Reliability of Point-of-Care International Normalized Ratio Testing in an Academic Family Medicine Clinic

Abstract A recent increase in the number of point-of-care (POC) international normalized ratio (INR) measurements greater than 4 and the need for venous confirmatory testing in an academic family medicine clinic has led to the further exploration of the accuracy of POC INRs greater than 4. The primary objective of this quality improvement project was to compare the POC INR measurements with confirmatory venipuncture INR measurements. Seventeen patients representing 41 separate encounters over a 20-month period were retrospectively reviewed. The median POC INR was 4.5 (interquartile range, 4.1 to 5.1) versus a median venous INR of 3.4 (interquartile range, 3 to 3.9) of the 27 patient encounters where a venous INR was available. This difference between POC and venous INR resulted in a change in therapy for over 80% of the patients, which is a higher-than-reported discrepancy between POC INRs and venous INRs in patients with a POC INR greater than 4. Clinicians using the results of these POC devices who are unaware of the potential positive bias in POC INR values compared with venous draw values are at risk for making warfarin dose reductions, which can put their patients at risk for thromboembolic events.