Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Subcutaneous heparin is more efficacious and at least as safe as continuous heparin infusion in deep-venous thrombosis

ACP J Club. 1992 May-June;116:76. doi:10.7326/ACPJC-1992-116-3-076

Source Citation

Hommes DW, Bura A, Mazzolai L, B├╝ller HR, ten Cate JW. Subcutaneous heparin compared with continuous intravenous heparin administration in the initial treatment of deep vein thrombosis. A meta-analysis. Ann Intern Med. 1992 Feb 15;116:279-84.



To evaluate the efficacy (prevention of extension and recurrence of venous thromboembolism) and safety (lack of major hemorrhage) of subcutaneous heparin administration compared with continuous intravenous heparin infusion, using meta-analysis.

Data sources

Citations were retrieved from MEDLINE for the years 1966 to 1991, the Science Citation Index (1989 to February 1991), and Current Contents (January through April 1991). Bibliographies and proceedings and abstract books of scientific meetings were also scanned to obtain additional references.

Study selection

Potential studies for inclusion were clinical trials directly comparing full-dose subcutaneous heparin administration with continuous intravenous heparin infusion in patients with venous thromboembolism confirmed by venography, plethysmography, pulmonary arteriography, or perfusion ventilation scanning. 8 studies met the criteria.

Data extraction

The studies were independently rated by 3 investigators for description of the location of thrombosis; the heparin dose; the primary methods for measuring outcomes; and the selection, description, and follow-up of patients.

Main results

6 of the 8 studies (comprising 81% of the patients) had adequate methods to be included in the meta-analysis. All were randomized trials. 1 trial limited entry to patients with calf-vein thrombosis, whereas in the other 5 studies 62% to 100% of patients had thrombosis extending into the proximal deep veins. The daily heparin dose varied from 24 384 to 37 000 international units. The methods used for measuring the efficacy of treatment were venography in 3 studies, ventilation-perfusion lung scanning in 1 trial, and confirmation of clinical events by either method in 2 studies. Fewer patients (25 of 388) in the subcutaneous heparin group had an extention or recurrence of the venous thromboembolism compared with those in the intravenous heparin group (40 of 382) {P = 0.05, relative risk reduction (RRR) 38%, 95% CI 0% to 62%}*. Major hemorrhage occurred slightly less often in the patients treated subcutaneously than in those receiving heparin intravenously {RRR 21%, CI -48% to 58%}*. The test for homogeneity was significant for efficacy (P < 0.001) but not for safety (P > 0.6).


Available data suggest that subcutaneous heparin infusion is more efficacious and at least as safe as continuous heparin infusion in the initial treatment of deep-vein thrombosis.

Sources of funding: The Educational Erasmus Program and the Royal Netherlands Academy of Arts and Sciences.

Address for article reprint: Dr. D.W. Hommes, Academic Medical Center, F4-139, Meibergdreef 9, 1105 AZ, Amsterdam (ZO), The Netherlands.

*Numbers calculated from data in article.


This review addresses the question of whether an alternative mode of heparin delivery, intermittent full-dose subcutaneous heparin (i.e., heparin given subcutaneously every 12 hours resulting in inhibition of clotting to a degree similar to continuous intravenous heparin infusion), is different in efficacy and safety from the standard continuous intravenous heparin infusion. Potential advantages of intermittent full-dose subcutaneous heparin include dispensing with the intravenous line and pump, as well as providing initial treatment of deep vein thrombosis in nonhospital settings such as long-term care facilities or even the home.

The authors identified 6 well-designed randomized trials (3 finding no significant differences, 2 favoring the subcutaneous route, 1 favoring the intravenous route). In order to obtain greater statistical power for finding a difference between the 2 treatment regimens, the authors combined the results from the 6 studies and concluded that intermittent subcutaneous heparin is significantly more effective than continuous intravenous heparin. These studies differed in many important respects, however, including extent of deep-vein thrombosis (e.g., proximal vs calf); primary method used for outcome measurement (e.g., extension of clot by venography vs "symptomatic pulmonary embolism" confirmed by lung scan); method of following the effect of heparin (e.g., activated partial thromboplastin time vs kaolin cephalin clotting time); timing of these tests (4, 6, or 12 hours after the heparin dose); and results. The test for homogeneity of the efficacy results found significant heterogeneity, which means the individual studies may not be comparable. Because of these differences, the authors may have overstepped the permissible bounds of meta-analysis by combining "apples and oranges."

A conservative conclusion is that the 2 treatments are similar in efficacy. Either is acceptable therapy in the hospital setting. However, before intermittent full-dose subcutaneous heparin is routinely used in another setting, further trials should be done to validate efficacy and safety.

John T. Philbrick, MD
University of VirginiaCharlottesville, Virginia, USA