Assessment of factors affecting the platelet yield using continuous flow cell separator

Senior Resident, Department of Transfusion Medicine, Post Graduate Institute of Medical Education and Research, Chandigarh Assistant Professor, Department of Transfusion Medicine, Government Medical College and Hospital, Chandigarh Professor, Department of Transfusion Medicine, Government Medical College and Hospital, Chandigarh Assistant Professor, Department of Pathology, Government Medical College and Hospital, Chandigarh

IJBR (2014) 05 (03) www.ssjournals.com segment to ensure representative product of the bag and subjected to determination of platelet count after appropriate dilution (1: 5) with sample diluents on an automated hematology analyzer or with ammonium oxalate for manual count on Neubauer Chamber to calculate the yield.

Statistical analysis
Influence of donor variables such as predonation platelet count and Hb% on the yield of platelets was studied by Pearson correlation coefficient and multivariate linear regression using SPSS Inc., Chicago, IL, and version 15.0 for Windows. All statistical tests were two-sided and performed at a significance level of α = 0.05. A p value of < 0.05 was taken as significant.

Results
During the study period, a total of 171 healthy donors (mean age 28.1 ± 7.6 years) weighing 72 ± 11.1 kg underwent plateletpheresis procedures on continuous flow cell separator (Amicus Automated Blood Collection System version 2.52/3.1). The mean blood volume processed was 3030.7 ± 633 ml over the mean duration of 66.8 ± 14.3 minutes by using 400.5 ± 67.1 ml of acid citrate dextrose (ACD). After each procedure , Hb concentration increased from 13.6 ± 1.6g/dl to 15.3 ± 1.2 g/dl and hematocrit ( hct) increased from 42.7 ± 3.4 % to 43.43.6 ± 3.6 %, the difference was found to be statistically significant (p < 0.001). The platelet count deceased from 241 ± 57.1 x 10 3 /µl to 170.28 ± 48.86 × 10 3 /µl after plateletpheresis and was found to be statistically significant (30%; p < 0.001). However, the decrease in MPV & PDW was not statistically significant. The changes in the donor hematological variables before and after plateleletpheresis are summarized in Table 1 Table 2 shows correlation between predonation platelet count and the yield of platelets. The mean platelet yield of all procedures was 3.8+ 1.65 X 10 11 /unit. Figure 1 shows a direct correlation between predonation platelet count and the platelet yield (r = 0.446, p < 0.001). The platelet count was > 3x 10 11 in 83% of procedures.

Fig.1 shows a correlation between Predonation platelet count and the yield of platelets
Platelet yield was negatively correlated with MPV (r = -0.447) and PDW (r = -0.396) and was not found to be statistically significant as shown in Figure 2 and Figure 3 respectively. Figure 4 shows that there was no significant correlation of Platelet yield with predonation Hb (r =-0.004, p = 0.954). Similarly, multiple linear regression analysis demonstrated no effect of donor age and weight on platelet yield.

Discussion
Plateletpheresis is becoming a routine procedure in most of blood centres in developing countries. There have been significant improvements in productivity and quality of apheresis platelets. However, donor safety issues with regards to reduction in hematological values have received little attention. The present study addressed the changes in hematological parameters after plateletpheresis and influence of donor hematological parameters on the platelet yield.
On evaluation of donor related parameters on platelet yield, we found that predonation platelet count has significant linear correlation with the platelet yield (r = 0.446, p < 0.001). The higher the platelet count means more platelets are available for collection. Out of the 171 donors assessed for predonation platelet count effect on platelet yield, 46 (26%) had a pre-donation platelet count < 200 x 10 3 /µl. The mean yield of product prepared from these donors was 3 x 10 11 /unit. 93 donors had a pre-donation count in the range of 200-300 x 10 3 /µl and the mean yield in the product from these donors was 3.7 x 10 11 /unit. 32 donors had pre-donation counts > 300 x 10 3 /µl and the mean yield in the products prepared was 5.3 x 10 11 /unit.
Chaudhary et al in a study of 94 plateletpheresis procedures found a mean yield of 2.8 ± 0.73 x 10 11 /unit and they found that when the predonation platelet count was greater than 300 x 10 3 /µl, the yield was greater than 3 x 10 11 platelets/unit in 80% of the products. They also reported that the mean yield was 2.5 ± 0.59 x 10 11 /unit when the pre-donation platelet count was < 200 x 10 3 /µl 9. Our results were in accordance with these observations and the platelet yield correlated linearly with the pre-donation platelet count of the donor Goodnough et al studied 708 plateletpheresis procedures and a direct correlation between platelet yield and pre-donation platelet count was observed in all the procedures. In 12% of the procedures, the mean yield was < 3 x 10 11 /unit when the pre-donation platelet count was < 200 x 10 3 /µl 4 . Many other studies have also demonstrated significant positive correlation between the donor platelet count and the yield of the product 10-12 . According to the AABB requirements 13 , 75% of the plateletpheresis products prepared must contain ≥ 3 x 10 11 platelets per unit, while the European guidelines (Council of Europe publishing, 2006) recommend platelet count of ≥ 2 x 10 11 /unit 14 . These levels have been determined from the studies to provide required haemostatic dose to the recipient. 129 (76%) numbers of procedures in our study have platelet yield of > 3x 10 11 /unit. We observed negative correlation between the donor Hb and platelet yield (r = -0.004) and it was statistically insignificant (p = 0.954). Other authors also reported no correlation between pre-donation Hb and the yield 15, 16 . Chaudhary et al found no correlation between the pre-donation Hb and the yield (r = -0.10, p > 0.005), but in three donors with Hb > 16 g/dl gave a lower yield as compared with donors having Hb < 16g/dl 9 . In contrast, an inverse relationship between the Hb and the yield has been demonstrated by Guerrero-Rivera et al and Enien et al 6,17 . They concluded that this might be related to the greater amount of plasma processed in donors with low Hb.
We also studied the effect of donor MPV on platelet yield. In our study, we found a direct negative correlation between the mean MPV of the donor and the platelet yield obtained (r = -0.447, p < 0.001). 51 (29%) of donors had MPV in the range of 7-9 fl and the mean platelet yield in these donors was 4.9 ± 1.7 x 10 11 /unit. Maximum number of donors (49%) had MPV in the range of 9-11fl and the mean platelet yield obtained in this range was 3.5 ± 1.4 x 10 11 /unit. 30 (17%) donors had MPV in the range of 11-13 fl and the mean yield obtained in this group was 2.95 ± 1.2 x 10 11 /unit. Hence, it was observed that higher MPV values corresponded to lower platelet yield. This could be explained on the basis of separation mechanism of the Amicus. The separation of platelets by automated cell separators like Amicus is based on cell size. As the size of platelets increase, these large platelets are excluded from collection mimicking as red cells. Hence, the smaller platelets are collected more efficiently by the machine and hence yielding a better product.
Although we expect a decrease in hematological values after plateletpheresis procedure, adverse clinical outcomes, such as thrombocytopenia and anemia as a result of these decreases should be prevented. In our study, the mean pre-donation platelet count of the donors was 241 ± 57.1 x 10 3 /µl with a range of 149 -409 x 10 3 /µl. The mean post-donation platelet count was 170.2 ± 48.8 x 10 3 /µl. We found that the post-donation platelet count showed a significant decrease in all of the procedures (p < 0.001). The mean decrease in the post-donation platelet count was 74.0 ± 36.6 x 10 3 /µl (30.7%). The postdonation platelet count was less than 100 x 10 3 /µl in 4 procedures, fortunately with no clinical manifestations. Significant reductions in platelet count after plateletpheresis have also been reported by other researchers 18,19 . Lazarus et al did not find any donors with clinical thrombocytopenia among 939 donors who underwent 11,464 procedures 20. After plateletpheresis procedure, we observed 1.7% increase in post donation Hb values and was statistically significant (p <0.001). This might be due to the fact that concentrated red cells were returned to the donor and 200-300 ml of plasma was retained at the end of the procedure. The increase in Hb may be apparent as the sample was withdrawn immediately after the procedure. Love et al reported an increase in Hb and hct after plateletpheresis 18 . Other studies demonstrated conflicting results i.e. decrease in Hb and hct 15,21 . This could be explained on the basis that there was a time-lag between the end of procedure and obtaining sample which allowed physiological mechanisms to come into action. Another factor causing variation in results may be the IJBR (2014) 05 (03) www.ssjournals.com cell separator used. Taking into account the above observations, post-donation changes in the donors should be reviewed in those donors whose hematological parameters are border-line or who undergo frequent plateletpheresis. We observed post-procedure change in donors MPV and PDW. The post-donation decrease in the MPV was 0.0752 (0.7%) which was not statistically significant (p = 0.116). The post-donation decrease in PDW was 0.115 (0.3%) which was not statistically significant (p > 0.500). Das et al conducted a study on a total of 477 plateletpheresis procedures using five different cell separators over a period of 26 months. It was observed that there was no significant change in MPV or PDW of the donor after each procedure (p > 0.5) 15 .
On evaluating the effect of donor clinical variables such as age and weight on yield, there was no significant correlation (r = 0.067, p > 0.01). Chaudhary et al also studied the effect of donor weight on platelet yield and did not find any significant correlation 9. Buchholz et al studied the quality of SDP in relation to low weight of the donors and demonstrated no effect of donor weight on platelet yield 22 .
There were certain limitations in our study. The donor sample was obtained soon after completion of the procedure. Hence, the change in the hematologic values observed may not be representative of changes taking place over a longer time period (1 or 2 days). In addition, in our study, biochemical parameters were not considered at all and it is possible that post procedure, there could be biochemical changes which have not been looked into. These issues are more important in determining the suitability of frequent plateletpheresis donors. The 1990 UK guidelines recommend that frequent apheresis donors should be assessed for blood counts, total protein and albumin 23 . Also, since double dose platelet collections (collection of two units from a single donor) were not performed in any of our cases, we could not address issues regarding donor safety and quality of products in these settings. The Drugs and Cosmetics Act, which is regulatory body in blood transfusion services in India, should establish guidelines for repeat plateletpheresis donors 24 .
There is a need that all transfusion centers should have their own plateletpheresis donor data based on demographic and hematological characteristics of the donor population. This will ensure donor safety, product quality and an effective donor registry. More prospective studies are required to establish guidelines for donor safety in apheresis, especially in the present trend of double dose platelet collections.