|
|
||||||||||
|
J Am Coll Cardiol, 2006; 48:1765-1770, doi:10.1016/j.jacc.2006.04.102
(Published online 16 October 2006). © 2006 by the American College of Cardiology Foundation |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

* Division of Cardiovascular Diseases, Department of Internal Medicine, Rochester, Minnesota
Section of Biostatistics, Mayo Clinic and Mayo Foundation, Rochester, Minnesota.
Manuscript received January 13, 2006; revised manuscript received April 20, 2006, accepted April 23, 2006.
* Reprint requests and correspondence: Dr. Abhiram Prasad, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905. (Email: prasad.abhiram{at}mayo.edu).
| Abstract |
|---|
|
|
|---|
BACKGROUND: Cardiac troponin T is a sensitive and specific marker of myonecrosis. There is little known about the incidence and prognostic significance of an isolated elevation of cTnT without a rise in creatine kinase (CK)-MB following PCI.
METHODS: We evaluated the outcomes of 1,949 patients from the Mayo Clinic registry who had normal pre-procedure cTnT and CK-MB, required nonemergency percutaneous coronary intervention (PCI), and had normal CK-MB after the procedure.
RESULTS: An elevation in cTnT (cTnT+) was observed in 383 patients (19.6%) (median 0.04 ng/ml, interquartile range 0.03 to 0.06 ng/ml). The TnT+ status was associated with adverse clinical and angiographic characteristics, and multivessel PCI. Over the median follow-up duration of 26 months, mortality (p < 0.001) and the combined rate of death and myocardial infarction (p = 0.004) were significantly higher in cTnT+ patients. Estimated 3-year survival for those with and without cTnT elevation was 86.9% and 93.2%, respectively. By multivariate analysis, an elevation in cTnT after PCI was an independent predictor of increased long-term mortality. A doubling in the post-PCI cTnT was associated with a partial hazard ratio of 1.20 (95% confidence interval 1.02 to 1.40; p = 0.023).
CONCLUSIONS: An isolated minor elevation in cTnT after PCI provides long-term prognostic information regarding mortality and myocardial infarction.
| ||||||||
The data for a relationship between CK-MB elevation and long-term survival is less consistent. Several studies suggest that any elevation of CK-MB is associated with reduced long-term survival, and that there is direct correlation between the magnitude of myonecrosis and mortality (1,5,6 ). In contrast, others have reported a threshold of CK-MB 5 to 8 times normal to be predictive of long-term survival (2,4,7 ). The differentiation of a continuous versus a threshold relationship between post-procedural myonecrosis and long-term survival is important to establish, because it provides insight into the pathophysiologic mechanisms for the association (8 ). A threshold of 5 to 8 times normal would support a direct effect of myocardial injury on survival, whereas a continuous relationship would be consistent with the CK-MB elevation being a marker of atherosclerotic burden and disease activity. We hypothesized that the relationship is continuous, even at very low levels of myonecrosis. The aim of this study was to evaluate whether, in patients with normal postprocedure CK-MB, an isolated elevation in cardiac troponin T (cTnT) predicts long-term survival.
| Methods |
|---|
|
|
|---|
The current study included all consecutive patients from the database for the period of August 1, 2000, when routine measurements of cTnT were initiated for our PCI patients, through October 31, 2003. Inclusion criteria were a normal preprocedure cTnT (below the 10% coefficient of variation [CV] value) and CK-MB. Patients were excluded if they suffered myocardial infarction (MI) in the week before PCI or had a post-procedure elevation in CK-MB. Hospital charts of each patient were reviewed to verify the data, and the study was approved by the Institutional Review Board. There were 5,505 PCIs performed on 4,760 unique patients. One hundred seventeen patients refused use of their records for research, yielding a sample size of 5,368 PCIs. Of these, 1,997 had elevated baseline cTnT, 52 had elevated baseline CK-MB, 733 had elevated post-PCI CK-MB, and 357 had missing values for at least 1 of these 3 measures. Fifty-two PCIs were excluded for pre-procedural shock or recent myocardial infarction (within 7 days before PCI). Of the remaining 2,177 qualifying PCIs, the first for each of the 1,949 patients was included.
Blood samples for cardiac biomarkers were collected before and at 8 and 16 h after PCI. The analysis was performed using a highly sensitive and precise third-generation assay (Elecsys; Roche Diagnostics, Indianapolis, Indiana). The upper limit of normal for the assay is <0.03 ng/ml. The following definitions were used for the database. Myocardial infarction during follow-up was defined by the presence of 2 of 3 criteria: chest pain lasting 20 min or longer, new ST/T-wave changes or Q waves on the electrocardiogram, and increased cardiac biomarker (CK or CK-MB) levels at least twice the upper limit of the normal range. The number of diseased coronary arteries was defined by the number of major coronary arteries with luminal diameter stenosis
70%. Patients with
50% stenosis in the left main coronary artery were considered to have 2-vessel disease if there was right dominance and 3-vessel disease if there was left dominance. Angiographic success was defined as PCI with residual stenosis <50% in at least one treatment site. Complete revascularization was achieved if there were no remaining stenoses
70%.
Statistical analysis. Data are presented as the mean values ± SD or as a frequency (percentage). Kaplan-Meier methods were used to estimate survival curves. Survival was analyzed in successful PCIs, with the discharge date as day 0. Comparisons between groups are made using Student t test, Pearson chi-squared statistic, Mann-Whitney rank sum test, and log rank test for continuous, nominal, ordinal (i.e., number of diseased vessels, number of vessels treated, and length of stay), and survival data, respectively.
Cox proportional hazards models were used to estimate partial hazards ratios. Covariate models for each follow-up end point were constructed in the following manner: simple associations with the end point were estimated for covariates with sufficient data (<10% missing values); covariates significant at the 0.15 level were incorporated into a full covariate model; a backward selection procedure was used (with 0.05 significance threshold) to produce the final covariate model. Cardiac troponin (base 2 logarithmic transformation) was then added to the covariate-only model. Ejection fraction was dichotomized at 40%, and the hazard ratio for age is based on a 10-year increment in age. All analyses presented here were performed using SAS software (SAS Institute, Cary, North Carolina).
| Results |
|---|
|
|
|---|
40%, and peripheral vascular disease and less likely to be current smokers. There was no difference in the use of beta-blocker, angiotensin-converting enzyme inhibitors, and lipid-lowering therapy between the groups.
|
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
The present study analyzes the prognostic significance of an isolated rise in cTnT after successful nonemergency PCI procedures. The findings are representative of contemporary practice, because the majority of patients were treated with stents (>86%), atherectomy devices were used infrequently (<3%), and glycoprotein IIb/IIIa inhibitors were used in approximately 50% of interventions. In-hospital event rates were extremely low regardless of cTnT status, but follow-up death and the combined end point of death/MI were significantly higher in cTnT+ patients. Elevation in cTnT was an independent predictor of mortality over a median follow-up period of 2.2 years.
Comparison with previous studies. Earlier studies have evaluated the relationship between cardiac troponin I (914 ) and troponin T (14 ) with long-term mortality. However, unlike the present investigation, all except one (11 ) did not exclude patients who had elevated CK-MB after the revascularization procedure. Several (10,13,14 ) did not exclude patients with MI who would have had abnormal cardiac biomarker levels at baseline. For these reasons, the reported frequency for post-procedure troponin elevation ranged from 17% to 48%, and although some found elevated troponin to be an independent predictor of survival (10,12,14 ), others did not (9,11,13 ). The inconsistent findings may relate to heterogeneity in the inclusion criteria, the sensitivity and specificity of the assay used to measure troponin, and the duration of follow-up. The study by Natarajan et al. (11 ) analyzed the impact of troponin I and excluded patients with elevated CK-MB and pre-procedure MI. The incidence of isolated troponin elevation was similar (17%) to the present study, but troponin I was not found to be a predictor of mortality. However, their study had a smaller sample size and a follow-up of only 1 year. Consistent with this fact is the finding in the present investigation that the Kaplan-Meier curves for survival and survival free of MI diverged after 1 year. To our knowledge, our study is the first to report that an isolated elevation in troponin is a predictor of outcomes.
Mechanisms for the adverse clinical outcomes. The relationship between long-term survival and very low levels of cTnT elevation following PCI is remarkable. This suggests that there may be a continuous, rather than a threshold relationship between the magnitude of rise in the biomarkers of myocardial injury and long-term outcomes. Two potential mechanisms may explain this observation. First, studies with magnetic resonance imaging have confirmed that an elevation in troponin after PCI is related to myocardial necrosis, and that there is a positive correlation between the magnitude of injury and the extent of troponin release (15 ). Thus, myocardial injury may impair left ventricular function and predispose to arrhythmias, which would influence survival. Alternatively, cTnT and CK-MB elevation may be a marker for severe atherosclerosis, increased plaque burden (16 ), presence of vulnerable plaques, endothelial dysfunction, microvascular injury, and inflammation. Inflammation, as measured by elevated C-reactive protein (CRP), is an independent predictor of adverse long-term outcomes after PCI (1 ). The present study supports the latter mechanism, because the patients with elevated cTnT were older and had more severe angina, multivessel disease, complex lesion morphology, coronary thrombus, left ventricular systolic dysfunction, peripheral vascular disease, and urgent procedures, all consistent with advanced or unstable coronary artery disease. Moreover, pharmacologic interventions that have antiinflammatory and antithrombotic effects, such as statins and glycoprotein IIb/IIIa inhibitors have been shown to reduce the incidence and magnitude of myonecrosis (17,18 ). It seems unlikely that the very low level of myonecrosis observed in the present study would have been adequate to impact left ventricular function. One can speculate that both mechanisms may be at play, with minor myonecrosis serving as a marker of advanced and active atherosclerosis and large increases in biomarkers diagnosing periprocedural MI that would directly impact prognosis.
Mechanisms for the troponin elevation. Patients with cTnT elevation had greater atherosclerotic burden, as suggested by a higher prevalence of multivessel disease and type C lesions and a greater number of stents used per patient. Despite these differences, procedural success and TIMI grade 3 flow was achieved in nearly all patients. However, cTnT+ patients had a higher incidence of branch vessel occlusion and coronary dissection. Abrupt closure was not different between the groups, and angiographic embolization was not detected in any patient. These observations are consistent with earlier studies that have implicated branch vessel occlusion, dissection, and embolization as the mechanisms for the ischemic injury and myonecrosis (11,12,19 ). However, angiographic complications were absent in approximately three-quarters of the patients with elevated cTnT. Impaired microvascular perfusion, due to plaque and thrombus embolization, and vasoconstriction may be underlying mechanisms in a significant proportion of patients (2022 ). This would be either undetectable by routine angiographic evaluation in the catheterization laboratory or present as slow reflow.
Study limitations. Although the data was collected prospectively, this is a retrospective single-center analysis and is subject to the limitations of such analyses. Furthermore, multiple regression models are unable to account for unobserved covariates that may be confounded with troponin. Independent core angiographic laboratory analysis to investigate myocardial perfusion was not performed and may have provided insight into the mechanism for the cTnT release in patients without angiographic complications. Markers of inflammation such as CRP were not routinely measured. Emerging data indicate that TnT values between the 10% CV value, which we used, and the 99th percentile value of 0.01 ng/ml have prognostic significance. This fact and the recent data concerning the potential significance of pre-PCI TnT levels for evaluating post-PCI myonecrosis (23 ) highlights the need for additional studies.
Conclusions and clinical implications. Patients with an isolated elevation in cTnT following nonemergency PCI represent a very low-risk group for in-hospital major adverse cardiac events, and their hospitalization need not be prolonged. However, an isolated elevation in cTnT is associated with higher rates of death and MI on follow-up, provides long-term prognostic information, and appears to identify, independent of well recognized markers of atherosclerotic burden and ventricular function, high-risk patients who may benefit from more intensive secondary prevention. Further studies using multiple biomarkers together with imaging techniques such as intravascular ultrasound and core angiographic laboratory analysis are required to improve our understanding of the mechanisms responsible for the myonecrosis, with the goal of improving risk stratification and ultimately developing novel treatment strategies.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Cuisset, M. Hamilos, N. Melikian, E. Wyffels, J. Sarma, G. Sarno, E. Barbato, J. Bartunek, W. Wijns, and B. De Bruyne Direct stenting for stable angina pectoris is associated with reduced periprocedural microcirculatory injury compared with stenting after pre-dilation. J. Am. Coll. Cardiol., March 18, 2008; 51(11): 1060 - 1065. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Schiano and P. Gabriel Steg Measurement and prevention of myocardial injury during percutaneous coronary intervention Heart, June 1, 2007; 93(6): 656 - 657. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. E. Cutlip, S. Windecker, R. Mehran, A. Boam, D. J. Cohen, G.-A. van Es, P. Gabriel Steg, M.-a. Morel, L. Mauri, P. Vranckx, et al. Clinical End Points in Coronary Stent Trials: A Case for Standardized Definitions Circulation, May 1, 2007; 115(17): 2344 - 2351. [Abstract] [Full Text] [PDF] |
||||
![]() |
Isolated cTnT Elevation After PCI Journal Watch Cardiology, December 6, 2006; 2006(1206): 3 - 3. [Full Text] |
||||
![]() |
N. S. Kleiman Measuring Troponin Elevation After Percutaneous Coronary Intervention: Ready for Prime Time? J. Am. Coll. Cardiol., November 7, 2006; 48(9): 1771 - 1773. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |