Prevention of deep vein thrombosis after hip replacement: randomised comparisons between unfractionated heparin and low molecular weight heparin.

Introduction

Patients who undergo total hip replacement are at high risk of developing postoperative venous thrombo-embolic disease. [1] Without prophylaxis deep vein thrombosis and fatal pulmonary embolism occur in 40-70% and 1-5% of patients respectively. [2-4] Moreover, it has been repeatedly shown that proximal and bilateral thrombi are much more common after total hip replacement than after general surgery. [5-10] This high risk has led to the widespread application of prophylactic measures in hip surgery, but considerable controversy exists about the optimal prophylactic approach. The four main considerations are: firstly, the efficacy of prophylaxis in reducing the incidence of venous thrombosis; secondly, but equally importantly, the bleeding enhancing potential of several prophylactic methods, which threatens the final outcome of the operation; thirdly, the incidence of non-haemorrhagic complications; and, finally, the complexity of giving some prophylactic methods.

Subcutaneous fixed doses of unfractionated heparin reduce by about a half the risk of phlebographically proved deep vein thrombosis after hip replacement without causing a clinically relevant increase in blood loss, [11] but a considerable risk of deep vein thrombosis of 20-35% remains. [12-15] The addition of dihydroergotamine mesylate to unfractionated heparin improves the antithrombotic protection but carries a potential risk of ischaemic disease. [16-20] Studies in which the doses of subcutaneous heparin were adjusted according to the partial thromboplastin time brought about a significant reduction in the rate of deep vein thrombosis to 10-16% without an increase in haemorrhagic complications. [20-23] Frequent adjustment of the heparin dose, however, is very labour intensive. [24]

Several studies using contrast phlebography for diagnosing deep vein thrombosis showed that subcutaneous low molecular weight heparins or heparinoids were significantly more efficacious in preventing thromboembolic complications than placebo and at least as efficacious or more efficacious than fixed doses of unfractionated heparin. The observed incidences of deep vein thrombosis varied between 10% and 19% and major haemorrhagic complications were rarely encountered. [25-29]

Interestingly, low molecular weight heparins seem to have a particularly favourable effect in preventing proximal deep vein thrombosis. The incidences of proximal deep vein thrombosis after hip replacement in studies using low molecular weight heparins and systematically phlebography in all patients were 4%, [25] 7.5%, [26] 6.5%, [29] and 6.5% [30] with enoxaparin; 2.4% [23] and 0% [28] with Fragmin; 3.6% [31] with Fraxiparine; and 8% [27] and 4.8% [32] with Lomoparan.

These favourable results, as well as the simple mode of administration, prompted us to compare treatment with a low molecular weight heparin and treatment by the efficacious but demanding "titrated dose heparin" method. [33] As lethal pulmonary embolism is thought to be due to migration of proximal thrombi, [2 34] the study design took into account primarily the expected incidences of proximal deep vein thrombosis.

Patients and methods

From September 1988 to May 1989, 1109 patients were admitted for elective total hip replacement to the 28 participating European departments of orthopaedic surgery. The following eligibility criteria were applied: body weight between 45 kg and 100 kg, age over 40 years, normal results on Doppler examination of the lower extremities, and having the operation under general anaesthesia.

[TABULAR DATA OMITTED]

In all, 700 patients (63%) were excluded before randomisation for reasons specified a priori in the protocol (table I). Hence 409 patients were randomised, by using a balanced assignment for each centre (by the sealed envelope method), to either titrated doses of unfractionated heparin thrice daily or how molecular weight heparin once daily. The study protocol was approved by the ethical committees of all the participating centres.

COMPOUNDS AND ASSAYS

A single batch of the low molecular weight heparin Fraxiparine (CY 216, Sanofi-Choay Laboratory, Paris) was used. It had an anti-factor specific activity of 27 IIa units (tested against the fourth international heparin standard) and an anti-factor Xa activity of 89 IU/mg when copmared with the international standard for low molecular weight heparins. [35] It was supplied as a concentrated solution containing 25 000 Institute Choay units (ICU)/ml, corresponding to 10 400 IU/ml and supplied in phials containing 0.6 ml. [36] The unfractionated heparin was a sodium salt of 10 000 IU/ml of heparin (Sanofi-Choay) supplied in ampoules containing 1 ml.

Blood was collected four hours after the morning injection into a Diatube H (Diagnostica Stago, Asnieres, France) containing 0-1 volume of citric acid (0.11 mol/1), theophylline (15 mmol/1), adenosine (3.7 mmol/1), and diphyridamole (0.198 mmol/1 mixture. [37] Activated thromboplastin times were compared with values for each laboratory's control plasma by using the general diagnostics automated activated partial thromboplastin time reagent (Organon Teknika, Fresnes, France). The control plasma was calibrated weekly with general diagnostics plasma verify H (Organon Teknika). The anti-factor Xa activity of the low molecular weight heparin was assayed by a …