How to Determine Bioburden: Key Points of ISO 11737-1

Answering the three most common questions first

Bioburden determination is not a release test. It is an input to sterilization validation. Equally, the sterility testing used within a sterilization validation is not there to prove that the batch is sterile; it is there to determine whether any microorganisms survive a given verification sterilization condition. The method basis for bioburden determination is ISO 11737-1, while the sterility testing used in sterilization validation has its own accompanying method document. The two are constantly treated as one thing, and once a report mixes them up there is very little that can be salvaged afterwards.

So when a manufacturer calls and says "I need a sterility test", the first thing we ask back is: is this report going to support a sterilization validation, or is it for product release? The two routes differ completely in sample quantity, protocol design and how the report conclusion is worded, and getting it wrong means a wasted round at review.

Third question: when is it mandatory? For any device placed on the market in a sterile state, whether the route is ethylene oxide (ISO 11135 / GB 18279), irradiation (ISO 11137 / GB 18280) or moist heat (ISO 17665), sterilization validation cannot avoid bioburden data. The only difference is how that data gets cited on each route, which the table below covers.

Who actually consumes the bioburden data

Many manufacturers treat bioburden as an isolated test item, file the certificate and move on. In reality it is the starting point of a whole chain of downstream reasoning, and if the data is not sound, every conclusion after it has to be redone. The table below answers one question only: once this data leaves your hands, who cites it in the sterilization validation file, and in what form?

Of the three routes, irradiation is the most sensitive to the bioburden count: the calculation logic for dose setting and dose audit is built on the count itself, so any bias in enumeration transfers straight into the dose conclusion, and once the dose has to be set high, material compatibility and packaging performance both need reassessing. The ethylene oxide route cares about something different - it is more concerned with whether the resistance of the bioburden is covered by the biological indicator, and the absolute count is not the whole story.

A boundary is worth stating here: how to choose between validation approaches on the ethylene oxide route, how to schedule validation and requalification, and how to design the process challenge device are decisions belonging to sterilization process validation itself and are not covered here. We have a dedicated article on EO sterilization validation that reads well alongside this one, and you can also look at the scope of our sterilization validation services. This article is only concerned with making the data that feeds those decisions solid. Method names, calculation rules and acceptance requirements are per the current valid version of the standard text.

The method itself: four steps, with the traps concentrated in the last two

Sampling: what to take, and when

The first frequent error is sampling the wrong state. Bioburden is meant to reflect the microbial load carried by the product as it enters the sterilization process, so the sample should be fully assembled, packed in its final packaging format, and not yet sterilised. Data from work-in-progress taken off the shop floor, or from already sterilised product because it was easier to get hold of, cannot be used.

The second is failing to define the sampling unit. For complex, bulky or expensive devices - assemblies with long catheters, large-area dressings, whole machines - the whole item is usually not tested. Instead a representative sample item portion (SIP) is taken. Once the SIP is fixed, the conversion relationship between it and the whole product has to be written into the protocol and supported by an engineering rationale: why does this portion represent the whole? Could contamination be concentrated in the portion you did not take? We have seen protocols in real projects that excluded the dirtiest fitting from the SIP - the data looked good and did not hold up.

When defining a SIP, there are three lines of enquiry for finding the heavily contaminated locations. First, locations handled frequently by hand: the points repeatedly gripped during assembly, leak checking, weighing and pouching. Second, locations that geometrically resist drying and tend to hold liquid: internal corners, snap-fit gaps, bonded and welded joints. Third, locations where the raw material itself carries higher microbial risk: naturally sourced materials, textile and nonwoven parts, components requiring wet processing. A SIP scheme that excludes all three will be challenged on representativeness at review, however rigorously the conversion is written.

The third is inadequate batch coverage. Single-batch data cannot describe the variability of the manufacturing process; you normally need to cover several independent production batches, and those batches should come from different material charges and different shifts. Splitting one batch into several portions does not count as several batches. The number of batches and the sample quantity are per the standard text and your own product risk assessment.

Elution and extraction: is the physical method compatible with the product

Transferring microorganisms from the product into liquid is the step most likely to introduce systematic bias in the whole determination. Common approaches include shaking, ultrasonication, vortexing, flushing and stirring, each with its own range of application:

• Rigid, simply constructed products that tolerate mechanical stress can take full physical intensity;
• Products with coatings, hydrogels or hydrophilic surface treatments may have the coating damaged by aggressive elution, and constituents leaching out of the coating can then inhibit growth, biasing the result artificially low;
• Lumened products cannot be cleaned by external shaking alone and need internal flushing to be considered;
• Highly absorbent dressing-type products soak up a large share of the eluent, and the losses at the recovery stage have to be accounted for.

Which approach to use is not a matter of instinct; it is supported by the recovery efficiency data generated in the next step.

Culture conditions: do not report a single "total bacterial count"

Bioburden is not the same thing as an aerobic count. Whether fungi also need to be examined, and whether anaerobes do, depends on the raw materials, the manufacturing environment and the process - naturally sourced materials, wet processing steps and intermediates held at room temperature for extended periods all raise fungal risk noticeably. When setting an irradiation dose, having only aerobic data for a product with obvious fungal risk is something technical review will pursue. Media types, incubation temperature and duration are per the current valid version of the standard text.

Recovery efficiency validation: the most commonly rejected element

This is where we send back the most protocols. A bioburden report that gives raw counts with no recovery efficiency validation is effectively no report at all.

The reason is straightforward: elution never transfers every organism off the product, so the measured count is inherently lower than the true value. The ISO 11737-1 approach is to estimate that transfer efficiency through a validation test and then use a correction factor to bring the measured value back close to the true level. Two methods are common:

• Repetitive recovery: elute the same item several times in succession, observe how the count declines from round to round, and derive the proportion recovered in the first round;
• Inoculated recovery: inoculate a known quantity of microorganisms onto the product, recover under the defined elution conditions, and compare the inoculated quantity with the recovered quantity.

Each has its own premises. With inoculated recovery in particular, note that the attachment state of the inoculum differs from natural contamination, so carrying the conclusion straight across needs an explanation. Whichever method is used, the protocol has to state how the correction factor is calculated and how it is applied to the final result.

There are three things to look at when judging whether a recovery validation was done properly. First, whether the conditions of the validation test match the formal determination: the same product and lot, the same disassembly and packaging state, the same eluent and mode of action. A validation run under mismatched conditions cannot be extrapolated to the formal determination. Second, whether the correction factor was derived on the whole product or on the SIP: mixing the two amplifies the bias again at the conversion step, and the protocol must state which level it applies to. Third, whether, when recovery efficiency comes out clearly low, the protocol goes back and changes the elution method rather than simply applying a large correction factor to pull the number up - the larger the correction factor, the more the final conclusion depends on the validation test itself, and the more likely it is to be questioned at review. Acceptable recovery levels, the number of replicates and the statistical treatment are per the current valid version of the standard text.

Products with antimicrobial activity: flag them before you ship

Products containing silver, quaternary ammonium compounds or chlorhexidine, carrying antimicrobial coatings, or formulated with alcohols or essential oils will sterilise the eluent and the medium along with everything else. The result reads zero, when in fact nothing was measured at all. These products need bacteriostasis screening first: demonstrating, through neutralisers, dilution, filtration or similar means, that the test system does not inhibit growth of the target organisms, and that the neutraliser itself is not toxic to them.

This work has to be completed before the formal determination; you cannot wait for a result that looks wrong and then add it. If a manufacturer ships samples without declaring the antimicrobial constituents in the formulation, the laboratory runs the standard protocol and the whole round of data is void. When you fill in the testing requirements, list the antimicrobial, preservative and disinfectant constituents in full. This is not a surrender of commercial confidentiality; it is what gives the method design a basis.

Where sterility testing sits within a sterilization validation

The sterility testing used in a sterilization validation serves the validation activity, not product release. Its typical uses include testing samples run under verification sterilization conditions on the ethylene oxide route, and verification testing at the dose audit stage on the irradiation route. The logic of the assessment is: under this deliberately reduced sterilization condition, do all samples still show no growth - which is then used to infer the lethality margin under normal conditions.

Three things must be in place for this kind of testing: method suitability validation (showing the test system does not inhibit microbial growth), negative and positive controls (if the positive control fails, a negative result means nothing), and reliable aseptic technique in sample handling (losses from false positives are more common than from false negatives, and a single contamination event can write off a whole dose audit). Sample numbers, incubation conditions and acceptance rules are per the current valid version of the standard text.

One distinction to keep clear: the sterility test used for product batch release has its own method basis and sampling logic, and a sterility test report generated for sterilization validation cannot be substituted for it. The purposes differ, and the wording of the report conclusion should differ accordingly.

Bioburden is not a one-off task

Another frequent misconception is that one report settles the matter permanently. All of the following should trigger re-evaluation:

• Change of raw material supplier, change of grade, or change in incoming cleanliness requirements;
• Relocation of the manufacturing site, change of cleanroom classification, or changes to line layout or personnel flow;
• Process changes, particularly the addition of wet processing steps or longer hold times for intermediates;
• Changes to product construction or packaging format;
• A rising trend in routine monitoring, or clear seasonal variation.

For changes you are unsure about, three self-check questions help. Has it changed the duration or manner of the product's exposure to the environment before sterilization? Has it introduced a new source of microorganisms - a new supplier, a new operation, a new personnel or material flow? Has it undermined the feasibility of the existing elution method, in that the construction, coating or packaging format has changed and the original method may no longer hold? If the answer to any one of them is yes, schedule a confirmatory determination rather than waiting for the next scheduled monitoring point.

The frequency of routine monitoring should be written into the quality system documentation and aligned with the periodic re-evaluation schedule of the sterilization process. How that re-evaluation itself is carried out, and what activities it comprises, is governed by the sterilization process validation documentation and is not resolved in the bioburden report. Irradiated products deserve particular attention here: whether a dose audit passes depends directly on the bioburden level in that period, and a gap in monitoring often only surfaces when an audit fails.

The relationship with packaging and the sterilization chain

The sterile barrier system (ISO 11607) is responsible for maintaining sterility after sterilization; it does not reduce bioburden before sterilization. But the two are coupled at project level: the packaging format determines whether you can hold the product in its final ready-to-sterilise state at sampling, and it also determines whether packaging material performance has to be reassessed if the irradiation dose ends up high. When planning a project, put bioburden, sterilization validation and packaging validation on one timeline rather than running each in isolation.

What to send with your samples

Sample quantity should be agreed with the laboratory during protocol review, because it depends simultaneously on product construction, how the SIP is defined and the intended sterilization route; there is no universal number. Several other things are worth arranging before shipping. Recovery efficiency validation and bacteriostasis screening normally consume additional samples, so hold back enough spares in one go - topping up later introduces a new batch difference. Samples must be protected from heat and moisture in transit and interim storage, and samples arriving with damaged packaging can only be scrapped and reshipped. The three items of antimicrobial constituents, intended sterilization method and intended use of the data should be filled in after R&D and regulatory affairs have confirmed them, rather than left to production to write from memory. And where one product has several sizes, state up front which size is intended to cover which, and write the representativeness reasoning into the protocol - far less work than supplementing it afterwards. For the full sample sequencing and document checklist, see the testing process.

Quick reference: common mistakes

• Running bioburden as a batch release item while never performing recovery efficiency validation;
• Measuring bioburden on already sterilised product, obtaining zero, and writing it into the sterilization validation;
• Testing aerobes only, on a product containing naturally sourced materials;
• Reporting "not detected" on an antimicrobial product without bacteriostasis screening;
• Defining the SIP so that it avoids the heavily contaminated locations;
• Continuing to use pre-change data to support a dose audit after a supplier change or a line relocation;
• Using a sterility test report generated for sterilization validation as a basis for release.

What we can do

SUNGO Lab is accredited by CNAS, CMA and IAS (USA), with laboratories in Shanghai and Hefei. To be clear, an accreditation mark only demonstrates the technical competence of the laboratory within its accredited scope; it is not a commitment as to market access in any target market. We can take on protocol design and recovery efficiency validation for bioburden determination, the sterility testing that accompanies sterilization validation, bacteriostasis screening, and the connected sterilization process validation, packaging and distribution validation work. We would rather be involved at the protocol stage, so that the sampling unit, batch coverage and intended use of the data are settled in one conversation instead of discovering after the fact that the data cannot be used. Call +86 132 4819 8029 with your product construction and intended sterilization route, or request a quote - we will work out who the data is going to first, and then decide how the protocol should be written.