Correspondence
Pulmonary Embolism
N Engl J Med 1998; 339:1555-1557November 19, 1998
- Article
To the Editor:
In his article on pulmonary embolism, Dr. Goldhaber (July 9 issue)1 states that unfractionated heparin “constitutes the cornerstone” of treatment, and he mentions fractionated, or low-molecular-weight, heparin only in the discussion of prevention. Nevertheless, there have been two randomized trials in patients with pulmonary embolism that show that low-molecular-weight heparin is as effective as unfractionated heparin in regard to the incidence of recurrence, major bleeding complications, and death. In one study, which involved 612 patients with symptomatic pulmonary embolism who did not require thrombolytic therapy or embolectomy, subcutaneous low-molecular-weight heparin (tinzaparin) given once daily in a fixed dose was compared with adjusted-dose, intravenous unfractionated heparin. There were no significant differences between the two groups with respect to a combined end point of recurrent thromboembolism, major bleeding, and death at day 8 (2.9 percent of patients in the group receiving unfractionated heparin and 3.0 percent in the group receiving low-molecular-weight heparin) or at day 90 (7.1 percent and 5.9 percent, respectively).2
In another randomized trial, 1021 patients with symptomatic venous thromboembolism were treated with fixed-dose, subcutaneous low-molecular-weight heparin (reviparin sodium) or adjusted-dose, intravenous unfractionated heparin. Approximately one third of the patients had associated pulmonary embolism. The incidence of outcome events over the subsequent 12 weeks was similar in the two groups: recurrent thromboembolic events, 4.9 percent in the group receiving unfractionated heparin and 5.3 percent in the group receiving low-molecular-weight heparin; major bleeding, 2.3 percent and 3.1 percent, respectively; and death, 7.6 percent and 7.1 percent, respectively.3 Therefore, initial therapy with subcutaneous low-molecular-weight heparin appears to be as effective and safe as intravenous unfractionated heparin in patients with acute pulmonary embolism.
3 ReferencesDirk Schrijvers, M.D.
Jan Van den Brande, M.D.
Jan B. Vermorken, M.D., Ph.D.
University Hospital Antwerp, B-2650 Edegem, Belgium1
Goldhaber SZ . Pulmonary embolism. N Engl J Med 1998;339:93-104
Full Text | Web of Science | Medline2
Simonneau G ,Sors H ,Charbonnier B , et al. A comparison of low-molecular-weight heparin with unfractionated heparin for acute pulmonary embolism. N Engl J Med 1997;337:663-669
Full Text | Web of Science | Medline3
The Columbus Investigators
Full Text | Web of Science | Medline
To the Editor:
Citing the trial by Decousus et al.,1 Goldhaber states, “Vena caval filters plus anticoagulation did not reduce the two-year mortality rate, as compared with anticoagulation alone.” This statement, however, may underestimate the role of vena caval filters in the prevention of early and, possibly, late mortality from pulmonary embolism.
The study by Decousus et al. involved 400 patients with proximal deep-vein thrombosis who were at high risk for pulmonary embolism.1 All the patients received anticoagulant therapy for at least three months, and half of them were also randomly assigned to receive vena caval filters. The primary outcome event was the occurrence of pulmonary embolism within the first 12 days after randomization. Secondary outcome events included pulmonary embolism, recurrent deep-vein thrombosis, death, major bleeding, and filter-associated complications. The study demonstrated that vena caval filters reduced the incidence of pulmonary embolism within the first 12 days after the diagnosis of deep-vein thrombosis; it also demonstrated an excess of recurrences of deep-vein thrombosis among patients treated with vena caval filters. Moreover, although the overall mortality rate during the first 12 days was the same in the two groups (five patients died in each group), vena caval filters prevented early death from pulmonary embolism, since four of the five patients in the no-filter group but none of the patients in the filter group died of pulmonary embolism (P=0.02 by Fisher's exact test). For the entire two-year period, however, the data presented do not allow a comparison of the two groups with respect to the incidence of death directly related to thromboembolism.
1 ReferencesUlisse Corbanese, M.D.
Clemente Possamai, M.D.
Ospedale S. Maria dei Battuti, I-31015 Conegliano, Italy1
Decousus H ,Leizorovicz A ,Parent F , et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. N Engl J Med 1998;338:409-415
Full Text | Web of Science | Medline
To the Editor:
Several issues raised in Goldhaber's excellent overview of pulmonary embolism are worthy of comment. The global recommendation of a “routine” evaluation of hypercoagulability is insufficient. Specifically, three points should be considered. First, evaluation for antithrombin III, protein C, and protein S is indicated in younger patients with pulmonary embolism — especially if there is a familial predisposition to thrombophilia. Second, the antiphospholipid-antibody syndrome is commonly associated with venous thromboembolic disease. A substantial portion of patients with this syndrome do not have a lupus anticoagulant but do have anticardiolipin antibody. Thus, tests for both these proteins should be performed. Third, the prothrombin 20210A allele mutation is present in 1 percent of whites and is clearly a predisposing cause of venous thromboembolic disease. Evaluation for this mutation1 should be a routine component of the evaluation of hypercoagulability in many patients.
Dr. Goldhaber recommends that oral anticoagulant therapy not be used to treat pulmonary embolism in pregnant women and that instead heparin should be used throughout pregnancy. This general recommendation is clearly accepted by most clinicians in view of the known teratogenic effects of warfarin — especially when it is used in the first trimester of pregnancy. However, it is important to note that patients with prosthetic valves — especially mechanical valves — often do poorly when they receive heparin for several months during pregnancy. In this important subgroup of patients who have a pulmonary embolism during pregnancy, consideration of warfarin therapy instead of heparin therapy is justified.2,3
3 ReferencesPhilip M. Blatt, M.D.
Christiana Care Health System, Newark, DE 197131
Poort SR ,Rosendaal FR ,Reitsma PH ,Bertina RM . A common genetic variation in the 3'-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88:3698-3703
Web of Science | Medline2
Tiede DJ ,Nishimura RA ,Gastineau DA ,Mullany CJ ,Orszulak TA ,Schaff HV . Modern management of prosthetic valve anticoagulation. Mayo Clin Proc 1998;73:665-680
CrossRef | Web of Science | Medline3
Oakley CM . Anticoagulation and pregnancy. Eur Heart J 1995;16:1317-1319
Web of Science | Medline
To the Editor:
We would like to clarify the issue of the effect of unfractionated heparin on the prothrombin time and the international normalized ratio (INR).
In his review, Dr. Goldhaber states, “the initial target . . . INR should be 3.0, because concomitant administration of unfractionated heparin usually prolongs the INR by an additional 0.5, thus yielding an effective INR due to warfarin alone of 2.5.” It should be noted, however, that many newer prothrombin-time reagents make the prothrombin time insensitive to heparin up to concentrations greater than those recommended for unfractionated-heparin therapy in patients with venous thromboembolism (0.2 to 0.4 U per milliliter by protamine sulfate titration or 0.35 to 0.7 U per milliliter by the anti-Xa chromogenic assay).1 Table 1Table 1
Absence of Effect of Unfractionated Heparin on the Prothrombin Time with Two Reagents Containing a Heparin Neutralizer. shows the prothrombin time in patients receiving heparin therapy, determined with the use of two human recombinant prothrombin-time reagents available in the United States. INR results were not affected by heparin with either of these reagents.On the basis of a 1998 College of American Pathology Proficiency Testing survey, these two reagents constituted at least 22 percent of the commercial prothrombin-time reagents used in the United States. These and several other reagents are manufactured to be insensitive to heparin at concentrations of up to 0.7 to 2.0 U per milliliter by the admixture of polybrene or other additives that neutralize heparin.
Clinical laboratories should provide clinicians with information about whether the prothrombin-time reagent used is sensitive to heparin and should specify the concentration of heparin above which the reagent is sensitive. This information can be obtained by performing prothrombin-time assays on normal pooled plasma to which unfractionated heparin has been added to yield concentrations of 0.25 to 1.0 U per milliliter and determining the concentration of heparin at which the prothrombin time is longer than the upper limit of the normal reference interval. Preferably, laboratories should use reagents that make the prothrombin time insensitive to therapeutic heparin concentrations.
Clinicians should know the sensitivity to heparin of the prothrombin-time reagents being used by the laboratories providing services for their patients. If the prothrombin time is insensitive to unfractionated heparin, the clinician should accept the INR value without correction. If the prothrombin time is sensitive to unfractionated heparin, the clinician may want an INR that is 0.5 unit above the target value, as Dr. Goldhaber states, although this is highly arbitrary and is not supported by appropriate safety and efficacy studies. If the patient is receiving any of the low-molecular-weight heparins, the therapeutic concentrations generally do not prolong the activated partial-thromboplastin time or prothrombin time, and the desired INR should be the target value.
1 ReferencesRichard A. Marlar, Ph.D.
Denver Veterans Affairs Medical Center, Denver, CO 80220J. Heinrich Joist, M.D., Ph.D.
St. Louis University School of Medicine, St. Louis, MO 63110Louis M. Fink, M.D.
John L. McClellan Veterans Affairs Medical Center, Little Rock, AR 722051
Hirsh J ,Raschke R ,Warkentin TE ,Dalen JE ,Deykin D ,Poller L . Heparin: mechanism of action, pharmacokinetics, dosing considerations, monitoring, efficacy, and safety. Chest 1995;108:Suppl:258S-275S
CrossRef | Web of Science | Medline
Author/Editor Response
Dr. Goldhaber replies:
To the Editor: Dr. Schrijvers and colleagues may have overlooked two citations (references 81 and 82) in my article. The important point is that despite the use of low-molecular-weight heparin for pulmonary embolism, all patients receiving this therapy remained hospitalized. No patients were assigned to an early-discharge or outpatient-management strategy.
Filter placement implicitly suggests a reduction of venous thromboembolism and an improvement in survival. Neither goal was achieved in the carefully executed randomized, controlled trial by the Prévention du Risque d'Embolie Pulmonaire par Interruption Cave investigators.1 I ask Drs. Corbanese and Possamai to consider how often filter placement provides a net clinical benefit instead of substituting one type of death for another. In the Brigham and Women's Hospital registry of 150 consecutive patients with deep-vein thrombosis, my colleagues and I found, much to our surprise, that 30 percent of those who received filters died within three months.2
As recently as a decade ago, it was common to order as many tests of hypercoagulability as possible in patients with newly diagnosed pulmonary embolism. Now we recognize the limitations of our resources and must choose tests parsimoniously and according to their effect on management. Dr. Blatt should bear in mind that even the triad of tests that I usually obtain has not been validated as a means of reducing morbidity or recurrences of pulmonary embolism.
I routinely prescribe twice-daily injections of low-molecular-weight heparin for pregnant women with prosthetic heart valves. I find this regimen is tolerated well and far more readily than thrice-daily injections of unfractionated heparin. Among pregnant women who require anticoagulant therapy, I always use heparin, not warfarin.
With regard to the letter by Dr. Marlar and colleagues, I should point out that warfarin often prolongs the activated partial-thromboplastin time. In a study by investigators at McMaster University, the activated partial-thromboplastin time increased by an average of 16 seconds for each increase of 1.0 in the INR.3 These investigators also found that heparin prolonged the INR. It appears that the relation between the activated partial-thromboplastin time and the INR is mediated predominantly by antagonism of vitamin K. Unfortunately, only in a minority of hospitals are prothrombin-time reagents selected that are insensitive to heparin.
3 ReferencesSamuel Z. Goldhaber, M.D.
Brigham and Women's Hospital, Boston, MA 021151
Decousus H ,Leizorovicz A ,Parent F , et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. N Engl J Med 1998;338:409-415
Full Text | Web of Science | Medline2
Piccioli A ,Prandoni P ,Goldhaber SZ . Epidemiologic characteristics, management, and outcome of deep venous thrombosis in a tertiary-care hospital: the Brigham and Women's Hospital DVT registry. Am Heart J 1996;132:1010-1014
CrossRef | Web of Science | Medline3
Kearon C ,Johnston M ,Moffat K ,McGinnis J ,Ginsberg JS . Effect of warfarin on activated partial thromboplastin time in patients receiving heparin. Arch Intern Med 1998;158:1140-1143
CrossRef | Web of Science | Medline
