• Robert A Reiss, Curtis E Haas, Deborah L Griffis, Bernadette Porter, and Mary Ann Tara.
• Department of Pharmacy Services, ViaHealth-Rochester General Hospital, New York, USA.
• Pharmacotherapy. 2002 Jun 1; 22 (6): 677-85.

Study ObjectiveTo compare point-of-care and standard hospital laboratory assays for monitoring patients receiving single or combination anticoagulant regimens.DesignProspective analysis.SettingNursing units and clinics at a large, community hospital.PatientsOne hundred fifty patients receiving anticoagulants for cardiac, vascular, orthopedic, or cancer indications. Thirty patients were enrolled into each treatment group: warfarin, enoxaparin, heparin, warfarin plus enoxaparin, and warfarin plus heparin.InterventionCapillary and venous blood samples were collected once in each patient for simultaneous measurement of international normalized ratio (INR) and activated partial thromboplastin time (aPTT) by both assays.Measurements And Main ResultsMean differences in paired INR and paired aPTT by point-of-care and standard assays were small, but 95% confidence intervals were wide. The INR differences were greater for the warfarin plus heparin group than for the warfarin group or warfarin plus enoxaparin group; clinical decision agreement was 47% for warfarin plus heparin, 73% for warfarin, and 93% for warfarin plus enoxaparin. The aPTT difference was greater for the warfarin plus heparin than for the heparin group; however, clinical decision agreement, 67% and 70%, respectively, was similar.ConclusionsPoint-of-care methods showed limited agreement with standard hospital laboratory assays of coagulation for all treatment groups. For INR values, significantly greater disagreement was noted between the assay methods for the warfarin plus heparin group compared with the warfarin group, but the agreement was similar for the warfarin and warfarin plus enoxaparin groups. Our data indicate that the point-of-care assays should not be considered interchangeable with standard laboratory assays.

15 keywords...

hide…