Atrial fibrillation: cardioversion and aspirin, followed by warfarin if AF recurred, was most cost-effective.

Clinical bottom line (level 2b)

  1. The most cost-effective strategies for treating patients with non-valvular atrial fibrillation were (in increasing cost and increasing effectiveness):
    • cardioversion and aspirin, swapping to warfarin if AF recurs
    • cardioversion, amiodarone and aspirin, swapping to warfarin if AF recurs
    • cardioversion, amiodarone and warfarin
Eckman et al: Archives of Internal Medicine 1998; 158: 1669-1677
Expires November 2003

The study

cost-effectiveness analysis based on a Markov decision analysis
Setting:

data from searching MEDline and bibliographies from selected articles, comparing antithrombotic and antiarrhythmic therapy for non-valvular atrial fibrillation. Strategies studied included cardioversion, aspirin or warfarin, sotalol, quinidine or amiodarone.

  • Viewpoint: third party e.g. HMO or NHS purchaser
  • Benefit assessment: death, long-term morbidity, resolution
  • Resources and costs: from US physician-fee references, a medical centre's cost accounting system and the in-patient costs (1995 $) for haemorrhagic or embolic events and ambulatory costs for routine follow-up at a major New England HMO. A discount rate of 3% per year was used.
  • Sensitivity analysis: : varied severity of symptoms, mortality rates and adverse effects with antiarrhythmics

    • The evidence

    intervention cost
    cardioversion plus aspirin 13800
    ( QALY 9.60 )
    aspirin 14000
    ( QALY 9.21 )
    cardioversion, aspirin; if AF recurs, warfarin 15900
    ( QALY 9.79. cost-effectiveness ratio$ per additional QALY 10800 )
    cardioversion, if AF recurs warfarin 16200
    ( QALY 9.69 )
    warfarin 17200
    ( QALY 9.51 )
    cardioversion and warfarin 17300
    ( QALY 9.82 )
    cardioversion, amiodarone and aspirin 20900
    ( QALY 9.86 )
    cardioversion, amiodarone, aspirin; if AF recurs, warfarin 22100
    ( QALY 9.98. cost-effectiveness ratio$ per additional QALY 33700 )
    cardioversion, amiodarone, if AF recurs, warfarin 22600
    ( QALY 9.79 )
    cardioversion, amiodarone , warfarin 24700
    ( QALY 10.0. cost-effectiveness ratio$ per additional QALY 92400 )

    Effect of sensitivity analysis: Options involving quinidine or sotalol were more expensive and less effective. If few or no AF symptoms: cardioversion followed by aspirin or warfarin alone is best. If high-risk of amiodarone pulmonary toxicity or adverse effects: cardioversion followed by warfarin is best

    Comments

    1. Patients failing to respond to one antiarrhythmic were assumed not to start another.
    2. Few details were given about the selection or appraisal of the articles used.

    Citation

    1. Eckman MH, Falk RH, Pauker SG: Cost-effectiveness of therapies for patients with nonvalvular atrial fibrillation. Archives of Internal Medicine 1998; 158: 1669-1677
    Contributor: Chris Ball and Clare Wotton, November 2000
    Reviewer:

    Clinical Question.
    Patient atrial fibrillation
    Intervention or Exposure cardioversion and drugs
    Outcome cost-effectiveness