Anticoagulation and bleeding: increased risk if INR >4.5 or started in the last three months.

Clinical bottom line (level 1b)

  1. Major bleeds rarely occurred in patients on warfarin.
  2. Patients were at increased risk for a major bleed if anticoagulated if:
    • actual INR > or = 4.5 (NNF = 20 for unknown)
    • <90 days since commencing therapy (NNF = 67 for unknown)
    • more than 70 years old (NNF = 150 for unknown)
    • arterial disease requiring anticoagulation (NNF = 140 for unknown)
Palareti et al: Lancet 1996; 348: 423-428
Expires June 2003

The study

Inception cohort study with objective outcomes, adjusted for confounding factors, not validated in an independent set of patients.

Setting: 34 anticoagulation clinics, Italy

2745 patients (aged range 8-93 years; mean 62, 57% male) studied from start of their anticoagulation therapy. Most frequent indications for anticoagulation was venous thromboembolism, atrial fibrillation and heart valves

Excluded if
  • pregnant
  • geographically inaccessible



    • Factors studied:
  • major bleed
  • actual INR > or = 4.5
  • <90 days since commencing therapy
  • aged > or = 70
  • arterial disease requiring anticoagulation


    • Warfarin was used in 64% of patients, acenocoumarol in the rest. Target INR < or = 2.8 in 71%, and >2.8 in the remainder

    Multivariate regression analysis was performed to adjust for confounding factors.

    100% followed for mean 267 days
    Outcomes studied:
  • major bleed fatal bleed, intracranial haemorrhage, ocular bleed (resulting in blindness), articular, retroperitoneal bleed, surgical or angiographic intervention required, or Hb fall of >2 g/dl or requiring transfusion of >2 units

    • The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)
    major bleed mean 267 days 28/2745 1.0%
    (0.6% to 1.4%)

    prognostic factor for
    major bleed     		time to outcome adjusted RR
    (95% CI)    		 NNF+
    (95% CI)
    actual INR > or = 4.5 mean 267 days 5.96
    (3.68 to 9.67)     		20
    (12 to 37)
    <90 days since commencing therapy mean 267 days 2.50
    (1.40 to 3.30)     		67
    (43 to 250)
    aged > or = 70 mean 267 days 1.69
    (1.21 to 2.37)     		150
    (73 to 480)
    arterial disease requiring anticoagulation mean 267 days 1.72
    (1.17 to 2.54)     		140
    (65 to 590)

    • Sex, target INR range and type of coumarin used were not significantly associated with a major bleed.
    • In patients with bleeding events, the INR was above the range 10% of the time.
    • 20/28 (71%) of bleeding events occurred at low anticoagulation intensity. But in four of these, the low INR on the day of the event had been preceded by a value of 4.5 which may indicate that erratic anticoagulation may have been a cause of bleeding in these cases.

    Comments

    1. 30% of patients stopped anticoagulation therapy during the study- reasons are unclear.

    Citation

    1. Palareti G, Leali N, Cocchari S, et al: Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Lancet 1996; 348: 423-428
    Search Terms: atrial fibr* in Cochrane
    Contributor: Chris Ball and Clare Wotton, June 2000
    Reviewer:

    Clinical Question.
    Patient being anticoagulated
    Intervention or Exposure warfarin or acenocoumarol
    Outcome major bleeding