The Doctor Weighs In

by Bill Bestermann, MD

In my last post, I discussed the untimely death of Wake Forest Basketball coach Skip Prosser and the relationship of vulnerable plaque to sudden cardiac death and myocardial infarction. Only 14% of heart attacks are caused by a fixed artery blockage of 70% or greater. For 70% of heart attack patients, the blockage in the coronary artery is less than 50% (non-obstructive). A non-obstructive plaque causes no symptoms and usually would not produce a positive stress test. Since the 50% blockage typically causes no symptoms, for 70% of myocardial infarction patients, the heart attack or sudden death is their first symptom.

We try to overcome this by using the Framingham risk score, assigning points for risk factors including HDL cholesterol, systolic blood pressure, age, total cholesterol, and smoking status. This is useful, and helps to identify some high-risk patients, but still we frequently miss people who go on to infarction. Our current system, based on risk scores, stress tests, coronary angiography, bypasses and stents has simply failed to identify too many patients with substantial risk.

Since the vast majority of heart attacks are not occurring at sites of fixed stenosis but rather at the site of a vulnerable plaque rupture, the question becomes-how do we identify these high-risk patients and treat them aggressively. Patients who have established atherosclerotic arterial disease at any site should be treated as if they have coronary artery disease. Arterial disease is a diffuse process and any blockage anywhere indicates that most of the arteries are involved with atherosclerotic plaque. There is a dramatic correlation between type 2 diabetes and arterial disease. The same holds true for patients with kidney damage. Both of these patient classes should be treated with the same level of aggression as the patient with established vascular disease. Patients with a high Framingham Risk Score should be aggressively managed. The risk factor management targets for these patients are lower than those we normally are aiming for. The blood pressure should be less than 130 systolic (top number). The LDL cholesterol should be less than 70. The hemoglobin A1c should be under 6.5.

There are many patients at risk who fit none of these categories and currently they are not being treated aggressively enough. Patients with strong family histories but low to intermediate risk scores are an example. Some people have intermediate risk scores but in actuality are very high risk—how do we identify them? Since the fundamental risk is the extent of plaque in the artery—specifically the amount of unstable plaque—the ideal way to identify high-risk patients would be to develop a methodology that allows us to identify patients with unstable plaque. The higher the amount of unstable plaque, the higher the risk.

The gold standard for directly examining the amount of plaque in the artery is coronary catheterization using intravascular ultrasound technology. This is an invasive technique that carries some risk and substantial expense. It is not routinely used even in patients having a heart catheterization. It is impractical for intermediate risk screening.

More studies are now available to help us understand the role of coronary artery calcium scoring. The American College of Cardiology and the American Heart Association have just published an expert consensus document on this technology. Atherosclerotic plaques are dynamic deposits in the arterial wall that go through progressive and predictable stages. Plaque “instability and rupture can be followed by calcification, perhaps to provide stability to an unstable plaque.” The authors state: “Radiographically detected coronary calcium can provide an estimate of total coronary plaque burden. The authors go on to provide a further rationale for the use of this technology: “Patients who have calcified plaque are also more likely to have non-calcified or “soft-plaque” that is prone to rupture and acute coronary thrombosis….coronary artery calcium scoring may be able to globally define a patient’s CHD (coronary heart disease) event risk by virtue of it’s strong association with total coronary atherosclerotic disease burden, as shown by correlation with pathologic specimens. Perhap even more convincing is the following:

“ Pathology studies have shown that the extent of coronary calcium within plaques tends to be related to healed plaque ruptures.” We cannot identify the vulnerable plaque but we can quantify ruptured plaque history which tells us his risk for future plaque rupture and thrombotic obstruction. We cannot identify the vulnerable plaque, but we can identify the “vulnerable patient.”

Even more impressive, when we combine the Framingham Risk Score and the Coronary Calcium Score, we have a system that is able to predict coronary risk in a very robust fashion as shown in the figure below. Any patient with a coronary calcium score over 100 should be considered to have coronary disease and should have risk factors reduced to those same aggressive targets.

Women are a special case here and for them this technology may be even more important. Women are less likely to form focal narrowings in the arteries and so they are even more likely to have an infarction with a non-obstructive plaque (narrowing less than 50%) Women tend to deposit their plaque in a concentric, symmetrical fashion up and down the artery. In fact, women with recurrent chest pain and a normal heart catheterization still have a 20% six-year risk of sudden death, myocardial infarction, stroke or congestive heart failure (WISE study). For this reason, I would not consider any woman’s cardiac workup to be complete until she had a calcium score. Too many women are told they have nothing to worry about after a normal heart catherterization. The woman with recurrent chest pain is still often high risk and in need of aggressive risk-factor management.

Here is the really amazing part. In spite of the extensive literature on the new science of risk assessment and the importance of vulnerable plaque, almost no insurance companies pay for the calcium score. In our group practice, we offer this test for $249.00. When you consider the information to be gained from the study, that seems very reasonable. This technology should be much more widely applied to identify high-risk patients and we should press the payors to allow this test in intermediate-risk patients.

This is Dr. Bill Bestermann's first post as one of the TDWI writers (A Big Welcome, Bill!).   Dr. Bestermann makes it clear that there is a long way to go, baby, before the diagnosis and treatment of heart disease in women gets to where it should be...hey! ladies, let's WISE up!  Here you go:

The evidence has become irrefutable that life-style change and medications aimed at vascular risk factors outperform bypass surgery and stenting in prevention of heart attack. Not only that, but non-invasive strategies have positive benefits on the entire vasculature that prevent stroke, nerve damage, eye damage, kidney damage and amputation.

Women are different

These differences may be even more important for females. For some time now, there has been discussion of women being different in the manifestation of vascular disease. Women are generally smaller in stature and so would have smaller vessels. They are protected by hormonal factors and experience fewer vascular events prior to the menopause. Now it seems that women really are quite different in how they deposit atherosclerotic plaque and that this difference is important in their diagnosis and treatment.

Men are more likely to develop focal stenoses in addition to more diffuse disease. These localized narrowings may interfere with flow and lead to the classic anginal symptom of chest pain relieved by rest. Women, on the other hand have more diffuse disease that does not cause a vascular narrowing as often. Therefore, their pains are more atypical, may not be related to exertion at all, and may indicate a tendency to increased vascular reactivity and spasm. Women develop plaque and their plaque burden may be quite high, but their deposition of cholesterol is concentric, symmetrical and diffuse with the result that the vessel appears “small”.

The WISE Study

The recent WISE (Women’s Ischemia Syndrome Evaluation) Study sponsored by the National Institute of Health-National Heart Lung and Blood Institute underscores these differences. Even today, the cardiac catheterization is considered the “gold standard” when testing for the severity of heart artery disease. A woman with chest pain who has a cardiac catheterization that is normal is considered “cleared” as far as coronary disease goes and told that there is some other cause such as stress, depression, or gastrointestinal disease.

In the WISE Study 673 women had cardiac catheterizations done for the evaluation of possible coronary artery disease. 45% of these women had persistent chest pain. Women with no coronary obstruction and no chest pain had a very low rate of coronary events. Women with recurrent chest pain and normal coronary arteries had a significant 20% six-years risk of developing heart attack, cardiovascular death, stroke or congestive heart failure. This reflects the fact that plaque in women is deposited in a symmetrical fashion along the length of the artery making it appear small. In this high-risk group, only 9% of women were on lipid lowering therapy and 14% on antihypertensive treatment one year after the catheterization. Obviously, for these patients medical therapy is their only hope and they are receiving the benefit of this treatment far less often than should be the case.

The Takeaway Message

The takeaway message is that the current method of evaluating coronary arteries in standard practice may give the patient a false sense of security if there is no blockage found. There is no way that a stress test or catheterization can adequatly reassure us that that patient is low risk. The woman with persistent chest pain should be evaluated with other modalities. If coronary intravascular ultrasound is available at the time of the catheterization then that is very helpful. If not, calcium scoring correlates directly with the amount of plaque in an artery. The calcium score coupled with the Framingham risk score is a practical way to get a much better idea of the risk in a stable patient

The data indicating a need to improve risk assessment and risk factor management in these patients is clear. Still the system continues to function much as it has for over three decades. This may be the best example of how actual practice lags changes in knowledge and technology. The pace of change in how we deal with these patients has been so slow that it is hard to explain. There is always a resistance to the new way and in this case perverse financial incentives aggravate the situation.

So, what should we do?

Patients (that means you) and payors (health plans, governmental agencies, and employers) are going to have to insist that they receive the benefits of the new science before we see improved care at reduced cost.

Bill Bestermann, MD

I have a file full of warranties on things I have purchased. They give me peace of mind. Sometimes, they even save me money. Now, a hospital system – The Geisinger Health System in central Pennsylvania -- is providing warranties for heart surgery. You don’t have to pay extra to get your defects related to surgery repaired if they occur within 90 days of the operation. Brilliant idea!

Only in health care, with its many perverse financial incentives, do you have a situation where a provider has the opportunity to make more money by providing shoddy care than providing good care. What do I mean by that? Let me give you an example. If you are admitted to the hospital and get an infection related to inadequate infection control. Treatment of that hospital-acquired infection prolongs your hospitalization and you use more expensive services, such as intravenous antibiotics. In a fee-for-service world, that means more revenue for the hospital. Not only does this generate extra costs, it also fails to provide an incentive to do things right the first time (and every time).

Starting in 2006, Geisinger Health System began exploring how to make elective heart bypass surgery flawless. And, to put there money where their mouth is, so to speak, they combined their quality improvement efforts with a 90 day warranty.

Doctors at Geisinger reviewed all of the steps in these elective surgeries and identified 40 as essential to achieve good outcomes. Then they created procedures to ensure these 40 steps are always followed, not matter who the surgeon or where the operation is performed. Hmmm….sounds like the checklists pilots use to make sure everything is in order before the plane takes off.

Here is the checklist:

1. Preadmission documentation:

a. Document the American College of Cardiology/American Heart Association indication for surgery

b. Screening for and consultation regarding IMI (inferior myocardial infarction)/RV

(right ventricular) involvement

c. Record treatment options and patient preferences

d. Determine and document the need for anticoagulation with warfarin – Anterior MI (myocardial infarction) or WMA (wall motion abnormality)

e. Record whether the patient is a current user of anticoagulation medications (clopidogrel or warfarin)?

f. Screen the patient for risk of stroke

g. Obtain a carotid doppler ultrasound exam (test for stroke) if the test is indicated

h. Obtain a vascular surgery consultation if indicated

i. Obtain and/or record the ejection fraction

j. Screening for need to use intra-aortic balloon pump (IABP)

k. Screening using epiaortic echo as indicated

l. Document that the patient did not take anticoagulation medications (clopidogrel/warfarin) in the 5 days before the operation

2. Operative documentation:

a. Did the patient receive the correct dosing of beta-blocker (pre-op)

b. Was there correct use of intra-aortic balloon pump (pre-op -->post-op)

c. Did the patient receive appropriate and timely pre-op antibiotics within 60 minutes of incision; if Vancomycin within 120 minutes)

d. Record any blood cardioplegia (on-pump patients)

e. Document epiaortic echo of the ascending aorta and the peer consult

f. Obtain and record intra-operative hyperglycemia screening

g. Apply correct insulin management (as indicated; per protocol)

h. Use of LIMA (left internal mammary artery) for LAD (left anterior descending)

grafting

3. Post-Operative patient documentation:

a. Anteroapical MI within prior 7 days: post-op echo

b. Monitoring for atrial fibrillation for >48 hours

c. Anticoagulation therapy (as indicated)

d. Were antibiotics administered post-op for 24-48 hours

e. Was aspirin given six hours post-op or 24 hours post-op

f. Was a beta-blocker given within 24 hours post-op

g. Was a statin administered post-op

h. Document any surgical debridement and revascularization of any sternal wound infection

i. Obtain a plastic surgery consult regarding ongoing management of sternal wound

j. Tobacco screening and counseling

4. Discharge documentation:

a. Referral to cardiac rehabilitation

b. Discharge medications (e.g., beta-blocker)

c. Discharge medication: aspirin

d. Discharge medication: statin

5. Post-Discharge documentation:

a. Is the patient correctly taking beta-blocker?

b. Is the patient correctly taking aspirin?

c. Is the patient correctly taking statin?

d. Is the patient correctly administering anticoagulant?

e. Did patient resume smoking?

f. Is the patient enrolled in cardiac rehabilitation?

Gisinger calls this new program “ProvenCare.” That is because the 40 items on their check list are supported by medical evidence. Imagine that…design protocols based on evidence and follow them to get better results.

According to a story about Geisinger in the NY Times, the system was only documenting the performance of all 40 steps 59% of the time before implementing ProvenCare. Now an operation is cancelled if any of the pre-operative measures have not been done. Recently, Geisinger’s surgical teams have had scored 100% in following the recommended steps before, during and after surgery. And it has paid off in terms of reducing complications:

• 35% of patients had any type of complication after compared with 39% before ProvenCare
• Only 16% of patients required supplemental blood products compared with 23% before
• 19% of patients were not able to be discharged directly to their homes before the program; only 9% after
• There were fewer readmissions within 30 days, fewer pulmonary complications, fewer re-operations for bleeding, fewer readmissions to the ICU, and a lower in-hospital death rate (in fact, it was 0 after implementing the program.

So how does the warranty work? Geisinger charges a fixed fee for the surgery and half of the historical costs of related care provided in the ensuing 90 days. If the patient has to be readmitted, Geisinger absorbs the costs. If they do a great job and there are no problems, they have a larger profit. Now that is aligning financial incentives with good patient outcomes.

Geisinger is developing similar approaches for other types of medical care, such as hip replacements. It will be interesting to see how far and how fast they are able to go with this new model. I hope this is not a one-shot wonder. The concept of standing behind your work by offering a warranty could prove to be a powerful driver of health care quality improvement and patient safety.

Great work, Geisinger

Pat Salber, MD

Every once in a while a medical research study is published that significantly alters clinical practice. The paper on percutaneous coronary intervention (PCI) in patients with stable coronary artery disease, published in the prestigious New England Journal of Medicine on March 26, 2007, is one such study.  The results show that PCI plus optimal medical management (lifestyle changes and drugs) are no more effective than optimal medical management alone at preventing heart attacks and death in individuals with stable coronary artery disease.

 In the NY Times’ story about the study, Steven E. Nissen, MD, President of the American College of Cardiology, describes the study as a blockbuster. And, indeed it could lead to some folks’ blocks being busted (particularly stent companies, such as Boston Scientific and the Cordis Cardiology division of Johnson and Johnson, as well as invasive cardiologists who perform these procedures).

The name of the study is COURAGE, an apt description for a study that is likely to be attacked vigorously by people who stand to lose in the high stakes arena of invasive cardiology. COURAGE stands for “Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation.”

Understanding plaque

In order to understand what these researchers were trying to determine, it is important to understand a bit about what causes heart attacks. Years ago, we used to think heart attacks were caused because one or more blood vessels supplying the heart became progressively more narrowed by atherosclerotic plaques, perhaps even closed off completely, thus, depriving the heart muscle of oxygen.

As we have learned more about the anatomy and physiology of coronary artery plaques, we have discovered that the plaques that cause heart attacks and other acute coronary syndromes, such as unstable angina, are prone to rupture, lead to clot formation, and, in that manner, block the coronary artery.

These vulnerable plaques, according to the NEJM article, tend to “have thin fibrous caps, large lipid cores, fewer smooth muscle cells, more macrophages [inflammatory cells], and less collagen” than stable plaques. They also tend to grow outward in the coronary artery wall, thus, before they rupture, they actually cause less blockage (stenos is) of the coronary artery lumen than stable plaques.

Stable plaques can cause symptoms of shortness of breath and chest pain with exertion, but they are less likely than unstable plaques to cause a heart attack or the worst outcome, sudden death.

This is an important point so I will repeat it by quoting from the article:

“…unstable coronary lesions that lead to myocardial infarction are not necessarily severely steno tic, and severely steno tic lesions are not necessarily unstable.”

The study

The study is a well designed randomized trial with enough participants, 2287 in all, to detect small differences between the two groups. 1149 patients were assigned to undergo PCI with optimal medical therapy and 1138 to receive optimal therapy alone. The primary outcome was non-fatal heart attack or death from any cause.  Participants were in the study between 2.5 to 7 years (median 4.6 years).

There were 211 of the above described primary events in the PCI group and 202 in the medical therapy group. This translated to a 4.6 year cumulative primary-event rate of 19% in the PCI group and 18.5% in the medical therapy group. There were no significant differences between the PCI group and the medical-therapy group in the composite of death, heart attack, and stroke, nor in the percent hospitalized for an acute coronary syndrome, such as unstable angina, nor in the percent that had myocardial infarctions.

The PCI group did have a greater decrease in symptoms (shortness of breath or chest pain) compared with the medical management group. But it is important to note that there was a substantial reduction in angina in both groups during follow up and, by 5 years, the percent free of angina was similar in both groups. Fewer individuals (21.1% vs 32.6%) had to undergo a subsequent revascularization for angina (chest pain) unresponsive to maximal medical therapy or when there was worsening ischemia (oxygen deprivation of heart muscle).

The critics

Advocates for PCI are going to try to find reasons why PCI should still be used to prevent heart attacks. In the NY Times article, Dr. David Kandzari of J&J’s Cordis Cardiology (stent company) points out that the individuals in the trial were largely from the Veterans Hospitals in the US and Canada and, therefore, usually get their medications for free. Yeah, so that is a reason for a $25,000 procedure instead of finding a way to get those at risk the relatively inexpensive generic medications used in this study.

Others will point out that bare metal stents were used in most of the patients in the study. Drug-eluting stents were not available until the end of the study. Perhaps, there would have been a greater difference in symptoms if the drug-eluting stents were used. Maybe, but there is no evidence that drug-eluting stents prevent deaths or heart attacks in stable patients and there is the recently described problem of late-forming clots that has led some cardiologists to switch back to bare stents.

Another important lesson from the study

I do think there is another very important take away lesson from this study. All you doctors out there should study Table 2 of this study carefully. These people were aggressively managed. This is not usual care. LDL levels were lowered from about 100 to the mid 70s after five years. Average blood pressure went from a 130/74 range to 122 to 124/70-72. HDLs increased slightly and Hemoglobin A1cs were maintained in the low 7s. There was a substantial increase in individuals following the American Heart Association diet and there was an increase in percent of people who participated in moderate physical activity.

Now, we all know that it is hard work for people with heart disease to take a bunch of medications and change their lifestyles. It is also hard work for docs and other clinicians to provide support for people as they try to make these changes. On the other hand, if we could take even a portion of the $25,000 spent on each PCI and apply it to buy support for people trying to make these changes; we probably could help folks in the general population achieve the same remarkable adherence to an optimal medical regimen. How many hours of nutritional counseling and personal training could be bought for even a fraction of that amount…week after week, for years? Not only would it be affordable in the long haul, there probably would be enough money left over to buy or subsidize the generic medications as well.

Isn’t it time we to start paying for the right stuff in healthcare?

Pat Salber, MD, MBA