Considerations in the laboratory assessment of haemostasis


Angus McCraw, Katharine Dormandy Haemophilia Centre & Thrombosis Unit, Royal Free Hospital, London, UK.
Tel.: +44 2078 302 228; fax: +44 2074 726 759;


Summary.  This review outlines a number of key issues when performing laboratory testing of homeostasis. The effect pre-analytical variables have on the reliability and consistency of screening tests is often forgotten due to a lack of understanding and awareness. This can be improved through educating healthcare professionals who are involved in taking blood for assessment. Recent advances in coagulation testing have not enabled laboratories to replace the Prothrombin Time (PT) and Activated Partial Thromboplastin Time (APTT) screening tests with more advanced assays and they continue to play an important role with the advantage of being easily automated. However, there are many analysers on the market, each with varying sensitivity to coagulation defects and it is important to keep this in mind when interpreting results.

The pre-analytical phase of testing encompasses everything that happens to a patient specimen up to the point of actual testing (analytical phase). A review of the literature by Bonini et al. [1] revealed that 32–68% of all laboratory errors occur in the pre-analytical phase. There is probably no other pathology discipline requiring greater understanding of how variations in sample preparation affect laboratory results that can have a significant impact on patient outcomes such as diagnosis, treatment and therapeutic monitoring than in coagulation testing. There have been a number of articles discussing this topic [2–4], yet it continues to be a problem for laboratories. Some of the issues associated with this are outlined below.

When ordering the test the requisition form, according to NCCLS guidelines H3-A5 (2003) [5], should contain the patient name and date of birth, hospital number, date and time the specimen was requested, the physician’s name as well as department or location. Critical to a coagulation laboratory is information on further clinical details like medications that may affect the test result outcomes particularly the anticoagulants heparin, Warfarin or anti-platelet therapies.

It is very important that the phlebotomist ensures that the patient on the requisition form is the person from whom the blood is to be drawn, by providing full name and/or some other unique identifier. Prior to the venipuncture, the phlebotomist should, immediately and in the patient’s presence, label each of the drawn tubes with the patient’s full name, hospital, date and time of collection.

Phlebotomy is the act of puncturing a vein for the purpose of withdrawing blood and is one of the most critical parts of the whole pre-analytical phase. During this step you are in fact causing injury by the very act of the venipuncture, which in itself initiates the haemostatic response and best explains the vulnerability of specimens for coagulation testing. Various blood collection systems may be used for obtaining a blood specimen; however, it should be noted that the larger the syringe, the greater the chance that activation may occur before it is mixed adequately with anticoagulant, and therefore volumes of <20 mL are recommended. Obtaining a specimen through a venous access device should be avoided to minimize heparin contamination. The collection device in widespread use is an evacuated blood collection tube [6] and care must be taken to follow manufacturer’s expiry dates as water can diffuse over time affecting blood to anticoagulant ratios. A question often asked is should the first draw of blood be discarded. Studies by Yawn et al. [7], Gottfried and Adachi [8], Adcock et al. [9] and Brigden et al. [10], showed that no statistical differences occurred for prothrombin time (PT), International Normalised Ratio (INR) and/or activated partial thromboplastin time (APTT) between a first and second draw tube. NCCLS guideline H21-A4 [11] indicates that it is acceptable practice to use the first draw tube if only PT, International Normalised Ratio (INR) or APTT are requested but for other coagulation tests there are no current published data to suggest that this practice is unnecessary. Should a patient require testing in addition to coagulation testing, then it would be sensible to draw blood for other pathological specimens first; however, when using winged blood collection sets or when obtaining blood from venous access devices, a discard tube or volume is necessary.

Coagulation cannot occur without calcium ions, and agents that bind calcium such as sodium citrate permit blood fluidity in the test tube. This process is reversed in coagulation testing such as the PT and APTT wherein plasma samples are re-calcified and the coagulation process is triggered; therefore different citrate concentrations will affect results. Values for PT and APTT are longer in 3.8% (129 mmol L−1) than 3.2% (109 mmol L−1) sodium citrate tubes. The Fourth Edition Approved Guideline H21-A4, NCCLS [11] states that only concentrations between 105 and 109 mmol L−1 (3.13–3.2%), respectively, should be used, which aligns with recommendations from the International Society on Thrombosis and Haemostasis and the European Committee for Clinical Laboratory Standards. It should also be noted that variances such as under filling of collection tubes or an elevation in patient haematocrit may also impact the citrate–calcium ratio.

How a specimen tube is handled from patient to laboratory is mostly out of laboratory control. One must consider that the means of transport, exposure to heat (reference laboratory collection boxes), vibration (pneumatic tube systems), position of tubes (upright preferable) and overall time to delivery can dramatically affect test results. For example, elevated temperatures enhance degradation of factors V and VIII whereas prolonged exposure to cold (>7 h) may activate factor VII. Traumatic handling such as shaking, vibration and agitation can lead to haemolysis and platelet activation causing falsely shortened clotting times. Specimen tubes should undergo physical inspection prior to and following centrifugation. Initial examination should note that the correct specimen tube has been used, the collection volume is appropriate for tube fill volume, and no clots or fibrin strands are observed. Subsequent to centrifugation, the plasma should be checked for haemolysis, icterus and lipaemia.

Platelet poor plasma used for coagulation testing is obtained by centrifuging specimens at 1500 gravity (g) relative centrifugal force (rcf) for no <15 min (either room or refrigerated temperature is acceptable). A rcf nomogram is provided in NCCLS document H18-A2 [12]. To minimize re-mixing of plasma and red cells, a swing-out bucket (angle) rotor should be used with minimal brake applied at the end of centrifugation. All tubes, whether whole blood or separated plasma, should be processed or stored with a cap or stopper to minimize loss of CO2, which causes pH to increase, leading to prolongation of PT and/or APTT. If testing cannot be performed immediately, plasma should be separated from cells using a plastic disposable pipette taking care not to disturb the platelet layer. According to Woodhams et al. [11], stability of coagulation proteins is best when sample volumes of ∼1 mL are stored in microtubes having a minimum of dead space. Plasma aliquots may be stored for 2 weeks at −20°C; however, it is imperative that a frost-free freezer not be used. The current NCCLS recommendations for long-term storage indicate that samples may be kept for up to 6 months at −70°C. Woodhams et al. [13] showed stability (defined as <10% loss of activity) of coagulation factors up to 18 months at this temperature, however, they also noted that FVIII and FXI showed deterioration above this level at 6 months. Prior to testing, frozen plasmas should be thawed rapidly at 37°C (to prevent denaturing fibrinogen) and tested immediately; however, it is acceptable to hold at 4°C for a maximum of 2 h.

One of the screening tests particularly sensitive to pre-analytical variables is the APTT which is activated and then recalcified with phospholipids under controlled conditions and these can very easily be disrupted during the pre-analytical phase. The test is used to detect various bleeding disorders caused by deficiencies in the intrinsic clotting system, i.e. fibrinogen, prothrombin, FV, FVIII, FIX, FX and FXI. It is invariably prolonged if one or more of these components are reduced to very low concentrations. It is important to note that deficiencies of any of the contact activation factors, i.e. high molecular weight kininogen (HMWK), prekallikrein or FXII, also leads to a prolongation of the APTT but is not linked to a bleeding diathesis. To detect less obvious bleeding disorders caused by mild to moderate factor deficiencies, it is important to choose a reliable and sensitive APTT reagent. Ideally, it should have a proven capacity to generate a prolonged APTT if a single or combined deficiency may lead to clinically important bleeding complications. The APTT reagent is also the test base for one-stage factor assays and the variable responsiveness is also propagated to these assays. The importance of choosing the correct APTT reagents for FVIII:C and FIX:C activity assays was recently shown by two investigations that illustrate how the diagnostic value can differ between reagents [14,15]. The laboratory phenotype known as discrepant mild haemophilia A is another example where the APTT-based one-stage FVIII:C assay may be poorly correlated to the bleeding phenotype. Specialized coagulation laboratories usually have insight about the variability of APTT reagents and can choose between reagents depending on the application. However, it is important to remember that in some patients, a normal APTT may not exclude the possibility of a mild bleeding disorder and further testing may be warranted. As the APTT reflects the sum activity of several clotting factors it can happen that a transient elevated level of one factor may mask a mild deficiency of another under certain conditions. In the case of a prolonged APTT, it is likewise important that an appropriate interpretation is made that may guide any further investigation, particularly in the absence of an obvious explanation for the prolongation. Therefore, every laboratory that performs APTT (and other screening assays) should be aware of their reagent characteristics and have defined a practical approach how to evaluate test results.

The APTT is a test to determine the intrinsic coagulation time and was first developed as the partial thromboplastin time test by Langdell et al. [16]. The driving force for the new test was that existing tests at that time, such as the PT test, which involved the addition of tissue thromboplastin to initiate plasma coagulation, could not be used to identify classical haemophilia caused by FVIII deficiency. Langdell et al. made the important observation that partial thromboplastin, in contrast to the complete thromboplastin used in the PT assay, could be used to initiate coagulation and with this reagent it was for the first time possible to differentiate between normal and haemophilic plasma samples. Today we know that the partial thromboplastin reagent is lacking tissue factor and merely contains the phospholipid component. The interesting story about the test and the people behind it was described in a recommended historical sketch by White to commemorate the 50th anniversary in 2003 [17]. Proctor and Rapaport [18] reported an important refinement in 1961 that led to the APTT test. Their modification involved the addition of kaolin to the partial thromboplastin reagent, which replaced the test tube glass surface as activator. This resulted in an improved and faster test through optimal activation of the first stage of the intrinsic coagulation system (Fig. 1). The APTT has since had a tremendous impact, in the clinical routine as well as in basic coagulation research.

Figure 1.

 The reactions of the activated partial thromboplastin time (APTT) test. The first stage involves addition of 1 part activator to 1 part citrated plasma and incubated 5 min at 37°C. During this stage, also known as contact activation phase, factor XII (FXII) is activated to FXIIa with the help of prekallikrein and high molecular weight kininogen (HMWK). The activator is added to provide a necessary surface for the optimal activation during this stage. In the next stage, the re-calcification phase, 1 part of calcium chloride solution is added and all the components of the Xase- and prothrombinase complexes are activated and assembled on phospholipid surfaces. When thrombin is generated it leads to the cleavage of fibrinogen and formation of the fibrin clot.

The main test protocol involves mixing of 1 part activator with 1 part citrate plasma, incubate 5 min at 37°C and then add 1 part CaCl2 that introduces calcium ions to the reaction mixture. The first stage during the 5 min of incubation, also known as the contact activation phase, involves the activities of prekallikrein, HMWK and factor XII (FXII). The activator is usually of the particulate type (kaolin, celite or silica) or the soluble type (ellagic acid) and provides negatively charged surfaces for the reactions to take place. The activator also contains phospholipids but these are not involved in the coagulation process until the re-introduction of calcium ions. After calcium has been added, the coagulation proenzymes (FXI, FIX, FX and prothrombin) and cofactors (FV and FVIII) are assembled on the phospholipid surface leading to efficient formation of thrombin. The generated thrombin will convert fibrinogen to fibrin, which will spontaneously polymerize to form a fibrin clot. The whole process will take about 30–40 s after re-calcification. The phospholipids are of great importance for the performance of the test and can be obtained from various species and organs. Vegetable and pure synthetic phospholipids can also be used. The diversity among APTT reagents, resulting from different contents and concentrations, is reflected by the wide variation in responsiveness to deficiencies of clotting factors, lupus anticoagulant and anticoagulant drugs. Furthermore, the type of clot detection system may also influence the APTT. Thus, any attempt for a common calibration procedure of the test has so far failed and it is important to ensure the reliability of the APTT using appropriate quality control system. It is also necessary to establish a local reference interval that reflects normalcy.

One of the main advantages of the APTT is its simplicity. It can be performed manually by tilt-tube technique or easily be automated using high throughput analysers. Whilst many countries are still dependent on manual coagulation techniques, automation, whether it be semi or fully automated are slowly becoming the norm. However, technologists should recognize that even with automated equipment they are ultimately in control of its use and maintenance. The following may give emerging countries some guidance on how to approach the transition from manual to automation.

Evaluation and selection of coagulation analysers

Automation in haemostasis is relatively recent. The original techniques used for coagulation studies were manual methods based on visual detection of the fibrin clot and were the most common form of clot detection right up to the 1970s when new semi-automatic equipment was invented based on photometric or mechanical principles to detect fibrin. Because of the increasing demand for high volume, routine coagulation screening tests such as PT, APTT, Clauss fibrinogen (FIB) and an increasing demand of budget management, fully automated coagulation analysers have become more and more popular.

These analysers have continued to be developed and as a result have become more sophisticated and coagulation testing results have become more than just a number expressed in seconds. For instance, modern photo-optical coagulometers collect optical data over the entire course of clot formation in the form of a reaction curve, thus providing additional information through alterations that may affect its shape and slope caused by the activities and reactions of coagulation factors and inhibitors.

Advantages of automation in a coagulation laboratory

Automation in a coagulation laboratory:

  • 1Improves the capacity and flexibility of time spent by a professional.
  • 2Improves test reproducibility.
  • 3Reduces costs in samples and reagents.
  • 4Facilitates data storage and recovery systems by means of computer programmes.
  • 5Provides automatic replay of results when mistakes are made in the first run.
  • 6Allows different tests to be run at the same time using a single sample.
  • 7Permits sampling from a closed tube, which improves safety and efficiency in coagulation tests.
  • 8Enables analysers to dilute samples, calibrators and controls.

There are basically two methods of end-point clot detection available:

  • 1Mechanical
  • 2Optical
  • 2.1 Photo-optical
  • 2.2 Nephelometric
  • 2.3 Chromogenic
  • 2.4 Immunological

These methodologies all have their advantages and disadvantages from the possibility to measure antigen-antibody reactions in proteins to optical checks for haemolysis, lipaemia and icteric samples as well as wave form analyses [19].

For routine assays, most instruments are sold in combination with coagulation reagents that are intended for use on those instruments. The reagents may vary greatly in their degree of sensitivity to detect factor deficiencies and coagulation inhibitors; therefore when selecting an instrument type, a specific instrument-reagent combination should be evaluated [20].

Some studies have suggested that optical and mechanical detection methods are equivalent in terms of correlation, accuracy and precision for coagulation testing [21]; however, these studies focused primarily on the performance of one particular instrument/reagent combination rather than a standard instrument/reagent combination as in the comparative method. On the other hand, some other studies have suggested that optical detection is superior to mechanical detection [22]. Consequently, the advantage of one detection method over the other remains unknown. It is anticipated that different failure rates for detecting coagulation in samples are observed through optical and mechanical methods depending on the nature of interfering factors. If a measurable end point is not detected, it is important for laboratories (with either a photo-optical or mechanical system) to check the sample with an alternative method of clot detection. It is also important to establish specific reference ranges for the particular system employed in a laboratory.

Newer trends in haemostasis testing and reagent/instrument manufacturing necessitate the development of an updated guideline for coagulometer evaluation. A large number of organizations, such as the clinical and laboratory standards institute (CLSI) provide protocols for the evaluation of clinical laboratory tests. It is worth pointing out that laboratories are responsible for trustworthy results and must choose the coagulation equipment that will generate appropriate results despite budget constraints. Naturally, equipment demands regular technical maintenance, permanent knowledge and system control, as a mistake or failure may definitely influence results.

To conclude, it is very important to understand that to guarantee the reliability of the results issued by coagulation tests, a series of activities are required to control and prevent errors that may occur from the time when the test is ordered until the results are interpreted. In addition, appropriate selection of reagents and equipment to use is also relevant to make sure that the delivered result reflects the true condition of the patient. However, this is not the only source of error and therefore an abnormal result is not necessarily caused by poor choice of an instrument-reagent system.


The authors stated that they had no interests which might be perceived as posing a conflict or bias.