GUSTO:
  Global Utilization of Streptokinase and tPA for Occluded Coronary Arteries
  ==========================================================================

þ 20,000+ patients enrolled; randomized to streptokinase or tPA; those
  receiving streptokinase further randomized to heparin IV or subQ.

þ front-loaded tPA slightly but significantly better than streptokinase in
  terms of mortality; this is maintained even past the initial 30-day period.
  BUT!!  see below <c-gusto.txt -retraction>

þ GUSTO IIb: shows angioplasty slightly better than thrombolysis.

was I the only person in the world who read the letter from the GUSTO
investigators "More on the GUSTO trial" NEJM:331;277-278?? In this letter
one year after the so-called 'results' of the trial the investigators report
that once they had analysed ALL the data the bit about 4 hours was no longer
valid!

Three other interesting facts often ignored:
1. the group treated within 2 hours show no difference between the two agents.
2. there was no difference in outcome SK vs tPA in the <55y group (mortality
2.0% nd 1.8% respectively)
3. the group receiving 'both' received subtotal doses of each agent

My conclusion from the results is that tPA MAY be the agent of choice for
those patients with anterior MI AND aged 55-75. This then excludes the 32%
of patients in the trial who were aged <55 and not just the meer 12% over
75. I couldn't find a reference to being Libra or Gemini ala ISIS-2 :-)

Isn't it wonderful how selective data reporting has influenced official dogma?

Dr Garry Wilkes MBBS FACEM
Department of Emergency Medicine
Royal Brisbane Hospital
Queensland, Australia
Email G.Wilkes@mailbox.uq.oz.au

----------------------------------------

I have eventually got around to chasing the reference & I cannot help
reading it as a retraction of the original Gusto findings.  Am I reading
these two paragarphs from the letter incorrectly or has there been a
mountain of medical literature written about an alleged survival difference
that never existed????

[snip]  In September 1993 we reported on the Global Utilization of
Streptokinase & Tissue plasminogen Activator for Occluded Coronary Arteries
(GUSTO) trial, which involved 41,021 patients with acute myocardial
infarction (Sept. 2 issue)*  In that report , we erred in stating that there
was a 99.9% level of completeness of "30 day mortality outcomes"; this
should have read "in-hospital outcomes."  Since our initial report, we have
received complete data on outcomes at 30 days in all but 98 patients (0.2 %)
and there has been refinement in the data base for variables for which
either data were missing or queries were pending.  With the exception of one
analysis, these represent only minor changes that have no substantive effect
on the data or their interpretation.  The final data have been deposited
with the National Auxiliary Publications Service (NAPS).

There is one change, however that deserves attention & is related to the
analysis of the study groups according to the time to treatment.  In the
original report, we found that the benefit of accelerated tissue plasminogen
activator (t-PA) treatment as compared with streptokinase treatment was
greater when the agent was administered early (within four hours) after the
onset of symptoms (i.e. we found a significant interaction between the time
to treatment & the reduction in mortality) Figure 1 shows the final data on
the group given accelerated t-PA, as compared with the pooled groups given
streptokinase (these data correspond to those in Figure 3 of the original
paper)  The interaction between time to treatment and the reduction in
mortality is no longer statistically significant (P=0.38).  The significant
changes in the time-to-treatment data are attributed to the more extensive
process involved in sorting this particular variable, for which data were
miscoded or initially left out.  The data base was finalized in March 1994
with respect to 30-day outcomes, and no further changes are anticipated.
[snip]

Tony Nocera
NRMA CareFlight & Nepean Hospital Sydney, Australia

------------------

Tony

I believe that you are misreading here.  What it says is that the
relation of benefit (of TPA vs Strepto) which initially appeared to be a
function of time is no longer time dependant.  I am not sure what this
says about the overall benefit in mortality but it seems to imply that
it still holds.
--
Richard O. Gray MD.
Assistant Professor Emergency Medicine
Hennepin County Medical Center

----------------------------