INTRODUCTION: The Di@pason scheme, introducing a point-of-care (POC) International Normalized Ratio (INR) measurement with the LabPad® was initiated in 2019 by the French National Health Agency. The aim of our study was to assess the analytical agreement between LabPad ® and laboratory INR results, especially at sub- and supratherapeutic levels. The allowable differences were based on the accuracy requirement defined by the International Standard ISO 17593: 2022 and analytical discordances versus INR Ranges.
METHODS: From February 2020 to August 2022, the agreement between POC and laboratory INR results was analyzed in 83 patients. All subjects were monitored on oral anticoagulant therapy (57 patients treated with fluindione (Previscan ®), 24 patients were treated with warfarin (Coumadin®) anticoagulant medication and acenocoumarol (Sintrom®) was used for two subjects).
RESULTS: The laboratory INR results ranged from 1.2 to 10 with a mean of 3.736±1.479, and LabPad® INR ranged from 0.8 to 7.3 with a mean of 3.818±1.599. Analysis of the graph demonstrated that the INR relationship between LabPad ® and STA-R Max3 ® did not remain linear above 7.3. An extended measurement area and Pearson correlation coefficient (r) showed a significant and strong (r=0.91 [p<0.001; 95% CI: 0.81–1.00]). The regression slope was 0.980 (p<0.001; 95% CI: 0.878–1.081) and the y intercept was 0.158 (p<0.001; 95% CI: −0.251–0.566). The concordance analysis showed that 93% of the results were within the accuracy requirement defined by the international standard ISO17593: 2022, but significant discrepancies appeared on the LabPad® for INR results above 4. According to analytical discrepancies and INR, ranges showed substantial agreement on these criteria with 88% and a Kappa coefficient of 0.67.
DISCUSSION AND CONCLUSION: The LabPad® and laboratory INR results were highly correlated within the therapeutic range, above this range, a venous checking is recommended to confirm the results.