Dental Sterilization Monitoring: Spore Testing and the Logs Inspectors Check

Sterilization monitoring is the single most commonly inspected item in a dental practice, and the one where practices most often discover their records do not say what they thought they said.

The requirement itself is simple. The documentation is where it goes wrong.

Three kinds of monitoring, and only one of them proves sterilization

Mechanical monitoring is the cycle readout — time, temperature, pressure. It tells you the machine reported correct parameters. It does not tell you anything was actually sterilized.

Chemical indicators are the strips and tape that change color. They confirm the package was exposed to the conditions of a sterilization cycle. They do not confirm that microorganisms were killed.

Biological indicators — spore tests — are the only method that directly verifies sterilization. They contain highly resistant bacterial spores, typically Geobacillus or Bacillus species. If the cycle killed those, it killed everything else.

All three have a role. Only the third one answers the question an inspector is actually asking.

How often, and with what

The CDC recommends spore testing sterilizers at least weekly. Some state dental boards write that frequency into rule; others treat it as a professional standard. Either way, weekly is the practical answer, and a practice testing monthly will struggle to explain the gap.

The test must be run with a matching control — a biological indicator from the same lot number that is not processed through the sterilizer. The control proves the spores were viable to begin with. Without it, a negative result is ambiguous: it could mean the cycle worked, or it could mean the indicator was dead on arrival.

A surprising number of practices run the test correctly and never run the control, which quietly undermines every result in the log.

When a spore test fails

A positive biological indicator means the cycle did not sterilize. Take the sterilizer out of service immediately.

Before returning it to service, the biological indicator should return negative results across three consecutive empty-chamber cycles. That is the standard remediation sequence, and it is the part practices most often shortcut — running one confirmatory test, getting a negative, and resuming.

You also need to consider the instruments processed since the last negative test. Depending on your protocol and the nature of the failure, those may need recall and reprocessing. Document the decision and the reasoning either way. An inspector reviewing a failed test will want to see that someone thought about the affected loads, not just the machine.

What the log needs to contain

A defensible sterilization log records, for every test:

• Date of the test
• Sterilizer identification, if you run more than one
• Lot number of the biological indicator
• Result of the test
• Result of the matching control
• Who read and recorded the result
• For failures: the corrective action taken, the three confirmatory cycles, and the disposition of affected instruments

Retain the records long enough to satisfy your state board and any local rules — requirements vary, and some states are considerably longer than practices assume.

The three failure modes

Gaps in the log. Weeks with no entry. Usually vacation, staff turnover, or a busy stretch. There is no way to reconstruct a test that was not run, which makes this the most expensive of the three to discover late.

Results recorded but never read. Mail-in services return results days later. If nobody closes the loop, a positive result can sit unnoticed while the sterilizer stays in service. Assign the responsibility to a named person, not to the practice generally.

No control documented. As above — the result is uninterpretable without it, and an inspector who knows the standard will notice.

Why this gets inspected first

Sterilization monitoring is attractive to inspectors because it is binary and it is dated. Either there is a test for that week or there is not. Unlike a policy, which can be written the night before, a log is a record of continuous behavior — and it cannot be produced retroactively.

That is also why it is worth automating the reminder rather than relying on someone remembering. The test takes minutes. Remembering to do it every week for three years is the hard part.