Current issues of ACP Journal Club are published in Annals of Internal Medicine


Primary angioplasty led to improved clinical outcomes at a cost similar to tissue plasminogen activator in acute MI

ACP J Club. 1997 Nov-Dec;127:58. doi:10.7326/ACPJC-1997-127-3-058

Related Content in the Archives
Immediate angioplasty was better than streptokinase in acute MI

Source Citation

Stone GW, Grines CL, Rothbaum D, et al., for the PAMI Trial Investigators. Analysis of the relative costs and effectiveness of primary angioplasty versus tissue-type plasminogen activator: the Primary Angioplasty in Myocardial Infarction (PAMI) Trial. J Am Coll Cardiol. 1997 Apr;29:901-7.



To determine the relative costs and effectiveness of primary percutaneous transluminal coronary angioplasty compared with tissue plasminogen activator (t-PA) in patients with acute myocardial infarction (MI).


Cost analysis based on data obtained from a randomized controlled trial (Primary Angioplasty in Myocardial Infarction [PAMI] trial).


11 clinical centers in the United States.


395 patients (mean age 60 y, 72% men) with acute MI of ≤ 12 hours duration. Exclusion criteria were absence of ST-segment elevation, presence of left bundle-branch block, cardiogenic shock, risk factors for bleeding after thrombolytic therapy, or inability to give informed consent. Cost data were available for 358 patients (91%).


177 patients were allocated to angioplasty and 181 to t-PA, 100 mg (or 1.25 mg/kg body weight for patients weighing < 65 kg) over 3 hours. Intravenous heparin was continued in all patients for 3 to 5 days.

Main cost and outcome measures

In-hospital mortality, reinfarction, recurrent ischemia, stroke, total mean hospital charges, and professional fees. Resource consumption during late follow-up was estimated by assessment of major clinical events and functional status.

Main results

Compared with t-PA, angioplasty led to reduced rates of in-hospital mortality (2% vs 7%, P = 0.03), recurrent ischemia (11% vs 29%, P < 0.001) and stroke (0% vs 4%, P = 0.02), and shorter hospital stays (7.6 vs 8.4 d, P = 0.04). Total mean hospital charges were U.S. $3436 lower per patient with angioplasty than with t-PA (U.S. $23 468 vs U.S. $26 904, P = 0.04). The difference primarily resulted from a reduction in adverse in-hospital outcomes in the angioplasty group. Professional fees were higher after angioplasty (U.S. $4185 vs $3183, P = 0.001), so that total costs did not differ between the groups (U.S. $27 653 vs U.S. $30 227, P = 0.21). At a mean follow-up of 2.1 years, no differences existed between the 2 groups for postdischarge events or New York Heart Association functional class, suggesting similar late resource consumption.


Primary percutaneous transluminal coronary angioplasty led to improved clinical outcomes at a total cost similar to that of tissue plasminogen activator in patients with acute myocardial infarction.

Source of funding: No external funding.

For article reprint: Dr. G.W. Stone, The Cardiovascular Institute, 2660 Grant Road, Mountain View, CA 94040, USA. FAX 415-969-2807.


Percutaneous angioplasty led to improved clinical outcomes at a cost similar to tissue plasminogen activator in acute MI

It is still unclear which patients benefit from primary angioplasty as a treatment for acute MI. In the report by Stone and colleagues, more clinical benefit is derived in high-risk patients. However, Zijlstra and colleagues also report benefit in low-risk patients. Although the numbers in the latter study were small, reinfarction occurred in 16% and more patients reached the combined primary end point after thrombolytic therapy than after primary angioplasty. Reinfarction rates were higher than in most previous studies, which is possibly related to chance in this relatively small investigation. Benefits associated with angioplasty were derived from lower re-infarction rates without statistically significant differences in mortality rates.

Although the use of primary angioplasty seemed to have no extra cost, it can be debated whether the use of this procedure to prevent reinfarction (but with no difference in mortality) is justified, considering the resources required for its widespread use. Perhaps a more cost-conscious way would be to carefully assess patients after thrombolysis and to consider intervention in those who fail to reperfuse (1, 2), so-called rescue angioplasty. Although the clinical criteria used for reperfusion may lack a high positive predictive value, it is clear that 20% to 30% of patients do not have Thrombolysis in Myocardial Infarction (TIMI) grade 2 or 3 perfusion and up to 40% do not achieve grade 3. Careful evaluation of the efficacy of the thrombolytic agent in each individual patient is currently widespread in clinical practice. Full use of early angioplasty in those who seem not to have reperfused could have led to early transfer of high-risk patients who were most likely to benefit in terms of mortality and morbidity to facilities equipped for angioplasty. This strategy of careful early evaluation for failure to reperfuse and invasive management in a targeted group could lead to increased acceptance of a role for angioplasty as a treatment for acute MI.

The cost-effectiveness of primary angioplasty compared with thrombolysis for acute MI remains a contentious issue. It has been fairly well established that, at least in relatively selected patients, primary angioplasty ensures TIMI grade 3 flow in a higher proportion of patients than can currently be achieved with the best thrombolytic therapy. Better TIMI flow has been shown to correspond to improvement in short- and long-term outcomes (3). It is less clear whether the results reported in the PAMI trial can be reproduced in most clinical settings and at what true cost. Should the goal of obtaining the best patency be paramount and, if so, how many balloons, stents, or newer agents, such as ReoPro (Centocor, Malvern, PA), will be required? Although these issues are the subject of ongoing studies, the trial by Stone and colleagues suggests that the outcome benefits have no additional cost; however, some issues that surround the data presented warrant cautious interpretation.

First, it is uncertain whether the high patency rates achieved by angioplasty in the PAMI study can be consistently attained if the technique becomes more widely used by less-skilled operators. Second, readers really want to know what the cost is, not what is charged. The authors acknowledge this potentially serious shortcoming in the discussion section of their paper. Finally, the extra cost of thrombolysis primarily resulted from the need for repeated catheterization and subsequent nonprotocol angioplasty and coronary surgery. Can we be sure that all these further interventions were done according to protocol and not under the pressures physicians may feel to provide a treatment initially denied by randomization in the trial? One important potential resource saving is the shortening of the hospital stay for patients known to have TIMI 3 flow in all arteries after angioplasty.

In the many clinical settings, primary angioplasty may cost more than is suggested by these studies because the start-up costs for new interventional catheter laboratories and the provision of personnel for 24-hour coverage must be included. These substantial funding requirements are likely to limit wide dissemination of primary angioplasty. Also, new trials will need to address improvements in angioplasty techniques as well as in pharmacologic therapies for acute MI (new thrombolytic agents and adjunctive therapies vs stents and ReoPro). Transfer of patients most likely to benefit from intervention may be the preferable option, at least in the short term, but it must be stated that this has not been shown in trials. In the meantime, in situations in which primary angioplasty is available, these studies show that it is a credible clinical option.

A. H. Gershlick, MD
University of LeicesterLeicester, England, UK


1. Klootwijk P, Langer A, Meij S, et al. Eur Heart J. 1996;17:689-98.

2. Zabel M, Hohnloser SH, Koster W, et al. Circulation. 1993;87:1542-50.

3. Anderson JL, Karagounis LA, Califf RM. Am J Cardiol. 1996;78:1-8.