(Circulation. 1998;97:1649-1650.)
© 1998 American Heart Association, Inc.
Association Between Thrombolytic Treatment and the Prognosis of Hemodynamically Stable Patients With Major Pulmonary Embolism
Emile Ferrari, MD;
; Marcel Baudouy, MD
Cardiology Department,
Hopital Pasteur,
Nice, France
To the Editor:
Konstantinides et al1 recently reported the results of a
major multicenter registry involving more than 700 patients with severe
pulmonary embolism (PE) but no clinical instability. They
deserve to be congratulated for their interesting data. However, we
would like to comment on these data.
We believe that it is particularly difficult to assess PE gravity using
only clinical criteria. When patients are prescribed bed rest for a few
hours, we have often encountered significant discrepancy between
"clinical status" and echocardiography,
pulmonary angiography, or right-side
catheterization results.2 After a few
hours' bed rest, patients who have suffered from syncope or transient
collapse before hospitalization are often quiet in their beds. Probably
the slightest exertion or stress (or the slightest recurrence
of PE) would lead to severe clinical signs.
As a result, we believe that in some cases, the lack of hypotension may
lead to misdiagnosing the seriousness of PE. In particular, in young
patients with no concomitant diseases, blood pressure may be maintained
in normal ranges by several compensatory mechanisms. The fact that 70%
of the patients reported presented with tachycardia
and with a mean oxygen partial pressure of 56 mm Hg strengthens
this observation. Consequently, hypotension should not be used as the
only clinical criterion to justify thrombolytic
treatment. In this regard, we totally agree with the results of
Konstantinides et al.
Another important point is the interpretation of the results of this
study. As the authors themselves clearly state, "even
multivariate analysis cannot be expected to
eliminate the biasing effect of all confounding factors." In
particular, it may be that because the treatment was not randomly
assigned, practitioners would have chosen
thrombolysis in those patients with low risk of
bleeding (ie, younger age, less major surgery, less history of stroke).
As a result, thrombolytic treatment may not
represent the only independent predictor of outcome but an
"artificial" way in which patients with lower risk of mortality and
major complications have been selected. Finally, the criterion chosen
for identifying the thrombolytic group was the use of
thrombolytic treatment in the first 24 hours after PE
diagnosis. However, in 125 patients (23%), the attending physicians
proceeded to provide thrombolytic treatment later. On
the one hand, the time lag between diagnosis and treatment seems less
interesting than the exact determination, although often very difficult
to state precisely, of the interval between the onset of PE and the
start of the treatment. The latter should be the actual interval to be
considered. On the other hand, it would have been interesting to know
whether results concerning thrombolytic treatment would
have changed if this subgroup of patients thrombolized later were taken
into account. Did patients with a severe PE diagnosis assessed 36 hours
previously derive benefit from thrombolytic
treatment?
References
1.
Konstantinides S, Geibel A, Olschewski M, Heinrich
F, Grosser K, Rauber K, Iversen S, Redecker M, Kienast J, Just H,
Kasper W. Association between thrombolytic
treatment and the prognosis of hemodynamically stable
patients with major pulmonary embolism.
Circulation.. 1997;96:882–888.[Abstract/Free Full Text]
2.
Ferrari E, Baudouy M, Cerboni P, Tibi T, Guigner A,
Leonetti J, Bory M, Morand P. Clinical
epidemiology of venous thromboembolic disease:
results of a French multicentre registry. Eur Heart
J.. 1997;18:685–691.[Abstract/Free Full Text]
Response
Stavros Konstantinides, MD;
; Wolfgang Kasper, MD
Universitaetsklinik Freiberg,
Freiberg, Germany
It is by no means suggested in our
article1 that the clinical findings alone can be
sufficient for severity assessment or for the choice of treatment in
patients with pulmonary embolism. On the contrary, the
interactions of the pathophysiologic factors leading to
hemodynamic compromise after an acute thromboembolic
event are known to be very complex.2 3 In our
registry, the vast majority of the patients had
echocardiographic or, to a lesser extent, invasive
(catheter-derived) evidence of pulmonary hypertension and/or
right ventricular dysfunction. We have previously reported
on the importance of echocardiography for risk
stratification of patients with clinically suspected acute
pulmonary embolism.4 In fact, the results
of the present registry support the thesis that patients with
echocardiographic (or clinical) evidence of acute right
ventricular failure might benefit from
thrombolytic treatment even in the absence of
arterial hypotension at presentation. However,
we totally agree with Ferrari et al that these results should be
confirmed by a prospective, randomized trial. Selection bias and the
confounding effect of unrecognized prognostic parameters
cannot be excluded in the design of a registry, as has been clearly
stated in the "Limitations" section of our
study.1
We chose the time of diagnosis of pulmonary embolism to define
thrombolytic treatment as early or late. This is a time
point that can be most reliably and precisely identified in the
patient's records. On the other hand, patients with
pulmonary embolism are often unable to determine the exact time
of symptom onset because symptoms can develop gradually over a period
of hours or even days.
Finally, the assignment of patients to the two treatment arms
(thrombolysis or heparin alone) was based, as stated in
our article, on the intention-to-treat principle. Considering patients
who were given thrombolytic agents later during the
hospital stay would substantially increase the likelihood of selection
bias. The reason is that the indication for
thrombolysis in this patient group is usually mandated
by failure of the initially chosen heparin treatment, which is
manifested as clinical or hemodynamic
deterioration.
References
1.
Konstantinides S, Geibel A, Olschewski M, Heinrich
F, Grosser K, Rauber K, Iversen S, Redecker M, Kienast J, Just H,
Kasper W. Association between thrombolytic treatment
and the prognosis of hemodynamically stable patients
with major pulmonary embolism: results of a multicenter
registry. Circulation. 1997;96:882–888.
2.
Moser KM. Venous thromboembolism. Am Rev Respir Dis. 1990;131:235–249.
3.
Lualdi JC, Goldhaber SZ. Right ventricular
dysfunction after acute pulmonary embolism: pathophysiologic
factors, detection, and therapeutic implications. Am Heart
J. 1995;130:1276–1282.[Medline]
[Order article via Infotrieve]
4.
Kasper W, Konstantinides S, Geibel A, Tiede N, Krause T, Just
H. Prognostic significance of right ventricular afterload
stress detected by echocardiography in patients
with clinically suspected pulmonary embolism. Heart. 1997;77:346–349.[Abstract/Free Full Text]
