Background and aim of the study: In this single-center French study, the conventional management of patients receiving vitamin K antagonists (VKAs) was compared with an International Normalized Ratio (INR) self-testing program. The aim was to determine the reliability of self-testing devices, and to estimate the variability of the self-measured INR within the therapeutic and target range.
Methods: A total of 206 patients who had undergone valve replacement with a mechanical prosthesis, with or without myocardial revascularization, between May 2004 and September 2007 was randomized into two groups. Group 1 patients (n = 103) underwent INR monitoring at a laboratory, while Group 2 patients (n = 103) underwent self-testing INR using either the CoaguChek (Roche) (Group 2A; n = 55) or INRatio (Hemosense) (Group 2B; n = 48) system. Patients in Group 1 underwent at least once-monthly INR measurement, while those in Group 2 carried out once-weekly self-testing, and also underwent once-monthly INR measurement at the laboratory. The large majority of patients (97.9%) were treated with fluindione.
Results: The mean follow up period was 49.0 +/- 10.3 weeks. Self-testing was reliable, with a correlation coefficient between device- and laboratory-measured INRs of 0.80 [CI: 0.78, 0.82] (p < 0.0001). The proportion of time spent within the INR target range was significantly higher for Group 2 (61.5 +/- 19.3% versus 55.5 +/- 19.9%; p < 0.05), while the absolute mean deviation of INR from the target range was higher in Group 1 (60.1 +/- 70.2% versus 47.4 +/- 51.5%). Adverse events were reported by seven patients in Group 1, but by no patients in Group 2 (p < 0.01).
Conclusion: INR self-testing devices are reliable and beneficial. Moreover, INR self-monitoring allows an enhanced stability within a target range, and also helps to prevent serious postoperative complications.