52 Articles: Lysis vs Transfer for PCI (NEJM 2003)

ST segments:
a lytic in my hand,
or transfer?

-Unknown, 2015

This is part of a recurring series examining landmark articles in Emergency Medicine, in the style of ALiEM’s 52 Articles.


Andersen et al., NEJM 2003.  A Comparison of Coronary Angioplasty with Fibrinolytic Therapy in Acute Myocardial Infarction.

Main Points:

  1. This prospective randomly-controlled trial investigated whether patients with STEMIs who presented to community hospitals would benefit from transport to specialized centers for percutaneous coronary intervention (PCI) rather than remaining at the local hospital for fibrinolysis. The authors demonstrated an almost 5% reduction in reinfarction rate at 30 days between the two groups.
  1.  STEMI patients that presented to community centers and received aspirin+metoprolol+heparin+PCI within 2 hours, including transfer time, had statistically significantly fewer reinfarctions at 30 days than the STEMI patients who remained at the initial facility andon alteplase+aspirin+metoprolol+heparin.  Ultimately, PCI was shown to be superior to fibrinolytic if available in a timely fashion.


It is now an established tenet that timely PCI in an experienced center by experienced interventionalists is the best treatment for STEMI.  Historically, with no PCI available at the presenting facility, STEMI patients received anticoagulation and fibrinolytics.  But there were obvious and dangerous concerns in applying a “PCI-or-bust” policy universally, as there are only so many PCI centers to drive patients to, and “time is myocardium.”  So the best treatment for the patient with a STEMI in the middle of the proverbial forest was unknown.  Lyse or transfer?  This study made significant progress towards helping us answer this question.


1572 patients were enrolled, 1129 from 24 referral hospitals and 443 from 5 PCI centers.  All of these patients were randomly assigned to receive lysis & anticoagulation (LA) or anticoagulation & PCI (APCI), with the patients from the referral hospitals therefore being transferred for their PCI.  And yes, patients who presented to PCI centers and were assigned LA therefore did not receive PCI.  Risk characteristics of all the patients were similar.  LA patients received 300mg aspirin PO, 20mg metoprolol IV, an “accelerated” alteplase treatment (15mg bolus + 0.75mg/kg over 30min + 0.5mg/kg over the next 60min), and unfractionated heparin for 48h (5000U bolus + infusion titrated to aPTT 70-90sec).  APCI patients received 300mg aspirin IV, 20mg metoprolol IV, unfractionated heparin (10,000U bolus + infusion during procedure to activated clotting time 350-450s).  In patients receiving PCI, GPIIb/IIIa blockers were given at discretion of proceduralist, arteries were treated only if stenosed >30% or if TIMI “flow grade” was <3, non-infarct-related arteries were NOT intervened on, and all patient received Ticlodipine (500mg) or Clopidogrel (75mg) for 1 month afterwards.  The primary endpoint was a composite of death from any cause, clinical reinfarction, or disabling stroke at 30 days follow up.


  • For patients presenting to PCI centers, there was a 45% reduction in composite outcome with PCI, from 12.3% in the LA group to 6.7% in the APCI group (P = 0.05).
  • For patients presenting to referral hospitals, there was a 40% reduction in the composite outcome with PCI, from 14.2% in the LA group to 8.5% in the APCI group (P = 0.002)
  • Driving the difference within the composite outcome was clinical reinfarction: 1.6% in all APCI vs 6.3% in all LA (P =  0.001).  Differences in death and stroke rates at 30 days were not statistically significantly different between all APCI and all LA patients, whether they were transferred to a PCI center or were already there.
  • The median distance from referral site to PCI center was 50 km
  • Transfer was highly protocolized, with emphasis on shaving-off every precious second of ischemic time.  This cooperation and efficiency resulted in 96% of referral patients landing on the nearest participating PCI table in 2 hours or less.
  • The study was stopped after the 1129 patients, when the third interim analysis showed referral APCIs were reinfarcting less than the referral LAs; this was built into the design as a study cutoff as it implies that APCI is better in both settings.


  • The bottom line is the balance between the proven benefits of PCI and the ischemic time on the myocardium.  There are many variables at play in the “lyse vs transfer (vs both)” decision (see below). These variables include: location and severity of the ischemia, comorbidities and stability of the patient, EMS capabilities, distance, receiving PCI center experience, available medications, etc.  The decision may not always be clear, every patient is probably different, and there is an overwhelming amount of research on this topic.  Do your best.
  • Implementing the airtight transfer logistics they executed here is an obstacle, science aside.
  • In hindsight, your heart has to go out (no pun intended) to the 12.3% of LA patients who presented to a PCI center, got heparin & alteplase, and had a poor outcome.  Their willingness to participate has certainly saved lives.
  • Did you notice that the referred APCI patients did not receive fibrinolysis, only asa, metoprolol, and UFH?  The older research on immediate PCI after fibrinolysis showed no benefit and increased adverse effects.  New and ongoing research on “facilitated PCI,” or PCI after fibrinolysis, using ever-advancing interventional equipment and techniques, may be changing this for some high risk patients for whom PCI is not available within 90 minutes.  For further reading, see these:

Herrmann HC, et al.  Benefit of facilitated percutaneous coronary intervention in high-risk ST-segment elevation myocardial infarction patients presenting to nonpercutaneous coronary intervention hospitals.  JACC Cardiovac Interv.  2009 Oct;2(10):917-24.

Ellis SG, et al.  Facilitated PCI in patients with ST-elevation myocardial infarction.  N Engl J Med. 2008 Maay 22;358(21):2205-17.

Ellis SG, et al. Facilitated percutaneous coronary intervention versus primary percutaneous coronary intervention: design and rationale of the Facilitated Intervention with Enhanced Reperfusion Speed to Stop Events (FINESSE) trial. Am Heart J. 2004; 147: E16.

Level of Evidence:

Based on the ACEP grading system for therapeutic questions this study was graded level I.


The study inclusion criteria was defined as the symptoms for greater than 30 minutes, but less than 12 hours with associated cumulative ST-segment elevation of at least 4mm in at least two contiguous leads.

Resident Reviewer: Dr. Kazakin
Faculty Reviewer: Dr. Siket

1 thought on “52 Articles: Lysis vs Transfer for PCI (NEJM 2003)

Leave a Reply

Your email address will not be published. Required fields are marked *