Coagulation, anticoagulation, and fibrinolysis

Background

A subset of hemostatic disorders are due to an imbalance in coagulation or fibrinolysis. Disorders of coagulation or fibrinolysis can be either inherited (e.g., hemophilia A) or acquired (e.g., liver disease), and the clinical presentation is typically either bleeding or thrombosis. Laboratory testing is used to diagnose and to monitor the therapy of patients with these disorders. Additionally, a growing list of anticoagulant medications need laboratory testing to adjust the dose or to assess the risk for bleeding or thrombosis.

Content

This chapter describes laboratory testing for disorders of coagulation and fibrinolysis. Prothrombin time (PT), activated partial thromboplastin time (aPTT), thrombin time, and fibrinogen are routine tests in coagulation laboratories and are used in the initial evaluation of bleeding or thrombosis. Mixing studies and specialized testing of the coagulation and fibrinolytic pathways are used in an algorithmic fashion to complete the diagnostic workup. d -Dimer testing is a marker of fibrinolysis used to assess disseminated intravascular coagulation or to exclude venous thromboembolism (VTE). Thrombotic disorders may be investigated with protein C, protein S, antithrombin, factor V Leiden (FVL) mutation, prothrombin G20210A mutation, and lupus anticoagulant (LAC) tests. Laboratory testing of patients on a variety of anticoagulant medications are also described.

Preanalytic variables for coagulation-based assays

Control of preanalytical issues in coagulation testing is paramount for good laboratory performance (see also “Hemostasis Testing” section in Chapter 5 ). In addition to the common issues of hemolyzed, icteric, or lipemic samples, some preanalytical factors of particular importance in coagulation testing include (1) clotted specimens, (2) improper blood-to-anticoagulant ratio, and (3) contamination with saline, heparin, or other anticoagulants. Traumatic venipuncture, activation of coagulation within the collection device, or improper mixing of the anticoagulant with blood may result in clotting, which consumes coagulation factors, making testing unreliable. Blood for coagulation testing should be collected by standard venipuncture techniques into 109 mmol/L (3.2%) trisodium citrate, such that the final proportion of blood to anticoagulant is 9 : 1. Blood is commonly collected into commercially available tubes with prealiquoted trisodium citrate and a line indicating the appropriate volume of blood to be drawn. Collection of a volume of blood less than the recommended volume (“a short draw”) will result in excess anticoagulant compared to plasma and prolonged clotting times. Likewise, samples with a high hematocrit (>55%) will require a decreased volume of anticoagulant because of a lower plasma volume ( Box 81.1 ).

Some coagulation testing, such as activated clotting time (ACT), may be performed on whole blood; however, most routine clot-based assays are performed on plasma, separated from the cellular components by centrifugation. Plasma prepared from citrate tubes is referred to as citrated plasma. Lipemia, hemolysis, and icterus are interferences in serum and plasma, affecting both chemistry and coagulation tests ; however, clot-based tests are sensitive to additional preanalytic factors (see Points to Remember box). To avoid interference from phospholipid, platelet poor plasma (PPP) with a platelet count less than 10 × 10 ⁹ /L is prepared by centrifugation, typically at 1500 × g for at least 15 minutes. Higher speeds and shorter times have been used to prepare plasma for PT, aPTT, and thrombin time (TT), because these tests may not be affected by platelet counts up to 200 × 10 ⁹ /L; this rapid processing is not suitable for heparin assays or LAC testing. If this rapid centrifugation approach is used, the coagulation tests should be performed immediately to prevent release of platelet phospholipid from activated platelets into the plasma. Ultimately, the laboratory should confirm that PPP is produced or that their method of plasma preparation does not affect coagulation results. Reagent phospholipids are important for the spatial orientation of coagulation molecules, so exogenous phospholipid may have a significant impact on the aPTT. Centrifugation and testing of the derived plasma should occur as soon as possible, usually within 24 hours for PT, or within 4 hours for aPTT or other clot-based tests. Plasma should be stored at room temperature for PT but may be stored at either room or refrigerated temperatures (2 to 8 °C) for aPTT. Whole-blood samples should be stored at 18 to 24 °C, whereas refrigerated temperatures should be avoided because of possible “cold activation” of factor VII. Refrigeration of whole blood also decreases factor VIII and VWF and may cause the misdiagnosis of hemophilia A or VWD. If the sample is centrifuged and the plasma aliquoted, cold storage may be acceptable for tests other than PT. Samples for monitoring unfractionated heparin (UFH) therapy should be centrifuged within 1 hour to avoid neutralization of heparin by platelet factor 4 (PF4) released from platelets. When coagulation testing is delayed beyond 24 hours, the plasma should be separated from the cells and kept below −20 or at −70 °C for longer storage. The freeze–thaw cycles disrupt platelet membranes, so PPP is especially important for frozen plasma. Therefore double centrifugation is the best practice to avoid this phospholipid contamination when freezing plasma. Household-grade freezers with auto-defrost cycles are not suitable. Frozen samples should be rapidly defrosted at 37 °C and then mixed to resuspend any precipitate that may contain coagulation proteins.

Optical versus mechanical end point detection

Clot-based tests (e.g., PT/INR, aPTT, and TT) detect the time interval from initiation of coagulation to clot formation. Detection of clot formation as an end point has been accomplished in a number of ways. Early methods used a tilt-tube technique that depended on visually identifying clot formation in plasma samples. A water bath was necessary to keep the temperature at 37 °C. Currently, this time-intensive manual method is used only with international reference thromboplastins.

As a result of high-volume testing, most coagulation testing is now performed on automated instruments that control the temperature of the reaction and detect end points by use of any one of several methods. Most methods detect either a change in physical/mechanical properties or a change in the light transmission produced by polymerized fibrin. Numerous approaches for mechanical end point detection have been developed. One mechanical method consists of a metal ball at the bottom of a sample cuvette that is sent into a back-and-forth motion by a magnet; the end point is detected when fibrin monomers polymerize into fibrin strands and impede the motion of the ball. Another mechanical detection system uses a magnet to hold a ball to the side of a rotating cuvette until fibrin strands physically displace the ball. Optical methods (usually nephelometric but occasionally turbidimetric; see Chapter 16 ) use the decrease in light transmission or increased light scatter as fibrin monomers are polymerized into fibrin strands. Optical end points may occur at preset thresholds or use the kinetics, such as maximum acceleration of fibrin polymerization, to define end points. Light sources have traditionally been halogen lamps or lasers, but newer instruments may use light-emitting diodes (LEDs) that increase longevity and allow measurement at wavelengths that have less overlap with interfering substances. A potential advantage of mechanical over optical end point detection is less interference from substances, such as hemoglobin, bilirubin, or lipid that interfere with optical methods. When an end point cannot be detected with an instrument using an optical end point method, laboratories should have a protocol, such as a backup mechanical method or a send-out laboratory, to obtain a result.

Chromogenic assays

Chromogenic assays have been used to bypass the complexity of the clotting cascade, including the effects of elevated coagulation factors, anticoagulant medications, and LACs (discussed later). In these assays, serine proteases (such as thrombin, factor Xa, factor IXa, factor XIa, and factor XIIa) of the clotting cascade cleave oligopeptide substrates, releasing chromogens that are then detected optically, most commonly at a wavelength of 405 nm:

The chromogenic method is commonly applied to assays that measure (1) factor VIII, (2) factor IX, (3) antithrombin, (4) protein C, (5) factor X, and (6) heparin concentration (anti-Xa method). However, numerous other applications are possible.

Prothrombin time

The PT is a clot-based assay that reflects the activity of extrinsic and common pathway factors of the coagulation cascade. The PT cannot be compared across different laboratories because the specific thromboplastin/instrument combination determines the responsiveness of the test. The INR, which is derived mathematically from the PT, allows harmonization of the PT across laboratories for the purpose of monitoring vitamin K antagonists (VKAs) (see “Therapy With Vitamin K Antagonist”).

The PT is performed on PPP at 37 °C. It is initiated by the addition of a thromboplastin reagent containing TF, calcium, and phospholipid. The TF in the reagent initiates the extrinsic pathway of the coagulation cascade (see Fig. 81.1 ), the phospholipids provide a surface for assembly of coagulation factors, and the calcium chloride counteracts the binding of plasma calcium by citrate, making calcium available for the coagulation process. The timer is started when reagent is added, and the end point occurs when fibrin monomer polymerization is detected.

The addition of TF activates the extrinsic pathway much like the natural process; however, clotting occurs before the intrinsic pathway is significantly propagated, as occurs with in vivo coagulation. Consequently, the PT is not sensitive to factor VIII or other intrinsic coagulation factors, and PT could be considered a test of the initiation phase of coagulation.

The PT is used to identify deficiencies or inhibitors of factors VII, X, V, and II and fibrinogen. Expected results of clotting assays, including PT, are shown in Table 81.1 for key inherited and acquired disorders. The sensitivity of PT reagents to the deficiency of coagulation factors varies with the specific reagent and instrument combination. Therefore it is useful for laboratories to determine the sensitivity of a given PT system to factors VII, X, V, and II (see Fig. 81.2 for illustration of the principle, although this example uses the aPTT test rather than PT). Alternatively, manufacturers may provide data regarding responsiveness to coagulation factors. If a reagent is overly sensitive to decreases in a factor (e.g., 0.5 IU/mL factor VII), the laboratory may detect clinically insignificant prolongations of the PT. Needless laboratory evaluations and delays in surgical interventions may be avoided if reagents are selected carefully.

The PT/INR has been used to monitor VKA therapy because it is sensitive to vitamin K–dependent factors, II, VII, and X. Monitoring during bridging or conversion of patients from heparin therapy to therapy with VKAs is made possible by the addition of heparin-neutralizing substances. VKA therapy and its monitoring are addressed in greater detail later.

Activated partial thromboplastin time

The aPTT is a clot-based assay initiated by activation of contact factors and reflects the activity of the intrinsic and common coagulation pathways (see Fig. 81.1 ). Activation of contact factors (high molecular weight kininogen [HMWK], prekallikrein, and factors XII and XI) is achieved by the addition of one of a wide variety of activators. With the exception of ellagic acid, these substances have negatively charged surface responsible for contact factor activation. Activators including kaolin, celite, and micronized silica have been used extensively when mechanical end points are determined by coagulometers (see earlier); however, if optical detection is used, micronized silica or the soluble chemical activator ellagic acid is used because they do not interfere with light transmission.

The aPTT is carried out in two stages. First, reagent containing a contact activator and phospholipids is added and incubated with the citrated plasma at 37 °C for several minutes (varies with the method). Second, the plasma is recalcified with calcium chloride, which is required for subsequent activation of downstream coagulation factors and the timer is started. The timer is stopped when the end point is detected.

Although aPTT, activated partial thromboplastic time, implies the addition of thromboplastin, this is not the case. This ambiguity is a result of use of the term partial thromboplastin by Langdell and associates to describe a group of reagents that produced clotting that was less rapid in hemophiliac plasma than in normal plasma. This was contrasted with “complete” thromboplastin reagents used in the PT assay, which did not discriminate normal and hemophiliac plasmas. Partial thromboplastin activity was achieved by replacing thromboplastin with phospholipid preparations, sometimes called cephalin; however, the overall effect was to dilute the thromboplastin activity, making the contribution of the intrinsic pathway measurable. Subsequently, the addition of activators led to the aPTT that we recognize today.

The aPTT is sensitive to activities of the intrinsic and common pathway factors (1) HMWK, (2) prekallikrein, (3) XII, (4) XI, (5) IX, (6) VIII, (7) X, (8) V, (9) II, and (10) fibrinogen. Common uses of the aPTT include monitoring of heparin therapy and screening for deficiencies or inhibitors of intrinsic and common pathway factors. The use of aPTT for monitoring heparin therapy is discussed later. The response of aPTT reagents to heparin, factor deficiency, specific factor inhibitors, and LACs varies widely. If a clinical laboratory intends to use aPTT to detect factor deficiencies, it is important to know the threshold of factor deficiency that prolongs the aPTT (see Fig. 81.2 ). The response of aPTT to factors VIII, IX, and XI is particularly important because of the prevalence of hemophilias and their clinically significant bleeding risk. In general, the desired response threshold (i.e., the level of factor deficiency associated with prolongation of aPTT) to factors VIII, IX, and XI should be approximately 30 to 40 IU/dL. Alternatively, manufacturers may provide data regarding responsiveness to coagulation factors. Very sensitive reagents may create unnecessary laboratory follow-up or treatment delay, whereas insensitive reagents may not detect mild factor deficiency.

The aPTT is prolonged in a variety of clinically significant and insignificant scenarios. Table 81.1 summarizes some causes of prolonged aPTT and/or PT. Six potential causes of isolated prolongation of aPTT include (1) a procoagulant deficiency or deficiencies that may be associated with a bleeding history, (2) a contact factor deficiency without bleeding risk (XII, prekallikrein, HMWK), (3) a specific inhibitor acquired as an alloimmune or autoimmune phenomenon (e.g., factor VIII inhibitor), (4) a nonspecific inhibitor such as a LAC, (5) a medication effect or contamination (e.g., heparin, direct thrombin inhibitor), and (6) a spurious result. A prolonged aPTT reflects clinically significant deficiencies of coagulation factors (factors XI, IX, and VIII) and is an important test for identifying these deficiencies. However, deficiencies of several intrinsic coagulation factors, HMWK, prekallikrein, and factor XII, prolong the aPTT but are not associated with bleeding.

Fibrinogen (clauss and derived method)

Coagulation ultimately depends on the conversion of fibrinogen to fibrin monomer by thrombin (see Fig. 81.1 ). The Clauss method is the most commonly used method to measure fibrinogen concentration. Initial dilution of plasma (usually 1:10) serves to dilute fibrinogen, and then excess thrombin (final concentration 30 to 100 U/mL) is added to the diluted plasma. With the enzyme (thrombin) at saturating concentrations and the substrate (fibrinogen) at limiting concentration, the rate of fibrin formation depends on the concentration of fibrinogen:

The fibrin end point is used to determine fibrinogen concentration from a calibration curve. Five dilutions of the fibrinogen calibrator are made, and a calibration curve is generated by plotting fibrinogen concentration against the rate of fibrin clot formation.

Another method of fibrinogen determination is PT-derived fibrinogen, also known as PT-fibrinogen. The fibrinogen concentration is proportional to the maximal change in absorbance in an optical PT measurement. A calibrator is used to create a calibration curve relating the change of absorbance of PT to the fibrinogen concentration. The PT-fibrinogen assay may overestimate fibrinogen concentration in patients with DIC, patients on fibrinolytic therapy, and many patients with dysfibrinogenemia. In addition, some fibrinogen calibrators contribute more turbidity and may not be optimal for PT-fibrinogen measurements. The Clauss method is considered by many to provide a more meaningful fibrinogen measurement because fibrinogen degradation products are not detected, whereas PT-fibrinogen methods are more sensitive to fibrinogen degradation products. FDPs prolong some clot-based tests by interfering with fibrin monomer polymerization, but they may also act as anticoagulants and contribute to hemorrhage in DIC.

Mixing studies

Mixing studies are performed on abnormally prolonged clot-based assays such as aPTT and PT. The purpose of mixing studies is to determine whether prolonged clotting times are due to factor deficiency or inhibitor activity. Mixing studies are useful for guiding the coagulation work-up, but they lack clinical sensitivity and specificity. Hence, the results of mixing studies are followed with LAC tests, factor assays, and possibly specific functional inhibitor studies for confirmation. Inhibitors are categorized as nonspecific or specific types. Nonspecific inhibitors (e.g., heparin, LAC) have activity against multiple procoagulants, whereas specific inhibitors are directed at a single factor. LACs are a heterogeneous group of immunoglobulins with phospholipid-dependent activity against coagulation factors; they are discussed in greater detail later. Specific factor VIII inhibitors are usually seen as alloantibodies in the setting of hemophilia A treated with factor VIII concentrates, but they may be seen in non-hemophiliac patients as an autoantibody. Although rare, specific inhibitors may occur against any factor, and a high degree of clinical suspicion is needed when bleeding is otherwise unexplained.

Mixing studies are useful when clotting time (e.g., aPTT, PT) is unexpectedly prolonged outside the reference interval. They are also recommended for LAC testing (see section on DRVVT and LAC studies later) to confirm the inhibitor effect. Patient plasma is mixed in a 1 : 1 ratio with NPP using the same anticoagulant, and then the clotting assay is repeated immediately . NPP should contain 100 (±20 IU/dL) of all factors. If a factor deficiency in the patient plasma exists, the NPP supplies the factors needed to correct the clotting time back into the reference interval. If an inhibitor is present, the mixed specimen will not correct the clotting time into the reference interval.

The utility of mixing studies largely depends on the patient population and the clinical setting. For example, aPTT mixing studies are performed more frequently than PT mixing studies because LACs, which are common, affect aPTT to a much greater extent than PT. When aPTT is prolonged and mixing studies fail to adequately correct into the reference interval, an inhibitor is suspected. If the patient is asymptomatic or has a history of thrombosis, the next step should be to look for LAC rather than measure factor VIII, IX, and XI concentrations. Conversely, if the patient has a bleeding presentation, a specific inhibitor may be suspected and factor activities should be measured. PT mixing studies may be less helpful because the differential diagnosis of an isolated prolongation of PT is limited.

Immediate aPTT mixing studies that correct to the reference interval require incubated mixing to assess for time-dependent inhibitors. In such studies, the patient’s plasma is mixed with NPP and they are incubated together for 1 or 2 hours at 37 °C. The aPTT is then repeated and compared with a control. The control consists of patient plasma and NPP incubated at 37 °C separately, followed by mixing and aPTT measurement. This control step is important because it controls for loss of labile factors during the incubation process ( Fig. 81.3 ). One approach to interpretation is to consider a 10% difference between the test and the control to indicate a time-dependent inhibitor. Acquired alloantibodies to factor VIII are frequently time dependent; however, up to 15% of LAC measurements are also time dependent. Considering the high prevalence of LAC relative to the low prevalence of acquired factor VIII inhibitors in hemophilia patients (1 to 2 per million of population per year), laboratories are more likely to encounter time-dependent LACs than time-dependent acquired factor VIII inhibitors ( Table 81.2 ).

The performance characteristics of mixing studies depend on the factor and phospholipid content of NPP, titer and strength of inhibitor, presence of multiple factor deficiencies, heparin contamination, and criteria for interpretation. Weak inhibitors or very sensitive aPTT reagents may cause substantial difficulties. Phospholipid-poor assay reagents are more sensitive to LACs, so perhaps weaker LACs are detected before mixing, but they are rapidly diluted out by 1 : 1 mixing. If the aPTT is only minimally elevated (e.g., < 5 seconds above the reference interval), 1 : 1 mixing studies are not useful, because the clotting times may correct despite the presence of an inhibitor. Some laboratories use an alternative mixing study with four parts patient plasma and one part NPP to improve sensitivity for these weak inhibitors. Mixing studies certainly are not quantitative, and the clinical significance of these “weak” inhibitors is not known. NPP should be prepared from at least 20 apparently healthy individuals to ensure 100 (±20 IU/dL) of all factors. It is also important that NPP be platelet poor because contamination, especially in frozen aliquots, may contribute to phospholipid content that will neutralize LACs. In some instances, 1 : 1 mixing may not correct when multiple factors are deficient, as may be seen in samples from patients who are supratherapeutic on VKAs. Heparin, direct thrombin inhibitors, or direct anti-Xa inhibitors also may behave like inhibitors in mixing studies. Additional clinical correlation and laboratory investigations may be needed to exclude these medications.

Interpretation of mixing studies is not always straightforward, and criteria for interpretation may vary somewhat across laboratories. Although laboratories have commonly considered correction of the immediate 1 : 1 mix to be within the reference interval (+2 standard deviations), many variations are used (e.g., +3 standard deviations of the mean aPTT). Because addition of NPP substantially dilutes weak LACs in test plasma, the requirement for correction into the reference interval may be too stringent and may result in false-negative interpretations. Rosner and colleagues suggested an index, and Chang and coworkers demonstrated improved performance when using a percent correction and a 4 : 1 mix. The performance of the percent correction and Rosner index has been verified for LAC-sensitive aPTT 1 : 1 mixing studies in a single laboratory. Currently, there is no uniformity in the performance of mixing studies, nor are there uniform criteria for interpretation; ultimately, the laboratory director must decide the appropriate approach and each result should be accompanied with an interpretation.

Factor assays

The most commonly performed factor assay is the one-stage factor assay. One-stage assays are based on the aPTT or the PT, depending on which factor is being measured. The assay is performed on dilutions of patient plasma that are then mixed 1 : 1 with factor-deficient plasma. Factor-deficient plasma is deficient in a single factor but contains essentially 100 IU/dL of all other factors. Plasma from patients with severe hemophilia has been used historically; factor-deficient plasmas manufactured by immunodepletion methods are now commercially available. Mixing of patient plasma with deficient plasma ensures that clotting times are dependent on the factor being measured. aPTT-based one-stage assays are used to measure factor activity of intrinsic factors, while PT-based assays are mostly used to measure factor VII and factors of the common pathway. Factor activity is then determined from a calibration curve created by plotting the clotting time (PT or aPTT) in seconds versus the concentration of factor ( Fig. 81.4 ).

The aPTT-based one-stage factor VIII assay consists of a dilutional series of patient plasma samples that are then mixed with factor VIII–deficient plasma. The aPTT changes minimally when factor concentration is near 100 IU/dL; thus the initial dilution (typically 1 : 10) prolongs the aPTT into a steeper part of the relationship between clotting time and factor activity (see Fig. 81.2 ). At least three dilutions are needed because clot-based tests are subject to interference from inhibitors. The aPTTs of the dilutional series are used to extrapolate factor VIII activity from the calibration line. The 1 : 10 dilution is considered the starting point for comparison with the calibration line, which also starts with a 1 : 10 dilution. Historically, calibrators were created locally by pooling large numbers of donor plasmas, and factor VIII was assumed to be 100%. However, studies demonstrated differences in factor VIII between populations. Currently, calibrators are assigned values in international units (IU) traceable to the international standard plasma established by the World Health Organization (WHO). The amount of factor VIII in 1 mL of fresh pooled plasma collected from a large number of donors was defined as 1 international unit when this unitage and the 1st WHO International Standard for FVIII was established. A calibration line is created by diluting the calibrator in a manner similar to the patient samples. Various mathematical transformations, most commonly log transformation, are employed to create a straight calibration line.

Various inhibitors such as lupus anticoagulant, anticoagulant medication, or specific factor inhibitor (e.g., factor VIII inhibitor), cause interference with the one-stage factor assay. Factor activity should be determined with at least three dilutions (e.g., 1:10, 1:20, and 1:40) to enhance accuracy and allow for detection of inhibitors by assessing parallelism between patient data and the calibration curve ( Fig. 81.5 ). Serial dilutions, when parallel, should return the same activities after multiplying by the dilution factors. If increasing activities are obtained as dilutions increase, the patient’s curve will be nonparallel to the calibration curve, and an inhibitor is suspected. Additional higher dilutions can be used in an attempt to dilute an inhibitor. When at least two consecutive dilutions produce similar activities, the inhibitor effect has been diluted out; the activity is then reported using the average of these activities, after correction for dilution. It may not be possible to dilute out strong or high titer inhibitors. The one-stage factor assay depicting elevated, reduced, and inhibited assays is shown in Fig. 81.5 .

The two-stage assay is only used by specialized laboratories as an alternative method to measure factor VIII activity. The first stage involves the production of factor Xa, and the second stage determines the amount of factor Xa produced. The first stage contributes exogenous coagulation factors needed for formation of the prothrombinase complex, except for factor VIII. In this first stage, the formation of Xa depends on the amount of factor VIII in the patient’s plasma. In the second stage, NPP is added to the reaction as a source of prothrombin and fibrinogen so that fibrin clotting will occur. The clotting time will be dependent on factor Xa generated in the first stage, which is dependent on factor VIII concentration; thus the clotting time in the second step is proportional to the amount of factor VIII in the patient sample. A calibration curve is then used to determine the factor VIII activity.

Chromogenic factor VIII assays are based on the two-stage assay ( Fig. 81.6 ). Similar to the two-stage assay, the first stage allows for the generation of factor Xa by the addition of reagents needed to the form intrinsic tenase complex. Calcium, phospholipid, excess purified factor IXa, and excess factor X are added such that the amount of factor Xa generated during this first stage depends on the amount of factor VIII in the patient sample. The second stage measures the absorbance developed from the enzymatic release of p- nitroaniline from a chromogenic substrate by Xa. A direct thrombin inhibitor is often added to the second stage to decrease cleavage of the chromogenic substrate by thrombin (factor IIa). A calibration curve relates the change in absorbance at 405 nm to factor VIII activity in the specimen.

Important differences have been observed in factor VIII activity depending on the method used. Two-stage assays and chromogenic assays are less sensitive to LACs because of the higher initial plasma dilutions employed. Some chromogenic assays include thrombin to fully activate factor VIII to VIIIa; others rely on feedback activation of factor VIII by thrombin generation analogous to the in vivo propagation phase of coagulation. One-stage assays sometimes will overestimate factor VIII activity compared with two-stage or chromogenic assays. ^, It is important to keep this in mind because a one-stage assay may not exclude a mild hemophilia A, and clinical correlation with bleeding symptoms and family history may be needed. The incubation times in the first stage of chromogenic assays are not uniform across different commercially available assays, and longer incubation times allow for the detection of some types of mild hemophilia A. Less frequently, two-stage assays will recover more factor VIII activity than one-stage assays.

Factor inhibitor assays

Inhibitors are quantified with a functional inhibitor assay. The Bethesda assay established the framework for functional inhibitor assays and defined the Bethesda unit (BU). The primary focus of this assay and unit was to create uniformity in the measurement of factor VIII inhibitors in patients with hemophilia A. This is accomplished by creating an incubated mix of patient plasma (containing the inhibitor) with NPP (containing ∼100 IU/dL factor VIII activity), and then measuring the remaining factor VIII activity. Strong inhibitors reduce the resultant factor VIII activity more than weak inhibitors do. Titration of this effect is possible using serial dilutions of patient plasma before mixing.

The Bethesda assay begins with undiluted patient plasma or dilutions of patient plasma with imidazole buffer. The patient sample is then mixed with NPP. The mixed sample is incubated for 2 hours at 37 °C, and then factor VIII is assayed. A reference mix, consisting of a 1 : 1 incubated mix of imidazole buffer and NPP, is crucial to control for loss of labile factors during incubation. The residual factor activity is calculated as follows:

Finally, the residual factor activity is translated into Bethesda units. One BU per milliliter is defined as the inhibitor activity producing a residual factor activity of 50% of the starting concentration of factor VIII in the reference mix. A value of 2 BU/mL is then the inhibitor activity producing a residual factor activity of 25%. A log-linear graph of factor activity and BU/mL is drawn according to this definition. For ease of use, a chart relating residual factor to Bethesda unit per milliliter is constructed. Parallelism of dilutions should be expected to confirm the precision of the result.

Nijmegen modifications of the Bethesda assay improve specificity near the low analytical limit of detection ( Fig. 81.7 ). The Nijmegen modifications include replacing imidazole buffer with factor-deficient plasma for dilutions and buffering the NPP at pH 7.4. Both of these changes reduce loss of factor VIII during incubation. Buffering the NPP prevents increasing pH, and diluting with factor deficient plasma normalizes protein concentrations in the reference plasma. An additional modification has included heating patient plasma to denature residual factor activities. Many laboratories in the North American Specialized Coagulation Laboratory (NASCOLA) survey were found to be using a hybrid of the classic Bethesda and Nijmegen procedures in which a commercially available buffered NPP was used, but imidazole buffer was used for dilutions of patient plasma. The high coefficient of variation (CV) and the variation in laboratory procedures make standard treatment guidelines based on BU/mL difficult. To assay other factor inhibitors, the factor VIII deficient plasma used in the assay is replaced with the corresponding factor deficient plasma.

The behaviors of inhibitors in laboratory tests are not uniform. Although most inhibitors demonstrate an expected increase in residual factor VIII activity with increasing dilutions (type I inhibitor), some inhibitors (type II inhibitors) lack this relationship. Type II inhibitors lack parallelism to the calibration curve and may be difficult to titer. In these cases, dilutions of patient plasma corresponding to approximately 50% residual activity are used to estimate BU/mL. In addition, some factor VIII inhibitors increase factor VIII clearance but do not affect factor VIII activity. These non-neutralizing factor VIII inhibitors cannot be detected with a mixing study, nor can they be titered with a functional inhibitor assay. ELISA assays for detecting factor VIII antibodies can be used to detect non-neutralizing types of inhibitors.

Factor XIII assays

Plasma factor XIII is a zymogen that is activated by thrombin in a calcium-dependent reaction. The active enzyme catalyzes the covalent cross-linking of fibrin molecules and polymers, producing a stable clot. Factor XIIIa is a transglutaminase that links glutamine residues to lysine residues via the transfer of an acyl group and the release of ammonia. Because noncovalently linked fibrin polymer is sufficiently stable to support the end point in these tests, factor XIIIa activity is not reflected in clot-based assays such as aPTT and PT.

The urea solubility screen, or a variation thereof, is a common screening method for factor XIII deficiency. Before the actions of factor XIIIa take place, fibrin polymer structure is held together by hydrogen bonds and is soluble when subjected to weak alkalis and acids or to high concentrations of solute such as urea. This is the basis of factor XIII screening methods. In the urea solubility screen, fibrin clot is formed from citrated patient plasma and then is suspended in 6 mol/L urea. The clot is visually observed at intervals for dissolution. Covalently linked fibrin clots do not dissolve. Noncovalently linked fibrin clots are dissolved after 1 to 3 hours in severe factor XIII–deficient plasmas. A fibrin clot prepared from NPP serves as the control. This method detects only very severe factor deficiency, as very little factor XIIIa activity is required to stabilize a fibrin clot.

Data from UK National External Quality Assessment Scheme (NEQAS) Surveys demonstrated variability in screening methods. Fibrin clots were achieved with calcium alone, calcium and thrombin, or thrombin alone; moreover, when thrombin was used, the concentration was not uniform. Acetic acid, urea, and monochloroacetic acid were used as solvents. Altogether, 15 different combinations of solvents and clot preparations were used, contributing to variable responses. Further evaluation of plasmas spiked with factor XIII confirmed that a combination of thrombin and acetic acid provided the best sensitivity, which was in the interval of 1 IU/dL to 5 IU/dL factor XIII. Positive solubility screens for factor XIII need confirmation by quantitative methods that may be factor XIII antigen assays or factor XIII activity assays.

Thrombin time

TT is a clot-based assay reflecting two steps: conversion of fibrinogen to fibrin monomer by thrombin and polymerization of fibrin monomers. A low concentration of thrombin (final concentration of 0.1 to 0.3 U/mL depending on source of thrombin) is added to citrated plasma, fibrin is generated, and the time it takes to form the clot is measured.

Note that TT is different from the Clauss fibrinogen measurement: in the Clauss fibrinogen method, patient plasma is diluted, and a higher concentration of thrombin is used. Although TT is still sensitive to low fibrinogen concentrations, TT is more sensitive to thrombin inhibitors, abnormal fibrinogen (dysfibrinogen), and substances that interfere with fibrin polymerization compared with the Clauss fibrinogen assay. Thrombin inhibitors prolong TT and are commonly encountered as therapeutic anticoagulants, including heparin and direct thrombin inhibitors. Because these inhibitors may interfere with other clot-based tests, many laboratories use TT as a tool to detect the unexpected presence of these therapeutic agents. If TT is used for this purpose, the laboratory should determine the responsiveness of its TT assay to direct thrombin inhibitors and heparin. TT assays with low concentrations of thrombin (0.1 U/mL) are exquisitely sensitive to subtherapeutic levels of unfractionated heparin (0.2 IU/mL), resulting in TT greater than 120 seconds. TT is also very sensitive to direct thrombin inhibitors like argatroban or dabigatran. TT is less responsive to low molecular weight heparin (LMWH), which produces mild to moderate elevations in TT.

Inhibitors to bovine thrombin and/or factor V occur in some individuals exposed to topical bovine thrombin during surgical procedures. Bovine thrombin combined with a source of fibrinogen, such as cryoprecipitate, is an effective hemostatic agent. However, topical bovine thrombin preparations also contain bovine factor V, so inhibitors to both bovine factors II and V can be produced. Bovine factor V inhibitors may cross-react with human factor V to produce bleeding, but cross-reactivity of bovine thrombin inhibitors with human thrombin is rare. There is, however, prolongation of the TT if bovine thrombin is used for TT assays instead of human thrombin; this will be confirmed by a normal TT using human thrombin.

TT may be prolonged in the presence of structurally abnormal fibrinogens, called dysfibrinogens. Congenital dysfibrinogenemia occurs only rarely, and it is much more common to see acquired dysfibrinogenemia. Individuals with dysfibrinogenemia are typically asymptomatic, but some may be at risk for bleeding or thrombosis. Both bleeding and thrombosis may occur in the same patient. Liver and biliary diseases are common causes of acquired dysfibrinogenemia and produce fibrinogen molecules with increased sialylation of carbohydrate moieties. As a result, an increase in negative charge retards fibrin polymerization and prolongs the TT. TT and the closely related reptilase time are considered initial screening tests for dysfibrinogenemia. The laboratory diagnosis of dysfibrinogenemia is confirmed by finding a discrepancy in the ratio of fibrinogen clotting activity to the fibrinogen antigen concentration.

Prolonged TTs occur in a wide variety of settings and are not specific for dysfibrinogenemia ( Table 81.3 ). For example, substances that prolong TT include paraproteins, fibrinogen degradation products, and high fibrinogen concentrations. Paraproteins affecting TTs may be of any heavy chain and are seen in the clinical setting of Waldenström macroglobulinemia or multiple myeloma. Paraproteins may affect other tests of hemostasis, including aPTT and bleeding time. These patients may even have clinical symptoms of bleeding, but it is not possible to predict bleeding based on any laboratory test. Increased FDPs and increased fibrinogen, by virtue of their structural similarities to fibrin, may prolong the TT by competing with normal fibrin monomers in the polymerization process. High concentrations of FDPs are seen in DIC and after fibrinolytic therapy. High concentrations of fibrinogen are commonly encountered as part of an acute-phase reaction in hospitalized patients. Thus fibrinogen assays are useful in the evaluation of prolonged TT.

Inherited coagulation disorders

The most common inheritable bleeding disorders resulting from coagulation factor deficiencies include (1) hemophilia A (factor VIII deficiency), (2) hemophilia B (factor IX deficiency), and (3) factor XI deficiency (historically referred to as hemophilia C). In the laboratory, deficiencies of the intrinsic pathway produce isolated prolongation of aPTT. Factor XII, prekallikrein, or HMWK deficiency are in the differential diagnosis of isolated prolongation of aPTT, but these contact factor deficiencies do not cause clinical bleeding. LACs prolong the aPTT, but are paradoxically associated with thrombosis. Only rare cases of LAC hypoprothrombinemia syndrome are associated with bleeding.

Inherited factor VII deficiency, or less likely mild factor X, V, or II deficiency, causes isolated prolongation of the PT without prolongation of the aPTT. Factor VII deficiency is the most common of these inherited deficiencies, with factor activities typically in the range of 0 to 20 IU/dL. Factor VII Padua, a factor VII variant, causes variable responsiveness to thromboplastins derived from rabbits, humans, and oxen. Factor VII Padua is probably not associated with a risk for hemorrhage, but it produces interlaboratory discrepancy in PT/INR and factor VII results. Consequently, laboratories using rabbit thromboplastin will find decreased factor VII activity, laboratories using ox thromboplastin will find normal factor VII, and laboratories using human thromboplastin will find intermediate factor VII activity. Factor VII Padua is considered a type 2 deficiency because factor VII activity is decreased (using rabbit reagent), but factor VII antigen is normal. Recognizing factor VII Padua is valuable to explain discrepant laboratory values and to avoid unnecessary treatment or procedure delays.

Clinically significant single deficiencies of factors X, V, II, or fibrinogen prolong both PT and aPTT. aPTT mixing studies show a factor deficiency, and factor activity is markedly decreased. Inherited afibrinogenemia or hypofibrinogenemia also causes prolongation of PT and aPTT, and mixing studies of patient plasma and normal pooled plasma demonstrates a factor deficiency. A functional fibrinogen assay shows markedly decreased fibrinogen. In all cases of inherited deficiency, clinical correlation with a bleeding and family history is needed to make a diagnosis.

Two rare inherited deficiencies of multiple factors are combined factor V and factor VIII deficiency (F5F8D) and vitamin K–dependent clotting factor deficiency (VKCFD). In F5F8D, LMAN1 or MCFD2 mutations disrupt the transport of coagulation cofactors from the endoplasmic reticulum to the Golgi apparatus. F5F8D is inherited in an autosomal recessive pattern and reduces cofactors to a range of 5 to 30 IU/dL. F5F8D causes epistaxis, menorrhagia, and bleeding associated with injury or surgery, but other phenotypic consequences have not been described in humans. VKCFD is caused by mutations in γ-carboxylase or vitamin K epoxide reductase that interfere with the vitamin K cycle. Vitamin K–dependent factors are not γ-carboxylated and do not obtain full activity. Intracranial hemorrhage in neonates has been described that is partially responsive to replacement of vitamin K. Antibiotics that decrease vitamin K–producing gut flora also can exacerbate deficiency and have been associated with reports of hemarthrosis and mucocutaneous hemorrhage. Skeletal abnormalities also have been described, presumably because of lack of γ-carboxylation of proteins involved in bone metabolism.

AT A GLANCE

Isolated Prolongation of Activated Partial Thromboplastic Time

Factor XIII deficiency is a rare disorder of secondary hemostasis with a bleeding phenotype and normal PT, aPTT, TT, and fibrinogen. A urea solubility test demonstrates dissolution of fibrin clot in 1 to 2 hours. Low factor XIII can be confirmed with antigen or activity assays. Platelet function and studies of fibrinolysis are normal.

Factor VIII deficiency (hemophilia A)

Factor VIII deficiency is an X-linked recessive genetic disorder with an incidence of 1 in 5000 live male births. Because one third of mutations occur spontaneously, a family history is not always present. Severity of bleeding correlates with the amount of factor VIII with activities of (1) less than 1 IU/dL, (2) 1 to 5 IU/dL, and (3) greater than 5 to 40 IU/dL associated with severe, moderate, and mild hemophilia A, respectively.

Laboratory studies of individuals with factor VIII deficiency show isolated prolongation of aPTT and normal PT. A normal TT is helpful to exclude heparin contamination as a cause for prolonged aPTT. Mixing studies should help identify a factor deficiency. Factor activities of the intrinsic pathway should be measured to confirm suspicion of an intrinsic pathway deficiency. In a male patient, factor VIII, IX, and XI activity should be measured. Female carriers may have mildly reduced factor VIII activity and mild bleeding symptoms, but investigation of factor XI deficiency and VWD is probably best considered first in females. Because factor VIII is elevated in response to stress and exercise, confirmation of carrier status is best confirmed with molecular studies. Decreased factor VIII activity with normal activity of factors IX and XI is expected in hemophilia A. Patients with normal factor VIII with a one-stage assay, but significantly lower values with a two-stage assay or chromogenic assay, most frequently present clinically like a mild hemophilia and have been referred to as mild discrepant hemophilia A. Up to 40% of mild hemophilia A patients may have discrepant results. ^, Patients with lower factor VIII with the one-stage assay generally have absent or mild bleeding; however, rare patients with a “reverse” discrepancy do have significant bleeding tendency. For these reasons, the World Federation of Hemophilia has recommended performing both the one-stage and chromogenic (or two-stage) assays at diagnosis.

AT A GLANCE

Isolated Prolongation of Prothrombin Time

Patients with other inherited bleeding disorders also present with a low factor VIII. Most VWD patients present with an isolated prolongation of aPTT and low factor VIII activity, so evaluation of VWD is warranted in some cases before a diagnosis is rendered. The distinction of hemophilia A from type 2N VWD is particularly difficult. Type 2N VWD is caused by decreased affinity of VWF for factor VIII, so patients have decreased factor VIII because it is not protected by binding to VWF. The distinction of type 2N VWD and hemophilia A can be accomplished with a careful family history, a VWF:VIII binding assay (which is performed only at very specialized laboratories), or genetic testing. VWD type 2N is inherited in an autosomal recessive pattern, rather than an X-linked inheritance pattern such as in hemophilia A. A family history of hemophilia may be absent in up to 30% of infant males with hemophilia A, because of spontaneous mutations. Genetic testing to identify a specific factor VIII mutation may be useful in prenatal testing of carrier females. Ultimately the family and clinical history is needed to make a confident diagnosis of an inherited bleeding disorder.

Factor VIII activity is used to monitor factor VIII replacement therapy—the mainstay of prophylaxis and of management of an acute bleeding event. When postinfusion recovery of factor VIII activity is lower than anticipated or the patient fails to respond clinically, a factor VIII inhibitor is suspected. The aPTT is prolonged, but in this case, mixing studies suggest an inhibitor. Most factor VIII inhibitors are time dependent ; thus an immediate aPTT mixing study may correct the prolonged aPTT. However, an incubated aPTT mixing study reveals prolongation compared with the appropriate control. Factor VIII inhibitor is quantified with a Bethesda assay, and the titer is used to guide therapeutic decisions. Patients with an inhibitor titer of less than 5 BU/mL may respond to high doses of factor replacement; conversely, patients with higher titers are not likely to respond and require alternative treatment. Anamnestic responses may occur, in which low or undetectable inhibitor titers rise 4 to 7 days after re-exposure to factor VIII. Inhibitors have been reported in 10 to 25% of patients and are associated with certain molecular defects in the F8 (for factor VIII) gene. Most inhibitors occur in severe factor VIII deficiency, but inhibitors may occur in mild disease. Alternative treatments for patients with inhibitors have included replacement with porcine factor VIII or products with factor VIII–bypassing activity, such as recombinant activated factor VIIa concentrate or activated prothrombin complex concentrates. These latter two products activate factor X without factor VIII or IX.