March 14th, 2013

Fibrinolysis Now or Primary PCI Later?

, and

CardioExchange’s Rick Lange and David Hillis ask Frans Van de Werf, principal investigator for the STREAM study, about his study group’s findings and their implications for clinical practice.


In the STREAM trial, investigators compared two treatment strategies in STEMI patients presenting early after symptom onset who were unable to undergo primary PCI within 1 hour: (1) prompt fibrinolysis (in a prehospital setting or the emergency room of a community hospital without PCI capability), followed by timely catheterization and appropriate rescue coronary intervention; and (2) transport for primary PCI.

The primary endpoint — a composite of  death, shock, congestive HF, and reinfarction within 30 days – occurred in 12.4% of the fibrinolysis group versus 14.3% of the primary-PCI group (relative risk, 0.86; P=0.21. Intracranial hemorrhage was more common with fibrinolysis than with primary PCI (1% vs. 0.2%, P=0.05), especially in the elderly. After the protocol-defined tenecteplase dose was cut in half for patients older than 75, the rate of intracranial hemorrhage was not significantly higher in the fibrinolysis group than in the primary-PCI group (0.5% and 0.3%, P=0.45).


Lange and Hillis: How does this trial differ from other trials of fibrinolysis versus primary PCI (e.g., DANAMI-2, CAPTIM, PRAGUE-2)?

Van de Werf: The inclusion of patients presenting much earlier after onset of symptoms and the much earlier start of lytic therapy compared with the invasive procedure are important differences from DANAMI-2. Our population, even after dose reduction of tenecteplase (n=1503), was also much larger than the referral population of DANAMI-2. Other important features of our trial as compared with the other trials are more–up-to-date antithrombotic co-therapy and the requirement of catheterization between 6 and 24 hours after randomization in patients with evidence of reperfusion after lytic therapy.

Lange and Hillis: With comparable efficacy in the two treatment arms but more intracerebral bleeding in the thrombolysis arm, is a recommendation for fibrinolysis difficult to defend?

Van de Werf: It all depends on how much ischemic myocardium you may be able to salvage by earlier reperfusion with a lytic, and thus on the system delay to primary PCI.

By the way, after the protocol amendment, there were more fatal strokes in the primary-PCI arm than in the fibrinolysis arm (4 vs. 3)!

Lange and Hillis: What about patients who present with symptoms more than 3 hours after symptom onset, or those who can undergo primary PCI within the 2-hour period recommended by the European and North American guidelines?

Van de Werf: In our study, we included only patients presenting within 3 hours and unable to undergo prompt primary PCI. I would assume that the benefit of earlier reperfusion in a patient presenting after 3 hours is much smaller. I don’t think these patients should get lytic therapy before transport, unless the ECG suggests that the occlusion occurred more recently than the onset of symptoms did. It is sometimes difficult to pinpoint the onset of a STEMI exactly.

Lange and Hillis: If you had an acute MI, and your local hospital didn’t have primary PCI capability, would you want prehospital fibrinolysis or transportation to a remote facility that is PCI-capable?

Van de Werf: It depends on how much time it would take to get the PCI, and who would do the procedure!

One Response to “Fibrinolysis Now or Primary PCI Later?”

  1. There is little doubt the STREAM trial was well designed. But when you look at DANAMI-2, PRAGUE-2, CAPTIM and STREAM, accepting there are big differences in trial design, 30 day mortality is not different between PCI and throbolysis groups, but the stroke rate is significantly different (15/2586 vs. 44/2561). If I had a low bleeding risk and a long transfer time, had presented early, and perhaps had an inferior infarct I would prefer thrombolysis though.