Anticoagulation: starting warfarin within 48 hours of heparin shortened hospital stay.

Clinical bottom line (level 1b)

  1. Patients anticoagulated with heparin who started warfarin within 48 hours were less likely to have infusion phlebitis, than those who started warfarin after 96 hours (NNT = 6 at unknown) .
  2. Hospital stays were a mean of 4 days shorter for patients who started warfarin within 48 hours.
  3. The effect on major bleeds or recurrent venous thromboembolism was unclear.
Mohiuddin et al: American Heart Journal 1992; 123: 729-732
Expires June 2003

The study

Unblinded concealed randomised trial without intention-to-treat
Setting: university hospital, USA

119 patients (aged mean 66 years, 59% male) deep vein thrombosis, pulmonary embolism or mural thrombus following myocardial infarction and requiring anticoagulation with heparin and warfarin

Excluded if
  • diagnosis of thromboembolic disease could not be confirmed


    • Control Group: (n = 56, 56 analysed): warfarin after 96 hours of starting heparin (mean 110 ± 9 hours)
    Experimental Group: (n = 63, 63 analysed): warfarin within 48 hours of starting heparin (mean 30 ± 9 hours)
    All patients had 5000 IU heparin bolus followed by a constant infusion of 15 to 25 IU/kg/hr to maintain aPTT 1.5-2.0. Warfarin was started at 10 mg po daily and adjusted so PT 1.2-1.5. No minimum heparin-warfarin overlap was required.
    100% followed for 6 months

    The evidence

    Outcome Time to outcome CEREERRRR
    (95% CI)    		ARR
    (95% CI)    		NNT
    (95% CI)
    recurrent clinical thromboembolism or major bleeding 3 months 3
    (5.36%)    		 2
    (3.17%)    		 41%
    (-242% to 90%)    		 2.18%
    (-5.13% to 9.50%)    		 46
    (NNT = 11 to infinity;
    NNH = 19 to infinity)
    infusion phlebitis unknown 10
    (17.9%)    		 1
    (1.59%)    		 91%
    (33% to 99%)    		 16.3%
    (5.77% to 26.8%)    		 6
    (4 to 17)
    mortality 3 months 3
    (5.36%)    		 0
    (0.00%)    		 100%
    (% to %)    		 5.36%
    (-0.54% to 11.3%)    		 19
    (NNT = 9 to infinity;
    NNH = 185 to infinity)

    Outcome Control Group
    (SD)    		 Experimental Group
    (SD)    		 Mean Difference
    (95% CI)
    length of hospital stay 13.0
    (4.31)     		8.62
    (2.50)     		4.38
    (3.12 to 5.64)

    Citation

    1. Mohiuddin SM, Hilleman DE, Destache CJ, et al: Efficacy and safety of early versus late initiation of warfarin during heparin therapy in acute thromboembolism. American Heart Journal 1992; 123: 729-732
    Contributor: Chris Ball and Clare Wotton, June 2000
    Reviewer:

    Clinical Question.
    Patient DVT, PE and requiring anticoagulation
    Intervention or Exposure warfarin within 48 hours of heparin
    Comparison warfarin within 96 hours of heparin
    Outcome recurrence