Microbiological sample incubation: conditions, timing and excursions

Microbiological incubation is a stage of the analytical method. Interpreting a result requires knowledge of the conditions actually experienced by the sample, rather than just the selected temperature or elapsed calendar time.

A measured excursion and a period without records raise different questions: the former involves characterising a deviation; the latter leaves part of the history uncertain. This article addresses the microbiological test. Incubator qualification and full mapping remain separate activities that supply some of the necessary evidence.

1. Start with the method conditions

Identify the test, matrix, medium, stage and applicable method version. Include the required conditions, sequences and supporting source in the procedure. Do not group different samples under a single setting simply because they share equipment.

Medium status and method suitability are separate prerequisites: see culture media performance and growth promotion testing. A satisfactory medium control does not authorise changing analytical conditions. Where a source allows alternatives, justify and implement the selection before use, without turning an occasional deviation into a new method.

2. Make timing and transfers verifiable

Define the events that start and end each stage and how transitions are recorded. Distinguish incubation time, holding time before entry and transfer time: they are not automatically interchangeable. A sequence requires a link between departure from the first environment and entry into the next.

The record should identify the sample, equipment, location, event, time and operator. Synchronise relevant time sources or document known differences. A later correction must preserve the original information and reason for the change; an operator’s memory does not replace a contemporaneous event record. The FDA’s 2018 data integrity guidance provides a framework for reliable, complete records.

3. Connect load and position with qualification evidence

Use zones and configurations supported by qualification. Consider container arrangement, load, airflow and access during operation. Having space on a shelf does not demonstrate that every possible arrangement is covered by the checks performed.

WHO good practices, 2011, sections 4.3.3–4.3.4 connect control of temperature-regulated equipment with its performance. For an individual test, record the relevant configuration and check that it falls within the authorised scope. If the load changes, first assess whether the evidence remains representative. An air probe does not automatically measure the temperature of every sample.

4. Retain the data needed to reconstruct exposure

Define what is measured, where, with which system, how continuously and who reviews it. Link sensor identity, metrological status, original data and the samples present. The setpoint describes a command; the record describes what the measurement system detected, within its limitations.

Assign responsibilities for alarms, faults, transfers and missing records, including outside normal working hours. Acknowledging an alarm does not demonstrate that the problem is resolved. Verify restoration and data availability while retaining abnormal events. Within its accreditation context, Cofrac LAB GTA 19, revision 02, §7.3.1 addresses the assessment of excursion magnitude, duration and missing evidence of temperature control.

5. Reconstruct excursions and interruptions

Begin the assessment by identifying the affected interval and the samples actually present. Reconstruct the available profile, duration, magnitude, analytical stage, access events and interventions. Distinguish equipment shutdown, temperature deviation and monitoring loss: they may overlap but are not the same event.

The matrix structures the investigation; it does not set acceptance criteria. Do not automatically compensate by adding time: the biological response is not a counter that can always be stopped and restarted without consequences.

6. Separate certainty about history from the result decision

Simulated case. Two sample series experience an abnormal event. For A, a short interruption is fully covered by continuous data, consistent timestamps and a known location. For B, a longer gap interrupts the record; only compliant readings before and after remain. A’s observed exposure can be bounded. For B, those two readings do not establish all intervening conditions.

Neither series is automatically valid or invalid. The responsible person assesses method requirements, sensor representativeness, the affected stage, controls and relevant recovery evidence, involving QA according to assigned responsibilities. Independent data can reduce uncertainty only when linked to time, location and samples. A reassuring final result does not erase a gap.

A 2026 primary environmental monitoring study describes differing recovery responses as conditions change. This illustrates biological specificity; it does not validate local excursions. Document the conclusion, limitations and consequences for results already used. Any further testing needs justification and does not eliminate earlier records.

7. Prevent recurrence through targeted controls

Review abnormalities by equipment, load, operational stage and event type. If they recur during transfers, investigate organisation and records; if they follow load changes, review representativeness and use practices. Distinguishing causes avoids maintenance or training that does not address the problem.

Before closing an action, verify that the new control works in actual operations. Keep current procedures, out-of-hours responsibilities and change criteria available. Both technical fault closure and the analytical impact decision must remain traceable.

Sources and access limitations

• WHO TRS 961, Annex 2, 2011; FDA Data Integrity, 2018; Cofrac LAB GTA 19, revision 02, applicable from 16 March 2020: public texts linked above, with differing scopes.
• USP 1117, 2022 record: public scope preview; consult the authorised current text for applicable details.
• Bailac and colleagues, 2026, PMID 42055940: indexed public abstract; controlled environmental monitoring study, not an excursion study. No universal incubation time or temperature is derived from these sources.