Author + information
- Received March 4, 1999
- Revision received June 18, 1999
- Accepted August 18, 1999
- Published online November 15, 1999.
- Shmuel Fuchs, MDa,b,
- Ran Kornowski, MD, FACCa,b,* (, )
- Roxana Mehran, MD, FACCa,b,
- Lowell F Satler, MD, FACCa,b,
- Augusto D Pichard, MD, FACCa,b,
- Kenneth M Kent, MD, FACCa,b,
- Mun K Hong, MD, FACCa,b,
- Steve Slack, BSca,b,
- Gregg W Stone, MD, FACCa,b and
- Martin B Leon, MD, FACCa,b
- ↵*Reprint requests and correspondence: Dr. Ran Kornowski, Cardiology Research Foundation, Washington Hospital Center, 110 Irving St. NW, Suite 4B-1, Washington, DC 20010.
To establish the role of early catheter-based coronary intervention among patients sustaining acute coronary syndromes (ACS) stratified according to admission plasma troponin I (Tn-I) levels.
The impact of early revascularization strategy on the clinical outcomes in patients with ACS stratified by plasma Tn-I levels has not been established.
In-hospital complications and long-term outcomes were assessed in 1,321 consecutive patients with non-ST elevation ACS undergoing early (within 72 h) catheter-based coronary interventions. Patients were grouped according to admission Tn-I levels. Group I (n = 1,099) had no elevated plasma Tn-I (<0.15 ng/ml), Group II (n = 95) had Tn-I level between 0.15 to 0.45 ng/ml and Group III (n = 127) had Tn-I > 0.45 ng/ml. In-hospital composite cardiac events (death, Q-wave MI, urgent in-hospital revascularization) and 8 months clinical outcomes (death, MI, repeat revascularization or any cardiac event) were compared between the three groups.
The rate of in-hospital composite cardiac events was 6.1% among patients with Tn-I > 0.45 ng/ml, 1.0% in patients with Tn-I between 0.15–0.45 ng/ml and 3.1% in patients without elevated admission Tn-I (p = 0.09 between groups). There was no difference in hospital mortality (p = 0.25). At eight months of follow-up, there was no difference in out-of-hospital death (3.5%, 3.8% and 1.8%, p = 0.17, respectively), MI (2.6%, 3.8% and 2.9%, p = 0.94) or target lesion revascularization (9.0%, 8.3% and 11.5%, p = 0.47), and cardiac event-free survival was also similar between groups (p = 0.66). By multivariate analysis, Tn-I > 0.45 ng/ml was independently associated with in-hospital composite cardiac events [odds ratio (OR) = 2.4, p = 0.04] but not with out-of-hospital clinical events up to eight months.
In patients with ACS, early (within 72 h) catheter-based coronary intervention may attenuate the adverse prognostic impact of admission Tn-I elevation during eight months of follow-up despite a trend towards increased in-hospital composite cardiac events.
Patients who present with non-ST elevation acute coronary syndromes (ACS) are a heterogeneous group with diverse short- and long-term prognosis (1). Thus, early risk stratification and individualization of medical or invasive treatment is mandatory to achieve optimal therapeutic results. Troponin I (Tn-I) is a relatively new, highly specific and sensitive marker for myocardial damage (2–4)which was found elevated in 24% to 40% of patients with unstable coronary syndromes (5–7). Recent studies in patients with ACS found abnormal Tn-I levels on admission to predict adverse prognosis (6,7). Specifically, patients with ACS and elevated admission Tn-I (5–8)or troponin T (Tn-T) levels have sustained adverse short- and long-term clinical outcomes (8–12). In this high-risk group of patients, medical approach, including utilization of new antithrombotic agents (6,13), showed inconsistent therapeutic effect. The role of early invasive percutaneous approach in ACS patients, stratified by Tn-I levels on admission, has not been established.
The purpose of this study was to determine the impact of early catheter-based coronary intervention on clinical outcomes in a large consecutive patient population presenting with non-ST elevation ACS and stratified according to admission Tn-I levels.
The patient cohort includes a consecutive series of patients who were admitted with a clinical diagnosis of acute non-ST elevation coronary syndromes [progressive or worsening of angina for less than six weeks, rest angina with or without objective evidence of ischemia such as electrocardiographic or hemodynamic changes and recurrence of angina within 6 weeks of an acute myocardial infarction (MI)] who underwent early (within 72 h) catheter-based coronary interventions between July 1, 1996 and June 30, 1998. Non-Q-wave MI was considered to be present when admission creatine kinase MB (CK-MB) fraction was ≥2 times upper normal range (≥8 ng/ml). All patients have had routine admission Tn-I levels measured. Patients were divided into three groups according to admission Tn-I levels: Group I (n = 1,099) had no elevation (Tn-I < 0.15 ng/ml), Group II (n = 95) had intermediate Tn-I levels between 1 to 3 times the upper normal level (0.15–0.45 ng/ml) and Group III (n = 127) had more than 3 times the upper normal level (>0.45 ng/ml). Patients in the last group were further stratified to 3 subgroups: Group IIIa (n = 35) had Tn-I levels between 3–5 times the upper normal level (0.45–0.75 ng/ml), Group IIIb (n = 37) had 5–10 times the upper normal level (0.75–1.5 ng/ml) and Group IIIc (n = 55) had more than 10 times the upper normal level (>1.5 ng/ml). Patients were excluded from analysis if on admission they had:
1. a diagnosis of acute ST-elevation MI,
2. documented MI in the previous 24 h,
3. new pathologic Q-waves in at least two contiguous electrocardiogram leads,
4. the index admission had occurred within 14 days of a previous percutaneous coronary intervention or bypass surgery.
Troponin-I and CK-MB measurements
Blood samples were routinely acquired from all patients on hospital admission and at 6 and 18 to 24 h after the index procedure. If postprocedural Tn-I or CK-MB levels were elevated, serial measurements every 8 h were performed and the peak level was recorded. Troponin I levels were measured by paramagnetic-particle, chemiluminescent immunoenzymatic assay (Beckman, Coulter Inc., California). Using this assay, the sample can be accurately measured within the reportable range of the lower limit of detection and the highest calibrated value (0.03 to 50 ng/ml). The upper limit of the 95% nonparametric range for a presumably healthy population is below the mean minimum detectable concentration of the assay (<0.03 ng/ml), and the diagnostic cutoff value (as specified by the manufacturer) for MI based on receiver-operating characteristic (ROC) plots (14)is >0.15 ng/ml. This diagnostic cutoff is routinely used in our institution as the upper normal value. Creatine kinase MB fraction enzyme levels were determined by a two-site immunoenzymatic (“sandwich”) assay (Beckman). Using this assay, the sample can be accurately measured within the reportable range of the lower limit of detection and the highest calibrated value (0.3 to 300 ng/ml). The upper limit of the 95% nonparametric range for a healthy population is 0.3 to 4.0 ng/ml. A diagnostic cutoff of 4.0 ng/ml is considered as the upper normal value. Accordingly, the upper normal values used in the current study were 0.15 ng/ml and 4.0 ng/ml for Tn-I and CK-MB, respectively.
Procedural and follow-up data
Baseline demographics and in-hospital complications were confirmed by independent hospital chart review. All patients underwent pre- and postintervention 12-lead electrocardiogram (ECG). The diagnosis of Q-wave MI was based on CK-MB elevation ≥3 times upper normal values with the appearance of new pathologic Q-waves on postinterventional ECG. The diagnosis of non-Q-wave MI was based on CK-MB elevation ≥5 times normal values in the absence of new pathologic Q-waves on postintervention ECG. The diagnosis of in-hospital recurrent ischemia was based on symptoms associated with transient ST-segment deviation. Angiographic success was defined as <50% residual diameter stenosis with Thrombolysis in Myocardial Infarction (TIMI) 3 flow. In-hospital composite cardiac events include death, Q-wave MI and urgent need for coronary revascularization (bypass surgery or repeat angioplasty). Clinical success was defined as angiographic success without in-hospital composite cardiac events. Clinical outcomes at 30 days and 8-month follow-up data were obtained by serial telephone interviews by research nurses. All clinical events (death, MI), target lesion revascularization (TLR) or any cardiac event (death, MI, coronary angioplasty or bypass surgery) were adjudicated by accompanying source documentation reviewed by research nurses. The diagnosis of Q-wave and non-Q-wave MI during follow-up was based on hospitalization records and documented discharge summaries with clinical diagnosis of MI with CK-MB rise of ≥3 times upper normal with or without the appearance of new pathologic Q-waves on surface ECG.
Continuous variables are presented as mean ± 1 standard deviation. In-hospital composite cardiac events and late clinical events were analyzed as hierarchical end points. Categorical data are presented as percent frequency and compared between groups using chi-square statistics or Fisher exact test. Multivariate analysis was performed by SAS (Cary, North Carolina) Logistic Regression Statistics. Survival curves for out-of-hospital survivors were calculated and displayed using the SAS LIFETEST procedure. Wilcoxon statistics were used for survival comparison between groups (normal, 1–3 and >3 times upper normal admission Tn-I levels). The means of continuous values were compared using the unpaired Student ttest. Regression analysis was used to correlate between admission Tn-I and CK-MB levels. A p value of <0.05 was accepted as statistically significant.
Table 1lists the baseline characteristics of all treated patients according to admission Tn-I levels (>3, 1–3 times upper normal and normal Tn-I levels). Patients with normal Tn-I had more frequent prior revascularization procedures and, less often, prior MI. They also had significantly higher left ventricular ejection fraction. Hypercholesterolemia was found more frequently in the group with normal Tn-I levels.
Troponin-I and CK-MB measurements
On hospital admission, any elevated level of Tn-I (>0.15 ng/ml) and Tn-I level >3 times upper normal occurred in 16.8% and 9.6% of the patients, respectively. Creatine kinase MB fraction ≥2 times upper normal (preintervention non-Q-wave MI) was elevated in 8.7% of all patients and in 66.7% of patients with Tn-I levels >3 times upper normal (Table 1). The mean Tn-I levels were 31.3 ± 16.4 ng/ml, 0.25 ± 0.09 ng/ml and 0.03 ± 0.03 ng/ml in the >3, 1–3 times upper normal and normal Tn-I level groups (p = 0.0001). The mean CK-MB levels were 78.9 ± 137 ng/ml, 5.8 ± 8.0 ng/ml and 2.2 ± 7.3 ng/ml in those groups, respectively (p = 0.0001). A regression model yielded a significant correlation (r = 0.78, p = 0.0001) between admission Tn-I and CK-MB levels with the following regression equation: CK-MBadm= 0.298 × Tn-Iadm+ 8.5.
Lesion location and characteristics
There were no differences in lesion location or characteristics between the three groups except for lower rate of right coronary artery lesions in patients with intermediate Tn-I level (30.9%, 21.0% and 29.3% in the >3, 1–3 times upper normal and normal Tn-I levels, respectively, p = 0.04). The overall rate of saphenous vein grafts involvement was 12.4% and was similar between groups (p = 0.78). There were no differences in pre- or postprocedural lesion diameter stenosis between the three groups (data not shown).
Percutaneous coronary intervention was performed earlier in the group with normal Tn-I (0.8 ± 1.0 days) compared with patients with “intermediate” Tn-I levels (1.2 ± 1.2 days) and more than 3 times normal Tn-I (1.1 ± 1.2 days), p = 0.001. The rate of stent use was similar among the three groups and rotational atherectomy was used less frequently in patients with highest Tn-I (Table 2). The periprocedural use of abciximab was more frequent in patients with Tn-I > 0.45 ng/ml (12.9% vs. 6.2% and 6.4% in >3, 1 to 3 times upper normal and normal Tn-I levels, p = 0.019).
Angiographic success was achieved in 98% of patients with similar rates among the three groups (Table 3). Clinical success was somewhat lower in the highest Tn-I group (p = 0.12). The rate of in-hospital mortality, Q-wave MI, urgent bypass surgery or repeat angioplasty was similar among the three groups (Table 3). Nonetheless, composite cardiac events (death, Q-wave MI and urgent revascularization) showed a trend towards increased clinical event rate in patients with more than 3 times upper normal Tn-I level (6.1% vs. 1.0% and 3.1% in >3, 1 to 3 times normal and normal Tn-I level, p = 0.09).
Between hospital discharge and 8 months, there was no difference in death, MI or TLR between the three groups (Table 3). Actuarial out-of-hospital event-free survival curves for any cardiac events up to 8 months (death, MI, bypass surgery or repeat angioplasty) are shown in Figure 1. Event-free survival was similar in all groups for both end points, (p = 0.66 for any cardiac event and p = 0.15 for TLR). However, cumulative mortality (in-hospital and out-of-hospital) tended to be higher in patients with elevated admission Tn-I (5.1%, 4.7% and 2.4% in the >3, 1–3 times upper normal and normal Tn-I levels, p = 0.09, respectively).
In-hospital composite cardiac events, 30 days and eight months out-of-hospital clinical outcomes were not significantly different among patients with normal, 1 to 3, 3 to 5, 5 to 10 and >10 times upper normal Tn-I level (Table 4).
Logistic regression analysis was used to identify independent predictors for in-hospital composite cardiac events. Variables in the model were expected to correlate with outcomes and included: age, gender, prior coronary angioplasty, prior bypass surgery, diabetes mellitus, left ventricular ejection fraction, saphenous vein graft (SVG) lesion, admission CK-MB ≥2 times upper normal, admission Tn-I levels >3 times upper normal, abciximab administration and stents implantation. Using this model, Tn-I level above 3 times upper normal (>0.45 ng/ml) on admission was found to be the only predictor associated with in-hospital composite cardiac events [odds ratio (OR) 2.4, 95% confidence interval (CI) 1.03–5.5, p = 0.04]. A second logistic regression analysis was performed to identify independent predictors of late clinical outcomes (death, MI and repeat revascularization) or TLR alone. A history of coronary angioplasty, diabetes mellitus or prior bypass surgery was found to predict any cardiac events. Prior coronary angioplasty and diabetes mellitus were associated with TLR (Table 5). Neither admission Tn-I nor CK-MB elevation predicted out-of-hospital to eight months clinical outcome or TLR in this analysis.
In a large consecutive group of patients with ACS who underwent early (within 72 h) catheter-based coronary revascularization, a trend was found for increased in-hospital composite cardiac events (death, Q-wave MI, urgent revascularization) in patients with admission Tn-I level more than three times normal (>0.45 ng/ml). This trend was not observed in patients with ACS and lower level of admission Tn-I elevation (0.15–0.45 ng/ml). This study also found that early coronary intervention was associated with favorable out of hospital to eight months clinical outcomes regardless of Tn-I level on admission. Moreover, admission Tn-I levels more than three times upper normal was found as a predictor for in-hospital composite end point but not for any adverse late clinical outcomes during the eight months follow-up period. Neither Tn-I level between 1–three times upper normal (0.15–0.45 ng/ml) nor a confirmed diagnosis of non-Q-wave MI on admission (defined as CK-MB level ≥2 times upper normal) was associated with increased in-hospital composite cardiac events, late adverse clinical outcomes or TLR at 8 months follow-up. Those results indicate that, in patients with ACS, early catheter-based coronary revascularization may attenuate the adverse prognostic impact of admission Tn-I elevation.
This study shows that admission Tn-I levels more than three times the upper normal in ACS patients undergoing early coronary intervention is an independent predictor for in-hospital adverse cardiac end points. This observation is in accord with previous studies of patients with ACS (5–8). However, unlike prior experiences which showed increased death and MI at 30 days (6), 42 days (7)and one-year follow-up (5)in patients with elevated admission Tn-I, this study demonstrated overall low mortality and MI rates which were not significantly different among the three groups. Furthermore, unlike previous studies (7,12)a subgroup analysis of the patients with elevated Tn-I >3 times upper normal (up to >10 times upper normal) showed similar short- and long-term clinical event rates regardless of the Tn-I levels. Compared with previous medical management trials in ACS, this study found similar event rates among patients with normal Tn-I but lower adverse cardiac events in patients with elevated Tn-I levels (5,7,13). This may be due to an early catheter-based coronary revascularization approach in ACS that has resulted in improving the clinical outcome of high risk patients with elevated Tn-I. Such early interventional strategies may have less impact on patients at lower risk as reflected by normal Tn-I levels on admission. In this respect, our results are in accord with the beneficial responses obtained by prolonged antithrombotic treatment (13)and preangioplasty abciximab administration (15)in patients with unstable coronary artery disease and elevated Tn-T levels.
There are several potential mechanisms by which early coronary revascularization may beneficially alter the clinical course of ACS, especially when associated with elevated Tn-I levels. Mechanical revascularization may preserve myocardial function and obtain electrical stability similar to the effect of early revascularization in acute Q-wave MI (16,17). In a substudy of the Global Use of Strategies to Open Occluded Coronary Arteries in Acute Coronary Syndromes (GUSTO IIB) trial, it was indeed found that patients with recurring ischemic episodes benefitted from an early strategy of coronary revascularization (18). Those data are in accordance with our experience, as patients with ACS and Tn-I elevation may reflect a higher degree of myocardial vulnerability during episodes of coronary occlusion or pre-existing myocardial damage due to coronary thrombotic microembolization (5,19). Ultimately, these patients can benefit most from an early strategy of revascularization and myocardial salvage.
The main limitation of our study is its retrospective design and the absence of a comparison to a medical management group. There is a considerable debate regarding the optimal management strategy of patients with ACS (20). Without the incorporation of a “conservative” treatment arm, this debate cannot be resolved by our own findings. Two recent randomized clinical trials addressing this issue have suggested a complimentary therapeutic effect for the interventional and medical management strategies without significant differences in short-term clinical outcomes between the two approaches (1,21,22). In addition, the overall use of abciximab in this study was relatively low and, thus, precludes analysis of the potential beneficial effect of the combined utilization of abciximab and early catheter-based coronary intervention. Such an approach has been recently shown to have a substantial beneficial effect on patients with refractory unstable angina and elevated Tn-T levels (15). Our study was also limited by its statistical power to detect significant differences in hospital (beta = 0.5) and long-term (beta = 0.3) composite cardiac events between the study groups. Assuming a similar ratio of 1:8 in patients with normal versus elevated Tn-I level more than three times normal, according to sample size calculation it would be required that 15,345 patients achieve an 80% statistical power in order to detect significant differences in long-term death rates between groups with similar patient demographics. Furthermore, the conclusions of this study may not to be generalized to those patients excluded from the analysis. Finally, it is possible that patients with large non-Q-wave MI or left ventricular dysfunction should be considered to be at higher risk regardless of Tn-I levels. The overall number of patients with large non-Q-wave MIs in this study was relatively small and, hence, precluded such subgroup analysis.
This study has several potential clinical implications. As previously suggested (5,7,8), routine admission and possibly serial measurements of Tn-I levels should be obtained in patients with non-ST elevation ACS and may be used for risk stratification and tailoring therapeutic strategies. Early coronary angiographies and “aggressive” revascularization strategies may be warranted to improve the clinical outcome, especially in patients with elevated admission Tn-I levels >0.45 ng/ml. In summary, according to our experience, patients with ACS and elevated Tn-I levels who undergo early catheter-based coronary revascularization have somewhat increased in-hospital complications, but their long-term (eight month) clinical outcomes are favorable regardless of Tn-I levels on admission.
- acute coronary syndromes
- confidence interval
- creatine kinase MB fraction
- myocardial infarction
- odds ratio
- target lesion revascularization
- troponin I
- troponin T
- Received March 4, 1999.
- Revision received June 18, 1999.
- Accepted August 18, 1999.
- American College of Cardiology
- Théroux P.,
- Fuster V.
- Adams J.E. III.,
- Bodor G.S.,
- Dávila-Román V.G.,
- et al.
- Galvani M.,
- Ottani F.,
- Ferrini D.,
- et al.
- TRIM study group,
- Lüscher M.S.,
- Thygesen K.,
- Ravkilde J.,
- Heickendorff L.
- Newby L.K.,
- Christenson R.H.,
- Ohman E.M.,
- et al.
- Stubbs P.,
- Collinson P.,
- Moseley D.,
- Greenwood T.,
- Noble M.
- FRISC study group,
- Lindahl B.,
- Venge P.,
- Wallentine L.
- Fragmin In Unstable Coronary Artery Disease (FRISC) study group,
- Lindahl B.,
- Venge P.,
- Wallentine L.
- Zweig M.H.,
- Campbell G.
- Hohnloser S.H.,
- Frank P.,
- Klingenheben T.,
- Zabel M.,
- Just H.
- White H.D.,
- Cross D.B.,
- Elliott J.M.,
- Norris R.M.,
- Yee T.W.
- GUSTO-IIb investigators,
- Armstrong P.W.,
- Fu Y.,
- Chang W.C.,
- et al.
- Abdelmeguid A.E.,
- Topol E.J.
- TIMI IIIB Investigators
- Veterans Affairs Non-Q-Wave Infarction Strategies In Hospital (VANQUISH) trial investigators,
- Boden W.E.,
- O’Rourke R.A.,
- Crawford M.H.,
- et al.