Correspondence
Treatment of Acute Myocardial Infarction
N Engl J Med 1993; 329:430-433August 5, 1993
- Article
To the Editor:
“Referral filter” bias can affect the generalizability of the results of any clinical trial1. The three clinical trials comparing angioplasty with thrombolytic therapy in the March 11 issue of the Journal did not mention the referral filter that led to the enrolled samples2-4. Physicians have no way of knowing whether the study populations are similar to their own patients without a clinical description of the patients who were excluded. Could the authors comment on the number of patients excluded and the reasons for exclusion?
4 ReferencesCharles Beauchamp, M.D., Ph.D.
Eric C. Westman, M.D.
Joseph A. Govert, M.D.
Durham Veterans Affairs Medical Center, Durham, NC 277051
Sackett DL, Haynes RB, Guyatt GH, Tugwell P. Clinical epidemiology: a basic science for clinical medicine. 2nd ed. Boston: Little, Brown, 1991.
2
Grines CL ,Browne KF ,Marco J , et al. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1993;328:673-679
Full Text | Web of Science | Medline3
Zijlstra F ,de Boer MJ ,Hoorntje JCA ,Reiffers S ,Reiber JHC ,Suryapranata H . A comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial infarction. N Engl J Med 1993;328:680-684
Full Text | Web of Science | Medline4
Gibbons RJ ,Holmes DR ,Reeder GS ,Bailey KR ,Hopfenspirger MR ,Gersh BJ . Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993;328:685-691
Full Text | Web of Science | Medline
To the Editor:
It is difficult to draw any firm conclusions about the role of coronary angioplasty in acute myocardial infarction from the report by Grines et al. The treatment of the patients in the thrombolytic-therapy group, however, is of particular concern. As a group, these patients were not particularly ill: their base-line blood pressures, heart rates, and left ventricular ejection fractions were normal. Only one third of the patients had an anterior-wall myocardial infarction, and the average Killip class was 1.2. Previous experience suggests that at least 60 percent of such patients require no more than a warm handshake and a smile. Yet, 63 percent of the patients who received thrombolytic therapy had an “unscheduled” cardiac catheterization before discharge, resulting in revascularization in 48 percent (percutaneous transluminal coronary angioplasty in 36 percent and coronary-artery bypass surgery in 12 percent). Thus, for an expected survival advantage of 2 to 3 percent,1,2 2 percent of the patients had an intracranial hemorrhage and died; 8 percent received a blood transfusion, with all the attendant concern regarding the administration of blood; and thrombolytic therapy somehow led to invasive interventions, for which the benefits are yet to be determined in this setting3. The authors' choice of tissue plasminogen activator (t-PA) as the thrombolytic agent “because it is the most commonly used agent in the United States” is also disturbing, since survival advantages for this drug have yet to be demonstrated4,5. In addition, the risk of intracranial hemorrhage appears to be higher with t-PA than with a less expensive thrombolytic drug5 -- an outcome that appears to be particularly germane to the findings in this report. The decision to use t-PA, coupled with the invasive approach taken, also seems to disregard the implications for health care costs.
5 ReferencesHoward S. Friedman, M.D.
SUNY Health Science Center at Brooklyn, Brooklyn, NY 112011
Gruppo Italiano per lo Studio della Streptochinasi nell'Infarto Miocardico (GISSI)
Web of Science | Medline2
ISIS-2 (Second International Study of Infarct Survival) Collaborative Group
Web of Science | Medline3
Alderman EL ,Bourassa MG ,Cohen LS , et al. Ten-year follow-up of survival and myocardial infarction in the randomized Coronary Artery Surgery Study. Circulation 1990;82:1629-1646
CrossRef | Web of Science | Medline4
Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto MiocardicoGISSI-2: a factorial randomised trial of alteplase versus streptokinase and heparin versus no heparin among 12 490 patients with acute myocardial infarction. Lancet 1990;336:65-71
Web of Science | Medline5
ISIS-3 (Third International Study of Infarct Survival) Collaborative Group
CrossRef | Web of Science | Medline
To the Editor:
In treating myocardial infarction, Gibbons et al.1 found little difference in outcome between immediate angioplasty and thrombolysis. Since thrombolysis is 30 percent more expensive and since alternatives to double-chain tissue plasminogen activator (duteplase) are available, it would be interesting to know how much of the cost of the thrombolysis was the cost of the drug.
1 ReferencesT.H. Hughes-Davies, F.R.C.P.
Breamore Marsh, Fordingbridge, Hampshire 2P6 2EJ, United Kingdom1
Gibbons RJ ,Holmes DR ,Reeder GS ,Bailey KR ,Hopfenspirger MR ,Gersh BJ . Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993;328:685-691
Full Text | Web of Science | Medline
To the Editor:
Dr. Gibbons and colleagues have developed an elegant means of assessing myocardial salvage resulting from reperfusion therapy for acute myocardial infarction. Their method may be biased against patients treated with thrombolytic therapy, however.
After successful thrombolysis, approximately 80 percent of patients will have clinically important residual stenosis of the infarcted vessel, and stunned myocardium is common1. In the presence of severe residual stenosis, a state of hibernating myocardium can exist and can persist for more than six weeks. Recent studies suggest that imaging with technetium-99m sestamibi of patients at rest significantly underestimates myocardial viability in the setting of hibernating myocardium, missing it entirely in up to 30 percent of patients2-4.
Gibbons et al. used technetium-99m sestamibi imaging. In their report, 31 of 51 patients treated with thrombolysis did not have angioplasty or bypass during their initial hospitalization. In view of the conservative strategy used, this finding presumably implies that such patients had negative results on thallium exercise tests before discharge. Nonetheless, a significant number of these patients, as compared with patients initially treated with angioplasty, had recurrent infarction or revascularization during six months of follow-up (14 vs. 4). This strongly suggests the presence of hibernating myocardium at hospital discharge. Therefore, the degree of myocardial salvage in patients treated with thrombolysis was significantly underestimated.
Until the frequency and quantity of hibernating myocardium after thrombolytic therapy are accurately assessed, the use of sequential sestamibi imaging for the evaluation of thrombolytic therapy will remain severely flawed.
4 ReferencesWilliam W. Rowe, M.D.
Skagit Valley Hospital, Mount Vernon, WA 982731
Rowe WW ,Simpson RJ Jr ,Tate DA , et al. Nonemergent cardiac catheterization and risk-stratified revascularization following thrombolytic therapy for acute myocardial infarction: a critical analysis of therapy in the community setting. Arch Intern Med 1989;149:1611-1617
CrossRef | Web of Science | Medline2
Maublant JC ,Citron B ,Lipiecki J , et al. Predictive value of Tc-99m-sestamibi tomographic imaging at rest for myocardial viability in hibernating myocardium. J Am Coll Cardiol 1993;21:Suppl A:282A-282A abstract.3
Taylor AM II ,Merhige ME . Detection of myocardial viability: sestamibi overestimates necrosis compared with thallium. J Am Coll Cardiol 1993;21:Suppl A:283A-283A abstract.4
vom Dahl J ,Altehoefer C ,Biedermann M ,Uebis R ,Buell U ,Hanrath P . Technetium-99m-methoxy-isobutyl-isonitriles as a tracer of myocardial viability? A quantitative comparison with F-18 fluorodeoxyglucose in 100 patients with coronary artery disease. J Am Coll Cardiol 1993;21:Suppl A:283A-283A abstract.
Author/Editor Response
The authors reply:
To the Editor: In response to Friedman, as shown in Table 4 of our article, the majority of catheterizations were performed for spontaneous or provokable ischemia. Catheterization was performed for chest pain without electrocardiographic changes in 5.5 percent of patients given t-PA and because of the physician's preference in an additional 14.5 percent of patients. Therefore, 43 percent of the patients in the t-PA group had good clinical indications for catheterization. Forty-five percent of the patients given t-PA underwent percutaneous transluminal coronary angioplasty, coronary-artery bypass grafting, or both. The rates of invasive procedures after thrombolysis in our study were actually lower than those reported in myocardial infarction registries1 and similar to those reported in other prospective thrombolytic trials2,3.
Our patients were not at lower risk than those in the Thrombolysis in Myocardial Infarction (TIMI) trial4. In fact, 27 percent of the patients in our trial were ineligible for the TIMI trial because of advanced age or because they presented more than four hours after the onset of symptoms. Unlike the TIMI investigators, we did not code an event as an anterior myocardial infarction if ST elevation was present in any chest leads on electrocardiography; for example, ST elevation in inferior leads and leads V5 to V6 was coded as an inferior myocardial infarction.
We believe that the use of t-PA was appropriate in our trial. This agent is the most widely used thrombolytic drug in the United States. In addition, the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) trial demonstrated that treatment with t-PA and intravenous heparin results in a lower mortality rate than therapy with streptokinase5. Even when compared with streptokinase therapy, angioplasty was associated with a reduced rate of reinfarction6. With regard to health care costs, Gibbons et al.3 demonstrated that primary percutaneous transluminal coronary angioplasty was less costly than intravenous thrombolytic therapy. In addition, preliminary data from our trial indicate that the two approaches are similar in cost.
In response to Beauchamp et al., since the Primary Angioplasty Myocardial Infarction Study Group was not funded externally, the majority of the participating centers did not keep detailed exclusion logs. Thus, we have only the data from sites that did keep track of such information. Of the 547 patients suspected of having an infarction during the recruitment period, 370 (68 percent) were excluded (Table 1Table 1
Reasons for the Exclusion of 370 Patients from the Study.). Of the patients excluded, 40 percent received thrombolytic therapy (usually administered at an outside hospital or as part of the Third International Study of Infarct Survival), 16 percent received immediate catheterization with angioplasty or bypass surgery when appropriate, and the remaining 44 percent of patients received no immediate reperfusion therapy. The outcomes in excluded patients included death in 14 percent, stroke in 3 percent, and congestive heart failure in 9.5 percent. Therefore, as in other trials of thrombolysis, the patients excluded from the trial were part of a high-risk population with mortality three times higher than that of the patients studied.6 ReferencesCindy L. Grines, M.D.
William W. O'Neill, M.D.
William Beaumont Hospital, Royal Oak, MI 480731
Nicod P ,Gilpin EA ,Dittrich H , et al. Trends in use of coronary angiography in subacute phase of myocardial infarction. Circulation 1991;84:1004-1015
Web of Science | Medline2
Muller DWM ,Topol EJ ,Ellis SG , et al. Determinants of the need for early acute intervention in patients treated conservatively after thrombolytic therapy for acute myocardial infarction. J Am Coll Cardiol 1991;18:1594-1601
CrossRef | Web of Science | Medline3
Gibbons RJ ,Holmes DR ,Reeder GS ,Bailey KR ,Hopfenspirger MR ,Gersh BJ . Immediate angioplasty compared with the administration of a thrombolytic agent followed by conservative treatment for myocardial infarction. N Engl J Med 1993;328:685-691
Full Text | Web of Science | Medline4
The TIMI Study Group
Full Text | Web of Science | Medline5
Topol EJ. Results of the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) trial. Presented at the American Federation for Clinical Research, Washington, D.C., April 29, 1993. abstract.
6
Zijlstra F ,de Boer MJ ,Hoorntje JCA ,Reiffers S ,Reiber JHC ,Suryapranata H . A comparison of immediate coronary angioplasty with intravenous streptokinase in acute myocardial infarction. N Engl J Med 1993;328:680-684
Full Text | Web of Science | Medline
Author/Editor Response
We agree with Beauchamp and colleagues that referral-filter bias can affect the generalizability of the results of a clinical trial. We registered all patients admitted to our hospital who fulfilled the inclusion criteria for our study, but who were excluded for other reasons. During the study period, six eligible patients were not randomized. Two patients declined to participate, one died in the emergency room before randomization, and one who lived outside the Netherlands was not randomized because of anticipated problems with follow-up. In two cases the attending physicians did not wish to include the patient in the study.
Felix Zijlstra, M.D., Ph.D.
Menko Jan de Boer, M.D.
Harry Suryapranata, M.D., Ph.D.
Ziekenhuis de Weezenlanden, 8011 JW Zwolle, the NetherlandsAuthor/Editor Response
Beauchamp et al. raise the important issue of referral-filter bias. As indicated in our article, the study group represented more than 90 percent of the patients who met the inclusion and exclusion criteria. The most common reason for the failure to enroll a patient was the patient's preference for one of the initial therapies. Although our entry criteria were similar to those of many other thrombolytic trials, an urban population-based study1 has clearly demonstrated that less than half of all patients with myocardial infarction will be eligible on the basis of such criteria. In patients ineligible because of contraindications to thrombolysis, immediate angioplasty should be considered as an alternative therapy.
In response to Hughes-Davies, a patient charge for 100 mg of commercially available t-PA ($3,400) was included in the costs for the thrombolytic-therapy group, although the drug was supplied free of charge by the manufacturer. When the patient charge for 1.5 million units of streptokinase ($314) was substituted, the difference between the groups was not significant.
Rowe correctly points out that imaging with technetium-99m sestamibi may underestimate myocardial viability in the setting of hibernating myocardium, a limitation that we have acknowledged previously2. Despite this potential limitation, the size of the resting perfusion defect on sestamibi imaging correlates closely with the amount of histologic fibrosis in hearts after cardiac transplantation3. Sestamibi imaging can reliably identify myocardial viability in regions with stunned myocardium4 and in regions with hibernating myocardium and mild-to-moderate reductions in resting blood flow. Its main limitation is in regions with hibernating myocardium and a severe reduction in resting blood flow, which will require a residual stenosis of more than 90 percent. Previous studies with thallium scintigraphy5 suggest that myocardial viability is present in less than 20 percent of such regions. In our study, of the 31 patients treated with thrombolysis who did not have angioplasty or bypass surgery during the hospitalization, 18 underwent coronary angiography; only 3 of the 18 had a stenosis of more than 90 percent in the infarct-related artery with evidence of myocardial viability on thallium scintigraphy with reinjection. The measured infarct size on sestamibi imaging in these patients averaged only 3 percent of the size of the left ventricle, suggesting that the underestimation of myocardial viability was small. We therefore doubt that hibernating myocardium had a substantial effect on our results.
5 ReferencesR.J. Gibbons, M.D.
D.R. Holmes, M.D.
G.S. Reeder, M.D.
Mayo Clinic, Rochester, MN 559051
Karlson BW ,Herlitz J ,Edvardsson N ,Emanuelsson H ,Sjolin M ,Hjalmarson A . Eligibility for intravenous thrombolysis in suspected acute myocardial infarction. Circulation 1990;82:1140-1146
CrossRef | Web of Science | Medline2
Christian TF ,Gibbons RJ ,Gersh BJ . Effect of infarct location on myocardial salvage assessed by technetium-99m isonitrile. J Am Coll Cardiol 1991;17:1303-1308
CrossRef | Web of Science | Medline3
Medrano R ,Wielbaecher D ,Young JB , et al. Assessment of myocardial viability with technetium-99m sestamibi in patients undergoing cardiac transplantation: a scintigraphic-pathologic study. Circulation 1992;86:Suppl I:I-108 abstract.4
Christian TF ,Behrenbeck T ,Pellikka PA ,Huber KC ,Chesebro JH ,Gibbons RJ . Mismatch of left ventricular function and infarct size demonstrated by technetium-99m isonitrile imaging after reperfusion therapy for acute myocardial infarction: identification of myocardial stunning and hyperkinesia. J Am Coll Cardiol 1990;16:1632-1638
CrossRef | Web of Science | Medline5
Bonow RO ,Dilsizian V ,Cuocolo A ,Bacharach SL . Identification of viable myocardium in patients with chronic coronary artery disease and left ventricular dysfunction: comparison of thallium scintigraphy with reinjection and PET imaging with 18F-fluorodeoxyglucose. Circulation 1991;83:26-37
Web of Science | Medline
