Author + information
- ↵*Reprint requests and correspondence:
Dr. Jean-Lucien Rouleau, Division of Cardiology, Toronto General Hospital, Eaton North 13-212, 200 Elizabeth Street, Toronto, Ontario, Canada M5G 2C4.
In this issue of the Journal, Mehta et al. (1) address the question as to whether patients suffering an acute myocardial infarction (MI) would be better served presenting to a hospital with coronary revascularization capabilities. To answer this question adequately in the year 2002, two separate issues need be addressed. The first relates to whether patients with an acute ST-segment elevation MI are better served by primary angioplasty (percutaneous coronary intervention [PCI]) rather than thrombolysis, and if primary PCI is superior, whether this needs to be done at the hospital of presentation or can patients be transferred to regional centers where large volumes assure good results (2). The second question, addressed by the investigators, is whether patients receiving thrombolysis are better served by presenting to a hospital with coronary revascularization capabilities.
Primary PCI versus thrombolysis: need for presentation to a hospital with coronary revascularization capabilities
The Primary Angioplasty in Myocardial Infarction (PAMI) study in 1993 (3) suggested that, in experienced hands, primary PCI was superior to thrombolysis in patients presenting with an acute ST-segment elevation MI. A meta-analysis by Yusuf and Pogue (4) re-enforced this concept, and the SHould we emergently revascularize Occluded Coronaries in cardiogenic shocK (SHOCK) trial (5) clearly established the benefits of early mechanical reperfusion with PCI or coronary artery bypass grafting (CABG) in patients with acute MI complicated by cardiogenic shock. Whether patients suffering an acute ST-segment elevation MI need to have primary PCI in their hospital of presentation or whether they can safely be transferred to a regional center to have PCI has recently been studied. The most comprehensive of these is the Danish Multicenter Randomized Study on Thrombolytic Therapy versus Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI-2) that was reported at the American College of Cardiology Scientific Sessions in March 2002. In that study, 1,572 patients suffering from an acute ST-segment elevation MI were randomized to front-loaded tissue-type plasminogen activator (t-PA) or PCI if they presented within 12 h of onset of their MI. A significant proportion of patients (n = 1,129) presented to a center without revascularization capabilities and required being transferred. All patients assigned to primary PCI received treatment within 3 h of randomization. The study was discontinued prematurely owing to clear benefit in the combined end point of death, reinfarction, and disabling stroke (40% reduction in the combined end point at 30 days, p = 0.003). Interestingly, there was little difference in delay for PCI whether a patient presented to an interventional facility or required transfer from a noninterventional hospital, and no deaths occurred during transfer. Importantly, no difference in benefit with the PCI strategy was noted between patients presenting to a hospital with or without coronary revascularization capabilities.
Two other studies support the results of the DANAMI-2 trial. The first is the PRimary Angioplasty in patients transferred from General community hospitals to specialized PTCA Units with or without Emergency thrombolysis (PRAGUE) study (6), in which transfer for primary PCI, thrombolysis, and immediate transfer for PCI and thrombolysis with ischemia-driven transfer were compared in 300 patients with ST-segment elevation MI. A significant reduction in 30-day death, MI, and stroke was reported for primary PCI compared to the other two strategies (8% vs. 15% vs. 23%, respectively, p < 0.02). Admittedly, streptokinase was the thrombolytic agent utilized in this trial.
Finally, the Air Primary Angioplasty in Myocardial Infarction (AIR PAMI) trial, cited by the investigators attempted to deal directly with the issue of routine transfer of acute MI patients to an interventional facility in the U.S. in a randomized trial of transfer for primary PCI versus thrombolysis. As Mehta et al. (1) point out, the trial was terminated early because of poor patient enrollment, and no difference in outcome was noted at that time. However, a 38% reduction in major adverse cardiac events was observed in the 138 patients randomized to PCI, despite a delay in treatment of 155 min. Taken together, these studies are highly suggestive that, if appropriate facilities for rapid patient transfer are available, the benefits of primary PCI over thrombolysis are maintained and equivalent regardless of whether a patient presents to a hospital with coronary revascularization capabilities or a center without such facilities.
Patients receiving thrombolysis: need to present to a hospital with coronary revascularization capabilities
Mehta et al. (1) conclude that patients receiving thrombolysis for a ST-segment elevation MI obtain no benefit from presenting to a hospital with coronary revascularization capabilities. They correctly cite data from registries performed in the U.S. between 1988 to 1996, all supporting their conclusions (7–10). Data from comparisons between the U.S. and Canada performed during that period yielded more complex results (11–13). Although still somewhat controversial, these studies suggest that there may be a short-term benefit associated with greater use of invasive revascularization procedures in the U.S., but that this benefit, at least as it relates to survival, disappears by one year.
How then do we reconcile the differences in outcome related to the greater use of invasive diagnostic and therapeutic procedures in Canada and the U.S. with the lack of difference reported by Mehta et al. (1) and various other studies done within the U.S.? The answer is certainly complex, possibly being related in part to differences in the use of medications. However, it may also be the result of a threshold below which there is a clear benefit from the use of invasive diagnostic and therapeutic interventions. In favor of this possibility is the very low use of coronary angiography in Canada versus the U.S. (6.7% vs. 34.9%, respectively) and the very low use of revascularization procedures (2.9% vs. 22.3%, respectively) in the only study that demonstrated a survival benefit post-MI in the U.S. (11). In the two other studies (12,13), the use of invasive diagnostic (25% and 35%) and revascularization procedures (13% and 14%) in Canada was greater than the one study showing a short-term difference in survival (a study done exclusively in patients over 65 years of age) (11). It should be noted, however, that even in the one study that demonstrated a difference in survival at 30 days (77.7% in Canada vs. 78.6% in the U.S., p = 0.03) this difference disappeared by one year (65.6% in Canada vs. 65.7% in the U.S.).
Previous reports evaluating the effect of on-site availability or use of coronary revascularization have all been registries and thus better reflect effects in the entire population. However, data from randomized trials add complementary information that, although less generalizable, controls for more variables and provides more precise information in the population studied. In this way, this analysis of the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries-1 (GUSTO-1) study is important. The investigators found a borderline improvement of 30-day survival in patients admitted to centers with coronary revascularization capabilities. When adjusted for the baseline risk factors found to be the most predictive of survival in the overall GUSTO-1 study—that is, age, systolic blood pressure, heart rate, Killip class, and MI location—this improvement became insignificant. The MITI (Myocardial Infarction Triage and Intervention) registry (9) also demonstrated a tendency for improved in-hospital survival in centers with coronary revascularization capabilities (relative risk [RR] 0.88, [95% confidence interval (CI) 0.71 to 1.09]), as did a large registry reported by Rogers et al. (7) at six months’ post-MI (RR 0.92, [CI 0.83 to 1.01]). This raises the question as to whether a real difference exists and whether their adjustments for severity of disease included all potentially important variables.
Mehta et al. (1) justify their model by telling us that these five variables make up approximately 90% of the total chi-square of their model. It is, however, very difficult to adjust for all differences between groups. In a previous study of consecutive patients admitted with an acute MI in Canada (14), despite adjusting for important variables, which included left ventricular ejection fraction, we were unable to eliminate an inverse relationship between patient acuity and survival. In the study by Mehta et al. (1), several potentially important differences in patient acuity were noted, most of which were not included in their adjustment model and would favor a widening of the RR reduction in centers with revascularization capabilities. These include more prior MI, hypertension, prior revascularization, and greater use of ventilators and pacemakers in patients from the centers with revascularization capabilities.
These concerns notwithstanding, the results of survival at one year in the present study clearly indicate no difference between groups such that if borderline differences existed in survival at 30 days, they do not persist. A loss of potential early differences in survival by one year was also found in the MITI registry (hazard ratio of 1.0 [95% CI 0.93 to 1.1]), and in the one Canada–U.S. comparison that demonstrated an association between greater use of invasive diagnostic and therapeutic interventions and survival. Thus, as outlined by the researchers (1), whether an early benefit in centers with coronary revascularization capabilities exists or not, the differences do not persist.
It bears mentioning that the GUSTO-1 study was performed a decade ago and that medical as well as PCI techniques have evolved. Mehta et al. (1) point out that few new interventions that would modify their results have appeared since then. This may be true for individual developments (except angiotensin-converting enzyme [ACE] inhibitors), but it is possible that, when combined, these could influence the results of such an analysis if it were done in 2002. These include ACE inhibitors, clopidogrel, the use of coronary stents, platelet glycoprotein IIb/IIIa receptor inhibitors, lipid-modifying agents, and low molecular weight heparin.
One final cautionary note with respect to the interpretation of the study by Mehta et al. (1). In the GUSTO-1 study, the number of patients transferred for invasive diagnostic and revascularization procedures was reasonably high in the centers without revascularization capabilities, 59.2%. Thus, the threshold for invasive diagnostic and therapeutic interventions was low, such that the conclusions of the investigators can only be said to hold true if there is liberal use of regional coronary diagnostic and revascularization facilities. Their results cannot be extrapolated to situations where there is less use of diagnostic and therapeutic procedures. Also, while the reported rates of cardiac catheterization were quite high in centers with and without revascularization capabilities (78.1% and 59.2%, respectively), the rates of revascularization were considerably lower in both types of institutions. It is unclear as to whether this represents rates based on knowledge of anatomy and ischemia or a less interventional bias of treating physicians in the early 1990s. It is unlikely that a similar pattern would be observed in contemporary practice.
At this time, the weight of the evidence supports the conclusions of Mehta et al. (1), namely that when suffering from an acute MI, when receiving thrombolysis, there is no survival benefit in presenting to a hospital with coronary revascularization capabilities. Recent studies documenting the benefits of primary PCI in patients presenting with an MI to a hospital without coronary revascularization capabilities would suggest that their conclusions may hold true for primary PCI as well, if proper emergent patient transfer is available. The study (1) in this issue of the Journal, however, does not address the possibility that there could be differences in the outcome of patients presenting to centers without revascularization facilities where timely (2 to 3 h) access to primary PCI is not possible. Nonetheless, because of resource allocation issues, in the majority of healthcare settings in the world today, thrombolysis will continue to be the only feasible primary reperfusion modality rendering the results of the Mehta et al. (1) study important.
↵* Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
- American College of Cardiology Foundation
- Mehta R.H.,
- Criger D.A.,
- Granger C.B.,
- et al.
- Widimsky P.,
- Groch L.,
- Zelizko M.,
- Aschermann M.,
- Bednar F.,
- Suryapranata H.
- Rogers W.J.,
- Canto J.G.,
- Barron H.V.,
- et al.
- Myocardial Infarction Triage and Intervention project investigators,
- Every N.R.,
- Larson E.B.,
- Litwin P.E.,
- et al.
- Every N.R.,
- Parsons L.S.,
- Fihn S.D.,
- et al.
- Rouleau J.L.,
- Talajic M.,
- Sussex B.,
- et al.