Objectives: This study was undertaken to test the hypothesis that the risk of perioperative cardiac events is not simply determined by the presence of myocardium at risk, but is directly related to the extent of myocardium at risk as reflected in thallium-201 myocardial imaging.
Background: The risk of perioperative cardiac events in patients undergoing noncardiac surgery has been related to the presence of transient defects on dipyridamole thallium-201 myocardial imaging, reflecting jeopardized viable myocardium.
Methods: The study cohort consisted of 231 consecutive patients who underwent noncardiac surgery and had a preoperative dipyridamole thallium-201 imaging study. Patients with vascular reconstruction or bypass constituted the largest surgical subgroup (n = 140). For thallium-201 imaging data, each of three planar projections was divided into three segments (total nine segments) and each segment was interpreted as normal or showing a transient or fixed defect. The ability of clinical and thallium-201 imaging data to predict perioperative cardiac events was compared with stepwise multivariate logistic regression analysis.
Results: Perioperative cardiac events occurred in 19 patients, including 5 with cardiac death, 7 with nonfatal myocardial infarction and 7 with unstable angina. For cardiac death or nonfatal myocardial infarction, the only significant multivariate predictors were the number of myocardial segments with transient thallium-201 defects (p < 0.0005) and a history of diabetes mellitus (p < 0.05). For all cardiac events, the only significant multivariate predictors were the number of myocardial segments with transient defects (p < 0.0001), diabetes mellitus (p < 0.05) and calcium channel blocker use (p < 0.05).
Conclusions: The probability of important cardiac events in patients undergoing noncardiac surgery is best predicted by the extent of myocardium at risk as reflected on thallium-201 myocardial perfusion imaging. A history of diabetes mellitus also has a significant influence on perioperative risk.