College of American Pathologists
Printable Version

  Q & A





cap today

April 2005

Richard A. Savage, MD, Editor

Q.  Document H21-A3 (1998) of the NCCLS, now CLSI, states that coagulation samples need to be respun at the adequate speed and for the proper length of time to produce platelet-poor plasma of less than 10,000 platelets.

Document H21-A4 (2003) has the added comment that prothrombin times and partial thromboplastin times may be performed on the first samples, continuing up to 200,000 platelets. Doesn’t this affect lupus anticoagulant and heparin testing?

A.  Platelets can be activated or disrupted by freeze-thawing, cooling, hemolysis, slow or difficult blood draws, movement through pneumatic tube systems, and other mechanisms. Activated or disrupted platelets can neutralize lupus inhibitors and heparin in plasma as well as cause other interference in coagulation assays. For this reason, it is routinely recommended that coagulation samples be free of hemolysis or visible clotting (fibrin strands) and centrifuged to remove platelets to a count of less than 10 x 109/L (10,000/µL).

The NCCLS document H21-A4 (section 5.1) indicates that "APTT, PT/INR, and TT [thrombin time] performed on fresh plasma samples are not affected by platelet counts of at least up to 200 x109/L (200,000/µL)." This was based on studies of platelet counts versus clotting times.1,2

While it is correct that on fresh, warm, unactivated platelet-rich plasma, routine APTT, PT/INR, and thrombin time were not significantly different than in platelet-poor plasma for the instrument-reagent combinations tested, these references state that this is not the case for samples being evaluated for heparin therapy or lupus inhibitors. In both of these cases, platelet-poor plasma is required. Because it is often difficult to ensure that a sample does not contain heparin, lupus anticoagulants, or activated platelets prior to running the sample for APTT, PT/INR, or thrombin time, the conservative approach is to centrifuge all samples to remove platelets. In addition, different instrument-reagent combinations may be more susceptible to interference from platelets.


  1. Carroll WE, Wollitzer AO, Harris L, et al. The significance of platelet counts in coagulation studies. J Med. 2001;32: 83-96.
  2. Barnes PW, Eby CS, Lukoszyk M. Residual platelet counts in plasma prepared for routine coagulation testing with the Beckman Coulter power processor. Lab Hematol. 2002;8:205-209.
Wayne Chandler, MD
Department of Laboratory Medicine
University of Washington

Member, CAP Coagulation Resource Committee

Q.  Does serum carbon dioxide content decrease when the assay tube sits unstoppered? If yes, how long does it take for a difference to occur, or how much of a decrease will occur for each hour that it is exposed? Is there a reference in the literature concerning this?

A.  Carbon dioxide levels can decrease if tubes are left unstoppered, according to a number of references in the literature. One study by Bandi noted that a mean of 4.1 mmol/L of CO2 was lost during routine laboratory processing. This study also reported losses of up to 8 mmol/L in open-cup instrument systems.1 Howse et al noted that specimens that were spun and stored at refrigerated temperatures did not lose as much CO2 as those stored at room temperature.2

Another factor that contributes significantly to underestimating CO2, especially in pediatric patients, is the underfilling of blood collection tubes, which causes evaporation loss of CO2 to the relatively large unfilled volume and aerobic condition in the unfilled tube. One author noted a 42 percent decrease in a significantly underfilled tube within four hours.3 Another study noted that carbon dioxide pressure declined within two hours of air exposure, but total CO2 did not deteriorate significantly within two hours of air exposure.4

The analysis of CO2 is optimized by minimizing the time tubes are left uncapped, centrifuging expeditiously, storing tubes at refrigerated temperature, and using the correct size tubes for the volume of collection.


  1. Bandi ZL. Estimation, prevention, and quality control of carbon dioxide loss during aerobic sample processing. Clin Chem. 1981;10:1676-1681.
  2. Howse ML, Leonard M, Venning M, et al. The effect of different methods of storage on the results of serum CO2 assays. Clin Sci (Lond). 2001;100:609-611.
  3. Williams AJ, Norris L, Coakley JC. Artefactual reduction in plasma total carbon dioxide concentration in children. Ann Clin Biochem. 2002;39:62-63.
  4. Kirschbaum B. Loss of carbon dioxide from serum samples exposed to air: effect on blood gas parameters and strong ions. Clin Chim Acta. 2003;334:241-244.
Kathleen Allen, MD
Medical Director
Quest Diagnostics

Member, CAP Chemistry Resource Committee