Coagulation testing in common clinical scenarios

Routine coagulation tests incorporate an activated partial thromboplastin time (APTT) and prothrombin time (PT). They provide insight into the clotting pathways of a patient. However, when to order them and what to do with an abnormal result, especially with many patients now being prescribed nonvitamin K oral anticoagulants (NOACs; also known as direct oral anticoagulants [DOACs]), is increasingly challenging. The utility and limitations of routine coagulation testing are explored in this article through three scenarios that focus on patients who are concerned about their bleeding risk.

Haemostasis overview

Haemostasis represents the delicate balance between maintaining circulatory flow within the vasculature and the need to cease bleeding at the site of any vascular injury. It comprises the complex interplay between blood vessels, platelets, coagulation factors and the fibrinolytic system. Damage to a vessel triggers recruitment of von Willebrand factor (VWF), which facilitates platelet attachment and activation. This process is known as primary haemostasis. Secondary haemostasis represents the stepwise activation of coagulation proteins leading to fibrin production, which reinforces the initial platelet ‘plug’ (Figure 1).

These processes are balanced with endogenous anticoagulants (protein C, protein S and antithrombin) and fibrinolytic pathways. Protein C inactivates clotting factors V and VIII and protein S potentiates protein C activity. Antithrombin’s main role is the inactivation of thrombin (factor II) and factor X. The fibrinolytic system acts to contain the clot to the site of injury and then dissolve the clot as the wound heals. If this delicate haemostatic system becomes unbalanced, it can lead to either thrombosis or haemorrhage.¹

Routine coagulation tests - APTT and PT

Routine coagulation testing is largely designed to detect specific coagulation factor deficiencies and should not be considered a measure of overall haemostasis as it excludes several key components. Although inherently integrated in vivo, the clotting cascade is traditionally depicted as consisting of intrinsic (contact) and extrinsic (tissue factor) pathways (Figure 2). In vitro, these pathways are represented by the APTT and PT coagulation tests, respectively. To ensure an accurate and useful result, the blood sample needs to be collected in the correct tube, filled to the correct volume (the sodium citrate in the tube is measured to provide a specific ratio to blood) and transported to the laboratory in a timely manner with the clinical indication for testing and the patient’s current anticoagulation therapy, if any, recorded on the request form.

For the APTT, the sample is centrifuged to obtain platelet-poor plasma and then combined with phospholipid, a contact activator (kaolin, silica or ellagic acid) and calcium (this reverses the effect of citrate and enables the blood to clot again). The test result is the time (in seconds) from the addition of calcium to the formation of a fibrin clot. The APTT test is sensitive to the presence and activities of factors II, V, VIII, IX, X, XI, XII and fibrinogen (factor I). It is important to discuss with your local laboratory whether their APTT reagent is sensitive or insensitive to the presence of a lupus anticoagulant.

The PT represents the time taken for the blood to clot after platelet-poor plasma is combined with thromboplastin (tissue factor and phospholipid) and calcium. The PT is sensitive to the presence and activity of factors II, V, VII, X and fibrinogen (Figure 1). The international normalised ratio (INR) is derived from the PT and adjusts for differences in the PT reagents over time and between laboratories (thereby allowing monitoring of patients regardless of where testing is performed). Most laboratories report both PT and INR values whenever a PT test is performed; however, the INR is applicable only for patients taking warfarin.

Causes of abnormal PT and APTT are outlined in Table 1. NOACs have intentionally been excluded here and are discussed below.

Scenario 1

A young woman presents to your practice. She describes her periods as heavy and has noticed a few bruises on her arms and legs and cannot remember how she got them. She has occasional nosebleeds in winter.

Minor bleeding, although sometimes alarming, is common in otherwise healthy people.² A thorough history should include the type of bleeding, when it was first noticed (e.g. menorrhagia since menarche is suggestive of a possible bleeding disorder), the frequency and the spontaneous or provoked nature of the bleeding including any relationship with haemostatic challenges (including trauma, surgery, dental extractions and pregnancy). Disorders of primary haemostasis are primarily due to abnormal levels and/or function of both VWF and platelets. They tend to manifest as mucocutaneous bleeding. Secondary haemostatic defects are primarily due to abnormal levels and/or function of clotting factors and tend to manifest as bleeding in joints or deep tissues.³

Standardised bleeding questionnaires have been generated to help assess bleeding symptoms using clear objective criteria to reduce reliance on patient interpretation. Haematologists commonly use the International Society on Thrombosis and Haemostasis Bleeding Assessment Tool in clinical practice.⁴ This assessment tool covers all areas of haemostasis in bleeding disorders such as bruising, epistaxis, dental extraction, menses, childbirth, muscle haematomas and gastrointestinal bleeding. The patient’s score increases if their bleeding has required medical consultation and further increases if intervention was required, particularly blood transfusion.

A review of all medications including over-the-counter medications, especially anti-inflammatories, and herbal supplements should be obtained. Family history is also vital in the assessment of a potential bleeding disorder. Finally, clinical examination for any bleeding signs (such as petechiae, ecchymoses and haemarthroses), joint hyperflexibility (associated with connective tissue disorders), stigmata of chronic liver disease and any evidence of underlying congenital abnormalities should be performed.⁵

If a bleeding disorder is considered possible, initial screening laboratory tests include a full blood count, blood film, coagulation profile (APTT and PT) and fibrinogen level. The FBC and film should identify thrombocytopenia as well as any abnormal platelet morphology, which may suggest a specific diagnosis. For example, the inherited platelet function disorder Bernard Soulier syndrome is characterised by thrombocytopenia and large platelets from childhood. Renal and liver function tests are also recommended, as renal failure can cause acquired platelet dysfunction and liver failure reduces the production of all procoagulant and anticoagulant proteins except factor VIII and VWF.

If the patient has a prolonged PT and/or APTT, the next step we recommend when investigating for a bleeding disorder is ordering a mixing study. This involves combining 50% of the patient’s plasma with 50% normal plasma and repeating the test. If the cause for the test prolongation is deficiency in one or more clotting factors, they will reach sufficient levels by adding in the normal plasma and hence the mixed sample will provide a normal result. If this is the case, the next step is to quantify the patient’s factor levels (i.e. factors VIII, IX, XI and XII for APTT) that contributed to the original abnormal test result. A mixing study that does not correct implies the presence of a factor inhibitor. If the patient is not taking an anticoagulant and testing is negative for lupus anticoagulant (which acts as an inhibitor in vitro but leads to thrombosis, not bleeding, in vivo), we would recommend discussion with a haematologist, as a factor inhibitor can represent a life-threatening condition.

If the patient’s symptoms are suggestive of a primary haemostatic defect and/or the factor VIII level is decreased, we also recommend assessing for von Willebrand disease. This comprises a panel of tests that assess the level and function of the patient’s VWF. It is worth noting that VWF levels fluctuate with inflammation, infection, pregnancy, hormonal therapy, cancer and acute stress. Therefore, if clinical suspicion is high, the tests should be repeated.⁶

All patients with an identified bleeding disorder should be referred to a haematologist for ongoing management. It is not uncommon that, despite these screening investigations, a cause is not immediately apparent. If suspicion remains high based on the patient’s personal and/or family history, a haematology referral should be considered. Further investigations may include platelet function analysis and more detailed coagulation testing, which are beyond the scope of this article.

Scenario 2

A 65-year-old patient recently changed from warfarin to apixaban for atrial fibrillation. He liked knowing his INR and asks if you can check this new medication ‘just to be sure’.

The NOACs currently include the direct thrombin inhibitor dabigatran and the activated factor X (factor Xa) inhibitors apixaban and rivaroxaban. Clinical trials and real-world data show that these medications have similar or superior efficacy and safety profiles to vitamin K antagonists. Combined with the fact that they do not require routine laboratory monitoring, they have understandably revolutionised the management of atrial fibrillation and venous thromboembolism.^7,8

However, there are disadvantages that should be acknowledged. These medications are more expensive, and a missed dose can result in a lack of appropriate anticoagulation for 12 to 24 hours (depending on the medication). Although dabigatran may be reversed with idarucizumab, there is no reversal agent available in Australia for apixaban or rivaroxaban. In addition, they should not be used in patients with prosthetic heart valves, antiphospholipid syndrome, severe renal impairment or severe liver disease, or who are pregnant or breastfeeding. Although there is accumulating evidence for the efficacy of these drugs in malignancy-associated venous thromboembolism and heparin-induced thrombocytopenia, we recommend discussion with a haematologist before prescribing in these contexts.^9,10

NOACs can cause confusion when routine coagulation tests are ordered. Importantly, although none of the NOACs are monitored with the traditional PT/INR or APTT, they will affect these tests in various ways (Table 2). The only way to ascertain a drug level is by performing a drug-specific assay, and discussion with the testing laboratory should take place before requesting this.

NOACs can also have a significant impact on more specialised coagulation testing including factor levels, protein C, protein S, antithrombin and lupus anticoagulant testing (depending on the method used). We strongly recommend discussion with your laboratory before ordering these tests on a patient taking a NOAC to avoid a misdiagnosis.

We do not routinely perform any laboratory studies to monitor patients who are taking NOACs. We also endeavour to order all necessary specialised testing before these drugs are started. In our experience, the most common real-world scenarios in which drug levels are ordered are when an urgent surgery or invasive procedure is required (to ensure the drug level is low/undetectable), when therapy appears to have failed, or in the case of an overdose.

Scenario 3

A 78-year-old woman presents to you before her knee replacement. She remembers bleeding when she had her tonsils removed as a child and wants to discuss the upcoming surgery.

Routine coagulation tests are frequently performed in the preoperative setting in the belief that they may identify patients who have bleeding disorders and thereby reduce the risk of perioperative bleeding. However, indiscriminate screening will only very rarely identify previously undetected individuals. In contrast, prolongation of the APTT may be due to mild factor XII deficiency, high molecular weight kininogen deficiency, prekallikrein deficiency, a collection issue or the presence of a lupus anticoagulant, none of which are associated with a bleeding phenotype. Also, defining the normal range as the mean plus two standard deviations in healthy subjects means that 2.5% of normal individuals will yield a prolonged result. Therefore, if coagulation testing were performed on every patient preoperatively, it would be likely to lead to unnecessary testing, patient concern, increased costs and procedural delay without any clinical relevance.¹¹

It is also dangerous to presume that all bleeding disorders will be detected by PT and APTT. Conditions that may not be identified include von Willebrand disease, platelet function disorders, factor XIII deficiency, connective tissue disorders and fibrinolytic pathway deficiencies. The British Committee for Standards in Haematology guideline therefore recommends against ordering coagulation tests to predict perioperative bleeding risk before surgery or other invasive procedures in patients without any bleeding history.¹² This recommendation is also shared by the guidelines of the European Society of Anaesthesiology and the French Society of Anaesthesia and Intensive Care.^13,14

We therefore advocate that a bleeding history including any evidence of excessive post-traumatic or postsurgical bleeding, family history and medication review be undertaken in all patients before surgery or invasive procedures. Only then, if there is concern for a bleeding disorder, should testing incorporating a coagulation profile (APTT and PT) and fibrinogen level be done. Preoperative interventions may then include antifibrinolytics such as tranexamic acid, platelets or clotting factor replacement depending on the underlying condition detected.

If a patient requires urgent surgery, concern for a potential bleeding disorder should not prevent lifesaving procedures. If surgery can be safely delayed until further work-up can be done, this is preferable. Other acute conditions that can influence haemostasis and warrant preoperative coagulation testing include liver disease, sepsis, diffuse intravascular coagulation and poor nutritional states (which can lead to vitamin K deficiency).

In summary, in the absence of a definitive bleeding history we do not recommend routine coagulation testing before surgeries/procedures. We also do not recommend routinely checking NOAC levels before elective surgeries. Rather, we recommend withholding the medication beforehand, basing the timing of cessation on the bleeding risk of the surgery, the half-life of the drug and the renal function of the patient (Table 3).

Conclusion

A thorough medical history remains the mainstay of assessing patients preoperatively and for possible bleeding disorders. If required, judicious coagulation testing can then provide significant insight into any underlying primary or secondary haemostatic condition. Routine testing for patients on NOACs is not required; however, these medications can influence the results. Therefore, coagulation results should always be interpreted within the individualised clinical context and with knowledge of the testing limitations. MT

References