Comparison of three D-dimer assays for the diagnosis of DVT:ELISA, latex and an immunofiltration assay (NycoCard D-Dimer).

Ninety-two consecutive patients referred for suspicion of deep venous thrombosis (DVT) were analyzed for D-dimer using ELISA, latex test, and a new immunofiltration method (NycoCard D-Dimer). Contrast venography verified the diagnosis in 40, and excluded the diagnosis in 52 patients. The sensitivity, negative predictive values, specificity and positive predictive values were, for ELISA 98%, 95%, 38% and 54, for NycoCard D-Dimer 100%, 100%, 42% and 57% and for the latex test 73%, 78%, 75%, and 69%, respectively. Sensitivity and specificity were inversely related with increasing pathological cut-off value. Comparison of test results by concentration category revealed a good agreement between ELISA and NycoCard D-Dimer, but to less extent between latex and the two other tests. It is concluded that NycoCard D-Dimer and D-dimer ELISA are well-suited as exclusion tests for DVT. A plasma sample is tested with NycoCard D-Dimer in less than 2 min. Thus, this test combines advantageous analytical properties comparable to the ELISA-test, with rapidity and simplicity comparable to the latex test.

Assay of D-dimer based on immunofiltration and staining with gold colloids.

In this immunofiltration assay of D-dimer in plasma samples, the antigens are captured by a monoclonal antibody on a porous membrane, and labeled with the same antibody conjugated to gold colloids. The assay time is < 2 min, and a color of intensity proportional to the concentration of D-dimer is left on the membrane. The reference range (mean +/- 2 SD) was 0.336 +/- 0.133 mg/L (n = 69). Linearity was found up to 10 mg/L. Comparison with ELISA results (x) for 198 patients' samples demonstrated a linear regression equation of y = 0.99(+/- 0.05)x + 0.68(+/- 0.07) and a mean square error of 0.503. Comparison of visual reading of the color signal (y) vs reflectometric measurements (x) for 220 patients' samples demonstrated a linear regression equation of y = 2.5(+/- 0.06)x -0.22(+/- 0.04) and a mean square error of 0.095. Bilirubin, hemoglobin, fibrinogen, soluble fibrin, and fibrinogen degradation products and freezing/thawing of samples did not interfere. Some interference from rheumatoid factor, heparin, and the presence of cells or large lipid particles was seen. The variance (CV) was 8-12% within run, 10-18% between runs, and 13-20% between persons. The new assay constitutes a rapid and reliable analytical tool combining simplicity equivalent to that of latex tests with analytical information approaching that of ELISA.

Rapid immunometric measurement of C-reactive protein in whole blood.

We examined an instrument-free test for C-reactive protein (CRP) in whole blood. The NycoCard CRP Whole Blood test uses a cell-solubilizing dilution liquid, a membrane-bound antibody that binds CRP, and a gold-conjugated antibody for making visible the bound CRP. We obtained essentially identical dose-response curves in citrate-, heparin-, and EDTA-treated blood. CVs were 6.7-12.5% within series and 10.1-14.7% between series. The detection limit was 12 mg/L. Intralipid added to blood increased measured CRP by 10-20%, whereas no change was seen with added bilirubin, added serum amyloid P component, or the presence of rheumatoid factor. In 234 patients' blood samples the results of the NycoCard Whole Blood test correlated well (r = 0.96) with those of a turbidimetric serum method. The test allows reliable measurement of CRP from a small volume of whole blood (25 microL) without using expensive equipment; it should be useful for decentralized testing in hospital departments, emergency units, and primary health care centers.

Turbidimetric determination of prothrombin time by clotting in a centrifugal analyzer.

Two thromboplastin reagents ("Thrombotest" and "Normotest Automated") were used in evaluation of an automated method for determination of prothrombin time based on turbidimetric measurement of clot formation in a centrifugal analyzer. We used 60 plasma samples from patients with various diseases or being treated with oral anticoagulant and 16 normal plasma samples. Prothrombin times were calculated by a computer connected to the analyzer, a reading being made at either a certain per cent increase in total absorbance or a fixed absorbance increase. Both correlated well with the manual method (r = 0.98-0.99). The reading points best fitting the manually obtained data were estimated by minimizing the residual sum of squares in regression analyses performed at various absorbance increases. The per cent reading was better in this respect. Normotest Automated could be nearly perfectly related to the manual method, whereas Thrombotest showed a (negligibly) small deviation. Reproducibility was good within run (CV less than or equal to 3.2%) as well as between batch of the reagents, as assessed from variation in INR (CV less than or equal to 4.9%). We conclude that turbidimetry of clot formation may be validly used in automation of the prothrombin-time test. The equipment needed and the total time per analysis are about as for chromogenic substrate methods, but reagent cost is considerably lower.

Utility of a modified calibration model for reliable conversion of thromboplastin times to international normalized ratios.

The combined thromboplastin reagent, Normotest, has been calibrated against the secondary international reference preparation for bovine thromboplastin, OBT/79. Three expert laboratories measured up to 62 patients on stabilized oral anticoagulant therapy and up to 20 normals in order to establish an INR-scale for Normotest. It was found that the model recommended by the WHO was less suited for the calibration of this thromboplastin. This is the first study in which three independent laboratories demonstrate a similar bias of the WHO calibration model. A modified model in which a correction factor is introduced was applied to the problem and proved to give a reliable calculation method for INR on Normotest. The mean coefficient of variation of INR calculated between measurements with Normotest and OBT/79 (scatter of data around calibration line) was 4.2-5.0% as compared to 5.1-5.7% for the WHO-method. A conversion scale for percent activities between Normotest and Thrombotest was established showing that the recommended therapeutic range of 5-10% Thrombotest (INR = 4.8-2.8) corresponds to 10-20% Normotest.

Glycoprotein IIb-IIIa complex in platelets of patients and heterozygotes of Glanzmann's thrombasthenia.

The concentration of the glycoprotein (GP) IIb-IIIa complex in thrombasthenic platelets of 8 patients of 6 families has been estimated. In the thrombasthenic platelets of 3 patients this complex is absent (thrombasthenia type I and subtype I). In 2 patients only traces are detectable and in 3 patients GP IIb-IIIa complex is strongly reduced (less than 5%). On the basis of the haemostatic data as well as the content of GP IIb-IIIa complex and platelet fibrinogen the classification of these types as subtypes of thrombasthenia type II is discussed. The diagnostic applicability of GP IIb-IIIa complex determination for heterozygote detection in types of thrombasthenia with absent or extremely reduced GP IIb-IIIa complex is shown.

Comparison of protein and lipid composition of the human platelet alpha-granule membranes and glycerol lysis membranes.

Platelet glycerol lysis membranes and alpha-granule membranes were compared with respect to protein and lipid composition. Crossed immunoelectrophoresis using antibodies against whole platelets, and sodium dodecyl sulphate polyacrylamide gel electrophoresis, revealed the presence of the glycoproteins IIb and IIIa, myosin and an antigen termed G4 in both membrane fractions. The glycoproteins Ia, Ib and IIIb, in addition to beta 2-microglobulin and actin, appeared specific for the glycerol lysis membranes, whereas two antigens, termed G8 and G18, were observed only in the alpha-granule membranes. The localization of glycoprotein IIa was inconclusive. Comparison with the surface-located proteins revealed that the glycerol lysis membranes represented a reasonable approximation to a plasma membrane preparation. Radioactively labelled immunoprecipitates obtained after crossed immunoelectrophoresis of 125I-labelled platelets were cut out and applied to sodium dodecyl sulphate electrophoresis on polyacrylamide slab gels. Autoradiography of the dried gels revealed that antigen G4 represented a protein with an average molecular weight of 146 000 in its unreduced state and 132 000 in its reduced state. Antigen G18 represented a protein of molecular weight 130 000-135 000 in the reduced as well as unreduced state. Quantitation of protein and lipids showed that the alpha-granule membranes contained about one-third as much cholesterol and 2-times as much protein in relation to phospholipids as compared to the glycerol lysis membranes. No significant difference between the two membrane preparations was found as regards the composition of their phospholipids.

Demonstration of 125I-labelled thrombin binding platelet proteins by use of crossed immunoelectrophoresis and autoradiography.

A possible receptor for thrombin on the platelet membrane has been identified. Whole platelets were treated with 125I-labelled thrombin followed by washing of the platelets, solubilization in Triton X-100, crossed immunoelectrophoresis and autoradiography. A heavily labelled antigen which migrated slightly more slowly than albumin was observed. No corresponding arc was seen on the same immunoplate when stained with Coomassie brilliant blue, indicating that the antigen possessed weak antigenic properties and/or was present in very small amounts. When 125I-labelled thrombin that had been inactivated by phenylmethylsulphonyl fluoride was used, no such labelled arc was seen. The radiolabelled immunoprecipitate does not represent any of the antigens identified hitherto in the immunoelectrophoretic patterns obtained with platelets or platelet material. The electrophoretic mobility of the antigen was influenced neither by neuraminidase treatment of the platelets prior to the 125I-labelled thrombin exposure nor by inclusion of concanavalin A, wheat-germ lectin or lentil lectin in the gel during the first-dimension electrophoresis. This suggests that the antigen does not represent a glycoprotein. Upon subcellular fractionation the radioactively labelled arc was observed in the cytosol fraction following crossed immunoelectrophoresis and autoradiography. Analysis of the secreted proteins after induction of the release reaction with 125I-labelled thrombin revealed labelling of immunoprecipitates representing thrombospondin, albumin and the 'line' form of platelet factor 4. This confirms that stable complexes of 125I-labelled thrombin and platelet proteins can exist in the presence of Triton X-100 and during electrophoresis.

Complex-formation between the fibrin-derived plasmic fragments DD and E demonstrated by crossed immunoelectrophoresis.

The formation of a complex between the fibrin fragments DD and E was studied by crossed immunoelectrophoresis using antibodies against human fibrinogen. The complex formation was seen by a common electrophoretic migration of the DD-fragment and part of the E-fragments. This effect was abolished by a further incubation with plasmin of the preparation containing the (DD) E-complex. This also led to an anodal shift in migration of the E-fragment indicating a transfer from E1 to E3.

Calcium-binding proteins from human platelets. A study using crossed immunoelectrophoresis and 45Ca2+.

Calcium-binding platelet proteins were examined by crossed immunoelectrophoresis of solubilized platelets against antibodies to whole platelets followed by incubation of the immunoplates with 45Ca2+ and autoradiography. When the immunoplates had been pretreated with EDTA at pH 9.0 in order to remove divalent cations, three immunoprecipitates were markedly labelled with 45Ca2+. These corresponded to the glycoprotein IIb-IIIa complex, glycoprotein Ia and a presently unidentified antigen termed G18. These antigens were membrane-bound and surface-oriented. When an excess of EDTA was introduced in the incubation media the results revealed that the glycoprotein IIb-IIIa complex and antigen G18, but not glycoprotein Ia, contained sites with a stronger affinity for calcium than has EDTA at pH 7.4. Immunoprecipitates of the separate glycoproteins IIb and IIIa both bound calcium in the same manner as the glycoprotein IIb-IIIa complex. As another approach, platelet-rich plasma was incubated with 45Ca2+ prior to crossed immunoelectrophoresis of the solubilized platelets. A single immunoprecipitate was weakly labelled. This did not correspond to any of the immunoprecipitates which were visible after staining with Coomassie blue. The labelling of this antigen was markedly increased when the platelet-rich plasma had been preincubated with EDTA and in this case a weak labelling of the glycoprotein IIb-IIIa precipitate also became apparent. No increased incorporation of calcium occurred in any of these immunoprecipitates when the platelets were aggregated with ADP in the presence of 45Ca2+.

Heparin-binding platelet proteins demonstrated by crossed affinity immunoelectrophoresis.

Platelet proteins that interact with heparin were studied using crossed affinity immunoelectrophoresis. Platelet proteins solubilized in Triton X-100 were applied to crossed immunoelectrophoresis against anti-platelet antibodies, and an intermediate gel containing heparin covalently linked to Sepharose 4B was inserted. Six immunoprecipitates were absent or showed an altered position compared to control immunoplates, indicating that the corresponding antigens were bound to the immobilized heparin. These precipitates represented platelet factor 4, thrombospondin, glycoprotein Ib, and three antigens termed G4, 17 and 25. The subcellular location of the heparin-binding proteins was either in the surface membrane (glycoprotein Ib and the antigens 17 and 25), or in the alpha-granules (platelet factor 4, thrombospondin and G4). both forms of platelet factor 4 appearing after crossed immunoelectrophoresis, i.e. a line-form and a peak-form, bound strongly to the heparin. Glycoprotein Ib showed a weak binding whereas its proteolytic split product glycocalicin did not significantly bind to the heparin in the present system. It is concluded that the platelets contain at least six heparin-binding proteins which are present on the cellular surface or are able to be exposed to the extracellular medium after the release-reaction has occurred.

Demonstration of a new glycoprotein Ib-related component in platelet extracts prepared in the presence of leupeptin.

The water-soluble protein glycocalicin is generated during platelet lysis by a proteolytic attack on the integral membrane glycoprotein GP Ib. However, only small amounts of glycocalicin are formed when platelets are solubilized by 1% Triton X-100. Crossed immunoelectrophoresis of such extracts using an antiserum to glycocalicin, shows a continuous immunoprecipitate consisting of two peaks, one representing glycocalicin and the other GP Ib. When leupeptin was present during solubilization, subsequent immunoelectrophoresis revealed yet another GP Ib-related component represented by a third, slow-migrating peak of the immunoprecipitate. During incubation of platelets with dibucaine followed by solubilization in the presence of leupeptin, a gradual transformation of this new form of GP Ib into the previously defined one took place prior to the formation of glycocalicin. An increase followed by a decrease in the agglutination response of the platelets to bovine von Willebrand factor occurred concomitant with these transformations. SDS-polyacrylamide gel electrophoresis of Triton X-100 extracts of platelets did not reveal any difference in the size of GP Ib whether or not leupeptin had been present during the solubilization.

A radioimmunoassay for thrombospondin, used in a comparative study of thrombospondin, beta-thromboglobulin and platelet factor 4 in healthy volunteers.

A radioimmunoassay was developed for the platelet alpha-granule protein thrombospondin; concentrations of thrombospondin as low as 3 ng ml-1 could be measured. There was no interference from other components of human biological fluids and no crossreactivity with beta-thromboglobulin (beta-TG) or platelet factor 4 (PF4). Plasma samples were stable when stored at -20 degrees C. Normal human plasma contained 105.0 +/- 31.0 ng thrombospondin ml-1 compared with beta-TG concentrations of 37.2 +/- 10.9 ng ml-1 and PF4 concentrations of 14.7 +/- 10.1 ng ml-1 when samples were carefully taken into a platelet inhibitor cocktail and processed at 0-4 degrees C. Release of thrombospondin during clotting of blood occurred at the same time as that of beta-TG and PF4 and resulted in a serum concentration of 17.5 +/- 5.5 micrograms ml-1. Assay of whole blood gave a platelet thrombospondin content of 89.1 +/- 28.3 ng/10(6) platelets. The concentration in normal urine fluctuated widely from 3 to 22.5 ng ml-1, and was unrelated to urine flow. The half-life of thrombospondin in vivo was about 9 h, much longer than that of either beta-TG or PF4. Unlike PF4, it was not released into the blood following an intravenous heparin injection. Bovine, ovine, canine and porcine sera contained thrombospondin which crossreacted immunologically with the human molecule; these species would be suitable animal models for the study of thrombospondin and its value as a platelet release marker.

Platelet factor XIII is an active enzyme after solubilization and crossed immunoelectrophoresis.

Crossed immunoelectrophoresis of platelets against antiplatelet antibodies has proved to be a valuable tool in the study of platelet proteins (1-8). The advantage of this separation system is that the proteins are separated under nondenaturating conditions and thus to some extent would be expected to maintain their functional properties. Previously, the binding of several proteins to immobilized thrombin (5) and immobilized heparin (9) during crossed immunoelectrophoresis of platelet proteins solubilized in a Triton X-loo-containing buffer has been described. Furthermore, it has been demonstrated that fibrinogen is able to bind to immunoprecipitates containing the glycoprotein IIb-IIIa-complex (7). These studies indicate that the proteins contained in the immunoprecipitates represent biologically active entities. In the present study we provide direct evidence for this by demonstrating enzymatic activity associated with the immunoprecipitate containing Factor XIII in immunoplates obtained after crossed immunoelectrophoresis of solubilized platelets against anti-platelet antibodies.

Diagnosis of heterozygotes in Glanzmann's thrombasthenia.

A study of a family with a propositus suffering from classical thrombasthenia type I has shown that the new immunochemical methods detect heterozygotes with high reliability. There was no overlapping between heterozygotes and normals, and the concentration of the glycoproteins IIb-IIIa-complex is remarkable constant around 50-60% in the heterozygotes. Furthermore, heterozygotes as a group show an increased bleeding tendency.

Further studies on the interaction between thrombin and GP Ib using crossed immunoelectrophoresis. Effect of thrombin inhibitors.

The platelet surface protein GP Ib (glycocalicin-related protein) has been shown to be retarded by thrombin-Sepharose 4B in a crossed immunoelectrophoresis system. The interaction between GP Ib and thrombin was abolished when thrombin was blocked either at the active serine site with tosyl-lysine-chloromethyl-ketone (TLCK) or phenylmethylsulfonylfluoride (PMSF) or at the fibrinogen binding site (macromolecular binding site) with N-bromosuccinimide (NBS) or heparin, indicating that both sites have to be freely accessible for the retention of the glycocalicin-related protein by thrombin.

Demonstration of variable forms of the platelet factor 4 immunoprecipitate using crossed immunoelectrophoresis.

Proteins with different electrophoretic properties were precipitated by a monospecific antiserum to platelet factor 4 either as a "line" or as a "peak" precipitate. The "line" form seen on crossed immunoelectrophoresis of whole platelets was retained when immobilized thrombin was included in the intermediate gel. The retention was partially abolished when thrombin had been blocked at the active serine site or at the fibrinogen binding site. The "peak" form seen on analysis of material secreted from platelets passed unaffected through thrombin-Sepharose. It is suggested that platelet factor 4 exists in the platelets in a state different from that observed extracellularly after platelet secretion.