Comparison of CoaguChek® XS INR and Laboratory INR in Patients With a Heartware® Continuous Flow Left Ventricular Assist Device

Background: Accuracy of point-of-care testing with CoaguChek® XS (Roche) has not been well validated in patients with Continuous Flow Left ventricular assist devices (CF-LVAD). Methods: Patients with a Heartware® CF-LVAD for end-stage heart failure (ESHF) between December 2013 and August 2015 as a bridge to transplant were included in the retrospective analysis. As part of standard care INR values as measured by CoaguChek® XS and the laboratory Stago's STA-R Evolution analyser were recorded on a single spreadsheet to confirm correlation prior to discharge. Blood samples for each of the testing methods was taken within a 4 hour period of the other. Results: 230 samples were obtained from 15 CF-LVAD patients with a mean age of 40 (+/-14) years. There was a moderate correlation between laboratory and CoaguChek® XS INR values with a correlation coefficient of 0.85 (r2=0.72, CI:0.81-0.88, p<0.0001; Fig 1a). The mean difference between the methods was 0.13 (+/-0.54) with CoaguChek® XS tending to overestimate INR (Fig 1b). When the laboratory INR was normal, CoaguChek® XS returned a normal result in 191 (91%) of occasions and 16 (94%) of the abnormal CoaguChek® XS results were an overestimation of a normal laboratory INR with only one being an underestimation of a normal laboratory INR. Conclusion: This study supports the use of point-of-care testing with CoaguChek® XS in patients with ESHF taking warfarin supported with the Heartware® CF-LVAD. However, CoaguChek® XS tended to overestimate INR in this setting.