Thanks Anton; this one is more interesting, but the clear message is that
chest pain severity is related to infarct size, and chest pain resolution is
related to myocardial salvage, which would "support" my tendency to be
conservative with painfree patients.

Note however ambiguity of usage of the term "significant" which frequently
occurs in such papers, and I quote: "Patients without chest pain at the
start of thrombolysis still demonstrated significant salvage (11 +/- 11% of
the left ventricle, p = 0.009)." Later however the paper states:
"Thrombolysis was most effective in the 20 patients (32%) with moderate or
severe chest pain and complete resolution of symptoms during thrombolysis
(salvage of 79% to 89% of the area at risk). In the remaining 32 patients
with chest pain, salvage of the area at risk was only 32%."

You see my point; the 32 patients with mild chest pain or unresolved chest
pain achieved myocardial salvage of "only" 32% according to the authors (as
they're comparing with other salvages of over 80%) while the pain-free
patients achieved salvage of 11%. Clearly 11%, while "significant" in a
statistical sense (as p = 0.009) is not considered by these authors to be
significant in a clinical sense. I would agree with them, as the 11% salvage
index probably only relates to a salvaged ejection fraction of some 2 or 3%.
And 2 or 3% ejection fraction is neither here nor there when it comes to
long term morbidity, and 2 or 3% is also perilously close to your
complication rate with thrombolytic agents.

This is certainly an area that deserves more work, although such patients
are in a very small minority and difficult to study prospectively. I suspect
such painfree patients have spontaneously reperfused, as we do see this from
time to time with their ECGs normalising as long as you're prepared to watch
and wait, which can be a very tough thing to do in acute medicine! I could
be wrong about all of this Anton, but that's my "feeling" at the moment,
pending further evidence.

Regards

Adrian Fogarty

----- Original Message -----
From: <[log in to unmask]>
To: <[log in to unmask]>
Sent: Sunday, March 09, 2003 2:33 AM
Subject: Re: Thrombolysis Question


> Good critique of the paper - you are quite right, Adrian, it suffers
predominantly from being a post-hoc analysis.
>
> Only other paper I could lay my hands on is a myocardium salvage study, in
which they note that patients without chest pain still demonstrate
myocardial salvage using Tech scanning, though the authors note that chest
pain severity is, unsurprisingly, still important. (Abstract follows, as
does comment by Olson.)  Can debate the outcome importance of myocardial
salvage, but I think is important for CCF development, which would be in
keeping with the retrospective 1994 Cox paper.
>
> The advice from our local cardiologists is to perform thrombolysis where
the patient has had a convincing history of cardiac pain but is presently
pain free (obviously with ST elevation).
>
> Anton
>
> J Am Coll Cardiol 1993 Nov 1;22(5):1311-6
>
> Severity and response of chest pain during thrombolytic therapy for acute
myocardial infarction: a useful indicator of myocardial salvage and infarct
size.
>
> Christian TF, Gibbons RJ, Hopfenspirger MR, Gersh BJ.
>
> Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic,
Rochester, Minnesota 55905.
>
> OBJECTIVES. The purpose of this study was to determine noninvasively
whether chest pain severity is predictive of the amount of myocardium at
risk and whether the response of pain during thrombolysis is associated with
myocardial salvage during acute myocardial infarction. BACKGROUND. The
perception of chest pain and response to reperfusion therapy during acute
myocardial infarction may provide important information for treatment
benefit. Previous studies have been limited by the inability to measure
myocardium at risk and myocardial salvage. METHODS. Sixty-two patients with
acute myocardial infarction received an injection of technetium-99m
sestamibi before thrombolysis and again at hospital discharge. Tomographic
imaging was performed 1 to 6 h later. Myocardium at risk, infarct size and
absolute myocardial salvage were derived from these images using previously
described techniques and were expressed as a percent of the left ventricle.
Salvage index was calculated by divid!
> ing myocardial salvage by the myocardium at risk. Chest pain severity was
graded before thrombolysis as none, mild, moderate or severe. Chest pain
response during thrombolytic therapy was graded as none, partial or
completely resolved. RESULTS. There was no association between chest pain
severity and myocardium at risk, but there was a weak trend toward greater
myocardial salvage and salvage index (p = 0.09 and p = 0.12, respectively)
for patients with more severe symptoms. There was a significant association
between chest pain response to therapy and both myocardial salvage (p =
0.03) and salvage index (p = 0.01). By multivariate analysis, chest pain
severity and response of chest pain during thrombolysis were significant
independent predictors of myocardial salvage, salvage index and infarct
size. CONCLUSIONS. These findings suggest that the assessment of chest pain
before and after thrombolytic therapy is a readily available, useful
indicator of the efficacy of the therapy.
>
>
> J Am Coll Cardiol 1993 Nov 1;22(5):1317-9
>
> Comment on:
> J Am Coll Cardiol. 1993 Nov 1;22(5):1311-6.
>
> Bedside cardiology and thrombolysis.
>
> Olson HG.
>
> The study by Christian et al. (37) provides useful clinical information
for the management of the patient with myocardial infarction with
thrombolytic therapy. 1) Complete resolution of chest pain during the
infusion of an intravenous thrombolytic agent suggests reperfusion and
myocardial salvage. Initial conservative medical management should be
considered in these patients especially if serial ECGs show a progressive
and rapid downward defection of the ST segment. 2) Patients presenting
within 6 h after the onset of infarction, who are pain free, may still
benefit from thrombolysis if their ECGs show persistent ischemia. 3) In the
remaining patients in whom chest pain does not resolve completely during
thrombolytic therapy, management, whether continued medical or invasive
strategy, should be individualized and dictated by the extent of myocardium
at risk (i.e., by the number of ECG leads showing ST segment elevation), the
response of the ST segment to thrombolysis and, mo!
> st important, the clinical status of the patient.
>
> In message <002701c2e5d9$c2227fe0$e177fea9@mydell> Adrian Fogarty
<[log in to unmask]> writes:
> > But this was a retrospective and uncontrolled study Anton, i.e. the
painfree
> > patients were ALL thrombolysed, so you can't draw any conclusions about
what
> > would have happened had they not been thrombolysed, which is the crux of
the
> > original question from Phil Munro. You have misinterpreted, or rather
> > misapplied the findings of this paper to Phil's patient, but this is
clearly
> > not applicable. The only way to "prove" your point would be to divide
the
> > 337 painfree patients into thrombolyse or non-thrombolyse and then
compare
> > outcomes; this paper did not do that.
> >
> > Basically this paper fails to demonstrate that the painfree patients
would
> > not have ended up with good ejection fractions (or less risk of failure
etc)
> > had they not received thrombolysis in the first place, as we just don't
know
> > this because they all were given thrombolysis from the outset. My
hypothesis
> > (the null hypothesis) is that these patients would have fared just as
well,
> > perhaps better, without thrombolysis, but this paper does not test this
> > hypothesis so it cannot be relied upon.
> >
> > Adrian Fogarty
> >
> >
> > ----- Original Message -----
> > From: <[log in to unmask]>
> > To: <[log in to unmask]>
> > Sent: Saturday, March 08, 2003 9:32 PM
> > Subject: Re: Thrombolysis Question
> >
> >
> > > Phil,
> > >
> > > Would still thrombolyse - these patients seem to develop less CCF than
> > their peers with chest pain at the time of thrombolysis, otherwise
similar
> > in other respects. (Abstract follows)
> > >
> > > Anton
> > >
> > > Am J Cardiol 1994 Apr 15;73(11):729-36
> > >
> > > Effect on outcome of the presence or absence of chest pain at
initiation
> > of recombinant tissue plasminogen activator therapy in acute myocardial
> > infarction. The Thrombolysis in Myocardial Infarction Investigators.
> > >
> > > Cox DA, Rogers WJ, Aguirre FV, Forman S, Solomon R, Zaret BL.
> > >
> > > University of Alabama Medical Center, Birmingham 35294.
> > >
> > > To ascertain whether the outcome of patients with suspected myocardial
> > infarction differs when chest pain is still present at initiation of
> > thrombolytic therapy, participants in the Thrombolysis in Myocardial
> > Infarction Phase II study, all of whom presented within 4 hours of
symptoms
> > onset, were retrospectively divided into 2 groups: (1) those with chest
pain
> > present at onset of intravenous thrombolysis, n = 3,000; and (2) those
who
> > were free of chest pain on beginning intravenous thrombolytic therapy, n
=
> > 337. Patients free of chest pain were older (58 vs 57 years, p = 0.01),
more
> > often women (23 vs 17%, p = 0.01), had fewer electrocardiographic leads
with
> > ST elevation (3.8 vs 4.1, p < 0.001), and the presenting event was
confirmed
> > less often as myocardial infarction than as chest pain without
infarction
> > (88 vs 96%, p < 0.001). There were no significant differences between
the 2
> > groups for in-hospital death, reinfarction, recurrent ischemic events,
> > stroke, overall hemorrhag!
> > > ic complications, coronary angioplasty or bypass surgery. At 6-weeks
> > follow-up, more pain-free patients had resting ejection fraction > 0.55
(35
> > vs 31%, p = 0.001) and fewer developed congestive heart failure (12 vs
20%).
> > At 1-year follow-up, fewer pain-free patients developed congestive heart
> > failure (15 vs 21%, p = 0.009), but no differences existed between the 2
> > groups in frequency of death, reinfarction, coronary angioplasty, bypass
> > surgery or anginal class. Thus, there are several observations in
patients
> > who were free of chest pain at onset of lytic therapy. (1) The majority
> > developed enzymatic or electrocardiographic evidence of acute myocardial
> > infarction.