Statistical prediction of cardiac risk in patients who undergo vascular surgery

Since August 1982, an extensive computerized data base has been developed on all patients admitted to the Division of Vascular Surgery at the Ottawa Civic Hospital. For each patient, 180 variables are recorded, including information about preoperative risk factors and postoperative complications. Since cardiac complications are a major cause of mortality and morbidity, the data file has been used to study postoperative cardiac complications in patients who undergo arterial reconstructive operations. Between August 1982 and December 1983, 353 artery repairs were performed, excluding ruptured aneurysms. Cardiac complications developed following surgery in 56 patients. Risk factors were initially studied using contingency table analysis. Four of these factors were found to be significant: electrocardiographic evidence of previous myocardial infarction (p = 0.0003), nonspecific ST-segment changes (p = 0.0007), New York Heart Association classification of symptoms (p = 0.0003) and age (p = 0.01). A further statistical study was based upon multiple logistic regression. The authors believe that the identification of a high-risk group, using these criteria, is helpful in selecting patients for intensive preoperative investigation, including coronary arteriography.