Dealing with these samples can be challenging, but understanding how to work with them will result in the best outcomes for patient care.
Medical errors are typically preventable events that may cause or lead to inappropriate medication use or patient harm. Diagnostic errors are an important part of medical errors and are defined as any defect from ordering tests. This includes errors from reporting of results, the appropriate interpretation of the tests and reaction of these results.1
Laboratory medicine is typically divided into three main phases (pre-analytical, analytical and post-analytical). Data has shown that problems directly related to specimen collection are the main cause of pre-analytical errors, which may be as high as 84.5%.2 The complete elimination of laboratory testing errors is unrealistic, especially those relating to the pre-analytical phases. Most of these processes are not under direct laboratory supervision and are harder to control.
Most vascular cell damage occurs during phlebotomy; this is the most frequent reason for specimen rejection. According to the College of American Pathology Q-probe survey, it occurs five-fold (54%) more frequently than insufficient specimens (21%).1 In vitro hemolysis, which can occur during phlebotomy, causes cell membrane disruption and leakage of hemoglobin into the surrounding fluid. It can occur from improper specimen collection due to a wrong needle, excessive mixing of the blood sample, inadequate storage temperatures or rough handling during specimen transport. Hemolysis also can occur when blood is drawn from a peripheral IV catheter.
In vivo hemolysis can occur from a pathological condition such as autoimmune hemolytic anemia or a transfusion reaction.3
The leakage of intracellular analytes that occurs from hemolysis can produce a wide variation in coagulation results (i.e., results can be falsely elevated or there can be dilutional effects).
It cannot be assumed, however, that because of hemolysis, results are either prolonged or shortened. It has been speculated that hemolysis provides a source of tissue factor that can cause coagulation factors to activate and shorten results. Another theory is that the process of hemolysis competes with coagulation reagents, resulting in the prolongation of coagulation results.4
The Clinical and Laboratory Standards Institute guidelines for prothrombin time (PT) and activated partial thromboplastin time (aPTT) testing states: “Samples with visible hemolysis should not be used because of possible clotting, factor activation and interference with endpoint measurement.”5
The best solution for hemolysis is to recollect the sample; however, that isn’t always possible. Many laboratories report results and alert clinicians to interpret the results in the presence of hemolysis. The most common methodologies to measure clotting are optical density and electromechanical. Optical density uses a light that passes through a specimen and measures the change in turbidity. Electromechanical measures the increase in viscosity as a clot forms. There is no advantage to one method of measurement since hemolysis is a process that can interfere with the plasma sample and, consequently, coagulation results. Using a result that may not truly represent a correct result can be dangerous in the diagnosis and management of a patient.
Lipemic plasma has large lipid particles that include lipoproteins and chylomicrons. As a result, these samples have increased sample turbidity and may result in the prolongation of coagulation results. Interference is variable among analyzers. Turbid samples cause attenuation of the intensity of light passed through a sample due to scatter, reflectance or absorption.6
Large lipid particles may be removed from samples by ultracentrifugation. However, this may result in the sedimentation of the large molecular protein masses of fibrinogen or factor VIII/von Willebrand complex. Another method that may clarify plasma is by adding n-hexane to plasma samples to clear the lipids. This would have to be verified by laboratories to determine if the analyte of interest is affected.7
The interference of lipemia may also be minimized by using higher dilutions. When using immunoturbidimetric methods for automated, quantitative D-dimer testing and von Willebrand factor activity and antigen, the higher the dilution that can be used, the less the chance of lipid-based interference. Samples diluted 1:20 may show lipid interference; this may be removed by increasing the dilutions up to 1:400.6
Icteric plasma contains high levels of bilirubin. Normal levels of bilirubin are about 0.5 mg/dL. In cases of hyperbilirubinemia, levels will exceed 1.5 mg/dL and plasma will become affected.
Icteric plasma samples have a high prevalence in samples from patients in the intensive care unit, as well as gastroenterology, surgical and pediatric patients. Concentrations of bilirubin greater than 2.5 mg/dL can lead to clinically relevant changes of anti-thrombin. Higher concentrations can interfere with other coagulation tests.8
Studies have demonstrated that using higher wavelengths at 570 nm on samples that are icteric demonstrate concordance to electromechanical methods, making this an alternative method to measure these samples.9
Optical vs. Mechanical Methods
Laboratories use either photo optical or mechanical methods to measure coagulation results. There have always been questions as to which methodology has an advantage over the other in samples with hemolysis and lipemia or icteric samples. Optical clot detection systems have the capability to use multiple wavelengths to try to eliminate issues associated with visual interferences.
A large study compared more than 2,000 samples using both methodologies. Of those samples, 26.5% had visual interferences. The study demonstrated that results obtained by the photo-optical detection system are as reliable and statistically equivalent as those obtained using the mechanical detection system. The correlation between the analyzers resulted in an r>0.96 for all samples and an r >0.98 on samples that were turbid.10
Another small study looked at lipemic samples (n=10) using a mechanical method of clot detection and a photo-optical method of clot detection following ultracentrifugation. There was no difference in APTT and PT results when comparing the ultracentrifuged samples read by the photo-optical method and the mechanical methodology.6
- Plebani M. Errors in clinical laboratories or errors in laboratory medicine. Clin Chem Lab Med 2006;44: 750-759.
- Lippi G, Guidi GC, Mattiuzzi C, Plebani M. Prenalytical variability: The dark side of the moon in laboratory testing. Clin Chem Lab Med 2006;44: 358-365.
- Arzoumanina, L. What is hemolysis? Tech Talk, Vol 2 No 2 Franklin Lakes, NJ: Decton Dickinson; October 2003).
- Laga AC, Cheves TA, Sweeney JD. Specimen hemolysis and coagulation testing: The effect of specimen hemolysis on coagulation test results. American Journal of Clinical Pathology 2006;126: 748-755.
- CLSI Document H3-A4 Procedures for the Collection of Diagnostic Blood Specimens by venipuncture;Approved Standard, 4th ed. Wayne, PA: 1998.
- Adcock D. Questions and Answers. CAP Today December 2002.
- Arambarri M, Oriol A, Sancho JM, et al. Interference in blood coagulation tests on lipemic plasma. Correction using n-hexane clearing. Sangre (Barc). 1998;43:13-19.
- Guder WG, da Fonseca-Wollheim F, Heil W, et al. The haemolytic, icteric and lipemic sample recommendations regarding their recognition and prevention of clinically relevant interferences. Recommendations of the Working Group on Preanalytical Variables of the German Society for Clinical Chemistry and the German Society for Laboratory Medicine Journal of Laboratory Medicine 2000;24(8).
- Junker R, Käse M, Schulte H, Bäumer R, Langer C, Nowak-Göttl U,. Interferences in coagulation tests-Evaluation of the 570-nm method on the Dade Behring BCS Analyzer. Clin Chem Lab Med 2005;43(2):244-52.
- Bai B, Christie D, Gorman R, Wu J. Comparison of optical and mechanical clot detection for routine coagulation testing in a large volume clinical laboratory. Blood Coag Fibrinolysis 2008; 19:569-576.