Venous thromboembolism: prophylaxis: major abdominal surgery: LMWH caused fewer wound haematomas.

Clinical bottom line (level 1b)

  1. In general surgical patients, low-molecular-weight heparin was not clearly better than unfractionated heparin for preventing venous thromboembolism or mortality.
  2. Low-molecular-weight heparin caused fewer wound haematomas (NNT = 78 at 8 weeks) , but had no clear effect on major bleeds.
Kakkar et al: Lancet 1993; 341: 259-265
Expires December 2003

The study

Unblinded concealed randomised trial with intention-to-treat
Setting: nineteen hospitals, UK

3809 patients (aged ?, ?% male) undergoing major gastrointestinal, gynaecological or urological surgery (defined as more than 30 minutes under general anaesthetic, and requiring at least six days in hospital)

Excluded if
  • <40 years old
  • known heparin allergy
  • on oral anticoagulants pre-admission
  • severe haemorrhagic episode in previous three months unrelated to proposed surgery
  • known bleeding diathesis (including jaundiced patients with a prolonged prothrombin time)
  • scheduled for re-operation during the study period
  • women of child-bearing age not actively avoiding pregnancy
  • any other heparin contraindication


    • Control Group: (n = 1915, 1915 analysed): 5000 units heparin subcutaneously given 1-4 hours pre-op, ten for at least five days post-op or until patient is fully mobile
    Experimental Group: (n = 1894, 1894 analysed): fragmin 2500 IU subcutaneously twice daily given 1-4 hours pre-op, then for at least five days post-op or until patient fully mobile

    100% followed for 8 weeks
    Outcome notes:
    • deep vein thrombosis : diagnosed clinically- venogram or ventilation-perfusion scan and/or angiogram respectively
    • pulmonary embolism : diagnosed clinically- as DVT
    • major bleed : blood loss during perioperative period requiring discontinuation of prophylaxis, bleeding attributable to trial drug, bleed requiring re-operation, wound haematoma whether drained or not

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    deep vein thrombosis 8 weeks 11
    (0.57%)    		 11
    (0.58%)    		 -1%
    (-130% to 56%)    		 -0.01%
    (-0.49% to 0.47%)    		 -16000
    (NNT = 210 to infinity;
    NNH = 210 to infinity)
    pulmonary embolism 8 weeks 14
    (0.73%)    		 13
    (0.69%)    		 6%
    (-99% to 56%)    		 0.04%
    (-0.49% to 0.58%)    		 2200
    (NNT = 170 to infinity;
    NNH = 210 to infinity)
    wound haematoma 8 weeks 52
    (2.72%)    		 27
    (1.43%)    		 48%
    (17% to 67%)    		 1.29%
    (0.39% to 2.19%)    		 78
    (46 to 260)
    major bleed 8 weeks 91
    (4.75%)    		 69
    (3.64%)    		 23%
    (-4% to 44%)    		 1.11%
    (-0.16% to 2.38%)    		 90
    (NNT = 42 to infinity;
    NNH = 610 to infinity)
    perioperative death 8 weeks 47
    (2.45%)    		 63
    (3.33%)    		 -36%
    (-97% to 7%)    		 -0.87%
    (-1.94% to 0.19%)    		 -115
    (NNT = 520 to infinity;
    NNH = 52 to infinity)
    post-operative death 8 weeks 9
    (0.47%)    		 10
    (0.53%)    		 -12%
    (-176% to 54%)    		 -0.06%
    (-0.51% to 0.39%)    		 -1720
    (NNT = 260 to infinity;
    NNH = 200 to infinity)

    Comments

    1. This study only addressed symptomatic venous thromboembolism unlike many which fail to make any distinction between asymptomatic and symptomatic VTE.

    Citation

    1. Kakkar VV, Cohen AT, Edmondson RA, et al: Low molecular weight versus standard heparin for prevention of venous thromboembolism after major abdominal surgery. Lancet 1993; 341: 259-265
    Contributor: Chris Ball and Clare Wotton, June 2000
    Reviewer:

    Clinical Question.
    Patient undergoing major surgery
    Intervention or Exposure LMWH
    Comparison unfractionated heparin
    Outcome DVT, PE