Chapter 19 Investigation of a thrombotic tendency

Introduction to thrombophilia 447

Tests for the presence of a lupus anticoagulant 448

Sample preparation 448

Dilute Russell’s viper venom time 449

Kaolin clotting time 451

Dilute thromboplastin inhibition test 451

Other acquired thrombophilic states 451

Platelet ‘hyperreactivity’ and activation 461

Platelet activation: flow cytometry 461

Homocysteine 461

Markers of coagulation activation 462

Global assays of coagulation 462

Introduction to thrombophilia

Investigations to identify an acquired or inherited increase in thrombotic tendency are frequently carried out in patients who develop venous or arterial thrombosis at a young age, in those who have a strong family history of such events or have thrombosis at an unusual site and in individuals of all ages with recurrent episodes of thromboembolism. In recent years, the utility of these tests, as judged by their ability to alter management, has come under scrutiny. In most cases the results of individual assays have a limited effect on decisions made on the basis of clinical history alone. This is partly because they are initiated in patients who have already had a thrombosis and have thus demonstrated their thrombotic tendency. Nonetheless, there is still a need to identify those individuals whose risk of further thrombosis is sufficiently high to warrant long-term anticoagulation and attention has turned to global tests of thrombotic potential and combinations of single traits as well as details of the clinical history. It should be remembered that many thromboses are almost entirely the result of circumstantial factors; these include trauma, fractures, operations and an acute-phase inflammatory response. Further investigation of coagulation is often unnecessary in these circumstances. The investigations described here are most commonly instituted in venous thrombosis, but some unexplained arterial events, especially in young people or when paradoxical embolism is suspected, are also studied. In general, the contribution of the inherited plasma coagulation factors is less evident for arterial than venous thrombosis because their effect is then often obscured by atherosclerosis.

In this chapter, the investigations to detect an acquired thrombotic tendency are presented first, followed by a simplified battery of tests needed to establish the diagnosis of the more important inherited ‘thrombophilias’. Although the number of coagulation factors known to contribute to a thrombotic tendency has increased greatly in the last few years, it remains clear that not all factors have been identified. Hence, the failure to detect one of the traits described does not imply that the individual’s risk of thrombosis is normal. An acquired thrombotic tendency is common and occurs in many conditions but is usually complex, multifactorial and not easily identifiable by a single laboratory test. The large number of traits identified, often with a small associated relative risk, makes their individual utility equally small. Until the interactions of these numerous factors are more completely understood, the clinical history remains a dominant factor in clinical management. The British Committee for Standards in Haematology has published guidelines on the investigation of inherited thrombophilia.¹

Tests for the presence of a lupus anticoagulant

The lupus anticoagulant (LAC) is an acquired autoantibody found in various autoimmune disorders and sometimes in otherwise healthy individuals.² LACs are immunoglobulins that bind to certain proteins when bound to phospholipid. The effective sequestration of phospholipid can then cause prolongation of phospholipid-dependent coagulation tests such as the prothrombin time (PT) or activated partial thromboplastin time (APTT). The name ‘anticoagulant’ is misleading because, despite the in vitro effects, patients do not have a bleeding tendency. Instead, there is a clear association with recurrent venous thromboembolism, cerebrovascular accidents and other arterial events and, in women, with recurrent abortions, fetal loss and other complications of pregnancy.³ Therefore, tests for the presence of the LAC should be carried out in all young individuals with unexplained venous or arterial thrombosis and also in women with recurrent-early or late pregnancy loss.⁴ Antibodies of this class are members of a larger group called antiphospholipid or anticardiolipin antibodies. (Although not precisely the same these terms are used interchangeably.) Tests for lupus anticoagulant are usually performed in parallel with tests for the presence of antiphospholipid antibodies, usually by ELISA. In general these tests are not performed by haematology laboratories and are therefore not described here. Serological tests for antiphospholipid antibodies are not standardized and agreement between laboratories is poor. A large number of target proteins have been described but the most important, and the only one for which there is evidence of a pathogenic effect, is β₂-glycoprotein 1.⁵ Increasingly, tests specifically for anti-β₂-glycoprotein 1 antibodies are performed and possible mechanisms for their prothrombotic activity are being elucidated.⁶

The presence of a LAC may be detected by the clotting screen and, depending on the reagents and methods used as well as on the potency and avidity of the antibody, either the PT or APTT may be prolonged. However, the sensitivity of both APTT and PT to LAC varies considerably, so that these tests may well be normal and, if clinically suspected, specific tests should always be performed.⁴ The unmodified test for activated protein C (APC) resistance (see p. 456) is also sensitive to the presence of a LAC.

Patients with a LAC may show other abnormalities, including thrombocytopenia, a positive direct antiglobulin test and a positive antinuclear antibody test. Another frequent target for antiphospholipid antibodies is prothrombin but only rarely are these antibodies sufficient to inhibit or deplete prothrombin activity. Such patients may have a bleeding tendency. A recent international guideline on detection of lupus anticoagulants has been published ⁴ and recommended the following tests:

1. Dilute Russell’s viper venom time (DRVVT) in conjunction with the platelet neutralization test.

2. An APTT test that has a low concentration of phospholipid and uses silica as an activator, thus making it sensitive to the presence of LAC.

There are a large number of additional tests which have in the past been successfully used for the detection of LAC, several of which are no longer recommended due to poor reproducibility, technical problems and lack of standardization.⁴ The kaolin clotting time (KCT) and the dilute thromboplastin inhibition test are retained here because they are still widely used and thought to have some advantages by some authors. Although no single test is sufficiently sensitive to detect all LAC, readers are counselled against performing an excessive (more than two) number of tests because a large number of false positives will be generated. The most recent guidelines for the optimal performance of testing for LAC have been well laid out and are summarized in Box 19.1.⁴

Box 19.1 Recommendations for the optimal laboratory detection of lupus anticoagulant (LAC)⁴

(D) Confirmatory test

1. Confirmatory test(s) must be performed by increasing the concentration of phospholipid content of the screening test(s)

2. Bilayer or hexagonal (II) phase phospholipid (PL) should be used to increase the concentration of PL

(E) Expression of results

Results should be expressed as ratio patient:PNP for all procedures (screening, mixing and confirm)

(F) Transmission of results

A report with an explanation of the results should be given

Sample Preparation

It is essential that all the samples of plasma tested for an LAC should be as free of platelets as possible. This is achieved by further centrifugation of plasma at 2500 g for 10 min. A platelet count of <10 × 10⁹/l should be achieved. The plasma is centrifuged at room temperature to avoid platelet activation because platelet microvesicles may also invalidate the test. After separation, the plasma should be frozen at −70°C as soon as possible to prevent deterioration. Prior to testing, the frozen sample should be rapidly warmed to 37°C in a water bath.

Dilute Russell’s Viper Venom Time

Platelet Neutralization Test

Principle

When an excess of phospholipid, originally in the form of lysed platelets, is added to clotting tests, the tests become insensitive to the presence of a LAC. This appears to be a result of the ability of the platelets/phospholipid to adsorb the LAC. Platelet neutralization reagents are available commercially and are usually provided in DRVVT kits. Commercial reagents are preferred for consistency but a method for preparation is provided below. To utilize this property of platelets, they must be washed to remove contaminating plasma proteins and activated or ‘fractured’ to expose their coagulation factor binding sites.

Reagents for preparation of platelet neutralization reagent

Commercial platelet extract reagent or washed normal platelets

Acid–citrate–dextrose (ACD) anticoagulant solution (see p. 619), pH 5.4, is required for washed platelets. For use, 6 parts of blood is added to 1 part of this anticoagulant

Na₂EDTA. 0.1 mol/l in saline

Calcium-free Tyrode’s buffer. Dissolve 8 g NaCl, 0.2 g KCl, 0.625 g Na₂HPO₄, 0.415 g MgCl₂ and 1.0 g NaHCO₂ in 1 litre of water. Adjust pH if necessary to 6.5 with 1 mol/l HCl.

Method

Collect normal blood into ACD and centrifuge at 270 g for 10 min. Pipette the supernatant platelet-rich plasma (PRP) into a plastic container and centrifuge again to obtain more PRP, which is added to the first lot. Dilute the PRP with an equal volume of the calcium-free buffer and add one-tenth volume of EDTA to give a final concentration of 0.01 mol/l. Centrifuge the mixture in a conical or round-bottom tube at 2000 g for 10 min and discard the supernatant. Gently resuspend the platelet pellet in buffer and 0.01 mol/l EDTA. Centrifuge again, discard the supernatant and resuspend the pellet in buffer alone. Then centrifuge the platelets a third time and resuspend the pellet in buffer without EDTA to give a platelet count of at least 400 × 10⁹/l. The washed platelets may be stored below −20°C in volumes of 1–2 ml. Before use, they must be activated by thawing and refreezing 3–4 times.

Use the washed platelets or the commercial reagent in the dilute RVV test or in the APTT in place of the usual phospholipid reagent. First, determine a suitable dilution by testing a range of doubling dilutions in the test system with control plasma. A suitable dilution gives a similar clotting time to that obtained using control plasma and the phospholipid reagent.

Interpretation

The addition of platelets or a commercial ‘confirm’ phospholipid reagent to the DRVVT system corrects the clotting time when a LAC is present. It does not correct the time when the prolongation is due to a factor deficiency or an inhibitor directed against a specific coagulation factor. However, the ability of different batches of platelets to perform this correction is variable and may vary further with storage. Accordingly, each time the test is performed a plasma sample known to contain a LAC should be tested in parallel to establish the efficacy of the platelets.

Many commercial kits are now available for performing the tests described above. As with all such tests, there is an inevitable trade-off between sensitivity and specificity. This varies with different techniques, kits and coagulometer. One survey of reagents found that the best discriminator of positivity was by using a normalized correction ratio (CR) of DRVVT clotting times as follows:

where P is patient’s clotting time and N is the clotting time of normal plasma and D represents the detection procedure and C represents the confirmation (platelet/phospholipid neutralization) procedure.⁷ A correction of >10% is regarded as positive, but care should be taken to establish a local normal range. Other calculations may also be used such as a simple ratio of P_D/P_C or percentage correction: [(P_D − P_C)/P_D] × 100%.⁴

False-positive results may be obtained in patients receiving intravenous heparin although some reagents contain neutralizing agents. Interpretation may be difficult in patients receiving oral anticoagulants; this can sometimes be overcome by performing the test on a 50:50 mix with normal plasma.

Interpretation of Tests for Lupus Anticoagulant

Detailed instructions for the interpretation of LAC testing have been published.⁴ No single test detects all lupus-like anticoagulants and, if suspected clinically, then two specific tests should be performed before concluding that a LAC is not present.^4,⁸ Conversely, a single positive test should be repeated 12 weeks later because a transient positive may arise as the result of intercurrent illness or medication. It is crucial to distinguish LAC from specific anti-factor VIII antibodies, which are more typically time dependent but may have some immediate effect as well. Specific factor assays can be useful in discrimination but note that a LAC may result in non-parallelism and spuriously low results in these assays. Similarly, some weak LAC are neutralized by 50:50 mixing with normal plasma and sometimes exhibit a time-dependent effect. Some transient non-specific coagulation inhibitors are not detected by tests for LAC. Tests may be falsely negative while taking warfarin.

Kaolin Clotting Time

Principle

When the APTT is performed in the absence of platelet substitute reagent, it is particularly sensitive to a LAC. If the test is performed on a range of mixtures of normal and patient’s plasma, different patterns of response are obtained, indicating the presence of a LAC, deficiency of one or more of the coagulation factors or the ‘lupus cofactor’ effect.⁹

There are commercially available kits based on the KCT, such as Kaoclot (Life Therapeutics USA). This method uses a low-turbidity colloidal kaolin solution, making this reagent slow to settle and therefore suitable for use on automated coagulation analysers. Kaoclot shows high sensitivity to LACs but is not suitable for testing patients undergoing heparin therapy

Method

Mix normal and patient plasma in plastic tubes in the following ratios of normal to patient’s plasma: 10:0, 9:1, 8:2, 5:5, 2:8, 1:9 and 0:10. Pipette 0.2 ml of each mixture into a glass tube at 37°C. Add 0.1 ml of kaolin and incubate for 3 min, then add 0.2 ml of CaCl₂ and record the clotting time.

Results

Plot the clotting times against the proportion of normal to patient’s plasma on linear graph paper as shown in Figure 19.1.

Figure 19.1 Curves obtained using the kaolin clotting time (KCT) to test for the presence of a lupus anticoagulant (see text).

Interpretation

The pattern obtained for each patient must be critically assessed. A convex pattern (Pattern 1) indicates a positive result, whereas a concave pattern (Pattern 4) indicates a negative result. Pattern 2 indicates a coagulation factor deficiency and a LAC. Pattern 3 is found in plasma that contains a LAC but is also deficient in a cofactor necessary for the full inhibitory effect. The initial rate of slope is important because a steep slope indicates a positive result. This allows the test to be simplified so that only the tests of 100% normal and of 80% normal/20% test plasmas need be performed. The slope can be calculated using the ratio of KCT at 20% test plasma and KCT at 100% normal control plasma (N). For a positive result the ratio at this point should be ≥1.2.

Thus,

A control KCT of <60 s may indicate contamination of the control plasma with phospholipid.

Dilute Thromboplastin Inhibition Test

Principle

When the thromboplastin used for the PT is diluted, the PT becomes prolonged. At a certain point (usually 1:50–1:500 dilution) the concentration of phospholipid is low enough for the test to become sensitive to phospholipid binding antibodies and when a LAC is present the ratio of the test plasma to normal plasma clotting time increases. This test is now considered more useful because some thromboplastin reagents (e.g. Innovin) are more sensitive to LACs. However, it should be noted that diluting thromboplastin makes the system sensitive to low levels of factor VIII as are encountered in mild haemophilia, acquired haemophilia and low levels of factor V or factor VII. Care should be taken that these disorders are not confused. In one study, the test was determined to be positive when the dilute PT ratio (test/mean normal) using Innovin at 1:200 dilution was >1.15.¹⁰

Other Acquired Thrombophilic States

There are numerous other disorders that are associated with an increased risk of thrombosis but are not usually diagnosed using coagulation-based tests. Appropriate tests for some of these such as myeloproliferative neoplasms and paroxysmal nocturnal haemoglobinuria are found elsewhere in this book. One of the most important factors precipitating thrombosis is malignancy. However, the value of extensive testing for possible malignancy in patients with thrombosis remains contentious; some studies have shown that a history and examination combined with a few simple tests detect the large majority of malignancy and other systemic disorders. Other studies suggest that more intensive screening including abdominal and pelvic scans are required to achieve this. Large numbers of ‘tumour marker’ analyses result in numerous false positives. Even when tests have been effective in detecting occult malignancy it is not clear there is any improvement in outcome.^11,¹²

Investigation of inherited thrombotic states

Testing for thrombotic syndromes remains frequent, despite doubts about its clinical utility.¹ Patients with disorders of pregnancy and those with thrombotic disorders are often referred for investigation. Prior to testing for thrombophilia consideration should be given to the likely benefits including alteration in management that can be achieved. This requires a carefully taken history, noting in particular the circumstances of any previous thrombotic event, a family history of thrombosis and identification of any coexisting disorders. The relevant tests are described below.

Antithrombin (AT)

AT ¹³ (previously known as antithrombin III) is the major physiological inhibitor of thrombin and factors IXa, Xa and XIa. AT deficiency is found in approximately 2% of cases of thrombosis and may be acquired or congenital. Various methods are available for measuring either functional activity or antigenic quantity of AT. The functional methods are based on the reaction with thrombin or factor Xa and can be coagulation based or chromogenic assays. A chromogenic assay is described below.