| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Dr. Davidson is from the Pulmonary-Critical Care Medicine Section, Virginia Mason Medical Center, and is Clinical Associate Professor of Medicine, Division of Pulmonary and Critical Care Medicine, University of Washington School of Medicine; Dr. Lensing is from the Department of Vascular Medicine, Academic Medical Center, University of Amsterdam, and Clinical Development Department, NV Organon, Inc., Amsterdam, The Netherlands.
Correspondence to: Bruce L. Davidson, MD, MPH, FCCP, Pulmonary-Critical Care Medicine (C7-PUL), Virginia Mason Medical Center, 1100 Ninth Ave, Seattle, WA 98101; e-mail: bruce.davidson{at}vmmc.org
Readers who think they know the answer to this question should read the carefully analyzed patient series of Hamel and colleagues published in this issue of CHEST (see page 120), and then think again.
In recent years, champions of thrombolytic therapy for patients with pulmonary embolism have identified a patient subset without gross hemodynamic compromise (shock) but having subclinical, echocardiographically identifiable, right ventricular (RV) dysfunction.1 2 3 Several articles1 2 have concluded that thrombolytic therapy saves lives in this group. Another thoughtful analysis4 concluded that evidence for benefit of thrombolytic therapy in the absence of shock is lacking. Can we draw a conclusion about what might be the truth? It is useful to break the question into two questions: (1) should identifying RV dysfunction (recently termed "submassive" pulmonary embolism) in the absence of shock ("massive pulmonary embolism") make a difference in patient management, and (2) is thrombolytic therapy that difference, ie, is it a necessary-unless-contraindicated management strategy for these patients?
It would be reasonable to deduce that a poorly functioning right ventricle at the time acute pulmonary embolism is discovered portends a worse prognosis, whether the functional impairment is completely new, partially new, or chronic and predating the embolism. Pulmonary hypertension with RV dysfunction identifies a sicker patient whether the disease of concern is congestive heart failure,5 respiratory insufficiency waiting for lung transplant,6 or ARDS.7 Grifoni et al,3 in a recent cohort series report of patients with proven pulmonary embolism, found that 22% of the presenting cohort had hypotension; the death rate from pulmonary embolism in this subset was 19%. Hypotensive patients with pulmonary embolism comprise the population that would ordinarily be considered for thrombolytic therapy if important contraindications were absent. Thirty-one percent of all patients presenting with pulmonary embolism were normotensive but had occult (echocardiographically demonstrable) RV strain at presentation. The death rate attributed by the authors to pulmonary embolism in this group was 5% (95% confidence interval [CI], 0 to 13%). For normotensive patients without RV strain, the death rate from pulmonary embolism was 0% (95% CI, 0 to 4%). Three percent of the normotensive patients without strain died anyhow in the short-term (95% CI, 0 to 9%), with death attributed by the authors to non-pulmonary embolism causes. Because of the sample sizes, the 95% CIs for these death rates widely overlap.
In this era where physicians are reporting sending some normotensive pulmonary embolism patients home (without obtaining diagnostic echocardiography) receiving low-molecular-weight heparin treatment,8 one could argue that echocardiography might be important to identify patients with occult right RV strain (40% of the normotensive patients in the Grifoni study!), in order to manage them differently. Perhaps echocardiography, followed by more intensive management in the hospital for patients with RV strain, should be the standard of care, but we don’t know that because of conflicting data.
Time for the second question: should normotensive patients (with submassive embolism) with RV strain receive thrombolytic therapy if important contraindications are absent? A previous registry study2 of patients with pulmonary embolism found RV overload or pulmonary precapillary hypertension (by echocardiography or right-heart catheterization) in 85% of the patients included. This registry found a death rate of 10% for those with, but 4% for those without, RV overload or pulmonary hypertension. Some of the patients with RV dilation received thrombolysis (4.7% died); some received only heparin (11.1% died). Since the heparin recipients were acknowledged to be sicker, sorting out what this means highlights the trouble with registries as a research resource. Hamel et al found very different results. Matching similar patients with RV dilation, they found fewer deaths among those receiving unfractionated or low-molecular-weight heparin (0% [0 of 64 patients; 95% CI, 0 to 5.6%]) than among those receiving thrombolytic therapy (6.3% [4 of 64 patients; 95% CI, 1.7 to 15%]). In the population of Hamel et al, no heparin/low-molecular-weight heparin patients had significant bleeding, but three thrombolysis recipients had intracranial bleeds (two patients died of them, including a 58-year-old patient), and six bleeds from thrombolysis were severe. They concluded that thrombolysis recipients had no better survival and appeared to bleed more and die at higher rates.
What holes could one poke in the report by Hamel et al? The authors describe it also as a registry (from one center), but we find it a notch better, because it includes consecutive patients. Patients were excluded to aid matching for the comparison, but the authors’ Table 4 demonstrates that the excluded patients don’t appear to bias the results. Three different thrombolysis regimens were used, including saruplase (pro-urokinase), not approved in Europe or the United States. But the regimens used have not otherwise been identified to lead to excessive bleeding in published reports. These authors used an echocardiographic criterion unfamiliar to many North American cardiologists, a ratio of > 0.6 for RV to left ventricular end-diastolic diameter, for determining RV dysfunction. This differs from the criteria of Goldhaber et al,1 Konstantinides et al,2 and Grifoni et al,3 but Hamel et al defend their criterion choice. Who can say which criteria are best? For example, does RV free wall "hinge sign" but not other RV dysfunction criteria successfully distinguish between which pulmonary embolism patients do and do not require thrombolysis? Which investigators are prepared to spend resources testing that hypothesis? And how many different echocardiographic criteria for RV dysfunction would partisans want them to test before accepting a "no difference" conclusion?
We conclude that the current report by Hamel et al calls into question whether RV strain is a proper criterion for deciding to choose thrombolytic therapy in a normotensive patient. The bleeding and death rates that result from thrombolysis, compared to those for heparin treatment, are still the key reasons for caution.2 4 9 Some doctors are using echocardiographic evidence of RV dysfunction to decide whether to use thrombolysis in normotensive pulmonary embolism patients. If you believe Goldhaber et al1 and Konstantinides et al,2 many patients will benefit. If you believe Hamel et al, there will be unnecessary deaths and intracranial bleeds. We simply don’t know what the truth is. Since registry data are not conclusively answering this question, a randomized clinical trial is in order.
References
This article has been cited by other articles:
|
|
 |
|
  M. ten Wolde, M. Sohne, E. Quak, M. R. Mac Gillavry, and H. R. Buller Prognostic Value of Echocardiographically Assessed Right Ventricular Dysfunction in Patients With Pulmonary Embolism Arch Intern Med, August 9, 2004; 164(15): 1685 - 1689. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Dalen The Uncertain Role of Thrombolytic Therapy in the Treatment of Pulmonary Embolism Arch Intern Med, December 9, 2002; 162(22): 2521 - 2523. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
