Preanalytic Variables: Effects on CBC and Serum Chemistry Results
In both human and veterinary medical practice, current trends indicate a move toward increased point-of-care capabilities. When point-of-care technologies are used efficiently, this translates into better customer service and enhances the practice of medicine. It also leads to improvements in practice financial health. However, these technologies depend on the skill and knowledge of the user to give accurate results. In veterinary medicine, one of the most critically important steps in laboratory analysis is blood sample collection. Determining the levels of the various cellular and chemical constituents of blood can provide valuable diagnostic information when test results are accurate.
Many preanalytic factors other than disease influence the results of diagnostic tests.1 Veterinary technicians must be familiar with each test methodology used to avoid errors caused by improper sample handling. This article covers a few of the preanalytic factors most commonly encountered in clinical practice.
The preanalytic period begins with the preparation of patients and materials for the sampling procedure and continues through sample collection and specimen handling up to beginning the specific laboratory analyses. Preanalytic factors may be biological or nonbiological. Biological variables are factors that are inherent to the patient, such as breed, age, and sex. Because these cannot be controlled, they must be considered when evaluating test results. Other biological variables involve factors that can be controlled when drawing the blood sample, such as ensuring the animal is properly fasted. Nonbiological variables are related to sample collection and handling. Preanalytic errors are significantly more common than analytic errors.2 The impact of preanalytic factors on test results depends on the analyzer and methods used (TABLE 1).
TABLE 1 Effects of Sample Compromise
|*Variable effect, depending on the analyte and test method used.
Courtesy of Elsevier. From Sirois M. Laboratory Procedures for Veterinary Technicians. 6th ed. St. Louis, MO: Mosby; 2015.
|Hemolysis/blood substitutes||Release of analytes||↑|
|Release of enzymes*||↑ ↓|
|Increased optical density (absorbance)||↑|
|Release of water||↓|
|Analyte binding*||↑ ↓|
|Volume displacement||↑ ↓|
|Medications||Reaction interference*||↑ ↓|
Lipemia, the presence of excessive lipoproteins in the blood, is common in postprandial blood samples. Lipemic blood samples are turbid (FIGURE 1) because of the presence of large lipid particles, which can absorb or scatter most wavelengths of light and thus interfere with many spectrophotometric test methods. Lipemia can also create a phenomenon known as volume displacement, in which lipoproteins comprise a larger than normal fraction of plasma, which reduces the fraction of plasma composed of water. This falsely decreases electrolyte levels when using analyzers that measure electrolyte concentration from the total plasma volume. Because lipoprotein concentrations in blood plasma can quickly reach levels that interfere with laboratory testing and persist for several hours after a patient eats, it is recommended that patients be fasted for 12 hours before blood collection.3
Hemolysis is perhaps one of the most commonly encountered sources of sample compromise and has an array of effects on both hematologic and clinical chemistry assays (FIGURE 2). Hemolysis increases the light absorbance of blood serum or plasma, which particularly interferes with chemistry tests that read in the ultraviolet/visible wavelengths.4 Increased free hemoglobin in plasma can also directly inhibit some chemical reactions. Additionally, hemolysis results in release of analytes and enzymes from red blood cells (RBCs), which can falsely elevate many test results.5 Destruction of RBCs yields a lower RBC count.1 Excess fluid released from the lysed RBCs also creates a dilution effect in serum and can result in an artifactually decreased packed cell volume.
Many factors can destroy RBCs; while nearly all are a result of improper collection and handling, some may be a result of patient disease processes, such as immune-mediated hemolytic anemia. Administration of certain blood products, such as hemoglobin-based oxygen carriers, can also cause hemolysis. This must be considered when choosing test methods because most chemistry analyzers are incapable of providing accurate readings on samples from these patients.
To avoid the common causes of hemolysis, blood should be collected and handled as atraumatically as possible. Repeated attempts to draw blood from the same vessel are undesirable. However, if this cannot be avoided, there are other methods to reduce physical trauma to the sample. Such practices include using the largest-gauge needle that the patient can tolerate, limiting the amount of negative pressure created when drawing back on the syringe plunger, and, most importantly, removing the needle from the syringe and the cap from the collection tube before transferring blood. Additionally, because isopropyl alcohol can cause sample hemolysis, collection sites should be clipped and excess alcohol allowed to evaporate before venipuncture is performed.
Veterinary technicians should alert the veterinarian about any sample characteristics that may interfere with analysis so that the veterinarian may better interpret the results. If a technician notices a sample to be lipemic or hemolyzed, it should be noted in the patient’s medical record. Varying intensities of lipemia or hemolysis can be described as slight, moderate, or marked.
Veterinary technicians must ensure that the anticoagulant chosen does not interfere with the blood constituent(s) being assayed. Citrate and oxalate anticoagulants can be used for plasma samples but interfere with some biochemical testing and damage RBCs. Fluoride inhibits in vitro glycolysis by RBCs and is a useful anticoagulant when preservation of glucose is critical.4
Ethylenediaminetetraacetic acid (EDTA) preserves cellular components for CBC and morphologic evaluation but falsely decreases calcium, phosphorus, and alkaline phosphatase levels and elevates potassium levels; therefore, EDTA plasma samples should not be used for chemical analysis. Conversely, lithium heparin is a generally safe anticoagulant for plasma chemistry and electrolyte analysis, but it distorts blood cells and should not be used for evaluation of cell morphology in samples from mammals.6
Clotting, Centrifugation, and Separation
To prevent the initiation of blood clotting processes, several guidelines should be followed when collecting whole blood or plasma samples. The coagulation cascade can be triggered by excess venous stasis (>1 minute) or by excessive probing with the needle. Excess venous stasis also alters the composition of the sample as a result of water and electrolytes moving from the intravascular to the extravascular space.3 It is important that samples be collected with minimal trauma to avoid these common errors. Anticoagulant tubes should be gently mixed by inverting several times immediately after collection. Failure to follow these guidelines can result in platelet clumping, which yields low platelet counts by automated machines and also makes it impossible to accurately estimate platelet numbers on a blood film. Additionally, formation of clots in a blood sample may reduce other cell counts. Whole blood samples with evidence of clotting should not be used for testing when the test utilizes whole blood.
When collecting blood for the purpose of obtaining serum, blood should be allowed adequate time to clot. If a blood sample is centrifuged before the clotting process has completed, the serum may retain fibrin strands that can alter analyzer readings.4 Unless the blood collection tube contains clot accelerators, the sample should be left undisturbed at room temperature for 20 to 30 minutes while the clot is forming. Immediately after clotting, a wooden applicator stick should be used to gently separate the clot from the walls of the tube (FIGURE 3) and the sample should then be centrifuged for 10 to 20 minutes at 1000g.
Immediately after centrifugation, plasma or serum should be removed with a pipette, transferred to a plain tube, and labeled. Numerous blood constituents will be affected if the cells are allowed to remain in contact with serum or plasma. Generally, decreases are seen in glucose and calcium levels while phosphorus and potassium are increased in affected samples.7 Even though barrier gels separate blood cells from fluid components, they are capable of absorbing certain hormones and drugs, such as progesterone and phenobarbital.8 In addition, centrifuging samples using a fixed-head rotor centrifuge can allow gaps to develop in the gel barrier.4 This is why it is recommended to separate serum and plasma samples after centrifugation regardless of whether a separator gel is present in the tube.
When multiple types of samples are required, the samples should always be collected with the vacuum system, as it ensures that an appropriate volume of each sample type is obtained. EDTA is hypertonic, and excess EDTA results in crenation of RBCs, which in turn drastically reduces the RBC count. The correct-size tube must be used to minimize damage to the sample and the possibility of collapsing the patient’s vein. If using the vacuum system, the tubes must be filled in a specific order to avoid the potential contamination of samples with additives from other tubes (TABLE 2).
TABLE 2 Order of Draw for Commonly Used Blood Collection Tubes
If processing is delayed, samples should be refrigerated or frozen according to laboratory protocols and type of tests ordered. Storing at cooler temperatures is generally preferred as it slows in vitro reactions of chemical components. However, all samples should be allowed to warm to room temperature before processing because cold temperatures can inhibit certain chemical reactions necessary for blood analysis. Additionally, chemistry samples should also be stored away from ultraviolet light because prolonged exposure reduces bilirubin levels.4
A variety of factors can influence the quality of CBC and serum chemistry results. Intrinsic patient factors, such as age, breed, and presence of disease, cannot be controlled with human intervention. This strengthens the argument for exercising all reasonable methods to reduce nonbiological preanalytic errors. Blood sample collection and handling protocols should be implemented and followed by all practice personnel to avoid variation in the accuracy of laboratory results. For example, to avoid lipemia, staff should not feed or offer treats to patients that may require blood testing during their visit. When scheduling appointments, receptionists should instruct owners to fast patients when necessary. Protocols to reduce hemolysis of samples should require staff to use appropriately sized needles when drawing blood and to remove needles from syringes before transferring samples to collection tubes. Consistency in preparation of samples can be enhanced with the use of automated instruments (FIGURE 4). Other guidelines related to method of collection, sample handling, and storage of collected samples should be followed to further help minimize effects of preanalytic errors on hematology and serum chemistry results (BOX 1).
- American Society for Veterinary Clinical Pathology (ASVCP). Quality Assurance for Point-of-Care Testing in Veterinary Medicine. Available at asvcp.org/pubs/qas/index.cfm. Accessed August 2015.
- Baron JM, Mermel CH, Lewandrowski KB, Dighe AS. Detection of preanalytic laboratory testing errors using a statistically guided protocol. Am J Clin Pathol 2012;138(3):406-413.
- Narayanan S. The preanalytic phase: an important component of laboratory medicine. Am J Clin Pathol 2000;113:429-452.
- Humann-Ziehank E, Ganter M. Pre-analytical factors affecting the results of laboratory blood analyses in farm animal veterinary diagnostics. Animal 2012;6(7):1115–1123.
- Bell R, Harr K, Rishniw M, Pion P. Survey of point-of-care instrumentation, analysis, and quality assurance in veterinary practice. Vet Clin Path 2014;43(2):185-192.
- Harvey JW. Veterinary Hematology. St. Louis, MO: Saunders; 2012.
- Joshi A. Variations in serum glucose, urea, creatinine and sodium and potassium as a consequence of delayed transport/processing of samples and delay in assays. J Nepal Med Assoc 2006;45(161):186-189.
- Dasgupta A, Dean R, Saldana S, et al. Absorption of therapeutic drugs by barrier gels in serum separator blood collection tubes. Volume- and time-dependent reduction in total and free drug concentrations. Am J Clin Path 1994;101:456-461.