The Danish Multicenter Randomized Trial on Thrombolytic Therapy vs. Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI-2) compared the outcomes of treating ST-elevation myocardial infarction (MI) with thrombolytic therapy vs. primary angioplasty (PCI), with transfer via ambulance within 3 hours, to an invasive center when necessary. The study was funded primarily by the Danish Heart Foundation and the Danish Medical Research Council and was designed to compare the 2 revascularization strategies on a national level in Denmark. The study included 24 referral hospitals (that did not perform PCI) and 5 invasive centers. In all, these 29 centers serve approximately 62% of the entire population of Denmark. Dr. Andersen pointed out that prior to DANAMI-2, primary angioplasty was performed infrequently in Denmark, and that primary angioplasty programs and protocols were established and centers accredited as part of the preparation for this study. The initial power calculation suggested that approximately 1900 patients, approximately half at referral hospitals and half at invasive centers, would need to be randomized to detect a difference between the treatment strategies.
Patients with acute MI were eligible for randomization if they demonstrated cumulative ST elevation of =4 mm at presentation, were within 12 hours of symptom onset, and, if they did not present to an invasive center, could be transferred to one within 3 hours of randomization. There were no age restrictions. Patients were excluded from randomization for contraindications to thrombolytic therapy, shock, or persistent life-threatening arrhythmia. All patients received intravenous aspirin and unfractionated heparin, and beta-blockade. No "upstream" glycoprotein IIb-IIIa inhibitors were given. The primary end point of the study was the composite of all-cause mortality, reinfarction, or disabling stroke at 6 months of follow-up. Each component of the primary end point was also evaluated separately.
Seven hundred eighty-two patients were randomized to thrombolytic therapy (100 mg of front-loaded tPA) and 790 to primary PCI. There were no significant differences between the groups with respect to baseline demographics. Of note, approximately 74% of patients were male and just more than half sustained anterior MI in each group. The median transport distance for patients presenting to local referral hospitals and randomized to PCI was 35 miles (with the maximum transfer distance of 95 miles). Overall, few complications occurred during transfer. These included atrial fibrillation (2.5%), ventricular tachycardia (0.2%), ventricular fibrillation (1.4%), and second- or third-degree heart block (2.3%). There were no deaths and no patient required endotracheal intubation during transfer. It is interesting to note that the mean "door-to-balloon" time was only 10 minutes longer for the patients requiring transfer than for patients initially presenting to an invasive center. Results were analyzed on an intention-to-treat basis. Of the patients randomized to fibrinolysis, 99% received the intended therapy. Of those randomized to PCI, 99% underwent coronary angiography and 87% went on to receive PCI. A total of 93% of the patients who received primary angioplasty also received stent implantation. There was a low crossover rate in this study, with only 2.5% of patients randomized to thrombolytic therapy undergoing "salvage" angioplasty within the first 24 hours.
The study was stopped prematurely after 1562 patients had been randomized (1129 at referral hospitals and 443 at invasive centers) due to superior outcomes in the patients receiving primary PCI. The PCI treated patients demonstrated a 40% relative risk reduction in the primary composite end point (8.0 vs 13.7%; P = 0.0003). While all individual components showed trends in favor of PCI, including death (6.6% vs 7.6%; P = 0.35) and stroke (1.2% vs 2.0%; P = 0.15), the difference in the composite end point was driven largely by the significant reduction in repeat MI (1.6% vs 6.34%; P < 0.0001) in the patients receiving PCI. Dr. Andersen stated that with this degree of risk reduction only 18 patients would need to be treated with primary angioplasty to prevent one adverse event. Furthermore, patients receiving primary PCI had better outcomes than those receiving thrombolytic therapy, whether they were transferred from a local referral hospital for the procedure (8.5 vs 14.2%; P = 0.002) or presented initially to the invasive center (6.7 % vs 12.3%; P = 0.048). The P value for the latter is less robust in large part because of the smaller number of patients randomized from invasive centers (n = 443) at the time the study was halted. In addition, Dr. Andersen also noted that the need for revascularization within the first 30 days of follow-up was higher for patients initially treated with thrombolytic therapy (16.6% vs 5.9%).
Comment by Sarah M. Vernon, MD
It is now clearly established that primary angioplasty is the superior revascularization strategy for many, if not all, patients presenting with acute ST-elevation MI. This is particularly true for the patients at highest risk for poor outcomes (large, anterior infarcts, cardiogenic shock) or complications of thrombolytic therapy. The ACC/AHA Guidelines for the Management of Acute Myocardial Infarction list primary PCI as a Class I recommendation for revascularization.1 This recommendation goes on to say that primary PCI is preferable in any patient when "performed in a timely fashion by individuals skilled in the procedure and supported by experienced personnel in high-volume centers," suggesting that primary angioplasty should be performed by interventionalists who do more than 75 angioplasties a year, and in centers that perform more than 200 PCI procedures annually. The difficulty lies in the fact that, even in the United States in the year 2002, only a small percentage of hospitals have the capability to perform coronary angioplasty, much less meet case volume criteria to be described as "invasive centers." Therefore, the reality in most hospitals is that any potential advantage for primary angioplasty (or risk of thrombolytic therapy) has to be carefully balanced against the potential risk of time delay in transferring the patient to another institution to receive mechanical perfusion.
So, the quandary persists: primary PCI may be the superior revascularization strategy, but in many cases, particularly when transfer to another institution is involved, it takes more time to implement—and as we all know: "time is muscle." But how much time (or muscle) is spent by putting a patient in the throes of an acute MI "back on the road," as Dr. Andersen described it, to be revascularized at another institution? More importantly, does this additional time affect clinical outcomes? DANAMI-2 may provide the answers to these questions. In this study, not only was it safe to transport patients with acute MI for PCI at a regional center of excellence, but also this strategy resulted in superior patient outcomes with respect to reinfarction and probably need for revascularization in the subsequent 6 months. In fact, patients transferred for mechanical reperfusion experienced equivalent benefit to those initially presenting to an invasive center. The reason for this lies, in part, in the time required to achieve an open artery, the "door-to-balloon time," which was only 10 minutes longer, on average, for patients requiring transfer. With a well-established protocol, good lines of communication, and ground transportation (albeit in a relatively small country with short distances between participating hospitals) it was possible to mobilize the catheterization laboratory in anticipation of the patient’s arrival, resulting in minimal delays to revascularization. This situation could be extrapolated to many community hospitals in the United States, who, with expedient mechanisms for referral and transport directly to the nearest available cardiac catheterization laboratory (by air if warranted), would likely be able to achieve favorable "door-to-balloon times" as well. This is because, even in the United States, which is known for a much more aggressive approach to revascularization for acute MI, the vast majority of invasive centers do not maintain laboratory personnel in-house after hours, and require on the order of 30-60 minutes to mobilize in the best of circumstances. A call ahead from the referring hospital will allow a fully staffed and prepared catheterization laboratory to be ready and waiting when the patient arrives for primary angioplasty.
Any interventional cardiologist knows that there is no teacher like experience, and recent evidence suggests that operators and institutions with extensive experience in performing acute MI angioplasty (which, as we all know, is a markedly different beast than stenting a type A lesion in a stable patient) are likely to achieve better procedural and clinical outcomes. While the interventional operators in DANAMI-2 were certainly well qualified, these were not centers with extensive (or, in some cases, any) experience with primary angioplasty prior to the initiation of this protocol. Despite this limitation, outcomes were superior for patients receiving PCI. One could ascertain that the differential benefit might be even greater at centers highly experienced in acute MI angioplasty, and would increase at other less-experienced institutions, such as these, with time, if volume was adequate. In addition, since transport was found to be quite safe and did not translate into substantive delays to treatment, DANAMI-2 would suggest that there should be less incentive to establish low-volume primary PCI programs or PCI programs lacking surgical backup in areas where patients could be transferred to a more experienced center in a timely fashion.
The DANAMI-2 investigators set out to design and implement a nation-wide strategy to optimize outcomes for patients with acute MI undergoing revascularization. To function effectively, as demonstrated in this study, the mechanism of transferring patients for acute MI intervention must be a "well-oiled machine." This is a far cry from the sporadic, late night, ER to ER transfer with incomplete clinical information or unclear diagnosis requiring time-consuming reevaluation at the tertiary center. This strategy is also infinitely preferable, in the eyes of an interventional cardiologist, to immediate administration of a thrombolytic agent at the local hospital and transfer for salvage angioplasty hours later when thrombolysis fails. The results from DANAMI-2 suggest that transfer of patients for primary angioplasty is the preferable strategy, and should prompt us to review and optimize our lines of referral and communication so that more patients in this country have access to the superior strategy of revascularization for acute myocardial infarction.
Dr. Vernon is Assistant Professor of Medicine, Director , VAMC Cardiac Catheterization Laboratory, University of New Mexico, Health Sciences Center, Albuquerque, NM.
1. Ryan TJ, et al. Circulation. 1996;94:2341-2350.