Mending a broken heart
What we’ve learned in 30 years
by Greg Curnew, MD and Chinedu Onochie, MD
Vol.17, No.01, January 2009

The past several decades have seen a dramatic decrease in the rate of death attributable to coronary heart disease. Canada’s CHD-attributable mortality rate has halved in men and fallen by more than a third in women. About half of this improvement has been attributed to reductions in dyslipidemia, hypertension and physical inactivity. The remainder of the benefit stems from improvements in post-MI prevention, heart failure management, acute coronary syndrome management and revascularization.

All of our treatments have improved, but it’s still not clear which is the best approach to revascularization in stable CHD. The options of angioplasty, stenting, bypass surgery (CABG) or medical management have been heavily studied over the years. But rapid changes in medical management, stent technology, and CABG techniques have rendered much of the data obsolete. A summary of the best evidence, from the 1970s to 2008, may help guide the primary clinician making tricky decisions on revascularization.

Medical management vs coronary artery bypass graft surgery

The Veterans Administration Cooperative Study, published in 1977, and the Coronary Artery Surgery Study (CASS), performed in the 1970s and ’80s, for years provided a basis for comparison of CABG vs medical therapy, testing the contemporary state of the art in both approaches. These trials showed no evidence of an overall mortality difference between the groups. At 10 years, CASS subjects experienced a slightly higher symptom rate of 47% in the surgical group, vs 42% in the medical group. The difference was largely attributed to the development of atherosclerotic disease in both bypass and native vessels.

Medical management vs angioplasty

The AVERT trial, published in 1999, compared medical therapy with aggressive lipid lowering to angioplasty, then becoming a truly widespread technique. Patients with stable angina and LDL cholesterol > 3.0 mmol/L were randomized to atorvastatin 80 mg per day, or percutaneous transluminal coronary angioplasty (PTCA) followed by less aggressive “usual care” lipid lowering therapy. Over 18 months’ follow-up, the rate of ischemic events was 36% lower in the atorvastatin group. This result fell short of statistical significance after adjustment for interim analyses, but aggressive lipid lowering with atorvastatin did result in a statistically significant increase in time to first ischemic event.

Angioplasty vs coronary artery bypass graft (CABG) surgery

A whole slew of studies in the later ’90s sought to measure angioplasty against the gold standard of bypass. The CABRI trial, the EAST trial, the RITA trial, the ERACI study, and the MASS study treated a lot of patients but produced no clear winner in terms of overall survival.

CABG tended to outperform angioplasty in combined endpoints that included revascularization and even nonfatal MI, but not in analyses that looked purely at death or cardiac death. Angioplasty patients were more likely to experience new or persistent angina, and much more likely to need second procedures. Nevertheless, the ERACI study, which compared costs, still found CABG to be the pricier option over three years, while RITA found no significant difference in total costs.

What conclusions can be drawn from these studies? First, angioplasty is less likely to control angina in the medium to long term. Secondly, patients receiving angioplasty are more likely to need repeat revascularization. Thirdly, these disadvantages of angioplasty don’t apparently translate into higher patient mortality or higher costs.

One other trial in the ’90s, the BARI study, showed a particular subset of patients who appeared to derive extra benefit from CABG. BARI was the largest trial of all, comprising 1,829 patients with stable angina and multivessel disease. Like the other trials, it showed no overall mortality benefit from CABG over angioplasty, though bypass again achieved lower rates of revascularization and angina.

Among particular high-risk groups, like patients with proximal LAD (left anterior descending artery) disease and left ventricular dysfunction, and those with an ejection fraction less than 50%, there was also no difference in mortality — with one exception.

That exception was patients with treated diabetes, in whom CABG showed a markedly lower mortality rate than PTCA, 19.4% vs 34.5%. There was another striking finding — in these diabetics, cardiac mortality was 2.9% when a LIMA (left internal mammary artery) graft was used, compared to 18.2% when only saphenous vein grafts were used.

Coronary artery bypass surgery vs stent therapy

The arrival of the stent posed new questions. These were addressed in a meta-analysis of data obtained from the ERACI II, SoS, ARTS, and MASS II trials, which compared CABG to stent therapy in patients with multivessel disease. After five years, there was no difference in the combined endpoint of death, stroke, myocardial infarction or repeat revascularization (16.7% stent vs 16.9% CABG, p = 0.69). Repeat revascularization rates were significantly higher in the stent group (p < 0.001). Although “major adverse cardiovascular or cerebrovascular events” were higher in the stent group (39.2% stent vs 23.0% CABG, p < 0.001), this difference was due to the higher rate of revascularization, which was counted as an adverse event. There were no differing trends observed in patient subgroups with diabetes or three-vessel disease. In short, stent angioplasty performed a lot like balloon angioplasty.

Modern medical therapy vs stent or CABG

Lastly, we’ll look at the MASS II and COURAGE trials. These trials compare medical management to revascularization in the modern medical, stent and CABG era. The MASS II trial compared the three main treatments in higher-risk patients — many had 3-vessel disease and almost all had LAD disease. The COURAGE investigators excluded several high-risk categories, but both trials ended up with the same, now predictable result: no significant difference in mortality. While COURAGE showed no significant differences in any major endpoint, MASS II showed lower risks of nonfatal MI, revascularization and refractory angina in the CABG patients. Revascularization rates were highest in stented patients, while nonfatal MI was most common in medical therapy.

Of course, not all stents are created equal. Debate has raged in recent years about the risks and benefits of the drug-eluting stent, originally hailed as a solution to high revascularization rates, but now questioned as a potential long-term thrombosis risk. The debate is more relevant south of the border, where about 80% of recently implanted stents are drug-eluting. Cost concerns have kept the proportion in Canada to about 30%.

Bare stents are prone to early complications (restenosis), while drug-eluting stents are prone to late complications (thrombosis). So research may prove less favourable to drug-eluting stents when follow-up is longer. At the moment, the clearest message is that antiplatelet therapy should last a good while after stent emplacement — at least 12 months according to the American Heart Association — whether the stent is bare or not.


The decision of whether to treat a patient medically, with stent therapy, or CABG is a complex one, but there are some clear trends in the evidence. The ACC/AHA have provided guidelines for stable angina management.

For the most part, it’s accepted that interventional therapy, usually in the form of stenting, should be performed in patients who experience insufficient symptom relief from drugs alone.

CABG is the accepted treatment — with a survival benefit — in patients with significant (> 50%) stenosis, left main, left main equivalent, or three-vessel disease with an impaired ejection fraction. CABG is also generally preferred in multivessel treated diabetics and patients with two-vessel disease involving the left anterior descending artery (proximal to the first septal perforator). CABG may be considered, too, in any patient who has received insufficient symptom relief from stent therapy. For people who don’t fit into one of the categories where surgical intervention has shown a mortality benefit, it’s generally accepted that medical management should be attempted.

Many of the studies on which current decisions are based are now old. They were conducted before we had the benefit of statins, beta-blockers, ace inhibitors, stents or internal mammary graft techniques. Many looked at a less heterogeneous population than we care for today. The COURAGE and MASS II trials sought to update our evidence for the modern medical era, but have made the decision-making process no clearer.

With a narrowing of the gap between therapies, patients must consider their own personal views regarding the relative inconvenience of post-operative surgical recovery and restrictions, post-stent repeat revascularization, and medication side effects. What is clear is that CABG and PTCA/PCI don’t treat the underlying progression of CAD but do bring most patients symptomatic relief. Over 80% of patients find less need for anti-angina agents for a limited period. About 95% of patients see improvement or complete relief of their angina immediately after surgery. About 85-90% of people remain angina-free at one to three years after surgery, and about 75% of people remain angina-free or free of major coronary events at five years after surgery.

Gregory P. Curnew, MD, FRCPC, is associate professor at McMaster University in Hamilton, ON, and director of the Coronary Care Unit at Hamilton General Hospital.

Chinedu Onochie, MD, is a resident at Hamilton General Hospital.

Postoperative CABG instructions

  • Driving is prohibited for six weeks after surgery.
  • If a patient is a passenger in a front seat equipped with shoulder harness, a pillow must be placed between the patient and the harness.
  • The patient may not lift more than 25 pounds for three months after surgery. Avoid heavy lifting and extremes of shoulder movement, i.e. as in tennis, baseball and golf, to allow for complete healing of the sternum.
  • Three months after surgery, the patient may return to work on “light” duties. At nine months the patient may resume full duties for physical work. The exercise portion of cardiac rehabilitation may begin at six weeks after surgery. There are several components to cardiac rehabilitation, including exercise, reducing risk factors and dealing with stress, anxiety and depression. The benefits of cardiac rehabilitation are seen only when this multifactorial approach is used.
  • Transient AF is seen between 5-30% of patients and for most is not a long-term problem.
  • Patients with an uncomplicated bypass usually go home about five days after surgery.
  • After bypass surgery, it’s common to start new medications for prevention, and to stop or withdraw previous medications for angina.
  • Most wound infections start to occur within 14 days post-bypass.
  • Sexual problems after a bypass surgery are common, occurring in one-half to three-quarters of patients. Both men and women may have less sex or feel less satisfied with sexual activity. A variety of factors may contribute, including side effects of drugs (such as beta blockers), depres- sion and fears about triggering a new heart attack.

subscription   |   advertising information   |   about us   |   contact us   |   privacy statement   |   legal terms of use   |   Doctors review
Oncology Exchange   |   Relay   |   Health Essentials   |   Our Voice   |   login