Home treatment of DVT with low-molecular-weight heparin

At-home treatment of proximal DVT with low-molecular weight heparin

In previous studies, subcutaneous administration of low-molecular-weight heparin has been shown to compare favorably to intravenous heparin for the in-hospital treatment of proximal deep-vein thrombosis of the legs. Because of the ease of subcutaneous administration, and particularly because LMW heparin is administered at a fixed, weight-based dosage, without adjustment based on blood tests, it should be well suited to treatment of DVT at home, either as initial therapy or after a shortened hospital stay. These two studies, one from Canada and one from Europe, Australia and New Zealand, examine this approach.

A comparison of low-molecular-weight heparin administered primarily at home with unfractionated heparin administered in the hospital for proximal deep-vein thrombosis

Authors: Levine M, Gent M, Hirsh J, Leclerc J, Anderson D, et al.
Source: New England Journal of Medicine 334:677-81. March 14, 1996.
Institutions: multi-institutional, in Canada.
Financial support: none indicated.

Treament of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous low-molecular-weight heparin administered at home

Authors: Koopman M, Prandoni P, Piovella F, Ockelford P, Brandjes D, et al.
Source: New England Journal of Medicine 334:682-7. March 14, 1996.
Institutions: multi-institutional in Holland, Italy, Australia and New Zealand.
Financial support: Sanofi-Winthrop and Dutch Heart Foundation.

Summary / Comparison of the two trials

	Canadian Study	European / Australian Study
Patients screened	2230 patients with acute DVT (popliteal or more proximal) confirmed by venogram or duplex ultrasound. Patients had to be candidates for initial treatment at home, or for early discharge.	692 patients with acute DVT (popliteal or more proximal) confirmed by venogram or duplex ultrasound. Patients were not required to be candidates for home therapy or early discharge.
Exclusions	67% excluded: 32% - home treatment excluded because of coexisting illness (cancer, infection, stroke) 5% - home treatment excluded because of geographic distance 10% - concurrent symptomatic PE 6% - already treated with heparin for 48 hours or more 5% - two or more prior DVT's or PE's 13% - other reasons (bleeding disorder, coagulopathy, pregnancy)	31% excluded: 5% - DVT or PE within prior two years 5% - suspected PE 4% - already treated with heparin for 24 hours or more 5% - geographic inaccessibility, making follow-up impractical 12% - other reasons (short life expectancy, combinations of the above reasons)
Randomization	739 patients offered randomization. 239 declined (128 because of desire to be admitted to hospital). Randomized: 500 patients. Randomization stratified according to center, to mode of diagnosis (ultrasound or venography) and according to "category" (outpatients with initial home therapy; outpatients treated initially with heparin; patients hospitalized for other reasons with in-hospital diagnosis of DVT).	476 patients offered randomization. 76 declined. Randomized: 400 patients. Randomization stratified according to center.
Treatment regimens	Heparin: In-hospital therapy; 5000 unit bolus and infusion adjusted to keep PTT 60-85 secs. LMW heparin: 1 mg/kg enoxaparin administered subcutaneously, twice daily. Warfarin (both groups): started on day one, daily PT measurements, adjusted to maintain INR 2.0-3.0, heparin stopped when two subsequent days of target INR reached (but at least 5 days of heparin). At least three months of therapy with warfarin.	Heparin: In-hospital therapy; 5000 unit bolus and infusion adjusted to keep PTT 1.5-2.0 times control. LMW heparin: nadroparin, adjusted to patient's weight, administered subcutaneously twice daily. Warfarin (both groups): Same protocol as in Canadian study.
Outcome measures	Patients assessed monthly for three months after randomization, urgently if any symptoms of recurrent DVT or PE. Principal outcome events: recurrent DVT or PE within 90 days after randomization, or bleeding during administration of study drug.	Patients contacted daily during initial treatment, then at 4, 12 and 24 weeks. Principal outcome events: recurrent DVT or PE within 6 months after randomization, or bleeding within three months after randomization. Also measured: a quality of life index, and utilization of resources in the two groups.
Results
Treatment at home and hospital stay	LMW heparin group: 49% were never admitted to the hospital; mean hospital stay: 1.1 days. Standard heparin group: mean hospital stay: 6.5 days	LMW heparin group: 36% were never admitted to the hospital; mean hospital stay: 2.7 days for patients without "events". Standard heparin group: mean hospital stay: 8.1 days for patients without "events".
Recurrent venous thromboembolism	During the 90 day f/u period, there was no significant difference in the incidence of recurrent DVT / PE between LMW heparin and standard heparin groups (13 vs 17 patients).	Similar results to Canadian study: no significant difference between LMW heparin and standard heparin during 6-month f/u (14 vs. 17 patients).
Hemorrhagic complications	During the study drug period plus 48 hours, there was no significant difference in major hemorrhagic complications between LMW heparin and standard heparin (5 patients vs. 3 patients).	No significant difference between LMW and standard heparin (1 patient vs. 4 patients with major bleeding).
Mortality	During the 90 day follow-up period, there was no significant difference in mortality between LMW heparin and standard heparin (11 patients vs 17 patients).	No significant difference between LMW and standard heparin during 6 month follow-up period (14 patients vs. 16 patients died).

Comment

In previous studies, subcutaneous low-molecular-weight heparin has been shown to compare favorably with standard, unfractionated heparin for the in-hospital treatment of proximal deep-vein thrombosis. The two studies presented here demonstrate that, in selected patients, subcutaneous low-molecular-weight heparin can be administered at home without compromising patient safety. Thus hospitalization can be shortened or avoided altogether. Before we say "take two enoxaparin and call me in the morning", however, a few caveats:

Patients with pulmonary emboli were excluded from both trials. Safety of at-home treatment cannot be assumed for these patients based on the results presented here.
Patients at significant risk for bleeding were excluded from the larger, Canadian study. Although this was not a formal exclusion from the European/Australian study, only 36 percent of patients from that trial were treated exclusively at home and 25 percent were not even discharged early from the hospital.
Overall, only 5.4% of patients screened for the first trial and 10.4% of those screened for the second trial were actually treated entirely at home. Even doubling these figures (since patients were randomized) yields a relatively small proportion of screened patients treated totally at home.

Home therapy of proximal DVT appears to be safe in a very selected population. It is sure to benefit from support by the increasingly active medical cost-cutting movement. This approach needs to be applied with discrimination and common sense, however, not automatically or by guideline.

3/26/96

Reader comments

Date: Thu April 18, 1996
From: Tom Huddleston <tlhcbh@ilinkgn.net>

Informative article. Will LMWH someday replace unfractionated Heparin for most uses? I'm an orthopaedic surgeon and I use LMWH for DVT prophylaxis.

For reviews of the efficacy of LMWH in DVT prophylaxis (Kakkar) and in the treatment of established thromboembolic disease (Hirsh) see the July, 1995 issue of Thromb Haemost 74 (1). See also a review by Tapson and Hull in the June, 1995 Clin Chest Med, 16(2). Low molecular weight heparins seem to be better drugs in some settings and are easier to use, but are more expensive than unfractionated heparin. Of course, certain associated costs are lower with their use (blood drawing / PTT determination / IV equipment / possibly hospital stay). Seems like LMWH use (currently FDA approved for DVT prophylaxis in the US) will expand, but I would be interested in other readers' comments on their future role. -- mj

Date: Sat, 20 Apr 1996
From: rhogan@interserv.com

Ultimately I think that many of the uses of heparin which are primarily limited to inpatient settings will migrate to the outpatient arena.

Robert Hogan M.D.

Subject: LMWH
Date: Tue, 23 Apr 1996
From: "Albert J. Kirshen, MD, MSc, FRCPC" <albert.j.kirshen@utoronto.ca >
Baycrest Centre for Geriatric Care & Faculty of Medicine, Univ. of Toronto

LMWH appears to be effective in this group. I don't have the article in front of me, but were the elderly not excluded? What about the educational level of the participants? From a policy standpoint who would pay for this if the person isn't in hospital?

-- Albert

There were no specific age cut-offs in either study. In the Canadian study, the mean age in the LMWH group was 57 years (+/- 17). In the European study, 59 (+/- 17).

The educational level of the participants wasn't specified, but in the Canadian study "likelihood of non-compliance" was an exclusion criterium; in the European study, initial therapy at home was not mandated.

Exclusions and limitations in studies such as this one are an important issue. A given course of action is validated under favorable conditions and then the conclusions are extended to patients some of whom are quite different from the original study group. Particularly when a treatment or diagnostic approach saves money, there is the risk that it will be applied indiscriminately. This needs to be borne in mind, particularly by those involved in promulgating guidelines.

Who will pay? This will vary from country to country, but presumably at-home treatment will turn out to be cheaper, even if visiting nurse service is required. Whoever ultimately foots the bill for medical therapy would pay for at home treatment (similarly to home IV antibiotics for conditions requiring prolonged therapy). -- mj

Subject: jrnlclb/a17.html#c3
Date: Mon, 29 Apr 1996
From: "Albert J. Kirshen, MD, MSc, FRCPC" <albert.j.kirshen@utoronto.ca>
Organization: Baycrest Centre for Geriatric Care & Faculty of Medicine, Univ. of Toronto

The articles aren't generalizable to my population of frail, infirm elderly.

Agreed. Even if a study is perfectly valid, it won't be applicable to every patient. Determining whether valid study results are generalizable to a specific patient or group of patients is a key part of "evidence-based medicine" (see the links to other sites for more about EBM). -- mj

August 29, 1996

Letters to the editor about these articles from the New England Journal of Medicine's website.

Journal Club on the Web home page

Submit a comment about this article