Why do we want to deep dive?

Deep diving is underwater diving to a depth beyond the norm generally deeper than 30 meters [98 ft.]. Divers needed to breathe special gas mixtures because they get exposed to very high ambient pressure [More than 50 times atmospheric pressure]. Deep dive is done to explore deepest parts of the ocean and for groundbreaking discoveries. In other context, deep dive helps to focus on high-level and go deeper search in the subject to get high level knowledge. Deep dive can tell you the full story of what’s going on. It takes lot of time, energy but has benefits.

In the earlier blog, particulates were covered. In this blog, we will take a deep dive in particulates, to go into the depth of the subject.

Introduction

Container integrity defects such as cracks, misplaced stoppers, or incomplete seals, any of which may compromise the sterility of the product as well as other product characteristics such as fill level, discoloration, or clarity may also be detected during visual inspection, and non-conforming units should be rejected. Inspection for these other quality attributes often occurs at the same time as the inspection for particles.

Inspection Process Capability

The detection process is probabilistic and depends upon the particle size, shape, color, density, and reflectivity. Human eye resolving power is reported as 85–100 µ. The probability of detection for a seeded sample with a single 50-µ particle in a clear solution contained in a clear 10 mL vial with illumination between 2,000 and 3,000 lux is only slightly greater than 0%. The detection probability increases to approximately 40% for a seeded sample with a 100-µ particle and typically exceeds 95% for particles that are 200 µm and larger. Thus, in a qualified visual inspection system, the vast majority of particles that might go undetected and be introduced into the pharmaceutical supply chain will be smaller than 200 µm. Changes to the container [e.g., increasing size and opacity], formulation [e.g., color and clarity], fill level, and particle characteristics beyond size [e.g., color, shape, and density] will all affect the probability of detection [POD].

Typical Inspection Process flow-100% inspection.

Visual Inspection should take place prior to labeling. Each unit may be examined manually with the unaided eye, or by using a conveyor to transport and present the containers to a human inspector [semi-automated inspection], or by means of light obscuration or electronic image analysis [automated inspection]. This inspection may be performed in-line with filling or packaging or in a separate, off-line inspection department. 100% inspection refers to the complete inspection of the container–closure system and its contents.

Supplemental testing is required when the nature of the product or container limits visual inspection of the contents [e.g., with a lyophilized cake or powder or with an amber glass or opaque container]. Samples for supplemental testing may be taken from any point in the process after 100% inspection. During 100% inspection, limits on typical rejection rates should be established to identify atypical lots. These limits may be established for categories of defects [e.g., critical, major, and minor] or for specific types of defects (e.g., particles. If a limit is exceeded, it should trigger an investigation. The investigation may include additional inspection or it may determine whether additional inspection is necessary

Acceptance Sampling and Testing

After 100% inspection, a statistically valid sample is taken from the units accepted by the inspection process. This may be a random sample or a representative sample (e.g., at fixed time intervals or a fixed number per tray). Defects may not be distributed equally over the lot, and therefore a sampling process that represents the whole lot is required. Typical sampling can be found in the ANSI/ASQ Z1.4 standard, ISO 2859 or JIS Z9015 standards. For batch release, the sampling plans listed as Normal II are typically used. The AQL [acceptable quality limit] is the defect rate at which 95% of the lots examined will be accepted and is a measure of falsely rejecting good batches.

Reinspection

Reinspection may be appropriate if the initial 100% inspection is not successful. This includes instances when the established 100% inspection failure rate(s) and/or the accept/reject number(s) associated with the chosen AQL values have been exceeded. Reinspection should only be conducted using a prior-approved procedure that addresses key parameters such as the inspection conditions [e.g., same as primary inspection or modified to enhance detection of a specific defect type], the number of times reinspection may be performed [this should be limited, and justified], and the acceptance criteria [e.g., same as primary inspection or tightened]. If reinspection is required often, consideration should be given to improving the sensitivity of the primary inspection process. Frequent and routine reinspection is not recommended.

Two stage inspection

In cases where an assignable cause, such as formation of air bubbles or specific container or closure variation, results in a high false-rejection rate [rejection of acceptable units], the use of a second inspection step may be considered. This is more common with automated inspection systems. Under these circumstances, the inspection system is adjusted to ensure acceptance of good units. Those not accepted are considered of uncertain disposition until inspected by another means [e.g., manual inspection following automated inspection]. Inspection conditions may be adjusted to provide greater sensitivity in this second inspection step [e.g., additional inspection time] to ensure a high probability that true defective units will be rejected. The limitations of the first inspection and the reason for conducting a second stage of inspection should be clearly defined and documented. The second inspection of these units by the same method [e.g., automated inspection after automated inspection] is generally not recommended. However, it may be suitable when the root cause is air bubbles in the solution and a study has been performed to establish an appropriate holding time to allow the bubbles to dissipate before performing the second inspection. It is recommended that each inspection stream [those accepted by the first stage and those accepted by the second stage] be sampled separately and evaluated against the sampling plan acceptance criteria before they are confirmed as accepted and recombined into a single batch.

Defect classification

Critical– Defects that may cause serious adverse reaction or death of the patient if the product is used. This classification includes any nonconformity that compromises the integrity of the container and thereby risks microbiological contamination of the sterile product.

Major – Defects carry the risk of a temporary impairment or medically reversible reaction, or involve a remote probability of a serious adverse reaction. This classification is also assigned to any defect which causes impairment to the use of the product.

Minor – Defects do not impact product performance or compliance; they are often cosmetic in nature, affecting only product appearance or pharmaceutical elegance.

Knapp’s methodology – is used worldwide as an industry common practice for rejecting particle defects.

The POD [Probability of Detection] at 70% or greater is known as the Reject Zone. Detection is that at this size threshold particles of the same size may routinely be missed or go undetected up to 30% of the time.

Lyophilized product

The solid, lyophilized cake can mask the presence of visible particles because they cannot be seen within the solid matrix. Because of these challenges in evaluating acceptability, a small sample of units is reconstituted and inspected for visible in addition to the 100% inspection of the cakes for visible particles. Typical sampling plans for this type of test can be found in the special sampling plans S-3 and S-4 in ANSI/ASQ Z1.4. The S-plans offer a practical compromise between sample size and statistical power and for most batch sizes between 3,201 and 150,000 suggest a sample size of 20 with an accept number of 0 (based on an AQL of 0.65%). Once inspection of these reconstituted samples has been performed, they may be used for other required testing, such as that for subvisible particles, potency, impurities, or other specified tests. If particles are detected in this relatively small sample, additional units may be reconstituted as part of an investigation and to assess the compliance of the entire batch.

Sterile powders should be reconstituted and inspected for visible foreign particles using an approach similar to that for lyophilized products.

For suspension products, a test dissolving the suspension or disruption of the emulsion that provides   for extrinsic and intrinsic particle detection is also recommended as part of destructive supplemental testing of a small sample as described above for lyophilized products.

For amber containers, plastic or translucent containers, increased light intensity [e.g., 8,000–10,000 lux] may be required to observe visible particles during inspection. Directional lighting from behind the container may also be beneficial. At the extreme, filled transferred into clean containers can be done.

Large-volume containers (>100 mL) may require additional time to complete a thorough inspection.

Critical process Parameters [CPP] in MVI [Manual Visual Inspection]

Light intensity

NLT 2,000–3,750 lux at the point of inspection for routine inspection of clear glass containers. Increased light levels are recommended for plastic containers or those made from amber glass.

Light intensity in each inspection station should be measured periodically to ensure continued compliance within the specified range. The frequency of monitoring should be based on historical experience with the type of light source in use. Use of both black and white backgrounds provides good contrast for a wide range of particulate.

Inspection rate

10 seconds / container [5 seconds each against both black and white backgrounds]. Larger or more complex containers may require additional time for inspecting all attributes. Recording the time spent inspecting each batch and then calculating a nominal inspection rate is a good way to confirm that the rate of inspection was within established limits.

Container handling and movement

Good techniques for manual inspection include a careful swirl or inversion of the liquid product within the container. Holding many containers by hand at once should be avoided. Magnification to avoid eye strain. Magnification can be helpful for critical examination during investigation.

Inspector fatigue and ergonomic considerations

It is recommended that inspectors be given a break from performing inspection at least every hour. This break should allow time to rest the eyes and mind, and may be achieved with a short rest [e.g., 5 min] or a longer meal break. This regular break may also be met through rotation to a non-inspection function, such as material handling or documentation. Temperature and humidity should be controlled for inspector comfort. Reduced ambient lighting is recommended to focus the inspection process and to reduce distraction from extraneous reflections.

Semi-Automated Visual Inspection

Semi-automated visual inspection combines automated material handling of the containers to be inspected with human vision and judgment to make the decision to accept or reject. These systems often use a conveyor equipped with rollers to transport the containers in front of the inspector inside an inspection booth or station. Mirrors may also be used to provide a clear view of the top and bottom of each container. Rejected units may be removed from the rollers by hand, and some systems are equipped with a remote rejection system that can be triggered by the inspector.

Critical process Parameters [CPP] in MVI [Semi-Automated Visual Inspection]

Spin speed for liquid products and rotation rate for all containers should be established during validation/qualification and maintained within the validated range for routine inspection.

Automated Visual Inspection

Automated visual inspection [AVI] combines automated material handling of the containers with electronic sensing of product appearance. AVI offers advantages in the areas of throughput and consistency, compared with MVI. Examples are Light obscuration and imaging methods.

Qualification and validation of Inspection processes

Preparing Defect Standards-Visual inspection standards may be identified from known production rejects, or may be created manually with characterized particulate material. A single particle/seeded container should be used when determining detection thresholds.

Particle types

Physically prepared particles can be sieved initially to target a specific size, and then the individual particles are measured using optical microscopy.

Once a well-defined defect standard is available, it is assigned a detection frequency or probability of detection [POD] by conducting a documented, manual human inspection qualification that is accomplished by repeated manual inspection. This repeated inspection is the basis for qualifying the defect standard. The Knapp methodology recognizes that the detection of particles is probabilistic, and repeated inspections with strict controls on lighting and inspection pacing/sequencing generate the statistical confidence to assign a reject probability to each standard unit. A manual, visual inspection POD of ≥0.7 or 70%, is required to assign the container to the Reject Zone for subsequent calculation of the reject zone efficiency [RZE].

Probabilistic data for particulate standards can be achieved with 30–50 inspections of each container. This is best achieved with multiple inspectors.

POD = [Number of times rejected] / [Number of times inspected]

Test sets

These sets are used to specifically challenge the particle detection technique of human inspectors, used as part of a defect test set (including container–closure defects) for human qualification, or for comparison during automated equipment qualification and validation. The test set should be prepared with duplicate product units per particle type and size to ensure that backup units are available. When using test sets, it is a good practice to verify the presence of particles before and after use, when a freely moving particle cannot be verified, the unit should not be used and the data should be excluded from subsequent calculations.

The number of defective units in each test set should be limited to approximately 10%. The accept containers will be identified as having a pre-determined manual, visual inspection POD of <0.3 or 30%. Any particle standards found to fall within the acceptable “grey zone”, indicating a manual inspection rejection probability ≥30% and <70%, may be included as an “acceptable unit” in a test set, if desired.

Written procedures should define the qualification criteria, appropriate storage conditions, periodic examination and requalification, expiration, and sample custody during use. Test sets should be approved by the quality unit.

Types of test sets

It is a standard curve of detection probabilities at various particle types and sizes in an approximate range of 100–500 µm [with recommended increments of 100 µ]). Fibers are typically observed in sizes >500 µm. for clear solutions in 10-mL tubing glass vials, past thresholds studies indicate that particles within the range of 150–250 µm [500–2000 µm for fibers] can be detected with a POD of 70% or greater. Results can differ due to differences in product formulation as well as container type and size. Detection threshold studies are typically the first step in evaluating the performance of any new inspection method. The samples should be blinded. UV ink [Invisible to the inspectors] may be used to mark all containers. Alternatively, bar codes or other coded labels may be used. Separate test sets should be prepared to represent each unique combination. A bracketing approach may be used with regard to different container sizes.

Training and Qualification of Human Inspectors

Before training, potential inspectors should be tested for near-vision performance should be the equivalent of 20/20 with no impairment of color vision. Initially, train the potential inspectors with defect photographs or a video library and clear written descriptions. Utilize SME to mentor and provide hands-on training with defect standards for the specified method. Qualification should be performed for each product type and package that the inspector will encounter. A bracketed or matrix approach can be used to simplify qualification of products.

Inspector Qualification Requirements

Three successful inspections of the test set are recommended to demonstrate consistent performance for initial qualification of new inspectors. Acceptance criteria for each defect class should be based on the POD [or RZE] observed during test set qualification. A limit is also needed for false rejection, with a recommended target of <5% falsely rejected good units.

Requalification

Inspectors should be requalified at least annually. Requalification includes a test of visual vision performance and testing with at least one product/test set configuration. A single successful inspection of the test set is sufficient for requalification. Requalification may also be necessary in the event that poor performance is observed during routine inspection or if the inspector has been away from the inspection operation for an extended period of time [e.g., 3 months].

If an inspector fails the requalification test, a retraining process should be initiated to identify the root cause and allow the inspector to receive additional instruction. After this process has been completed, the inspector may attempt to meet the acceptance criteria one additional time. If the inspector fails, he or she may attempt to qualify again after a specified time period.

Products in distribution

“If it becomes necessary to evaluate product that has been shipped to customers [e.g., because of a complaint or regulatory concern], sample and inspect 20 units. If no particles are observed in the sample, the batch is considered essentially free of visible particulates. If available, additional units may be inspected to gain further information on the risk of particulates in the batch.”

Conclusion

Visual inspection for particles and other visible defects continues to be an important part of the manufacturing process for injections. Inspectors must be trained to ensure consistent, high-quality performance. Alternative inspection methods either semi-automated or fully automated, may be used in place of manual inspection methods. Where machine methods are used, the equipment must be validated to demonstrate equivalent or better performance when compared to manual inspection. The use of test sets that contain standard defects is an important element in inspector training and qualification. Good product development will lead to a stable product with a lower risk of particle formation. Identification of the type or types of particles found during product development and routine manufacturing is an important aid in source identification and reduction. Inspection results should be trended to further aid in continuous process improvement with the ultimate goal of defect prevention.

References

USP <1790> Visual inspections of Injections.

Note-The images given for representation in this blog are taken from Google Images. Many thanks for Google.

“More the practical and efficient controls, more we will inch towards the Quality for the safety of any living forms in this Earth”.

Muthu