Verifying Instrument Cleaning: Protein Residue Tests, TOSI and Cavitation Methods Compared

Quick facts

• Inspection and verification are different steps. Inspection looks at the instrument; verification measures something about it, or about the process that cleaned it.
• Four method families get confused with each other: protein residue tests (swab, strip or elution), haemoglobin-specific tests, process soil monitors such as TOSI, and ultrasonic energy or cavitation monitors.
• Each family proves one thing and is silent about the rest. A cavitation check proves the machine delivered energy; it says nothing about whether the instruments in the basket came out clean.
• Sensitivity figures are method-specific and rarely comparable. Detection limits in published and vendor documentation range from tenths of a microgram of blood protein on a swab to microgram-per-millilitre ranges in colourimetric assays.
• Peer-reviewed audits of "ready-for-use" instruments have found measurable residual protein on a large share of them. Clean-looking is not the same as measured clean.

A cleaning step that looks finished is not evidence that it is finished. Verification answers a narrower and more useful question than inspection does: what did we measure, with what method, against what threshold, and what did we do when it did not pass. This guide compares the main method families used in dental, clinic and sterile-processing settings, sets out what each one proves and what it does not, and lists the sensitivities that published and vendor documentation actually claim.

What does "cleaning verification" mean, and what is it not?

In the Canadian guidance that most facilities work to, inspection is an explicit step in the reprocessing sequence and not a substitute for measurement. Ontario's provincial reprocessing guidance lists inspection alongside disassembly, cleaning, disinfection or sterilization, rinsing and drying as a step that must be based on the manufacturer's recommendations and established guidelines, and it describes the inspection itself as "visually inspecting the equipment/device for discolouration or soil; if present, the item is removed from service and reprocessed". The same document requires that newly purchased non-sterile critical and semicritical devices "first be inspected and reprocessed according to their intended use" (Public Health Ontario, reprocessing guidance for all health care settings, 2013).

That gives a facility two different obligations, and they are often run together in practice:

1. Inspection — an eye-and-hand check of every item, with a defined consequence if the item is not clean.
2. Verification — a measurement that produces a result you can record, trend and act on, usually on a sample rather than on every item.

Verification is not the same as sterility assurance either. A protein residue result describes the state of an instrument after cleaning, before sterilization. It is not a substitute for the chemical and biological monitoring that covers the sterilization step itself.

What does ISO 15883-5 actually specify?

The standard that defines how cleaning efficacy is demonstrated for washer-disinfectors is ISO 15883-5:2021, Washer-disinfectors — Part 5: Performance requirements and test method criteria for demonstrating cleaning efficacy (ISO catalogue entry). Its introduction is unusually candid about the problem this article is about: "The cleaning efficacy of washer-disinfectors has historically been demonstrated by referring to different test soils and methods that have been used in several different countries." The standard exists to give those different national approaches a common, standardised set of methods.

Its scope covers "procedures and test methods used to demonstrate the cleaning efficacy of washer-disinfectors (WD) and their accessories intended to be used for cleaning of reusable medical devices" (ISO 15883-5:2021). The first edition replaced the 2005 technical specification, harmonised its terms with ISO 11139:2018, moved the examples of protein detection methods into an informative annex, added examples of haemoglobin detection methods, and placed the immersion test protocol for a protein-based test soil in a normative annex together with worksheets (ISO 15883-5:2021 preview, foreword).

The structure is worth knowing because it explains which numbers are comparable:

ISO 15883-5 element	Status	What it contains
Clause 4.4, cleaning efficacy test criteria	Normative	Visual examination, assay criteria and process residual criteria
Clause 5.1, cleaning test method validation	Normative	Load soiling method, detection method, and the validation of the pair
Annex B, protein-based test soil performance assessment	Normative	The immersion test protocol and the assessment of a protein-based test soil
Annex C, examples of test methods for residual proteinaceous contamination	Informative	Example protein detection methods, transferred from ISO 15883-1:2006
Annex D, examples of test methods for haemoglobin detection	Informative	Example haemoglobin detection methods
Annex E, test soil performance assessment results sheets	Informative	Sample worksheets for laboratories running the test

The distinction matters. Clause 4.4 and Annex B carry the test criteria; the detection chemistry is largely in informative annexes, which is exactly why two facilities running "protein residue testing" can be running different chemistries with different limits.

Table 1 — what each verification method proves, and what it does not

Method family	Example implementation	What it proves	What it does not prove
Visual inspection	Inspection step in provincial reprocessing guidance	That an item shows no visible discolouration or soil, and that visible failures were removed from service	That residual protein, biofilm or process residues are absent; it is a screening step, not a measurement
Protein residue test, swab or strip	Colorimetric swab and strip systems; laboratory colorimetric assays	That protein above the method's detection limit is, or is not, present on the sampled area	That the whole instrument is clean; results depend on where and how the sample was taken and how much protein the swab recovered
Blood-protein specific test	Peroxidase-based swab test for insoluble blood protein	That blood-derived protein residue is present on the sampled surface, down to the method's stated detection limit	That non-blood organic soil, detergent residue or biofilm is absent
Process soil monitor	Prepared test object with simulated blood soil, including a cannulated version	That the cleaning process could (or could not) remove a defined, reproducible soil under the conditions of that run	That every instrument in the load is clean; it is a process challenge, not an instrument-by-instrument result
Ultrasonic energy or cavitation monitor	Indicator that reacts to cavitation energy	That the bath delivered sufficient cavitation energy within the cycle time, and that transducers are performing	That instruments were cleaned; energy present is not the same as soil removed
Cavitation distribution check	Aluminium foil test, or a calibrated cavitation meter in the tank	That cavitation is occurring and how it is distributed across the tank	Instrument cleanliness, or chemical cleaning efficacy, or rinse quality
Residual water check	Swab test for residual water in endoscope channels	That a channel is dry to the method's detection limit	That the channel is clear of residual protein or biofilm
Washer-disinfector performance testing	ISO 15883-5 test soils and criteria	That a machine and a cleaning cycle meet defined cleaning-efficacy criteria under the standard's test conditions	Routine cleanliness of a specific working day's instruments, unless the same method is applied routinely

Two of these deserve a closer note because they are the most misread.

Process soil monitors are deliberately standardised. One widely used test object is described by its distributor as a stainless steel coupon with simulated blood soil in a clear holder, supplied to challenge the cleaning efficacy of mechanical cleaning equipment and proteolytic detergents (Healthmark TOSI product listing). The European distributor documentation describes the read-out as more than pass or fail: a clean result indicates good cleaning efficacy, a fibrin residue points to low chemical efficacy, and haemoglobin residue points to low mechanical efficacy (PEREG TOSI product documentation). A cannulated version applies the same idea inside a simulated hollow instrument, for channels cleaned in a washer-disinfector or an ultrasonic bath (PEREG TOSI LumCheck). An independent validation report describes the design goal as correlation of the test soil with human blood and visibility of the result (validation report, test object surgical instruments). What none of that establishes is that a specific patient's instrument came out clean: the monitor challenges the process.

Cavitation monitors have the same shape of limitation. A pass on an ultrasonic energy monitor means the bath produced enough cavitation for the reaction to change colour within the cycle time; the vendor documentation notes that insufficient energy, overloading, water level and degassing all lengthen the time to a colour change, and that a major problem produces no change at all (PEREG SonoCheck documentation). That is a machine performance result. A separate, widely used field check — suspending aluminium foil in the bath and reading the perforation pattern — is described as a way to check cavitation efficiency and uniformity (field procedure for testing an ultrasonic bath); it is qualitative, operator-dependent, and says nothing about chemistry, rinse quality or the instruments themselves.

How sensitive are these methods, and where does the detection limit bite?

Detection limits are where comparisons usually go wrong, because they are quoted in different units, for different sample types, and by different kinds of source. The figures below are reproduced as they are documented, with the kind of source identified so you can weigh them.

Method	Documented sensitivity or published reading	Unit basis	Source type
Laboratory residual-protein analysis (acid stripping, hydrolysis, total amino acid analysis)	Median residual protein per instrument across five tray groups: 267, 260, 163, 456 and 756	µg per instrument	Peer-reviewed audit of 120 ready-for-use instruments from five hospital sterile service departments in England and Wales
Rapid colorimetric protein assay on a moving-drop platform	Linear range 5–20	µg/mL	Peer-reviewed method-development study
Peroxidase-based blood-protein swab test	Down to 0.1	µg of blood-protein residue	Vendor technical documentation
Prepared test soil monitor	Visual result; no numeric detection limit published as a pass threshold	Not applicable	Vendor and validation documentation
Ultrasonic energy monitor	Colour change within the recommended cycle time; time to change lengthens with inadequate energy	Not applicable	Vendor technical documentation
Residual water swab	0.05	µl of residual water in a channel	Vendor technical documentation

Sources: J Hosp Infect 2006, quantitative analysis of residual protein on reprocessed surgical instruments; J Hosp Infect 2024, smart-detection approach for protein residues; PEREG HemoCheck-S; PEREG HydroCheck-E.

The laboratory audit is the figure to keep in mind when someone argues that cleaning failures are hypothetical: on instruments returned as ready for use, median residual protein per instrument ranged from 163 µg to 756 µg depending on the tray group, and the highest levels were found on instruments used for tonsillectomy and adenoid surgery (J Hosp Infect 2006). That study used a laboratory reference method, not a rapid swab, which is precisely the point: rapid methods are chosen because they can be run routinely, not because they are the most sensitive available.

Where do false negatives come from?

A false negative is a cleaning failure the method did not catch. Six sources account for most of them.

1. Sampling recovery. A swab recovers only what it touches and releases. Areas hidden from the swab — crevices, joints, box locks, lumen walls — are under-sampled, and recovery efficiency varies with the chemistry being tested.
2. Dried and fixed soil. Protein that has dried or been fixed on a surface behaves differently from fresh soil. Peer-reviewed work on residual protein assessment notes that interpretation of indirect elution-based measurements depends on factors beyond analytical sensitivity (J Hosp Infect, residual protein assessment in medical device processing).
3. Non-protein soil. Protein is used as a surrogate marker for organic contamination. Lipids, carbohydrates, detergent residue and mineral scale are not protein, and a protein test is silent about them.
4. Process residues. Residual detergent can interfere with the disinfection or sterilization step downstream (Public Health Ontario reprocessing guidance, 2013). Rinse water quality is part of the same chain: the same guidance lists limiting values for final rinse water including pH 6.5 to 8, evaporation residue not more than 15 mg/L, conductivity not more than 50 µS/cm and hardness not more than 0.1 mmol/L.
5. Energy present, cleaning absent. A cavitation monitor can pass while the chemistry is exhausted, the load is overloaded, or the instruments are stacked so that surfaces are shadowed.
6. Recording the wrong thing. "Cleaning checked" is not a result. A record that does not name the method, the sampling site and the threshold cannot be trended, and cannot show a drift before it becomes a failure.

Studies that run several methods side by side are the ones that show this. A 2026 before-and-after study in a central sterile supply department used visual inspection, ATP testing and protein residue testing together as its routine audit methods across more than 54,000 inspected instruments (Risk Management and Healthcare Policy, 2026). A 2025 systematic review and meta-analysis of laparoscope decontamination defined its primary outcome using "predetermined thresholds for each detection method (visual cleanliness, protein…)", which is another way of saying that the threshold comes from the method, not from a universal standard (Frontiers in Cellular and Infection Microbiology, 2025).

Is any one method required in Canada?

No. This is the single most important correction to make when a supplier pitch or a staff-room conversation implies otherwise.

• Provincial guidance sets the reprocessing steps, including inspection, and requires records; it does not mandate a specific protein test kit or cavitation monitor (Public Health Ontario reprocessing guidance, 2013).
• The CSA Z314 series is the Canadian standards family for reprocessing of reusable medical devices. Its clause text sits behind a paywall, so it is cited here by number only, without a version year.
• ISO 15883-5 sets out test soils and criteria for demonstrating the cleaning efficacy of washer-disinfectors, aimed at equipment performance and validation work (ISO 15883-5:2021).
• Vendor method documentation describes what a specific product does and what its detection limit is. It is product information, not regulation.

The practical consequence is that a facility chooses its verification programme. That freedom cuts both ways: nothing stops a clinic from running no measurement at all, and nothing stops a clinic from running a defensible programme built from one routine method plus an annual process challenge.

What should a verification record contain?

If the programme is going to be worth the time it takes, the record has to be able to answer an inspector's questions without the operator who ran it in the room. A minimum set of fields:

• date and time, and which cleaning equipment and cycle was used;
• the load or the specific instruments sampled, identified as precisely as the workflow allows;
• the method used, including the product or the standard it follows, and the sampling site;
• the result as read, and the threshold the facility applies to that method;
• the action taken if the result did not pass, and the re-run or rework outcome;
• who performed the test and who reviewed the result.

The same detail is why a record sheet matters more than a memory. A printable blank form for cleaning and sterilization records, with defined fields and no account required, is available from the compliance log centre, and the fields above map onto it directly.

Two pieces of equipment carry most of the routine cleaning load in a dental or clinic setting. The 6 L benchtop ultrasonic cleaner is sized so that a full instrument cassette fits the tank, and the multi-enzyme ultrasonic cleaning concentrate is the chemistry that sits in it; both pages state their measured dimensions, capacities and dilution rather than an assumed value. A facility building a monitoring programme alongside the cleaning step can review the sterilization monitoring collection, which lists the indicator formats stocked for Canadian clinics, and evaluate a new routine with a free five-pack sample of biological indicators.

Have a question about your own facility? Send it in and you will get a written answer specific to your setup, with the regulation or standard it is based on cited. Ask your compliance question.

Buying for a facility rather than a single practice? Request a B2B quote and we will price your volume with Canadian warehouse delivery. Reselling? become a distributor.

Related reading

• Protein residue testing after ultrasonic cleaning: what counts as clean for a dental clinic
• Cleaning verification and sterility assurance: what each monitor proves
• Ultrasonic cleaner comparison for Canadian clinics: 8 models, 10 parameters

Frequently Asked Questions

Is visual inspection a valid way to verify cleaning?

Inspection is a required step and a useful screen: provincial guidance describes visually inspecting each device for discolouration or soil and removing anything that fails from service. It is not a measurement, and it cannot detect residual protein, biofilm or process residues below the eye's limit. Treat it as the last line of the cleaning workflow and as a separate line in the record, not as verification.

What does a TOSI test actually prove?

It proves that the cleaning process could remove a defined, reproducible simulated-blood soil from a prepared test object under the conditions of that run, and it can indicate whether a failure points to chemistry or to mechanical action. It does not prove that any individual instrument in the load is clean, because the monitor challenges the process rather than the instrument.

Does a passed aluminium foil test mean my ultrasonic cleaner is cleaning instruments properly?

No. The foil test is a qualitative check that cavitation is occurring and how evenly it is distributed across the tank. It says nothing about the cleaning chemistry, the rinse step, the loading pattern or the instruments themselves. It is a machine check, not a cleaning result.

Can ATP testing replace protein residue testing?

They measure different things. ATP reflects total biological material and can degrade during cleaning, so it is used mainly as a rapid hygiene indicator; protein testing targets a specific class of soil and is used where a defined residue question is being asked. Studies that audit cleaning routinely run more than one method rather than choosing between them.

Why do different sources quote such different detection limits?

Because they are quoting different methods in different units: tenths of a microgram of blood protein on a swab, a microgram-per-millilitre linear range in a colourimetric assay, or a microgram-per-instrument reading from a laboratory reference analysis. A detection limit is only meaningful when the sample type, the chemistry and the unit are stated together.

Is any protein residue test or cavitation check required by law in Canada?

No. Provincial guidance sets the reprocessing steps and the record-keeping, and it does not name a specific verification product or method. CSA Z314 is the Canadian standards family for reprocessing, and ISO 15883-5 sets out test methods and criteria for demonstrating cleaning efficacy of washer-disinfectors. A facility chooses its programme from those options.

How often should a clinic run cleaning verification?

There is no single Canadian figure, because the requirement is set by the facility's own programme and by what the equipment manufacturer specifies for validation and performance testing. What is defensible is a written programme that states the method, the sampling plan, the threshold and the action on failure, and that is applied consistently enough to show a trend.

What is the difference between cleaning verification and sterility assurance monitoring?

Cleaning verification measures the state of an instrument after cleaning and before sterilization, using methods such as protein residue testing or a process soil monitor. Sterility assurance monitoring covers the sterilization step itself through physical, chemical and biological indicators. Both records are needed, and neither substitutes for the other.

CliniEco Medical distributes instrument cleaning and sterilization monitoring supplies in Canada under MDEL #35334. Method names and product names in the comparison tables are listed for identification only; inclusion is not an endorsement of one method or supplier over another.