ISO/DIS 11737-1:2026(en)

ISO/TC 198

Secretariat: ANSI

Date: 2026-01-09

Sterilization of health care products — Microbiological methods — Part 1: Determination of a population of microorganisms on products

© ISO 2026

All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below or ISO’s member body in the country of the requester.

ISO copyright office

CP 401 • Ch. de Blandonnet 8

CH-1214 Vernier, Geneva

Phone: +41 22 749 01 11

Email: copyright@iso.org

Website: www.iso.org

Published in Switzerland

Contents

Foreword vii

Introduction viii

1 Scope 1

2 Normative references 1

3 Terms and definitions 1

4 General 4

5 Selection of products 5

5.1 General 5

5.2 Sample item portion (SIP) 5

6 Methods of determination and microbial characterization of bioburden 6

6.1 Determination of bioburden 6

6.1.1 Selection of an appropriate method 6

6.1.2 Neutralization of inhibitory substances 6

6.1.3 Recovery of microorganisms 6

6.1.4 Detecting viable microorganisms 7

6.1.5 Enumeration of microorganisms 7

6.2 Microbial characterization of bioburden 7

7 Validation of the method for determining bioburden 7

7.1 General 7

7.2 Validation 8

8 Routine determination of bioburden and interpretation of data 8

8.1 General 8

8.2 Limits of detection and plate counting 8

8.3 Microbial characterization 8

8.4 Use of bioburden data for determination of the extent of treatment 8

8.5 Bioburden spikes 8

8.6 Bioburden levels 9

8.7 Data analysis 9

8.8 Statistical methods 9

9 Assessment of change for bioburden determination 9

9.1 Changes to the product or manufacturing process 9

9.2 Changes to the method for determining bioburden 9

9.3 Requalification of the method for determining bioburden 9

9.4 Review of appropriateness of bioburden levels 9

Annex A (informative) Guidance on the determination of a population of microorganisms on products 10

Annex B (informative) Guidance on methods to determine bioburden 26

Annex C (informative) Establishment of bioburden recovery efficiency 37

Annex D (informative) Guidance on bioburden nethod suitability testing for health care products 46

Annex E (informative) Guidelines for counting plates and recording results 49

Annex F (informative) Bioburden excursions 53

Annex G (informative) Typical assignment of responsibilities 56

Annex ZA (informative) Relationship between this European standard and the General Safety and Performance Requirements of Regulation (EU) 2017/745 aimed to be covered 58

Annex ZB (informative) Relationship between this European standard and the General Safety and Performance Requirements of Regulation (EU) 2017/746 aimed to be covered 62

Bibliography 65

Foreword

ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been established has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.

The procedures used to develop this document and those intended for its further maintenance are described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types of ISO documents should be noted. This document was drafted in accordance with the editorial rules of the ISO/IEC Directives, Part 2 (see www.iso.org/directives).

ISO draws attention to the possibility that the implementation of this document may involve the use of (a) patent(s). ISO takes no position concerning the evidence, validity or applicability of any claimed patent rights in respect thereof. As of the date of publication of this document, ISO had not received notice of (a) patent(s) which may be required to implement this document. However, implementers are cautioned that this may not represent the latest information, which may be obtained from the patent database available at www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.

Any trade name used in this document is information given for the convenience of users and does not constitute an endorsement.

For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions related to conformity assessment, as well as information about ISO's adherence to the World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/iso/foreword.html.

This document was prepared by Technical Committee ISO/TC 198, Sterilization of health care products in collaboration with the European Committee for Standardization (CEN) Technical Committee CEN/TC 204, Sterilization of medical devices, in accordance with the Agreement on technical cooperation between ISO and CEN (Vienna Agreement).

This fourth edition cancels and replaces the third edition (ISO 11737-1:2018), which has been technically revised.

The main changes are as follows:

— incorporation of the amendment content.

A list of all parts in the ISO 11737 series can be found on the ISO website.

Any feedback or questions on this document should be directed to the user’s national standards body. A complete listing of these bodies can be found at www.iso.org/members.html.

Introduction

A sterile health care product is one that is free of viable microorganisms. International Standards that specify requirements for the validation and routine control of sterilization processes require, when it is necessary to supply a sterile health care product, that adventitious microbiological contamination of a health care product prior to sterilization be minimized. Such products are non-sterile. The purpose of sterilization is to inactivate the microbiological contaminants and thereby transform the non-sterile products into sterile ones.

The kinetics of inactivation of a population of microorganisms by physical or chemical agents used to sterilize health care products can generally best be described by an exponential relationship between the numbers of microorganisms surviving and the extent of treatment with the sterilizing agent. Inevitably, this means there is always a finite probability that a microorganism can survive regardless of the extent of treatment applied. For a given treatment, the probability of survival is determined by the number and resistance of microorganisms and by the environment in which the microorganisms exist during treatment. It follows that the sterility of any one product in a population subjected to sterilization processing cannot be guaranteed and the sterility of a processed population is defined in terms of the probability of there being a viable microorganism present on a product item.

Generic requirements of the quality management system for design and development, production, installation and servicing are given in ISO 9001 and particular requirements for quality management systems for medical device production are given in ISO 13485. The standards for quality management systems recognize that, for certain processes used in manufacturing, the effectiveness of the process cannot be fully verified by subsequent inspection and testing of the product. Sterilization is an example of such a process. For this reason, sterilization processes are validated for use, the performance of the sterilization process is monitored routinely and the equipment is maintained.

International Standards specifying procedures for the validation and routine control of the processes used for the sterilization of health care products have been prepared (see, for example, ISO 14937, ISO 11135, the ISO 11137 series, ISO 13004, ISO 17665, ISO 14160, ISO 20857, ISO 22441 and ISO 25424). However, it is important to be aware that exposure to a properly validated and accurately controlled sterilization process is not the only factor associated with the provision of assurance that the product is sterile and, in this respect, suitable for its intended use. Furthermore, for the effective validation and routine control of a sterilization process, it is important to be aware of the microbiological challenge that is presented in the process, in terms of number, characteristics and properties of microorganisms.

The term “bioburden” is used to describe the population of viable microorganisms present on or in a product and a sterile barrier system. A knowledge of bioburden can be used in a number of situations as part of the following:

— validation and requalification of sterilization processes;

— routine monitoring for control of manufacturing processes;

— monitoring of raw materials, components or packaging;

— assessment of the efficiency of cleaning processes;

— evaluation of change in a manufacturing process or location;

— an overall environmental monitoring programme.

Bioburden is the sum of the microbial contributions from a number of sources, including people, raw materials, manufacturing of components, assembly processes, manufacturing environment, assembly/manufacturing aids (e.g. compressed gases, water, lubricants), cleaning processes and packaging of finished products. To control bioburden, attention should be given to the microbiological status of these sources.

It is not possible to enumerate bioburden precisely and, in practice, a determination of bioburden is relative to the defined method used. Definition of a single method for use in determining bioburden in all situations is not practicable because of the wide variety of designs and materials of construction of health care products. Nor is it possible to define a single technique to be used in all situations for the extraction of microorganisms in preparation for enumeration. Furthermore, the selection of culture conditions for enumeration of microorganisms will be influenced by the types of microorganism likely to be present on or in health care products.

This document specifies the requirements to be met for the determination of bioburden. In addition, it gives guidance in the annexes to provide explanations and methods that are deemed suitable to conform with the requirements. Methods other than those given in the guidance may be used, if they are effective in achieving conformity with the requirements of this document.

Sterilization of health care products — Microbiological methods — Part 1: Determination of a population of microorganisms on products

This document specifies requirements and provides guidance on the enumeration and microbial characterization of the population of viable microorganisms on or in a health care product, component, raw material or package.

NOTE 1 The nature and extent of microbial characterization is dependent on the intended use of bioburden data.

This document does not apply to the enumeration or characterization of viral, prion or protozoan contaminants. This includes the extraction and detection of the causative agents of transmissible spongiform encephalopathies, such as scrapie, bovine spongiform encephalopathy and Creutzfeldt-Jakob disease.

NOTE 2 Guidance on inactivating viruses and prions can be found in ISO 22442‑3, ICH Q5A(R1) and ISO 13022.

NOTE 3 ISO/TS 22456 provides specific guidance for bioburden testing for biologics and tissue-based products where this testing is conducted in relation to product sterilization.

This document does not apply to the microbiological monitoring of the environment in which health care products are manufactured.

There are no normative references in this document.

For the purposes of this document, the following terms and definitions apply.

ISO and IEC maintain terminology databases for use in standardization at the following addresses:

— ISO Online browsing platform: available at https://www.iso.org/obp

— IEC Electropedia: available at https://www.electropedia.org/

3.1

action level

value from monitoring that necessitates immediate intervention

[SOURCE: ISO 11139:2018, 3.5]

3.2

alert level

value from monitoring providing early warning of deviation from specified conditions

[SOURCE: ISO 11139:2018, 3.11]

3.3

batch

defined quantity of a product (3.18) intended or purported to be uniform in character and quality produced during a specified cycle of manufacture

[SOURCE: ISO 11139:2018, 3.21]

3.4

bioburden

population of viable microorganisms on or in a product (3.18) and/or sterile barrier system (3.24)

[SOURCE: ISO 11139:2018, 3.23]

3.5

bioburden correction factor

numerical value applied to a viable count to compensate for incomplete removal of microorganisms from a product (3.18) and/or failure to culture microorganisms

[SOURCE: ISO 11139:2018, 3.24]

3.6

bioburden estimate

value established (3.11) by applying a bioburden correction factor (3.5) to a bioburden (3.4) count

Note 1 to entry: The recovery efficiency can also be used to determine the bioburden estimate.

[SOURCE: ISO 11139:2018, 3.25, modified — “bioburden” has been added to the referenced term “correction factor.”]

3.7

bioburden method suitability

<microbiological> assessment of the test method to demonstrate its ability to allow microbial growth

[SOURCE: ISO 11139:2018, 3.168, modified — “bioburden” has been added to the term.]

3.8

bioburden spike

individual bioburden (3.4) value that is significantly greater than other bioburden values in a set

[SOURCE: ISO 11139:2018, 3.26]

3.9

corrective action

action to eliminate the cause of a nonconformity and to prevent recurrence

Note 1 to entry: There can be more than one cause for a nonconformity.

Note 2 to entry: Corrective action is taken to prevent recurrence whereas preventive action (3.17) is taken to prevent occurrence.

[SOURCE: ISO 9000:2015, 3.12.2, modified — Note 3 to entry has been deleted.]

3.10

culture condition

combination of growth media and manner of incubation used to promote germination, growth, and/or multiplication of microorganisms

Note 1 to entry: The manner of incubation can include the temperature, time, and any other conditions specified for incubation.

[SOURCE: ISO 11139:2018, 3.70]

3.11

establish

determine by theoretical evaluation and confirm by experimentation

[SOURCE: ISO 11139:2018, 3.107]

3.12

excursion

data exceeding an established level

Note 1 to entry: Bioburden results are typically evaluated as averages of a number of individual values.

[SOURCE: AAMI TIR106:2024, 3.2]

3.13

facultative microorganism

microorganism capable of both aerobic and anaerobic metabolism

[SOURCE: ISO 11139:2018, 3.114]

3.14

health care product

medical device, including in vitro diagnostic medical device, or medicinal product (3.18), including biopharmaceutical

[SOURCE: ISO 11139:2018, 3.132]

3.15

microbial characterization

process by which microorganisms are grouped into categories

Note 1 to entry: Categories can be broadly based, for example, on the use of selective media, colony or cellular morphology, staining properties, or other characteristics.

[SOURCE: ISO 11139:2018, 3.170]

3.16

obligate anaerobe

organism that only lives and grows in the absence of molecular oxygen

[SOURCE: ISO 11139:2018, 3.186]

3.17

preventive action

action to eliminate the cause of a potential nonconformity or other potential undesirable situation

Note 1 to entry: There can be more than one cause for a potential nonconformity.

Note 2 to entry: Preventive action is taken to prevent occurrence whereas corrective action (3.10) is taken to prevent recurrence.

[SOURCE: ISO 9000:2015, 3.12.1]

3.18

product

tangible result of a process

EXAMPLE Raw material(s), intermediate(s), sub-assembly(ies), health care product(s) (3.15).

[SOURCE: ISO 11139:2018, 3.217]

3.19

product family

group or subgroup of product (3.18) characterized by similar attributes determined to be equivalent for evaluation and processing purposes

[SOURCE: ISO 11139:2018, 3.218]

3.20

recovery efficiency

<bioburden> measure of the ability of a specified technique to remove, collect, and/or culture microorganisms from a product (3.18)

[SOURCE: ISO 11139:2018, 3.225]

3.21

requalification

repetition of part or all of validation (3.25) for the purpose of confirming the continued acceptability of a specified process

[SOURCE: ISO 11139:2018, 3.220.5]

3.22

sample item portion

SIP

specified part of a health care product (3.14) that is tested

[SOURCE: ISO 11139:2018, 3.240, modified — acronym SIP has been added.]

3.23

sterile

free from viable microorganisms

[SOURCE: ISO 11139:2018, 3.271]

3.24

sterile barrier system

SBS

minimum package that minimizes the risk of ingress of microorganisms and allows aseptic presentation of the sterile (3.23) contents at the point of use

[SOURCE: ISO 11139:2018, 3.272]

3.25

terminally sterilized

condition of a product (3.18) that has been exposed to a sterilization process in its sterilized barrier system

[SOURCE: ISO 11139:2018, 3.296]

3.26

validation

confirmation process, through the provision of objective evidence, that the requirements for a specific intended use or application have been fulfilled

Note 1 to entry: The objective evidence needed for a validation is the result of a test or other form of determination such as performing alternative calculations or reviewing documents.

Note 2 to entry: The word “validated” is used to designate the corresponding status.

Note 3 to entry: The use conditions for validation can be real or simulated.

[SOURCE: ISO 9000:2015, 3.8.13, modified — “process” has been added to the definition.]

4.1 The development, validation and routine control of bioburden are critical elements in the realization of some types of health care products. To ensure the consistent implementation of the requirements specified in this document, the necessary processes shall be established, implemented and maintained. Processes of particular importance in relation to the development, validation and routine bioburden control of a process include but are not limited to:

— control of documentation, including records;

— assignment of management responsibility;

— provision of adequate resources, including competent human resources and infrastructure;

— control of product provided by external parties;

— identification and traceability of product throughout the process; and

— control of non-conforming product.

NOTE 1 ISO 13485 covers all stages of the lifecycle of medical devices in the context of quality management systems for regulatory purposes. National or regional regulatory requirements for the provision of health care product can require the implementation of a full quality management system and the assessment of that system by a recognized conformity assessment body.

NOTE 2 See Annex G for information on typical assignment of responsibilities.

4.2 A process shall be specified for the calibration of equipment, as applicable, used in meeting the requirements of this document (e.g. instrumentation for test purposes).

5.1.1 The procedures for the selection and handling of products for the determination of bioburden shall ensure that the selected product is representative of routine production, including primary packaging materials and processes, but prior to the terminal sterilization process, if applicable. Terminally sterilized product should not be used for bioburden testing.

5.1.2 If product(s) are grouped in a product family for the purpose of the determination of bioburden, the rationale for inclusion of a product within a product family shall be documented. The rationale shall include criteria to ensure that bioburden determined for a product selected from the product family is representative for the whole product family.

5.1.3 Consideration shall be given to the timing of the determination of bioburden relative to manufacturing because bioburden can change with the passage of time.

5.2.1 The determination of bioburden may be performed on:

a) the entire product (SIP = 1,0);

b) a portion of the product, e.g. half of the product for an SIP of 0,5, or;

c) the sterile fluid path for which sterility is claimed (SIP = 1,0).

5.2.2 The selection of sample portions for an SIP <1,0 shall be based on the following criteria and should be reviewed periodically:

a) if the bioburden distribution is known and is evenly distributed, the SIP may be selected from any portion of the item;

b) if the bioburden distribution is not known or is known and is unevenly distributed the SIP shall include:

1. portions of the product selected that proportionally represent each of the materials (e.g. 30 %) composing the product; or

2. alternatively, only where the bioburden distribution is known and is unevenly distributed, the SIP can be selected from the portion of the product that contains the most severe microbial challenge (numbers and types) to the sterilization process.

5.2.3 The SIP can be calculated on the basis of measurable physical characteristics, such as length, mass, volume or surface area (see Table A.1 for examples).

NOTE Some standards specify requirements for validation and routine control of the sterilization process stipulate criteria for the adequacy of the SIP, e.g. the ISO 11137 series.

The method shall be appropriate to the purpose for which the data are to be used. The method(s) shall comprise techniques for the following:

a) neutralization of inhibitory substances, if needed;

b) extraction of microorganisms, if appropriate;

c) a means of detecting viable microorganisms e.g. culture or non-culture based;

d) enumeration of microorganisms.

If the physical or chemical nature of the product or SIP is such that substances can be released that adversely affect the detection of the product bioburden, then a system shall be used to neutralize, remove or, if this is not possible, minimize the effect of any such released substance. The effectiveness of such a system shall be demonstrated.

NOTE Annexes B and D describe techniques that can be used to assess the release of microbicidal or microbiostatic substances.

6.1.3.1 For a product that cannot be directly cultured and where the extraction of microorganisms is part of the method, the need for recovery studies (or bioburden recovery efficiency) shall be considered using risk-based approach and the outcomes of this consideration documented. Consideration shall, at least, be given to the following:

a) ability of the technique to remove microorganisms;

b) possible type(s) of microorganism(s) and their location(s) on the product;

c) effect(s) of the extraction technique on the viability of microorganisms;

d) the physical or chemical nature of the product under test.

6.1.3.2 For a product where the extraction of microorganisms is not part of the method (e.g. direct culture of a product), consideration shall, at least, be given to the following:

a) possible type(s) of microorganism(s) and their location(s) on or in the product;

b) the physical or chemical nature of the product under test.

Culture conditions shall be selected based on consideration of the types of microorganisms likely to be present and the physical or chemical nature of the product to be tested.

The technique for enumeration shall be selected after consideration of the types of microorganisms likely to be present.

6.2.1 Appropriate techniques for microbial characterization of bioburden shall be selected.

NOTE Microbial characterization supports detecting a change to the product bioburden that can affect some aspects of the use of bioburden data (e.g. establishing or maintaining a sterilization process). Furthermore, knowledge of the types of microorganisms can be helpful for identifying sources of contamination.

6.2.2 Bioburden shall be characterized using one or more of the following technique(s):

a) colony morphology;

b) cell morphology;

c) differential staining;

d) culture using selective or differential conditions;

e) biochemical properties;

f) genotypic analysis, e.g. pattern or fingerprint-based techniques or sequence-based techniques;

g) proteomic methods, e.g. mass spectrometry;

h) other technologies.

Additional guidance on microbial characterization of bioburden for radiation-sterilized product is provided in ISO 11137-1.

The method(s) for determining bioburden shall be validated and documented.

NOTE See A.7.1 for information on validation and the use of classic microbiological methods.

Validation shall consist of the following:

a) if applicable, assessment of test method suitability to demonstrate lack of inhibition of growth in the test (see 6.1.2);

NOTE 1 Several techniques can support the absence of inhibition of growth such as data from bioburden recovery efficiency testing, if an inoculated product was used, or information from method suitability testing from the test of sterility (bacteriostasis/fungistasis). See B.8 and Annex D for guidance on necessity of inhibition tests for information on test method suitability.

b) assessment of the technique for the extraction of microorganisms from a product or SIP if the extraction is part of the method (i.e. bioburden recovery efficiency), if appropriate for the purpose for which the data are being generated;

NOTE 2 Annex C provides information on the determination of bioburden recovery efficiency.

c) assessment of the technique for the enumeration of microorganisms, including culture conditions and microbiological counting techniques;

d) assessment of the suitability of the technique(s) of microbial characterization to the extent it is used.

Routine determination of bioburden shall be performed by employing documented sampling plan(s) that specify sample size, sampling frequency, sample preparation (e.g. SIP), where applicable, and the test method to be performed. Bioburden data shall be analysed over time to determine and monitor changes in numbers and types (i.e. microbial characterization) of microorganisms.

The determination of bioburden shall be performed using the method(s) specified for a product or a product family (see 5.1.2). The method selected shall consider factors that will affect the results, such as the limits of detection and plate counting.

Microbial characterization shall be performed. The extent of characterization is dependent on the intended use of the data.).

If bioburden data are to be used to establish or maintain the extent of treatment with a sterilization process (i.e. bioburden-based method), any requirements applicable to the use of bioburden data, specified in the appropriate standard for the development, validation and routine control of the sterilization process, shall be met.

If bioburden data demonstrate a test result that is unexpectedly greater than other values within the data set (bioburden spike), these data shall be evaluated to determine possible causes and for impact as appropriate depending on the purpose for the data.

Acceptable levels for bioburden on or in a product or group of products shall be specified (e.g. alert and action levels). Acceptable levels shall be reviewed periodically and revised as necessary per a documented procedure. The results of the review shall be documented.

If alert or action levels are exceeded, a predetermined response shall be carried out in accordance with a documented procedure.

Data derived from the determination of bioburden obtained over a period of time, including microbial characterization, shall be used to identify trends.

If applied, statistical methods can be used to define sample size, sampling frequency or acceptable levels.

Changes to the product or manufacturing process shall be reviewed to determine whether they are likely to alter the ability of the test method to measure bioburden with consideration to the purpose for which bioburden data are to be used. The results of the review shall be documented. If there is potential for alteration of bioburden, specific determinations of bioburden shall be performed to evaluate the extent and nature of any effect of the change.

NOTE The assessment of the change can indicate that the previous method suitability and bioburden recovery efficiency are still applicable.

Any change to a routine method of bioburden determination shall be assessed. This assessment shall include evaluation of the effect of the change on the outcome of determination. The results of the assessment shall be documented.

NOTE The assessment of the change can indicate that the previous method suitability and bioburden recovery efficiency are still applicable.

The original validation data (see 7.2) and any subsequent requalification data shall be reviewed at specified intervals in accordance with a documented procedure. The outcome of the review and any requalification undertaken shall be documented.

When such changes are made that can impact bioburden alert and action levels, the continued appropriateness of bioburden levels shall be assessed and revised if appropriate.

1. 
   (informative)
   
   Guidance on the determination of a population of microorganisms on products

NOTE For ease of reference, the numbering in this annex corresponds to that used in the main body of this document.

• 1. Related to the Scope

This annex gives guidance on the implementation of the requirements specified in this document. The guidance given is not intended to be exhaustive, but to highlight important aspects to which attention should be given.

Methods other than those given in this annex may be used, but these alternative methods should be demonstrated as being effective in achieving conformity with the requirements of this document.

This annex is not intended as a checklist for assessing conformity with the requirements of this document.

Though microbiological characterization is a tool to understand product bioburden, its information is relative to its intended use, and this document does not specify the impact of characterized microorganisms on the effectiveness of the sterilization process.

• 1. Related to the Normative references

No guidance offered.

• 1. Related to the Terms and definitions

No guidance offered.

• 1. General

A.4.1 No guidance offered.

A.4.2 No guidance offered.

• 1. Selection of products
     1. General

A.5.1.1 When selecting and handling product samples the introduction of microbiological contamination and significant alterations to the numbers and types of microorganisms in the sample should be avoided. Sampling techniques should be consistent and should allow for event-based and time-based comparisons of bioburden.

In choosing product samples for the determination of bioburden, there are several possibilities including

a) select samples from routine product (at random or at a specified frequency);

b) manufacture a product specifically for bioburden testing using the routine manufacturing procedures;

c) select samples from product that is not suitable for sale, which can be scrapped or otherwise rejected.

Product selection can depend on a number of factors, but the first prerequisite is that the product selected should possess bioburden representative of routine production. If the decision is made to utilize a rejected product, that product should have undergone all the applicable stages of production that have the potential to affect bioburden, including possible cleaning and placement in the sterile barrier system (SBS).

When sampling for determination of bioburden, the product should be contained, at a minimum, in the SBS representative of that used routinely. Typically, it is sufficient to perform a bioburden determination on a product after its removal from its SBS and to omit the SBS from the determination. Depending upon the sterile label claim, internal packaging components, such as a tray or product insert, might need to be tested based upon factors such as

— if the SBS or internal packaging components are an integral part of the process for using the product (e.g. for mixing a powder and solution); or

— if the components are being specifically evaluated.

In general, the intention is to test the entire contents of the SBS. However, specific contents can be excluded based on a risk assessment.

A.5.1.2 The use of bioburden data (e.g. control of raw materials, acceptance of incoming components, evaluation of process steps, validation of a sterilization process) should be considered when establishing product families for bioburden determination. The following should be considered when establishing product families for bioburden determination:

a) nature and source of raw materials;

b) nature and source of components;

c) complexity of manufacturing process, i.e. degree of handling, number of process steps;

d) types of manufacturing processes used;

e) manufacturing or assembly environment;

f) product design and size;

g) manufacturing equipment;

h) manufacturing location.

In addition, the numbers and types of microorganisms can influence the selection of a bioburden test method for the product family. For each product family, a master product or representative product(s) should be selected for the routine determination of bioburden. The selection of the master product should be based on a documented rationale.

If the products within a family are considered equivalent then a representative product can be selected for the determination of bioburden. The selected product can be monitored routinely or the other members of the group can be chosen on a rotational basis. If a selected product is monitored routinely, the continued equivalence of other products in the family should be periodically monitored or a rationale provided.

It should be understood that there are various types of product families, and the product families established for bioburden testing might not be the same as product families established for other purposes such as bacterial endotoxins testing or sterilant residuals testing, or for applying a sterilization process to product (which is better defined as a processing category).

A.5.1.3 If data from bioburden determinations are to be used to establish or maintain a sterilization process, the period of time that elapses between the completion of manufacturing to the initiation of bioburden testing should generally reflect the time period between the completion of manufacturing and sterilization of the product. The primary concern is the change in bioburden that can occur with the passage of time.

While it is preferred that the time period from manufacturing to the initiation of bioburden testing is representative of routine processing, it might not be practical to do so in many cases. The criticality of controlling this time period should be determined based on product characteristics and the intended use of the data.

For products that do not support microbial growth, bioburden can decrease over time. The types of microorganisms that die from environmental conditions (e.g. desiccation, lack of nutrients) are not likely to challenge a terminal sterilization process. In these cases it might not be critical to control the time period from manufacturing to the determination of bioburden. Conversely, for products that do support microbial growth (e.g. products that contain liquids), bioburden levels might increase over time. In these cases, it might be necessary to monitor and control the storage conditions and duration of time from manufacture to the initiation of bioburden testing to ensure bioburden counts remain in control and are representative of product at the time of sterilization.

• • 1. Sample item portion (SIP)

A.5.2.1 Whenever practicable, the determination of bioburden should utilize the whole product or fluid path (SIP = 1,0), as applicable, although this might not be possible if the product cannot be accommodated in available laboratory testing vessels; in this case an SIP of less than 1,0 is used. Consideration should be given to the distribution of bioburden across the whole of the product. If the distribution is expected to be uneven across the product, one option is to determine the area of the product with the greatest microbiological challenge. This area should be included in the SIP selected and calculations applied to determine the bioburden of the SIP relative to the bioburden of the entire product. The SIP should be representative and not over or underestimate the product bioburden.

A.5.2.2 As large a portion of the product as possible should be used for the SIP. The SIP should be representative so that the bioburden of the whole product or the portion of the product for which sterility is claimed can be determined. Careful selection of the SIP of the product is necessary when large products, such as surgical gowns or external drainage kits, are tested to ensure adequate bioburden representation by the SIP relative to the entire product.

When the bioburden distribution is known, and an SIP < 1,0 is being tested, the following applies:

a) if the bioburden is evenly distributed on or in the item, the SIP may be selected from any portion of the item;

b) if the bioburden is not evenly distributed, the SIP shall include either

1. portions of the product selected that proportionally represent each of the materials (e.g. 30 %) from which the product is made; or

2. the portion of the product that contains the most severe microbiological challenge (numbers and types) to the sterilization process.

When selecting the portion that contains the most severe microbiological challenge, the relationship of the bioburden of the SIP tested to the entire product bioburden should be established.

If the bioburden distribution is not known, and an SIP < 1,0 is being tested, the SIP shall consist of portions of the product selected that proportionally represent each of the materials (e.g. 50 %) from which the product is made.

Consideration should be given to aspects of manufacturing that contribute to the distribution of microorganisms on products.

Examples of an SIP that can be selected from the device with a more severe challenge to the sterilization process are tubing sets with connections or stopcocks.

A.5.2.3 Examples of products for which various bases for SIP calculation are employed are given in Table A.1.

When preparing or assembling an SIP, care should be taken during manipulations of products. If portions are to be separated from products, this should be done under aseptic conditions in a controlled environment (e.g. inside a laminar flow cabinet) in order to avoid adding contamination.

Table A.1 — Examples of an SIP calculation

When testing powder aliquots from a bulk substance, the entire bulk substance might not be considered an SIP of 1,0. In this situation the SIP of 1,0 could be considered the amount of powder that constitutes a single finished product. The same approach can be used for liquid bulk substances.

• 1. Methods of determination and microbial characterization of bioburden
     1. Determination of bioburden
        1. Selection of an appropriate method

Figure A.1 is a decision tree that has general application in the initial stages of the selection of a method of bioburden determination.

For a product with high bioburden and for which a culture method is employed, ensure that a sufficient number of dilutions are carried out to obtain countable results and to prevent issues such as the masking of colonies or too numerous to count (TNTC) plates (see Annex E).

NOTE 1 This decision tree does not preclude the use of alternative, rapid microbiological methods to determine bioburden (e.g. autofluorescence, flow cytometry, direct epifluorescence, filter technique and solid phase cytometry).

NOTE 2 This decision tree does not encompass all types of products that can be tested or all types of testing that can be used.

Figure A.1 — Decision tree for selection of a method of bioburden determination

For a product with very low bioburden, it might not be possible to recover detectable bioburden from individual product units even though a suitable bioburden test method with established bioburden recovery efficiency is used. Caution should be exercised in respect to estimation of average bioburden where zero colonies are detected to avoid overestimation of the true product bioburden. The desired limit of detection (LOD) for the bioburden test method should reflect the intended use of bioburden data, and, if necessary, the bioburden test method should be designed to minimize the limit of detection to increase the test sensitivity as much as reasonably practicable.

To optimize the bioburden determination methods for low bioburden products, it can be necessary to consider use of an alternative approach. Examples are given in the following:

a) Pooled sample approach: where multiple product units are combined into a single test. A bioburden recovery efficiency should be determined for this approach. To estimate the colony forming units (CFU) per product unit, the total recoverable CFU for the pooled sample is divided by the number of pooled units. Pooling of units can permit estimation of a low number of CFUs per unit; however, it provides no information about bioburden distribution or variability on individual units that comprise the pooled sample.

b) Most probable number (MPN) approach: see B.3.3.

c) Combining and eliminating tests for specific groups of microorganisms: for many types of products it is not necessary to split the extract into portions for separate tests, such as aerobes, anaerobes, spores and fungi. If a rationale or evaluation has shown that testing for anaerobes is not indicated, then it is not necessary to include it or this test can be eliminated in the future. Additionally, if aerobic spores are detected in the aerobic bacterial count and fungal counts are not likely to mask bacterial or yeast colonies on the plate, then it is possible to combine the aerobic bacterial, bacterial spores and fungal tests into a single test. For example, filtration of the entire volume of extraction fluid through a single filter that is placed on a suitable general purpose culture medium, which is then incubated at two different temperatures (e.g. 30 °C to 35 °C and 20 °C to 25 °C). Other examples include the use of a single incubation at 30 ± 2 °C, or incubation at other temperature ranges suitable for detecting a specific microbial population. Elimination of dilution factors in this way (provided the elimination is justified) can increase the sensitivity (reduce the LOD) of the bioburden test.

d) One half limit of detection approach: this assists in calculating a bioburden average when “less than” values are present in the bioburden results. This approach provides a lower bioburden result when a lesser number of plates are at 0 CFU per plate (for more information refer to Reference [31]).

e) Poisson-based substitution for “less than” values approach: this assists in calculating a bioburden average when “less than” values are present in the bioburden results. This approach provides a lower bioburden result with a greater number of plates are at 0 CFU (for more information refer to Reference [31]).

Bioburden is typically not distributed on products throughout the manufacturing process in a Poisson distribution fashion. Applications of the use of Poisson distribution to bioburden should be carefully considered in relationship to the intended use of the information. (For more information refer to References [27], [31] and [47].)

The selection of a method for determining bioburden should consider the possible occurrence of biofilm on or in the product. Biofilm can form on or in the product when in prolonged contact with aqueous liquids unless appropriate microbiological control measures are taken. Health care products incorporating tissue have a potential for biofilm occurrence.

• • • 1. Neutralization of inhibitory substances

See Annexes B and D.

• • • 1. Recovery of microorganisms

See Annex B.

• • • 1. Detecting viable microorganisms

The nature of raw materials, the method of manufacture and the conditions under which the product manufacture occurs, are factors that influence product bioburden and should be considered when choosing the culture media and incubation conditions. Unless fastidious microorganisms are likely to be present, general purpose, non-selective culture media and incubation conditions are appropriate. The recommendations of the laboratory, with the input of the manufacturer, for the use of standard bioburden culture conditions can suffice as the consideration and rationale.

Culture media used should be tested for growth promotion, whether by the manufacturer or the laboratory.

When selecting culture media and incubation conditions, the following, at least, should be considered:

a) no single combination of medium and incubation conditions can support the growth of all microorganisms; the choice of conditions should minimize the potential to overestimate the average bioburden due to double counting (e.g. spores and facultative anaerobes that also form CFUs on the aerobic bacterial plates);

b) the assessment of the technique for the enumeration of microorganisms can require the use of a wider range of culture media and conditions of incubation than those used routinely;

c) likely microbial contamination sources and the types of the microorganisms that can be encountered, bearing in mind that some contamination sources can vary seasonally.

Health care products manufactured from synthetic material are unlikely to be contaminated with obligate anaerobes. Health care products manufactured from tissue-based products can be at risk of contamination with obligate anaerobes.

Examples of culture media and incubation conditions are given in Table A.2. Rapid microbiological methods will have media types and incubation conditions that are specific to that method and will be determined during the method validation.

Table A.2 — Examples of culture media and incubation conditions^a

• • • 1. Enumeration of microorganisms

The laboratory can specify the technique for enumeration, which will suffice as the consideration and rationale. See also B.6.

• • 1. Microbial characterization of bioburden

A.6.2.1 The degree of characterization necessary is dependent on the nature of the product, diversity of the detected population, and the use of the data (e.g. validated sterilization process ). As it is understood that no single combination of media and incubation conditions can support the growth of all microorganisms, it should not be expected that all microorganisms undergo microbial characterization. Generally, it is sufficient to perform characterization on a subset of isolates (e.g. top three or five isolates), as these data would be sufficient for identifying trends in the predominant types of microorganisms that are present. A greater degree of characterization can be necessary for specific purposes, such as when there is a need to screen for specific objectional organisms, or to aid in an investigation.

A.6.2.2 A wide range of methods can be used to characterize microorganisms comprising the bioburden on or in a health care product. Typical microbial characterization methods for bioburden include colony morphology, cellular morphology, staining properties, selective culturing and microbial identification. Details regarding these methods are as follows.

a) Colony morphology is simple to record when the colony count is obtained. Describing the colony morphology is somewhat subjective and includes colour, shape, size, texture, margin, elevation and other physically observable characteristics of the colony. This information alone is not conducive to trending (see A.8). It can usually be used to distinguish between bacterial and mould isolates and to initially determine if the colonies on a plate are likely to be the same microorganism. Further characterization in order to identify sources of contamination requires more specific methods.

b) Cellular morphology and staining techniques, such as a wet mount and Gram stain, are often used to characterize microorganisms. The benefits of these methods are that they require minimal equipment and time, and can provide valuable information regarding the general characteristics of the microorganisms. Characterization of fungi (i.e. mould and yeast) via a physical description and a wet mount can be sufficient for the majority of isolates.

c) Selective culturing and differential media can be used to inhibit the growth of particular microorganisms, select for certain microorganisms, or assist in differentiating some microorganisms from others (e.g. colour of the colony on specific media) which can be useful in characterizing the microorganism.

d) Microbial identification can be performed using phenotypic or genotypic methods, or a combination of both. Classical phenotypic tests, such as colony and cell morphology, Gram and spore stain reactions, ability to grow aerobically or anaerobically, and simple biochemical reactions (e.g. catalase, oxidase, indole), usually provide some indication of the group or genus to which a bacterium belongs. More complex biochemical and serological tests, or genotypic or proteomic methods can identify a bacterium to genus, species or strain level. A similar approach can be taken with yeast and mould. A combination of morphological and physiological properties can be used to establish genera, with biochemical assimilations used to differentiate species.

Table A.3 provides information on common bioburden characterization methods.

Table A.3 — Attributes of common bioburden characterization methods

• 1. Validation of the method for determining bioburden
     1. General

Bioburden determination is a collection of methods, guidance, and recommendations, which in sum lead to an individual testing of a product, including individual conditions. These classical microbiological methods or methods described in national and international standards and pharmacopoeias usually do not require validation (see References [34] and [35]). These methods should only need to be verified for their accuracy and reliability under their unique conditions of use. Such actions are usually sufficient to confirm the validity of the determination of bioburden. Additionally, this document (see introduction) mentions explicitly that other methods than those described in the standards can be applicable (see References [41] and [44]).

In the validation of bioburden test methods on a specific product there are two aspects to consider. The first is the ability to neutralize inhibitory factors in the test system to allow microorganisms to replicate (bioburden method suitability) and the second is the ability to remove and culture microorganisms from a product (bioburden recovery efficiency).

When methods for determining bioburden include extraction of the microorganisms from a product, it is the efficiency of the extraction process that is of greatest concern (see Annex C for details).

• • 1. Validation

Bioburden method suitability

Bioburden method suitability testing is used to demonstrate that the product does not prevent the growth or detection of microorganisms. The product can contain substances that are inhibitory to microorganisms in bioburden test conditions.

Dilution or suitable inactivation/neutralization methods should be used in testing products that contain antimicrobial substances.

The composition of the device should be investigated for ingredients with known risks for inhibitory effects (e.g. antibiotics, intentional antimicrobial properties). Further risks can result from the use of materials, which are not common for medical devices or with potential impact to biocompatibility (e.g. copper, silver).

Inhibitory effects of substances eluted from the product, when these substances are part of the product (e.g. antimicrobial coating) or if a specific risk is identified, should be investigated in product-specific preliminary experiments, to evaluate whether the substances can cause inhibition to microorganism growth in bioburden test conditions. A documented rationale can be used if the device does not contain known risks for inhibitory effects and comprises only materials that are known or have been demonstrated to be inert.

Bioburden method suitability should be considered

a) when there are new or modified products;

b) whenever there is a change in the conditions of the test (e.g. incubation conditions, extraction media).

The application of methods with given suitability for microbicidal or microbiostatic substances (e.g. membrane filtration with an established membrane rinsing procedure) might not require a product specific bioburden method suitability test except when known risks of inhibitory effects have been identified.

The direct application of a method suitability test from the test of sterility (bacteriostasis/fungistasis) might be applied to a bioburden test. Information such as incubation temperature(s), times and culture media can provide direction for neutralization in the bioburden test, if they are needed. However, any differences in the culture media, or volumes of the extract versus the growth media in the test of sterility should be considered.

Additional guidance for method suitability testing for pharmaceuticals can be found in current Pharmacopeias (see References [34], [35], [43], [44] and [45]). See Annex D for guidance on method suitability testing for health care products.

Assessment of technique for extraction of microorganisms

There are essentially two traditional approaches available for establishment of the efficiency of the extraction of a microorganism from health care products (see C.1.4). These approaches are

— repetitive recovery: the repetitive extraction of a product sample followed by quantitative assessment of the extent of recovery; or

— inoculated product: a product inoculated with known levels of microorganism(s), followed by quantitative assessment of the extent of recovery.

The first of these approaches has the advantage of utilizing the naturally occurring microorganisms but usually needs a moderate to high initial bioburden. If this is the case, then the first approach can be preferred based on the product and configuration. The second approach creates a model system for testing purposes but raises questions as to how it compares to recovering natural microorganisms. For additional information see Table C.1.

Some products, including certain non-traditional products (e.g. complex or complicated products containing powders, liquids, antimicrobial agents, multiple components), or low bioburden products where it is difficult to obtain an adequate recovery efficiency from a single extraction, can use a combination of the above methods to assess bioburden recovery efficiency. In this case a combination of the above approaches known as inoculated repetitive recovery can be used. For this approach, a product is inoculated with microorganism(s), followed by repetitive extraction of a product sample, and quantitative assessment of the extent of recovery. The correction factor is calculated as detailed for repetitive recovery. This method can be applied to product that has been exposed to a sterilization process or to product prior to a sterilization process. When product is used prior to a sterilization process, during the enumeration step both the natural bioburden count and the inoculated bioburden count can be used to calculate the recovery efficiency. When using this method for recovery efficiency it should be considered whether multiple extractions are also needed during routine testing (e.g. when the first extraction results in a very low recovery efficiency, and it is determined that a higher recovery efficiency is needed for the intended use of the data). Refer to Annex C for additional information.

For a liquid product that is filtered, or when the MPN method is used, determination of a bioburden recovery efficiency and calculation of a bioburden correction factor are not necessary. However, test method suitability for enumeration should still be assessed.

Assessment of the technique for enumeration and culture conditions

For further guidance on enumeration, see B.6.

The culture conditions (i.e. media and incubation conditions), selected for use in determination of bioburden cannot be expected to detect all potential microorganisms. In practice, therefore, it is likely that bioburden will be underestimated. Nevertheless, a decision on appropriate culture conditions will have to be made.

One approach to the assessment of culture conditions consists of selecting the culture conditions based on a knowledge of the manufacturing process, environment, materials and the microorganisms expected to be present. If specific product characteristics indicate that additional assessment is needed, the microorganisms enumerated under typical culture conditions are compared to those detected by alternative culture conditions. If this approach indicates that a low proportion of the bioburden is being detected in the typical culture conditions, the alternative culture conditions should be reconsidered to optimize the determination. This is of particular concern for health care products where antimicrobials can affect microbial growth and neutralizers can be added to the growth medium.

Assessment of the technique for microbial characterization

When selecting techniques for use in the microbial characterization of microorganisms, consider the following:

— risk to the manufactured product considering the mode of sterilization validation;

— previously available data;

— the purpose for generating the data;

— the nature of the manufacturing process (e.g. water involved, manual, automated) and the product.

• 1. Routine determination of bioburden and interpretation of data
     1. General

In order to demonstrate that effective control of microbiological quality has been implemented and maintained, a programme of monitoring the product or components should be developed.

It is common practice to use a sample size of between three to ten items for routine monitoring of bioburden levels.

Where bioburden data are used to satisfy the requirements of another standard (e.g. the ISO 11137 series), sample size and test frequency can already be predefined by that standard, which would supersede the sample size recommended here.

A rational choice of sample size primarily depends upon two factors.

a) The potential impact of a change in bioburden on the product.

This will depend upon the consequences associated with a change (either increase or decrease) in bioburden level and how the bioburden information is being applied. For early detection of a small change in the mean bioburden level, a large sample size can be needed.

b) The item-to-item variability in bioburden.

The degree of this variability will determine the sample size necessary to detect a given change. If the product bioburden has been demonstrated to have little variability, a smaller sample size can be appropriate. If the product bioburden has been demonstrated to have greater variability, a larger sample size can be appropriate in order to better understand and trend the variability.

It should be recognized that the manner in which bioburden data are used can influence the desired level of confidence in detecting a change of a given magnitude. A rational choice of the magnitude of change to be detected and the probability of achieving that detection should be made.

A rational choice for the frequency of monitoring should be made, considering a variety of factors including the following:

— the availability of historical data;

— the purpose for generating the data;

— the nature of the manufacturing process;

— the production frequency for the product;

— the criticality of detecting bioburden changes in a timely fashion;

— seasonal and environmental variations.

Sampling can be performed at a frequency based on time (e.g. monthly, quarterly), or on production volume (e.g. alternate batches). However, in order to establish baseline levels, it is common practice to determine bioburden at a higher frequency during the initial production of a new product and for this frequency to be reduced as a knowledge of bioburden develops.

The frequency of determinations of bioburden should be influenced by the intended use of the data, and should allow detection of changes in bioburden, for example, due to seasonal variations, manufacturing changes or changes in materials.

• • 1. Limits of detection and plate counting

Limits of detection (LOD) for bioburden test methods should be taken into account in determining the bioburden value. For microbiological reporting, when a portion of the extract is tested for bioburden, and zero colonies are recovered, the results are typically reported as less than “X” where “1/X” is representative of the fraction of the portion tested. For example, if a product is extracted in 400 ml and one-fourth of the extract is filtered, results of zero colonies will be reported as less than 4 CFU (i.e. < 4 CFU). Therefore, the LOD for this example is 4. In microbiological reporting, a result of < 4 CFU means that it is possible that the entire extract contains either 0, 1, 2 or 3 CFU, but microbiological reporting rules require that it be reported as < 4 CFU.

Individual bioburden results are reported in whole numbers because the number is representative of a CFU. Averages or other mathematical calculations using bioburden data are typically reported to one decimal place.

LOD can be improved by the following:

a) modification to the test method (e.g. filtering a larger portion of the extract);

b) pooling multiple samples;

c) utilizing another test method, such as MPN.

For plate counting and reporting guidance, see Annex E. For excursions, see Annex F.

Counts that are above the countable or estimated range can be semi-quantitated if that value can be approximated based on the presence of discernible colonies. However, if this cannot be done, a result of TNTC should be assigned. In some circumstances it is an acceptable practice to omit TNTC results from the average for a group of samples. If repeated TNTC results are observed they should be addressed. The minimum number of samples, excluding TNTC omissions, should meet the requirements the applicable sterilization standards (e.g. ISO 11137-2).

When duplicate plate counts, dilution factors or aliquots are used, the plate counts should be adjusted accordingly to obtain the count for a single product.

• • 1. Microbial characterization

If, on microbial characterization, types of microorganisms are recovered that are not part of the normal bioburden, consideration should be given to assessing the relevance of the presence of these isolates. Examples of what can be assessed when this occurs can include determining if the new types of microorganisms are potentially concerning from a microbiological quality standpoint, determining if the new types indicate a problem with the manufacturing process, or determining if the new types are a result of a change in the process or in supplied materials.

• • 1. Use of bioburden data for determination of the extent of treatment

No additional guidance.

• • 1. Bioburden spikes

Bioburden data can demonstrate a value that is significantly greater (commonly called a bioburden spike) than other values within a set of values. This bioburden spike can occur in one of two situations:

a) the value is not a normal and consistent part of the bioburden distribution;

b) the value can be a normal and consistent part of the bioburden distribution.

It can be determined that the bioburden spike is not a normal and consistent part of the bioburden distribution by reviewing historical data and through an investigation into the manufacturing practices, microbiological testing, and handling of samples. Bioburden results that are less than a one-log difference (i.e. a factor of 10) are usually not considered to be microbiologically significant.

It can be determined that the bioburden spike is a normal and consistent part of the bioburden distribution by reviewing historical data. Historical data can demonstrate a periodic occurrence of a greater value that is within expectations making it a consistent part of the bioburden. If these data are comprised of microorganisms typically found on the product but present in greater numbers, this is likely a normal part of the bioburden. These spikes should be included when determining the extent of treatment of a sterilization process. For example, a bioburden spike can occur due to raw materials that are not consistent, or manufacturing processes that involve significant handling.

In the example given in Table A.4, there are 3 batches (batches 2, 5, 6) in 10 batches that contain individual values that are significantly greater than the batch average and comprise of microorganisms typically found on the product (in this example, five or more times the batch average). It was determined that these high values are a normal and consistent part of the bioburden. Consequently, the high values or the batch averages for those batches that contain high values might need to be taken into consideration when establishing the overall average bioburden for determining the extent of treatment for a sterilization process.

Table A.4 — Example of bioburden data containing bioburden spikes

• • 1. Bioburden levels

Historic bioburden data are used to establish bioburden levels that are commonly defined as alert levels and action levels. Establishment of these levels should take into consideration the approach used based on the intent of the use of the information. For example, levels can be used to evaluate raw material suppliers, qualify or demonstrate the continued effectiveness of the sterilization process or assess the efficacy of environmental control in a manufacturing process.

The specified levels used for bioburden are commonly based upon historical data for a product and the purpose for which the data are to be used. Prior to the collection of historical data, if it is desired to establish temporary levels, then these can be set after evaluating the first three or more batches of a given product. Temporary bioburden levels can also be established based on how the data are used in sterilization validation activities. Historical data from similar products, manufacturing processes or manufacturing environments can also be used when setting temporary levels for new product lines. Based upon successive test results, bioburden data should be re-evaluated after a period of time to verify whether the original levels are appropriate. It is not expected that these microbiological data are precise. Rather it is common that a substantial range is present in microbiological data for bioburden. Seasonal humidity or temperature levels/changes can also alter the types and numbers of microorganisms in the bioburden. It is also not expected or necessary that microbiological data for bioburden fit a statistical distribution.

One common method for determining bioburden alert and action levels is by using standard deviations. In this instance, the standard deviation calculation is used to understand the dispersion of the data and it is less critical whether the bioburden data fits a particular statistical distribution.^[47] Another example of a way is to use log variations from the average such as a 0,5 log variation for the alert level and a one-log variation for the action level.

A predetermined course of action should be taken when specified levels are exceeded, e.g. alert and action (see AAMI TIR106). If corrective actions lead to changes to the process that affect the bioburden, new data should be obtained and new levels established for the product. Data identified as unusually high or low, or as atypical of the trend, should be investigated. Atypical data, with identified cause (e.g. laboratory error, occasional high values found in the manufacturing process), can be omitted from calculations in setting the levels for bioburden monitoring. When bioburden data are analysed for use in a quality-related decision, individual test outcomes, such as “no growth” or “TNTC”, are included in the analysis.

• • 1. Data analysis

Graphical representation of data collected over time can be useful in distinguishing actual trends from sampling variability. Graphical representation can also indicate that a significant change in the microbiological population has occurred even though the bioburden values reside within the pre-set levels.

Before statistical calculations can be performed on data derived from bioburden determinations, especially where many observations are documented it can be necessary to manipulate the data in such a way that the significant features are revealed. This can be done in a qualitative manner by grouping the measurements to form frequency tables and charts. Upon completion, the data can be examined for trends.

There are a number of techniques for trending that can be applied to bioburden. These trending techniques can be, but are not limited to, trending of bioburden averages or bioburden estimates, Shewhart control charts (ISO 7870‑2), control based on range, or cumulative sum charts (see ISO 7870-4). Each of these different techniques can be used to establish a possible shift from the usual random spread of results and to highlight deviations.

In some instances, it can be appropriate to utilize more than one of these techniques to determine whether or not action is to be taken based upon the available data set or whether additional data are required.

• • 1. Statistical methods

ISO 13485 requires the planning and implementation of appropriate methods of measurement and analysis, including selecting suitable statistical methods. The examination of data derived from determinations of bioburden for a wide range of products illustrates the variability of such data. Determinations from a group will vary within the group of items, and, therefore, analyses of data generally use means. Clearly, these means can take high, intermediate or low values, and mean values will vary over time. Furthermore, the types of microorganism that comprise the bioburden can also vary.

• 1. Assessment of change for bioburden determination
     1. Changes to the product or manufacturing process

Changes to the product or manufacturing process should also be reviewed for any potential effects to the efficacy of the method for determining bioburden. The results of the review should be documented. In some cases, it can be necessary to change or requalify the method of bioburden determination.

• • 1. Changes to the method for determining bioburden

No guidance offered.

• • 1. Requalification of the method for determining bioburden

Several small changes can accumulate and might necessitate a revalidation. The review should consider at least the following points:

— Determination of the recovery efficiency, as applicable;

— Method suitability test;

— Appropriateness of SIP.

• • 1. Review of appropriateness of bioburden levels

No guidance offered.

1. 
   (informative)
   
   Guidance on methods to determine bioburden
   1. General

B.1.1 Bioburden determinations can be employed in a variety of situations. The individual responsible for the conduct of such determinations should take into account to what extent method development and validation needs to be performed. In addition, the particular circumstances under which the determinations are made should be considered, e.g. deciding sampling rates, the method to be used, nature of the culture media and relevant incubation conditions.

B.1.2 The sequence of key steps of the process for determining bioburden is illustrated in Figure B.1. The individual responsible for the conduct of such determinations should use knowledge of the raw materials, components, manufacturing environment, production processes and the nature of the product to select appropriate techniques for the various steps. For proper method development and validation, it is possible that a combination of different methods might need to be employed initially in order to establish the method(s) most suitable for routine use.

Figure B.1 — Sequence of key steps of the process for determining bioburden

• 1. Methods where removal of microorganisms by extraction is used
     1. General

B.2.1.1 Several methods described in this annex can be combined to increase the number of microorganisms recovered and reduce variability.

B.2.1.2 The degree of adhesion of microorganisms to surfaces varies with the nature of the surface, the microorganisms involved and other materials present (e.g. lubricants). The origin of the contamination will also influence the degree of adhesion. To remove microorganisms, extraction consist of rinsing together with some form of physical force or direct surface sampling. A surfactant can be used to enhance recovery but it should be recognized that surfactants at high concentrations can be inhibitory to the growth of microorganisms.

B.2.1.3 For materials in prolonged contact with non-sterile fluids, microorganisms can occur as a biofilm unless appropriate microbiological control measures are taken. A biofilm is a structure in which microorganisms are encapsulated in an extracellular matrix that adheres strongly to surfaces. Microorganisms in biofilms can exhibit increased resistance to sterilization processes. Biofilms can initiate quickly and can develop to a much greater extent on health care products incorporating tissue or on reusable devices. In such instances, consideration should be given to the potential for biofilm formation and it should not be assumed that the extractions outlined in B.2.2 would be appropriate for liberating microorganisms completely from a biofilm. An indication that a biofilm is present can be obtained during establishment of the extraction technique if repeated high microbial counts are recorded during repetitive recovery. A high level of endotoxin can also be an indication of biofilm.

B.2.1.4 Any extraction technique used during bioburden determination should be reproducible and should avoid conditions that are likely to affect the viability of microorganisms, such as excessive cavitation, shear forces, temperature rise or osmotic shock.

B.2.1.5 Some extraction techniques are easier to control than others. The variables and ways of controlling them should be considered when selecting a technique and devising suitable conditions of extraction. For example, for a given extraction, the time can be extended or the nature of mechanical agitation modified to increase the removal of microorganisms.

B.2.1.6 Certain methods of extraction can disaggregate the product under test (e.g. disintegration, processing in a stomacher and vortexing). The presence of disaggregated material can render enumeration of microorganisms difficult. Additional extraction, for example to separate the disaggregated material from the eluent, can be necessary. Care should be taken to ensure that the counts obtained are representative. Certain types of microorganisms are more prone to aggregation/reaggregation than others based primarily on their relative hydrophobicity.

B.2.1.7 Every effort should be made to transfer items for testing to the laboratory as quickly as possible. If delay in transfer is unavoidable, the conditions under which the items are stored should be selected to minimize changes in the microbial population.

• • 1. Extraction techniques
       1. Shaking (mechanical or manual)

B.2.2.1.1 The test item is immersed in a known volume of eluent within a suitable vessel and shaken on a mechanical shaker (e.g. reciprocating, orbital or wrist action) using setpoints such as time, number of cycles, RPM, to assist the extraction of microorganisms. Manual shaking can be used but its effectiveness can vary depending on the operator.

B.2.2.1.2 The time and frequency of shaking should be defined.

B.2.2.1.3 Glass beads of a defined size can be added to increase surface abrasion and thereby bioburden recovery efficiency. The size of added glass beads, together with the time and frequency of shaking, should not be such as to cause possible damage to the microorganisms. Use of a bead beater or tissue homogenizer is not recommended.

NOTE The addition of glass beads will increase the surface area to which microorganisms can adhere.

• • • 1. Ultrasonication

B.2.2.2.1 The test item is immersed in a known volume of eluent within a suitable vessel. Either the vessel and contents are treated in an ultrasonic bath or an ultrasonic probe is immersed in the contained eluent. The sonication method should be assessed in accordance with B.9.

B.2.2.2.2 The nominal frequency of sonication and duration of extraction should be defined. Furthermore, the position(s) in which items are placed in an ultrasonic bath should be defined. Consideration should be given to limiting the number of items to be processed concurrently as some of the sonication power can be reduced through shielding.

B.2.2.2.3 The technique is particularly suitable for products with complex shapes and when breaking up clumps of microorganisms is needed.

• • • 1. Vortex mixing

B.2.2.3.1 The test item is immersed in a known volume of eluent in a closed container that is placed on the rotating pad of the vortex mixer so that a vortex is created. The vortex produced will depend upon the pressure applied manually. Variations in the vortex can cause variable extraction.

B.2.2.3.2 The container to be used, the time of mixing and the speed at which the mixer is set should be defined.

B.2.2.3.3 The method is quick and simple to perform but is mainly suitable for small items in small containers. Variations in extraction should be assessed among different individuals operating the vortex mixer.

• • • 1. Processing in a stomacher

B.2.2.4.1 The test item and a known volume of eluent are enclosed in a sterile stomacher bag. Reciprocating paddles operate on the bag, forcing the eluent through and around the item.

B.2.2.4.2 The time of extraction should be defined.

B.2.2.4.3 This method is particularly suitable for soft, fibrous or absorbent materials but is also unsuitable for any materials that would puncture the bag (e.g. products containing needles or rigid items).

• • • 1. Flushing (fluid pathway)

B.2.2.5.1 The eluent is passed through the item (e.g. internal lumen). Liquid flow can be induced by gravity or pumping. Alternatively, the product can be filled with the eluent, sealed and agitated.

B.2.2.5.2 The time of contact between the device and eluent, the rate of flushing and the volume of fluid should be defined.

B.2.2.5.3 Device configurations and lumen sizes can limit the physical forces necessary to remove microorganisms completely from internal surfaces.

• • • 1. Blending (disintegration)

B.2.2.6.1 The test item is immersed in a known volume of eluent within a suitable vessel. The item is blended or chopped for a specified time.

B.2.2.6.2 The specified time depends on the item and the blender but should not be extended such as to cause overheating of the eluent and possible damage to the microorganisms.

B.2.2.6.3 This technique provides a way of dividing an item into small enough parts so that the microorganisms can be enumerated by a direct culture technique.

• • • 1. Swabbing

B.2.2.7.1 Swabs consist of absorbent material which is usually mounted on some form of stick or handle. The sampling material can be soluble or insoluble.

B.2.2.7.2 The normal method of use is to moisten the swab with eluent and wipe a pre-determined surface area of the item. The bioburden recovery efficiency can be improved in some circumstances by first moistening the surface and then swabbing with a dry swab. The swab is transferred to diluent and agitated to remove microorganisms from the swab. Alternatively, in the case of soluble swabs, the swab is dissolved in diluent.

B.2.2.7.3 Swabs are a useful method of sampling irregularly shaped or relatively inaccessible areas. They are also useful when a large area is to be sampled.

B.2.2.7.4 This technique is particularly prone to errors due to variation in the way the swab is manipulated. Furthermore, it is unlikely that all microorganisms on the surface will be collected by the swab. Some of the microorganisms that are collected can become trapped in the matrix of the swab itself. Because of these issues, recovery using this method is generally low.

• • 1. Eluents and diluents

B.2.3.1 During bioburden determination, eluents can be used to remove microorganisms from the product. Diluents are used to obtain suspensions containing microorganisms in countable numbers.

B.2.3.2 The nature of the eluents and diluents can have a marked influence on the overall efficiency of the method used. In selecting a diluent or eluent, consideration should be given to its composition (e.g. constituents and their concentrations, osmolarity and pH). Ideally, the composition should be such that proliferation or inactivation of microorganisms does not occur; however, it might not be possible to ascertain this for all potential contaminants.

B.2.3.3 When a liquid is used for extraction of microorganisms from solid surfaces, the incorporation of a mild surfactant can be considered, see Table B.1.

B.2.3.4 Eluents and diluents commonly used include those given in Table B.1.

Table B.1 — Examples of eluents and diluents

• 1. Methods where removal of microorganisms by extraction is not used
     1. Contact plating

B.3.1.1 Contact plates or slides are means by which solidified culture medium can be applied to a surface with the intention that microorganisms will adhere to the surface of the medium. The plate or slide can then be incubated to produce colonies that are enumerated.

B.3.1.2 Such systems have the advantage of being easy to use. Results are directly related to the area in contact with the solidified culture medium.

B.3.1.3 The natural clumping of cells on surfaces, spreading of colonies at the agar interfaces, and drying out of the agar are potential disadvantages.

B.3.1.4 This method should be used only when other methods are not applicable because the efficiency is generally low. Contact plates and slides are generally only useful on flat or at least regular surfaces.

• • 1. Agar overlaying

B.3.2.1 Agar overlaying involves coating the surfaces of a product with a molten agar medium and allowing it to solidify, followed by incubation to produce visible colonies. This method is not commonly used but can be applicable when the bioburden is low and the product configuration suitable.

B.3.2.2 The natural clumping of cells on surfaces, spreading of colonies at the agar interfaces, and drying out of the agar are potential disadvantages.

• • 1. Most probable number (MPN) approach

B.3.3.1 The MPN approach can be used for estimating the number of microorganisms in a product, or SIP in which the microorganisms are randomly distributed.^[46] Where bioburden is not homogeneously distributed, an SIP, where used, should be representative of the entire product. This approach enables the culturing and estimation of the MPN bioburden as it resides in or on product (i.e. without the need for bioburden extraction necessary in other bioburden determination methods); therefore, it can be useful in hard-to-extract bioburden situations and is particularly appropriate for a product having a low average bioburden.

B.3.3.2 The approach consists of taking replicate samples of a product or SIP (by volume or weight) which contain, on average, the same number of microorganisms in each sample/subsample (hence the requirement for randomness of distribution) and assessing each sample for growth (e.g. turbidity, sediment) that indicates the presence of microorganisms by means of transferring to liquid culture media, incubating, and interpreting the results. Generally, the same culture media and conditions referred to in tests of sterility per ISO 11737‑2, for seven days, is appropriate. A range of dilutions (could be aseptic product fractionation into an SIP) can be inoculated into nutrient medium such that a fraction of the inoculated media does not produce visible growth on subsequent incubation. At least one test sample of the number tested needs to be negative for growth (fractional growth) to perform the necessary MPN calculations; to facilitate this, it can be practical to test multiple dilutions or SIP sets simultaneously. From the frequency of the occurrence of positive tests within a set of replicates, an estimate is made of the number of microorganisms present in the sample or the bulk product from which the sample has been taken; the 95 % confidence limits about the estimate are relatively wide. The estimate and its confidence limits are derived from published MPN Tables^[29] which have been developed on the assumption that numbers of microorganisms present in replicate samples are distributed around a mean number in accordance with Poisson distribution. The FDA BAM Annex 2^[36] includes a spreadsheet for calculation of MPN 95 % confidence levels.

B.3.3.3 The key requirement for the application of the MPN approach is the random distribution of microbial population throughout the product or SIP under investigation. Accordingly, the MPN method can have value for the determination of bioburden for liquid health care products, viscous fluids, powders, low bioburden products (see B.3.3.4) or in situations where the bioburden is being estimated in a liquid used as an eluent for a single product.

B.3.3.4 The MPN approach for 10 samples of a single dilution is shown in Table 5 of the FDA BAM.^[36] This single dilution method does not incorporate additional dilutions that could provide further information about the number of microorganisms producing a positive sample. In the case of testing a non-liquid product (e.g. solid medical devices, semisolids or powders), dilutions can be performed by testing different SIPs that have been properly qualified. However, when testing an SIP of 1,0 and when most or all test samples are negative for growth, performing dilutions will not provide further information as they will likely all be negative for growth. In this situation, if further information is desired, a larger sample size or multiple batches can be tested.

Formula (B.1) can be used for individual samples or SIPs to determine a most probable number. Formula (B.1) is a simplified version of the original formula from Cochran^[32].

(B.1)

where

sd is for single dilution;

ln represents natural log;

n is the total number of samples tested;

s is the number of samples negative for growth.

B.3.3.5 Results of MPN per product can be equated to results of CFU per product for bioburden enumeration and calculations.

B.3.3.6 If microbicidal or microbiostatic substances are present, the considerations outlined in B.8 will apply.

• 1. Transfer to culture medium
     1. General

B.4.1.1 Extraction will usually produce a suspension of microorganisms. Enumeration of the microorganisms in the suspension can be undertaken using one of the techniques described below.

B.4.1.2 Prior to transfer to culture medium, additional extraction can be necessary in order to disrupt aggregates of microorganisms and thereby reduce underestimation. In some cases, the technique used to remove the microorganisms from the item under test can also disrupt such aggregates.

B.4.1.3 The presence of microbicidal or microbiostatic substances will influence the choice of culture method. If microbicidal or microbiostatic substances are present in the eluent, these can be reduced to an ineffective concentration by dilution, removed by filtration or chemically inactivated.

• • 1. Membrane filtration

B.4.2.1 Filtration of an eluent, followed by incubation of the filter on an appropriate growth medium to give visible colonies, is an effective means of enumerating microorganisms. A filter of appropriate nominal pore size not greater than 0,45 μm is generally adequate to capture microorganisms [58].

B.4.2.2 A vacuum, or in some instances pressure, source is usually required. Care should be exercised in order to avoid excessive backpressures, which can cause distortion of or damage to the membrane filter.

B.4.2.3 Membrane filtration of eluents containing particulates, such as remnants of fibrous products, can be difficult, as the particulates can block the filter.

B.4.2.4 For incubation, the membrane filter can be placed either on an agar surface or on an absorbent pad saturated with liquid nutrient medium. Colonies produced on the surface of the membrane filter are counted and isolated for microbial characterization.

B.4.2.5 Membrane filtration is particularly useful for suspensions of low concentrations of microorganisms because a large volume of fluid can usually be filtered.

B.4.2.6 Filtration is also useful when the liquid substrate is suspected of containing microbicidal or microbiostatic substances, as the microorganisms are removed from the eluent and can be washed on the membrane filter prior to incubation. Some types of membrane can absorb or release substances that can inhibit the growth of microorganisms, so it is important that only membrane filters suitable for enumeration of microorganisms be used. The membrane filter and the eluent should be compatible.

• • 1. Pour plating

B.4.3.1 With a pour plate technique, separate aliquots of a suspension are mixed with molten agar medium at a temperature of approximately 45 °C; the mixture is then allowed to solidify in the petri dish. Plates are incubated and the colonies are counted.

B.4.3.2 Pour plating does not separate microorganisms from the eluent. If microbicidal or microbiostatic substances are present, the considerations outlined in B.8 will apply.

B.4.3.3 The amount of eluent that can be pour-plated is limited. Therefore, this method might not have the desired sensitivity for suspensions with low concentrations of microorganisms.

B.4.3.4 It is desired to keep the agar temperature as low as possible to avoid damage to microorganisms.

• • 1. Spread plating

B.4.4.1 With a spread plate technique, an aliquot of suspension is spread on the surface of a culture medium using a spreading device.

B.4.4.2 The aliquot of suspension that has been spread on the surface of the medium should be absorbed so that discrete colonies can develop; the need for absorption governs the volume of the aliquot that can be processed using one plate.

B.4.4.3 If microbicidal or microbiostatic substances are present, the considerations outlined in B.8 will apply.

B.4.4.4 The amount of eluent that can be spread-plated is limited. Therefore, this method might not have the desired sensitivity for suspensions with low concentrations of microorganisms.

• • 1. Spiral plating

B.4.5.1 The spiral plating technique uses automated equipment, which deposits an aliquot of a suspension on the surface of solid medium. The suspension is spread at a decreasing rate in a spiral track from the centre of the culture plate to the periphery. After suitable incubation, the count of microorganisms in the original suspension is established using a particular counting grid and counting technique when either total plate or sector counts are the basis for calculations.

B.4.5.2 The spiral plating technique has been shown to give reproducible results that correlate very well with those using conventional serial dilution and surface spreading techniques. Due to the design of the apparatus and the use of capillary tubing and small volumes, spiral plating primarily lends itself to inoculating suspensions that are well dispersed, free from aggregates of material and contain a high concentration of microorganisms

B.4.5.3 If microbicidal or microbiostatic substances are present, the considerations outlined in B.8 will apply.

• 1. Incubation (culture media and incubation conditions)

B.5.1 Examples of some culture media and incubation conditions are given in Table A.2. This list is not all inclusive, and determination of the type(s) of bioburden microorganisms present on products, including by molecular means, can trigger the inclusion or exclusion of these or many other media for microbial culture.

B.5.2 The range of microorganisms recovered can vary considerably with different culture media and incubation conditions.

• 1. Enumeration (counting colonies)

For plate counting and reporting guidance, see Annex E.

B.6.1 In an enumeration technique using colony counts, procedures should be established to address various situations, such as

a) detecting small colonies (e.g. using magnification);

b) counting and reporting unusual colonies (e.g. spreaders);

c) enumerating and reporting crowded plates (e.g. obscured colonies or TNTC plates);

d) reporting counts from serial dilutions.

B.6.2 In the enumeration technique using colony counts, consideration should be given to the number of colonies produced on a plate. This number should be such that each microorganism is able to express itself as a visible colony without being affected adversely by its near neighbours.

B.6.3 Standard plate counting practice normally specifies a lower limit for the number of colonies on a plate. This limit is based on the availability of multiple dilutions from which to choose. Multiple dilutions do not necessarily apply to bioburden determinations for health care products where the bioburden is low, thus there is no need to establish a lower limit for the number of colonies on a plate.

B.6.4 When counting plates the variability in results between technical personnel should be assessed.

B.6.5 The presence of fibres or other material on the surface of the filter or agar can prevent the formation of discrete colonies and thereby make enumeration difficult.

B.6.6 The use of an agar layer poured carefully over the surface of the test plate can provide a test result that is easier to enumerate after incubation when conditions for the potential for spreading microorganisms are present. This is different from the agar overlay procedure mentioned in B.3.2 in that in this case the agar is poured over the surface of the test plate rather than being poured over the surface of the product.

B.6.7 For automated enumeration methods, validation of the system should be performed in accordance with ISO/IEC 17025.

B.6.8 If multiple test conditions are used (e.g. aerobic count from one plate and fungal count from another plate), and there are no colonies recovered, the LOD values are cumulative. For example, if the aerobe count is < 2 CFU and the fungal count is < 2 CFU, then the total count is < 4 CFU.

B.6.9 Spore formers, if characterized by heat shock, can also be reflected within aerobic counts, and can result in overestimation of the bioburden if the spore counts are added to the aerobic counts.

B.6.10 If anaerobic counts are added to aerobic counts, it could result in overestimation of bioburden from double counting aerobic and facultative anaerobic growth. Double counting can be avoided by subculturing the colonies observed from the anaerobic incubation in aerobic conditions to determine if the isolates are facultative or obligate anaerobes.

• 1. Other techniques for detecting microorganisms

Techniques other than colony counts can be used for determining bioburden. These include the determination of microbiological growth prior to being visible, and the measurement of metabolic activity (e.g. impediometry or epifluorescence). Methods that measure metabolic activity are termed “indirect” because, to have a meaning relative to the numbers of microorganisms as defined previously, they have to be calibrated against colony counts. Alternative techniques should be of adequate sensitivity to detect low levels of microorganisms.

Some rapid microbiological methods (e.g. bioluminescence, enzymatic, cytometry) can provide detailed information as to the range and relative numbers of microorganisms present in bioburden and allow assessment of the variability that can occur. They can also provide bioburden information more rapidly than direct culturing.

• 1. Screening for substances affecting bioburden determinations

NOTE Annex D describes techniques that can be used to assess the release of microbicidal or microbiostatic substances

B.8.1 Screening is aimed at investigating the effects on viability of microorganisms due to substances that can be released from the product into an eluent. It is an example of an approach that may be used to assess a technique for conformity with 6.1.2. Annex D provides details on how this an be accomplished.

B.8.2 Products are selected and each should be subjected to the technique for extraction of microorganisms to be used routinely. If the extraction technique uses an eluent, B.8.3 can apply, whereas, if the product is introduced directly into medium, B.8.4 can be more appropriate.

B.8.3 The eluent and any substances that might have been released from the product should not inhibit the growth of microorganisms removed from the product when subjected to the technique for extraction of microorganisms to be used routinely. This is a quantitative approach.

B.8.4 If the product is to be introduced directly into the growth medium (for example, as in an MPN estimation; see B.3.3), the method suitability test described in pharmacopoeias or that has been used in support of test of sterility per ISO 11737-2 can be used. This is a qualitative approach. The same suitability test can qualify both MPN and test of sterility, provided the following are met:

— the media type is the same between the two tests;

— the worst-case sample-to-media volume ratio, if different between the two tests, is used (i.e. either more sample or less media, or both is the worst-case test system with respect to stasis testing).

In this suitability test, the product is introduced into the medium together with low numbers of microorganisms and incubated under the same conditions as proposed for routine bioburden determination. The number of microorganisms used should not be more than 100 CFU. See B.8.5 for the assessment of results. After a defined period, the medium is examined for visible growth.

If a health care product incorporates an antimicrobial substance that can be released slowly into the medium, then it is appropriate to challenge the product-medium combination with a low number of microorganisms at the end of the incubation period.

B.8.5 If the number of microorganisms inoculated and the number recovered differ appreciably (in the quantitative approach), or no growth of the microorganisms is observed (in the qualitative approach), the technique for bioburden determination should be reconsidered. It can be necessary to introduce a dilution, neutralization or filtration stage to reduce, inactivate or remove the inhibitory substance(s). For additional information, see Annex D.

If the effects of the eluent prior to contact with the product need to be assessed, known numbers of microorganisms can be inoculated into both the eluent and a control solution for a time approximating that proposed for routine bioburden determination. After incubation, the recovered microorganisms from the eluent compared to the counts from the control solution.

• 1. Screening for the adverse effects of physical stress

Physical forces can be used to remove microorganisms from the product (see B.2.2). The effects of these forces on the bioburden determination should be taken into consideration. If the effects of physical forces need to be assessed, known low numbers (not more than 100 CFU) should be exposed to the physical forces to be used in the absence of the device. Enumeration of the microorganisms compared to a control gives a measure of the effects of the physical forces.

1. 
   (informative)
   
   Establishment of bioburden recovery efficiency
   1. General

The extraction technique should be justified and defined for each product, or parts thereof, or product group. The documented rationale should include the product, sample size, and choice of recovery technique. If the typical extraction techniques in this document are used, the documented rationale can consist of a documented test method that is specific to the product or product family and that is approved by the manufacturer and laboratory. If extraction techniques are used that are not discussed in this document, additional documentation for the rationale might be needed.

When bioburden data are being used solely for trending purposes then bioburden recovery efficiency might not be required ^[25].

• • 1. Grouping of products for purposes of bioburden recovery efficiency

Products, or parts thereof, that are similar can be grouped together as a product group and a representative product chosen for the bioburden recovery efficiency test. Evaluation criteria for inclusion can include similar types of raw materials, design and size, and manufacturing processes. The results of the bioburden recovery efficiency test can then be applied to all products in the group for future testing.

• • 1. Sample size

C.1.2.1 The number of products, or parts thereof, for which the bioburden recovery efficiency is to be determined, should be selected.

C.1.2.2 Common approaches are to utilize three to ten products for bioburden recovery testing. The sample size should be based primarily on the purpose for which the testing is being performed (e.g. in support of substantiation of a radiation sterilization dose, or an overkill sterilization cycle). When reviewing bioburden recovery efficiency results, a review of the consistency of results, or lack thereof, can indicate that a different extraction method should be applied. Additionally, a larger sample size can provide a more accurate determination of bioburden recovery efficiency.

• • 1. Guidance on selection of bioburden recovery efficiency approach

C.1.3.1 Bioburden recovery efficiency testing is performed to establish a percent recovery or bioburden correction factor that can be applied to bioburden data to account for microorganisms that remain on the product after the extraction technique. Bioburden data that have been adjusted by inclusion of a recovery efficiency or correction factor are understood to more accurately represent the true bioburden count; this is called a bioburden estimate. A bioburden recovery efficiency test can also be used to compare bioburden test methods.

C.1.3.2 The primary determining factor in the selection of a bioburden recovery efficiency approach (i.e. repetitive recovery versus inoculated product) is the level of naturally occurring product bioburden. Generally, the repetitive recovery method is best for products with a higher product bioburden and the inoculated product method is best for products with a lower product bioburden. Bioburden recovery efficiency results and the corresponding bioburden correction factor can differ based on bioburden extraction parameters (e.g. number and type of extractions for repetitive recovery, or use of an inoculated product versus repetitive recovery). Therefore, it is important to consider the reason why bioburden data are being collected and the purpose of the bioburden recovery efficiency determination.

C.1.3.3 Table C.1 summarizes typical product and method characteristics that should be considered when selecting an appropriate bioburden recovery efficiency approach.

Table C.1 — General considerations for selecting a bioburden recovery efficiency approach

C.3.1.4 Complex products with different types of components (e.g. kits, powders) can require more than one type of bioburden recovery efficiency determination if the product is tested in separate containers or different parts are tested using different techniques. This can require the application of more than one bioburden correction factor for items tested using different methods.

C.3.1.5 If the bioburden is low, and if a larger sample size of tested products is desired, then multiple products can be tested together as a pooled sample. In this situation, the bioburden distribution on individual products is not observable. If products are intended to be routinely pooled for testing the bioburden recovery efficiency determination should be performed in the same manner. For example, if five products will be pooled during routine testing, the bioburden recovery efficiency determination should be performed with five pooled products.

Bioburden recovery efficiency results of a test of pooled products can be unique to the number of products pooled. If the number of products pooled changes then a new assessment of bioburden recovery efficiency should be considered.

• 1. Repetitive recovery test
     1. General

The repetitive recovery approach uses the bioburden as it occurs naturally on the product. Sometimes it is referred to as “exhaustive recovery”.

Devices submitted for repetitive recovery bioburden testing can also be used for assessing the natural bioburden of the device (quantity and type). In order to determine the total bioburden for each device, the total number of microbes extracted from each rinse would be added together to determine the total bioburden count for the device.

C.2.1.1 The underlying principle of this approach is that the method of bioburden determination is repeated on the same products until there is a significant decrease seen in the number of microorganisms recovered. After each extraction, the eluent is totally recovered from the product or product portion and enumerated. Results accumulated from the consecutive recoveries are compared. It should be noted, however, that this method is not necessarily precise. The exact relationship between the number of microorganisms recovered and the actual number on the product cannot always be demonstrated.

The exact number of extractions applied will depend upon a number of factors including the nature of the product, the microorganisms that comprise the bioburden and the initial bioburden level. Preliminary experiments or experience with testing similar products can be used to establish the number of repetitions of the extraction process to be applied.

C.2.1.2 The number of colonies counted after initial application of the extraction method is expressed as a fraction of the total number of colonies from all repetitions (i.e. bioburden recovery efficiency).

C.2.1.3 Using the aerobe count for repetitive recovery is and has been industry standard. The aerobe count typically constitutes the majority of microorganisms on a health care product, therefore, in most cases it is a valid representation of the recovery properties for other types of counts.

• • 1. Examples to illustrate calculation of a bioburden recovery efficiency and correction factor

C.2.2.1 In this example, a set of data for validation by repetitive extraction is shown in Table C.2. The data in this example relate to ten replicate health care products and include five extractions in the repetitive recovery tests.

C.2.2.2 From the data in Table C.2, the recovery efficiencies can be calculated as shown in Table C.3.

Table C.2 — Example of repetitive recovery data

Table C.3 — Example of repetitive recovery data (continued)

C.2.2.3 Using average recovery by first extraction and appropriate rounding, the bioburden correction factor for bioburden recovery efficiency would be as shown by Formula (C.1):

(C.1)

For some applications, it can be appropriate to use the lowest recovery percentage value in order to reflect the worst case. This decision can be dependent on the purpose for which the bioburden estimates are to be used. For the data presented in Table C.2, the worst-case bioburden correction factor including appropriate rounding, would be as shown by Formula (C.2):

(C.2)

• 1. Inoculated products test
     1. General

C.3.1.1 An artificial bioburden can be created by inoculating a known number of a selected microorganism onto the product in order to establish bioburden recovery efficiency. The microorganisms can be vegetative cells but the most common approach utilizes aerobic bacterial spores. The use of vegetative microorganisms is difficult in practice because loss of viability can occur on drying.

Microbial inoculation has limitations, such as encrustation, adhesion or non-adhesion of the suspension, and clumping and variation in the level of the inoculum. These limitations should be considered when inoculating products. The inoculated repetitive approach mentioned in A.7.2.2 can help to overcome these limitations.

The devices used during product inoculation method when establishing the extraction efficiency typically are not suitable for also determining the natural bioburden (quantity or type) for the test samples submitted. This is due to the devices typically being previously exposed to a sterilization process prior to use and artificially inoculated with microbes for the purpose of determining the efficiency of the extraction method.

C.3.1.2 A suspension of the microorganisms with which the product is to be inoculated should be prepared and its count determined.

C.3.1.3 Preliminary experiments can be necessary to establish the appropriate dilution. Typically it is appropriate to deposit a known level of microorganisms on the product which will result in a countable range during the plate count step.

C.3.1.4 A number of products, or parts thereof, for which the bioburden recovery efficiency is to be determined should be selected. Consideration should be given on a case-by-case basis as to whether a sterile product is necessary. Each product is inoculated with a volume of the suspension of microorganisms and, if appropriate for the particular product, allowed to dry under laminar airflow conditions. The count of the inoculum is determined at the time of inoculation.

The suspension should be distributed on the product in such a way that the part from which it is most difficult to remove natural bioburden is included. The various material types of the product should also be considered for inoculation.

Inoculation of the microorganisms is typically performed prior to preparation of the product since the preparation steps, such as disassembly, cutting, blending (e.g. shear forces) could influence the recovery. However, certain products might require preparation prior to inoculation.

C.3.1.5 The proposed extraction technique is used to determine the number of inoculated microorganisms that are removed from the product.

C.3.1.6 The number of microorganisms removed is expressed as a fraction of the number inoculated on to the product. This fraction can be calculated for each product and used to establish a bioburden recovery efficiency.

C.3.1.7 To address the potential of a quantitative decrease during the drying of the inoculum, an inoculum control can be used. The inoculum control can be performed by

a) inoculating materials with low adhesion properties followed by drying, extraction, and plating; or

b) inoculating a membrane filter followed by drying and plating.

• • 1. Examples to illustrate calculation of a bioburden correction factor

C.3.2.1 In this example, a set of data for the inoculated recovery test is shown in Table C.4. These data relate to three replicate product items.

C.3.2.2 For establishment of the bioburden recovery efficiency, a product inoculation method was selected because preliminary experiments indicated that the bioburden was very low.

C.3.2.3 An aqueous suspension of Bacillus atrophaeus was prepared and the count of the suspension was determined.

C.3.2.4 A dilution of the suspension was prepared such that 0,1 ml aliquots contained an average of 100 spores. Each device was inoculated with 0,1 ml of this diluted suspension and allowed to dry under laminar airflow.

C.3.2.5 The inoculated products were subjected to the chosen extraction technique and the mean number of spores removed was 76, with a range from 68 to 83.

Table C.4 — Sample data for inoculated recovery test

C.3.2.6 The bioburden correction factor for bioburden recovery efficiency, including appropriate rounding, would be as shown by Formula (C.3):

(C.3)

In some applications the lowest value of the range of percentage recovery in order to reflect the worst case. For the above data the worst-case bioburden correction factor, including appropriate rounding, would be as shown by Formula (C.4):

(C.4)

• • 1. Example to illustrate comparisons of two bioburden recovery efficiency methods

C.3.3.1 In this example, two sets of data for recovery methods are shown in Table C.5. The company established internal acceptance criteria for bioburden recovery efficiency based on risk. These data sets relate to five products that were tested with one technique (initial test), where the average recovery was below the established criteria. Consequently, an additional step was added to the existing technique to determine if the bioburden recovery efficiency was improved (second test).

Table C.5 — Comparison of bioburden recovery efficiency percentage for two recovery methods

C.3.3.2 After modifying the original technique, the bioburden recovery efficiency did improve and met the established criteria. The purpose for bioburden data and the accuracy needed will influence whether or not more consistent data are required or if the bioburden recovery efficiency should be higher in order to better estimate the recovered bioburden.

• 1. Bioburden recovery efficiency for complex product testing

C.4.1 In the example shown in Table C.6 multiple recovery methods are needed to estimate the bioburden level of a complex product. This example shows how two different bioburden correction factors can be applied to the respective groups of bioburden data. In order to establish the bioburden level of this complex product, all three of the bioburden estimates should be added together.

C.4.2 When testing tissue and biological products, additional guidance can be obtained from AAMI TIR37. Products that intentionally break down over time (e.g. drug eluting or bio-absorbable products) should account for these alterations in development of the bioburden recovery efficiency test method.

Table C.6 — Complex product bioburden estimate determined by utilizing two bioburden correction factors and MPN result

• 1. Data analysis and application of bioburden correction factor

C.5.1 Due to the variability of design, materials, product configurations, manufacturing processes , it is not required by this document that a particular bioburden recovery efficiency result be obtained. However, if bioburden recovery efficiency results fall below a target or desired value, another technique should be attempted (e.g. addition of another extraction method or lengthening the current extraction method) to determine if better results can be obtained.

Items that can be considered in determination of a desired bioburden recovery efficiency value for a health care product include the following:

a) sterilization validation approach (e.g. overkill versus bioburden based);

b) use of bioburden data (e.g. support sterilization validation approach, raw material screening, trending);

c) the type of product or material being tested (e.g. plastics and metals versus absorbent materials);

d) product properties (e.g. simple geometry without any cavity versus complex geometry with cavities, smooth versus rough surfaces);

e) robustness of recovery method used (e.g. ultrasonication, shaking, or a combination of both).

Based on these concepts, a low bioburden recovery efficiency (e.g. for an absorbent or complex product) can be considered acceptable. Consideration for use of the lowest recovery percentage value to reflect the most conservative worst case estimate can be appropriate, as described in C.2.2.3 and C.3.2.6. Also, it should be noted there are times that determination of bioburden recovery efficiency is not necessary (e.g. component or raw material screening, or if the product is a liquid in which the entire contents are being filtered).

In microbiological test methods it is expected to obtain more variability than is typically observed in more predictable physical science test methods (e.g. chemistry or physics). This greater variability is largely due to the fact that microorganisms are viable and the number of microorganisms can change over time depending on conditions. Other factors that also influence bioburden recovery efficiency can include clumping of microorganisms, consistency of microorganisms depositing on the product surface, product’s surface characteristics (e.g. coatings with specific silicone material, high porous surface areas), incubation conditions, and/ or inherent limitations in the ability to detect or measure the microorganisms.

Nevertheless, unexpectedly low or widely distributed bioburden recovery efficiency might not be appropriate depending on the criticality and purpose of the bioburden data, and, if this is the case, further improvement of the extraction technique (e.g. enhanced by disassembling, more intensive mechanical shaking, active rinsing of cavities, prolongation of rinsing time, modification of eluent) should be investigated. An example where criticality and purpose of the bioburden data can warrant more effort and resources to obtain better recovery results is when the bioburden data are used to establish a “bioburden-based” sterilization process (e.g. radiation sterilization, and, in particular, dose establishment methods that require low bioburden counts). An example where criticality and purpose of the bioburden data might not warrant more effort and resources to obtain better recovery results can be for the application of bioburden component screening.

C.5.2 In calculating a bioburden recovery efficiency apply either the limit of detection (e.g. a “less than” value) or the plate count value (e.g. 0 CFU).

C.5.3 When reviewing bioburden recovery efficiency results it is appropriate to round all values to one decimal place.

C.5.4 The bioburden can be adjusted by either the recovery efficiency percentage or the correction factor. The bioburden recovery efficiency is applied to bioburden data by dividing the bioburden average by the recovery efficiency as a decimal (e.g. 0,75 for 75 %). The bioburden correction factor is applied to bioburden data by multiplying the bioburden average by the correction factor. When the bioburden has been adjusted using either option the resulting value is termed bioburden estimate. In some applications, it can be decided to apply the lowest bioburden recovery efficiency value of the range obtained to determine the bioburden correction factor to reflect the worst case. This decision will be influenced by the use of the data.

1. 
   (informative)
   
   Guidance on bioburden method suitability testing for health care products
   1. General

D.1.1 Validation of the method for determination of bioburden as specified in 6.1.2 requires an assessment of test method suitability to demonstrate lack of inhibition of growth. Demonstration of lack of inhibition of growth for a specified bioburden method can be done by providing a rationale or by testing.

D.1.2 If classical microbiological methods are used for the bioburden test, no official validation of the overall test method itself is needed. See A.7.1. As the bioburden test methods described in this document are considered classical microbiological methods, a method suitability test is appropriate rather than a test method validation.

D.1.3 Bioburden method suitability is assessed when a product contains antimicrobials or some inhibitory substance/properties that can interfere with recovery and culturing in the bioburden test. However, health care products made of traditional synthetic materials, such as polymers, nonwovens, metals, or ceramics, typically do not exhibit inhibitory properties, unless the product incorporates antimicrobial substances. Where a rationale cannot be provided about the materials that comprise the product, consider performing testing to demonstrate this lack of inhibitory properties.

D.1.4 When a rationale is used for bioburden method suitability testing, it can be supported in several ways, including

a) a materials and manufacturing process assessment;

b) a sterility suitability test (bacteriostasis/fungistasis) having comparable test parameters to those in the bioburden test method (e.g. media with similar ingredients but different properties due to the presence of agar, incubation temperatures intended for growth of aerobic, mesophilic type organisms); or

c) an inoculated product recovery efficiency test showing a high percent recovery.

NOTE It is possible that spores, if used, are not sensitive to inhibitory substances during the extraction process.

D.1.5 In cases where there is little or no expectation of inhibitory properties, a bioburden method suitability test using a single microorganism type could be used to assess whether a product has inhibitory substances that would be manifested in the bioburden test. For those health care products that could potentially have inhibition in the bioburden test, the bioburden method suitability test can also indicate the need for further/more in-depth testing (e.g. using more than one microorganism type).

D.1.6 There are two aspects of bioburden method suitability that can be assessed. They can be assessed separately or together. The first aspect is the effect on microorganism viability of being in the extraction media for a period of time in the presence of potential inhibitory substances from the product. The second is the effect on microorganism replication of being on or in the growth medium in the presence of potential residual inhibitory substances.

Due to the many factors involved, the laboratory needs to provide a rationale for the approach selected, including the microorganism type(s). This annex provides a simplified approach using a single microorganism type that has been rationalized and documented by the laboratory.

D.1.7 It is recommended that products used in the simplified bioburden method suitability test have been subjected to a sterilization process prior to commencement of the test. By using products that have undergone a sterilization process, the risk of enumerating colonies from the product’s natural bioburden is reduced or eliminated. If the product is expected to have a very low natural bioburden (e.g. < 10 CFU) utilizing non-sterilized products in this test could be appropriate.

• 1. Simplified bioburden method suitability test

A recommended simplified bioburden method suitability procedure is as follows.

• • 1. Number of samples

In performing a simplified bioburden method suitability test, at least duplicate or triplicate samples should be tested, and the results should be reported as the average product value.

• • 1. Microbiological challenge

The microorganism selected for the screening procedure outlined here should be selected based on a rationale. Consideration should be given to the genus to be used, i.e. using a microorganism that can be expected to be part of the product’s natural bioburden or an appropriate challenge to the test system ^[37][42].

The choice of microorganism(s) should be based on factors such as the following:

a) a well-characterized microorganism that can reliably be cultured and quantified or purchased;

b) a microorganism that is used in many challenge studies for microbiological purposes and is typically one of the main choices for a panel of microorganisms in studies measuring inhibitory qualities;

c) a microorganism(s) in a prepared solution that can be purchased in a stable form with a standardized titre, making it easy to maintain and use.

• • 1. Inoculation procedure

To perform a simplified bioburden method suitability test, the extract solution containing the product should be inoculated. The inoculum level should be representative of the expected bioburden level but should be approximately 1 to 2 x 10² CFU.

The product itself can be inoculated instead of the extract, but this is considered a worse case situation and factors associated with recovery efficiency can interfere.

• • • 1. Inoculating the extract solution

The product is put into the extract solution as would be done for the bioburden test, and then the extract solution containing the product is inoculated. The bioburden test would then be performed as specified. This is preferable and is particularly applicable in situations where recovery efficiency can interfere with the recovery of vegetative cells in the suitability test. If membrane filtration is being used, it is possible to inoculate the extract solution after extraction and placement in a funnel for filtration.

If recovery efficiency is not known, this technique is a more logical choice.

• • • 1. Inoculating the product

The product itself is inoculated, allowing contact for a few minutes and then is placed into the extract solution in the same manner as the bioburden test. Inoculating the product is considered a worse case compared to inoculating the extract, as inoculating the product involves potential inhibition from contact with the product before it is in the extract solution, as well as potential factors associated with recovery. This is more applicable in situations where a product has a high recovery efficiency as it does not significantly impact recovery in the suitability test.

Using vegetative cells in inoculation and recovery always poses the potential for loss of viability upon inoculation and drying. For this reason, it is not recommended in the simplified bioburden method suitability to dry a vegetative cell inoculum. The subsequent reduction in count might only be due to loss of viability (bacterial death) upon drying and not due to product related inhibition of microbiological growth in the performance of the test.

• • • 1. Test controls

As in all testing, there should be positive and negative controls. At least duplicate or triplicate controls should be tested in line with the number of samples tested. Results should be evaluated as the average product value. If inoculation of the product itself is used, an inoculum contact control should be done, consisting of an inert object (such as a coupon) showing the level of the inoculated microorganism after the contact process, to evaluate any loss of viability or extraction dynamics that can occur during the contact process.

An inoculum contact control does not apply where the extract solution is inoculated.

• • 1. Comparing the inoculum level to the recovered level

The recovery of the challenge microorganism is compared to the inoculum level. For comparison of the inoculum control and the test results, averages should be used.

• 1. Evaluation of results

D.3.1 Due to the variability of design, materials, product configurations, manufacturing processes, it is not required by this document that a particular recovery for bioburden suitability be obtained. However, if bioburden suitability results fall below a target or desired value, a modification to the bioburden test might be warranted to determine if better results can be obtained.

D.3.2 Acceptance criteria that can be considered are as follows:

a) criteria found in current pharmacopeial chapters on Microbiological Examination of Nonsterile Products: Microbial Enumeration Tests, (e.g. USP <61> and EP 2.6.12) where 50 % recovery is considered acceptable; or

b) other appropriate criteria.

Regardless of the criterion chosen, a bioburden method suitability test showing an acceptable recovery is considered an indication that there are no appreciable inhibitory properties, meaning acceptability of the suitability of the specified bioburden test method. For results that fall below the criterion selected, it is possible that some degree of inhibition is indicated. Modification to or neutralization of the bioburden test method should be considered, followed by another bioburden method suitability test using the modified test to assess the effectiveness of the modifications. Alternatively, further, in-depth testing per one of the previously mentioned pharmacopeial methods can be performed ^[30].

1. 
   (informative)
   
   Guidelines for counting plates and recording results

NOTE Content adapted from AAMI TIR106:2024, Annex C. Different country-specific limits and calculation principles can also be applied, such as given by EN 13727:2012, 5.6.2.2 and EN 13624:2013, 5.6.2.2.

• 1. General

The plate counting and reporting rules specified in classic microbiological manuals were originally intended to be used for microbiology tests related to samples such as food and water. Because of this, the rules were typically based on a sample having

a) a significant microbial load;

b) several dilutions from which to choose;

c) duplicate or triplicate plates at each dilution.

For bioburden testing of health care products and related items (e.g. components), the microbial load can be very low and therefore only one dilution (or aliquot) may be necessary. Additionally, duplicate, or triplicate plates at each dilution are not always performed. These differences necessitate using plate counting and reporting rules that are appropriate for the microbial expectations for bioburden testing.

• • 1. Guidelines for counting plates

Direct counting of all colonies on the plate should be done whenever possible. If actual counts of each colony cannot be performed, portions of the plate can be counted followed by calculations (see E.2.2). A colony counter, stereomicroscope, or automated plate counting software/technology (when appropriately validated) can be used to facilitate counting.

• • 1. Ideal counts (countable range)

Ideal counts for plates were originally developed for a situation where there were several dilutions from which to choose. The ranges for ideal counts, such as 25 to 250 or 30 to 300 CFU/plate were chosen so that the lower limit would be statistically valid when choosing which plates to use from a series of dilutions, and the upper limit would be microbiologically valid in terms of colony crowding.

• • • 1. Lower ideal count

In cases where serial dilutions are performed, the typical ideal countable range can apply to assist in selecting which results to report. For bioburden test results where a single dilution or volume has been tested, a lower ideal count does not apply.

• • • 1. Upper ideal count

The upper ideal count applies for any type of sample due to the potential for crowding of colonies. Crowding can present a problem for nutrient competition as well as potentially interfering with accurate counts.

The upper countable range in most literature for filters is 200 CFU, which is based mostly on the size of the filter. In many cases, especially when spreaders, fungi, or large colonies are present, 200 CFU can be too high for countability.

The upper countable range in most literature for pour plates and spread plates is 250 to 300 CFU, which is based on countability. In many cases, especially when spreaders, fungi, or large colonies are present, 250 to 300 CFU can be too high for countability.

Regardless of the specified limit for the upper ideal count, the plate can be counted or estimated (see E.2.2), depending on the use of the data.

• 1. Plate counting guidelines
     1. Counts that are below the upper countable range and do not have to be estimated

Count all the colonies directly using any method that facilitates accurate counts (tally counter, marking plates).

• • 1. Counts that are above the upper countable range, but countable

Count all the colonies, directly if possible.

NOTE Colonies that are not evenly distributed are counted directly, and the process described below does not apply.

If colonies are evenly distributed, but it is not possible to count all colonies, count a portion of the plate or filter and multiply appropriately to obtain the count for the whole plate. For example, count one-half of the plate and multiply by two (2) to obtain the total, or count one-fourth of the plate and multiply by four (4) to obtain the total.

If colonies are evenly distributed, but it is not possible to count the whole plate or a fraction of the plate (e.g. one-half or one-fourth of the plate), estimate the counts by counting the colonies in at least five (5) squares using the background grid of a suitable colony counter and averaging the count per square; multiply by the appropriate factor to get an estimate for the whole plate.

For filters, multiply the average count per square by an appropriate factor to get an estimate of the whole filter.

Any estimated counts should be documented and reported as an estimate.

• • 1. Counts that are beyond the countable range but can be semi-quantified

If possible, a plate should be counted, even if the count is based on an estimation. A count that is beyond the counting range can often be assigned a semi-quantitative value if that value can be approximated based on the presence of discernible colonies. This helps put the count in perspective, rather than implying that the count is an undefinable number.

Reporting a TNTC result without any type of quantification can be misconstrued as a count well in excess of what it really is (i.e. millions versus thousands). If a plate is beyond the countable range, but still can be semi-quantified, it should be.

For example, if there are small discernible colonies and it appears the maximum number of colonies that could be counted in a small square on a colony counter is 12 CFU, then results for that sample could be estimated as follows:

12 CFU x 9 (number of small squares in a large square) = 108 CFU per large square

108 CFU per large square x 58 (number of large squares) = 6 264 CFU per whole plate

A count of 6264 should not be reported in this case since it is estimated and could never be this precise. However, the estimated count can be reported in several ways to provide a semi-quantitation of a count that might otherwise be reported as TNTC. For the above example, some reporting options are:

a) (EST 6,3 x 10³ CFU);

b) (EST < 10,000 CFU);

c) (EST 6,000 – 7,000 CFU);

d) (EST 5,000 – 10,000 CFU).

• • 1. Counts that are TNTC

If the colonies on a plate are too numerous to be estimated or semi-quantified by E.2.2 or E.2.3, record the count for that plate as TNTC. Reporting TNTC without a value will apply when there is no possible way to discern individual colonies, and when it is determined that the overgrowth is due to the presence of many unique colonies that have grown together rather than being a spreader.

Consideration should be given to whether TNTC results should be treated as excursions. This will depend on the purpose for which the data are to be used e.g. the effectiveness of the sterilization process.

TNTC results are omitted from the average for a group of samples since there is no numerical value to use. This is an acceptable practice, especially if it is determined that the TNTC result is an anomaly or due to potential lab contamination. The minimum number of samples, excluding TNTC omissions, should meet the requirements of all applicable standards (e.g. ISO 11137-2).

• • 1. Spreaders

Spreading colonies (or “spreaders”) are those that exhibit over-growth that can influence the accuracy of the plate count. Spreaders come in several forms including a chain, a lawn, or a film at the edge of a filter or the agar.

If a spreader(s) is present, but it does not obscure other colonies (due to size or opaqueness), count the other colonies, and count the spreader as one colony.

NOTE Plate counting prior to the final enumeration can help when spreaders are present (i.e. a count at 24 or 48 hours), in addition to the final enumeration. Additionally, an agar-overlay technique can be useful when dealing with spreaders to suppress the mode of spreading.

If the spreader(s) obscures other colonies, but it covers ≤ 50 % of the plate/filter, count half of the plate/filter (portion not including the spreader) and multiply by 2 to obtain the estimated total.

If the spreader(s) obscures other colonies, and it covers > 50 % of the plate/filter, count the portion of the plate/filter that is not obscured (e.g. 20 %) and adjust the total count by the approximate value of that portion to obtain an estimated total.

If a spreader obscures the entire plate/filter, such that counting or estimation is not possible, report as an uncountable spreader. The count can indicate that the spreader is > 1 CFU.

Uncountable spreader results are omitted from the average for a group of samples since there is no numerical value to use. This is an acceptable practice, especially if it is determined that the spreader is an anomaly or due to potential lab contamination. The minimum number of samples, excluding TNTC omissions, should meet the requirements of all applicable standards (e.g. ISO 11137-2).

Any countable plates with spreaders should be documented as containing a spreader(s).

• • 1. Laboratory error

If a plate has obvious contamination, report accordingly; if there is a replicate plate, use the count from that replicate plate only.

An experienced microbiologist might be able to determine if there is obvious contamination based on the type(s) of microorganisms, the sample itself, the colonies on other plates from the same set, knowledge of the procedure, and technical experience. Indications can include:

a) significant level of a single colony morphology on a plate or replicate set, where the other samples in the group have a variety of colony morphologies or the history of that type of sample shows a variety of colony morphologies;

b) one plate in a replicate set with significantly more colonies than the other(s);

c) an excess of microorganisms that are part of the lab’s environmental flora but not the product’s flora (e.g. fungus);

d) a spreader on one plate or replicate set with no other spreaders on other samples in the group;

e) one sample in a group of samples with a significantly higher count (e.g. 2 or more logs higher), with no history of similar bioburden spikes;

f) a higher dilution that is not consistent with the results of the sample without dilution or at a lower dilution.

• 1. Calculating sample results

For all plates/filters, average the replicate plates (if applicable) and multiply the average by the appropriate dilution factor to obtain the count for that aliquot or dilution.

When more than one aliquot or dilution has plates within the ideal counting range, use the least diluted aliquot.

When all aliquots or dilutions are greater than the ideal counting range, use the plates from the highest dilution to calculate results.

When all aliquots or dilutions have zero colonies, report as less than the LOD of the test.

• 1. Reporting results

Report each sample result to the number of significant figures deemed to be appropriate based on the data from the test. Individual bioburden results are usually reported in whole numbers.

For bioburden estimates, report the mathematical average to one decimal place. If the number is > 1 000 CFU, report the average as a whole number. Under some radiation sterilization circumstances, the average may need to be reported to two decimal places.

For recovery efficiency values, report individual results and averages to one decimal place.

1. 
   (informative)
   
   Bioburden excursions
   1. General

F.1.1 The bioburden monitoring program should include procedures for investigating excursions. The bioburden excursion investigation program should identify the correct procedures and personnel to be notified of the excursions to ensure appropriate actions are taken in a timely manner.

F.1.2 The components of a bioburden excursion investigation might include many factors such as a laboratory investigation, analysis of data, and an investigation of the manufacturing process.

• 1. Investigating bioburden excursions

F.2.1 Due to the time-sensitive nature of certain components of an investigation, it is common practice to initiate and perform different components of the investigation in parallel. In cases where a probable or definitive root cause has been identified, some aspects of an investigation may be omitted.

F.2.2 Single-batch averages exceeding a calculated alert level are not necessarily evidence that a significant bioburden shift has occurred but can provide a signal that the trending of bioburden data should be reviewed to determine the extent of the excursion. When an average result exceeds an alert level multiple times over a short period of time, an investigation should be performed. It should be noted that the investigation might not trigger corrective action if no root cause can be determined for an alert level excursion.

F.2.3 Bioburden excursions over the action level represent a potential shift in bioburden and usually requires an investigation to determine root cause and corrective action, if possible.

Investigation of a bioburden excursion can include one or more of the following activities:

a) review of previous results to determine if alert/action levels have been recently exceeded;

b) characterization of recovered microorganisms;

c) evaluation of laboratory controls and monitors;

d) review of changes to sampling and testing procedures;

e) review of changes to the manufacturing process;

f) additional testing (same or different batch);

g) review of cleaning and disinfection of the manufacturing area;

h) review of changes in raw materials and suppliers;

i) review of controls and monitors (e.g. water systems, environmental).

• • 1. Laboratory investigation

F.2.4.1 The initial step of a bioburden excursion investigation typically involves a review of the laboratory data to ensure that the data are valid. Laboratory quality management systems can already include this as part of standard processes prior to release of the laboratory result. In this case, it is possible that further laboratory investigation is not necessary.

F.2.4.2 The test laboratory providing bioburden results has the responsibility to have a process that does not contribute bioburden to the product. The laboratory should be able to show conformance to procedures and established practices, such as the use of proper aseptic technique, suitable equipment, media, and reagents, and necessary competency.

• • 1. Data analysis

In addition to verifying the validity of the data, during the initial stage of the investigation, a review of the bioburden data should be performed to determine if there is a possible adverse trend. If no trend is observed, the scope and extent of the investigation can be reduced to address the isolated excursion event.

• • 1. Manufacturing investigation

F.2.6.1 A detailed evaluation of the manufacturing process and controls is typically performed when the initial steps of the investigation have determined that this is necessary.

F.2.6.2 An investigation of the manufacturing process can consider the following.

a) Microbial characterization to determine the potential source of contamination and evaluate the potential resistance of the microorganisms to the sterilization method.

b) Additional sampling to determine if the excursion is an isolated or persistent event.

c) Raw material, component, or subassembly bioburden testing to identify the point in manufacturing where a potential change might have occurred.

d) A review of the manufacturing process to evaluate potential contributors from the manufacturing process. Areas of investigation can include:

1) changes to equipment or process;

2) material and process controls;

3) relocation of equipment or processing lines;

4) activity in or around the manufacturing environment (e.g. construction or other disruptive activity);

5) adverse events (e.g. power outage) leading to a loss of one or more engineering controls in the manufacturing environment at time of excursion;

6) changes in the number of personnel located in and around the manufacturing process.

e) Review training of manufacturing personnel on contamination control procedures and processes and determine if training is adequate.

f) Review environmental monitoring results and documentation of cleaning for the manufacturing environment.

• • 1. Conclusions of the investigation

Document the results of the investigation. If potential root cause(s) are determined, indicate actions to be taken. It is common that a definitive root cause for a bioburden excursion cannot be identified.

• 1. Impact assessment

F.3.1 The impact assessment for the bioburden excursion should take into consideration how the data are used, and the risk to either product or process quality. An impact on quality is uncommon for product sterilized using non-bioburden based (e.g. overkill) processes. For sterilization processes validated using bioburden-based methods, this can require more information and potentially more testing before any conclusions can be made.

F.3.2 If a shift in bioburden has occurred and it is determined in the investigation that it does not affect the Sterility Assurance Level (SAL) for the product, a review of the alert and action levels can be conducted to determine if these levels are still appropriate or should be revised. In some instances, a change in alert and action levels might have regulatory implications.

F.3.3 If warranted, the impact assessment can also indicate the implementation of modifications to the monitoring program (i.e. frequency or sample size).

1. 
   (informative)
   
   Typical assignment of responsibilities

The manufacturer and laboratory should have an agreement that assigns responsibilities for the completion of the requirements as defined in this document. Ultimately, the manufacturer is responsible to ensure that the requirements are met. This annex gives information on typical assignments. The requirements given in Table G.1 are abbreviated. See the specific clause for details regarding each requirement.

Table G.1 — Typical assignment of responsibilities

Annex ZA
(informative)

Relationship between this European standard and the General Safety and Performance Requirements of Regulation (EU) 2017/745 aimed to be covered

NOTE Annex ZA is not included in the final ISO publication.

This European standard has been prepared under M/575 to provide one voluntary means of conforming to the General Safety and Performance Requirements of Regulation (EU) 2017/745 of 5 April 2017 concerning medical devices [OJ L 117] and to system or process requirements including those relating to quality management systems, risk management, post-market surveillance systems, clinical investigations, clinical evaluation or post-market clinical follow-up.

Once this standard is cited in the Official Journal of the European Union under that Regulation, compliance with the normative clauses of this standard given in Table ZA.1 and application of the edition of the normatively referenced standards as given in Table ZA.2 confers, within the limits of the scope of this standard, a presumption of conformity with the corresponding General Safety and Performance Requirements of that Regulation, and associated EFTA Regulations.

Where a definition in this standard differs from a definition of the same term set out in Regulation (EU) 2017/745, the differences are indicated in Table ZA.3. For the purpose of using this standard in support of the requirements set out in Regulation (EU) 2017/745, the definitions set out in this Regulation prevail.

Where the European standard is an adoption of an International Standard, the scope of this standard can differ from the scope of the European Regulation that it supports. As the scope of the applicable regulatory requirements differ from nation to nation and region to region, the standard can only support European regulatory requirements to the extent of the scope of the European regulation for medical devices (EU) 2017/745).

NOTE 1 Where a reference from a clause of this standard to the risk management process is made, the risk management process needs to be in compliance with Regulation (EU) 2017/745. This means that risks have to be ‘reduced as far as possible’, ‘reduced to the lowest possible level’, ‘reduced as far as possible and appropriate’, ‘removed or reduced as far as possible’, ‘eliminated or reduced as far as possible’, ’removed or minimized as far as possible’, or ‘minimized’, according to the wording of the corresponding General Safety and Performance Requirement.

NOTE 2 The manufacturer’s policy for determining acceptable risk must be in compliance with General Safety and Performance Requirements 1, 2, 3, 4, 5, 8, 9, 10, 11, 14, 16, 17, 18, 19, 20, 21 and 22 of the Regulation.

NOTE 3 When a General Safety and Performance Requirement does not appear in Table ZA.1, it means that it is not addressed by this European Standard.

NOTE 4 When a General Safety and Performance Requirement does not appear in Table ZA.1, it means that it is not addressed by this European Standard.

Table ZA.1 — Correspondence between this European standard and Annex I of Regulation (EU) 2017/745 [OJ L 117] and to system or process requirements including those relating to quality management systems, risk management, post-market surveillance systems, clinical investigations, clinical evaluation or post-market clinical follow-up

Table ZA.3 — Terms and definitions from clause 3 of this document which differ from a definition of the same term set out in Regulation (EU) 2017/745

WARNING 1: Presumption of conformity stays valid only as long as a reference to this European standard is maintained in the list published in the Official Journal of the European Union. Users of this standard should consult frequently the latest list published in the Official Journal of the European Union.

WARNING 2: Other Union legislation may be applicable to the product(s) falling within the scope of this standard.

Annex ZB
(informative)

Relationship between this European standard and the General Safety and Performance Requirements of Regulation (EU) 2017/746 aimed to be covered

NOTE Annex ZB is not included in the final ISO standard.

This European standard has been prepared under M/575 to provide one voluntary means of conforming to the General Safety and Performance Requirements of Regulation (EU) 2017/746 of 5 April 2017 concerning in vitro diagnostic medical devices [OJ L 117] and to system or process requirements including those relating to quality management systems, risk management, post-market surveillance systems, performance studies, clinical evidence or post-market performance follow-up.

Once this standard is cited in the Official Journal of the European Union under that Regulation, compliance with the normative clauses of this standard given in Table ZB.1 and application of the edition of the normatively referenced standards as given in Table ZB.2 confers, within the limits of the scope of this standard, a presumption of conformity with the corresponding General Safety and Performance Requirements of that Regulation, and associated EFTA Regulations.

Where a definition in this standard differs from a definition of the same term set out in Regulation (EU) 2017/746, the differences are indicated in Table ZB.3. For the purpose of using this standard in support of the requirements set out in Regulation (EU) 2017/746, the definitions set out in this Regulation prevail.

Where the European standard is an adoption of an International Standard, the scope of this standard can differ from the scope of the European Regulation that it supports. As the scope of the applicable regulatory requirements differ from nation to nation and region to region, the standard can only support European regulatory requirements to the extent of the scope of the In vitro Diagnostic Regulation (EU) 2017/746).

NOTE 1 Where a reference from a clause of this standard to the risk management process is made, the risk management process needs to be in compliance with Regulation (EU) 2017/746. This means that risks have to be ‘reduced as far as possible’, ‘reduced to a level as low as reasonably practicable’, ‘reduced to the lowest possible level’, ‘reduced as far as possible and appropriate’, ‘removed or reduced as far as possible’, ‘eliminated or reduced as far as possible’, ‘prevented’ or ‘minimized’, according to the wording of the corresponding General Safety and Performance Requirement.

NOTE 2 The manufacturer’s policy for determining acceptable risk must be in compliance with General Safety and Performance Requirements 1, 2, 3, 4, 5, 8, 10, 11, 13, 15, 16, 17, 18 and 19 of the Regulation.

NOTE 3 When a General Safety and Performance Requirement does not appear in Table ZB.1, it means that it is not addressed by this European Standard.

Table ZB.1 — Correspondence between this European standard and Annex I of Regulation (EU) 2017/746 [OJ L 117] and to system or process requirements including those relating to quality management systems, risk management, post-market surveillance systems, performance studies, clinical evidence or post-market performance follow-up.

Table ZB.3 — Terms and definitions from clause 3 of this document which differ from a definition of the same term set out in Regulation (EU) 2017/746

WARNING 1: Presumption of conformity stays valid only as long as a reference to this European standard is maintained in the list published in the Official Journal of the European Union. Users of this standard should consult frequently the latest list published in the Official Journal of the European Union.

WARNING 2: Other Union legislation may be applicable to the product(s) falling within the scope of this standard.

Bibliography

[1] ISO 7870‑2, Control charts — Part 2: Shewhart control charts

[2] ISO 7870‑4, Control charts — Part 4: Cumulative sum charts

[3] ISO 9000:2015, Quality management systems — Fundamentals and vocabulary

[4] ISO 9001, Quality management systems — Requirements

[5] ISO 11135, Sterilization of health-care products — Ethylene oxide — Requirements for the development, validation and routine control of a sterilization process for medical devices

[6] ISO 11137 (all parts), Sterilization of health care products — Radiation

[7] ISO 11139:2018, Sterilization of health care products — Vocabulary of terms used in sterilization and related equipment and process standards

[8] ISO 11737‑2, Sterilization of health care products — Microbiological methods — Part 2: Tests of sterility performed in the definition, validation and maintenance of a sterilization process

[9] ISO 13004, Sterilization of health care products — Radiation — Substantiation of selected sterilization dose: Method VD_max^SD

[10] ISO 13022, Medical products containing viable human cells — Application of risk management and requirements for processing practices

[11] ISO 13485, Medical devices — Quality management systems — Requirements for regulatory purposes

[12] ISO 14160, Sterilization of health care products — Liquid chemical sterilizing agents for single-use medical devices utilizing animal tissues and their derivatives — Requirements for characterization, development, validation and routine control of a sterilization process for medical devices

[13] ISO 14937, Sterilization of health care products — General requirements for characterization of a sterilizing agent and the development, validation and routine control of a sterilization process for medical devices

[14] ISO 15189, Medical laboratories — Requirements for quality and competence

[15] ISO/IEC 17025, General requirements for the competence of testing and calibration laboratories

[16] ISO 17665, Sterilization of health care products — Moist heat — Requirements for the development, validation and routine control of a sterilization process for medical devices

[17] ISO 20857, Sterilization of health care products — Dry heat — Requirements for the development, validation and routine control of a sterilization process for medical devices

[18] ISO 22442‑3, Medical devices utilizing animal tissues and their derivatives — Part 3: Validation of the elimination and/or inactivation of viruses and transmissible spongiform encephalopathy (TSE) agents

[19] ISO 22441, Sterilization of health care products — Low temperature vaporized hydrogen peroxide — Requirements for the development, validation and routine control of a sterilization process for medical devices

[20] ISO 25424, Sterilization of health care products — Low temperature steam and formaldehyde — Requirements for development, validation and routine control of a sterilization process for medical devices

[21] ISO/TS 22456, Sterilization of healthcare products — Microbiological methods— Guidance on conducting bioburden determinations and tests of sterility for biologics and tissue-based products

[22] EN 13624:2013, Chemical disinfectants and antiseptics - Quantitative suspension test for the evaluation of fungicidal or yeasticidal activity in the medical area - Test method and requirements (phase 2, step 1)

[23] EN 13727:2012, Chemical disinfectants and antiseptics - Quantitative suspension test for the evaluation of bactericidal activity in the medical area - Test method and requirements (phase 2, step 1)

[24] AAMI TIR37, Sterilization of health care products — Radiation — Guidance on sterilization of biologics and tissue-based products

[25] AAMI TIR106:2024, Microbiological Methods — Understanding and use of product bioburden data

[26] ICH Q5A(R1), Viral Safety Evaluation Of Biotechnology Products Derived From Cell Lines Of Human Or Animal Origin

[27] Bailey, M. Notes on bioburden distribution metrics: The log-normal distribution. Panel on Gamma and Electron Beam. March 2010

[28] Briglia C. The story behind the 0,45 um membrane pore size rating. PDA Letter, September 2015

[29] Bryans, T., Hansen, J. The Bioburden Estimate: Not Just Math, But Microbiology. AAMI Industrial Sterilization Research from the Field. 2013

[30] Bryans, T., Leckwart, C., Santos, H. Bioburden Method Suitability: A practical solution to screening for inhibition. BI&T. 2023, 57 (3) pp. 81–86

[31] Bushar, H.F., Kowalski, J.B., Mosley, G. Estimation of average bioburden values for low-bioburden products. MD+DI. July 2011, 33 (7)

[32] Cochran, W. Estimation of Bacterial Densities by Means of the Most Probable Number. Biometrics. 1950, 6 pp. 105-116

[33] DeMan, J.C. M.P.N. Tables Corrected. European J. Appl. Microbiol. 1983, 17 pp. 301–305

[34] European Pharmacopoeia 11, 2.6.12, Microbiological Examination of Non-Sterile Products (Total Viable Aerobic Count). European Directorate for the Quality of Medicines & HealthCare. 11^th edition, effective 1 January 2023.

[35] European Pharmacopoeia, Chapter 5.1.6 Alternative Methods for Control of Microbiological Quality. European Directorate for the Quality of Medicines & HealthCare. 11^th edition, effective 1 January 2023.

[36] FDA Bacteriological Analytical Manual, Annex 2, 2023

[37] Gould, G.W., Hurst, A., The Bacterial Spore. Academic Press, London. 1969

[38] International Conference on Harmonization (ICH) Validation of Analytical Methods: Definitions and Terminology (CPMP/ICH/381/95)

[39] International Conference on Harmonization (ICH) Validation of Analytical Methods: Methodology (CPMP/ICH/281/95)

[40] McDonnell, G., Hansen, J., ed. Block’s Disinfection, Sterilization, and Preservation, Chapter 64. 6th ed. Philadelphia: Wolters Kluwer, 2021

[41] PDA Technical Report No 33. Evaluation, Validation and Implementation of Alternate and Rapid Microbiological Testing Methods. 2013

[42] Sonenshein, A.L., Hoch, J.A., Losick, J.A. Bacillus subtillis and Other Gram-Positive Bacteria. American Society for Microbiology, Washington. 1993

[43] US Pharmacopoeia, USP-NF 2023, Issue 1, effective 1 May 2023, <61> Microbiological Examination of Nonsterile Products: Microbial Enumeration Tests, United States Pharmacopeial Convention Inc.: Rockville, MD.

[44] US Pharmacopoeia, USP-NF 2023, Issue 1, effective 1 May 2023, <1223> Validation of Alternative Microbiological methods, United States Pharmacopeial Convention Inc.: Rockville, MD.

[45] US Pharmacopoeia, USP-NF 2023, Issue 1, effective 1 May 2023, <1225> Validation of Compendial Procedures and <1226> Verification of Compendial Procedures. United States Pharmacopeial Convention Inc., Rockville, MD.

[46] Winters, M. Consider the Most Probable Number Method for Bioburden Testing. MD&DI. 22 October 2018

[47] Winters, M., Patch, E., Wangsgard, W., Bushar, H., Ferry, A. Establishing Bioburden Alert and Action Levels. MD&DI. 31 May 2013