Following coronary artery bypass graft (CABG) surgery, 20% to 40% of patients have been shown to have a significant increase in cardiac biomarker levels.^1,2 Markedly increased biomarker levels are associated with more postoperative complications; however, long-term clinical outcomes are not known. The prognostic importance of increased creatine kinase-myocardial band (CK-MB) levels is still being debated.^1-5

Few studies have been done on the incidence and prognostic value of an increase in CK-MB levels among patients with acute coronary syndromes (ACS) undergoing CABG surgery, despite the fact that approximately 15% to 20% of patients with ACS undergo surgery.^6-8 Increased CK-MB levels following CABG surgery may be clinically important, as approximately 50% of ACS patients have evidence of myocardial necrosis as indicated by increased serum biomarker levels. No agreement has been reached regarding the level of CK-MB that should be used to define perioperative myocardial infarction (MI), although many clinical trials include MI events after CABG surgery as end points. As a result, differing CK-MB levels are employed in different studies.^2,3,8 We combined data from 4 ACS studies to evaluate the significance of increased CK-MB levels on clinical outcomes following CABG surgery in subjects with ACS.

Subjects and methods

We used the data from the Global Use of Strategies to Open Occluded Coronary Arteries in Acute Coronary Syndromes (GUSTO-IIb), Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT), Platelet IIb/IIIa Antagonism for the Reduction of Acute Coronary Syndrome Events in a Global Organization Network (PARAGON)-A, and PARAGON-B studies.^6,9-11 The enrollment criteria and study design have been previously reported.^6,9-11 We included subjects who underwent CABG surgery during the initial hospitalization and excluded those whose CK-MB levels were not measured within 24 hours of surgery, as well as those with a CK or CK-MB level > 1 × the upper limit of normal (ULN) 24 hours before surgery. All-cause mortality over a period of 6 months was the primary end point. For statistical purposes, subjects who had at least 1 CK-MB measurement taken within 24 hours of CABG surgery were grouped according to the following peak CK-MB ratios (peak CKMB value divided by the ULN at the local institution): 0-1, > 1-3, > 3-5, > 5-10, and > 10 x ULN.

For each category of peak CK-MB ratio, we ascertained Kaplan-Meier event rates through 6 months. Using log-rank statistics, we compared unadjusted clinical event rates among the various peak CK-MB groups. We tested continuous covariates eligible for inclusion in the multivariate model for the 6-month mortality model. A continuous form of MB ratio was used in the multivariate model because there was no appropriate cutoff value for categorization of the MB ratio, and the linearity assumption of CK-MB truncated at 30 was satisfied. This method produces several benefits, including the fact that the standard error of the continuous MB ratio is smaller compared with CK-MB categories, leading to an increase of the inferential potential.

To create a model in the CABG surgery population, we used a multivariate regression model of 6-month mortality based on the PURSUIT database.¹² The potential risk profile differences across the various CK-MB groups was accounted for by the adjustment for baseline characteristics from the PURSUIT model and other applicable clinical characteristics. CK-MB was added to the Cox proportional hazards 6-month mortality model as a continuous variable defined by the peak CKMB ratio after CABG surgery.

Results

A total of 26 465 subjects with non–ST-segment elevation (NSTE) ACS were included in the 4 trials. Of these, 17.5% (n = 4626) underwent CABG surgery during the index hospitalization. We eliminated 3220 subjects because their CK-MB levels were increased in the 24 hours before CABG surgery or because CK-MB levels were not measured after the surgery. This left 4401 subjects from the total population or 1406 of all subjects undergoing CABG surgery for inclusion in our study.

Table 1. Baseline and angiographic characteristics by peak CK-MB ration category. Click to enlarge.

The baseline and angiographic characteristics by peak CK-MB ratio category are shown in Table 1. Data are also given for subjects who underwent CABG surgery but were not included in our study because of missing CK-MB information.

Table 2 shows the unadjusted clinical outcomes by peak CK-MB ratio category. The Kaplan-Meier curves for 6-month mortality with CK-MB ratio as both a continuous and dichotomous variable are shown in the Figure. We ascertained the CK-MB increase associated with a 10%, 20%, and 25% increase in relative risk (RR) of mortality. The adjusted RR of 1.10 (95% confidence interval [CI], 1.02-1.18) was associated with a 1.7-unit increase in CK-MB ratio, the adjusted RR of 1.21 (95% CI, 1.04-1.42) was associated with a 3.5-unit increase in CK-MB ratio, and the adjusted RR of 1.25 (95% CI, 1.04-1.49) was associated with a 4.0-unit increase in CK-MB ratio.

Table 2. Unadjusted clinical outcomes by peak CK-MB ration category. Click to enlarge.

The following variables were shown to be independent predictors of 6-month mortality: age (RR = 1.04; 95% CI, 1.01-1.08; heart rate (RR = 1.02; 95% CI, 1.01-1.04); peak CK-MB ratio (RR = 1.05; 95% CI, 1.01- 1.10); time to CABG surgery (RR = 1.001; 95% CI, 1.000-1.002); prior angina (RR = 3.58; 95% CI, 0.87- 14.85); signs of congestive heart failure (RR = 1.99; 95% CI, 0.91-4.35); 3-vessel coronary disease (RR = 2.26; 95% CI, 0.85-6.01); 2-vessel coronary disease (RR = 1.96; 95% CI, 0.69-5.60); enrollment infarction (RR = 0.78; 95% CI, 0.45-1.34); ST-segment depression (RR = 1.26; 95% CI, 0.73- 1.16); female sex (RR = 0.85; 95% CI, 0.54-1.32); experimental treatment (RR = 1.19; 95% CI, 0.70-2.01); and systolic blood pressure (RR = 0.10; 95% CI, 0.99-1.01). Of inpatients with the highest CK-MB increase (> 10 x ULN), only 72% were treated with aspirin, 30% with ß blockers, 14% with angiotensin-converting enzyme inhibitors, and 10% with lipid-lowering agents at hospital discharge after CABG surgery.

Figure. Kaplan-Meier curves for (A) continuous unadjusted relationship between peak creatine kinase-mycordial band (CK-MB) as a multiple of the upper limit of normal (ULN) and 6-month mortality; and (B) unadjusted mortality after coronary artery bypass graft (CABG) surgery up to 6 months. (Adapted with permission from Mahaffey KW, Roe MT, Kilaru R, et al. Creatine kinase-MB elevation after coronary artery bypass grafting surgery in patients with non-ST-segment elevation acute coronary syndromes predict worse outcomes: results from four large clinical trials. Eur Heart J. 2007;28[4]:425-432.)

Discussion

Results of our study indicate that among subjects with NSTE ACS, there is a definite association between elevated CK-MB levels following CABG surgery and an increased risk of mortality over a 6-month period. We used the largest number of subjects ever assessed for this association in our analysis, and we also included a higher-risk population, which had not been fully examined previously. The model we used was wide-ranging and included clinical characteristics, medications, angiographic data, and peak CK-MB ratios as categorical and continuous variables; thus, the associations revealed appear to be valid. Our results concur with and add to the findings of prior studies.

Because a large number of patients with increased perioperative CK-MB levels have an elevated risk of longterm negative outcomes, greater myocardial protection should be supplied during ischemia and reperfusion throughout CABG surgery. Potential benefits have recently been shown with complement inhibition² and off-pump surgery.¹³ Long-term treatment with evidence-based therapies may also help to improve outcomes in this group of patients, even with revascularization.

Despite the fact that the relationship between worse outcomes and perioperative infarctions based on an increase in CK-MB levels has been clearly shown in the current and prior studies, controversy exists because no specific mechanism has been identified. Nearly half of post-CABG surgery patients have shown indications of MI in several small studies, as exhibited by hyperenhancement on magnetic resonance imaging (MRI) scans taken within 6 days of surgery.^14,15 Infarct size has been shown to be significantly related to the extent of CK-MB increase, and instead of patchy necrosis, more than 50% of patients have shown indications of focal or transmural endocardial injury.¹⁶ Results from these studies indicate that isolated myocardial necrosis is contributory and that an increase in CK-MB level following CABG surgery is probably not strictly the result of cardiac manipulation or incomplete cardioplegia.

Because of an improved understanding of the significance of increased levels of cardiac biomarkers, the definition of MI has changed over the years. Although the American College of Cardiology/European Society of Cardiology does not offer clear recommendations and thresholds for defining MI following CABG surgery, it does acknowledge the importance of these events.

The results of our study, coupled with results from MRI studies, have shown that elevated CK-MB levels are associated with larger areas of myocardial necrosis and, therefore, poorer long-term outcomes. These findings are similar to those shown for patients with NSTE ACS¹⁷ and for patients with NSTE ACS undergoing percutaneous coronary intervention.¹⁸ These studies, however, have not proven a cause-and-effect relationship.

Conclusions

Results of our study showed that increased CK-MB levels are commonly found among patients with NSTE ACS following CABG surgery, and they correlate with a higher rate of mortality. These findings highlight the importance of measuring CKMB levels following CABG surgery in patients with NSTE ACS to identify perioperative MI events. They also underscore the importance of this end point in clinical trials. The precise threshold of CK-MB elevation for use in defining clinically important MI is not yet clear, and although our data suggest that any increase in MB elevation is prognostically important, more research is needed to identify a validated cutpoint and to assess methods for decreasing the long-term outcomes of intraoperative events.

Acknowledgment

The author would like to acknowledge the editorial assistance of Amanda McMillan.

References