Selection of the candidate for myocardial revascularization; a profile of high risk based on multivariate analysis

A survey of 60 patients who died from cardiac related causes after vein or artery bypass operations alone (1967 to 1973) was made with respect to 26 clinical, angiographic, and operative variables. These factors were compared with identical characteristics of 1,188 survivors operated upon in 1973. Through discriminant analysis, the various characteristics, isolated or multiple in any combination, have been converted into risk related to operative death. The distinctive features of the mortality group were vastly different from those in the surviving group. Ten patients (16.67 per cent) of the mortality group were in the ninety-ninth percentile of risk, whereas these factors or variables of similar weight produced an equivalent risk of only 0.34 per cent of the survivors; thus, operative death in these circumstances could be predicted with an estimated 98.0 per cent assurance. Each of 6 patients with mortality risks above 0.99999 had (1) marked cardimegaly, (2) uncompensated congestive heart failure (CHF), (3) triple vessel coronary artery disease and/or obstruction of the left main coronary artery, (4) generalized impairment of left ventricular contraction or segmental left ventricular scar, and (5) evelated left ventricular end-diastolic pressure. As a single factor, congestive heart failure (CHF) exerted the most influence on the probability of dying. A new and more desctiptive statistical interpretation of the factors presumed to affect risk is presented.