As you may recall, a member reported that her blood collection facility had encountered an increased incidence of WBC contamination of plateletpheresis products produced off their automated platelet collection equipment. This member expressed interest in hearing from other collecting facilities that collect plateletpheresis products using automated equipment. The member was especially interested in knowing what percent of the products tested were failing QC (that is having WBC counts that did not meet the leukocyte reduction standards). The member was interested in knowing if any trends had been identified as to the causes of the failures? The member also wanted to know:

• how are platelet products being monitored for contamination?
• how are products used for Quality Control being selected (by piece of equipment, random selection, sampling plan?)
• how many products (percent of total draw) are used for QC monthly?
• what corrective actions are being taken for product failures? (i.e., are the actions directed at the product - like filtration, or at the equipment - such as changes to equipment setting.)
• is there a mechanism to detect activated platelets?

To which the following replies were received:

1. I don't think one needs to point fingers, but it would be useful to know if these products are being leukocyte reduced by filtration (in-line or otherwise, from whatever the source) or not. I would think the fix for the problem would be radically different depending on the methodology even in this most general of terms. In the latter case, machine settings might be at the root of the problem. In the former case, I would think not.

2. Our QC failure (too many WBCs in final product) over the first quarter of 2000 was 14 units out 1207 sent for QC, or approximately 1.1%. The numbers are not calculated for the 2nd quarter, but I have not "heard" that the rate of failure has increased over the past 3 months. As far as "monitoring" for contamination, we perform our monthly QC (1% of previous month's collections, or a minimum of 4 products) per location, per instrument type (i.e., Amicus versus COBE). QC samples are selected at random by the Collections staff at the beginning of the collection process and are then sent to the QC lab for sampling. Our SOPs require investigation of any QC failures, including evaluation of potential sampling error, error in testing, error in collection process (donor count, length of run, alarms), handling and storage of components, checking for air contamination, fluids for possible contamination, kit defects, outdate, etc. Most of our failures are probably attributed to sampling problems. Once the investigation is complete, QC needs to be repeated. If problems still occur, if a particular instrument is isolated, that instrument needs to be taken out of service and the vendor contacted. As far as detecting activated platelets, there are three inspections for aggregates - at collection, at component production and prior to distribution. No other method is used.

3. At our facility, leukocyte-reduced plateletpheresis products are collected using Gambro.BCT Spectra and Trima instruments. The Trima instruments have been operational only since June 2000. A review of one year of WBC counts (7/99 - 6/00) showed the following: 0% - QC failures for meeting the 5 x 10⁶ standard and 3.8% - QC failures for meeting the 1x 10⁶ standard (quoted by Gambro.BCT as their standard and one which we use internally). We noted a cluster of QC failures in late 1999 that the vendor responded to by sending their staff to assess our collection, product sampling and Nageotte counting techniques. The vendor's staff made no significant observations, but they did offer a few technique tips and subsequently we have had no further problems other than an occasional QC failure. QC sampling is done on one product per instrument per month. Products are randomly sampled from procedures that did not have excessive alarms or WBC measure messages. The sample of one product per instrument per month equals roughly 2.5 - 3.0% of all products collected. Corrective action for QC failures includes testing of an additional product from the same instrument that was associated with a QC failure. Based on supervisor judgment, the QC results from testing the additional product may cause the implicated instrument to be removed from service with a request for the vendor to check function. Products with failed QC are filtered. We do not have a mechanism to detect activated platelets.

4. 2 of 293 Cobe LRS units tested in the last 12 months (0.7%) failed WBC QC (the same machine was implicated in both failures which occurred within a 3 week period). Units are tested according to a Quality Control sampling plan: a minimum of 4 units per center per month (6 total centers) are tested for WBC count and platelet recovery; this is roughly 3.5% of total Cobe LRS units processed (based on 8400 total LRS units processed per year). Units that fail WBC QC are labeled and released as regular platelets (i.e., not leukocyte reduced); NO filtration is performed subsequent to obtaining unacceptable WBC count results in order to achieve leukocyte reduction. When a WBC QC failure occurs, an investigation is immediately initiated and the involved machine is quarantined. The vendor is also called in to investigate. No mechanism is used to detect activated platelets other than gross visual inspection for aggregation. In addition to the Cobe LRS failures noted above, there have been some WBC QC failures with the Amicus machines.

5. We are in the process of implementation of Trima equipment. We are having what sounds like what may be a similar issue in certain collections and are working with Cobe-Gambro to resolve the issue.

Editor's NOTE: It is interesting that some blood collection agencies will filter the plateletpheresis units that contained too many WBCs and therefore failed QC. It is also interesting that other collection agencies will not filter products that failed leukocyte reduction QC, but will still distribute them without labeling them as Leukocyte-reduced. While these practices might be safe and efficacious, in our current regulatory and litigious environment, I wonder how many transfusion services would accept a plateletpheresis product if they knew the product had failed Quality Control (for any reason)? Do any transfusion services have a vendor qualification that prohibits shipment of products that failed QC?

Please submit comments to the e-Network Forum.

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