Evaluation of quality issues in haemostasis testing for the identification of bleeding and thrombotic disorders.

Abstract

Haemostasis is a complicated and regulated process, which when disrupted (e.g., by a factor deficiency, excess presence of anticoagulants, presence of thrombophilia) can lead to bleeding or thrombosis. Laboratories help to assess haemostasis for investigated patients by performing a range of tests, and test quality is dependent on various processes including internal quality control (IQC) external quality assessment (EQA). Once a thrombosis has occurred, medical intervention is usually required in the form of anticoagulant therapy. Heparin, either unfractionated (UFH) or low molecular weight heparin (LMWH) is usually administered to patients to prevent or treat thrombosis. In Chapter 3 (Heparin anticoagulant therapy) the monitoring of UFH by the activated partial thromboplastin time (APTT) is explored, with variation in reported results attributed to different APTT reagent sensitivities and the many reagent/instrument combinations available. Compared to UFH, LMWH has a more predictable response and, in general, does not require monitoring. However, under certain circumstances, such as, impaired renal function, monitoring using the anti-Xa assay may be required to improve efficacy. Therefore, the APTT and anti-Xa assay, as performed by any given laboratory, must be done at the highest level of accuracy and reliability as possible, in order that clinicians can properly manage their patients based on these test results. Several international EQA programs have highlighted assay variations observed with these tests and data published in Chapter 3 provides comparative data from our geographical region. For many years, the progression from heparin therapy has been towards vitamin K antagonist (VKA) therapy (e.g., warfarin). These cause a prolongation of the Prothrombin time (PT) by diminishing the functional activity of Factors II (FII), VII (FVII), IX (FIX) and X (FX). VKA therapy has many limitations including, a narrow therapeutic range, interaction with many drugs and dietary influence. These issues mandate regular monitoring of patient anticoagulant status, which is usually achieved using the International Normalised Ratio (INR). This is a simple calculation (INR=(PT/MNPT)ISI) requiring the PT, mean normal prothrombin time (MNPT) and international sensitivity index (ISI). Despite international standardisation, there remains considerable concern regarding ongoing high levels of inter-laboratory variation, as generated by different laboratories using the same homogeneous plasma sample. This issue is highlighted in Chapter 4 (Vitamin K Antagonist anticoagulant therapy) and significant discrepancies continue to be evidenced in EQA environments, prompting additional investigations to determine causes and to identify potential inconsistencies of practice. Alternatives to traditional VKA therapy are now available for the treatment and prevention of arterial thrombosis (AF) and venous thromboembolism disease (VTE). The direct oral anticoagulants (DOACs) dabigatran, rivaroxaban and apixaban have been developed as not requiring laboratory monitoring; however, under certain clinical situations, laboratory testing may be indicated. Two separate studies were performed to assess the interference of these DOACs on a wide range of haemostasis tests. Specific assays are used to predict the level of DOAC, while in other assays these drugs caused interference and sub-optimal assay performance, leading to false positive or negative results. This is explored in Chapter 5 (Anticoagulant therapy using DOACs). One of the more complicated examples of clinical diagnosis of a thrombophilia disorder is represented by the Lupus Anticoagulants (LA), an important component of the autoimmune disorder antiphospholipid syndrome (APS). Challenges occur due to the heterogeneity of laboratory tests available and the differing approaches undertaken. Chapter 6 (Lupus Anticoagulant) evaluates EQA data for a complex but highly positive LA sample. Only 58% of survey participants correctly identified this sample as being positive for LA. Of the 34.7% participants who identified the sample as (falsely) LA negative, most incorrectly identified a (false) factor inhibitor. The debate over whether to utilise mixing studies in LA testing, with patient plasma mixed with normal pooled plasma before testing, is also discussed in Chapter 6. Progressing from thrombosis to bleeding disorders, Chapter 7 (Common Bleeding disorders) highlights the difficulty in diagnosis of von Willebrand Disease (VWD) and also the problems in the performance of FVIII inhibitor assays, as used to monitor patients with congenital and acquired Haemophilia. VWD is the most common inherited bleeding disorder, but its diagnosis can be difficult. Due to the heterogeneous nature of VWD, a panel of tests is required and these must be performed with highest degree of accuracy and reliability. Tests include both antigenic (VWF:Ag) and functional (VWF:RCo, VWF:Act and VWF:CB) assays, required for identification of the six subtypes of VWD. VWD testing is explored extensively in Chapter 7 (Common Bleeding disorders). The initial detection of factor inhibitors by laboratories is essential in clinical management of affected patients. The FVIII inhibitor is the most common and is directed against factor VIII coagulant (FVIII:C), resulting in an increased risk of bleeding. High interlaboratory variation (CVs ranging from 20-50%) in the factor inhibitor assay (including FVIII) have been previously reported by many EQA programs and this is extensively explored in Chapter 7 (Common Bleeding disorders). One of the contributing factors to the large interlaboratory variation is the vast array of methodologies used by laboratories. It has been previously identified that no two laboratories in our geographic area use identical methods for their inhibitor assay, thus increasing the potential for cross laboratory variation. Standardisation of the FVIII inhibitor assay over the past decade appears to be slow with no apparent improvement seen to reduce the interlaboratory variation. Overall, this thesis touches on many areas of haemostasis testing for the identification of bleeding and thrombotic disease, but specifically highlights the many quality issues in the performance and interpretation of these tests. Errors in patient test results can occur at multiple stages in the test process. Quality issues in haemostasis testing encompass pre-analytical, analytical and post analytical issues, and this thesis reports issues related to all these areas. However, given that laboratories are generally sent identical lyophilised samples for testing, together with the high degree of automation available, reduces the potential for analytical issues to occur, thereby facilitating primary focus on certain pre- and post-analytical issues. For example, many post analytical issues, particularly with VWD and LA testing, can be identified, where interpretations based on overall test results, can be offered by the laboratory to provide clinical guidance and enable clinicians to make the best possible decisions for patient clinical management of both bleeding and thrombotic disorders.