Questions & Answers on Manual Visual Inspection (MVI)

The visual inspection of parenteral products is described in detail in pharmacopoeias. Nevertheless, there remains room for interpretation in practical application, which regularly gives rise to questions. The questions and answers compiled below from conferences address common uncertainties that arise in practice and provide guidance based on current regulatory requirements and best practices.

 Q: How can a start-up without trained visual inspection personnel establish a manual visual inspection process from scratch?
A: As a startup, it's, it's always good to ask acquaintant companies or companies where their employees have previously worked to give you a kind of defect library, so that people can be trained with real defects. So, ask other companies to give you discarded containers according to your primary packaging material if possible and train your inspectors with these. Later, once the first development batches have been produced, you can gradually replace the "foreign" defects with your own defects and thus build up your defect library. A startup can also use a contract manufacturer for carrying out the visual inspection and take over after some time.

Q: There's currently a lack of clarity concerning the required number of qualification runs for qualification of inspectors. How many runs are recommended?
A: You should conduct standardized training for your employees, then it is sufficient to carry out the qualification once. If employees pass, they prove that they have completed the training and are approved for routine inspections. So it is perfectly fine to carry out this qualification only once.

Q: Should the annual qualification / requalification be conducted using the worst-case product and performed at the end of the production shift?
A: The annual qualification or requalification of manual visual inspectors should be performed using the defined qualification kit, not the "worst-case product." EU GMP Annex 1 and USP require requalification to be conducted on a specific commercial product presentation. Instead, both require qualification using appropriate samples from the manufacturer's defect library, considering worst-case conditions such as inspection time, fatigue, container size, and defect difficulty.
The qualification kit serves as a standardised reference set, ensuring reproducibility and comparability of performance over time. It should be designed to represent worst-case detectability scenarios (e.g. smallest or lowest-contrast particles, borderline defects). Consistency of the qualification kit is essential to allow meaningful performance trending.
Regarding timing, Annex 1 (8.31) requires consideration of factors such as fatigue. Conducting requalification at the end of a production shift can therefore be considered a realistic approach, as it reflects routine operating conditions. However, the regulation does not explicitly mandate this.

Q: Would you recommend using Tyndall lighting during inspection to help identify glass container defects?
A: There is no regulatory requirement to use Tyndall lighting for routine visual inspection. The pharmacopoeias define standardised inspection conditions (illumination, background, inspection time), and compliance should be demonstrated under these conditions.
If routine inspection performance is stable and market data do not indicate issues with glass defects, there is no need to introduce additional lighting techniques. Enhanced methods such as Tyndall lighting may be considered if recurring glass-related defects are observed and standard inspection controls prove insufficient.

Q: Do you recommend a time limitation during qualification?
A: Yes. Qualification should be performed under conditions that simulate routine inspection practice as closely as possible, including realistic time constraints.
EU GMP Annex 1 requires that inspector qualification considers worst-case conditions such as inspection time, line speed, container size, and fatigue. USP <1790> similarly emphasizes that qualification should reflect actual operating conditions.
Therefore, if inspectors normally perform visual inspection for defined periods between breaks (e.g., 30 minutes), the qualification exercise should be conducted within the same time frame. The number of units included in the qualification set should correspond to what the operator would typically inspect during that period.
The objective is not to test performance under artificial or unlimited time conditions, but to confirm that defect detection capability is maintained under realistic working conditions.

Q: How should inspection times be defined for products that are difficult to inspect (e.g., powders or suspensions)? Should the standard 5 seconds per background be maintained?
A: The standard inspection time described in pharmacopoeial procedures (e.g., approximately 5 seconds against a white background and 5 seconds against a black background per Ph. Eur. 2.9.20 and USP <790>) represents a harmonised reference condition for routine products. However, these chapters do not prohibit reasonable adjustment of inspection time where justified.
For difficult-to-inspect products-such as lyophilised powders, suspensions, or products that are not fully clear after resuspension-strict adherence to 5 seconds may not always ensure adequate defect detection. In such cases, extending the inspection time can be justified, provided it remains controlled and defined.
Unlimited inspection time is not advisable. Inspection must remain standardised and reproducible. Instead, companies should:

• Perform a risk-based assessment of detectability.
• Validate the inspection time during qualification.
• Monitor reject rates, complaint data, and trending results.
• Adjust inspection duration if data indicate insufficient detectability.

It is also important to distinguish between different inspection aspects. While particulate detection in turbid or resuspended products may require extended observation, other attributes (e.g., stopper position, crimp integrity, glass defects) may remain suitable for standard inspection time.