Author + information
- Received August 4, 2004
- Revision received October 14, 2004
- Accepted October 18, 2004
- Published online February 1, 2005.
- C. Michael Gibson, MS, MD*,†,* (, )
- Ajay J. Kirtane, MD*,†,
- Keith Boundy, BA†,
- Hung Ly, MD†,
- Dimitrios Karmpaliotis, MD*,
- Sabina A. Murphy, MPH†,
- Robert P. Giugliano, MD, SM†,
- Christopher P. Cannon, MD†,
- Elliott M. Antman, MD†,
- Eugene Braunwald, MD†,
- TIMI Study Group
- ↵*Reprint requests and correspondence:
Dr. C. Michael Gibson, 350 Longwood Avenue, First Floor, Boston, Massachusetts 02115
Objectives We hypothesized that <0% residual stenosis (RS) after rescue/adjunctive percutaneous coronary intervention (PCI) following fibrinolytic administration in ST-segment elevation myocardial infarction (STEMI) would be associated with improved outcomes.
Background Prior studies have associated larger lumen diameters after PCI with reduced rates of restenosis and target vessel revascularization.
Methods Data were drawn from 748 patients with open epicardial arteries and with optimal luminal results (RS <20%) following rescue/adjunctive PCI after fibrinolytic administration in six STEMI trials. Patients were divided into two groups: 1) <0% RS and 2) 0% to 20% RS.
Results A RS <0% was associated with greater gains in lumen diameter and smaller reference diameters after PCI (p < 0.001 for each), with a trend toward less frequent Thrombolysis In Myocardial Infarction flow grade (TFG) 3. A RS <0% was associated with a greater incidence of abnormal post-PCI Thrombolysis In Myocardial Infarction myocardial perfusion grades (TMPGs) (odds ratio 2.6 [1.2 to 5.9] for TMPG 0/1/2, p = 0.02), even when the analysis was restricted to patients with post-PCI TFG 3.
Conclusions A RS <0% following rescue/adjunctive PCI after fibrinolytic therapy for STEMI was independently associated with a reduction in the frequency of normal myocardial perfusion. Potential mechanisms of this finding include greater downstream embolization, increased stimulation of arterial stretch receptors with resultant coronary vasoconstriction, and increased vessel-wall injury after PCI. These findings suggest that additional prospective studies are needed to assess optimal RS that minimizes long-term restenosis without adverse effects.
Larger post-procedure minimum lumen diameters (MLDs) have been associated with procedural success and reduced rates of restenosis and target vessel revascularization among patients undergoing percutaneous coronary intervention (PCI) (1,2). These findings have led to the dictum that “bigger is better” in PCI, which, clinically, often translates into a goal of achieving minimal and even <0% residual stenoses (RSs) to achieve larger MLDs. Although the relationship between larger post-procedure MLD and reduced rates of restenosis has been well-established among lower-risk patients, the relationship between oversizing and angiographic and clinical outcomes has not been well-characterized in higher-risk patients.
Several studies in ST-segment elevation myocardial infarction (STEMI) have demonstrated that the restoration of normal myocardial perfusion is associated with improved outcomes (3,4). The goal of the present study was to evaluate the associations of achieving <0% RS during rescue/adjunctive PCI with both angiographic measures of myocardial perfusion and clinical outcomes in STEMI.
Data were drawn from STEMI patients who underwent successful rescue/adjunctive PCI fibrinolytic therapy or combination fibrinolytic therapy in the Thrombolysis In Myocardial Infarction (TIMI)-10A, -10B, and -14; Integrilin and Tenecteplase in Acute Myocardial Infarction (INTEGRITI); Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction Treatment (ENTIRE-TIMI)-23; and Fibrinolytics and Aggrastat with ST segment Resolution (FASTER) trials.
Angiography was performed at 60 or 90 min after fibrinolytic therapy. All angiograms were assessed offline by an angiographic core laboratory. Pre-PCI angiograms were assessed at the 60-min time point unless these data were not available, in which case 90-min data were used. Quantitative coronary angiography was performed, and assessment of Thrombolysis In Myocardial Infarction flow grades (TFGs), corrected Thrombolysis In Myocardial Infarction frame counts (CTFCs), and Thrombolysis In Myocardial Infarction myocardial perfusion grades (TMPGs) were performed as previously described. Residual stenosis was defined as the difference between the normal reference segment (REF) diameter and the MLD of the culprit lesion after PCI, divided by the REF. The REF was calculated as the average diameter from adjacent normal segments proximal and distal to the lesion. For the TIMI-10 and -14 trials, a normal segment proximal to the lesion was used as the REF unless it could not be assessed, in which case the distal diameter was used. The balloon/artery ratio was defined as the maximum balloon diameter divided by the pre-PCI REF.
A total of 1,400 patients underwent rescue/adjunctive PCI at the discretion of individual operators. A total of 1,305 patients had a patent epicardial artery (TFG 2/3) after PCI. Among these patients, 748 had an optimal post-PCI RS by American College of Cardiology (ACC)/American Heart Association (AHA) guidelines (<20% RS) (5). In the primary analysis, RS was analyzed as a categorical variable in 2 groups: <0% RS (n = 111) (Fig. 1)and 0% to 20% RS (n = 637). For further comparison, 434 patients with patent epicardial arteries after PCI but with suboptimal post-PCI luminal results (20% to 50% RS) from the same trials were included.
All analyses were performed using Stata version 7.0 or later (Stata Corp., College Station, Texas). Where data were not normally distributed as assessed by the Shapiro-Wilk test, continuous variables are reported as medians (25%, 75% interquartile range), and the Wilcoxon rank sum test was used for comparisons between groups. The chi-square test was used for the analysis of categorical variables. A logistic regression model was constructed with post-PCI TMPG 0/1/2 as the dependent outcome, including differences in baseline characteristics (p < 0.10 in univariate analyses) as well as other known covariates of post-PCI TMPG in the dataset. In the analysis of clinical outcomes, a multivariate model was constructed including known correlates of mortality in the dataset.
Baseline pre-PCI characteristics
Compared to patients with 0% to 20% RS, patients with <0% RS had longer times from symptom onset to treatment (3.3 [2.3, 4.8] h vs. 2.8 [2.0, 3.9] h, p = 0.003) and were older (61 [50, 69] vs. 56 [49, 65] years, p = 0.04). There were no differences in gender, history of diabetes, hypertension or hypercholesterolemia, prior myocardial infarction, smoking history, heart rate and blood pressure on admission, or Killip class. Angiographic characteristics before PCI are presented in Table 1.Pre-PCI MLD, REF, RS, artery patency, TFG 3, and abnormal TMPG were not appreciably different among patients with <0% RS compared to patients with 0% to 20% RS.
Association of <0% RS with angiographic and clinical outcomes
A RS <0% was associated with larger balloon diameters (in the 347 patients with available balloon diameter) and post-PCI MLD (Table 2).A RS <0% was also associated with a smaller REF after PCI (p < 0.001). The mean change in REF from before to after PCI varied with REF decrease of 0.10 mm among patients with <0% RS vs. REF increase of 0.04 mm among patients with 0% to 20% RS (Fig. 2).Post-PCI TFG 3 was observed slightly less frequently among patients with <0% RS (p = 0.09). There were no differences in rates of angiographically evident residual thrombus, dissection, distal embolization, or occurrence of no reflow.
A RS <0% was associated with a greater incidence of abnormal myocardial perfusion after PCI (69.1% TMPG 0/1/2 vs. 52.7%, p = 0.042). Even among patients with TFG 3 after PCI, TMPG was abnormal in 68.7% of patients with <0% RS versus 50.0% of patients with 0% to 20% RS (p = 0.036) (Fig. 3).A RS <0% was independently associated with abnormal TMPG (odds ratio [OR] for TMPG 0/1/2 of 2.6 (1.2 to 5.9), p = 0.02) after adjustment for age, gender, time to treatment, smoking status, left anterior descending coronary artery (LAD) lesion location, abnormal Killip class, three-vessel disease, pre-PCI percent stenosis, pre-PCI TMPG 3, and pre-PCI epicardial patency. The independent association between <0% RS and abnormal TMPG (OR 2.8 [1.2 to 6.8], p = 0.02) remained even when stratified in patients with TFG 3 after PCI.
When RS was analyzed as a continuous variable within the study population of patients with an optimal PCI result (<20% RS), the median post-PCI RS was lower among patients with abnormal TMPG (5% vs. 8%, p = 0.01). In the multivariate model, there was a trend for a lower RS to be associated with abnormal TMPG (p = 0.08).
A RS <0% was independently associated with both in-hospital and 30-day mortality (OR for 30-day mortality 2.8 [1.0 to 7.6], p = 0.042; OR for in-hospital mortality 3.1 [1.1 to 9.3], p = 0.039) in a multivariate model adjusting for covariates previously associated with mortality in this dataset (3): age, gender, time to treatment, LAD lesion location, and Killip class. Other in-hospital and 30-day clinical outcomes were similar among groups. There was no association between RS analyzed as a continuous variable and clinical outcomes.
In additional analyses, there appeared to be a U-shaped curve to the rates of abnormal TMPG among all patients with RS <50%, with the lowest frequency of TMPG 0/1/2 among patients with 0% to 20% RS (52.7%) and higher rates among patients with <0% RS (69.1%) and 20% to 50% RS (58.0%) (p = 0.09). Among all patients with <50% RS, the lowest incidence of death was among patients with 0% to 20% RS (2.5%), with a greater incidence among patients with <0% RS (6.3%) and 20% to 50% RS (4.2%) (p = 0.08).
Larger post-procedure MLD after PCI has been associated with reduced rates of restenosis and target lesion revascularization, leading to the assertion that “bigger is better” (1,2). However, improvements in post-procedure blood flow may be as important as achieving larger MLD in reducing adverse events, suggesting that both “bigger and faster is better” (6–8). In this analysis, we initially hypothesized that a <0% RS (i.e., a MLD < REF) would be associated with improved outcomes in high-risk patients with STEMI. Contrary to our hypothesis and to other studies in lower-risk patients, we found that there may be a limit to the optimal RS during PCI. In this analysis, we found that <0% RS was associated with impaired myocardial perfusion and increased mortality after rescue/adjunctive PCI in STEMI.
There are several mechanisms that may contribute to potentially deleterious effects of achieving a <0% RS during PCI, particularly in higher-risk patients such as those with STEMI. Increased mechanical stretch and endothelial injury could result in the release of vasoactive substances (9–11) such as serotonin, thromboxane A2, thrombin, and endothelin-1, with resulting vasoconstriction. The adrenergic system has also been implicated during PCI, and treatment with alpha-antagonists after stent placement attenuates vasoconstriction and post-ischemic left ventricular dysfunction (11,12). Percutaneous coronary intervention that achieves a <0% RS may exacerbate these processes and lead to diffuse vasoconstriction with impaired myocardial perfusion. Indeed, the decline in REF after PCI among patients with <0% RS in this study is consistent with this phenomenon.
Patients with STEMI have a greater thrombotic and inflammatory burden than patients with stable angina, and the performance of PCI resulting in <0% RS could further exacerbate platelet activation and aggregation, resulting in increased inflammation and vasoconstriction. It could also be speculated that in patients with <0% RS after PCI, greater disruptions of plaque and thrombus may have led to increased distal microembolization and microvascular dysfunction, or that patients in whom <0% RS was achieved may have had more thrombotic lesions that were more likely to embolize distally or cause vasospasm.
To this end, intravascular ultrasound (IVUS) studies have demonstrated that greater reductions in plaque area during PCI among patients with acute coronary syndrome (ACS) are independently associated with increased creatinine kinase release (13), worsened epicardial flow (14), and worsened myocardial perfusion (15). In contrast to stable angina, the presence of ACS has been independently associated with a greater particle size of embolic debris during PCI (16). These findings support the role of distal embolization during PCI in higher-risk patients, and our findings of an association between <0% RS and worsened myocardial perfusion add further data to support the occurrence of this phenomenon.
Further studies are warranted to prospectively validate our findings and to identify an optimal degree of vessel dilation in high-risk patients that minimizes restenosis yet at the same time does not adversely influence outcomes. These findings are especially important in the context of drug-eluting stents, which reduce restenosis independent of MLD. As demonstrated in an IVUS substudy of the Sirolimus-Eluting Balloon Expandable Stent in the Treatment of Patients With De Novo Native Coronary Artery Lesions (SIRIUS) trial, treatment with a drug-eluting stent reduced the biologic variability of the restenotic response, and increased predictability of long-term stent patency with a considerably lower optimal minimum stent area threshold compared to bare-metal stents (17). Thus, provided drug-eluting stents are adequately expanded, these devices may allow us to reduce restenosis without excessively dilating coronary arteries. Nonetheless, it may be difficult in practice to target an optimal degree of RS, as unmodifiable patient and lesion characteristics may influence the RS as much as operator technique.
The data were drawn from pooled analyses of several large randomized trials, are retrospective, and could have been influenced by both identified and unidentified confounders. Although the inclusion and exclusion criteria in these trials were generally uniform, PCI was performed at the discretion of individual operators and was not randomized, which could have resulted in selection bias.
A RS <0% following rescue/adjunctive PCI after fibrinolytic therapy for STEMI was associated with a greater frequency of abnormal myocardial perfusion and with increased mortality. Potential mechanisms for this finding may include greater distal embolization, increased mechanical stretch with release of vasoactive substances, triggering of autoregulatory reflexes, and increased vessel-wall injury.
This work was supported in part by grants from Genentech, Inc. (TIMI-10A and -B); Millennium Pharmaceuticals and Schering-Plough Research Institute (INTEGRITI); Merck and Co. (FASTER); Aventis Pharma (ENTIRE-TIMI-23); and Centocor and Eli Lilly, Inc. (TIMI-14).
- Abbreviations and acronyms
- American College of Cardiology
- acute coronary syndrome
- American Heart Association
- corrected Thrombolysis In Myocardial Infarction frame count
- intravascular ultrasound
- left anterior descending coronary artery
- minimum lumen diameter
- odds ratio
- percutaneous coronary intervention
- normal reference segment
- residual stenosis
- ST-segment elevation myocardial infarction
- Thrombolysis In Myocardial Infarction flow grade
- Thrombolysis In Myocardial Infarction
- Thrombolysis In Myocardial Infarction myocardial perfusion grade
- Received August 4, 2004.
- Revision received October 14, 2004.
- Accepted October 18, 2004.
- American College of Cardiology Foundation
- Kuntz R.E.,
- Safian R.D.,
- Carrozza J.P.,
- et al.
- Kuntz R.E.,
- Gibson C.M.,
- Nobuyoshi M.,
- et al.
- Gibson C.M.,
- Cannon C.P.,
- Murphy S.A.,
- et al.
- Stone G.W.,
- Peterson M.A.,
- Lansky A.J.,
- et al.
- Smith S.C. Jr.,
- Dove J.T.,
- Jacobs A.K.,
- et al.
- Serruys P.W.,
- di Mario C.,
- Piek J.,
- et al.
- Stankovic G.,
- Manginas A.,
- Voudris V.,
- et al.
- Gibson C.M.,
- Dotani M.I.,
- Murphy S.A.,
- et al.
- Hasdai D.,
- Holmes D.R. Jr.,
- Garratt K.N.,
- et al.
- Gregorini L.,
- Marco J.,
- Kozakova M.,
- et al.
- Prati F.,
- Pawlowski T.,
- Gil R.,
- et al.
- Sato H.,
- Iida H.,
- Tanaka A.,
- et al.
- Angelini A.,
- Rubartelli P.,
- Mistrorigo F.,
- et al.
- ↵Sonoda S, Morino Y, Ako J, et al. Impact of final stent dimensions on long-term results following sirolimus-eluting stent implantation: serial intravascular ultrasound analysis from the SIRIUS trial. J Am Coll Cardiol 2004;43:1959–63.