Author + information
- James E. Tcheng, MD, FACC⁎ ( and )
- Kurt G. Kinney, MD
- ↵⁎Reprint requests and correspondence:
Dr. James E. Tcheng, Duke University Medical Center, Department of Medicine, 7021 Duke North Pavilion, 2400 Pratt Street, Durham, North Carolina 27705.
Truth lies within a little and certain compass, but error is immense.(1)
The best known story about the Cynic philosopher Diogenes is of him carrying a lighted lantern in broad daylight through the streets of Athens searching for an honest man. The legend suggests a tortured trek where personal viewpoints were valued more highly than scientifically proven findings. In an analogous fashion, while acute myocardial infarction is one of the more widely studied disease states, there remains a lack of clarity concerning the myriad of management strategies espoused over the years. The clinician thus is beset with an array of decision points, complicated by both antecedent events and subsequent options that constrain potential choices (Fig. 1).Collet et al. (2) addressed this complexity of intertwined options in this issue of the Journal, specifically focusing on cardiac catheterization subsequent to fibrinolysis.
Clinical trials of fibrinolysis as first-line therapy in acute ST-segment elevation myocardial infarction (STEMI) have demonstrated clinical benefit when therapy is administered within 12 h of symptom onset (3). An alternative approach, primary (or direct) percutaneous coronary intervention (PCI) during acute STEMI, has demonstrable superiority to fibrinolysis in virtually all studies of this approach (4). Unfortunately, resource constraints, availability, access, and the logistics of emergency patient transport dictate that many patients will be ineligible for primary PCI, particularly within the 90-min window recommended by current guidelines. Fibrinolytic therapy remains the most widely used reperfusion therapy in the world; even in the U.S., a large proportion of acute STEMI patients are still primarily treated with fibrinolytic therapy (5). Although the decision to refer a patient for primary PCI is a singular (and well-studied) decision point, management of the patient subsequent to fibrinolysis vis-á-vis the intent and timing of cardiac catheterization remains complex, imprecise, and less systematically studied.
As successful implementation of an appropriate management strategy after fibrinolysis is a key to subsequent recovery, a plethora of options inevitably results in a lack of consistency in process and, by extension, suboptimal patient outcomes. The analysis by Collet et al. (2) attempts to discern the “truth” via a unique approach: the assembly of a collection of related meta-analyses, with each meta-analysis remaining true to proper inclusion and exclusion criteria for meta-analytic reporting. In doing so, the authors have avoided the trap of comparing multiple different strategies head to head via one single, overall meta-analysis. However, because the series of complex condition- and time-dependent decisions demands an algorithmic, decision-tree approach, collecting multiple meta-analyses under an umbrella concept (catheterization after fibrinolysis) was necessarily required to remain true to the clinical situation. An apt description is that the authors are providing guidance along the path of the search for “truth”; although the report is nominally structured around 3 specific comparisons, in reality the approach is designed to produce an overall “best practice” algorithm of care.
Delving into the details of the report, 3 specific meta-analyses were constructed, addressing the following explicit questions. First, should rescue PCI be performed on an emergency basis whenever reperfusion failure is suspected? Second, is an “early aggressive” or “early conservative” approach (i.e., cardiac catheterization within 24 h of fibrinolysis vs. a delayed and/or ischemia-guided strategy) better after (presumed successful) clinical reperfusion? And third, should fibrinolytic therapy be administered to facilitate primary PCI (i.e., fibrinolysis on the way to the catheterization laboratory)? As a subplot, analyses were stratified per interventional era, specifically the “balloon era” and the “stent era.”
Taken from the perspective of the stepwise decision options illustrated in Figure 1, the first of the questions is actually the role of fibrinolysis in facilitating emergency PCI. The concept of facilitated PCI is straightforward—to reduce “door-to-patency” time in a proportion of patients, thereby improving outcomes of the entire population. Represented as “Decision B” in Figure 1, the issues can be distilled to the following: 1) can the patient undergo emergency primary PCI; and 2) if the answer is “yes,” should the patient be given fibrinolytic therapy on the way to the catheterization suite? Whereas studies of facilitated PCI have documented infarct-related artery patency rates of 30% to 59% (6–8), an overall detrimental effect was observed in patients receiving fibrinolysis. In the report, a mortality trend along with a significant difference in reinfarction favoring primary PCI alone (p = 0.0076) was identified. Note that the authors (appropriately) elected not to analyze the question of primary PCI versus conventional (not facilitated) fibrinolysis (Fig. 1, Decision A). Seminal trials including DANAMI-2 (Danish Trial in Acute Myocardial Infarction-2) and PRAGUE-2 (PRimary Angioplasty in patients transferred from General community hospitals to specialized PTCA Units with or without Emergency thrombolysis-2) have demonstrated the superiority of primary PCI over conventional fibrinolysis, and other trials have provided remarkably consistent findings (4). Taken together, the answer to the first clinical dilemma appears quite clear: fibrinolytic therapy has no role when primary PCI can be accomplished. Note that adjunctive facilitation with alternative agents, particularly blockade of the platelet adenosine diphosphate receptor and the glycoprotein IIb/IIIa integrin, remains promising and thus deserves further study—these agents thus should notbe grouped (via meta-analysis) with studies of fibrinolytic therapy.
Should primary fibrinolysis be preferred, a critical decision point follows shortly after the administration of fibrinolytic therapy dependent on whether clinical reperfusion has been achieved (Fig. 1, Decisions C and D). The second meta-analysis (rescue PCI for failed fibrinolysis) provides clarity concerning Decision C: whether to refer a patient for rescue PCI. Again, the concept is straightforward; emergency mechanical reperfusion in patients failing to reperfuse will provide additional incremental benefit compared with leaving the artery closed. Interestingly, whereas rescue PCI is a class I recommendation per the current American College of Cardiology/American Heart Association guidelines (9) in patients who develop shock or have severe heart failure, for patients without shock, rescue PCI is considered a IIa (level of evidence C) indication. This latter recommendation stems largely from “balloon era” studies where rescue PCI failed to reduce mortality (10,11). Collet et al. (2) demonstrated a strong trend at 30 days in mortality reduction favoring rescue PCI (p = 0.055) along with a statistically significant reduction in death and reinfarction. At long-term follow-up, an overall 34% relative risk reduction again favored rescue PCI. The newfound advantage of rescue PCI was driven by the more recent “stent era” trials. Although this might superficially appear to be an endorsement of stent over balloon technologies, changes in the pharmacotherapeutic environment, including the liberal use of thienopyridine and glycoprotein IIb/IIIa integrin blockade therapies, undoubtedly contribute. Finally, although bleeding was increased in the rescue arm, this did not appear to contribute to mortality. Thus, the report supports the global use of rescue PCI in patients failing fibrinolysis. The decision to refer a patient to the catheterization suite should be made early and resources mobilized expeditiously whenever this circumstance is identified.
While the elements of the decision tree to this point have been categorical, there are a number of permutations related to cardiac catheterization of the patient who has (presumably) reperfused. Should the patient undergo catheterization? When should catheterization be performed? Should catheterization be delayed until the artery stabilizes? Should catheterization be performed only in higher-risk patients? Interestingly, routine catheterization is again considered a class IIb indication in the guidelines (9). With the report, clarity again is shed on this issue. Early systematic catheterization after fibrinolysis in the “balloon era” was associated with worse outcomes favoring the conservative approach. These findings are largely reversed in the “stent era,” with outcomes favoring early and systematic cardiac catheterization with revascularization. With balloon PCI, high rates of thrombosis, recurrent myocardial infarction, and restenosis were the norm; in the “stent era,” these issues have been largely addressed through the combination of stent implantation and improved pharmacotherapy. Thus, if the cardiology community were still practicing “balloon era” intervention, a delayed, ischemia-driven approach would be appropriate. Fortunately for patients, the “stent era” data supports the use of early and systematic catheterization, with an approximate 42% relative risk reduction in death or myocardial infarction at 30 days. Combining balloon and stent data to reach a single conclusion via meta-analysis would thus have not only been inappropriate but clinically anachronistic. The authors have shown that early systematic catheterization with stenting to be the appropriate, modern therapeutic strategy. The ultimate ramification of this finding is that regardlessof success or failure of fibrinolysis, patients with an acute STEMI should be transferred to a tertiary referral center for cardiac catheterization as soon as possible, with the decision to proceed to catheterization managed by the interventional cardiologist, and not the facility of patient origin.
In conclusion, meta-analysis can, when constructed and applied appropriately, help guide serial clinical decisions, even when smaller and dissimilar studies address only a limited scope of the decision tree and are individually underpowered to be definitive. Combining the consensus about primary PCI with the findings of the meta-analyses presented herein, we believe the decision points illustrated in Figure 1to have the following answers:
1. Fibrinolytic therapy should not be administered before primary PCI.
2. Fibrinolytic therapy should be administered when referral for primary PCI is not a reasonable (or available) option.
3. After fibrinolysis, emergency referral for rescue PCI is indicated for fibrinolysis failure.
4. After fibrinolysis, systematic referral for “urgent” cardiac catheterization (within 24 h) is indicated even after clinically successful fibrinolysis.
5. Stent PCI (accompanied by modern pharmacotherapy) is preferable to balloon PCI.
Patient management is rarely straightforward, typically consisting of a sequence of decisions, with dependencies on both preceding decisions and downstream capabilities. The work of Collet et al. (2) has clearly and succinctly illuminated the path of the search for “truth” in the management of the patient with an acute STEMI.
↵⁎ Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- St. John H.,
- Viscount Bolingbroke
- Collet J.P.,
- Montalescot G.,
- Le May M.,
- Borentain M.,
- Gershlick A.
- Appleby P.,
- Baigent C.,
- Collins R.,
- et al.,
- Fibronolytic Therapy Trialists Collaborative Group
- McNamara R.L.,
- Herrin J.,
- Bradley E.H.,
- et al.,
- NRMI Investigators
- Ross A.M.,
- Coyne K.S.,
- Reiner J.S.,
- et al.,
- PACT Investigators
- Widimsky P.,
- Groch L.,
- Zelízko M.,
- et al.,
- PRAGUE Study Group Investigators
- Van de Werf F.,
- ASSENT-4 Investigators
- Antman E.M.,
- Anbe D.T.,
- Armstrong P.W.,
- et al.
- Vermeer F.,
- Oude Ophuis A.J.,
- vd-Berg E.J.,
- et al.
- Ellis S.G.,
- Riberiero da Silva E.,
- Heyndrickx G.,
- et al.