Angiographic criteria for entry were two- or three-vessel disease (greater than 50 per cent obstruc- tion); left main disease was optional, and left ventricular function had to be essentially normal, with an ejection fraction of at least 50 per cent. Seven-hundred-and-sixty-eight patients were rando- mised: 395 to surgical, and 373 to medical treatment. At five years, 30 (7.6%) patients in the surgical group had died, compared with 61 (16.3%) of the medical group [9], This 53 per cent difference in overall mortality (95% confidence limits 30% and 71%) was highly significant both statistically and, more importantly, clinically.
1
—
—