… but, Is Suppression of Postoperative ST Segment Depression an Important Outcome

Over the past decade, considerable research effort has been directed at reducing perioperative cardiac morbidity in patients undergoing noncardiac surgery. Numerous strategies have been proposed, including coronary revascularization prior to noncardiac surgery, optimization of postoperative pain management and pharmacologic interventions to control hemodynamics [1-4]. In the current issue of Anesthesia & Analgesia, Beattie et al. [5] evaluate the efficacy of continuous postoperative ketorolac infusion with intravenous patient-controlled morphine analgesia after total hip or knee arthroplasty on reduction in the severity of ST segment depression. In evaluating the value of the proposed strategy, it is important to understand the etiology of perioperative myocardial infarction (MI) and cardiac death and the potential benefit of ketorolac. The etiology of a perioperative MI is often unknown. However, increases in myocardial oxygen demand or decreases in oxygen supply can lead to myocardial ischemia, which, if prolonged, may result in myocardial necrosis. This pathophysiology is supported by several perioperative studies that have demonstrated a relationship between the duration of ischemia and the occurrence of MI [6,7]. Alternatively, perioperative MI may be the result of acute coronary thrombosis. In the nonsurgical population, critical stenoses may progress to complete occlusion without morbidity, while thrombosis of a noncritical stenosis can lead to myocardial necrosis or death [8]. As early as 1938, it was shown that patients who die of cardiac causes during the perioperative period demonstrate significant coronary atherosclerosis and acute thrombosis on autopsy, which suggests that antithrombotic therapies (e.g., antiplatelet agents) have a role in the perioperative period [9,10]. As demonstrated in Beattie et al.'s study, optimal pain relief can reduce myocardial oxygen demand by decreasing heart rate. Optimal pain relief also reduces postoperative hypercoagulability. Decreased coagulability is associated with postoperative epidural analgesia; the reduction in catecholamines with these regimens may result in reduced platelet aggregability and, therefore, also decrease the rate of thrombosis [2,11]. Ketorolac is unique among the drugs presently used for postoperative pain management because it possesses a direct antiplatelet effect. Since the authors' previous study demonstrated the safety of a postoperative ketorolac infusion and the current study demonstrates a reduction in ST segment depression, such a strategy holds great promise for improving outcome in high-risk patients on two of the pathways (increased demand, thrombosis) that can lead to cardiac morbidity [12]. Yet, the question remains regarding the importance of the outcome measure, i.e., ST segment depression, used in the current study. In neither this nor previously published trials of postoperative analgesia has there been a reduction in irreversible cardiac outcomes, i.e., acute MI or cardiac death. Even in patients undergoing coronary artery bypass grafting, in whom the use of a continuous sufentanil infusion into the postoperative period was associated with reductions in the severity of ST segment depression, no difference in the incidence of perioperative MI or death was detected [13]. Although atenolol results in decreased heart rate and decreased demand, the only positive finding in the recently published randomized trial in high-risk noncardiac surgery patients was improved survival upon subsequent two-year follow-up, with no difference in cardiac morbidity during the perioperative period [3]. The low incidence of cardiac morbidity in all of these studies, including the current one (1.5% incidence of perioperative MI), precludes the statistical power to demonstrate a difference in major cardiac outcome. For example, assuming a power of 80% and an alpha value of 0.05, a population of 10,798 patients would be required to detect a 50% reduction in the incidence of MI in the current study. In the absence of a sample size sufficient to detect differences in cardiac outcomes, Beattie et al. and others have used ST segment depression as the meaningful outcome. The value of suppressing perioperative ST segment changes, both in the nonoperative and perioperative setting, is controversial [7,14]. This issue is currently of significant interest in the nonoperative population and was the basis for the National Institutes of Health-sponsored Asymptomatic Cardiac Ischemia Pilot [14]. This pilot study randomized patients with both silent and symptomatic ischemia to either pharmacologic therapy or surgical revascularization. At one year, revascularization was associated with significantly better outcome (combined incidence of death, MI, nonprotocol revascularization, and hospital admissions) than medical strategy. Importantly, silent ischemia-guided therapy was not superior to angina-guided therapy. The authors concluded that more definitive assessment of the value of suppressing silent ischemia on outcome requires a larger trial. Consequently, the proper treatment of silent myocardial ischemia remains unclear in 1997. In the perioperative period, virtually all myocardial ischemia is silent. Therefore, there are two issues related to the importance of perioperative ST segment changes as an outcome measure. The first concerns the relationship between ST segment depression and myocardial ischemia. The relationship between the ST segment depression and myocardial ischemia is a function of the prevalence of coronary artery disease in the population. The use of ST segment monitoring should be considered in the same manner as any noninvasive test, i.e., that a positive result in a low-risk population is most likely a false positive [15]. Recently, we demonstrated that ST segment changes in patients at low to moderate risk of coronary artery disease (only one risk factor without known coronary artery disease) are rare and do not correlate with ischemia on a stress test [16]. In a similar manner, the ketorolac study enrolled an unselected population of patients undergoing total hip and knee arthroplasty. Half of the patients with perioperative ST segment depression had no risk factors of coronary artery disease, which suggests that many of these changes were not due to myocardial ischemia or related to coronary artery disease. Second, if the observed ST segment depression does reflect true myocardial ischemia, then will suppression of ischemia lead to decreased myocardial necrosis? Previous studies demonstrating a relationship between duration of ST segment depression and cardiac morbidity have been interpreted to suggest that prolonged supply/demand mismatches can lead to morbidity [6,7]. An alternative interpretation is that the prolonged ST segment changes represent an evolving infarction due to an acute coronary thrombosis. This phenomenon is an event distinct from the other episodes of supply/demand-mediated myocardial ischemia. Therefore, short episodes of ST segment changes that reflect myocardial ischemia may simply be a normal manifestation of an underlying critical coronary stenosis. In contrast, the event (thrombosis) that leads to myocardial necrosis may be distinct and via a different mechanism. Treatments that reduce the former events will show a significant reduction in ST segment depression but will not lead to improved outcome. In summary, Beattie et al. evaluate a novel analgesic regimen that effectively reduced postoperative ST segment changes. However, in determining whether to incorporate any new modality or strategy into perioperative care, suppression of ST segment depression should not be assumed to be a important surrogate marker for cardiac morbidity. Larger clinical trials that have statistical power adequate to examine truly morbid outcome measures need to be performed.