German Angioplasty vs. Bypass surgery Investigation - GABI


CABG vs. PTCA for mortality in multivessel coronary disease.


Whether CABG or PTCA is more effective in the treatment of symptomatic (angina pectoris), multivessel CAD, one year after revascularization.

Study Design

Study Design:

Patients Screened: 8,981
Patients Enrolled: 359
NYHA Class: not given
Mean Follow Up: not given
Mean Patient Age: not given
Female: CABG = 20%; PTCA = 21%
Mean Ejection Fraction: not given

Patient Populations:

Symptomatic multivessel coronary disease (Canadian Cardiovascular Society [CCS] class >II)
Stenosis >70% diameter
Revascularization of at least two major coronary arteries supplying different myocardial regions (the left anterior descending, left circumflex, and right coronary arteries) had to be clinically necessary and technically feasible, based on clinical and angiographic criteria


Totally occluded vessels (Thrombolysis in Myocardial Infarction [TIMI] grade 0)
Lesions of the left main coronary artery (stenosis >30% in diameter)
Approximately 50% of the left ventricular circumference (that might be compromised should abrupt closure of one of the target vessels occur)

Morphologic criteria:
long lesions (>2 cm)
diffuse peripheral coronary disease

Myocardial infarction (MI) during the previous four weeks
Previous CABG or PTCA

Primary Endpoints:

Presence of symptoms (angina pectoris CCS class

Secondary Endpoints:

Incidence of major cardiovascular events (MI or death), procedure-related complications, and the rate of further interventions

Drug/Procedures Used:


Concomitant Medications:

Calcium antagonists
Antianginal medication

Principal Findings:

Both treatments were highly effective in relieving angina:
At 3-month follow-up, freedom from angina (CCS class At 12-month follow up, the difference in relieving angina disappeared (74% for CABG; 71% for PTCA). Class iii or iv angina was present in only 7% of those randomly assigned to CABG and 8% of those randomly assigned to PTCA.

Symptomatic relief was accompanied by improved capacity to exercise:
Median workload increased from 350 to 475 watt-minutes for CABG and from 375 to 450 watt-minutes for PTCA.
Median change in workload was 75 watt-minutes after CABG and 100 watt-minutes after PTCA.

Use of antianginal drugs was reduced in both treatment groups:
At 12-month follow up, 22% of the CABG group and 12% of the PTCA group were not using any antianginal medication

There was a greater reduction in the use of nitrates and calcium antagonists in the CABG group at 6 and 12 months.
After discharge from the hospital, 71 patients in the PTCA group needed more interventions (45 underwent repeated PTCA; 26 underwent CABG). In contrast, only 6 in the CABG group had subsequent PTCA; none had repeated CABG.

13% of the vein grafts occluded and 7% of the internal-thoracic-artery anastomoses did not function at 6 months in the CABG group
16% of the revascularized vessels were occluded or markedly restenosed (> 70%)

Six-month follow-up angiograms were available for 102 CABG patients and 117 PTCA patients. There was total occlusion of 94 native arteries (36.9%) in the CABG group and of six arteries (2.5%) in the PTCA group (P << 0.001). The rate of occluded native vessels did not correlate significantly with the severity of the lesion before bypass surgery. In the CABG group 31 bypass grafts (12.2%) were found to be occluded at the 6 month follow-up examination (29/225 vein grafts [12.9%]; 2/30 mammary artery grafts [6.7%] There were comparable rates of high-grade lesions in the main pathways of both treatment groups.


CABG and PTCA are equally effective in relieving angina, and had equal angiographic results on native vessels at 6 months.


1. N Engl J Med 1994;331:1037-43. Final results
2. Eur Heart J 1996;17:1192-8. Angiographic follow-up

Keywords: Workload, Nitrates, Mammary Arteries, Coronary Disease, Constriction, Pathologic, Coronary Vessels, Coronary Artery Bypass, Angioplasty, Balloon, Coronary, Calcium

< Back to Listings