Desbiens N A
Department of Medicine, Marshfield Clinic, WI 54449.
J Gen Intern Med. 1992 Mar-Apr;7(2):131-9. doi: 10.1007/BF02598001.
To address a patient care problem suggested by a quality assurance study: Should physicians of patients with atrial fibrillation who have not had echocardiography performed be encouraged to obtain echocardiograms in order to discover heretofore undetected mitral stenosis, which could be treated with anticoagulation, in order to avert strokes?
Decision analysis using a Markov-chain technique and quality-adjusted life expectancy.
Outpatient clinics.
Patients aged 40-80 years with chronic atrial fibrillation.
The following strategies were studied: 1) obtain echocardiograms for all patients and anticoagulate those with mitral stenosis; 2) anticoagulate patients with classic auscultatory sounds of mitral stenosis and obtain echocardiograms for patients with other murmurs; if an echocardiogram reveals mitral stenosis, anticoagulate; 3) anticoagulate all patients with murmurs; 4) anticoagulate all patients; 5) observe all patients (natural history).
The base analysis, using data available in the literature and on site, indicated that though performing echocardiography on all patients with atrial fibrillation gave the best quality-adjusted life expectancy, the differences among the five strategies studied were small. In addition, sensitivity analysis revealed that the decision was sensitive to many of the variables used in the analysis, especially the efficacy and the risks of anticoagulation.
A recommendation that cardiac ultrasonography be done for all patients with atrial fibrillation in order to detect clinically unrecognized mitral stenosis cannot be made.