Skip Navigation

  • PRINT  | 

Bypass Surgery Results in Lower Death Rates for Diabetic

Clinical Alert Issued to U.S. Physicians

Embargoed for Release:
September 21, 1995

Patients on drug treatment for diabetes who had blockages in two or more coronary arteries and were treated with coronary artery bypass surgery (CABG) had at 5 years a markedly lower death rate than similar patients treated with angioplasty, announced the National Heart, Lung, and Blood Institute (NHLBI) of the National Institutes of Health as it released findings from a major clinical trial and issued an alert to U.S. physicians. 

The clinical alert recommends CABG over standard angioplasty for patients on drug therapy for diabetes who have multiple coronary blockages and are first-time candidates for either bypass surgery or the artery-widening procedure called angioplasty. 

The clinical alert was prompted by an analysis of 5-year mortality data from the NHLBI-funded, international multi-center clinical trial known as BARI (Bypass Angioplasty Revascularization Investigation). The results, based on analyses of many subgroups of patients, show that in a subgroup of 353 drug-treated diabetic patients, those who were randomly assigned to have coronary angioplasty had a 35 percent death rate after 5 years, compared with a 19 percent death rate among those assigned to have CABG. By contrast, five-year mortality rates were 9 percent for the remaining 1,476 non-diabetic and non-drug-treated diabetic patients in both the bypass and angioplasty groups. 

The higher mortality rate of those with diabetes was not due to an acute complication of the procedure itself, and may be related to the less complete revascularization of the coronary arteries with angioplasty compared to CABG. 

The BARI study is the largest randomized clinical trial to compare bypass surgery and angioplasty and it is the first study to report on the outcome of these procedures among diabetic patients. 

"The findings from this study will have a major impact on the care of diabetes patients with coronary heart disease," said Dr. Claude Lenfant, NHLBI director. 

There are 14 million people with diabetes (diagnosed and undiagnosed) in the U.S. Diabetes is a risk factor for coronary heart disease, which affects more than 13 million Americans and is the number one killer in the U.S. Sixty-five percent of those with diabetes die of some form of heart or blood vessel disease. 

Both bypass surgery and angioplasty play an important role in the treatment of severe coronary artery disease. The goal of both "revascularization" procedures is to relieve or bypass blockages in coronary arteries, improving blood flow to the heart. In angioplasty, a catheter with a tiny balloon at one end is used to flatten the buildup of plaque against the artery wall. This opens up the artery and stretches it so that bloodflow is improved. In a coronary artery bypass operation, a blood vessel, usually taken from the leg or freed from the chest, is grafted onto a blocked artery, bypassing the obstruction. If more than one artery is blocked, a bypass can be done on each. The blood is rerouted around the obstruction to the heart. 

About 670,000 revascularization procedures were performed in 1993 in the U.S--362,000 angioplasties and 309,000 bypass surgeries. The average cost of an angioplasty is approximately $15,000; the average cost of a CABG is $30,000. 

The BARI study compares the long-term safety and efficacy of these two methods in patients with multivessel coronary artery disease who were eligible for either bypass or angioplasty. The primary endpoint of the study is mortality after 5 years of followup. BARI's other endpoints include repeat revascularizations, heart attacks, angina, quality of life, and total costs. 

Between 1988 and 1991, the BARI study enrolled 1,829 patients who were randomly assigned to either bypass surgery or angioplasty. When the study began, the mean patient age was 61 years old, 27 percent of the patients were women, 25 percent had diabetes (of these, 76 percent were being treated for diabetes with drugs or insulin), 60 percent had 2-vessel disease, 40 percent had 3-vessel disease, and 98 percent had angina (of these, 64 percent had unstable angina and 17 percent had severe stable angina.) 

On September 13, the BARI study's independent Data and Safety Monitoring Board (DSMB) held an urgent meeting to review the 5-year mortality data. The DSMB called for prompt public and physician notification of these results. The clinical alert has been distributed to physicians via the National Library of Medicine's Medlars Broadcast System and has been faxed to relevant profesional associations, state medical societies, and state health departments. The diabetes findings along with more results from BARI will be presented on November 16 at the annual meeting of the American Heart Association. 

The clinical alert emphasized that BARI's findings can only be applied to patients with diabetes who are on oral hypoglycemic agents or insulin, who have multivessel disease and are being considered for an initial revascularization procedure. The BARI trial did not study patients with previous bypass or angioplasty procedures. The angioplasty performed was done with the standard balloon technique. The alert advised patients with diabetes who have already had angioplasty and experience symptoms of heart disease such as angina to undergo evaluation by their physicians to determine their treatment options, including bypass surgery. People with diabetes and no symptoms of heart disease should continue to see their physician on a regular basis. 

Diabetics are approximately 2 to 4 times more likely to have heart disease than individuals without diabetes. Close physician monitoring is particularly important for diabetic patients with heart disease since they may not experience symptoms such as angina during periods of reduced blood flow to the heart. 

The NHLBI advises all diabetic patients to control other risk factors for heart disease--high blood pressure, high blood cholesterol, smoking, obesity, and physical inactivity. 

The complete text of the clinical alert will be available online ( or gopher:// 

B-roll from the press conference will be available via satellite transmission for local television news stations on September 21 from 2:00 p.m. to 2:30 p.m. Eastern. The coordinates will be: C-Band: Galaxy 3/Transponder 7/Audio 6.2 & 6.8.