8.47 It is the purpose of performance qualification to establish the standard of performance expected with a particular operating cycle and loading condition, so that subsequent production cycles can be judged by that standard. The evidence for this performance is provided by the indicated, recorded and measured cycle variables, and it is necessary to determine how much each variable will be permitted to vary from cycle to cycle while still conforming to that standard.
8.48 A starting point is the limits prescribed for the cycle variables in the
commissioning and PQ tests described in this SHTM. Other than in exceptional circumstances, these limits should be regarded as absolute, and a failure to meet them implies a failure of the test and a gross failure of the sterilizer. These limits originate from European and British Standards and from
operational experience. They are set to accommodate a wide range of sterilizer models and designs of operating cycles. However, an individual sterilizer
should be able to repeat a cycle well within these limits, and the permitted tolerances for PQ purposes should be correspondingly smaller.
8.49 It is important that the tolerances are set with careful consideration of the likely range of variation from cycle to cycle. If set too tight, acceptable production loads may be erroneously rejected as non-sterile, and automatic control and PRQ tests may fail unnecessarily. However, it would be a mistake to set an over-generous tolerance, since that may disguise variations signalling a
developing malfunction of the sterilizer. The following paragraphs give guidance on determining the permitted tolerances. The authorised person should be consulted in cases of doubt.
8.50 PQ tests (or commissioning tests providing PQ data) collect indicated, recorded and measured data (see paragraph 7.2-7.20 for an explanation of these terms). The three sets of data serve different purposes and may require different
tolerances:
a. indicated data are available to the user for production cycles on all types of sterilizer, but cannot be regarded as definitive. Except for sterilizers without a recorder, PQ tests require indicated values to be noted only during the holding time to ensure that they comply with the sterilization conditions;
b. recorded data are available to the user for production cycles on most types of sterilizer and can be regarded as definitive for routine production control. The permitted tolerances are normally marked on a master process record (see paragraph 8.58). The user should be aware of any calibration
c. measured data are not available for production cycles and so play no part in routine monitoring. However, they are to be regarded as definitive for the purposes of performance requalification. Measured variables are more reliable than Indicated or recorded values, and the permitted tolerances should reflect this.
8.51 A further consideration is the intended use of the PQ data:
a. PQ data valid for a single loading condition: where the PQ data are to be used for one loading condition only, the variation between cycles is essentially random (that is, due to uncontrolled variables or the intrinsic performance limits of the sterilizer) and the permitted tolerances can be tight. This is appropriate, as such cases are often used for loads which would be damaged if the limits were broader. The tolerances should be set by experience of the sterilizer and of the cycle. The three replicate
thermometric PQ tests (see paragraph 8.13) will give some indication of what variation to expect;
b. PQ data valid for a range of loading conditions: where the PQ data for a single loading condition is judged to be valid for a range of loading
conditions, the variation between cycles will contain a systematic variation related to the differing loading conditions and therefore the permitted tolerances will be greater. The choice of loading conditions for which the data is valid should take into account whether this greater tolerance is acceptable;
c. PQ data obtained from commissioning tests: for many loads, especially on porous load sterilizers, PQ tests are not normally necessary and data from the thermometric commissioning tests are used to establish
performance standards for a wide range of loading conditions. In these cases, data from the small-load and full-load tests can be used to establish the limits of variation for production loads which fall between these two extremes. The permitted tolerances will be broader than either (a) or (b). 8.52 Note that the permitted tolerances during the holding time of an operating cycle
will generally be tighter than those allowed during the preceding and following stages. In no circumstances should these tolerances permit the cycle variables to depart from the sterilization conditions specified in Table 8, unless the operating cycle has been designed with that intention.
8.53 Tolerances are normally expressed as a permitted variation either side of a central value, either in absolute terms or as a percentage. In some cases the tolerances may be expressed as an upper or lower limit, with the variables permitted to take any value on the safe side of the limit.
PQ report
8.54 All the data collected during PQ tests should be filed in a PQ report, a copy of which should be kept with the plant history file. The report should contain or refer to the complete specification for the sterilization process. The
specification should be detailed enough to allow the loading condition, the operating cycle and the test itself to be replicated on any future occasion. The report should include the following:
a. a specification of the loading condition, defined either by the nature and number of load items, items of chamber furniture, and their distribution within the chamber, or by a coded reference to a detailed specification held elsewhere;
b. a specification of the operating cycle, defined either by the settings for the cycle variables or by a coded reference to a detailed specification held elsewhere;
c. a specification of any preconditioning, conditioning and degassing process (this is essential for EO sterilizers);
d. all the indicated, recorded and measured data from the test, drawing attention to the values and permitted tolerances of elapsed time and of the indicated, recorded and measured cycle variables at all significant points of the operating cycle, for example at the beginning and end of each stage or sub-stage (the tolerances in recorded variables should also be marked on the master process record);
e. for loads which require the removal of air before sterilization, the method used to verify whether the minimum conditions of steam penetration into the load are attained (for porous load sterilizers, this is by use of the air detector);
f. the original of the master process record derived from the test.
8.55 EO sterilizers require extensive additional data for safety and quality control purposes and these are listed in Table 11.
8.56 Immediately following the PQ tests, the test person should prepare PQ
summary sheets (see Appendix 3) and working copies of any necessary master process records. These should be given to the user and kept with the sterilizer process log.
8.57 If PQ tests are not required, the PQ summary sheet should contain data from the thermometric test for a full load and be marked accordingly.