Correspondence
A Comparison of Thrombolytic Therapy with Primary Coronary Angioplasty for Acute Myocardial Infarction
N Engl J Med 1997; 336:1103-1104April 10, 1997
- Article
To the Editor:
The report by Every et al. (Oct. 24 issue)1 on the analysis of data from the Myocardial Infarction Triage and Intervention (MITI) registry underscores an important debate regarding the current standard of care for acute myocardial infarction. However, the conclusions require clarification. The authors claim that there is a savings of $3,000 per patient if thrombolysis is used instead of primary angioplasty. Unfortunately, only the inpatient costs were analyzed, instead of the total medical cost. Noninvasive outpatient evaluations (e.g., echocardiographic studies and stress tests) were not included. Outpatient assessment of ischemia is routine after thrombolysis in patients with myocardial infarction. Utilization rates for noninvasive outpatient procedures would probably be higher for such patients. The authors also conclude that the rate of repeated catheterization and angioplasty is higher in the primary-angioplasty group because more patients in this group had restenosis. It is more likely that the difference reflects the practice styles of the cardiologists and institutions involved. No standardized protocol was used to determine the presence of recurrent ischemia or the need for repeated angiography. Thus, the original biases that led the subgroup of cardiologists to recommend primary angioplasty initially would have been present in determining the need for repeated catheterization.
Such bias is well documented. Di Salvo et al. found that patients under the care of invasive cardiologists were more likely to undergo angioplasty than those under the care of noninvasive cardiologists, even when the extent of disease was similar.2 It would be informative to know what proportion of patients in each group in the MITI trial underwent noninvasive evaluation for recurrent symptoms instead of proceeding directly to catheterization. This information would also directly affect the cost calculations.
These biases may also be present at the institutional level. From Table 2 in the report by Every et al., it appears that 76 percent of the patients in the primary-angioplasty group (801 of 1050) were treated at three of the high-volume hospitals. In contrast, nearly half the patients in the thrombolytic-therapy group seem to have been admitted to the eight hospitals without catheterization laboratories. Since hospital costs and patient outcomes, as well as resource-utilization rates, differ significantly among hospitals, could the differences found in the MITI study reflect the maldistribution of patients rather than the revascularization strategy? In fact, the MITI investigators as well as others have noted that patients admitted to hospitals with on-site catheterization laboratories were much more likely to undergo angiography than patients admitted to hospitals without such laboratories.3,4 Outpatient decision making would probably suffer from the same biases. A breakdown of costs, rates of repeated catheterization, and outcomes at the various hospitals in the study may elucidate this matter.
4 ReferencesJoon Sup Lee, M.D.
University of Pittsburgh Medical Center, Pittsburgh, PA 152131
Every NR ,Parsons LS ,Hlatky M ,Martin JS ,Weaver WD . A comparison of thrombolytic therapy with primary coronary angioplasty for acute myocardial infarction. N Engl J Med 1996;335:1253-1260
Full Text | Web of Science | Medline2
Di Salvo TT ,Paul SD ,Lloyd-Jones D , et al. Care of acute myocardial infarction by noninvasive and invasive cardiologists: procedure use, cost and outcome. J Am Coll Cardiol 1996;27:262-269
CrossRef | Web of Science | Medline3
Every NR ,Fihn SD ,Maynard C ,Martin JS ,Weaver WD . Resource utilization in treatment of acute myocardial infarction: staff-model health maintenance organization versus fee-for-service hospitals. J Am Coll Cardiol 1995;26:401-406
CrossRef | Web of Science | Medline4
Blustein J . High-technology cardiac procedures: the impact of service availability on service use in New York State. JAMA 1993;270:344-349
CrossRef | Web of Science | Medline
To the Editor:
Every et al. did not find any significant difference in mortality between the primary-angioplasty group and the thrombolytic-therapy group. However, there are several limitations, some mentioned in the report, that preclude drawing any firm conclusions from this study.
It is particularly important to point out that the management of myocardial infarction in patients living in Seattle, where the study was carried out, is quite different from the management in many other places. Therefore, the conclusions drawn from this study cannot be generalized to other settings. Seattle has an established and well-developed mobile coronary care unit run by paramedics, with an average response time of eight minutes.1 In addition, because of an ongoing program of myocardial-infarction trials in the 19 participating hospitals, most patients receive more intense attention from very experienced medical personnel than they might otherwise.1 Presumably, the patients participating in the original MITI study,1 which compared thrombolysis initiated before hospitalization with thrombolysis initiated in the hospital,1 were also included in the current study. Ninety-seven percent of the patients in the original study received treatment within three hours after the onset of symptoms.1 Furthermore, in the current study, more than 25 percent of the patients who received thrombolytic therapy underwent coronary revascularization on the day of admission.
These factors have led to a shorter time to thrombolysis (a median of one hour) and a higher level of care among patients treated for acute myocardial infarction in Seattle than among those treated elsewhere. This difference is reflected in lower mortality rates both in the original MITI study1 and in the current study, even in the high-risk group. In the original MITI study, the in-hospital mortality rate among the patients treated with thrombolytic therapy within 70 minutes after the onset of chest pain was only 1.2 percent.1 It is hard to believe that a lower mortality rate could be achieved by any other means. In the current study, overall mortality in the thrombolytic-therapy group was 5.6 percent, and a prospective, randomized study would require nearly 12,000 patients to show a difference in mortality of 20 percent. Even with a mortality rate of 8.1 percent in the high-risk subgroup, the number of patients required would be more than 8000. The numbers in the current study are clearly not large enough.
1 ReferencesFarzin Fath-Ordoubadi, M.B., B.Chir.
Kevin J. Beatt, Ph.D.
Hammersmith Hospital, London W12 0NN, United Kingdom1
Weaver WD ,Cerqueira M ,Hallstrom AP , et al. Prehospital-initiated vs hospital-initiated thrombolytic therapy. JAMA 1993;270:1211-1216
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We appreciate the issues raised by Dr. Lee and Drs. Fath-Ordoubadi and Beatt and have, in fact, considered most of them as limitations of our study. It is important to remember that the goal of our cohort study was to evaluate the outcome and costs of primary angioplasty as compared with thrombolysis in the community setting and to see whether the results of randomized trials could be replicated, not to draw firm conclusions about which therapy is more efficacious. In such a study design there will indeed be bias, but this does not invalidate our finding that in the community setting, these therapies result in similar outcomes.
In response to Dr. Lee's comments, we agree that a limitation of the study was that outpatient costs were not included in the calculations of total cost. If there were substantial differences between outpatient office visits and procedures between patients treated with thrombolysis and those treated with primary angioplasty, this may have biased our findings. However, we are aware of no data that would support this assertion. In addition, if there was a bias to use more noninvasive testing in patients treated with thrombolysis, we should have captured this finding in our analyses of hospital charges, since much of this testing would have been done before discharge. We also agree that greater use of procedures after discharge in the primary-angioplasty group may have been a result of a more invasive strategy employed by physicians who favor primary angioplasty. In our view, however, this is not a limitation but a reflection of potentially higher costs associated with primary angioplasty in the community setting.
In response to Drs. Fath-Ordoubadi and Beatt: there were indeed 180 patients treated in the prehospital setting as part of the MITI trial who were included in our thrombolysis group. In these patients, the mean time from the onset of symptoms to treatment was 92 minutes and in-hospital mortality was 5.3 percent — not substantially different from the findings for the 1915 other patients included in the thrombolysis cohort. Although the time from the onset of symptoms to treatment in the Seattle hospitals was relatively short, the mean time to treatment in the thrombolysis cohort (1.0 hour) was not substantially different from that reported in the National Registry of Myocardial Infarction1 and thus should be generalizable to many urban areas in the United States.
We believe it is important for readers to recognize that the mortality among patients treated with aggressive thrombolysis was almost 2 percent lower than in the larger Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) trial and that the mortality among those treated with primary percutaneous transluminal coronary angioplasty was somewhat higher than in the early randomized trials. We believe that the two treatments result in excellent and similar outcomes.
1 ReferencesNathan R. Every, M.D., M.P.H.
Veterans Affairs Puget Sound Healthcare System, Seattle, WA 98108W. Douglas Weaver, M.D.
Henry Ford Health Care System, Detroit, MI 482021
Maynard C ,Weaver WD ,Lambrew C ,Bowlby LJ ,Rogers WJ ,Rubison RM . Factors influencing the time to administration of thrombolytic therapy with recombinant tissue plasminogen activator (data from the National Registry of Myocardial Infarction). Am J Cardiol 1995;76:548-552
CrossRef | Web of Science | Medline
