors of TIME magazine included the results of the JUPITER trails in their list of top 10 medical breakthroughs of 2008.

Well, maybe.

A closer look at the results of the JUPITER trials

Let’s take a closer look at the results. Below I’ve set out the first line of the table of results in what I hope is a more understandable form.

			Rosuvastatin Group		Placebo Group
1.	Number of patients		8,901		8,901
2.	Number of patients who experienced onset of cardiovascular disease		142		250
3.	Percentage of patients experiencing onset of cardiovascular disease per year		0.77		1.36
4.	“Hazard ratio” (100 * 0.77 / 1.36)		56%
5.	Confidence interval of hazard ratio		46% - 69%
6.	Probability that these results could have happened by chance		Less than 1 in 100,000

There were 8,901 patients in each group. 142 patients in the Rosuvastatin group experienced the onset of cardiovascular disease compared to 251 in the placebo group. So far so simple.

The next line is a little trickier. The trial ran for five years but not all patients were enrolled for the full term. Line 3 shows the likelihood of a patient getting cardiovascular disease per year.

What line 3 shows is that 0.77% of the patients in the Rosuvastatin group and 1.36% of the patients in the placebo group experienced the onset of cardiovascular disease.

Here is another way of understanding line 3. Taking Rosuvastatin reduced the likelihood of getting a cardiovascular condition by 0.59%. (1.36 – 0.77) per year. A 0.59% reduction sounds less dramatic than a 44% reduction. What happened to the 44%?

Well 0.59% is 44% of 1.36%. The percentage of those getting cardiovascular conditions dropped 44% (from 1.36% to 0.77%) per year.

The difference between relative risk and absolute risk

Here we see the difference between relative risk and absolute risk. The relative risk dropped 44%. But, since the risk per year of getting a cardiovascular condition was low for both groups, the absolute effect is quite small.

How many people to we need to treat to avert just one instance of cardiovascular disease?

Here is yet another way of understanding line 3.

0 .59% of 170 = 1 (approximately)

So you needed to treat 170 people with Rosuvastatin to avert a single incident of cardiovascular disease per year. In any one year only one participant in 170 benefited from taking Rosuvastatin. This is considerably less dramatic than saying you’ve cut the risk almost in half.

Of course the risks of getting a cardiovascular condition mount up over the years. The norm is to look at the risk reduction over a five year period. How many people do you have to treat with Rosuvastatin in order to avert just one onset of a cardiovascular condition?

According to the authors the answer is 25. 25 people have to pop pills for 5 years in order to avert one of them getting a cardiovascular condition.

The risks associated with statins

Cost aside, this might not matter if taking a statin was risk free. But it isn’t. Far from it. As reported in the study, 270 people in the Rosuvastatin group got type 2 diabetes versus only 216 in the placebo group. A very rough “back of the envelope” calculation indicates that this represents an increase in absolute risk of approximately 0.28% per year - about half the reduction in absolute risk of getting a cardiovascular condition.

Extrapolating we have the following result. If 50 people take Rusuvastatin for five years we can expect to prevent 2 cases of cardiovascular disease at the cost of one extra case of type 2 diabetes. Type 2 diabetes is a serious condition. It is itself associated with increased risk of cardiovascular disease and dementia. Had the trial continued the complications resulting from the additional cases of type 2 diabetes may well have tilted the balance against Rosuvastatin.

However diabetes is not the only risk facing patients who take statins. As Dr. Beatrice Golomb of the University of California, San Diego School of Medicine has documented, the adverse effects of taking statins range from muscle problems to cognitive disorders which is medicalese for “statins can effect your brain.”

Yeon-Kyun Shin, professor of biophysics at the Iowa State University has shown how statins may lessen brain function by inhibiting the brain’s ability to manufacture the cholesterol it needs. Professor Shin’s research will appear in Proceedings of the National Academy of Sciences, a leading peer reviewed scientific journal.

Are the benefits worth the risks?

All medical treatment is trade-off between benefits and risks. I have no doubt that there are many instances in which the benefits of statin treatment outweigh the risks. However, in the case of the people who participated in the JUPITER trails, the benefits appear to be slight. It takes 25 people popping pills for 5 years to avert one instance of cardiovascular disease. The risks are far from negligible.

Do the risks really outweigh the benefits?

Is there a better way?

Here is a more pertinent question. Are there better ways of averting the onset of cardiovascular disease than popping statins – ways that do not carry the risks inherent in statin usage? There are and I shall be discussing them in future posts.

In my next post I examine the position that Healthy Skepticism takes on these issues and explain why we have a difference of opinion.

Healthy Skepticism does have a comment on the JUPITER trials. You can read it here.

Keep watching this space.

Tags: astrazeneca, cardiovascular disease, crestor, healthy skepticism, heart, inflammation, jupiter, mansfield, rosuvastatin, stroke

This entry was posted on Sunday, June 14th, 2009 at 3:46 pm and is filed under cadiovascular disease. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.