6.2 Setting and maintaining specifications - Evaluation and manufacture of blood components

The wide variability of the source material from which blood components are prepared makes it difficult to set stringent limits. Nevertheless, realistic minimum specifications should be set and complied with.

Concessionary release limits are also set for certain components that are subject to non-destructive quality monitoring, such that components that are excessively out of specification are only used therapeutically under specific circumstances and subject to formal clinical approval (see Table 6.1.1 to Table 6.1.13).

All blood components tested and found outside concessionary release limits should have the testing repeated to confirm the original result and the production process should be reviewed as necessary.

Some abnormal results could have implications for the health of the donor and should be reviewed by a Designated Clinical Support Officer. Further samples or referral to an appropriate clinical service may be required for the donors of:

• Red Cells in Additive Solution, Leucocyte Depleted, where the Hct is >0.70.
• Platelet components collected by apheresis, where the platelet count is below any lower concessionary release limit.
• For Fresh Frozen Plasma, Leucocyte Depleted where the FVIII is <0.3IU/mL.

6.2.1: Concessionary release limits

6.2.1.1: Red cell components

Table 6.1.1: Red Cells in Additive Solution, Leucocyte Depleted

Parameter	Lower limit	Upper limit
Haemoglobin	30 g/unit	None
Volume	210 mL	375 mL
Haematocrit	0.40 L/L	0.70 L/L

Table 6.1.2: Red Cells, Washed, Leucocyte Depleted

Parameter	Lower limit	Upper limit
Haemoglobin	30 g/unit	None
Volume	210 mL	375 mL
Haematocrit	0.40 L/L	0.70 L/L
Residual protein	None	0.50 g/unit

Table 6.1.3: Red Cells for Intrauterine Transfusion, Leucocyte Depleted

Parameter	Lower limit	Upper limit
Haemoglobin	40 g/unit	None
Volume	150 mL	350 mL
Haematocrit	0.70 L/L	0.85 L/L

Table 6.1.4: Red Cells for Exchange Transfusion (not Whole Blood), Leucocyte Depleted

Parameter	Lower limit	Upper limit
Haemoglobin	40 g/unit	None
Volume	220 mL	420 mL
Haematocrit	0.50 L/L	0.60 L/L

Table 6.1.5: Red Cells in Additive Solution for Neonates and Infants, Leucocyte Depleted

Parameter	Lower limit	Upper limit
Haemoglobin	30 g/unit per number of splits manufactured	None
Haematocrit	0.40 L/L	0.70 L/L

6.2.1.2: Platelet components

Table 6.1.6: Platelets, Pooled (see note 1)

Parameter	Upper limit	Lower limit
Platelet yield	160×10⁹/unit	Defined by pack type (see note 2)
Volume	150 mL	380 mL

Table 6.1.7: Platelets, Apheresis, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Platelet yield	160×10⁹/unit	Defined by pack type (see note 1)
Volume	150 mL	380 mL

Table 6.1.8: Platelets in Additive Solution, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Platelet yield	160×10⁹/unit	Defined by pack type (see note 1)

Table 6.1.9: Platelets for Intrauterine Transfusion, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Volume	50 mL	120 mL
White blood cells	None	2.5×10⁶/unit

Table 6.1.10: Platelets, Neonatal Use, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Platelet yield	40×10⁹/unit	Defined by pack type (see note 1)
Volume	30 mL	95 mL

6.2.1.3: Plasma components

Table 6.1.11: Fresh Frozen Plasma, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Volume	200 mL	340 mL
Factor VIII	0.3 IU/mL	None
Residual platelet count	None	100×10⁹/L (see note 1)

Table 6.1.12: Fresh Frozen Plasma, Pathogen Reduced, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Residual platelet count	None	100×10⁹/L (see note 1)

Table 6.1.13: Fresh Frozen Plasma for Neonates and Infants, Leucocyte Depleted

Parameter	Upper limit	Lower limit
Factor VIII	0.3 IU/mL	None
Residual platelet count	None	100×10⁹/L (see note 1)

6.2.2: Statistical process monitoring

Component and process quality monitoring results should be subjected to statistical analysis so that trends can be identified. Guidance on appropriate approaches to statistical process monitoring is given in the Council of Europe guide and in Beckman et al [Beckman, 2009].

A flowchart adapted from Beckman et al, to aid in selection of appropriate methods, is reproduced as Figure 6.1.

Description of Figure 6.1: Algorithm for the selection of quality monitoring methods

1. Is the parameter influenced by the collection or production process?
   1. If no, continue to step 2.
   2. If yes, continue to step 3.
2. Are specification limits or performance criteria defined?
   1. If no, quality monitoring is not indicated.
   2. If yes, conformance to specification monitoring should be selected as a quality monitoring method.
3. Is criticality ‘low’?
   1. If no, continue to step 4.
   2. If yes, conformance to specification monitoring should be selected as a quality monitoring method.
4. Are specification limits or performance criteria defined?
   1. If no, define limits then continue to step 5.
   2. If yes, continue to step 5.
5. Are data variable or attribute?
   1. If variable, continue to step 6.
   2. If attribute, a statistical process control (SPC) chart for attributes or scan statistics should be selected as a quality monitoring method.
6. Are data normally distributed (after removing special cause failure data)?
   1. If no, transform data then continue to step 7.
   2. If yes, continue to step 8.
7. Are data now normally distributed?
   1. If no, a statistical process control (SPC) chart for attributes or scan statistics should be selected as a quality monitoring method.
   2. If yes, continue to step 8.
8. Is subgroup sampling applicable?
   1. If no, an individual/moving range (I/MR) chart or cumulative sum (CUSUM) control chart should be prepared before continuing to step 9.
   2. If yes, an X-bar/R chart or cumulative sum (CUSUM) control chart should be prepared before continuing to step 9.
9. Do the control charts reveal any reproducible special cause variation?
   1. If no, continue to step 11.
   2. If yes, continue to step 10.
10. Identify root cause and implement corrective action then continue to step 11.
11. Is the process capable?
    1. If no, audit the process design and control and identify and make improvements. Consider a post-improvement review. If improvements cannot be made, review the specifications and then continue to step 12.
    2. If yes, continue to monitor.
12. Is it possible to redefine specifications?
    1. If no, refer to legislation and guidelines setting committees. Conformance to specification monitoring should be selected as a quality monitoring method.
    2. If yes, implement the refined specifications and continue to monitor.

If the results of analyses show a consistent trend towards the minimum requirements specified in chapter 7, the cause should be investigated. The criteria to be investigated must be detailed in the relevant SOP together with the corrective action to be taken.

The steps to be considered should include the following:

• An investigation of the collection, testing, production and distribution procedures as appropriate.
• Checking that procedures are up to date and are not being deviated from.
• Checking the operation of equipment and storage conditions (this may include reviewing validation documentation and/or revalidation).

The person responsible for quality assurance and/or production may initiate investigations beyond the scope of written procedures.

References

Beckman N, Nightingale MJ, Pamphilon D (2009). Practical guidelines for applying statistical process control to blood component production. Transfusion Medicine, 19, 329–339. https://doi.org/10.1111/j.1365-3148.2009.00942.x