Pathogen reduction technologies and the shelf life of platelet concentrates

Author: Mikel Lezaun is professor at the University of the Basque Country – UPV/EHU where he is director of the Mathematical Technology Transfer Group

Production centers for blood components have the job of supplying hospitals with the products that they need in time and strive to discard as few units as possible due to outdating. In the case of platelet concentrates (PC) it is hard to meet these goals for several reasons. Their short shelf life (a maximum of five days, extendable to seven if a bacterial detection method is used or they are treated with a pathogen reduction technology) and growing demand due to an aging population and increasingly aggressive pharmacological treatments.

Pathogen reduction technology (PRT) for PC first emerged at the end of the 20th century to increase protection against infection, since it inactivates viruses, parasites, and bacteria. This technique allows the shelf life of PC to be extended to seven days and can replace irradiation because it inactivates residual leucocytes. However, in spite of the studies published on it to date, its use is not yet widespread due to cost.

Platelets are used for cancer patients during chemo. They have a very short shelf life and regular donations are needed to keep people alive. Photo: ABC Hobart / Carol Rääbus

In this work 1, we study:

  1. The reduction in outdates when average shelf life is extended from five to seven days.

  2. The increase in the remaining average useful lifetime of PCs when they are dispatched to hospitals.

  3. The effect of distributing treatment across different weekly patterns of production days when, for budgetary reasons, only part of the PCs can be treated.

The reduction in outdates brought about by extending the shelf life of PC from five days to seven is estimated using a mathematical simulation. Two production centers in Spain are considered: the CVTTH (Basque Centre for Transfusions and Human Tissue) and the BSTA (Aragon Blood & Tissue Bank). The CVTTH produces 11.500 PCs per annum and the BSTA 6.200. Each is the only production center in its region.

Both centers produce PCs on only five days each week (Monday to Friday at the CVTTH and Tuesday to Saturday at the BSTA). In CVTTH, orders produced on one day are incorporated into stocks for shipment to hospitals the next day with an age of one day, except in the case of those produced on a Friday, which go into stocks on the following Monday with three days. The main difference of the BSTA with the CVTTH is that does not carry out bacterial cultures, so units whose production are ordered from Tuesday to Friday can be sent to hospitals on the same day of their production, with an age of one day. The units produced on a Saturday become available as stock on Monday, because the laboratory does not run the relevant tests over the weekend.

In both scenarios, the model parameters are calculated using the demand data at the BSTA in 2016, taking advantage of the fact that the demand of each of day of the week fits into a normal distribution. Once the models are defined, they are applied to the real data of demand at BSTA from hospitals in 2017. This result in values for the number of outdates, the minimum daily stock, and the average remaining shelf life of PC sent to hospitals.

As usual in inventory systems, which are regularly reviewed, an order-up-to production policy is used. This consists of setting an order-up-to value for each day of the week. Once demand for the day has been met, an order is placed for the production of sufficient PC to reach the order-up-to value of the day. The order-up-to value is where and are the mean and the standard deviation of the demand to be covered by the order and k is constant used to increase the safety stock. For supplies a FIFO (first in, first out) policy is assumed.

The models are implemented on an easy-to-use Excel spreadsheet. The procedure entails calculating production day by day throughout the year, in line with the demand to be met. Implementation is immediate. The results are very similar in the two centers.

Days of treating

% PCs treated

Outdate %

Min. stock

Average remaining lifetime in days

Mon to Fri

100%

0,27%

10

3,23

Mon to Fri

45%

0,48%

10

2,17

Mon to Wed

3,77%

4

2,46

Thu and Fri

1,26%

10

2,21

Mon to Fri

25%

1,29%

10

1,77

Mon to Wed

3,94%

8

1,99

Thu and Fri

1,26%

10

1,76

Mon to Fri

12%

2,40%

10

1,64

Mon to Wed

4,09%

8

1,78

Thu and Fri

1,65%

10

1,57

Mon to Fri

6%

3,55%

10

1,58

Mon to Wed

4,26%

8

1,66

Thu and Fri

2,75%

10

1,52

Mon to Fri

0%

4,54%

8

1,55

This table shows the results obtained for the whole year in the CVTTH with on Monday, Tuesday and Wednesday and on Thursday and Friday. The “Min. stock” column shows the minimum daily stock obtained over the whole year. The “Average remaining lifetime in days” shows the mean remaining lifetime of PCs when they are sent to hospitals. The percentage indicates the annual percentage of PC treated. The days of the week indicate on which days the treatments are carried out. In those days, not all PCs have to be treated, only those that correspond to them to obtain the defined annual percentage.

The results show that a 2-day extension in the shelf life of PC results in drastic reductions in outdates at the production centers analyzed. In fact, if all the PCs produced are treated, expiration is reduced from a 4,54% until almost a 0%. In addition, the average remaining lifetime of PCs increases by two days.

It may be the case for budgetary reasons that only part of the PCs produced can be treated. This being so, we show that if the proportion treated per annum exceeds 25% the best option is to treat part of the output every day, otherwise, it is preferable to concentrate treatment on the last two production days of the week.

References

  1. C. Gorria, G. Labata, M. Lezaun, F. J. López, A. I. Pérez Aliga & M. A Pérez Vaquero: Impact of implementing pathogen reduction technologies for platelets on reducing outdates. Vox Sanguinis 2020; 115:167-173 doi: 10.1111/vox.12860

Written by

Leave a Reply

Your email address will not be published. Required fields are marked *