Capillary Whole Blood Testing by a New Portable Monitor Comparison With Standard Determination of the International Normalized Ratio

Transcription

1 Coagulation and Transfusion Medicine / CPILLRY WHOLE LOOD TESTING Y NEW PORTLE MONITOR Capillary Whole lood Testing by a New Portable Monitor Comparison With Standard Determination of the International Normalized Ratio Dunia de Miguel, MD, Carmen urgaleta, MD, PhD, Eduardo Reyes, PhD, and Teresa Pascual, MD Key Words: Oral anticoagulant therapy; mbulatory patients; Capillary blood; Portable monitor DOI: 0.0/0PDJQGDR bstract We evaluated a new portable monitor (vosure PT PRO, Menarini Diagnostics, Firenze, Italy) developed to test the prothrombin time in capillary blood and plasma by comparing it with the standard laboratory determination. We studied patients receiving acenocoumarol therapy. The international normalized ratio (INR) in capillary blood was analyzed by methods: vosure PT PRO and Thrombotrack Nycomed nalyzer (xis-shield, Dundee, Scotland). Parallel studies were performed in plasma samples by a reference method using the ehring Coagulation Timer (ehring Diagnostics, Marburg, Germany). Plasma samples also were tested with the vosure PT PRO. Correlation was good for INR values for capillary blood and plasma samples by vosure PT PRO and our reference method (R = 0.) and for capillary blood samples tested by the vosure PT PRO and Thrombotrack Nycomed nalyzer (R = 0.). The correlation for INR in capillary blood and plasma samples by vosure PT PRO was 0. (P <.000). Capillary blood determinations are rapid and effective for monitoring oral anticoagulation therapy and have a high correlation to plasma determinations. vosure PT PRO is accurate for controlling INR in plasma and capillary blood samples, may be used in outpatient clinics, and has advantages over previous portable monitors. Oral anticoagulation therapy is the treatment of choice for ambulatory patients with previous venous thrombotic episodes, for patients undergoing long-term anticoagulation after myocardial infarction, and for patients with atrial fibrillation., Patients receiving oral anticoagulant therapy must undergo careful monitoring to reduce the risk of hemorrhagic complications due to excess anticoagulation and of new thrombotic episodes due to inadequate anticoagulation. Oral anticoagulant therapy usually is monitored by determining the prothrombin time (PT) with the international normalized ratio (INR) on citrated plasma samples. Patients receiving oral anticoagulant therapy require regular and frequent monitoring of their PT values to ensure the desired therapeutic range. Traditional anticoagulation clinics require a high degree of medical compliance; in addition, for patients who live far from the hospital or who have poor venous access, anticoagulation control may be difficult. In recent years, the increased number of patients receiving oral anticoagulant therapy has contributed to the development of instruments for measurement of the INR in noncitrated whole blood samples., This type of monitor uses a drop of blood obtained by finger stick and allows near instantaneous measurements of the INR and PT, permitting an immediate change of dosage, if needed. In this context, portable monitors that perform coagulation tests from capillary blood offer advantages for control in clinics and for home testing. new generation of small instruments for the determination of PT and INR in capillary blood samples has been developed for monitoring oral anticoagulant therapy at home. Prospective multicenter studies have evaluated the accuracy and precision of several portable PT monitors and concluded m J Clin Pathol 00;0:- DOI: 0.0/0PDJQGDR

2 Coagulation and Transfusion Medicine / ORIGINL RTICLE that they can be used to monitor oral anticoagulant therapy. -0 Randomized clinical trials have shown that home INR testing and dose self-adjustment are feasible and safe for patients receiving oral anticoagulant therapy. The aim of our study was to evaluate the accuracy of a new portable PT monitor (vosure PT PRO, Menarini Diagnostics, Firenze, Italy) developed to test INR in whole blood and plasma samples. The INR values obtained with this portable system in a group of patients undergoing anticoagulation attending the ambulatory anticoagulation clinic were compared with the INR values obtained for plasma samples with the reference method used in our laboratory. The INR results obtained in capillary blood samples by a previous generation monitor also were evaluated. Materials and Methods Patients The study included patients receiving oral anticoagulant therapy who regularly attended the anticoagulant ambulatory clinic in the hematology service of our institution. Fifteen had a prosthetic heart valve;, atrial fibrillation;, mitral stenosis;, venous thromboembolic disease;, ischemic stroke; and, acute myocardial ischemia. ll patients received acenocoumarol. The median age was years (range, - years), and the sex distribution was men and women. Present guidelines recommend maintaining the INR between and (in a study of patients) or between and.0 (in a study of patients with prosthetic heart valves).,- Informed consent was obtained from patients. Methods The INR was tested simultaneously in whole capillary blood and plasma samples by methods. Determination of the INR in Capillary lood Samples lood samples were obtained by finger stick and analyzed by using the vosure PT PRO and the Thrombotrack Nycomed nalyzer (xis-shield, Dundee, Scotland). The vosure PT PRO portable monitor requires test strips containing dried recombinant rabbit brain thromboplastin for INR determination and control. drop of blood is placed directly on the test strip, and the meter detects the resistance drop between the electrodes and initiates test timing. This system uses an asymmetric polysulfone membrane to separate plasma from RCs and works with samples as small as 0 µl. The membrane contains calcium and thromboplastin and permits occurrence of the reactions of the complete extrinsic pathway with minimal distortion from membrane surface interactions. Thrombin generation is monitored optically using a rhodamine 0 based fluorescent thrombin substrate. Fluorescence kinetics are analyzed to produce a PT-equivalent parameter that can be converted to an INR value. The system provides results that correlate well with conventional liquid phase PT assays (R = 0.)., The handling of this meter is simple. blood drop is placed directly on the test strip, eliminating the need for capillary tubes and specialized collection devices and reducing exposure to blood contamination. The results are expressed as INR, without the need for a conversion table. It also permits the use of plasma samples and offers the possibility for connecting on line with a computer to print results from the monitor, avoiding transcription errors. The vosure PT PRO has liquid phase controls and an electronic control for meter quality verification. The device was tested with electronic controls every day before performing the anticoagulation controls. The vosure liquid control solutions, level and level, contain lyophilized nonhuman plasma with predetermined INR values. The liquid control solution is applied to a test strip in the same manner as the patient s blood sample. The level control is equivalent to a sample with a low INR value (.-.), and the level control is equivalent to a sample with a high INR value (-.). The thromboplastin has an approximate average international sensitivity index (ISI) of. The ISI of the thromboplastin in each lot is held for each one, using an internal signal placed on each cartridge. vosure PT PRO is calibrated by materials traceable to World Health Organization International Reference Preparations CT/ and rtf/. The same lot of cartridges was used for INR determinations with the vosure PT PRO analyzer throughout the study. In the Thrombotrack Nycomed nalyzer, clot formation is monitored by change in blood viscosity through magnetic particle oscillation. lood was obtained by finger stick and aspirated with a pipette before being deposited on the cartridge, which contained rabbit brain thromboplastin. The absence in particle movement is detected by a reduction in reflectance. The resulting clotting time is recorded and displayed as a PT ratio, and a conversion table is necessary to obtain INR values. This has been the method applied in our institution since to test INR in ambulatory patients receiving oral anticoagulant therapy (our outpatient clinic performs more than 0,000 determinations per year). The results are compared regularly with determinations in plasma samples by the reference method detailed in the following section. Determination of the INR in Plasma Samples lood was obtained by venipuncture and was collected in.% sodium citrate anticoagulated tubes (ecton Dickinson m J Clin Pathol 00;0:- DOI: 0.0/0PDJQGDR

3 de Miguel et al / CPILLRY WHOLE LOOD TESTING Y NEW PORTLE MONITOR [Erembodegem-alst, elgium] Vacutainer Systems NC 0. M), regularly used for INR determination, as well as for other coagulation studies in European countries. The manufacturer in the United States (vocet Medical, San Jose, C) used lower (.%) citrate concentrations for PT testing, and the lower concentrations previously have shown good correlation between plasma and capillary blood samples. The sample was centrifuged at,0g for minutes at C to obtain platelet-poor plasma. The PT was analyzed in plasma samples in the ehring Coagulation Timer (CT) (ehring Diagnostics, Marburg, Germany) as a reference method. The PT in citrated plasma was determined by a coagulometric method testing by optical density using a suspension of thromboplastin of rabbit brain in calcium chloride solution,. g/l (Thromborel-S reagent, ehring Diagnostics). The thromboplastin ISI value assigned by the manufacturer for this reagent on the CT coagulometer was. The INR in plasma samples also was determined by using the vosure PT PRO (plasma mode). The accuracy of the portable monitor measurements was evaluated by examining the agreement between the INR values obtained with vosure PT PRO monitor and the analyzer from the clinical hematology laboratory. Samples for comparative measures were obtained at the same time by a nurse. Statistical nalysis The INR values obtained by the different methods were compared by analysis of variance for repeated measures. Correlations between parameters were evaluated by the Pearson correlation and the Spearman nonparametric method. The relationships between the monitor INR values and the reference laboratory values are expressed by linear regression equation. We also determined the % confidence intervals. Concordance between results also was compared by using the land-ltman plot method (SD between the difference, ±.). nalysis was performed using SPSS for Windows (version.00, Chicago, IL). Clinically Relevant greement greement between results from the vosure PT PRO and CT coagulometer and between the Thrombotrack Nycomed nalyzer and CT coagulometer was defined for each dual INR measurement using criteria considered to be clinically relevant. Standard agreement as defined previously was achieved if any of the following conditions was present: () INR results for the vosure PT PRO and CT coagulometer were within the patient s targeted therapeutic range; () vosure PT PRO and CT coagulometer measurements were both above or below the targeted therapeutic range; () vosure PT PRO and CT coagulometer INR measurements were within 0. unit of each other. Results Monitor Measurements We analyzed blood specimen pairs using the monitor and the laboratory analyzer. The INR values showed correlations of R = 0. between plasma samples (CT coagulometer) and capillary blood samples (vosure PT PRO) and R = 0. between plasma samples (CT coagulometer and vosure PT PRO) Figure and Figure. INR p CT INR cb vosure INR cb vosure INR p CT (INR cb vosure + INR p CT)/ Figure International normalized ratio (INR) for plasma (p) samples on the ehring Coagulation Timer (CT) analyzer and capillary blood (cb) samples on the vosure PT PRO analyzer., Regression values; % confidence interval (hatched lines). Identity line, y = x (dotted line). Solid line represents the mean. y = 0.x + 0.; R = 0.; P <.00., land-ltman plot values. The mean (solid line) ± SD =. (dotted lines) is shown. For proprietary information, see the text. 0 m J Clin Pathol 00;0:- 0 DOI: 0.0/0PDJQGDR

4 Coagulation and Transfusion Medicine / ORIGINL RTICLE INR p CT INR p vosure INR p vosure INR p CT (INR p vosure + INR p CT)/ Figure International normalized ratio (INR) for plasma (p) samples on the ehring Coagulation Timer (CT) and the vosure PT PRO analyzer., Regression values; % confidence interval (hatched lines). Identity line, y = x (dotted line). Solid line represents the mean. y = 0.x + 0.0; R = 0.; P <.000., land-ltman plot values. The mean (solid line) ± SD =. (dotted lines) is shown. For proprietary information, see the text. The correlation between INR values in capillary blood samples (vosure PT PRO and Thrombotrack Nycomed nalyzer) was R = 0. (P <.00) Figure. The correlation between capillary blood and citrated plasma samples (vosure PT PRO) was R = 0. (P <.00) Figure. Clinically Relevant greement Clinically relevant agreement was obtained between capillary blood results (vosure PT PRO) and results from the reference method in plasma samples from patients (%). The proportions of patients with INR values within, above, or below the therapeutic range were % (/), % (/), and % (/), respectively, with both methods. For patients (%), the results in capillary blood and plasma samples were different and the dosage of acenocoumarol was modified, using the plasma result as optimal (dosage reduced for patients; dosage increased for patients). For the patients receiving anticoagulant therapy owing to the presence of metallic prosthetic heart valves, INR results were in the same range with the methods (vosure, Thrombotrack Nycomed, and CT). INR cb Thrombotrack INR cb vosure INR cb Thrombotrack INR cb vosure (INR cb vosure + INR cb Thrombotrack)/ Figure International normalized ratio (INR) for capillary blood (cb) samples on the vosure PT PRO analyzer and the Thrombotrack Nycomed nalyzer., Regression values; % confidence interval (hatched lines). Identity line, y = x (dotted line). Solid line represents the mean. y = 0.x + 0.; R = 0.; P <.00., land-ltman plot values. The mean (solid line) ± SD =. (dotted lines) is shown. For proprietary information, see the text. m J Clin Pathol 00;0:- DOI: 0.0/0PDJQGDR

5 de Miguel et al / CPILLRY WHOLE LOOD TESTING Y NEW PORTLE MONITOR INR p vosure INR cb vosure INR cb vosure INR p vosure (INR cb vosure + INR p vosure)/ Figure International normalized ratio (INR) for capillary blood (cb) and plasma (p) samples on the vosure PT PRO analyzer., Regression values; % confidence interval (hatched lines). Identity line, y = x (dotted line). Solid line represents the mean. y = x +.0; R = 0.; P <.00., land-ltman plot values. The mean (solid line) ± SD =. (dotted lines) is shown. For proprietary information, see the text. Discussion Oral anticoagulant therapy is used widely in the treatment and prevention of thromboembolic diseases. Previous studies have demonstrated the accuracy and precision of other portable instruments designed for monitoring oral anticoagulation., We evaluated the statistical agreement and the clinically relevant agreement between a new portable device for fingerstick INR control (the vosure PT PRO) and the laboratory reference method (CT coagulometer). The INR values also were analyzed and compared with results from the Thrombotrack Nycomed nalyzer (a portable monitor) Good correlation between results from the vosure PT PRO and CT coagulometer was found. Similar results also were obtained for the INR in capillary blood samples determined by using the Thrombotrack Nycomed nalyzer and by using the vosure PT PRO. The results are in accordance with the findings of previous studies performed with other portable monitors. Differences between results for the INR obtained by different methods were clinically significant and required dosage changes in a small proportion of patients. However, among those with INR ranges between and, similar results were obtained for patients who have a mechanical prosthetic valve and are undergoing anticoagulation. Those results are in contrast with previously reported results obtained by other methods, 0 although the number of patients from the present group was small. The present study included patients treated with acenocoumarol, a drug currently used for oral anticoagulation in many European countries, and confirms the results obtained with warfarin., The vosure PT PRO has clear advantages over the Thrombotrack Nycomed nalyzer in terms of manipulation and security for nurse operators and patients. lthough the current study was performed in an anticoagulation clinic, previous studies have shown the accuracy of portable PT monitors and satisfactory results for self-testing at home. We conclude the that vosure PT PRO, used to test the INR in capillary blood and plasma samples, has an acceptable level of accuracy for oral anticoagulation control as one of a new generation of portable prothrombin monitors and has advantages over other portable monitors. From the Service of Hematology, Príncipe de sturias University Hospital, and Department of Medicine, lcalá University, lcalá de Henares, Madrid, Spain. Supported in part by grant FIS 00/ from the National Health Service, Spain, and a grant from Menarini Diagnostics, Firenze, Italy. ddress reprint requests to Dr urgaleta lonso de Ozalla: Servicio de Hematología, Hospital Universitario Príncipe de sturias, 0 lcalá de Henares, Madrid, Spain. References. Hirsch J. Oral anticoagulant drugs. N Engl J Med. ;:-.. Geerts WH, Heit J, Clagett GP, et al. Prevention of venous thromboembolism. Chest. 00;( suppl):s-s.. Rosendal FR. The Scylla and Charybdis of oral anticoagulant treatment. N Engl J Med. ;:-.. Loeliger E, van den esselaar M, Lewis SM. Reliability and clinical impact of the normalization of the prothrombin times in oral anticoagulant control. Thromb Haemost. ;:-. m J Clin Pathol 00;0:- DOI: 0.0/0PDJQGDR

6 Coagulation and Transfusion Medicine / ORIGINL RTICLE. Tripodi, Chantarangkul V, Mannucci P. Near-patient testing devices to monitor oral anticoagulant therapy [review]. r J Haematol. 00;:-.. White RH, McCurdy S, von Marensdorff H, et al. Home prothrombin time monitoring after the initiation of warfarin therapy. nn Intern Med. ;:0-.. Lucas FV, Duncan, Jay R, et al. novel whole blood capillary technique for measuring the prothrombin time. m J Clin Pathol. ;:-.. nsell JE. nticoagulation management as a risk factor for adverse events: grounds for improvement. J Thromb Thrombolysis. ;(suppl ):S-S.. reukink-engbers WG. Monitoring therapy with anticoagulants in the Netherlands. Semin Thromb Hemost. ;:-. 0. Palareti G. guide to oral anticoagulant therapy. Italian Federation of nticoagulation Clinics. Haemostasis. ;(suppl ):-.. Cosmi, Palareti G, Carpenedo M, et al. ssessment of patient capability to self-adjust oral anticoagulant dose: a multicenter study on home use of a portable prothrombin time monitor (COGUCHECK). Haematologica. 000;:-.. Navarro JL, César JM, García Frade LJ, et al. Checking oral anticoagulation in capillary blood [in Spanish]. Sangre. ;:-0.. Zweig SE, Meyer G, Sharma S, et al. Membrane-based, dryreagent prothrombin time tests. iomed Instrum Technol. ;0:-.. Hyers TM, gnelli G, Hull RD, et al. ntithrombotic therapy for venous thromboembolic disease. Chest. 00;( suppl):s-s.. onomi, Tripodi, Chantarangkul V, et al. Near-patient testing devices to monitor oral anticoagulant therapy [review]. r J Haematol. 00;; -.. van den esselar MHP. comparison of INRs determined with a whole blood prothrombin time device and two international reference preparations for thromboplastin. Thromb Haemost. 000;:0-.. Gosselin R, Owings JT, White RH, et al. comparison of point-of-care instruments designed for monitoring oral anticoagulation with standard laboratory methods. Thromb Haemost. 000;:-0.. Schmitz LL, Olson SL, McCormick SR, et al. Failure to generate comparable international normalized ratio values using five different thromboplastin reagents in parallel studies of patients receiving warfarin. Clin ppl Thromb Hemost. ;:-0.. Gosselin RC, Owings JT, Larkin E, et al. Monitoring oral anticoagulation therapy with point-of-care devices: correlations and caveats. Clin Chem. ;:-. 0. Cosmi, Palareti G, Moia M, et al. ccuracy of a portable prothrombin time monitor (Coagucheck) in patients on chronic oral anticoagulant therapy: a prospective multicenter study. Thromb Res. 000;00:-.. McCurdy S, White RH. ccuracy and precision of a portable anticoagulation monitor in a clinical setting. rch Intern Med. ;:-. m J Clin Pathol 00;0:- DOI: 0.0/0PDJQGDR