Procedural Apheresis Instrumentation Algorithm Minimizes Platelets Loss during Peripheral Blood Stem Cells Collection.

Introduction: Peripheral blood hematopoietic progenitor cell collection is a more popular source of stem cells for transplantation than bone marrow harvest. A number of automated cell separators exist which can be used to perform mononuclear cell collections. These separators collect varying amounts of other cells during the collection process, including them in the mononuclear cell product. Because of this, collection of mononuclear cells can result in significant reductions in platelet count and loss of red blood cells during the procedure. In order to minimize platelet transfusion support, we developed an algorithm to optimize mononuclear cell collections while minimizing platelet losses using the Gambro COBE Spectra and the Baxter Amicus Methods: Previously, patients were assigned randomly to any available apheresis machine. Starting August 2nd 2002, patients with a WBC count higher than 35×10 (9)/L were preferably assigned to Amicus for 5 hour collection with processing 1000 ml/cycle or as second choice Spectra for 5 hours run with maximum draw rate 65 ml/Min., for both instruments. If the WBC was lower than 35×10 (9)/L, the preferred machine would be Amicus for 5 hours run at 1400 ml/cycle with maximum draw rate 90ml/Min. or Spectra for 5 hours run with maximum draw 100 ml/Min. We reviewed the difference of pre and post procedure platelet counts of non Hodgkin’s Lymphoma patients undergoing stem cells collection for 6 months before and after the implementation of the above protocol from February 2002 till February 2003. All CBC were performed on the same Coulter analyzer. Results: 14 patients (2 on Amicus and 12 on Spectra) underwent stem cells collection before August 2nd, 2002. 13 patients (8 on Amicus and 5 on Spectra) were collected after this date. The Fisher’s exact test was used for this study. The median and mean differences in the first group were 141.5 ×10(9)/L and 135 ×10 (9)/L respectively (95% CI, p 0.0183). The second group had a median and mean of 45 ×10 (9)/L and 51 × 10(9)/L (95% I, p 0.0183). The minimum and maximum differences were 36 × 10(9)/L and 200 × 10(9)/L respectively in the first group and −19 ×10(9)/L and 121×10(9)/L for the second group. There was no difference in the numbers of the collections days in both groups in regard to the type of automated cell collector. Discussion: The implemented algorithm showed a net advantage. Even though none of the patients required blood transfusions, there was lesser drop of platelets counts with the new settings.