Statistics from Altmetric.com
The success of fibrinolysis in the treatment of myocardial infarction has been attributed to reperfusion of the occluded vessel, however, it has become clear that it is not just reperfusion but restoration of normal flow—defined angiographically as TIMI-3 flow—in the infarct related artery that dictates mortality. Analysis of the different treatment arms of GUSTO-I,1 the PAMI trials,2 3 and the primary angioplasty registry,4 demonstrates a very clear inverse linear relation between mortality and the rate of TIMI-3 flow achieved in the infarct related artery. The randomised trials of primary angioplasty in acute infarction2 5 6 all point to it being superior to thrombolytic therapy in achieving this goal, although in the GUSTO-IIB substudy7 the benefit was less marked.
The trials of thrombolytic therapy versus placebo all demonstrated a time dependent benefit, the shortest “pain to needle” times having the lowest mortality with a cut off at approximately six hours. This has led to Department of Health guidelines, frantic (albeit appropriate) efforts to keep door to needle times as short as possible, and the consideration of prehospital thrombolysis.8
The observed decline in benefit of thrombolytic agents with time not only relates to continuing loss of potentially salvageable myocardium but also to the lower potency of agents on the more established thrombus. The same time dependent decline is seen in the trials of primary angioplasty within the high risk categories such as cardiogenic shock but is far less marked in the non-shocked patients. Indeed one of the striking features of the PAMI-2 trial was the consistently lower mortality seen in the angioplasty group even when pain to balloon time was six hours or more. This may be explained by the fact that even at this late stage angioplasty achieves TIMI-3 flow in 90% or more patients compared with 28% for thrombolytic agents. Such a preserved response to treatment suggests there may be some leeway in the time to treatment if angioplasty is the chosen treatment modality, maybe enough leeway to allow time to transport patients if angioplasty facilities are not immediately available.
Transporting patients for primary angioplasty is not new. The Mid-American Heart Institute under the direction of Hartzler has, since 1981, treated thousands of acute myocardial infarction patients from considerable distances reporting great success.9 10Lubbock Hospital in Texas currently takes patients from a radius of up to three hours flying time—an equivalent area would encompass almost the whole of the British Isles—with remarkable speed and effectiveness. In this edition the Zwolle group, in a retrospective analysis of its primary angioplasty practice over the past few years, reports equivalent outcomes in patients transferred from up to 60 miles away and those admitted directly.11
Could such transfer arrangements be translated to Great Britain with its hard pressed ambulance service and sparse interventional centres? The Zwolle group reports 104 patients transferred from 14 surrounding hospitals in a five year period, a referral rate of 1.5 patients per year per centre, a remarkably small number. These patients were a selected population being ineligible for thrombolysis and/or high risk patients as defined by clinical criteria. Based on our experience in the Exeter primary angioplasty pilot study (EXPAPS)12 the equivalent clinical definition would produce a minimum of 80 patients aged 79 or younger per year for a catchment population of 330 000. The southwest region (typical of many areas of Great Britain) has a population of 3.5 million and would produce 800 such cases annually for its only interventional centre, some travelling up to 200 miles—not a practical proposition. But considering transfer of such patients ignores the fact that it is the immediate angiogram that defines the high risk patient far better than the clinical presentation as was demonstrated by the PAMI-2 study13 and supported by our own experience. If angiographically high risk patients were included it would increase the above by at least 50%.
Zijlstra et al use their data to recommend and support a large randomised trial comparing locally administered thrombolysis with transfer for primary angioplasty in clinically high risk patients.11 There seems little doubt that such a trial will show a benefit in the angioplasty arm, just as the preliminary results from the equivalent US trial (AIR-PAMI) are already suggesting; however, this does not mean that we should adopt such a strategy as routine. That could result in a suboptimal service with a large number of eligible patients at risk being excluded.
If it is to be accepted that primary angioplasty is the superior reperfusion strategy for acute myocardial infarction then it should be available on a nationwide basis to all patients. If it is deemed prudent to offer it only to high risk patients then we must make sure we can define high risk adequately by the use of expedient angiography. A simple guideline target should suffice—for example, TIMI-3 flow should be established within 90 minutes of presentation in 60% and within 120 minutes in 90% of patients presenting with acute myocardial infarction. Local geographical considerations will then determine what is done to achieve the proscribed and easily audited target. After all if we can apply rules about the availability of lifesaving emergency surgical reperfusion in the event of an iatrogenic coronary occlusion,14 surely we should apply similar rules about emergency reperfusion of the much more common spontaneous variety.