Whole blood international normalization ratio measurements in children using near-patient monitors

Purpose: To report a comparison of international normalization ratio (INR) measurements on four near-patient (point-of-care or bedside) whole blood INR monitors in children. Patients and Methods: The INR results from 19 ambulatory pediatric subjects (30 hospital visits) receiving warfarin sodium were analyzed on four near-patient monitors and compared with plasma INR measurements on the laboratory CA-1000 Analyze. The instruments evaluated were CoaguChek, Hemochron Jr. Signature, ProTime Microcoagulation System, and RapidpointCoag. Results: The INR measurements ranged from 1.05 to 5.25. Over the entire INR range, the near-patient instrument with the least bias relative to the CA-1000 was the RapidpointCoag (r(2) = 0.923). The correlations (r(2)) of the CoaguCheck, Hemochron Jr., and ProTime were 0.877, 0.834, and 0.885, respectively. Precision studies involved repeated analysis of one nonmedicated adult (mean CA-1000 INR = 0.908) and one adult receiving oral anticoagulation therapy (mean CA-1000 INR = 2.42). The coefficient of variation on the near-patient monitors for both adult volunteers ranged from 4.9% to 22.3%. Bilirubin levels up to 20 mg/dL did not interfere in any of the methods. Conclusions: Near-patient testing whole blood INR monitors offer acceptably accurate and precise measurements. Values obtained on near-patient monitors may vary considerably from the reference method, and data obtained should serve as a supplement to, but not a replacement for, routine clinical laboratory measurements.