Author + information
- David Antoniucci, MD⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. David Antoniucci, Division of Cardiology, Careggi Hospital, Viale Morgagni, I-5034, Florence, Italy
The OAT (Occluded Artery Trial) study randomized 2,166 patients with subacute myocardial infarction and a persistently occluded infarct-related artery (IRA) to percutaneous coronary intervention (PCI) or to medical therapy (1). The study did not show any benefit of PCI as compared with medical therapy and did not confirm the so-called open artery hypothesis. The negative and unexpected results of the OAT study raised several concerns. Among these, the low risk of the enrolled population that prevented the demonstration of the hypothesized benefit of PCI. Major exclusion criteria of the OAT study included left main or 3-vessel disease, overt signs of heart failure, angina at rest, and severe inducible myocardial ischemia. Thus, the resulting profile of an OAT patient was defined by the lack of symptoms and a relatively low-risk of 1- or 2-vessel disease not associated with severe inducible myocardial ischemia. The exclusion criteria at least in part explain the recruitment problems encountered by the investigators that resulted in a low enrollment rate and in the reduction of the original planned number of patients from 3,200 to 2,166.
The study presented in this issue of JACC: Cardiovascular Interventions by the OAT investigators assessed the impact of randomization to PCI on outcome according to the patient's risk (2). The patient cohort was divided into risk tertiles (low-, medium-, and high-risk) according to the clinical outcome, and then the impact of mechanical infarct artery recanalization was assessed in the 3 subgroups. The primary end point rate (composite of death, reinfarction, New York Heart Association functional class IV heart failure), and mortality rate were, respectively, 8.6% and 5.5% for patients within the low-risk tertile, 13.4% and 6.9% for patients within the medium-risk tertile, and 30.6% and 22.6% for patients within the high-risk tertile. The study shows that there is no benefit of PCI as compared with medical therapy irrespective of the patient's risk with regard to the primary end point of the study as well as to the estimated mortality at 5-year follow-up. Moreover, in the highest risk patient subgroup there was a trend toward more adverse events in patients randomized to PCI as compared with medical therapy alone (33.9% vs. 27.3%, p = 0.10).
This study not only confirms the unexpected and negative results of the main study (no benefit of PCI) but also extends the results to all patients regardless of their risk and suggests that PCI may be responsible for a worse prognosis for patients within the high-risk tertile who could theoretically derive major benefit from revascularization as compared with low- or medium-risk patients.
At first sight, the negative and in some way provocative results of this study could be scaled back by the fact that the overall OAT patient cohort is not high risk, and the stratification into 3 tertiles cannot overcome the criticism that the benefit of revascularization could not be demonstrated even in the highest risk tertile because it does not include “true” high-risk patients. This explanation however is not tenable when considering that the risk stratification method used by the investigators is rigorous and effective and allowed the identification of a subgroup of patients with a remarkable 5-year mortality of 22.6%.
However, are the results of the OAT study really unexpected? Two aspects of the main study as well as of the current report were not sufficiently addressed.
First, the definition of a successful PCI was in several ways arbitrary and may be easily challenged. The OAT investigators include in the definition of successful PCI an open IRA with a Thrombolysis In Myocardial Infarction (TIMI) flow grade 1 or 2. All primary PCI studies, including the studies performed in the primary stenting supported PCI era, have shown that a TIMI flow grade <3 is a strong predictor of mortality and major adverse events (3,4). A post-procedure TIMI flow grade <3 in patients without cardiogenic shock should be considered as the marker of an ineffective reperfusion due to an already disrupted microvessel network before PCI, or more likely due to atherothrombotic embolization. The restoration of TIMI flow grade 3 by PCI, which is a prerequisite for an effective myocardial reperfusion but does not guarantee it, was achieved in only 80% of patients. Thus, using current standard criteria for the definition of PCI success, primary procedure failure occurred in at least 20% of patients. Moreover, it is likely that in most of these patients, PCI was not only ineffective but also deleterious because of embolization into the microvessel network with subsequent no-reflow, decreased retrograde flow from collateral circulation, and definite loss of viable myocardium. This hypothesis seems to be confirmed by: 1) the high percentage of large areas of viable myocardium (in the OAT viability substudy, the large majority of the patients—69%—had evidence of viable myocardium in the area supplied by the occluded IRA); and 2) the increased incidence of periprocedural elevation of cardiac enzymes (10% vs. 3.3%, p < 0.001) and of reinfarction (1.6% vs. 0.6%) in patients randomized to PCI as compared with patients randomized to medical therapy (1).
The BRAVE (Beyond 12 Hours Reperfusion Alternative Evaluation Trial)-2 study has demonstrated that residual viability is very frequent in asymptomatic patients with subacute myocardial infarction, and more importantly, that PCI may be effective in salvaging viable myocardium (5). In this study, 365 asymptomatic patients with acute myocardial infarction presenting between 12 and 48 h from the symptom onset were randomized to an invasive (PCI) or a conservative treatment strategy. The primary end point of the study was infarct size as assessed by sestamibi scintigraphy performed between 5 and 10 days after randomization. The median time from symptom onset to hospital admission was 22 h. In the invasive group, 98.6% of patients had mechanical revascularization. The final infarct size was smaller in the invasive group as compared with the conservative group (median 8% vs. 13%, p < 0.001). The scintigraphic substudy that included 261 patients who had paired studies (before randomization and 5 to 10 days after randomization) showed that the final smaller infarct size in the invasive strategy group as compared with the conservative strategy group was due to increased myocardial salvage in the former (salvage index: 0.44 and 0.23, respectively; p < 0.001) because the baseline areas at risk were similar in the 2 groups (6).
There is growing evidence that effective prevention of embolization during PCI in the setting of acute myocardial infarction results in a better myocardial reperfusion, increased myocardial salvage, and improved survival. This evidence is supported by several trials that used thrombectomy devices (7,8). The largest of these studies, the TAPAS (Thrombus Aspiration During Percutaneous Coronary Intervention in Acute Myocardial Infarction) trial, demonstrated a significant increase in survival in patients randomized to manual thrombus aspiration as compared with standard PCI (9), and it is likely that the result of ongoing trials will support the routine use of thrombectomy before IRA stenting.
In the setting of subacute myocardial infarction with a persistently thrombotic occlusion of IRA, the embolization of an occlusive partially organized thrombus during PCI may have a worse impact on microvessel circulation than fresh acute thrombus embolization because the former is more refractory to spontaneous or pharmacological lysis. The negative impact of PCI failure due to embolization on clinical outcome is well established from several PCI studies and, in the setting of acute myocardial infarction, is a strong predictor of mortality. It is reasonable to conceive of a major impact of PCI failure due to embolization in high-risk patients, as suggested by the OAT results.
A second aspect that deserves a specific comment is that in the subset of 379 patients with 2-vessel disease, a 2-vessel PCI was performed in only 75 patients, leaving most of them with an incomplete revascularization. Multivessel disease and incomplete revascularization have a negative impact on clinical outcome after PCI (10–12), and these 2 variables have the potential for some effect in the OAT results.
These 2 variables, PCI success defined according to standard criteria and completeness of revascularization in patients with 2-vessel disease, were not considered in the multivariable models, leaving undefined their role in the OAT patient cohort.
Uncertainties regarding the potential benefit of routine late PCI for subacute myocardial infarction on left ventricular remodeling and clinical outcome have long been a matter of debate. It is easy to predict that the debate will not end after the results of the OAT study. The opposite results provided by a recent meta-analysis of 10 randomized studies (the OAT study was included) comparing PCI with medical therapy in subacute myocardial infarction showed a strong benefit of PCI in terms of improved survival (13).
Based on the evidence currently available, and thanks also to the results of the OAT study, 2 statements can be made: 1) no significant benefit can derive from mechanical IRA reopening in the absence of viable myocardium; and 2) viable myocardium can be salvaged if PCI restores an adequate flow at the myocardial level. As a consequence, PCI, if performed, should not be driven by the interventionalist's oculo-occluded reflex, but supported by the assessment of a correct risk/benefit ratio. The most reasonable approach to an asymptomatic patient with subacute myocardial infarction with a persistently occluded IRA is to perform PCI only if myocardial viability or inducible ischemia is shown. The goal of the procedure should be effective reperfusion at the myocardial level. From a technical point of view, the procedure may be challenging, due to the high risk of embolization and disruption of the microvessel network that may jeopardize the collateral flow. The use of techniques that decrease the risk of embolization should be considered in all cases. Otherwise, the procedure could become detrimental to the patient and prove that an occluded artery is better than an open one.
↵⁎ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
- American College of Cardiology Foundation
- Kruk M.,
- Kadziela J.,
- Reynolds H.R.,
- et al.
- Heggunje P.S.,
- Harjai K.J.,
- Stone G.W.,
- et al.
- Van den Brand M.J.,
- Rensing B.J.,
- Morel M.A.,
- et al.
- Hannan E.L.,
- Racz M.J.,
- Djavad T.,
- et al.
- Sorajja P.,
- Gersh B.J.,
- Cox D.A.,
- et al.
- Abbate A.,
- Biondi-Zoccai G.G.L.,
- Appleton D.L.,
- et al.