- Email: [email protected]
Prognostic value of cardiac troponin-I levels following catheter-based coronary interventions
Prognostic Value of Cardiac Troponin-I Levels Following Catheter-Based Coronary Interventions Shmuel Fuchs, MD, Ran Kornowski, MD, Roxana Mehran, MD, Alexandra J. Lansky, Lowell F. Satler, MD, Augusto D. Pichard, MD, Kenneth M. Kent, MD, Chester E. Clark, MD, Gregg W. Stone, MD, and Martin B. Leon, MD
MD,
This study has examined the prognostic significance of troponin-I (Tn-I) levels after catheter-based coronary interventions in coronary arteries and saphenous vein grafts lesions. Tn-I and creatine kinase-MB (CK-MB) fraction levels were measured at 6 and 18 to 24 hours after catheter-based coronary intervention in 1,129 consecutive patients with normal preintervention plasma levels of Tn-I, and CK-MB levels below the cutoff for myocardial infarction. Patients were stratified according to maximal postangioplasty Tn-I levels. Group I (n ⴝ 784) had no elevated Tn-I (<0.15 ng/ml), group II (n ⴝ 170) had Tn-I at 0.15 to 0.45 ng/ml, and group III (n ⴝ 175) had Tn-I elevation >0.45 ng/ml. Major in-hospital complications (death, Q-wave infarction, and emergent coronary bypass grafting) and out-of-hospital intermediate-term (8 months) outcomes were compared between the 3 groups. Tn-I elevation >0.45 ng/ml was associated with increased risk of mortality (group III, 1.6%; group II, 0.6%; and group I, 0.1%; p ⴝ 0.019) and major in-hospital complications (3.2%, 1.7%, and 0.5%;
p ⴝ 0.004). There was no difference in death (1.8%, 3.2%, and 2.4%; p ⴝ 0.74), Q-wave infarction (0.6%, 0%, and 0.3%; p ⴝ 0.66), or target lesion revascularization (10.1%, 9.0%, and 9.3%; p ⴝ 0.86) between the 3 groups at follow-up. Cardiac event-free survival was similar between groups (p ⴝ 0.3). By multivariate analysis, Tn-I >0.45 ng/ml was an independent predictor for major in-hospital complications (odds ratio 2.1, 95% confidence interval 1.2 to 3.9, p ⴝ 0.01). The degree of risk was also associated with the conjoint elevation of Tn-I and CK-MB levels (odds ratio 1.1, 95% confidence interval 1.02 to 1.2, p ⴝ 0.01). We conclude that Tn-I levels >3 times the normal limit and conjoint elevation of Tn-I and CK-MB levels after coronary angioplasty are associated with increased risk of major in-hospital complications, but have no incremental risk of adverse intermediate-term (8 months) clinical outcomes. 䊚2000 by Excerpta Medica, Inc. (Am J Cardiol 2000;85:1077–1082)
ardiac troponin-I (Tn-I) is a sensitive and specific marker for detecting myocardial damage. C Previous studies have found Tn-I elevation to predict
Accordingly, we studied the prognostic significance of Tn-I levels after catheter-based coronary interventions in a large consecutive patient population.
1– 4
short- and long-term adverse clinical outcomes in acute coronary syndromes, including unstable angina and acute myocardial infarction.4 – 6 Elevation of cardiac troponins was detected in 30% to 44% of patients undergoing catheter-based coronary interventions.7–9 This rate may be higher than creatine kinase (CK) and creatine kinase-MB (CK-MB) elevation, which has been reported in 5% to 26% of patients undergoing percutaneous transluminal coronary angioplasty.7–10 Myocardial necrosis after coronary angioplasty, as determined by CK-MB elevation, is associated with up to a 30% increased risk for cardiac death and myocardial infarction during 3-year follow-up.10 However, the prognostic importance of myocardial damage after percutaneous coronary intervention stratified by Tn-I levels has not been established. From the Cardiovascular Research Foundation, The Washington Hospital Center, Washington, DC. This study was supported by a grant from the Cardiology Research Foundation, The Washington Cardiology Center, Washington, DC. Manuscript received August 3, 1999; revised manuscript received and accepted November 18, 1999. Address for reprints: Ran Kornowski, MD, Cardiovascular Research Foundation, Washington Hospital Center, 110 Irving St. NW, Suite 4B-1, Washington, DC 20010. E-mail: [email protected]. ©2000 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 85 May 1, 2000
METHODS
Study population: The patient cohort includes a consecutive series of 1,129 patients who underwent percutaneous coronary interventions between July 1, 1996, and March 31, 1998, in whom routine serial postintervention Tn-I was measured. Patients were divided into 3 groups according to maximal postintervention Tn-I levels. Group I (784 patients) had no elevated troponin (Tn-I ⬍ 0.15 ng/ml), group II (170 patients) had intermediate Tn-I levels between 1 to 3 times the upper limit of normal (0.15 to 0.45 ng/ml), and group III (175 patients) had ⬎3 times the upper normal limit (⬎0.45 ng/ml). Patients were excluded if preintervention serum markers were indicative for myocardial infarction before admission (Tn-I ⬎0.15 ng/ml or CK-MB ⬎8 ng/ml). Overall, 138 patients (14.7%) with a diagnosis of unstable angina who underwent percutaneous coronary interventions were excluded because of elevated preprocedural Tn-I levels. Tn-I and CK-MB measurements: Blood samples were routinely obtained from all patients at 6 and 18 to 24 hours after the procedure. If Tn-I or CK-MB levels 0002-9149/00/$–see front matter PII S0002-9149(00)00699-8
1077
Out-of-hospital clinical outcomes for up to 8 months of follow-up were Tn-I Level (patients) obtained by serial telephone interviews by research nurses. All clinical ⬍ ⫻1 ⫻1–3 ⬎ ⫻3 events (death, Q-wave infarction), Characteristics (n ⫽ 784) (n ⫽ 170) (n ⫽ 175) p Value target lesion revascularization, or Mean age (yrs) 64 ⫾ 11 65 ⫾ 12 66 ⫾ 11 0.049 any cardiac event (death, Q-wave inMale gender (%) 69.0 67.1 70.3 0.80 farction, coronary angioplasty, or byUnstable angina (%) 69.6 74.9 73.0 0.30 Systemic hypertension (%) 62.5 62.9 68.6 0.29 pass surgery) were adjudicated by Diabetes mellitus (%) 28.8 28.0 28.3 0.98 accompanying source documentation Hypercholesterolemia (%) 74 74.9 73.4 0.96 reviewed by research nurses. In adPrior myocardial infarction (%) 47.0 48.4 47.5 0.94 dition to target lesion revascularizaPrior bypass surgery (%) 39.5 40.2 48.1 0.095 Prior angioplasty (%) 51.4 48.0 46.5 0.40 tion, repeat revascularization is reEjection fraction (%) 48 ⫾ 13 45 ⫾ 13 46 ⫾ 13 0.10 ported for each patient (as any repeat revascularization) and includes all target lesion and vessels for single were elevated, serial measurements every 8 hours and multiple vessel interventions. The diagnosis of were performed and the peak level was recorded. Tn-I Q-wave infarction during follow-up was based on levels were measured by paramagnetic-particle, hospitalization records and documented discharge chemiluminescent immunoenzymatic assay (Beck- summaries with clinical diagnosis of infarction with man, Coulter Inc., California). With use of this assay, CK-MB increase ⱖ3 times the normal value and the the sample can be accurately measured within the appearance of new pathologic Q waves on the elecreportable range of the lower limit of detection and the trocardiogram. highest calibrated value (0.03 to 50 ng/ml). The upper Statistics: Continuous variables are presented as limit of the 95% nonparametric range for a presum- mean ⫾ 1 SD. Categorical data are presented as perably healthy population is below the mean minimum cent frequency and compared between groups using detectable concentration of the assay (⬍0.03 ng/ml) chi-square statistics. Major in-hospital complications and the diagnostic cutoff value for myocardial infarc- and late clinical events were analyzed as hierarchical tion based on receiver-operating characteristic plots is end points. Multivariate analysis was performed by ⬎0.15 ng/ml.11 This diagnostic cutoff has been used SAS Logistic Regression Statistics (Cary, North Caroroutinely in our laboratory as the upper normal Tn-I lina). Survival curves were calculated and displayed value. CK-MB levels were determined by a 2-site using the SAS LIFETEST procedure. Wilcoxon staimmunoenzymatic (“sandwich”) assay (Beckman). tistics were used for survival comparison between Using this assay, the sample can be accurately mea- groups. The means of continuous values were comsured within the reportable range of the lower limit of pared using the unpaired Student’s t test. Regression detection and the highest calibrated value (0.3 to 300 analysis was used to correlate between Tn-I and ng/ml). The upper limit of the 95% nonparametric CK-MB levels. An interaction test was performed range for a healthy population is 0.3 to 4.0 ng/ml. A between maximal postintervention CK-MB and Tn-I diagnostic cutoff of 4.0 ng/ml is considered the upper levels. To correct for differences in scales between normal value. Accordingly, the upper normal values peak Tn-I (0 to 50 ng/ml) and peak CK-MB (0 to 500 used in the present study were 0.15 and 4.0 ng/ml for ng/ml), a logarithmic transformation was performed. Tn-I and CK-MB, respectively. A p value ⬍0.05 was considered statistically signifiPeriprocedural and follow-up data: Baseline demo- cant. graphics and in-hospital complications were confirmed by independent hospital chart review. All patients underwent pre- and postintervention 12-lead RESULTS Baseline demographics: Table I lists the baseline electrocardiography. The diagnosis of Q-wave myocardial infarction was based on CK-MB ⱖ3 times the characteristics of treated patients. Patients with ⬎3 normal values, with appearance of a new pathologic Q times Tn-I levels were slightly older. Otherwise pawave on the postintervention surface electrocardio- tients’ demographics were similar among the 3 gram. The diagnosis of non–Q-wave infarction was groups. Tn-I and CK-MB measurements: Baseline and postbased on CK-MB elevation ⱖ5 times the normal values, in the absence of new pathologic Q waves on procedural Tn-I and CK-MB levels were available in the surface electrocardiogram. The diagnosis of in- all patients. Any postprocedural elevation of Tn-I hospital recurrent ischemia was based on symptoms (⬎0.15 ng/ml) occurred in 30.6% of patients, whereas associated with transient ST-segment deviation. An- CK-MB (⬎4 ng/ml) occurred in 40.8% of patients. giographic success was defined as ⬍50% residual Tn-I levels ⬎3 times the upper limit of normal were diameter stenosis without compromised anterograde elevated in 15.4%, whereas CK-MB levels ⬎3 times flow. Clinical success was defined as angiographic the upper limit of normal were elevated in 22.5% of success without in-hospital complications (death, Q- patients. However, in 19% of the patients with Tn-I wave infarction, emergent coronary bypass surgery, or levels ⬎3 times the upper normal limit, the CK-MB ischemia-driven repeat coronary angioplasty). level was ⬍3 times upper normal values. TABLE I Baseline Characteristics of the Study Population
1078 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
MAY 1, 2000
tients (90%) with 1 to 3 times normal Tn-I, and in 152 of 175 patients Tn-I Level (lesions) (87%) with ⬎3 times upper normal Tn-I levels. All clinical outcomes ⬍ ⫻1 ⫻1–3 ⬎ ⫻3 were similar among the 3 groups (n ⫽ 1,562) (n ⫽ 370) (n ⫽ 436) p Value (Table IV). Actuarial event-free surTarget coronary artery vival curves for any cardiac event up Right (%) 31.9 30.3 23.1 0.002 to 8 months are shown in Figure 1. Left anterior descending (%) 31.0 33.5 33.6 0.43 Left circumflex (%) 26.1 23.8 25.2 0.65 CK-MB measurements: In a sepaSaphenous vein graft (%) 11.0 12.4 18.1 0.001 rate analysis, patients were stratified Lesion location according to maximal postangioOstial (%) 10.4 7.8 10.2 0.33 plasty CK-MB levels. Group I (n ⫽ Proximal (%) 33.1 37.8 34.0 0.22 Mid (%) 39.4 36.5 38.9 0.58 668) had no CK-MB elevation (⬍4.0 Distal (%) 15.2 16.8 14.8 0.70 ng/ml), group II (n ⫽ 204) had Lesion % diameter stenosis CK-MB levels between 1 to 3 times Preprocedure (%) 68 ⫾ 14 63 ⫾ 9 63 ⫾ 16 0.67 the upper limit (4.0 to 12.0 ng/ml), Postprocedure (final) (%) 15 ⫾ 9 11 ⫾ 17 17 ⫾ 12 0.66 and group III (n ⫽ 257) had CK-MB levels ⬎3 times the upper normal level (⬎12.0 ng/ml). Patients in TABLE III Interventional Procedures group III had a trend for higher inTn-I Level (lesions) hospital mortality (1.1%) than those in groups II (0.5%) and I (0.1%), p ⫽ ⬍ ⫻1 ⫻1–3 ⬎ ⫻3 0.08. Major in-hospital complication (n ⫽ 1,562) (n ⫽ 370) (n ⫽ 436) p Value occurred more frequently in group Procedure type III (2.5%) than in groups II (0.9%) Balloon angioplasty (%) 96.1 98.4 96.1 0.076 and I (0.6%), p ⫽ 0.023. Overall Rotational atherectomy (%) 20.5 21.1 16.5 0.14 Eximer laser angioplasty (%) 6.9 10.3 12.4 0.001 clinical success was lower in group Directional atherectomy (%) 1.5 2.2 1.4 0.59 III (90.5%) than in patients in groups Stent (%) 73.1 74.1 70.9 0.61 II (96.2%) and I (97.4%), p ⫽ 0.001. Intravascular ultrasound (%) 75 75 70 0.18 At 8-month follow-up, death (1.2%, Procedural complications Procedural dissection (%) 2.3 1.1 5.3 0.001 4.0%, and 2.3%, p ⫽ 0.14), Q-wave Abrupt closure (%) 0.2 0.0 2.1 ⬍0.001 infarction (3.2%, 3.0%, and 2.3%, No reflow (%) 0.0 0.0 0.7 0.01 p ⫽ 0.70), or any repeat revascularization (19.1%, 18.7%, and 14.5%, p ⫽ 0.15) were similar between the 3 Lesion location and characteristics: Lesion location groups. Similar event-free survival was noted in paand characteristics are described in Table II. Patients tients with ⬎3 times the normal level (24.2%) comin the highest Tn-I level group underwent interven- pared with patients with intermediate (21.5%) and tions to the right coronary artery less often, and inter- normal (17.3%) CK-MB levels (p ⫽ 0.063). ventions to a vein graft more often. The locations of Multivariate analysis: Logistic regression analysis the lesions within the target artery and the pre- and was used to identify independent predictors of major postprocedural percent diameter stenosis were similar in-hospital complications (death, Q-wave infarction, between the 3 groups. and emergent bypass surgery). Variables included in Angiographic and procedural results: Procedural the model were age, gender, unstable angina, history type and complications are summarized in Table III. of coronary angioplasty or bypass surgery, diabetes Notably, there was no difference in the rate of stent mellitus, left ventricular ejection fraction, vein graft implantation between groups. Procedural complicalesion, abrupt closure, and CK-MB and Tn-I levels ⬎3 tions were more frequent in the group with the highest times the upper normal limit. Tn-I level ⬎3 times the Tn-I levels. Overall angiographic (98.1%) and clinical upper normal limit was found to be associated with success rates (95.6%) were high. Patients in the highmajor in-hospital complications (odds ratio 2.1, 95% est Tn-I level group had higher in-hospital mortality, Q-wave, and non–Q-wave infarction, repeat lesion confidence intervals 1.2 to 3.9, p ⫽ 0.01). We also angioplasty, and overall major in-hospital complica- found an interaction between the log of maximal Tn-I tions (Table IV). Patients in the highest Tn-I level and CK-MB levels (chi-square 6.36, p ⫽ 0.01). When group received abciximab (11.6%) more often than the interaction term (log [CK-MBmax ⫹ 1] ⫻ log patients with intermediate (4.6%) or normal (4.0%) [Tn-Imax ⫹ 0.01]) was entered into the multivariate Tn-I levels, p ⫽ 0.001. Abciximab was used mostly model, it was found to be an independent predictor for on a “rescue” basis (82.3%, p ⫽ 0.12 between groups major in-hospital complications, whereas CK-MB ⬎3 times the normal limit (⬎12 ng/ml) was only of marfor prophylactic vs rescue administration). Intermediate-term outcomes: Out-of-hospital clini- ginal significance in this model (Table V). In a sepacal follow-up at 8 months was available in 478 of 542 rate logistic regression analysis, unstable angina and patients (88%) with normal Tn-I, in 153 of 170 pa- diabetes mellitus were found to be independent preTABLE II Location Data and Angiographic Results of Treated Lesions
CORONARY ARTERY DISEASE/TROPONIN I AFTER CORONARY ANGIOPLASTY
1079
survival (76.2%, 80.8%, and 80.5%; p ⫽ 0.83) were similar between the 3 groups.
TABLE IV In-Hospital and Eight-Month Events in the Studied Patients Tn-I Level (patients)
In-hospital results Angiographic success (%) Clinical success (%) Overall major complications (%) Death (%) Q-wave infarction (%) Emergent coronary bypass (%) Non–Q-wave infarction (%) CK-MB ⬎3 times normal (%) Repeat lesion angioplasty (%) Follow-up events (8 mo) Death (%) Q-wave infarction (%) Target lesion revascularization (%) Any repeat revascularization (%) Any cardiac event (%)
⬍ ⫻1 (n ⫽ 784)
⫻1–3 (n ⫽ 170)
⬎ ⫻3 (n ⫽ 175)
98.2 97.7 0.5 0.1 0.0 0.4 1.4 6.1 0.4
98.1 94.3 1.7 0.6 1.7 0.0 22.3 48.6 3.4
97.4 87.8 3.2 1.6 0.5 1.1 76.2 82.0 6.9
2.4 0.3 9.3 13.8 15.5
3.2 0.0 9.0 16.4 19.0
dictors for out-of-hospital composite cardiac events (Table V). Coronary artery and saphenous vein graft lesions:
Clinical outcomes were compared between patients with coronary artery lesions and normal Tn-I levels (n ⫽ 676), and patients with 1 to 3 times (n ⫽ 142) and ⬎3 times upper normal Tn-I levels (n ⫽ 136). Clinical success rate was lower in patients with the highest Tn-I level (88.3%) than in patients with intermediate (94.9%) and normal (97.4%) Tn-I levels, p ⫽ 0.001. In-hospital repeat revascularization was higher in patients with the highest Tn-I level (7.8%) than in patients with intermediate (3.2%) and normal (0.38%) Tn-I levels, p ⫽ 0.001, whereas major in-hospital complications were similar between the 3 groups (1.9%, 1.3%, and 0.5%; p ⫽ 0.16). At 8-month follow-up, death (2.2%, 2.8%, and 2.2%; p ⫽ 0.87), Q-wave infarction (0%, 0%, and 0.3%; p ⫽ 1.0), target lesion revascularization (9.1%, 8.5%, and 9.2%, p ⫽ 0.94), and out-of-hospital event-free survival (82.2%, 81.7%, and 84.2%; p ⫽ 0.68) were also similar between the 3 groups. Similar subgroup analysis was performed in patients with saphenous vein graft lesions and normal Tn-I values (n ⫽ 106), in those with 1 to 3 times upper normal values (n ⫽ 26), and in patients with ⬎3 times upper normal Tn-I levels (n ⫽ 42). Patients with the highest Tn-I level had significantly lower clinical success (89.4%) than patients with intermediate (93.1%) and normal (99.2%) Tn-I levels, p ⫽ 0.001. In-hospital repeat revascularization rate was similar between the 3 groups (2.1%, vs 3.4%, and 0.8%; p ⫽ 0.55). Major in-hospital complications were significantly higher in patients with ⬎3 times upper normal Tn-I levels (6.4%) than in patients with 1 to 3 times upper normal (3.4%) and normal (0%) Tn-I levels, p ⫽ 0.03. However, at 8 months, death (2.4%, 3.8%, and 5.1%; p ⫽ 0.75), Q-wave infarction (2.4%, 0%, and 0%; p ⫽ 0.18), target lesion revascularization (14.1%, 11.9%, and 10.1%; p ⫽ 0.67), and out-of-hospital event-free 1080 THE AMERICAN JOURNAL OF CARDIOLOGY姞
p Value
DISCUSSION
This study assessed the significance of myocardial damage defined and stratified by Tn-I levels after catheter-based coronary interventions in a large consecutive group of patients. Tn-I elevation ⬎3 times the upper normal limit (⬎0.45 ng/ml) was a strong independent predictor of major in-hospital complications and was associated with lower pro1.8 0.74 cedural success, but not with an in0.6 0.66 creased rate of later clinical events 10.1 0.86 17.1 0.49 up to 8 months. This study also ex19.4 0.3 plored the interaction between postintervention maximal Tn-I and CK-MB values, showing that elevated levels of both markers may have an additive prognostic risk for major in-hospital complications. Furthermore, we found that neither Tn-I level between 1 to 3 times the upper normal limit nor CK-MB level ⬍3 times the upper normal limit were associated with increased major in-hospital complications or late adverse out-of-hospital outcomes up to 8 months follow-up. The definition of myocardial necrosis following coronary interventions has been traditionally based on CK and/or CK-MB levels.10,12 The actual cutoff level that may carry prognostic value is a matter of controversy, yet several recent investigators have recommended a value of CK-MB,10,12 or total CK ⬎3 times the upper normal range13 as a cutoff associated with incremental adverse prognostic risk. Previous studies have suggested that CK or CK-MB elevation after coronary angioplasty is associated with increased late mortality.13–15 However, other studies have found no adverse long-term clinical outcomes in patients with CK or CK-MB elevation.16,17 Moreover, conflicting results concerning the prognostic value of cardiac enzyme elevation were reported in studies comparing directional coronary atherectomy with balloon angioplasty.15,16 In those studies, the significance of myocardial damage was not assessed by Tn-I, which is probably a more sensitive and specific marker of myocardial damage.2,3 Tn-I was found as an independent predictor for early and late adverse clinical outcomes in patients with a variety of acute ischemic coronary syndromes, including acute chest pain,18 non-ST elevation acute coronary syndromes,6 and unstable angina.4,5 Our study indicates that Tn-I elevation is not associated with intermediate-term (8 months) out-of hospital adverse prognosis. However, because our follow-up was limited for 8 months, it is possible that this would not be the case during a longer follow-up period. This may be of particular significance because the increased late mortality in patients with elevated CK or CK-MB levels was previously observed after 1
VOL. 85
0.81 ⬍0.001 0.004 0.019 0.003 0.31 ⬍0.001 ⬍0.001 ⬍0.001
MAY 1, 2000
FIGURE 1. Kaplan-Meier plots of any out-of-hospital cardiac event-free survival (death, Q-wave infarction, coronary angioplasty, or bypass surgery) at 8 months after catheterbased coronary interventions in patients grouped by troponin-I (CT-I) values. The p value was 0.86 for survival comparison between the 3 groups.
TABLE V Independent Predictors of Any Cardiac Event (death, Q-wave infarction, coronary angioplasty, or bypass surgery) at Eight Months Predictive Variables Major in-hospital complications Tn-I ⬎0.45 ng/ml Log (CK ⫺ MBmax) ⫻ log (Tn-Imax ⫹ 0.01) CK-MB⬎12 ng/ml Any cardiac event (8 mo) Unstable angina Diabetes mellitus
Odds Ratio
95% Confidence Limits
p Value
2.1 1.1
1.2–3.9 1.02–1.2
0.01 0.01
1.8
0.99–3.30
0.053
1.55 1.43
1.04–2.31 1.00–2.04
0.03 0.049
year of follow-up.13,15,19 In the current analysis, similar to previous studies,16 only Q-wave infarction was part of late clinical outcomes. However, combining large non–Q-wave (CK-MB ⬎5 times normal level) and Q-wave infarctions might be more clinically relevant. Finally, it may be that improved angioplasty and stenting techniques have attenuated the potential long-term deleterious effects of periprocedural myocardial damage. In this study we also found that Tn-I was a stronger predictor than CK-MB for increased in-hospital major complications. Because Tn-I is a more sensitive and specific marker than CK-MB, patients at highest procedural risk should be stratified by Tn-I measurement in addition to CK-MB levels. In the present study, 19% of the patients with Tn-I levels ⬎3 times the upper normal range had CK-MB ⬍3 times the upper normal range. Such discrepancy can explain the potential advantage of Tn-I as a more sensitive predictor for adverse clinical events. The present study was not designed to evaluate the possible different mechanisms for Tn-I and CK-MB release after coronary
angioplasty interventions. However, the ability to detect even small increments of Tn-I compared with CK-MB may be the result of significantly (13 times) higher myocardial Tn-I concentration, a substantial narrower normal range, or perhaps a different response to ischemic episodes that are not associated with myocardial necrosis.4 There are several potential clinical implications for this study. Routine Tn-I measurement may be worthwhile undertaken after catheter-based coronary interventions. Patients with elevated Tn-I levels ⬎3 times the upper normal limits should be closely monitored; a longer in-hospital follow-up may be necessary regardless of CK-MB levels. Such an approach is in agreement with the recent recommendation for patients with CK or CK-MB ⬎3 times the normal limits after catheter-based coronary interventions to be followed as patients sustaining myocardial infarction.10,12 However, after hospital discharge, these patients can be followed in a manner similar to patients without Tn-I level elevation, and reassured that their outcomes would probably be comparable to those without Tn-I elevation. 1. Hamm CW, Katus HA. New biochemical markers for myocardial cell injury. Curr Opin Cardiol 1995;10:355–360. 2. Bertinchant JP, Larue C, Pernel I, Ledermann B, Fabbro-Peray P, Beck L, Calzolari C, Trinquier S, Nigond J, Pau B. Release kinetics of serum cardiac troponin I in ischemic myocardial injury. Clin biochem 1996;29:587–594. 3. Adams JE III, Bodor GS, Davila-Roman VG, Delmez JA, Apple FS, Ladenson JH, Jaffe AS. Cardiac troponin I, a marker with high specificity for cardiac injury. Circulation 1993;88:101–106. 4. Galvani M, Ottani F, Ferrini D, Ladenson JH, Destro A, Baccos D, Rusticali F, Jaffe AS. Prognostic influence of elevated values of cardiac troponin I in patients with unstable angina. Circulation 1997;95:2053–2059. 5. Luscher MS, Thygesen K, Ravkilde J, Heickendorff L. Applicability of cardiac troponin T and I for early risk stratification in unstable coronary artery disease. TRIM Study Group. Thrombin inhibition in myocardial ischemia. Circulation 1997;96:2578 –2585. 6. Antman EM, Tanasijevic MJ, Thompson B, Schactman M, McCabe CH, Cannon CP, Fischer GA, Fung AY, Thompson C, Wybenga D, Braunwald E.
CORONARY ARTERY DISEASE/TROPONIN I AFTER CORONARY ANGIOPLASTY
1081
Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342–1349. 7. Shyu KG, Kuan PL, Cheng JJ, Hung CR. Cardiac troponin T, creatine kinase, and its isoform release after successful percutaneous transluminal coronary angioplasty with or without stenting. Am Heart J 1998;135:862– 867. 8. La Vecchia L, Bedogni F, Finocchi G, Mezzena G, Martini M, Sartori M, Castellani A, Soffiati G, Vincenzi M. Troponin T, troponin I, and creatine kinase-MB mass after elective coronary stenting. Coron Artery Dis 1996;7:535–540. 9. Karim AM, Shinn M, Oskarsson H, Windle J, Deligonul U. Significance of cardiac troponin T release after percutaneous transluminal coronary angioplasty. Am J Cardiol 1995;76:521–523. 10. Abdelmeguid AE, Topol EJ. The myth of the myocardial ‘infarctlet’ during percutaneous coronary revascularization procedures. Circulation 1996;94:3369 – 3375. 11. Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 1993;39:561–577. 12. Califf RM, Abdelmeguid AE, Kuntz RE, Popma JJ, Davidson CJ, Cohen EA, Kleiman NS, Mahaffey KW, Topol EJ, Pepine CJ, Lipicky RJ, Granger CB, Harrington RA, Tardiff BE, Crenshaw BS, Bauman RP, Zuckerman BD, Chaitman BR, Bittl JA, Ohman EM. Myonecrosis after revascularization procedures. J Am Coll Cardiol 1998;31:241–251. 13. 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– 466.
1082 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
14. Abdelmeguid AE, Topol EJ, Whitlow PL, Sapp SK, Ellis SG. Significance of mild transient release of creatine kinase-MB fraction after percutaneous interventions. Circulation 1996;94:1528 –1536. 15. Elliott JM, Berdan LG, Holmes DR, Isner JM, King SB, Keeler GP, Kearney M, Califf RM, Topol EJ. One-year follow-up in the coronary angioplasty versus excisional atherectomy trial (CAVEAT I). Circulation 1995;91:2158 –2166. 16. Baim DS, Cutlip DE, Sharma SK, Ho KK, Fortuna R, Schreiber TL, Feldman RL, Shani J, Senerchia C, Zhang Y, Lansky AJ, Popma JJ, Kuntz RE. Final results of the Balloon vs Optimal Atherectomy Trial. Circulation 1998;97:322– 331. 17. Oh JK, Shub C, Ilstrup DM, Reeder GS. Creatine kinase release after successful percutaneous transluminal coronary angioplasty. Am Heart J 1985; 109:1225–1231. 18. Polanczyk CA, Lee TH, Cook EF, Walls R, Wybenga D, Printy-Klein G, Ludwig L, Guldbrandsen G, Johnson PA. Cardiac troponin I as a predictor of major cardiac events in emergency department patients with acute chest pain. J Am Coll Cardiol 1998;32:8 –14. 19. Topol EJ, Ferguson JJ, Weisman HF, Tcheng JE, Ellis SG, Kleiman NS, Ivanhoe RJ, Wang AL, Miller DP, Anderson KM, Califf RM. Long-term protection from myocardial ischemic events in a randomized trial of brief integrin beta3 blockade with percutaneous coronary intervention. EPIC Investigator Group. Evaluation of Platelet IIb/IIIa Inhibition for Prevention of Ischemic Complication. JAMA 1997;278:479 – 484.
MAY 1, 2000
MD,
This study has examined the prognostic significance of troponin-I (Tn-I) levels after catheter-based coronary interventions in coronary arteries and saphenous vein grafts lesions. Tn-I and creatine kinase-MB (CK-MB) fraction levels were measured at 6 and 18 to 24 hours after catheter-based coronary intervention in 1,129 consecutive patients with normal preintervention plasma levels of Tn-I, and CK-MB levels below the cutoff for myocardial infarction. Patients were stratified according to maximal postangioplasty Tn-I levels. Group I (n ⴝ 784) had no elevated Tn-I (<0.15 ng/ml), group II (n ⴝ 170) had Tn-I at 0.15 to 0.45 ng/ml, and group III (n ⴝ 175) had Tn-I elevation >0.45 ng/ml. Major in-hospital complications (death, Q-wave infarction, and emergent coronary bypass grafting) and out-of-hospital intermediate-term (8 months) outcomes were compared between the 3 groups. Tn-I elevation >0.45 ng/ml was associated with increased risk of mortality (group III, 1.6%; group II, 0.6%; and group I, 0.1%; p ⴝ 0.019) and major in-hospital complications (3.2%, 1.7%, and 0.5%;
p ⴝ 0.004). There was no difference in death (1.8%, 3.2%, and 2.4%; p ⴝ 0.74), Q-wave infarction (0.6%, 0%, and 0.3%; p ⴝ 0.66), or target lesion revascularization (10.1%, 9.0%, and 9.3%; p ⴝ 0.86) between the 3 groups at follow-up. Cardiac event-free survival was similar between groups (p ⴝ 0.3). By multivariate analysis, Tn-I >0.45 ng/ml was an independent predictor for major in-hospital complications (odds ratio 2.1, 95% confidence interval 1.2 to 3.9, p ⴝ 0.01). The degree of risk was also associated with the conjoint elevation of Tn-I and CK-MB levels (odds ratio 1.1, 95% confidence interval 1.02 to 1.2, p ⴝ 0.01). We conclude that Tn-I levels >3 times the normal limit and conjoint elevation of Tn-I and CK-MB levels after coronary angioplasty are associated with increased risk of major in-hospital complications, but have no incremental risk of adverse intermediate-term (8 months) clinical outcomes. 䊚2000 by Excerpta Medica, Inc. (Am J Cardiol 2000;85:1077–1082)
ardiac troponin-I (Tn-I) is a sensitive and specific marker for detecting myocardial damage. C Previous studies have found Tn-I elevation to predict
Accordingly, we studied the prognostic significance of Tn-I levels after catheter-based coronary interventions in a large consecutive patient population.
1– 4
short- and long-term adverse clinical outcomes in acute coronary syndromes, including unstable angina and acute myocardial infarction.4 – 6 Elevation of cardiac troponins was detected in 30% to 44% of patients undergoing catheter-based coronary interventions.7–9 This rate may be higher than creatine kinase (CK) and creatine kinase-MB (CK-MB) elevation, which has been reported in 5% to 26% of patients undergoing percutaneous transluminal coronary angioplasty.7–10 Myocardial necrosis after coronary angioplasty, as determined by CK-MB elevation, is associated with up to a 30% increased risk for cardiac death and myocardial infarction during 3-year follow-up.10 However, the prognostic importance of myocardial damage after percutaneous coronary intervention stratified by Tn-I levels has not been established. From the Cardiovascular Research Foundation, The Washington Hospital Center, Washington, DC. This study was supported by a grant from the Cardiology Research Foundation, The Washington Cardiology Center, Washington, DC. Manuscript received August 3, 1999; revised manuscript received and accepted November 18, 1999. Address for reprints: Ran Kornowski, MD, Cardiovascular Research Foundation, Washington Hospital Center, 110 Irving St. NW, Suite 4B-1, Washington, DC 20010. E-mail: [email protected]. ©2000 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 85 May 1, 2000
METHODS
Study population: The patient cohort includes a consecutive series of 1,129 patients who underwent percutaneous coronary interventions between July 1, 1996, and March 31, 1998, in whom routine serial postintervention Tn-I was measured. Patients were divided into 3 groups according to maximal postintervention Tn-I levels. Group I (784 patients) had no elevated troponin (Tn-I ⬍ 0.15 ng/ml), group II (170 patients) had intermediate Tn-I levels between 1 to 3 times the upper limit of normal (0.15 to 0.45 ng/ml), and group III (175 patients) had ⬎3 times the upper normal limit (⬎0.45 ng/ml). Patients were excluded if preintervention serum markers were indicative for myocardial infarction before admission (Tn-I ⬎0.15 ng/ml or CK-MB ⬎8 ng/ml). Overall, 138 patients (14.7%) with a diagnosis of unstable angina who underwent percutaneous coronary interventions were excluded because of elevated preprocedural Tn-I levels. Tn-I and CK-MB measurements: Blood samples were routinely obtained from all patients at 6 and 18 to 24 hours after the procedure. If Tn-I or CK-MB levels 0002-9149/00/$–see front matter PII S0002-9149(00)00699-8
1077
Out-of-hospital clinical outcomes for up to 8 months of follow-up were Tn-I Level (patients) obtained by serial telephone interviews by research nurses. All clinical ⬍ ⫻1 ⫻1–3 ⬎ ⫻3 events (death, Q-wave infarction), Characteristics (n ⫽ 784) (n ⫽ 170) (n ⫽ 175) p Value target lesion revascularization, or Mean age (yrs) 64 ⫾ 11 65 ⫾ 12 66 ⫾ 11 0.049 any cardiac event (death, Q-wave inMale gender (%) 69.0 67.1 70.3 0.80 farction, coronary angioplasty, or byUnstable angina (%) 69.6 74.9 73.0 0.30 Systemic hypertension (%) 62.5 62.9 68.6 0.29 pass surgery) were adjudicated by Diabetes mellitus (%) 28.8 28.0 28.3 0.98 accompanying source documentation Hypercholesterolemia (%) 74 74.9 73.4 0.96 reviewed by research nurses. In adPrior myocardial infarction (%) 47.0 48.4 47.5 0.94 dition to target lesion revascularizaPrior bypass surgery (%) 39.5 40.2 48.1 0.095 Prior angioplasty (%) 51.4 48.0 46.5 0.40 tion, repeat revascularization is reEjection fraction (%) 48 ⫾ 13 45 ⫾ 13 46 ⫾ 13 0.10 ported for each patient (as any repeat revascularization) and includes all target lesion and vessels for single were elevated, serial measurements every 8 hours and multiple vessel interventions. The diagnosis of were performed and the peak level was recorded. Tn-I Q-wave infarction during follow-up was based on levels were measured by paramagnetic-particle, hospitalization records and documented discharge chemiluminescent immunoenzymatic assay (Beck- summaries with clinical diagnosis of infarction with man, Coulter Inc., California). With use of this assay, CK-MB increase ⱖ3 times the normal value and the the sample can be accurately measured within the appearance of new pathologic Q waves on the elecreportable range of the lower limit of detection and the trocardiogram. highest calibrated value (0.03 to 50 ng/ml). The upper Statistics: Continuous variables are presented as limit of the 95% nonparametric range for a presum- mean ⫾ 1 SD. Categorical data are presented as perably healthy population is below the mean minimum cent frequency and compared between groups using detectable concentration of the assay (⬍0.03 ng/ml) chi-square statistics. Major in-hospital complications and the diagnostic cutoff value for myocardial infarc- and late clinical events were analyzed as hierarchical tion based on receiver-operating characteristic plots is end points. Multivariate analysis was performed by ⬎0.15 ng/ml.11 This diagnostic cutoff has been used SAS Logistic Regression Statistics (Cary, North Caroroutinely in our laboratory as the upper normal Tn-I lina). Survival curves were calculated and displayed value. CK-MB levels were determined by a 2-site using the SAS LIFETEST procedure. Wilcoxon staimmunoenzymatic (“sandwich”) assay (Beckman). tistics were used for survival comparison between Using this assay, the sample can be accurately mea- groups. The means of continuous values were comsured within the reportable range of the lower limit of pared using the unpaired Student’s t test. Regression detection and the highest calibrated value (0.3 to 300 analysis was used to correlate between Tn-I and ng/ml). The upper limit of the 95% nonparametric CK-MB levels. An interaction test was performed range for a healthy population is 0.3 to 4.0 ng/ml. A between maximal postintervention CK-MB and Tn-I diagnostic cutoff of 4.0 ng/ml is considered the upper levels. To correct for differences in scales between normal value. Accordingly, the upper normal values peak Tn-I (0 to 50 ng/ml) and peak CK-MB (0 to 500 used in the present study were 0.15 and 4.0 ng/ml for ng/ml), a logarithmic transformation was performed. Tn-I and CK-MB, respectively. A p value ⬍0.05 was considered statistically signifiPeriprocedural and follow-up data: Baseline demo- cant. graphics and in-hospital complications were confirmed by independent hospital chart review. All patients underwent pre- and postintervention 12-lead RESULTS Baseline demographics: Table I lists the baseline electrocardiography. The diagnosis of Q-wave myocardial infarction was based on CK-MB ⱖ3 times the characteristics of treated patients. Patients with ⬎3 normal values, with appearance of a new pathologic Q times Tn-I levels were slightly older. Otherwise pawave on the postintervention surface electrocardio- tients’ demographics were similar among the 3 gram. The diagnosis of non–Q-wave infarction was groups. Tn-I and CK-MB measurements: Baseline and postbased on CK-MB elevation ⱖ5 times the normal values, in the absence of new pathologic Q waves on procedural Tn-I and CK-MB levels were available in the surface electrocardiogram. The diagnosis of in- all patients. Any postprocedural elevation of Tn-I hospital recurrent ischemia was based on symptoms (⬎0.15 ng/ml) occurred in 30.6% of patients, whereas associated with transient ST-segment deviation. An- CK-MB (⬎4 ng/ml) occurred in 40.8% of patients. giographic success was defined as ⬍50% residual Tn-I levels ⬎3 times the upper limit of normal were diameter stenosis without compromised anterograde elevated in 15.4%, whereas CK-MB levels ⬎3 times flow. Clinical success was defined as angiographic the upper limit of normal were elevated in 22.5% of success without in-hospital complications (death, Q- patients. However, in 19% of the patients with Tn-I wave infarction, emergent coronary bypass surgery, or levels ⬎3 times the upper normal limit, the CK-MB ischemia-driven repeat coronary angioplasty). level was ⬍3 times upper normal values. TABLE I Baseline Characteristics of the Study Population
1078 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
MAY 1, 2000
tients (90%) with 1 to 3 times normal Tn-I, and in 152 of 175 patients Tn-I Level (lesions) (87%) with ⬎3 times upper normal Tn-I levels. All clinical outcomes ⬍ ⫻1 ⫻1–3 ⬎ ⫻3 were similar among the 3 groups (n ⫽ 1,562) (n ⫽ 370) (n ⫽ 436) p Value (Table IV). Actuarial event-free surTarget coronary artery vival curves for any cardiac event up Right (%) 31.9 30.3 23.1 0.002 to 8 months are shown in Figure 1. Left anterior descending (%) 31.0 33.5 33.6 0.43 Left circumflex (%) 26.1 23.8 25.2 0.65 CK-MB measurements: In a sepaSaphenous vein graft (%) 11.0 12.4 18.1 0.001 rate analysis, patients were stratified Lesion location according to maximal postangioOstial (%) 10.4 7.8 10.2 0.33 plasty CK-MB levels. Group I (n ⫽ Proximal (%) 33.1 37.8 34.0 0.22 Mid (%) 39.4 36.5 38.9 0.58 668) had no CK-MB elevation (⬍4.0 Distal (%) 15.2 16.8 14.8 0.70 ng/ml), group II (n ⫽ 204) had Lesion % diameter stenosis CK-MB levels between 1 to 3 times Preprocedure (%) 68 ⫾ 14 63 ⫾ 9 63 ⫾ 16 0.67 the upper limit (4.0 to 12.0 ng/ml), Postprocedure (final) (%) 15 ⫾ 9 11 ⫾ 17 17 ⫾ 12 0.66 and group III (n ⫽ 257) had CK-MB levels ⬎3 times the upper normal level (⬎12.0 ng/ml). Patients in TABLE III Interventional Procedures group III had a trend for higher inTn-I Level (lesions) hospital mortality (1.1%) than those in groups II (0.5%) and I (0.1%), p ⫽ ⬍ ⫻1 ⫻1–3 ⬎ ⫻3 0.08. Major in-hospital complication (n ⫽ 1,562) (n ⫽ 370) (n ⫽ 436) p Value occurred more frequently in group Procedure type III (2.5%) than in groups II (0.9%) Balloon angioplasty (%) 96.1 98.4 96.1 0.076 and I (0.6%), p ⫽ 0.023. Overall Rotational atherectomy (%) 20.5 21.1 16.5 0.14 Eximer laser angioplasty (%) 6.9 10.3 12.4 0.001 clinical success was lower in group Directional atherectomy (%) 1.5 2.2 1.4 0.59 III (90.5%) than in patients in groups Stent (%) 73.1 74.1 70.9 0.61 II (96.2%) and I (97.4%), p ⫽ 0.001. Intravascular ultrasound (%) 75 75 70 0.18 At 8-month follow-up, death (1.2%, Procedural complications Procedural dissection (%) 2.3 1.1 5.3 0.001 4.0%, and 2.3%, p ⫽ 0.14), Q-wave Abrupt closure (%) 0.2 0.0 2.1 ⬍0.001 infarction (3.2%, 3.0%, and 2.3%, No reflow (%) 0.0 0.0 0.7 0.01 p ⫽ 0.70), or any repeat revascularization (19.1%, 18.7%, and 14.5%, p ⫽ 0.15) were similar between the 3 Lesion location and characteristics: Lesion location groups. Similar event-free survival was noted in paand characteristics are described in Table II. Patients tients with ⬎3 times the normal level (24.2%) comin the highest Tn-I level group underwent interven- pared with patients with intermediate (21.5%) and tions to the right coronary artery less often, and inter- normal (17.3%) CK-MB levels (p ⫽ 0.063). ventions to a vein graft more often. The locations of Multivariate analysis: Logistic regression analysis the lesions within the target artery and the pre- and was used to identify independent predictors of major postprocedural percent diameter stenosis were similar in-hospital complications (death, Q-wave infarction, between the 3 groups. and emergent bypass surgery). Variables included in Angiographic and procedural results: Procedural the model were age, gender, unstable angina, history type and complications are summarized in Table III. of coronary angioplasty or bypass surgery, diabetes Notably, there was no difference in the rate of stent mellitus, left ventricular ejection fraction, vein graft implantation between groups. Procedural complicalesion, abrupt closure, and CK-MB and Tn-I levels ⬎3 tions were more frequent in the group with the highest times the upper normal limit. Tn-I level ⬎3 times the Tn-I levels. Overall angiographic (98.1%) and clinical upper normal limit was found to be associated with success rates (95.6%) were high. Patients in the highmajor in-hospital complications (odds ratio 2.1, 95% est Tn-I level group had higher in-hospital mortality, Q-wave, and non–Q-wave infarction, repeat lesion confidence intervals 1.2 to 3.9, p ⫽ 0.01). We also angioplasty, and overall major in-hospital complica- found an interaction between the log of maximal Tn-I tions (Table IV). Patients in the highest Tn-I level and CK-MB levels (chi-square 6.36, p ⫽ 0.01). When group received abciximab (11.6%) more often than the interaction term (log [CK-MBmax ⫹ 1] ⫻ log patients with intermediate (4.6%) or normal (4.0%) [Tn-Imax ⫹ 0.01]) was entered into the multivariate Tn-I levels, p ⫽ 0.001. Abciximab was used mostly model, it was found to be an independent predictor for on a “rescue” basis (82.3%, p ⫽ 0.12 between groups major in-hospital complications, whereas CK-MB ⬎3 times the normal limit (⬎12 ng/ml) was only of marfor prophylactic vs rescue administration). Intermediate-term outcomes: Out-of-hospital clini- ginal significance in this model (Table V). In a sepacal follow-up at 8 months was available in 478 of 542 rate logistic regression analysis, unstable angina and patients (88%) with normal Tn-I, in 153 of 170 pa- diabetes mellitus were found to be independent preTABLE II Location Data and Angiographic Results of Treated Lesions
CORONARY ARTERY DISEASE/TROPONIN I AFTER CORONARY ANGIOPLASTY
1079
survival (76.2%, 80.8%, and 80.5%; p ⫽ 0.83) were similar between the 3 groups.
TABLE IV In-Hospital and Eight-Month Events in the Studied Patients Tn-I Level (patients)
In-hospital results Angiographic success (%) Clinical success (%) Overall major complications (%) Death (%) Q-wave infarction (%) Emergent coronary bypass (%) Non–Q-wave infarction (%) CK-MB ⬎3 times normal (%) Repeat lesion angioplasty (%) Follow-up events (8 mo) Death (%) Q-wave infarction (%) Target lesion revascularization (%) Any repeat revascularization (%) Any cardiac event (%)
⬍ ⫻1 (n ⫽ 784)
⫻1–3 (n ⫽ 170)
⬎ ⫻3 (n ⫽ 175)
98.2 97.7 0.5 0.1 0.0 0.4 1.4 6.1 0.4
98.1 94.3 1.7 0.6 1.7 0.0 22.3 48.6 3.4
97.4 87.8 3.2 1.6 0.5 1.1 76.2 82.0 6.9
2.4 0.3 9.3 13.8 15.5
3.2 0.0 9.0 16.4 19.0
dictors for out-of-hospital composite cardiac events (Table V). Coronary artery and saphenous vein graft lesions:
Clinical outcomes were compared between patients with coronary artery lesions and normal Tn-I levels (n ⫽ 676), and patients with 1 to 3 times (n ⫽ 142) and ⬎3 times upper normal Tn-I levels (n ⫽ 136). Clinical success rate was lower in patients with the highest Tn-I level (88.3%) than in patients with intermediate (94.9%) and normal (97.4%) Tn-I levels, p ⫽ 0.001. In-hospital repeat revascularization was higher in patients with the highest Tn-I level (7.8%) than in patients with intermediate (3.2%) and normal (0.38%) Tn-I levels, p ⫽ 0.001, whereas major in-hospital complications were similar between the 3 groups (1.9%, 1.3%, and 0.5%; p ⫽ 0.16). At 8-month follow-up, death (2.2%, 2.8%, and 2.2%; p ⫽ 0.87), Q-wave infarction (0%, 0%, and 0.3%; p ⫽ 1.0), target lesion revascularization (9.1%, 8.5%, and 9.2%, p ⫽ 0.94), and out-of-hospital event-free survival (82.2%, 81.7%, and 84.2%; p ⫽ 0.68) were also similar between the 3 groups. Similar subgroup analysis was performed in patients with saphenous vein graft lesions and normal Tn-I values (n ⫽ 106), in those with 1 to 3 times upper normal values (n ⫽ 26), and in patients with ⬎3 times upper normal Tn-I levels (n ⫽ 42). Patients with the highest Tn-I level had significantly lower clinical success (89.4%) than patients with intermediate (93.1%) and normal (99.2%) Tn-I levels, p ⫽ 0.001. In-hospital repeat revascularization rate was similar between the 3 groups (2.1%, vs 3.4%, and 0.8%; p ⫽ 0.55). Major in-hospital complications were significantly higher in patients with ⬎3 times upper normal Tn-I levels (6.4%) than in patients with 1 to 3 times upper normal (3.4%) and normal (0%) Tn-I levels, p ⫽ 0.03. However, at 8 months, death (2.4%, 3.8%, and 5.1%; p ⫽ 0.75), Q-wave infarction (2.4%, 0%, and 0%; p ⫽ 0.18), target lesion revascularization (14.1%, 11.9%, and 10.1%; p ⫽ 0.67), and out-of-hospital event-free 1080 THE AMERICAN JOURNAL OF CARDIOLOGY姞
p Value
DISCUSSION
This study assessed the significance of myocardial damage defined and stratified by Tn-I levels after catheter-based coronary interventions in a large consecutive group of patients. Tn-I elevation ⬎3 times the upper normal limit (⬎0.45 ng/ml) was a strong independent predictor of major in-hospital complications and was associated with lower pro1.8 0.74 cedural success, but not with an in0.6 0.66 creased rate of later clinical events 10.1 0.86 17.1 0.49 up to 8 months. This study also ex19.4 0.3 plored the interaction between postintervention maximal Tn-I and CK-MB values, showing that elevated levels of both markers may have an additive prognostic risk for major in-hospital complications. Furthermore, we found that neither Tn-I level between 1 to 3 times the upper normal limit nor CK-MB level ⬍3 times the upper normal limit were associated with increased major in-hospital complications or late adverse out-of-hospital outcomes up to 8 months follow-up. The definition of myocardial necrosis following coronary interventions has been traditionally based on CK and/or CK-MB levels.10,12 The actual cutoff level that may carry prognostic value is a matter of controversy, yet several recent investigators have recommended a value of CK-MB,10,12 or total CK ⬎3 times the upper normal range13 as a cutoff associated with incremental adverse prognostic risk. Previous studies have suggested that CK or CK-MB elevation after coronary angioplasty is associated with increased late mortality.13–15 However, other studies have found no adverse long-term clinical outcomes in patients with CK or CK-MB elevation.16,17 Moreover, conflicting results concerning the prognostic value of cardiac enzyme elevation were reported in studies comparing directional coronary atherectomy with balloon angioplasty.15,16 In those studies, the significance of myocardial damage was not assessed by Tn-I, which is probably a more sensitive and specific marker of myocardial damage.2,3 Tn-I was found as an independent predictor for early and late adverse clinical outcomes in patients with a variety of acute ischemic coronary syndromes, including acute chest pain,18 non-ST elevation acute coronary syndromes,6 and unstable angina.4,5 Our study indicates that Tn-I elevation is not associated with intermediate-term (8 months) out-of hospital adverse prognosis. However, because our follow-up was limited for 8 months, it is possible that this would not be the case during a longer follow-up period. This may be of particular significance because the increased late mortality in patients with elevated CK or CK-MB levels was previously observed after 1
VOL. 85
0.81 ⬍0.001 0.004 0.019 0.003 0.31 ⬍0.001 ⬍0.001 ⬍0.001
MAY 1, 2000
FIGURE 1. Kaplan-Meier plots of any out-of-hospital cardiac event-free survival (death, Q-wave infarction, coronary angioplasty, or bypass surgery) at 8 months after catheterbased coronary interventions in patients grouped by troponin-I (CT-I) values. The p value was 0.86 for survival comparison between the 3 groups.
TABLE V Independent Predictors of Any Cardiac Event (death, Q-wave infarction, coronary angioplasty, or bypass surgery) at Eight Months Predictive Variables Major in-hospital complications Tn-I ⬎0.45 ng/ml Log (CK ⫺ MBmax) ⫻ log (Tn-Imax ⫹ 0.01) CK-MB⬎12 ng/ml Any cardiac event (8 mo) Unstable angina Diabetes mellitus
Odds Ratio
95% Confidence Limits
p Value
2.1 1.1
1.2–3.9 1.02–1.2
0.01 0.01
1.8
0.99–3.30
0.053
1.55 1.43
1.04–2.31 1.00–2.04
0.03 0.049
year of follow-up.13,15,19 In the current analysis, similar to previous studies,16 only Q-wave infarction was part of late clinical outcomes. However, combining large non–Q-wave (CK-MB ⬎5 times normal level) and Q-wave infarctions might be more clinically relevant. Finally, it may be that improved angioplasty and stenting techniques have attenuated the potential long-term deleterious effects of periprocedural myocardial damage. In this study we also found that Tn-I was a stronger predictor than CK-MB for increased in-hospital major complications. Because Tn-I is a more sensitive and specific marker than CK-MB, patients at highest procedural risk should be stratified by Tn-I measurement in addition to CK-MB levels. In the present study, 19% of the patients with Tn-I levels ⬎3 times the upper normal range had CK-MB ⬍3 times the upper normal range. Such discrepancy can explain the potential advantage of Tn-I as a more sensitive predictor for adverse clinical events. The present study was not designed to evaluate the possible different mechanisms for Tn-I and CK-MB release after coronary
angioplasty interventions. However, the ability to detect even small increments of Tn-I compared with CK-MB may be the result of significantly (13 times) higher myocardial Tn-I concentration, a substantial narrower normal range, or perhaps a different response to ischemic episodes that are not associated with myocardial necrosis.4 There are several potential clinical implications for this study. Routine Tn-I measurement may be worthwhile undertaken after catheter-based coronary interventions. Patients with elevated Tn-I levels ⬎3 times the upper normal limits should be closely monitored; a longer in-hospital follow-up may be necessary regardless of CK-MB levels. Such an approach is in agreement with the recent recommendation for patients with CK or CK-MB ⬎3 times the normal limits after catheter-based coronary interventions to be followed as patients sustaining myocardial infarction.10,12 However, after hospital discharge, these patients can be followed in a manner similar to patients without Tn-I level elevation, and reassured that their outcomes would probably be comparable to those without Tn-I elevation. 1. Hamm CW, Katus HA. New biochemical markers for myocardial cell injury. Curr Opin Cardiol 1995;10:355–360. 2. Bertinchant JP, Larue C, Pernel I, Ledermann B, Fabbro-Peray P, Beck L, Calzolari C, Trinquier S, Nigond J, Pau B. Release kinetics of serum cardiac troponin I in ischemic myocardial injury. Clin biochem 1996;29:587–594. 3. Adams JE III, Bodor GS, Davila-Roman VG, Delmez JA, Apple FS, Ladenson JH, Jaffe AS. Cardiac troponin I, a marker with high specificity for cardiac injury. Circulation 1993;88:101–106. 4. Galvani M, Ottani F, Ferrini D, Ladenson JH, Destro A, Baccos D, Rusticali F, Jaffe AS. Prognostic influence of elevated values of cardiac troponin I in patients with unstable angina. Circulation 1997;95:2053–2059. 5. Luscher MS, Thygesen K, Ravkilde J, Heickendorff L. Applicability of cardiac troponin T and I for early risk stratification in unstable coronary artery disease. TRIM Study Group. Thrombin inhibition in myocardial ischemia. Circulation 1997;96:2578 –2585. 6. Antman EM, Tanasijevic MJ, Thompson B, Schactman M, McCabe CH, Cannon CP, Fischer GA, Fung AY, Thompson C, Wybenga D, Braunwald E.
CORONARY ARTERY DISEASE/TROPONIN I AFTER CORONARY ANGIOPLASTY
1081
Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342–1349. 7. Shyu KG, Kuan PL, Cheng JJ, Hung CR. Cardiac troponin T, creatine kinase, and its isoform release after successful percutaneous transluminal coronary angioplasty with or without stenting. Am Heart J 1998;135:862– 867. 8. La Vecchia L, Bedogni F, Finocchi G, Mezzena G, Martini M, Sartori M, Castellani A, Soffiati G, Vincenzi M. Troponin T, troponin I, and creatine kinase-MB mass after elective coronary stenting. Coron Artery Dis 1996;7:535–540. 9. Karim AM, Shinn M, Oskarsson H, Windle J, Deligonul U. Significance of cardiac troponin T release after percutaneous transluminal coronary angioplasty. Am J Cardiol 1995;76:521–523. 10. Abdelmeguid AE, Topol EJ. The myth of the myocardial ‘infarctlet’ during percutaneous coronary revascularization procedures. Circulation 1996;94:3369 – 3375. 11. Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 1993;39:561–577. 12. Califf RM, Abdelmeguid AE, Kuntz RE, Popma JJ, Davidson CJ, Cohen EA, Kleiman NS, Mahaffey KW, Topol EJ, Pepine CJ, Lipicky RJ, Granger CB, Harrington RA, Tardiff BE, Crenshaw BS, Bauman RP, Zuckerman BD, Chaitman BR, Bittl JA, Ohman EM. Myonecrosis after revascularization procedures. J Am Coll Cardiol 1998;31:241–251. 13. 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– 466.
1082 THE AMERICAN JOURNAL OF CARDIOLOGY姞
VOL. 85
14. Abdelmeguid AE, Topol EJ, Whitlow PL, Sapp SK, Ellis SG. Significance of mild transient release of creatine kinase-MB fraction after percutaneous interventions. Circulation 1996;94:1528 –1536. 15. Elliott JM, Berdan LG, Holmes DR, Isner JM, King SB, Keeler GP, Kearney M, Califf RM, Topol EJ. One-year follow-up in the coronary angioplasty versus excisional atherectomy trial (CAVEAT I). Circulation 1995;91:2158 –2166. 16. Baim DS, Cutlip DE, Sharma SK, Ho KK, Fortuna R, Schreiber TL, Feldman RL, Shani J, Senerchia C, Zhang Y, Lansky AJ, Popma JJ, Kuntz RE. Final results of the Balloon vs Optimal Atherectomy Trial. Circulation 1998;97:322– 331. 17. Oh JK, Shub C, Ilstrup DM, Reeder GS. Creatine kinase release after successful percutaneous transluminal coronary angioplasty. Am Heart J 1985; 109:1225–1231. 18. Polanczyk CA, Lee TH, Cook EF, Walls R, Wybenga D, Printy-Klein G, Ludwig L, Guldbrandsen G, Johnson PA. Cardiac troponin I as a predictor of major cardiac events in emergency department patients with acute chest pain. J Am Coll Cardiol 1998;32:8 –14. 19. Topol EJ, Ferguson JJ, Weisman HF, Tcheng JE, Ellis SG, Kleiman NS, Ivanhoe RJ, Wang AL, Miller DP, Anderson KM, Califf RM. Long-term protection from myocardial ischemic events in a randomized trial of brief integrin beta3 blockade with percutaneous coronary intervention. EPIC Investigator Group. Evaluation of Platelet IIb/IIIa Inhibition for Prevention of Ischemic Complication. JAMA 1997;278:479 – 484.
MAY 1, 2000