Key result
EDTA plasma showed high correlation with citrate plasma for prothrombin time estimation using the Owren method, allowing the use of a single EDTA tube for multiple assays.
Why the study?
Does EDTA plasma provide comparable Prothrombin Time measurements to citrate plasma using the Owren method in hospital and health center patients?
Observational (n=123)
Does EDTA plasma provide comparable Prothrombin Time measurements to citrate plasma using the Owren method in hospital and health center patients?
Answers are grounded in this paper and cite their sources.
EDTA plasma can be used to measure PT with the Owren procedure if calibrated with normal EDTA plasma, potentially allowing a single tube for both hematologic and coagulation assays.
Prothrombin time (PT) tests usually are performed on blood collected with citrate. Because blood samples for other common tests use EDTA as anticoagulant, the use of EDTA plasma for PT tests could be convenient, and we wished to explore this possibility. PT tests are needed at regular intervals for control of anticoagulant therapy with vitamin K antagonists (1), and they are the most commonly performed coagulation test in our laboratory. Two major reagent systems are used to measure PT: the Quick method and the Owren method (combined thromboplastin reagent). The latter is the main method in Nordic countries, Benelux, and Japan. The Owren method measures only coagulation factors II, VII, and X because fibrinogen and factor V are in the reagent. We studied only the Owren method, and our results may not apply to the Quick method, in which the sample constitutes a much larger portion of the final reaction volume. The effects of preanalytical factors on PT depend, in part, on the units in which PT results are expressed. In the Nordic countries, percentage has previously been the unit used. When percentages or seconds are used as the unit, the result depends on the reagent. Clinical laboratories are moving toward the use of a new International Normalized Ratio (INR) unit. The use of INR units, INR = (samples/normals)ISI, where “normal” is a plasma pool obtained from healthy persons, ISI is the International Sensitivity Index, is designed to eliminate dependence on sensitivity of the reagents so that the international result is the same and recommendations for care are uniform. INR units are not without problems: reagents do not behave in the same way, and in some cases, the citrate concentration in the sample affects results (2)(3)(4). In our study, we expressed results as percentages, seconds, and INR units. The effects of preanalytical factors on coagulation test precision and accuracy constitute a challenge (5)(6). For most PT measurements, 9 parts of blood are collected with 1 part of citrate solution [0.109 mol/L (3.2%) or 0.129 mol/L (3.8%)]. This sample is usable only for coagulation tests. The use of isocitrate as anticoagulant for coagulation tests has been studied by Rånby et al. (7). The aim of the present study was to establish how well an EDTA sample is suited for PT measurement with the combined thromboplastin reagent (Owren PT). We compared EDTA and citrate samples for PT estimation; we studied 123 paired patient samples, chosen without conscious bias from hospital and health center patients. In both the coagulation tube (citrate) and the EDTA tube, PT was measured and reported in seconds, percentage, and INR units. Measurement commenced within 2 h of blood collection. For 31 of the same patients, Ca2+ concentrations were measured in the reaction mixtures for both sample types. The citrate coagulation tube (Vacutainer, cat. no. 367702, 9NC; Becton Dickinson) contained 0.35 mL of 0.129 mol/L citrate solution and had a blood volume of 3.15 mL. The EDTA sample tube (Vacutainer, cat. no. 367652; Becton Dickinson) contained 0.072 mL of 75 g/L K3EDTA. The blood volume was 3.0 mL, thus providing an EDTA concentration of 1.5 mg/mL of blood. The EDTA solution was 2.34% and blood was 97.66% of the final total volume. The sample needle (Precision Glide, cat. no. 360213; Becton Dickinson) was 0.8 × 38 mm. Sample tubes were centrifuged at 1560g for 10 min at 12 °C to separate the plasma. The ACL 1000 (Instrumentation Laboratory) is a fully automatic microcentrifugal analyzer. We measured PT in the ACL analyzer with a coagulation reagent (Nycotest PT, cat. no. 10021340; ISI = 1.21) and a diluent, citrate-barbiturate buffer (cat. no. 1002881), both from Nycomed Pharma. Each reaction contained 8 μL of sample, 42 μL of diluent, and 100 μL of reagent. We used the IL Test Reference (cat. no. 97569-00) as the internal standard for ACL. The ACL 1000 was calibrated with Normal kontrollplasma 160 [NKP 160, cat. no. GHI 160, lot nos. Y502P 17.8 s (INR 1.0) and Z235P 17.4 s (INR 1.0); Global Hemostasis Institute]. The ACL must be calibrated for percentages and INR units, and the calibration was the same for both sample types. The ACL PT calibration data (n = 6) were as follows: for the 100% calibrator, 17.4 s, CV = 0.48%; for the 50% calibrator, 24.8 s, CV = 0.34%; for the 25% calibrator, 35.8 s, CV = 1.1%, R2 = 0.994. For ACL quality control during 1 week of routine use of the control plasma, the mean was 16.9 s (n = 70; within-run CV = 2.3%). The within-run CV was 0.61% (n = 10) at a mean of 16.4 s. We measured Ca2+ with an IL 1640 blood gas analyzer, using IL Test contrIL PLUS (cat. no. 98321-30, lot no. I0798233) as control. Quality-control results for the analyzer during routine use were as follows: level 1 control, mean Ca2+ = 0.70 mmol/L (n = 9; target, 0.63–0.75 mmol/L; CV = 1.6%); level 2 control, mean Ca2+ = 1.08 mmol/L (n = 8; target, 1.00–1.12 mmol/L; CV = 1.3%); level 3 control, mean Ca2+ = 1.49 mmol/L (n = 7; target, 1.40–1.58 mmol/L; CV = 1.2%). For ionized Ca2+, we used IL Test Cal 1 (cat. no. 98318-00), Test Cal 2 (cat. no. 98342-00) as calibrators, and Test BGE Flush (cat. no. 98320-00) as the cleansing solution. For ionized Ca2+, the reaction mixture was prepared with volumes threefold larger than those for the ACL (24 μL of sample + 126 μL of diluent + 300 μL of reagent); after coagulation, the solution was centrifuged at 12 700g for 10 min, and Ca2+ was measured in the supernatant by the IL 1640 analyzer. We used Microsoft Excel 5.0 to obtain the correlation function. The mean Ca2+ concentration in the reaction mixture with citrate samples was 1.52 mmol/L (SD = 0.07 mmol/L); with EDTA samples, the mean was 1.71 mmol/L (SD = 0.07 mmol/L). The concentration with the EDTA sample thus was 0.183 mmol/L (12%) higher. In each pair of samples, the Ca2+ concentration was higher with EDTA. PT results were between 15.8 s (136%; 0.89 INR units) and 110 s (6%; 9.34 INR units). Regression analysis yielded R2 >0.99, so the linear least-squares method was used. The regression equations for PT estimations from citrate plasma (y) and EDTA plasma (x) were: for seconds, y = 1.17x − 2.37 s (R2 = 0.998; Fig. 1 ); for percentage, y = 0.898x − 0.785% (R2 = 0.994); for INR units, y = 1.20x − 0.162 INR (R2 = 0.998). Despite the high correlation, the results differed numerically. Two differences must be considered: the anticoagulant itself and the dilution of sample by the anticoagulant. The difference between results depended mainly on sample dilution by citrate solution; the EDTA sample was diluted only minimally. When the PT test is calibrated with normal EDTA plasma, these two differences disappear, so that it is possible to use EDTA plasma to measure PT with the Owren procedure. The coagulation reagent contains sufficient Ca2+ to allow coagulation with either sample type. The physiological concentration of Ca2+ in plasma is ∼1.3 mmol/L. The Ca2+ concentrations in the reaction mixtures are higher for both sample types (1.525 mmol/L for citrate and 1.708 mmol/L for EDTA). The main reason for the lack of a problem with EDTA as anticoagulant may reside in the assay procedure used. In the Owren method, only ∼5% of the reaction mixture is sample, whereas it is 33% in the Quick PT method. Difficulties in correctly balancing the Ca2+ concentration are thus much smaller with the Owren PT method than with the Quick PT. In an EDTA sample, blood cells preserve their cell morphology well (or the changes in morphology are known and do not cause diagnostic confusion), and there are minimal nonphysiologic effects on the cells (8). When tubes with solid EDTA are used, there is no dilution of the sample. In this study, tubes were used in which EDTA was in liquid form, but the dilution was ∼2.34%. It is important that the calibrators (for results in percentages and INR units) have the same dilution as the samples; calibrator manufacturers should inform the user of the diluent and dilution rate. When seconds are the unit, no calibration is involved. According to the regression equation, coagulation times are shorter with EDTA plasma because of the dilution. EDTA sample results (in percentages and INR units) can be converted by equation to citrate plasma results or vice versa. If we calibrate (percentages and INR units) our coagulation analyzer with normal EDTA plasma, we immediately obtain the correct result, and the citrate-plasma reference and therapeutic range for oral anticoagulant treatment are the same as for EDTA plasma. Use of a common tube for hematologic and coagulation assays is a good aim. Hematological analyzers use ∼150–250 μL of a blood sample. After hematological measurements are made, the sample can be centrifuged to obtain plasma for PT testing. This order of proceeding does not alter the relationship between plasma and cells. Use of the same EDTA sample for hematological and coagulation estimations has many advantages: sampling is faster in one tube, material and waste costs are lower, and the overall number of tubes needed in the laboratory is smaller, which saves natural materials. Moreover, the small sample volume needed is important for different patient groups and children. In addition, laboratories usually have an adequate supply of normal EDTA plasma for calibration. PT analysis for EDTA and citrate samples, in seconds. The equation for the line is as follows: y = 1.17x − 2.37; R2 = 0.9978. I thank the staff of the Laboratory at the Valkeakoski District Hospital for their assistance.
No takes yet. Share an insight, caveat, or question.
Juha Horsti (2000) conducted an observational in Patients requiring prothrombin time (PT) estimation (n=123). EDTA plasma vs. Citrate plasma was evaluated on Prothrombin time (PT) estimation correlation (R2 > 0.99). EDTA plasma showed high correlation with citrate plasma for prothrombin time estimation using the Owren method, allowing the use of a single EDTA tube for multiple assays.
Synapse has enriched 10 related papers. Consider them for comparative context: