Quality Control Specialist | Industry Experience
✓ Reviewed by: Pankaj Sharma - Quality Control Specialist
Reviewed for Quality Control accuracy, laboratory practices, analytical methods, and technical relevance
📅 Last Updated: October 30, 2026
Media fill incubation conditions and inspection decide whether an aseptic process simulation (APS) can detect contamination. EU GMP Annex 1 (2022) fixes no temperature and no number of days. It requires incubation conditions and duration that are scientifically justified and validated. PIC/S PI 007-6 describes a generally accepted schedule of 20–25 °C for at least 7 days, then 30–35 °C, with 14 days in total. FDA’s 2004 guidance asks for at least 14 days within 20–35 °C. Trained, qualified staff then inspect every integral unit, and any contaminated unit fails the APS. Your approved protocol governs the actual execution.
What Are Media Fill Incubation Conditions and Inspection?
They are the last two stages of an aseptic process simulation (APS), the test in which sterile growth medium replaces product on the filling line. Annex 1 uses the term APS, and media fill is a synonym (PI 007-6 definitions). Once filling ends, you hold the units at set temperatures for a set time, then examine each one for visible microbial growth.
For the full overview, see What is Media fill in Pharmaceutical. This article covers only what happens after the filling stops.
Four documents shape the topic. EU GMP Annex 1 was published on 25 August 2022 and applied from 25 August 2023, except point 8.123 on lyophilizer loading, which applied from 25 August 2024. WHO Technical Report Series 1044, Annex 2 (2022) is a revision made with the EU and PIC/S. PIC/S PI 007-6 (1 January 2011) says its advice is “not mandatory for industry”. FDA’s 2004 guidance contains nonbinding recommendations. Annex 1 is the standard an EU inspector measures you against; the other two are reference points you can adopt if you justify them.
Why Do You Incubate and Inspect APS Units?
The growth medium stands in for the product. Any microorganism that entered during the simulated process will multiply in it and turn it cloudy, given time and a suitable temperature. Incubation supplies the time, and inspection is how a trained person sees the result.
If either step is weak, a real contamination route can go unseen and the APS passes by accident. Annex 1 9.32 reminds us that an APS is a periodic verification of the controls already in place. Process design, the pharmaceutical quality system (PQS), training and monitoring data carry the main weight.
How Do You Prepare and Identify Units Before Incubation?
Handling before the incubator sets the quality of the whole read. Follow the order and labelling in your protocol, and cover these points.
- Label and trace. Mark each tray or container so every unit links back to the run, line and filling time. PI 007-6 5.1.8 suggests numbering vials or keeping them in chronological order, which helps you find the source if one turns positive.
- Agitate, swirl or invert the filled units so the medium touches every interior surface, including the closure (Annex 1 9.41).
- Incubate all integral units, including those with cosmetic defects or those that passed non-destructive in-process checks (Annex 1 9.41). FDA’s 2004 guidance rejects units that lack integrity and asks that any erroneously rejected unit return promptly for incubation.
- Handle discards with care. Annex 1 9.41 accepts non-incubated discards only when they match routine practice and the production SOPs name the same circumstances, line location and number of units. Such units are typically incubated separately and not necessarily counted in the acceptance criteria.
- Use clear containers. If the product container is amber or opaque, substitute a clear one of identical configuration. If that is impossible, develop and validate a detection method (Annex 1 9.43).
- Incubate without unnecessary delay (Annex 1 9.44).
Fill volume matters as well. Annex 1 9.36 (iv) asks for a volume that wets every surface that could contaminate the product and leaves enough headspace to support growth and show turbidity. PI 007-6 5.3.2 adds that containers should not be completely filled, so obligate aerobes get oxygen, and should not be overlaid with inert gas. Annex 1 9.33 (iii) also replaces inert gas with air unless an anaerobic simulation is intended.
What Incubation Temperature and Duration Do the Guidelines Describe?
Annex 1 9.44 gives no fixed figures. It requires conditions and duration that are scientifically justified and validated to give an appropriate sensitivity of detection. The numbers people quote come from other documents, and their status differs.
| Source | Status | Temperature | Duration |
|---|---|---|---|
| EU GMP Annex 1 (2022), 9.44 | EU GMP expectation | Justified and validated by the site | Justified and validated by the site |
| PIC/S PI 007-6 (2011), 5.3.1 | Non-mandatory recommendation | 20–25 °C, then 30–35 °C | At least 7 days at the first stage; 14 days in total |
| FDA guidance (2004), media fill section | Nonbinding recommendation | Never outside 20–35 °C; within ±2.5 °C of the target | Not less than 14 days; at least 7 days at each temperature if two are used, lower first |
PI 007-6 adds that other schedules should rest on supporting validation data. So the 14-day, two-temperature scheme is a widely used reference, not an Annex 1 rule. In my experience, teams often copy it into a protocol without the validation data behind it.
Several items are site-defined, because neither Annex 1 nor PI 007-6 sets them. These are incubator mapping, temperature tolerance, shelf loading and stack height, lighting level, background and any intermediate reads. Your approved protocol and incubator qualification define them. FDA’s ±2.5 °C is a recommendation, and a site may set a tighter tolerance.
Annex 1 does not state whether trays sit upright or inverted during incubation. The protocol decides, provided the medium has already wetted the closure. If the temperature leaves its range, raise a deviation under the site procedure, because the documents do not describe how to assess an excursion.
How Do You Inspect Incubated Media Fill Units?
Annex 1 9.45 (i) says trained and qualified personnel inspect the units, under conditions that help them identify contamination. The site fills in the rest.
- Reader training. PI 007-6 8.8 asks for specially trained evaluators with routine eyesight tests, and training that mixes contaminated units among sterile ones.
- Reference unit. Compare each unit with a known sterile container. Some growth shows only as a faint haze that is hard to see alone (PI 007-6 5.4.1). Faint haze causes more arguments between readers than anything else, and a reference unit settles most of them.
- Lighting and background. Neither document gives a lux value or background colour. Write the values your qualification supports into the protocol. Your routine visual inspection SOPs, such as this SOP for manual visual inspection of filled vials, are a useful layout reference. The APS read still follows its own protocol.
- QC oversight. If QC staff do not inspect, FDA’s 2004 guidance wants QC oversight throughout, and suspect units go straight to the QC microbiologist.
- Hazy or opaque plastic. Reading under room lighting is not enough. PI 007-6 4.2.2.2 says to remove and examine the whole contents, and Annex 1 9.43 allows a validated detection method instead.
For example, we can see signs of growth are turbidity, sediment, flocs or a surface film. Any other detection method, including instruments, must be scientifically justified to detect contamination reliably (Annex 1 9.36 vii).
Two controls are separate. Growth promotion of the medium happens before use (Annex 1 10.9). Positive control of incubated units happens after incubation, by inoculating samples with reference organisms and local isolates (Annex 1 9.45 ii). PI 007-6 5.2.5 describes growth promotion testing on completion of incubation, with growth shown within 5 days at the temperature used for the simulation.
What Do You Do With Positive, Damaged or Unclear Units?
Set a suspect unit aside at once and treat it as a potential failure until the investigation decides otherwise.
- Segregate the unit, record its identity and position, and tell the QC microbiologist immediately.
- Examine the container and closure for damage that could have let organisms in (PI 007-6 6.1).
- Identify the isolate to species level when practical (Annex 1 9.43). PI 007-6 5.3.3 accepts genus as a minimum.
- Open a deviation and start the investigation.
Annex 1 9.46 (also in WHO TRS 1044 Annex 2) sets the target at zero growth. Any contaminated unit fails the APS, and these actions follow:
- Investigate the most probable root cause and apply corrective measures.
- Run successful, consecutive repeat APS, normally at least three, to show the process is back in control.
- Review records since the last successful APS and assess the risk of sterile breaches in those batches.
- Quarantine products made on the line since the failure until it is resolved, and include other unreleased batches in the investigation.
- Limit the operator’s activities until retraining and requalification, where the root cause points to operator activity.
- Resume production only after successful revalidation.
FDA’s 2004 guidance and PI 007-6 list older tiers by run size, where one contaminated unit in a large run triggers an investigation only. Annex 1 and WHO state the single zero-growth target, so use that.
Damaged containers need careful wording. PI 007-6 6.1 says damaged containers are not counted as failures. FDA says a unit found damaged after incubation starts stays in the data unless the exclusion is fully justified and the deviation explained in the report. Annex 1 treats any contaminated unit as a failure. Do not exclude a unit without written justification approved by QA.
Annex 1 does not define an unclear unit. Your protocol should say who decides, usually a second qualified reader and the microbiologist, and how the decision is recorded. An aborted run needs a written procedure that treats commercial lots the same way, plus a documented investigation (Annex 1 9.48).
What Must You Record, and What Does QA Review?
Annex 1 9.47 says every APS run must be fully documented. The record should hold:
- A reconciliation of units: filled, incubated and not incubated, with a justification for each non-incubated unit.
- Every intervention, with its start and end time and the persons involved.
- All microbial monitoring data and other testing data, kept in the APS batch record. Environmental data from a qualified area, as in this AHU performance qualification protocol, belongs here.
- Incubator identity, start and end dates, and temperature records as the protocol defines.
- Each inspection: date, inspector, reference unit used, result, and the outcome for every suspect unit.
- Growth promotion results for the medium, positive control results and isolate identifications.
FDA defines yield as units incubated divided by units filled and expects criteria for yield and accountability. A count that does not reconcile is a deviation. Most reconciliation gaps I have seen began with units pulled after an intervention and never written down.
QA reviews the whole package against the protocol, closes the deviations and approves the report. Records should follow ALCOA+ data integrity principles, so a reviewer can see who inspected what, and when. A failed APS also feeds the PQS, because it can change the disposition of batches made since the last good run.
Recent Technology Trends
The biggest change is regulatory. Annex 1 was published on 25 August 2022 and applied from 25 August 2023. Its paragraphs 9.41 to 9.48 state one zero-growth target and require justified, validated incubation. The older run-size tiers in PI 007-6 and FDA’s 2004 guidance sit below that standard. WHO’s 2022 Annex 2 carries matching APS wording for 9.46 to 9.48.
What Common Errors Weaken APS Results?
- Treating the PI 007-6 schedule as an Annex 1 requirement, or changing it without validation data.
- Discarding integral units, or removing more units at an intervention than routine production would.
- Skipping the swirl or inversion, so the closure surface never meets the medium.
- Overfilling the container or overlaying it with inert gas, which starves aerobes.
- Reading without a sterile reference unit, or reading opaque containers under room light.
- Using readers who were never tested with contaminated units.
- Leaving temperature excursions or late incubator loading unrecorded.
- Excluding damaged units without justification, or closing a run with unreconciled counts.
These gaps appear often in inspections, as the top audit findings in pharmaceutical manufacturing show for documentation in general. A self-inspection programme can catch them before an auditor does.
Our Expert Tips
- Name the source. Write in the protocol which document each incubation value comes from and where the validation data sits.
- Map under real load. Qualify the incubator with the load pattern the APS will use, and file the mapping report with the run.
- Keep a sterile reference. Place a known sterile unit beside the tray at every read.
- Write pulled units down at once. Log every unit removed during an intervention at the moment it happens.
- Test your readers. Mix spiked units into reader training, as PI 007-6 8.8 describes, and repeat it at set intervals.
- Define doubtful in advance. State who decides on an unclear unit and how the decision is recorded.
Frequently Asked Questions
Annex 1 fixes no number of days. It requires a scientifically justified and validated duration. PIC/S PI 007-6 describes 14 days in total for its generally accepted schedule, and FDA’s 2004 guidance says not less than 14 days. Your approved protocol states the actual duration, and the validation file supports it.
Annex 1 sets no temperature. PI 007-6 describes 20–25 °C for at least 7 days, then 30–35 °C. FDA recommends staying within 20–35 °C and within ±2.5 °C of the target, starting with the lower temperature if two are used. The site protocol decides the exact schedule.
Annex 1 9.41 requires units to be agitated, swirled or inverted before incubation so the medium wets every interior surface, including the closure. It does not state the position during incubation. The protocol defines it, as long as closure contact was achieved beforehand and the position is applied and recorded consistently.
Annex 1 9.45 requires inspection on completion of incubation by trained, qualified personnel. PI 007-6 5.3.1 allows a reading between the two temperature stages. Intermediate reads are otherwise a site decision. Whatever your plan, write it into the protocol and record every read.
Under Annex 1 9.46, any contaminated unit means a failed APS. Required actions include a root cause investigation, corrective measures, and normally at least three consecutive successful repeat APS. You also review batches since the last successful APS and quarantine product made on the line since. Isolates are identified to species level when practical.
Only with written justification. PI 007-6 says damaged containers are not counted as failures. FDA’s 2004 guidance keeps units damaged after incubation starts in the data unless exclusion is fully justified and the deviation explained. Annex 1 treats any contaminated unit as a failure. Document the damage and obtain QA approval.
Annex 1 9.36 (vii) accepts a detection method only if it is scientifically justified to detect contamination reliably. Annex 1 does not name laser headspace or automated readers. A site would need its own validation, and usually a discussion with its inspectorate, before replacing trained visual inspection.
Summary
Annex 1 asks for justified, validated incubation, trained inspection and zero growth. PI 007-6 and FDA supply reference schedules, including 14 days at 20–35 °C. Wet the closure, incubate every integral unit, read against a sterile reference, investigate any positive, and reconcile every unit before QA review.
References
- EU GMP Annex 1: Manufacture of Sterile Medicinal Products (2022), European Commission
- PI 007-6 Recommendation on the Validation of Aseptic Processes (2011), PIC/S
- Guidance for Industry: Sterile Drug Products Produced by Aseptic Processing, Current Good Manufacturing Practice (2004), US FDA
- WHO good manufacturing practices for sterile pharmaceutical products, Annex 2, TRS 1044 (2022), World Health Organization
- Automated Media Fill Inspection (2025), Wilco
Pankaj Sharma — Quality Control Specialist
Pankaj Sharma is a pharmaceutical QC professional who reviews our technical content for GMP accuracy, laboratory practices, and regulatory relevance.