Mortality Risk Conferred by Small Elevations of Creatine Kinase-MB Isoenzyme After Percutaneous Coronary Intervention

Journal of the American College of Cardiology Vol. 42, No. 8, 2003 2003 by the American College of Cardiology Foundation ISSN 0735-1097/03/$30.00 Published by Elsevier Inc. doi:10.1016/s0735-1097(03)01044-1 Mortality Risk Conferred by Small Elevations of Creatine Kinase-MB Isoenzyme After Percutaneous Coronary Intervention Interventional Cardiology John P. A. Ioannidis, MD,* Evangelia Karvouni, MD, Demosthenes G. Katritsis, MD, PHD, FACC Ioannina and Athens, Greece; and Boston, Massachusetts OBJECTIVES BACKGROUND METHODS RESULTS CONCLUSIONS The aim of this study was to assess whether small creatine kinase-mb isoenzyme (CK-MB) elevations after percutaneous coronary intervention (PCI) affect the subsequent mortality risk. Several studies have evaluated the relationship of CK-MB levels after PCI with the subsequent risk of death. While there is consensus that elevations exceeding 5 times the upper limit of normal increase mortality significantly, there is uncertainty about the exact clinical impact of smaller CK-MB elevations. We performed a meta-analysis of seven studies with CK-MB measurements and survival outcomes on 23,230 subjects who underwent PCI. Data were combined with random effects models. Mean follow-up was 6 to 34 months per study. By random effects, 19% (95% confidence interval [CI], 16% to 23%) had one- to five-fold CK-MB elevations, while only 6% (95% CI, 5% to 9%) had 5-fold elevations. Compared with subjects with normal CK-MB, there was a dose-response relationship with relative risks for death being 1.5 (95% CI, 1.2 to 1.8, no between-study heterogeneity) with one- to three-fold CK-MB elevations, 1.8 (95% CI, 1.4 to 2.4, no between-study heterogeneity) with three- to five-fold CK-MB elevations, and 3.1 (95% CI, 2.3 to 4.2, borderline between-study heterogeneity) with over five-fold CK-MB elevations (p 0.001 for all). Any increase in CK-MB after PCI is associated with a small, but statistically and clinically significant, increase in the subsequent risk of death. (J Am Coll Cardiol 2003;42:1406 11) 2003 by the American College of Cardiology Foundation Microvascular coronary embolization, either spontaneous or iatrogenic, is now a well-recognized clinical entity (1). Creatine kinase-mb isoenzyme (CK-MB) elevation is common after percutaneous coronary intervention (PCI) and, if sought, can be detected in up to 25% of cases (2) or even a larger percentage if more sensitive markers, such as troponin T or I, are used (3,4). Extensive cardiac enzyme See page 1412 release after PCI has been correlated with late mortality in several studies (5,6). Still, however, the amount of increase that is clinically relevant and the threshold level of CK-MB that has prognostic significance remain elusive (7). Post- PCI CK-MB isoenzyme increases up to 5 times the upper limit of normal have typically not been found to be statistically significant for predicting survival, and their clinical meaning has been questioned (7,8). Nevertheless, the majority of post-pci CK-MB elevations are in this From the *Clinical Trials and Evidence-Based Medicine Unit, Department of Hygiene and Epidemiology, University of Ioannina School of Medicine and the Biomedical Research Institute, Foundation for Research and Technology, Hellas Ioannina, Greece; the Division of Clinical Care Research, Tufts-New England Medical Center, Tufts University School of Medicine, Boston, Massachusetts; and the Department of Cardiology, Athens Euroclinic, Athens, Greece. Manuscript received February 26, 2003; revised manuscript received April 4, 2003, accepted April 10, 2003. range; most are actually in the range of 1 to 3 times the upper limit of normal (5). Previous studies have had inadequate power to evaluate the exact impact of such enzyme elevations on the subsequent risk of death. Thus, we performed a comprehensive meta-analysis of all pertinent studies to clarify the clinical significance of small CK-MB elevations after PCI. METHODS Eligibility and search strategy. The meta-analysis included studies reporting data on post-pci CK-MB values in relationship to the subsequent risk of death during follow-up. Abstracts were not eligible. All types of PCI were eligible including percutaneous transluminal coronary angioplasty (PTCA), stent placement, atherectomy (directional, rotational, extraction), and excimer laser. Whenever study reports included both patients with and without PCI, only the first group was used. The meta-analysis focused on the strata of one- to three-fold elevation and three- to five-fold elevation compared with normal CK-MB, but also examined data on subjects with over five-fold CK-MB elevations to address whether there is a clear dose-response relationship. Because it is unclear whether the potential prognostic effect of CK-MB elevations may change in the very long-term due to competing causes of death or may be affected by extensive missing information and censoring, we

JACC Vol. 42, No. 8, 2003 October 15, 2003:1406 11 Ioannidis et al. Small CK-MB Elevations Post-PCI 1407 Abbreviations and Acronyms CI confidence interval CK-MB creatine kinase-mb isoenzyme EPIC Evaluation of IIb/IIIa Platelet Receptor Antagonist 7E3 in Preventing Ischemic Complications trial EPILOG Evaluation in PTCA to Improve Longterm Outcome with Abciximab Glycoprotein IIb/IIIa Blockade trial IMPACT-II Integrilin (eptifibatide) to Minimize Platelet Aggregation and Coronary Thrombosis-II study PCI percutaneous coronary intervention PTCA percutaneous transluminal coronary angioplasty focused on studies with mean follow-up between three months and three years; whenever reports with different follow-up were available from the same study, we selected the data closest to six months. We excluded studies with CK data without additional information on deaths according to MB isoenzyme strata. Potentially eligible studies were carefully scrutinized for overlap of subjects. Reports providing information on a subgroup of a larger population described in another report were excluded, while, in cases of considerable partial overlap of subjects between reports, we generally retained only the largest study with available data to avoid double counting. We identified eligible studies in MEDLINE and EM- BASE (last search updated February 2003) using the keywords creatine kinase AND angioplasty or stent. We also screened bibliographies of retrieved studies and communicated with experts. Data. For each study we recorded study design, the types of PCI employed, and the number of subjects in CK-MB strata (normal, 1- to 3-fold elevation, 3- to 5-fold elevation, 5-fold elevation). The outcome of interest was mortality during the available follow-up, excluding acute events during the PCI. Deaths were recorded per stratum. Whenever exact numbers were not provided, we approximated risk ratios and 95% confidence intervals (CI) from Kaplan- Meier curves and other presented information. Two independent investigators extracted data, and discrepancies were resolved with consensus. Analysis. Proportions of the subjects in each CK-MB stratum across studies were synthesized with random effects models. We evaluated separately the risk ratio for mortality with one- to five-fold, one- to three-fold, three- to fivefold, and 5-fold elevation versus normal CK-MB. Risk ratios for mortality were estimated in each study, and between-study heterogeneity was estimated using the Q statistic (significant for p 0.10) (9). Risk ratios were then combined using the general variance method, weighting each log-transformed risk ratio by the inverse of its variance (fixed effects model) or by the inverse of the sum of its variance plus the between-study variance (random effects model) (9). In the absence of between-study heterogeneity, the two models coincide, while random effects are more appropriate when there is between-study heterogeneity. Absolute risk differences for mortality across different CK-MB strata were calculated by multiplying the random effects relative risk increase (risk ratio 1) times the observed death rates in the stratum of subjects with normal CK-MB after PCI. Typical observed death rates were based on Kaplan-Meier estimates at one and two years of follow-up in studies that provided such information. Analyses were conducted in SPSS 10.0 (SPSS Inc., Chicago, Illinois) and in Meta-Analyst (Joseph Lau, Boston, Massachusetts). The p values are two-tailed. RESULTS Eligible studies. We identified 23 potentially eligible reports (5,6,8,10 29). Fifteen were excluded up front (CKbased strata [n 4] [15 18]; entirely included in larger studies [n 6] [19 24]; partial overlap with other larger studies with available data [n 5] [25 29]). Another large study (8) provided neither separate information for all the pertinent CK-MB strata nor absolute numbers of deaths, and it largely overlapped with another study that provided detailed pertinent data for the meta-analysis. We were unable to obtain additional clarifications from the primary Table 1. Characteristics of Included Studies Study (yr) n* Age (yrs) Male (%) Procedures Involved (% of Total Number) Reported Protocol for Obtaining CK-MB Measurements Brener (2002) 3,573 64 (mean) 71 Various (not specified) At 8 h, next morning and with suspected ischemia Akkerhuis (2002) 8,838 60 (median) 69 PTCA (81%), stent (8%), DCA (6%), Variable, used peak value within 48 h other (5%) Stone (2001) 7,147 64 (mean) 70 Stent (37%), atheroablation (37%), both (17%), PTCA (9%) At baseline, 8 12 and 16 24 h; if abnormal, also every 8 12 h until normalized Kini (1999) 1,675 64 (mean) 68 Stent (29%), HRSA (25%), both At baseline, 6 8 and 16 24 h (32%), PTCA (10%), other (4%) Baim (1998) 956 58 (mean) 79 DCA (50%), PTCA (50%) At baseline, 4 6 h, and hospital discharge or 24 h Harrington (1995) 483 no data no data DCA (58%), PTCA (42%) At 12 and 24 h and with suspected ischemia Kugelmass (1994) 558 60 (mean) 82 Stent (51%), DCA (49%) At baseline and 24 h; if abnormal, repeated until peak *With available CK-MB data; composite data from five trials (PURSUIT, IMPACT-II, EPIC, EPILOG, CAPTURE). CK-MB creatine kinase-mb isoenzyme; DCA directional coronary atherectomy; HSRA high-speed rotational atherectomy; PTCA percutaneous transluminal coronary angioplasty.

1408 Ioannidis et al. JACC Vol. 42, No. 8, 2003 Small CK-MB Elevations Post-PCI October 15, 2003:1406 11 Table 2. Distribution of CK-MB Values and Follow-Up Percent With Death Rate CK-MB Increase Mean (Normal CK-MB) Follow-Up Study (yr) None 1 3 3 5 >5 (months) 1 Year (%)* 2 Years (%)* Brener (2002) 61.5 21.9 5.8 10.7 34 3.7 6.0 Akkerhuis (2002) 74.1 15.6 3.4 6.9 6 ND ND Stone (2001) 62.7 19.4 6.1 11.8 17 3.4 7.0 Kini (1999) 81.3 12.6 3.7 2.4 13 ND ND Baim (1998) 74.7 14.1 4.6 6.6 12 ND ND Harrington (1995) 69.6 11.2 11.4 7.8 12 ND ND Kugelmass (1994) 87.3 5.8 2.8 4.1 24 4.0 8.0 *Estimated risk of death based on Kaplan-Meier plots for patients who had normal CK-MB after percutaneous coronary intervention; composite data from five trials (PURSUIT, IMPACT-II, EPIC, EPILOG, CAPTURE); subjects with up to two-fold elevation are included in the None category. CK-MB creatine kinase-mb isoenzyme; ND no data available. investigators. Nevertheless, this study was included in a sensitivity analysis for the comparison of one- to five-fold CK-MB elevation versus normal controls. The seven eligible studies with complete pertinent data (5,6,10 14) totaled 23,230 subjects (Table 1). Four of the eligible studies clearly stated that they excluded patients with major complications during catheterization (6,10,11,13), while three studies (5,12,14) did not comment on whether any such patients were included. All studies included a preponderance of males with a mean or median age between 58 to 65 years. A variety of PCI types had been employed in each study (Table 1). The protocol for measuring CK-MB in each study is shown in Table 1. The seven eligible studies included 16,133 subjects with normal CK-MB, 3,981 subjects with one- to three-fold, 1,124 subjects with three- to five-fold, and 1,992 subjects with over five-fold CK-MB elevations (Table 2). The percentage of subjects with one- to five-fold CK-MB elevation (range, 8.6% to 27.7%) exceeded the percentage of subjects with larger CK-MB elevations (range, 2.4% to 11.8%) across all studies. By random effects, the overall percentage was 19% (95% CI, 16% to 23%) for one- to five-fold CK-MB elevations versus 6% (95% CI, 5% to 9%) for CK-MB elevations exceeding five-fold the upper limit of normal. Meta-analysis. There was no significant between-study heterogeneity in the risk ratio estimates in any of the contrasts of CK-MB strata (p 0.10), with the exception of the comparison between subjects with over five-fold CK-MB elevations against subjects with normal CK-MB where borderline between-study heterogeneity was seen (p 0.10). One- to five-fold CK-MB elevations conferred a significant increase in the risk of mortality with an overall risk ratio of 1.5 (95% CI, 1.3 to 1.8, p 0.001; Fig. 1A). There was a dose-response: the summary risk ratio was 1.5 for one- to three-fold elevation (95% CI, 1.2 to 1.8, p 0.001; Fig. 1B) and 1.8 for three- to five-fold elevation (95% CI, 1.4 to 2.4, p 0.001; Fig. 1C). Fixed and random effects coincided, and the results were very consistent across studies. The dose-response was also clear when subjects with over five-fold CK-MB were considered. Their risk ratio compared with normal CK-MB subjects was 3.1 (95% CI, 2.3 to 4.2, p 0.001) by random effects calculations (Fig. 1D). There was a suggestion that for this contrast the risk ratio estimates tended to be higher in three (5,10,12) of the four (5,10,12,14) studies with relatively shorter follow-up. A sensitivity analysis including Ellis et al. (8) in the oneto five-fold versus normal CK-MB analysis yielded similar results (summary risk ratio 1.6, 95% CI, 1.3 to 1.8, p 0.001). There was no significant difference in the risk ratio estimates of small versus larger studies or in early published versus recent studies in any of the considered contrasts (not shown). Based on Kaplan-Meier plots, the one-year and two-year mortality risks among patients with normal CK-MB was 3% to 4% and 6% to 8%, respectively, in studies that provided such data (Table 2). For a cohort with a death rate of 3.5% per year among subjects with normal CK-MB, the absolute increase in the mortality risk among patients with one- to three-fold, three- to five-fold, and 5-fold CK-MB elevations after PCI would be 1.7%, 2.8%, and 7.4% per year, respectively. DISCUSSION This meta-analysis based on data from over 23,000 subjects clarifies the clinical significance of small CK-MB elevations after PCI. Any CK-MB increase is associated with a potential increase in the subsequent risk of death during follow-up. One- to three-fold CK-MB elevations increase the risk of death by approximately 50%. In a step-wise fashion, the risk is increased by 80% with three- to five-fold CK-MB elevations and is tripled with over five-fold CK-MB elevations. Prior investigations had clearly stressed the adverse prognosis of subjects with CK-MB increases exceeding 5 times (or even 8 times) the upper limit of normal (6,8). However, such CK-MB elevations are on average three times less frequent than elevations in the one- to five-fold range. Thus, the impact of small CK-MB elevations on excess mortality on a population basis is not negligible when

JACC Vol. 42, No. 8, 2003 October 15, 2003:1406 11 Ioannidis et al. Small CK-MB Elevations Post-PCI 1409 Figure 1. Relative risk of death with various creatine kinase-mb isoenzyme (CK-MB) elevations. Each study is shown by the risk ratio estimate and the corresponding 95% confidence interval. Also shown is the summary estimate for all studies. (A) One- to five-fold elevation versus normal CK-MB. (B) Three- to five-fold elevation versus normal CK-MB. (C) Three- to five-fold elevation versus normal CK-MB. (D) 5-fold elevation versus normal CK-MB. compared against the impact of the more unusual high-level increases. Approximately one in five subjects undergoing PCI will have an elevation of CK-MB by one- to five-fold, and one in 15 subjects will have an even larger increase. Stent and atheroablative procedures may confer a higher risk of CK-MB release than PTCA (10,12,13), and combined procedures may increase the risk even further (6). Saphenous vein graft interventions may also have a higher risk of CK-MB release (27). In certain clinical settings, such as diabetes mellitus and states of elevated C-reactive protein or other inflammatory markers, patients are particularly prone to coronary microembolization, either spontaneous or iatrogenic (1). We could not address separately the risk of death in these subgroups according to CK-MB strata. However, it is possible that the absolute excess mortality may be even larger in high-risk subpopulations. We should acknowledge that the studies included in the meta-analysis used a considerable variety of revascularization procedures. Even with 23,000 patients, the metaanalysis is not fully powered to examine whether there are any differences on what subclinical CK-MB elevations mean for different types of PCI. However, the absence of between-study heterogeneity in the risk ratios suggests that any such differences, if present, may not be very prominent. Another issue is whether the impact of the CK-MB elevations on mortality risk remains constant over long durations of follow-up. This seems to be the case at least for two to three years with minor CK-MB elevations; CK-MB may be an index of myocardial damage that carries prognostic information in the long-term even with small increases. For high CK-MB elevations, other investigators have noted a more prominent adverse prognostic impact in the early months after PCI (6), and our data, although not definitive, are also consistent with this perspective. Finally, it would be interesting to evaluate also with large-scale studies the ability of other myocardial enzymes (3,4) to predict long-term outcomes after PCI. It has been speculated whether periprocedural embolization carries the exact same adverse prognostic implications as with spontaneous myocardial necrosis (5). Other investigators observed two- to 2.5-fold increases in the risk of death with one- to five-fold CK-MB elevations after spontaneous infarction and suggested that iatrogenic and spontaneous CK-MB elevations may have similar implications (5). Although our risk ratio estimates are somewhat smaller,

1410 Ioannidis et al. JACC Vol. 42, No. 8, 2003 Small CK-MB Elevations Post-PCI October 15, 2003:1406 11 and the CIs exclude a doubling in mortality risk with oneto five-fold CK-MB elevations, any enzyme release post- PCI does seem to affect prognosis. The level of post-pci CK-MB elevation that carries an adverse prognosis has been debated (7). Based on our findings and in concordance with the recent redefinition criteria of myocardial infarction (30), we conclude that any increase in CK-MB post-pci is associated with a small, but significant, increase in the subsequent risk of death. Given that minor elevations are far more common than more pronounced CK-MB increases, their mortality impact may be considerable in the population of patients undergoing PCI. We should acknowledge that our meta-analysis focuses on the importance of mostly asymptomatic elevations of CK-MB after PCI as contrasted to major symptomatic periprocedural myocardial infarctions. We have clearly documented an increasing long-term mortality risk with increasing levels of CK-MB elevation. Nevertheless, this risk would have to be weighted against the anticipated benefitof the PCI and should not lead to abandoning PCI, when this is clearly indicated. For example, the risk conferred from small CK-MB elevations may be negligible compared with the benefit obtained from revascularization in a patient with tight proximal left anterior descending stenosis with unstable angina and a positive stress test. Risks and benefits should be carefully weighted in each case. Moreover, we should caution that the observed association between CK-MB elevation and subsequent mortality risk does not necessarily prove causality for this relationship. The CK-MB elevation may indeed reflect direct myocardial damage in some cases. However, in other cases it may simply be a surrogate for more extensive disease or more vulnerable plaques, and the subsequent increased mortality may not be directly linked to the original PCI-related microinfarction. New strategies should be considered to try to minimize the risk of cardiac events and death after PCI. For example, use of platelet glycoprotein IIb/IIIa receptor antagonists has recently been shown to decrease the risk of death both in the short- and long-term in patients undergoing PCI (31). Reprint requests and correspondence: Dr. John P. A. Ioannidis, Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina 45110, Greece. E-mail: jioannid@ cc.uoi.gr. REFERENCES 1. Topol EJ, Yadav JS. Recognition of the importance of embolization in atherosclerotic vascular disease. Circulation 2000;101:570 80. 2. Califf RM, Abdelmeguid AE, Kuntz RE, et al. Myonecrosis after revascularization procedures. J Am Coll Cardiol 1998;31:241 51. 3. Johansen O, Brekke M, Stromme JH, et al. Myocardial damage during percutaneous transluminal coronary angioplasty as evidenced by troponin T measurements. Eur Heart J 1998;19:112 7. 4. Bonz AW, Lengenfelder B, Strotmann J, et al. Effect of additional temporary glycoprotein IIb/IIIa receptor inhibition on troponin release in elective percutaneous coronary interventions after pretreatment with aspirin and clopidogrel (TOPSTAR trial). J Am Coll Cardiol 2002; 21:40 :662 8. 5. Akkerhuis KM, Alexander JH, Tardiff BE, et al. Minor myocardial damage and prognosis: are spontaneous and percutaneous coronary intervention-related events different? Circulation 2002;105:554 6. 6. Stone GW, Mehran R, Dangas G, et al. Differential impact on survival of electrocardiographic Q-wave versus enzymatic myocardial infarction after percutaneous intervention: a device-specific analysis of 7147 patients. Circulation 2001;104:642 7. 7. Colombo A, Stankovic G. Nothing is lower than 0, and 3 is closer to 0 than to 5 medicine is not arithmetic. Eur Heart J 2002;23:840 2. 8. Ellis SG, Chew D, Chan A, et al. Death following creatine kinase-mb elevation after coronary intervention: identification of an early risk period: importance of creatine kinase-mb level, completeness of revascularization, ventricular function, and probable benefit of statin therapy. Circulation 2002;106:1205 10. 9. Pettiti D. Meta-Analysis, Decision Analysis and Cost-Effectiveness Analysis. 2nd edition. New York, NY: Oxford University Press, 1999. 10. Kini A, Kini S, Marmur JD, et al. Incidence and mechanism of creatine kinase-mb enzyme elevation after coronary intervention with different devices. Catheter Cardiovasc Intervent 1999;48:123 9. 11. Brener SJ, Lytle BW, Schneider JP, Ellis SG, Topol EJ. Association between CK-MB elevation after percutaneous or surgical revascularization and three-year mortality. J Am Coll Cardiol 2002;40:1961 7. 12. Harrington RA, Lincoff AM, Califf RM, et al. Characteristics and consequences of myocardial infarction after percutaneous coronary intervention: insights from the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT). J Am Coll Cardiol 1995;25: 1693 9. 13. Kugelmass AD, Cohen DJ, Moscucci M, et al. Elevation of the creatine kinase myocardial isoform following otherwise successful directional coronary atherectomy and stenting. Am J Cardiol 1994;74: 748 54. 14. Baim DS, Cutlip DE, Sharma SK, et al. Final results of the Balloon vs Optimal Atherectomy Trial (BOAT). Circulation 1998;97:322 31. 15. Kong TQ, Davidson CJ, Meyers SN, Tauke JT, Parker MA, Bonow RO. Prognostic implication of creatine kinase elevation following elective coronary artery interventions. JAMA 1997;277:461 6. 16. Abdelmeguid AE, Ellis SG, Sapp SK, Whitlow PL, Topol EJ. Defining the appropriate threshold of creatine kinase elevation after percutaneous coronary interventions. Am Heart J 1996;131:1097 105. 17. Abdelmeguid AE, Topol EJ, Whitlow PL, Sapp SK, Ellis SG. Significance of mild transient release of creatine kinase-mb fraction after percutaneous coronary interventions. Circulation 1996;94:1528 36. 18. Abdelmeguid AE, Whitlow PL, Sapp SK, Ellis SG, Topol EJ. Long-term outcome of transient, uncomplicated in-laboratory coronary artery closure. Circulation 1995;91:2733 41. 19. Saucedo JF, Mehran R, Dangas G, et al. Long-term clinical events following creatine kinase-myocardial band isoenzyme elevation after successful coronary stenting. J Am Coll Cardiol 2000;35: 1134 41. 20. Tardiff BE, Califf RM, Tcheng JE, et al. Clinical outcomes after detection of elevated cardiac enzymes in patients undergoing percutaneous intervention: IMPACT-II Investigators. Integrilin (eptifibatide) to Minimize Platelet Aggregation and Coronary Thrombosis-II. J Am Coll Cardiol 1999;33:88 96. 21. Simoons ML, van den Brand M, Lincoff M, et al. Minimal myocardial damage during coronary intervention is associated with impaired outcome. Eur Heart J 1999;20:1112 9. 22. Gruberg L, Mehran R, Waksman R, et al. Creatine kinase-mb fraction elevation after percutaneous coronary intervention in patients with chronic renal failure. Am J Cardiol 2001;87:1356 60. 23. Fuchs S, Kornowski R, Mehran R, et al. Prognostic value of cardiac troponin-i levels following catheter-based coronary interventions. Am J Cardiol 2000;85:1077 82. 24. Dangas G, Mehran R, Feldman D, et al. Postprocedure creatine kinase-mb elevation and baseline left ventricular dysfunction predict one-year mortality after percutaneous coronary intervention. Am J Cardiol 2002;89:586 9. 25. Lincoff AM, Tcheng JE, Califf RM, et al. Sustained suppression of ischemic complications of coronary intervention by platelet GP IIb/ IIIa blockade with abciximab: one-year outcome in the EPILOG trial:

JACC Vol. 42, No. 8, 2003 October 15, 2003:1406 11 Ioannidis et al. Small CK-MB Elevations Post-PCI 1411 Evaluation in PTCA to Improve Long-term Outcome with abciximab GP IIb/IIIa blockade. Circulation 1999;99:1951 8. 26. Narins CR, Miller DP, Califf RM, Topol EJ. The relationship between periprocedural myocardial infarction and subsequent target vessel revascularization following percutaneous coronary revascularization: insights from the EPIC trial: Evaluation of IIb/IIIa platelet receptor antagonist 7E3 in Preventing Ischemic Complications. J Am Coll Cardiol 1999;33:647 53. 27. Hong MK, Mehran R, Dangas G, et al. Creatine kinase-mb enzyme elevation following successful saphenous vein graft intervention is associated with late mortality. Circulation 1999;100:2400 5. 28. Jeremias A, Albirini A, Ziada KM, et al. Prognostic significance of creatine kinase-mb elevation after percutaneous coronary intervention in patients with chronic renal dysfunction. Am Heart J 2002;143: 1040 5. 29. Brener SJ, Ellis SG, Schneider J, Topol EJ. Frequency and long-term impact of myonecrosis after coronary stenting. Eur Heart J 2002;23: 869 76. 30. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction redefined a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36: 959 69. 31. Karvouni E, Katritsis DG, Ioannidis JP. Intravenous glycoprotein IIb/IIIa receptor antagonists reduce mortality after percutaneous coronary interventions. J Am Coll Cardiol 2003;41:26 32.