A physician in Ohio asks "What is the appropriate practice to quantify that the number of platelets in an apheresis unit meets the required 3x10¹¹? Since the platelet count in an undiluted aliquot of the blood product is so much greater than the normal reference range used by the hematology lab, would it be appropriate to dilute the sample to an acceptable range testing on the instrument? If so how are the dilutions QC'd? Are there any instruments available to measure counts in platelet rich plasma without first diluting the sample?"

James P. AuBuchon, MD of Dartmouth-Hitchcock Medical Center and the Chair of the CAP Transfusion Medicine Resource Committee (attribution used with permission) comments that the issue of counting platelets at the higher concentrations found in platelet concentrates remains a problematic one. He states,

"Most automated hematology analyzers are not specifically approved for counting platelet-rich plasma (PRP), and few have documented linearity into the concentrations encountered. (The future use of platelet additive solutions, where apheresis platelets are collected as a hyperconcentrated PRP, will exacerbate this problem.) One has only to review the results of CAP surveys in which the reported mean platelet concentration in PRP varies between instruments by 30-60%! At present, there is no standard means of dealing with this dilemma. (The CAP Transfusion Medicine Resource Committee (CAP TMRC) attempted to work with instrument manufacturers to address this problem two years ago, but the relatively small size of the market involved may have limited their interest in developing FDA-approvable protocols for PRP counting.) One approach that could be tried is to dilute the specimen until the count drops into the validated range. However, dilution carries with it its own inaccuracies, and the entire procedure would need to be validated for both accuracy and precision, of course. The degree of activation of the platelets may also affect their ability to be recognized and counted accurately (Hervig T et al. Vox Sang 2004;87:196-203.). The Biomedical Excellence for Safer Transfusion (BEST) Collaborative has also worked on this problem -- look for a publication in the near future. Unfortunately, the problem does not end with accurate platelet enumeration in the unit to be transfused: The accuracy of the platelet count that indicates whether a prophylactic platelet transfusion is necessary for a thrombocytopenic patient is also an important variable , and the use of different instruments can lead to very different decisions about when to transfuse (Segal HC et al. Br J Haematol 2005;128:520-5.). The CAP TMRC is hoping to (partially) address these issues by including in a future version of the CAP TRM accreditation checklist an item specifically requiring that platelet counting methods be validated for the range in which they are being applied (which, in my opinion, is just good laboratory practice). Therefore, the unquestioning acceptance of a platelet count from an automated instrument is inaproppriate without validation against a defined reference method."

ADDENDA Sept. 14, 2005

1. Gary Moroff, PhD of the Jerome H. Holland Laboratory of the American Red Cross in Rockville, MD (attribution used with permission) reports that in his experience hematology analyzers have the capability to directly measure platelet concentration (/µL) with precision in platelet products without initial dilution of the samples (Moroff, G, Sowemimo-Coker, SO, Finch S, et al., for the Biomedical Excellence for Safer Transfusion (BEST) Collaborative; The Influence of Various Hematology Analyzers on Component Platelet Counts. Transfusion Medicine Reviews 2005;19:155-166). He states: "The maximum platelet concentration that can be measured without initial dilution varies with hematology analyzer make/model and ranges from 1 x 10⁶/µL to 5 x 10⁶/µL. Linearity of platelet concentration measurements is a prerequisite for establishing a maximum concentration for a specific hematology analyzer."

Dr. Moroff continues: "Initial dilution of samples becomes necessary when the measured platelet concentration is above the maximum specified concentration as indicated in the use instructions for a specific hematology analyzer. With maximum specified levels of 2 x 10⁶/µL to 5 x 10⁶ /µL the need to dilute may be limited. A level of 2 x 10⁶/µL, for example, is equivalent to a total platelet number of 6 x 10¹¹ if the volume of the apheresis product or the pool of platelet concentration is 300 mL. Examples of hematology analyzers that have maximum platelet concentration levels equal to or above 2 x 10⁶/µL are: Cell Dyn 3700, 2 x 10⁶/µL; Advia 120, 3.5 x 10⁶/µL; Micros 60, 4 x 10⁶/µL, depending on software version; XE 2100, 5 x 10⁶/µL; and Baker 9100 series, 5 x 10⁶/µL."

Dr. Moroff adds that "Even with hematology analyzers that can measure relatively high platelet concentrations, dilutions can be made with platelet concentrations below the maximum specified levels. The rationale, however, for not diluting samples if such a procedure is not needed, is that dilution can add sources of error when determining platelet concentration/µL and hence total platelet number in a platelet product. When dilutions are made, the diluent of the hematology analyzer can be utilized.

The capability to measure high platelet concentrations should be documented by periodically determining the linearity range. Commercially available kits, usually consisting of 5 or 6 samples of simulated human platelets stabilized at specified levels, are available to document a linear range of platelet concentrations for a specific hematology analyzer. These samples are utilized without initial dilution (Moroff, G, Kreutter A, Parker J, Counting platelets in apheresis components. Transfusion 1995;35:1S, (abstract supplement).

The maximum specified level is one technical issue that needs to be considered when using a hematology analyzer. In addition, it needs to be realized that the specific make/model of a hematology analyzer can influence the measured platelet concentration/µL (Moroff et al, Transfusion Medicine Reviews report cited previously; College of American Pathologists, Results with PRP samples provided for testing; Lord, RA, Smith GA, Nightingale MJ et al, Platelet counting using plasma platelet concentrate samples. Transfusion Medicine 1992;2:201-205)."

2. Dr. Jerard Seghatchian of Blood Components Technology & Haemostasis/Thrombosis Consultancy in London (attribution used with permission), who is one of the editors of the book 'PLATELET THERAPY; current status and future trends' suggests that those who wish to count platelets in concentrates or plateletspheresis units should look at Chapter 13 pp 279-319 of the aforementioned book for a report on the RESIDUAL PLATELET AND LEUCOCYTE COUNTING by L Dumont & G Moroff. Dr Seghatchian also thinks that chapter 14 (which he authored himself) is germane to this discussion, since that chapter reports on practical aspects of platelet storage and quality. Finally, he suggests that colleagues look at the publication in TRANSFUSION APHERESIS SCIENCE 2002;26: 83-90. "A platelet quality assessment scheme for comparing the performance of quality monitoring laboratories on platelet concentrates produced in national transfusion blood service". In his view, there will be always some dilution error (10-15%) when counting platelet concentrates. This error is almost equivalent to counting undiluted platelet samples with modern counting instruments.

ADDENDA Sept. 20, 2005

3. Jennifer Rhamy MBA, MA, MT(ASCP)SBB, HP, who is Vice President of Laboratory Services at Indiana Blood Center (attribution used with permission) reports that her laboratory's quality assurance experience has been that the Cell Dyn 3500 will count platelets in a linear fashion for samples taken from platelet products, up to 2,000,000 platelets per µl, and that the samples taken from the products do not require dilution before testing. She adds that the equipment manufacturer's claim in the Operators' Manual is that the device has a linear range for platelet counting of 0-2000 K/µl with +/- 10.0 or 7% acceptable limit (page 4-17). Furthermore, she reports that their most recent annual linear range determination for the equipment, which was done during June 2005, validated the linearity of that range. She concludes by saying that their validation data match up nicely with the use of a Trima System, which in their hands produces plateletpheresis units with a platelet concentration of between 1,000,000 - 2,000,000/µl.

Please submit comments to the e-Network Forum.

Ira A. Shulman, MD
CBBS e-Network Forum Editor & Moderator