What Must a Dental Laboratory Do When a Biological Indicator Fails?
A biological indicator fails in a dental laboratory and the instinct is to look at the sterilizer. That is the smallest part of the problem. The load that failed is still in the building, which makes it the one thing that is easy to control. The loads that ran before it are the difficulty, because a laboratory sterilizes its own burs, wheels, articulators and returned prostheses several times a day and dispatches cases between cycles. A failed indicator in a laboratory is therefore a recall exercise before it is an equipment fault, and the whole exercise turns on a number that either exists or does not: which case was in which load.
The rules that apply are written for health care settings generally and they fit a laboratory without modification. A load containing a positive biological indicator is, by definition, inadequately reprocessed. A written procedure for the recall and reprocessing of improperly reprocessed devices must exist. All devices in each processed load must be recorded so that tracking is possible. And the recall must include the failed load plus everything processed in that sterilizer since the last successful one. The consumables side of the routine is where the speed comes from: a 24-hour biological indicator with a lot number on the record, a Class 5 integrator in the package so a load can be judged before it is opened, and a Class 4 pouch read at the temperature its maker specifies, which is the difference between a result in 24 hours and a result next week.
What does a positive biological indicator actually tell a laboratory?
That the process did not achieve the conditions the test organism needed to be killed, in that load, on that cycle. It does not tell the laboratory which item failed, whether any patient was exposed, or what the cause was. It is a signal about a process, and the guidance treats it the way it treats an incorrect reprocessing method, a printout showing that the cycle missed its parameters, and a chemical indicator that did not change colour: all four are listed together as inadequate reprocessing, and all four trigger the same recall obligation.
That grouping matters for how a laboratory responds. A negative chemical indicator on one package, a printout that shows a short exposure time, and a positive spore test are not three different problems with three different procedures. They are one procedure, applied at three different sensitivities. A laboratory that has exercised the recall path on a failed chemical indicator knows how to run it on a failed biological indicator, because the scope question — what was in that load, and what else ran since the last clean one — is identical.
What is the true recall scope in a dental laboratory?
Wider than most laboratories assume, and it widens with time rather than staying fixed. Table 1 sets out the five layers, and the second one is the one people get wrong: the scope is not the failed load, it is the failed load plus everything sterilized in that unit since the last successfully sterilized load. In a laboratory running six cycles a day with a 24-hour read, a result read on Tuesday morning can reach back through Monday afternoon and every case dispatched in between.
| Layer of the recall | What falls inside it | Where the laboratory finds it |
|---|---|---|
| The failed load | Every device and every case that was in the sterilizer for that cycle, whether it went to the bench, to the spray booth, or into a dispatch box an hour earlier. | The load record, if contents were traced to case numbers when the cycle ran. Without that field the laboratory knows it ran a cycle and not what was in it. |
| Everything since the last successful cycle | The provincial checklist is explicit: where a recall follows a failed biological indicator, the recall shall include the devices in the failed load as well as all other devices processed in the sterilizer since the last successfully sterilized load. | The biological log, read backwards from the failed result to the last negative one, with the load numbers in between. |
| Any load containing implantable devices | Implants are quarantined until the biological result is in, which means they should still be inside the building. If the quarantine step was skipped, they are in the top two layers as well. | The load record, plus whatever passes for a quarantine shelf in the laboratory. |
| Cases already in a clinic | Dispatched devices that were sterilized in an affected load. These cannot be quarantined by the laboratory, so the clinic has to quarantine them. | The dispatch record: case number, clinic, date, carrier, and the load number the case was sterilized in. This is the layer a laboratory without a dispatch register cannot reconstruct. |
| The devices already in a patient’s mouth | A prosthesis that has been inserted cannot be recalled in the same sense. What happens instead is a risk assessment and, where indicated, notification — which is the clinic’s decision and the clinic’s conversation. | The clinic, from the case numbers the laboratory supplies. The laboratory’s job is to make that list complete and fast. |
Two consequences follow. The first is that the scope grows while the result is pending, which is why the quarantine rule exists: devices in a processed load should not be released until the biological result is available, and implants must be quarantined until it is. A laboratory that releases on the integrator and the printout — which the guidance permits where quarantine is not possible — is making a decision that it must be able to defend the morning a result comes back positive. The second is that the dispatch record carries as much weight as the load record, because the layers that leave the building are the ones the laboratory cannot fix by itself.

How does a laboratory find the cases that were in the affected loads?
Only one way: the load number has to be on the case sheet, and the case number has to be on the load record. Everything else is reconstruction. A laboratory that numbers its loads sequentially, writes the load number on the pouch label or the bag tag, and records the case numbers in that load has a recall list in twenty minutes. A laboratory that records “tray of burs, 14:30” has a recall that depends on the memory of whoever ran the cycle.
The practical fix costs nothing and takes a week to become habit. One column on the load sheet headed “case numbers in this load”. One field on the dispatch record for the load number. A rule that a case cannot go into the dispatch box until its case number is written on the load sheet, which is the same discipline that keeps a pouch from being sealed without an internal indicator. The companion guide to what an infection control auditor checks about dental instrument reprocessing sets out the full family of records this depends on; the biological indicator guide to storage, shelf life and lot tracking covers the other half, which is knowing which tests were run with which lot.
What should the laboratory tell the clinics?
Five facts per case, in writing, the same day: the case number, the date the case was sterilized, the fact that the cycle’s biological indicator was positive, what the clinic should do with the device, and who at the laboratory is handling the investigation. Nothing about cause, because the cause is not known yet, and nothing about risk to a patient, because the risk assessment is not the laboratory’s to make — the guidance puts the assessment of client and patient risk inside the recall procedure and puts the notification of physicians, clients and other facilities alongside it.
What the clinic does with the message is quarantine the device if it has not been used, and hold it until the laboratory says what happens next. If it has been used, the clinic owns the clinical judgement and, where indicated, the conversation with the patient. A laboratory that sends the five facts within the day has given the clinic what it needs to make that judgement; a laboratory that sends a paragraph of reassurance and no case numbers has given it nothing.
What happens to recalled prostheses and appliances?
Devices recalled because of a reprocessing failure are to be reprocessed before use, which is straightforward for a tray of burs and less so for a finished appliance. Three destinations, decided by material and by where the device is. An item that never left the building, or came back unopened, is reprocessed by the route its maker validates and returned to the queue with a new load record. An item that was already inserted cannot be reprocessed, and becomes a clinical question for the clinic. An item that was seated but not cemented, or delivered but not yet fitted, is usually remade rather than re-sterilized, because the material tolerates one cycle and not two.
That last point is a costing decision the laboratory should make in advance rather than in the middle of an event, because it changes the message to the clinic. A laboratory that knows it will remake rather than reprocess a heat-sensitive appliance can say so in the first notification, which saves a week of the clinic waiting to hear. The CliniEco guide to what to do when a biological indicator test fails covers the same decision path from the clinic side, and the provenance of the test itself — the unexposed control, the incubator, the read time — is where any event investigation starts.
Which checks come before the service technician is called?
Eight, and three of them are the technician’s first questions anyway. The control: an unexposed indicator from the same lot, incubated alongside the failed one, has to read positive, or the incubator or the lot is the problem. The temperature: a block that has drifted below the specified incubation temperature can produce a false positive reading by failing to support growth properly, or a false negative by overheating, and the incubator has its own verification record for exactly this reason. The lot and the date: an expired indicator or one stored in a warm room is a test failure rather than a process failure. The match: an indicator chosen for the cycle type, read at the temperature that indicator specifies, because a failure caused by the wrong indicator belongs to the test, not to the sterilizer.
Then the four that belong to the process. Was the load light enough and packed so that steam could reach the indicator, and was the indicator in the least accessible part of its challenge device rather than on top of the load? Was the packaging one the maker validates, with the internal indicator inside rather than taped to the outside? And was the printout actually checked at the end of the cycle, showing the time, temperature and pressure the cycle required, rather than filed unread? And was anything stacked on the chamber floor, where it would sit in the steam path? A laboratory that answers those eight questions before calling has either found the cause or has a service request the technician can act on.

| Window | What has to happen | Who does it | What is left behind |
|---|---|---|---|
| 0–10 minutes | Read the result again with the control in front of you: the unexposed control from the same lot, the incubation temperature, the read time, and the lot expiry. A failed result on an expired or mishandled indicator is a test problem before it is a sterilizer problem. | The person who read the result | The indicator itself, kept. Do not discard a positive indicator. |
| 10–20 minutes | Stop the sterilizer. Label it out of service, physically, and take the failed load out of circulation without unwrapping or re-running anything. | The bench lead | A written tag on the machine with the time and the load number. |
| 20–40 minutes | Identify the scope from the records: the failed load, then every load back to the last negative biological indicator, then the cases those loads contained. | Whoever owns the load records | A list of load numbers and case numbers, written as it is assembled. |
| 40–60 minutes | Notify the person named in the recall procedure. Open the event record with the time the result was read, not the time the investigation ends. | The designated person | The event record, with the notification names and times in it. |
| Same day | Tell the clinics whose cases are in scope, in writing, with the case numbers and what the laboratory needs them to do. Quarantine returned devices on arrival. | The designated person | Sent notifications, filed with the event record. |
| Before the unit runs again | Correct the cause, then requalify: three consecutive cycles with the appropriate biological and chemical challenges in an empty sterilizer, plus at least one cycle challenged with a full test load. A dynamic air removal unit also needs three consecutive air removal tests. | Service provider and the bench lead | The requalification results, attached to the event record and to the machine’s file. |
When can the sterilizer go back into service?
Not when the next cycle runs clean. The provincial guidance requires qualification before routine service with at least three consecutive cycles carrying the appropriate biological and chemical challenges in an empty sterilizer, plus at least one cycle challenged with a full test load, and three consecutive air removal tests where the unit is dynamic air removal type. A positive indicator during qualification means the unit is not approved for use. A laboratory that puts a failed unit back into rotation after one clean cycle has skipped the evidence that the repair worked.
Unexplained sterility failures are also on the list of events that require full re-qualification rather than a quick retest, alongside major repairs, construction or relocation, changes to the steam supply, and work on the emission control system. That is the honest reason to insist on a root cause: if the cause is a failed gasket, the requalification is a formality. If the cause is unknown, the laboratory is in the re-qualification category whether it likes it or not, and the record should say so.
Does one failed indicator mean the monitoring system is broken?
No, and treating it that way wastes a week. A single positive with an identifiable cause — an expired lot, an overloaded chamber, a unit that missed its parameters — is a control failure that was caught, which is what monitoring is for. What a positive result does prove is that the laboratory’s recall path either works or does not, and that is worth knowing before an inspection asks.
The pattern to watch is repetition: the same cycle type, the same position in the chamber, the same day of the week, the same operator. That is a process problem rather than a test problem and it belongs in a different conversation, which is why the event record should capture the load position, the operator and the cycle type even when the immediate cause seems obvious. The CliniEco guide to spore testing frequency in Canada versus the United States explains why the Canadian baseline is a daily test rather than a weekly one, and it is the reason a laboratory sees a failure in days rather than in months. If a result has to be turned around faster than 24 hours, a 1-hour rapid-readout indicator exists for that purpose, at a price that belongs in the same order review as the pouches.
What records does the event generate?
One event file with five contents, and it should be opened within the hour rather than at the end. The result itself, with the indicator kept and the control result attached. The scope list: load numbers and case numbers, written as they are identified with the time each was added. The notifications: who was told, when, and by what means. The investigation: the eight checks, what each returned, and what the service provider found. The closure: the corrective action, the requalification results, and the date the unit returned to service.
That file is also the honest answer to a question an inspector will ask months later, which is not “did you have a failure” but “what did you do about it”. A laboratory with a complete event file, three requalification cycles and a recall list can answer in one sentence. A laboratory that remembers the incident and has nothing on paper is starting an investigation at the moment it is being inspected. The six-step investigation protocol and its documentation duties are set out in positive biological indicator investigations, and the quarantine-and-release rules that apply while the investigation is open are in failed spore tests: quarantine, retest and release.
What should a laboratory change so the next failure is faster?
Four things, none of which needs a budget approval. Number the loads and write the number on the bag tag, so a case can be found from a load and a load from a case. Put one column on the dispatch record for the load number, so the outbound direction closes. Write the recall procedure with names in it — the guidance asks for the department and staff responsible to be designated, and a job title is not a designation when the person holding it is on holiday. And keep one notification template with the five facts in it, so the clinics hear on the day rather than when the investigation finishes.
A laboratory that has all four can run the whole exercise, from reading the result to notifying clinics, in a morning. A laboratory that has none of them will spend the morning reconstructing what happened and the week explaining why it could not. The same two-way handover that makes the routine work — described in how cases are packed and tracked between practices and laboratories and in the laboratory reprocessing routine — is what makes the failure survivable, because in both directions the thing being managed is a case number.
Primary sources. The definition of inadequate reprocessing, the recall scope, the notification duties, the quarantine rule and the requalification requirements above are quoted from the provincial reprocessing best practices and the provincial recall checklist that accompanies them; a laboratory in a province with its own college or ministry requirement should treat that requirement as the floor.
- Public Health Ontario — Steam Sterilization Monitoring and System Failures: Recalls — the list of situations that count as inadequate reprocessing including a positive biological indicator, the requirement that all devices in each load be recorded to enable recall tracking, the recall scope (failed load plus everything since the last successful load), the recall procedure contents, and the rule that recalled devices are reprocessed before use.
- Public Health Ontario — Best Practices for Cleaning, Disinfection and Sterilization in All Health Care Settings (May 2013) — the qualification and re-qualification requirements (three consecutive cycles with challenges plus a full test load, three air removal tests, a positive indicator meaning the unit is not approved for use), the events that require re-qualification including unexplained sterility failures, the release and quarantine rules, and the written recall procedure requirement [CSA Z314.0].
- Royal College of Dental Surgeons of Ontario — Standard of Practice: Infection Prevention and Control in the Dental Office — the Dental Laboratory Asepsis section (pp. 33–34), which is where the office-to-laboratory handover of impressions, prostheses and appliances is defined, and therefore which party a recalled case has to be traced through.
Where does a dental laboratory take this next?
Three of the tools here are usable rather than readable. The printable compliance log carries the fields a receiving check, a lot trace and a reprocessing record actually need, so the paperwork can start without inventing a format. The clinic supply planner turns a month of cycle volume into biological indicator and integrator quantities, so a daily test never waits for a delivery. The sterilization monitoring calculator works out how many indicators a year of cycles consumes, and the provincial self-check lists what an inspector asks for in each province. Multi-site or purchasing-group enquiries go through the wholesale account route. Reselling? become a distributor.
Questions that go past a template — a record format a particular inspector asked for, or a retention period your college is challenging — belong with the compliance consultation desk rather than a generic contact form.
If you would rather test the routine than read about it, the 24-hour biological indicator sample request is the simplest way to confirm that the incubator, the records and the staff habit all work together. The sterilization compliance hub collects the rest of the guides for this setting.
Related Reading
- Positive Biological Indicator Investigations: A Six-Step Protocol and the Documentation Duties
- Failed Spore Test: Quarantine, Retest and Release in Canada
- How Are Cases Packed and Tracked Between Canadian Dental Practices and Laboratories?
Frequently Asked Questions
How far back does a recall go after a positive biological indicator?
The provincial checklist sets the scope as the devices in the failed load plus all other devices processed in that sterilizer since the last successfully sterilized load. The biological log defines that interval: read backwards from the failed result to the last negative one, and every load in between is inside the recall. In a laboratory running several cycles a day, that can be a day and a half of work.
Does a laboratory have to recall prostheses that have already been sent to a clinic?
The laboratory has to identify them and tell the clinic, which then quarantines the device if it has not been used. The laboratory’s obligation is the trace: case number, clinic, dispatch date and the load the case was sterilized in. Devices that were already inserted cannot be recalled in the same sense; that becomes a risk assessment and, where indicated, notification, which the clinic owns.
Can a laboratory release a load before the 24-hour biological indicator result is back?
The guidance says items in a processed load should not be released until the result is available, and allows one alternative: where quarantine is not possible, evaluation of a Class 5 or 6 chemical indicator together with the cycle’s physical parameters may justify releasing routine loads. Implantable devices must be quarantined in every case. A laboratory relying on that alternative is accepting that a positive result will mean recalling devices it has already dispatched.
What is the first thing to check when a biological indicator turns positive?
The control. An unexposed indicator from the same lot, incubated with the failed one, has to read positive; if it does not, the incubator, the read time or the lot is the problem rather than the sterilizer. After that: the incubation temperature against the specified range, the expiry date and storage conditions, whether the indicator matched the cycle type, whether the load allowed steam access, and whether the printout actually showed the required parameters.
How many clean cycles before the sterilizer can go back into use?
Three consecutive cycles carrying the appropriate biological and chemical challenges in an empty sterilizer, plus at least one cycle challenged with a full test load, and three consecutive air removal tests on a dynamic air removal unit. A positive indicator during qualification means the unit is not approved for use. One clean cycle is not requalification, and a laboratory that restarts early has no evidence the repair addressed the cause.
Is a single positive biological indicator a reportable event?
It is an event the laboratory has to document and resolve internally, and the recall obligation is triggered by the result itself rather than by an external reporting threshold. The provincial guidance requires a written recall procedure with designated responsibility, identification of the devices affected, a risk assessment, notification where indicated, quarantine pending investigation, and the involvement of the facility’s risk manager where one exists. Whether an external body has to be told depends on the jurisdiction and on what the investigation finds.
What happens to a returned prosthesis after a recall?
Devices recalled because of a reprocessing failure are to be reprocessed before use, so an unopened or returned item follows its maker’s validated route and gets a new load record. Where the material tolerates only one cycle — common with acrylics and printed resins — the device is remade instead. Deciding which of those two applies before an event happens is what lets the laboratory tell the clinic the answer in the first notification.
How should the laboratory communicate with clinics during the investigation?
In writing, the same day, with five facts per case: case number, date sterilized, the fact that the cycle’s biological indicator was positive, what the clinic should do with the device, and who at the laboratory owns the investigation. Cause and patient risk do not belong in that first message, because the cause is not yet known and the risk assessment is the clinic’s to make.
CliniEco Medical holds Health Canada MDEL #35334. The recall scope, notification duties and requalification requirements described here follow the provincial reprocessing guidance and recall checklist that set them; where a province, college or accrediting body sets a higher requirement, that requirement governs. This page is general information about reprocessing equipment and consumables, not clinical or legal advice. Follow the instrument and sterilizer manufacturers' validated instructions; where this page and a maker's instructions differ, the maker's instructions govern. Health Canada MDEL #35334 covers our establishment licence; it is not a statement about any individual product.