Repeatability Testing for Visible Particulate Matter Detection: A Technical Guide for Pharmaceutical

Visible particulate matter detection repeatability testing is a critical method-validation study that verifies the consistency and reliability of inspection procedures used to identify visible foreign particles in injectable and ophthalmic drug products. By demonstrating that repeated inspections under identical conditions produce reproducible accept/reject decisions, manufacturers can confirm compliance with pharmacopoeial standards and protect patient safety. This article explains the technical principles, sample types, test protocols, acceptance criteria, and practical applications of repeatability studies for visible particulate matter detection.

Technical Overview

Visible particulate matter, also referred to as visible foreign matter, is any insoluble contaminant present in an injection or ophthalmic preparation that can be observed by the unaided eye under standardized inspection conditions. Such contaminants typically fall in the size range of approximately 50 micrometers and above, and may include glass fragments, rubber closure particles, fibers, metal shavings, lubricant residues, crystallized material, and protein aggregates in biological products. Because even a single visible particle in a parenteral product represents a potential patient-safety risk, pharmacopoeial standards such as USP <790>, USP <1790>, and the corresponding European and Japanese Pharmacopoeia chapters require 100 percent inspection of every unit.

Repeatability, as defined in ISO 5725-1, is the degree of agreement between independent test results obtained with the same method, on identical test items, in the same laboratory, by the same operator, using the same equipment, within short intervals of time. In the context of visible particulate detection, repeatability testing quantifies whether the inspection procedure, whether manual or automated, yields consistent results when the same units are examined multiple times. This is essential because visual detection is inherently probabilistic: the probability of detection depends on particle size, shape, contrast, mobility, container geometry, illumination intensity, and observer performance. A repeatability study converts this uncertainty into measurable statistics, allowing the inspection process to be qualified, validated, and continuously monitored.

• Repeatability evaluates within-laboratory, short-term consistency of detection results
• It provides statistical evidence that the inspection method is fit for its intended purpose
• It supports qualification of inspectors, verification of automated inspection systems, and batch release decisions
• It aligns with method validation principles outlined in ICH Q2 and with the risk-based expectations of USP <1790>

Testing Samples

Small-Volume Injectable Solutions

Liquid injectable products packaged in ampoules or vials, typically ranging from 1 mL to 20 mL, represent the most common sample type. These units are inspected directly without reconstitution, and repeatability is assessed by inspecting the same units repeatedly under controlled conditions.

Sterile Powders and Lyophilized Products

Freeze-dried and powder-filled vials are first reconstituted with the specified diluent or with water for injection, allowed to equilibrate, and then inspected. The repeatability protocol must confirm that the reconstitution and swirling steps themselves do not introduce variability into the detection outcome.

Large-Volume Parenterals and Ophthalmic Preparations

Infusion bags and large-volume bottles, as well as ophthalmic solutions and suspensions in plastic or glass containers, require adapted inspection parameters such as longer observation times and adjusted illumination due to container size and transparency characteristics.

Seeded Test Sets and Reference Units

For qualification and validation purposes, accredited third-party laboratories and manufacturers use test sets containing units deliberately seeded with characterized particles, such as glass fragments, stainless steel fragments, rubber pieces, and fibers of known size ranges. These defect libraries enable quantitative measurement of detection rates and false-negative behavior.

Testing Items

The core objective of the repeatability study is to evaluate the stability of detection results across replicate inspections. Typical test items and reported parameters include:

• Detection consistency rate: the proportion of repeated inspections in which a given defect unit is correctly rejected
• Probability of detection (POD) for seeded particles of defined size and composition
• False-negative rate: defect-containing units incorrectly accepted as conforming
• False-positive rate: good units incorrectly rejected, which drives yield loss
• Repeatability of enumerative particle counts, expressed as relative standard deviation (RSD), when quantitative counting is performed
• Inter-unit and inter-vial detection uniformity across the entire test set
• Repeatability of automated inspection machine zone thresholds and rejection decisions across consecutive runs

Testing Methods

Manual Light-Inspection Repeatability Protocol

The classical manual method follows the pharmacopoeial light-inspection procedure: units are equilibrated to room temperature, held against a non-glare white background and a matte black background, gently inverted or swirled to mobilize potential particles, and observed under controlled illumination, typically in the range of 2,000 to 3,750 lux depending on container type. For repeatability evaluation, each unit in the test set is inspected a defined number of times, commonly ten to twenty replicate inspections, by the same trained inspector under unchanged conditions. Every accept/reject decision is recorded independently, without the inspector knowing prior results. The detection proportion for each unit is then calculated, and the overall consistency of the inspection process is expressed as the agreement rate among replicates.

Automated Inspection Repeatability

For automated visual inspection systems, repeatability is established using the industry-standard Knapp test methodology. Each vial in a defect test set is passed through the inspection machine a sufficient number of times, for example ten or more consecutive passes, and the number of rejections per vial is recorded. From these data, the average rejection probability of each unit is computed, and the machine detection threshold is set so that critical defects achieve an average rejection rate at or above a predefined target, commonly 70 percent or higher in accordance with accepted industry practice. Repeatability is further verified by repeating the test set on different days, after recalibration, and across operator shifts, with results compared statistically to confirm stability.

Data Evaluation and Acceptance Criteria

Detection data are summarized as proportions with 95 percent confidence intervals, using methods such as the Wilson interval for binomial data. Acceptance criteria are risk-based and defined in the validation protocol before testing begins. Typical criteria include a minimum agreement rate among replicate inspections, a maximum permitted false-negative rate for seeded critical defects, and a bounded false-positive rate for good units. For enumerative tests, the relative standard deviation of replicate counts should fall within predefined limits. Documented acceptance criteria, raw data, and statistical analysis form the technical report issued by accredited third-party laboratories or in-house quality control departments.

Applications

Repeatability studies for visible particulate detection are applied throughout the product lifecycle and across multiple quality functions:

• Validation and periodic verification of the visual inspection method described in batch release procedures
• Qualification and routine requalification of automated visual inspection machines before and after line changes
• Training, certification, and periodic performance assessment of human inspectors
• Support of regulatory submissions, including method validation packages and inspection process descriptions
• Investigation of customer complaints, product recalls, and out-of-specification results related to visible contamination
• Stability programs, where containers are inspected at defined time points and inspection consistency must be demonstrated
• Evaluation of container-closure systems and packaging components from suppliers for their propensity to generate particles
• Biologics development, where subvisible-to-visible protein aggregation requires robust and repeatable detection strategies

FAQ

What particle size is considered visible?

Particles of approximately 50 micrometers are generally regarded as the threshold of visibility, and such particles have roughly a 50 percent probability of detection under standardized inspection conditions according to USP <790>. Reliable, near-certain detection requires larger particles, and inspection conditions such as illumination and container clarity strongly influence the outcome.

How many replicate inspections are required in a repeatability study?

There is no single universal number; protocols typically specify ten to twenty replicate inspections per unit for manual studies, and ten or more machine passes per vial for automated systems using the Knapp methodology. The number of replicates should be justified statistically based on the desired confidence level and the target detection probability.

What is the difference between repeatability and intermediate precision?

Repeatability assesses results obtained by the same operator, same equipment, and same conditions within a short time frame. Intermediate precision evaluates the method under varied conditions, such as different days, different analysts, and different instruments, and both are typically required for a complete method validation.

What is the Knapp test?

The Knapp test is a statistical methodology used to characterize the rejection probability of an inspection system, either manual or automated. Each seeded defect unit is inspected multiple times, the average rejection probability per unit is calculated, and the inspection threshold is set so that the system achieves the required detection performance for critical defects.

What acceptance criteria are typically applied?

Acceptance criteria are risk-based and must be predefined. Common criteria include a minimum replicate agreement rate, a minimum average rejection probability for seeded critical defects, a maximum false-negative rate approaching zero for large particles, and a controlled false-positive rate to avoid unnecessary yield loss.

Can repeatability testing be performed by external laboratories?

Yes. Accredited third-party laboratories routinely conduct repeatability studies using standardized test kits, controlled inspection environments, and documented statistical protocols, providing independent verification that a manufacturer's inspection process meets pharmacopoeial and regulatory expectations.