TBS preanalytical phase working group survey study – preanalytical phase in coagulation laboratories

Patient status

For coagulation tests, some additional information (e.g., fasting for 8–12 h, non-smoking in this period, avoidance of physiological stress, blood collection after resting of 10–15 min etc.) is needed in addition to the standard blood collection requirements. Although the CLSI H21-A5 standard contains detailed information on the sampling, transportation, preparation and storage conditions for coagulation and molecular hemostasis tests, it provides no information about fasting and the time of day when the sample should be taken [7]. There are a limited number of studies in this respect in the literature [18], [19]. According to Lippi et al., AT III activity increases at the first and fourth hours after a meal, while aPTT decreases at the second hour after a meal as distinct from basal level [19]. Nonetheless, the European Federation of Clinical Chemistry and Laboratory Medicine Preanalytical Phase Working Group recommends taking blood samples between 7 am and 9 am. and standardising the fasting periods according to 12 h [20]. According to the survey results, 87.5% of laboratories admitted accepting samples at any time of the day without considering fasting status (Table 2).

Tourniquet time

During the blood collection period, the use of a tourniquet is the most common method to increase venous visibility and, select the most suitable vein. A tourniquet should be applied on a point that is 7–10 cm above the blood collection site and, it should remain applied for less than 1 min [21]. A longer tourniquet time can cause variations in the coagulation test results due to venous stasis due to hemoconcentration, and the infiltration of the tissue fluid by blood [22], [23]. Despite the fact that this may be an important factor in causing errors, 10.64% of the laboratories indicated that the tourniquet time is not crucial in coagulation tests. In comparison, 65.96% of the laboratories suvrveyed emphasized that <1 min and a further 23.4% stated that <30 s is essential for reliable coagulation test results.

Citrate as anticoagulant

Sodium citrate is recommended as an anticoagulant for coagulation test samples and usually, a 9:1 blood to sodium citrate ratio is used with a buffer range between 3.2% (105–109 mM) and 3.8% (129 mM). In the 3.8% sodium citrate tube, the excess sodium citrate can potentially bind more calcium ions present, for example, as added back into the clot-based assay and result in longer PT and aPTT times [7]. According to studies, if coagulation tests reference ranges are determined for sample tubes with lower citrate concentrations, increased PT and aPTT and, decreased fibrinogen results may be obtained from higher citrate concentrations tubes (129 mM, 3.8%) [24],  25]. This difference would be more significant in higher PT/INR results obtained from patients on warfarin therapy [7]. The CLSI H21-A5 standard [7] recommends the use of sample tubes with lower citrate concentrations (105–109 mM, 3.2%). According to the survey results, most of the laboratories (83.3%) are used 3.2% citrated tubes. However, it is interesting that 6.25% of the laboratories use 3.2% and 10.42% of the laboratories use both 3.2% and 3.8% citrate tubes (Table 2). Since there may be reference range variances between these two concentrations, laboratories should standardise their tubes and determine the reference intervals accordingly.

Hematocrit correction

A high hematocrit level (>55%) would change the anticoagulant/plasma ratio. This change can adversely affect coagulation test results [26]. CLSI H21-A5 recommends correction of citrate concentration for >55% hematocrit levels [7]. However, according to the survey results, 90% of the participating laboratories did not make such a correction (Table 2). The guide recommends the use of the formula “citrate concentration = [1.85 × 103] X [100 – hematocrit] X [blood volume]” for the hematocrit correction [7]. As a simpler practice, a protocol is provided indicating that before drawing the blood, 0.1 mL citrate is drawn from 5 mL tubes containing citrate of 3.2% [7].

Order of draw

The CLSI GP41-A6 and the TBS national venous blood collection guideline recommend the specific blood order of draw and drawing blood in the citrated tube in the first-order [21], [27]. Contamination of the coagulation test samples with other anticoagulants such as Ethylenediaminetetraacetic acid (EDTA) and lithium heparin and thrombin clot activators may result in serious problems for test results [7]. We asked the blood order of draw for the coagulation tube when biochemistry, whole blood count, sedimentation and coagulation tests are requested in the survey. However, the answers of the participants were quite impressive. Laboratory compliance with guidelines recommendation was 61.11%. In contrast, 10.2% of the laboratories indicated that there is no such order of draw, while 18.37% said they draw the sample in the second-order and, 10.2% said they draw the sample in the third place for coagulation tests (Table 2).

Sample tubes filling

The sample/anticoagulant ratios and filling the samples up to the volume as indicated on the sample tubes are critical for coagulation tests. According to CLSI H21-A5 standard, blood collection of ±10% of the nominal fill line would be sufficient for coagulation tests [7]. Answers to the question in the survey in this respect prove that 64% of laboratories comply with the respective guides’ recommendations concerning the tube nominal fill ratio. However, 34% of the laboratories accept a fill ratio of ≤80% and, 2% of the laboratories stated that they run coagulation tests without considering the tube filling ratio (Table 2). It should be kept in mind that an underfill of tubes might cause dilution of the sample and longer coagulation time due to the capacity of citrate to bind excessive calcium. In respect of the determination of the tube fill ratios, manufacturer recommendations and scientific articles are as important as the respective national and international guides. According to survey results, laboratory specialists are taken into consideration both national and international guidelines and manufacturers’ recommendations about the coagulation tube filling ratios. One of the remarkable findings of the survey was the high use of individual experiences for tube filling ratios (27.91%).

The participant laboratories have different practices for under or overfilling sample tubes. 76.6% of the laboratories rejected these samples according to their sample acceptance and rejection criteria. In contrast, 4.26% of laboratories stated that they ran these samples and reported the results and 10.64% indicated that they reported the results with comments and 8.51% said that they report the results when they are within the reference ranges.

Mixing of sample tubes

It is critical to mix the tube gently by inversion 3–5 times to provide adequate mixing of the anticoagulant and blood and prevent clot formation after blood sampling. The most important preanalytical error source in both hematology and coagulation laboratories are clotted samples. Besides, rigorous shaking would cause both in vitro hemolysis and artificial coagulation factor activation [28], [29]. As reported by two independent studies, if the tubes are adequately filled, and the tests analysed quickly, during the blood collection procedure the mixing of blood and anticoagulant may be adequate to enable reliable results for routine coagulation tests [28], [29]. According to our study, 64.1% of the participants mix the samples 3–6 times after blood sampling, while 35.9% of them mixed just 1–3 times (Table 2). In this respect, it would be appropriate for the laboratories to act following their conditions and the experiences. We believe that clotted sample rejection rates should be determined as indicators. Following these ratios at specified intervals and applying corrective and preventive action according to the predetermined target value will be a useful guide for laboratories.

Transport time and temperature

According to international guidelines [7], [9], whole blood coagulation samples should be transferred to the laboratory as soon as possible (ideally within 1 h) at ambient temperature (15–22 °C). Transfer of these samples in the cold environment is not recommended due to the activation of F VII, loss of the von Willebrand factor and, thrombocytes degradation. After sample collection, aPTT should be run within 4 h, and the PT/INR within 24 h [30], [31]. However, if the aPTT test is used for heparin therapy monitoring, the sample should be centrifuged within 1 h because of the neutralization of heparin by platelet factor 4 [32], [33]. Delay in transportation would result in extended coagulation time and loss of in vitro factor activities [34].

Both excessively cold and hot transport conditions should be avoided for coagulation tests. Transfer time is critically important, especially for unstable coagulation factors (F V, F VIII). Our survey results showed that 57.14% of the laboratories transport their coagulation test samples at the temperature and time suggested by the guides. Besides, 2% of the laboratories stated that they transport the samples at 4–8 °C within 1 h, while 8.16% of the laboratories transport the samples at 15–22 °C within >2 h.

Transport time and temperature is important for specific coagulation tests, especially labile factors (FV and FVIII). According to the survey, the sample transport temperature and duration of the laboratories running the specific coagulation test are varied. 26.92% of laboratories transport samples at 15–22 °C within 1 h, 19.23% of laboratories transport samples at 4–8 °C within 1 h; 7.70% of laboratories transport samples at 4–8 °C within >2 h and 7.69% of laboratories transport samples at 15–22 °C within >2 h 11.54% of laboratories use the pneumatic transportation system, and 11.54% of laboratories said that they draw the samples for specific coagulation tests in the laboratory.

A pneumatic system can be used for sample transport in hospitals. According to the survey result, 32.65% of the laboratories use a pneumatic system to transport the coagulation samples. Due to vibration during the transportation of samples with a pneumatic system, foaming, which causes platelet activation and protein denaturation may occur [7].

Centrifugation

Centrifugation is the most critical step in the process of obtaining platelet free plasma (PFP) for coagulation tests. Plasma is called as PFP when the number of platelets is <10 × 109/L (or 10 000/μL). In medical laboratories, swing-bucket and fixed-angle rotors are the most preferred centrifuge types for the sample preparation [34], [35]. However, swinging-bucket rotors are the most suitable in order to minimize the contamination of the plasma with platelets and other blood cells [7]. The ideal centrifugation conditions (at ambient temperature (15–22 °C), 1500 × g and 10–15 min) to obtain PFP are stated in the respective guidelines; however, the centrifugal force and time can be determined according to the laboratory conditions [7]. A shorter centrifugation time may be used for routine coagulation tests. However, these samples must not be used for additional specific coagulation tests. Higher (>1500 × g) centrifugal force is not preferred since it may cause platelet activation and hemolysis [36]. A double step centrifugation protocol may also be used to obtain PFP. According to survey results, 78% of the laboratories prepare their plasma samples with a single step centrifugation protocol whereas 22% use double step centrifugation protocol (Table 2).

In centrifuges, repulsion generated during the rotational motion acts radially from the center to the particles in the sample. It is known as the relative centrifugal force (RCF) or Gravity (g). RCF is expressed as multiples of gravitational acceleration (× g). RPM (number of rounds per minute) is the number of rotations per minute in the centrifuge and the speed indicator of the centrifuge. RCF and RPM are often used interchangeably and mixed. Since different centrifuges have a different radius, even if they have the same RPM, the RCF values will be different. Therefore, RCF provides the same acceleration and centrifugal force standardisation in different centrifuges [35]. Interestingly, although laboratories were asked to provide centrifugal force as RCF, 42% of them indicated it as RPM. It is essential using refrigerated-centrifuges for coagulation tests, but it can always be challenging to achieve. Non-refrigerated centrifuges can be used without being allowed to overheat [7]. Answers to the questions in the survey concerning the centrifugal force, temperature and time were significantly varied between laboratories, which causes one to think that there are essential problems in particular with the standardization of the centrifugation procedure used in the preparation of samples. We are considering organizing periodic training for users according to national and international guidelines in order to help in the standardisation of centrifugation procedure.

Hemolysis

Hemolysis occurs when the erythrocytes are lysed and, the content of the erythrocyte is released into the plasma. Inappropriate venous blood collection and sample preparation result in vitro hemolysis [7]. Some disease conditions, such as autoimmune or hemolytic anaemia, serious infections, intravascular disseminated coagulation, transfusion reactions, hereditary and aquired or iatrogenic conditions, may cause in vivo blood cell lysis as well [36, 37]. Hemolysis can cause both analytical and biological interference in coagulation tests [37, 38]. Free hemoglobin causes high absorbance readings in optical methods, while cytoplasmic and plasma membrane molecules such as tissue factors, proteases, phospholipids and ADP can cause erroneous results by inducing blood coagulation, and pseudo-activation in platelets [37, 38]. Grossly hemolysed samples should be rejected. However, if the patient has in vivo hemolysis and the test must be pursued, the mechanical endpoint analysis method is recommended, and the potential effect of activation should also be noted [36]. According to survey results the participant laboratories have different practices for hemolysed samples. 55.3% of the laboratories rejected these samples. In contrast, 4.3% stated that they run these samples and report the results, 36.2% indicated that they report the results with comment and 4.3% said that they report the results when these are within reference ranges (Table 2).

Lipemia

Hypertriglyceridemia can cause interference in the coagulation tests. This interference can occur as light scattering, volume displacement, as well as direct interaction of lipid particles with hemostasis reactions [37, 38]. Regarding lipemic plasma samples, out of the laboratories participating in the survey; 47.06% of them reject the sample; 50% of them report the results with a comment and 2.94% of them report the results without any explanation if the tests are urgent.

Icterus

Interference from icterus causes spectral interference [37, 38]. In modern coagulometers, special wavelengths are used, and icterus does not cause interference up to 20 mg/dL [37, 38]. In our study, the participant laboratories have different practices for icteric samples. 34.3% of the laboratories reject these samples. In contrast, 8.57% of laboratories stated that they run these samples and report the results and 54.29% indicated that they report the results with a comment and 2.86% said they provide a report for the results without any explanation thereon if the tests are urgent.