|
|
||||||||||
|
J Am Coll Cardiol, 1999; 33:73-78 © 1999 by the American College of Cardiology Foundation |



* Department of Medicine, Brigham & Womens Hospital, Harvard Medical School, Boston, Massachusetts, USA
Department of Medicine, St. Louis University, St. Louis, Missouri, USA
Department of Medicine, University Hospitals of Cleveland, Case Western Reserve University, Cleveland, Ohio, USA
Biogen Inc., Cambridge, Massachusetts, USA
|| Ocala Heart Institute, Ocala, Florida, USA
Manuscript received April 3, 1998; revised manuscript received July 7, 1998, accepted September 10, 1998.
Address for correspondence: Robert N. Piana, MD, Cardiovascular Division, Brigham & Womens Hospital, Boston, Massachusetts 02115
RNPiana{at}BICS.BWH.Harvard.Edu
| Abstract |
|---|
|
|
|---|
Background. The acute complications of angioplasty are related to baseline clinical and angiographic variables, and early complications adversely affect long-term outcome. However, the predictors of enduring success after uncomplicated angioplasty are less well defined.
Methods. Of 4,098 patients undergoing angioplasty in the Hirulog Angioplasty Study, 3,899 (95%) had a successful procedure without in-hospital death, emergent bypass surgery or clinical evidence of myocardial infarction. Baseline and procedural variables for these 3,899 patients were examined.
Results. Major adverse cardiac events occurred in 22% of the patients with initially successful procedures at 6 months: death in 1%, myocardial infarction in 2% and repeat revascularization in 21%. Univariable predictors of increased events included successful salvage from abrupt vessel closure (p < 0.001), emergency stenting (p < 0.001), multilesion angioplasty (p < 0.001), diabetes (p = 0.02), target lesion in the left anterior descending artery (p = 0.02), unstable angina (p = 0.03) and smaller final luminal diameter (p = 0.04). There was a trend toward increased events among patients with prior angioplasty (p = 0.08), but asymptomatic elevation of the creatine kinase was not predictive (p = 0.5). In a multivariable model, abrupt vessel closure was the strongest independent predictor of major adverse cardiac events at 6 months (p < 0.001; odds ratio [95% confidence interval] = 3.6 [2.5 to 5.1]), while multivessel angioplasty, target lesion in the left anterior descending artery and diabetes also remained independent predictors (all p
0.02).
Conclusions. This analysis suggests that "uncomplicated" abrupt vessel closure is a powerful predictor of adverse clinical outcome following successful angioplasty. Improved techniques to reduce abrupt closure during angioplasty are thus urgently needed, and patients who experience "uncomplicated" closure require closer surveillance during follow-up.
| ||||||||||||||
See page 79(1724) have focused on predictors of enduring clinical success during the ensuing 6 months, a more relevant end point inclinical practice. Long-term success is clearly reduced when major ischemic complications occur at the time of the initial procedure (2,3). The predictors of 6 month clinical outcome among successfully treated patients without in-hospital complications remain uncertain, however. We utilized data from the recent Hirulog Angioplasty Study (25) to examine the predictors of clinical outcome at 6 months following successful percutaneous coronary revascularization.
| Methods |
|---|
|
|
|---|
Hospital outcome was analyzed for all 4,098 patients. In this substudy, 6-month clinical outcome was analyzed for the 3,899 of 4,098 patients (95%) who underwent successful coronary intervention without a major ischemic complication in-hospital. Because bivalirudin did not produce a significant treatment effect in the study (25), all patients were combined as a single group for the purposes of this analysis.
Definitions. Major ischemic complications were defined as in-hospital death, MI or emergent coronary artery bypass graft surgery (CABG). The definition of MI was a rise in the total serum creatine kinase (CK) concentration above twice the upper limit of normal with an MB fraction of at least 4.0% in association with either ischemic chest pain lasting longer than 30 min, or a new two-step Q wave change (according to the Minnesota code) (26), persistent ST segment or T wave changes or a new left bundle branch block. Reelevation of the total or CK-MB to above the previous nadir in association with ischemic chest pain or the defined ischemic electrocardiographic changes was also classified as reinfarction. Event-free survival at 6 months was defined as freedom from major adverse cardiac events (MACE), defined as death, MI or repeat revascularization.
A procedure was defined as successful if it was performed without a major in-hospital ischemic complication. Abrupt vessel closure (AVC) included both threatened and established closure. Threatened closure after initially successful dilation of the target lesion was defined as a recurrent stenosis of more than 50% with flow less than Thrombolysis in Myocardial Infarction trial (TIMI) grade 3 (27) requiring intracoronary stenting, thrombolytic therapy or repeat cardiac catheterization (28,29). Established AVC after initially successful dilation was defined as total or subtotal occlusion of the target lesion with impaired flow to TIMI grade 0 or 1 (2). "Complicated" AVC was defined as threatened or established closure that led to a major in-hospital ischemic complication. "Uncomplicated" AVC was a closure that was successfully reversed without a resultant major in-hospital ischemic complication. In the angiographic analysis, thrombus was defined as a filling defect, haze or ulceration at the lesion site, or a total occlusion.
Data collection. Clinical follow-up was obtained at 3 and 6 months postprocedure. Measurement of the total CK was performed at 8 and 16 h after angioplasty, with MB determinations done if the total CK was elevated. Of 4,098 patients in the trial, 4,068 (99%) had at least one CK measurement after angioplasty. All data were submitted to the data coordinating center (ClinTrials Research, Research Triangle Park, NC) and were verified for accuracy against source documents by clinical monitors from ClinTrials.
Angiograms were reviewed at the core angiographic laboratory by experienced angiographers blinded to treatment assignment using electronic digital calipers according to previously validated methods (30). Of 4,098 patients in the trial, 4,056 (99%) had films reviewed at the core laboratory. Angiograms were analyzed on 368 of the 378 (97%) patients who experienced AVC during the study.
Statistical analysis. Categorical variables were compared using Fisher exact test or Pearson chi-square test where appropriate. Student t test was used to compare continuous variables between groups. Logistic regression, incorporating univariates with at least borderline significance (p < 0.10), was used to determine predictors of ischemic complications during the 6 months after successful angioplasty. Statistical analysis was performed using SPSS. Baseline data were unavailable on 3 of 4,098 patients (0.1%) who underwent angioplasty, and follow-up data were incomplete for 195 patients (5%). Angiographic data for 42 patients (1%) were unavailable.
| Results |
|---|
|
|
|---|
|
|
|
Among the 3,899 patients with successful angioplasty, 155 (4%) developed an elevation of the CK to greater than twice the upper limit of normal without ischemic symptoms or electrocardiographic changes. These were not classified as MI events according to the predetermined definition. Asymptomatic elevations of CK were detected in 139 of 3,649 (0.4%) successful procedures without an episode of threatened or established AVC, compared with 16 of 250 (6%) procedures with uncomplicated AVC (p = 0.04).
Follow-up after successful angioplasty. Clinical status at 6 months was available for 3,704 of the 3,899 patients (95%) with successful index procedures. The rate of follow-up was equal for those with and without AVC (both 95%) during the index procedure. Patients without follow-up were younger (59 ± 12 vs. 62 ± 11 years; p < 0.001) and had a lower prevalence of prior angioplasty (17% vs. 27%; p = 0.002) compared to those with follow-up. Otherwise the two groups were similar.
Major adverse cardiac events occurred by 6 months in 825 of 3,704 (22%) patients, including death in 44 (1%), MI in 82 (2%) and repeat revascularization in 770 (21%). Clinical and angiographic predictors of an increased incidence of MACE in follow-up (Table 3) included AVC (p < 0.001), requirement for stent placement during the index procedure (p < 0.001), multilesion angioplasty (p < 0.001), diabetes (p = 0.02), angioplasty of the left anterior descending artery (p = 0.02), unstable angina as opposed to post-MI angina (p = 0.03) and a smaller postprocedural luminal diameter (p = 0.04). There was also a trend for increased MACE among patients who had undergone prior angioplasty (p = 0.08). Lesion length, reference diameter, final percent stenosis, residual stenosis of greater than 50%, persistent dissection at completion of the procedure and asymptomatic elevation of the CK were not statistically significantly associated with increased events in follow-up. In a multivariable analysis incorporating the univariable predictors, AVC was the strongest independent predictor of MACE at 6 months (p < 0.001; odds ratio [95% confidence interval] = 3.6 [2.5 to 5.1]), while multilesion angioplasty, target lesion in the left anterior descending artery and diabetes also remained independent predictors.
|
|
| Discussion |
|---|
|
|
|---|
The prognostic significance of transient vessel closure during angioplasty has previously been suggested by data from the 19851986 National Heart, Lung and Blood Institute Registry. At 2 years of follow-up, patients with successfully reversed AVC had reduced survival (92% vs. 96%) and increased repeat revascularization rates (42% vs. 31%) compared to those without transient occlusion (2). Tanaglia also documented an increased rate of death, MI and repeat angioplasty by 1 year among patients salvaged from abrupt closure at Duke (33).
Repeat revascularization procedures accounted for the large increase in adverse events among patients with uncomplicated AVC in the present study (44% vs. 19% among those without AVC; p < 0.001). This 44% rate is higher than the 13% incidence at 1 year observed by Abdelmeguid in a smaller group of 88 patients with transient vessel closure (34). However, Abdelmeguids single center, retrospective analysis excluded patients treated with bailout stenting, which was required in 44% of our patients with uncomplicated AVC, despite being officially discouraged by the study. The decision to deploy a stent may have reflected more complicated vessel disruption, portending a higher risk prognosis than for AVC manageable with repeat balloon angioplasty alone. Abdelmeguids report was also not restricted to patients with unstable angina. The 1-year repeat revascularization rate of only 11% among patients without AVC in Abdelmeguids study contrasts with the 20% to 28% incidence at 6 months observed in the Hirulog study and many other recent multicenter trials of balloon angioplasty for unstable angina (22,35,36), suggesting possible differences in the patient populations.
Several explanations for the adverse prognostic significance of uncomplicated AVC are possible. Patients with "uncomplicated" AVC had more complex target lesions and achieved less optimal postprocedural angiographic results than patients who had successful angioplasty with no abrupt closure event. Unplanned stenting was also required in 44% of patients with "uncomplicated" AVC. Bailout stenting is highly successful in the acute setting, but it is accompanied by higher rates of subacute thrombosis and restenosis than after elective stenting (21,3740). Periprocedural myonecrosis also occurred significantly more frequently following "uncomplicated" AVC. Even minor elevations of the CK after coronary intervention have been found by some investigators to predict reduced event-free survival (34,4143).
The significant increase in the rate of MACE among patients with uncomplicated AVC cannot be fully explained by adverse morphologic features or more complex interventions, however. In the present study, neither lesion morphology nor elevation of the CK to above twice the upper limit of normal among patients with successful angioplasty was predictive of MACE at 6 months. In addition, multivariable analysis demonstrated that the AVC event itself was the most powerful independent predictor of MACE in follow-up, with an odds ratio of 3.6. Thus, although clinical and angiographic characteristics are important in assessing patients risk following angioplasty, the occurrence of transient coronary occlusion during the procedure appears to have a significant impact on outcome independent of these other data.
Limitations. This was a retrospective analysis of subgroups from the Hirulog Angioplasty Trial, which was limited to patients with unstable angina and was completed prior to widespread use of glycoprotein IIb/IIIa inhibitors in coronary intervention. The prognostic implications of "uncomplicated" AVC may be different in stable angina or when IIb/IIIa antagonists are utilized. Elevation of the CK was detected in only 0.4% of non-AVC patients and 6% of AVC patients following successful angioplasty, possibly reflecting less stringent surveillance than in some other studies. Stricter surveillance and routine measurement of the CK-MB isoform may have uncovered a greater prognostic significance of enzyme elevation. Given the high prevalence of salvage stenting among patients with uncomplicated abrupt vessel closure, these results cannot necessarily be extrapolated to a population with transient occlusion that can be reversed with balloon angioplasty alone.
Conclusions. Abrupt vessel closure during angioplasty for unstable or post-MI angina has a significant, independent, adverse effect on clinical outcome at 6 months. This holds true even when the closure does not cause a major ischemic event acutely, and is thus traditionally considered "uncomplicated" by the operator at the time of the procedure. Abrupt closure may thus serve as a general marker that incorporates unfavorable clinical characteristics, adverse features of the lesion or a less satisfactory angiographic result when the closure is salvaged. Fortunately, such uncomplicated abrupt closure events do not appear to increase subsequent rates of myocardial infarction or death. They do portend a significant increase in the need for repeat revascularization by 6 months, however. As interventional techniques are applied to more severely ill patients with increasingly complex coronary artery disease, the potential exists for the rates of transient or established vessel closure to rise. Acute complication rates may not reflect these instances of transient occlusion, because two thirds of such events can be reversed using the current interventional armamentarium without a recognized major ischemic event. Nevertheless, our analysis suggests that these salvaged patients constitute a population at higher risk for subsequent ischemic events. Thus, in approaching increasingly difficult scenarios for coronary intervention, techniques designed to reduce rates of abrupt closure should be applied when possible, and the potential medical and economic implications of even transient occlusion should be considered prior to attempts at percutaneous revascularization.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T Nageh, R A Sherwood, B M Harris, and M R Thomas Prognostic role of cardiac troponin I after percutaneous coronary intervention in stable coronary disease Heart, September 1, 2005; 91(9): 1181 - 1185. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. W. Stone, R. Mehran, G. Dangas, A. J. Lansky, R. Kornowski, and M. B. Leon Differential Impact on Survival of Electrocardiographic Q-Wave Versus Enzymatic Myocardial Infarction After Percutaneous Intervention: A Device-Specific Analysis of 7147 Patients Circulation, August 7, 2001; 104(6): 642 - 647. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. E. Kimmel, A. R. Localio, C. Brensinger, C. Miles, J. Hirshfeld, H. L. Haber, and B. L. Strom Effects of Coronary Stents on Cardiovascular Outcomes in Broad-Based Clinical Practice Arch Intern Med, September 25, 2000; 160(17): 2593 - 2599. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |