ACCURACY OF A PORTABLE INTERNATIONAL NORMALIZATION RATIO MONITOR IN OUTPATIENTS RECEIVING LONG-TERM ORAL ANTICOAGULANT-THERAPY - COMPARISONWITH A LABORATORY REFERENCE-STANDARD USING CLINICALLY RELEVANT CRITERIA FOR AGREEMENT

The accuracy of a new, portable INR monitor (CoaguChek, Boehringer-Man nheim, Indianapolis, IN) was evaluated by comparing INR results from t he portable monitor with results obtained by a laboratory-based method . Dual INR measurements (portable monitor, laboratory) were performed in 163 consecutive outpatients receiving warfarin. Agreement in dual I NR measurements was defined based on clinically-relevant expanded and narrow criteria and statistical criteria. Agreement in dual INR measur ements also was evaluated as a function of increasing INR. The proport ion of dual INR measurements that satisfied the clinically-relevant ex panded, and narrow agree ment criteria was 90%, and 86%, respectively. Seventy-nine percent of all dual measurements were within 0.5 INR uni ts. The accuracy of the portable monitor was greatest for INR values l ess than 3.0; above this INR level, the portable monitor underestimate d laboratory INR values. The proportion of dual INR measurements withi n 0.5 INR units for laboratory INR ranges of <2.0, 2.0-3.0, 3.1-4.0, a nd >4.0 was 98%, 87%, 57%, and 21%, respectively. We conclude that the portable INR monitor achieved a clinically acceptable level of accura cy when compared to the traditional laboratory method and provides a s uitable alternative method of monitoring the INR in patients receiving warfarin. (C) 1998 Elsevier Science Ltd.