ASTM E3263-22e1

Standard Practice for Qualification of Visual Inspection of Pharmaceutical Manufacturing Equipment and Medical Devices for Residues

SIGNIFICANCE AND USE
4.1 Application of the approach described within this practice applies the science-based, risk-based, and statistics-based concepts and principles introduced in Guides E3106 and E3219.  
4.2 Application of the approach described within this practice provides a science-, risk-, and statistical-based approach for qualifying the inspection of equipment for cleanliness in accordance with 21 CFR 211.67(b)(6) and is in accordance with FDA Process Validation Guidance Life Cycle approach.  
4.3 Application of the approach described within this practice provides a science-, risk-, and statistical-based approach for qualifying the visual inspection of equipment for cleanliness in accordance with European Medicines Agency (EMA) Annex 15.  
4.4 Application of the approach described within this practice provides a science-, risk-, and statistical-based approach for qualifying the visual inspection of equipment for cleanliness in accordance with the EMA’s Q&A Guidance (Q&A’s #7 and #8) (2).  
4.5 Visual Inspection used as described in 4.4 should only be used in situations where there is a suitable safety margin between the VRL and MSSR and robust detectability at the VRL.  
4.6 Application of the approach described within this practice applies the risk-based concepts and principles introduced in ICH Q9. As stated in ICH Q9, the level of effort, formality, and documentation for validation (including cleaning validation) should also be commensurate with the level of risk.  
4.7 Application of the approach described within this practice provides a science-, risk-, and statistical-based approach for releasing manufacturing equipment and manufactured medical devices or cleanliness that is compatible with the U.S. FDA Guidance for Industry, PAT – A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance.  
4.8 Key Concepts—This practice applies the following key concepts: (1) visual inspection, (2) quality risk management, (3) science-based appr...
SCOPE
1.1 This practice provides statistically valid procedures for determining the visual detection limit of residues and the qualification of inspectors to perform the visual inspection of pharmaceutical manufacturing equipment surfaces and medical devices for residues.  
1.2 This practice applies to pharmaceuticals (including active pharmaceutical ingredients (APIs); dosage forms; and over-the-counter, veterinary, biologics, and clinical supplies) and medical devices following all manufacturing and cleaning. This practice is also applicable to other health, cosmetics, and consumer products.  
1.3 This practice applies to many types of chemical residues (including APIs, intermediates, cleaning agents, processing aids, machining oils, and so forth) that could remain on manufacturing equipment surfaces or medical devices that have undergone all manufacturing steps including cleaning.  
1.4 This practice applies only to equipment or devices that have been justified through a Quality Risk Management program to have an acceptable hazard analysis, have cleaning processes that are repeatable and validated and where Visual Inspection can be relied upon to determine the cleanliness of the equipment at the residue limit justified by the HBEL.  
1.5 The values stated in International System of Units (SI) units are to be regarded as standard. No other units of measurement are included in this standard.  
1.6 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety, health, and environmental practices and determine the applicability of regulatory limitations prior to use.  
1.7 This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the Development of International Standards, Guides and Recommend...

General Information

• Status: Published
• Publication Date: 30-Apr-2022
• ICS: 11.020.01 - Quality and environmental management in health care
• ICS: 11.120.01 - Pharmaceutics in general
• Technical Committee: E55 - Manufacture of Pharmaceutical and Biopharmaceutical Products
• Drafting Committee: E55.14 - Measurement Systems and Analysis

Relations

• Refers: ASTM G121-18 - Standard Practice for Preparation of Contaminated Test Coupons for the Evaluation of Cleaning Agents
• Effective Date: 01-Dec-2018
• Refers: ASTM E3106-18 - Standard Guide for Science-Based and Risk-Based Cleaning Process Development and Validation
• Effective Date: 01-Sep-2018
• Refers: ASTM E3106-17 - Standard Guide for Science-Based and Risk-Based Cleaning Process Development and Validation
• Effective Date: 01-Dec-2017
• Refers: ASTM E2782-11 - Standard Guide for Measurement Systems Analysis (MSA)
• Effective Date: 15-Nov-2011
• Refers: ASTM E2782-11e1 - Standard Guide for Measurement Systems Analysis (MSA)
• Effective Date: 15-Nov-2011
• Refers: ASTM E2782-10 - Standard Guide for Measurement Systems Analysis (MSA)
• Effective Date: 01-Oct-2010
• Refers: ASTM G121-98(2010)e1 - Standard Practice for Preparation of Contaminated Test Coupons for the Evaluation of Cleaning Agents
• Effective Date: 01-Sep-2004
• Refers: ASTM G121-98(2015)e1 - Standard Practice for Preparation of Contaminated Test Coupons for the Evaluation of Cleaning Agents
• Effective Date: 01-Sep-2004
• Refers: ASTM G121-98 - Standard Practice for Preparation of Contaminated Test Coupons for the Evaluation of Cleaning Agents
• Effective Date: 10-Sep-1998
• Refers: ASTM G121-98(2004) - Standard Practice for Preparation of Contaminated Test Coupons for the Evaluation of Cleaning Agents
• Effective Date: 10-Sep-1998

Overview

ASTM E3263-22e1 is an international standard published by ASTM International, providing a comprehensive practice for the qualification of visual inspection methods to detect residues on pharmaceutical manufacturing equipment and medical devices. This standard ensures that visual inspection processes are robust, reliable, and meet global quality and regulatory requirements, utilizing science-based, risk-based, and statistics-based approaches. The practice is applicable to pharmaceuticals, medical devices, and related health, cosmetics, and consumer products, emphasizing compliance with regulations such as 21 CFR 211.67(b)(6), FDA Process Validation Guidance, and EMA Annex 15.

Key Topics

• Statistical Qualification of Visual Detection: Describes statistically valid methods for determining the visual detection limit (Visual Residue Limit, VRL) and the qualification of inspectors.
• Scope of Application: The practice applies to all stages of manufacturing and cleaning for pharmaceuticals (including APIs, dosage forms, clinical supplies, and more) and medical devices, as well as other industries where equipment cleanliness is critical.
• Types of Residues: Considers various chemical residues, such as active pharmaceutical ingredients, cleaning agents, processing aids, and intermediates that may persist after cleaning.
• Risk Management Integration: Visual inspection qualification must be justified through a quality risk management program, with acceptable hazard analysis and validated cleaning processes.
• Safety and Compliance: Ensures that visual inspection as a standalone method is appropriate only where a suitable safety margin exists and detectability at the established VRL is robust.
• Lighting and Environmental Conditions: Specifies the importance of consistent lighting conditions and proper handling of surrogate surfaces to ensure meaningful qualification results.

Applications

• Pharmaceutical Manufacturing: Used to qualify visual inspection of residue on equipment used for drugs, APIs, biologics, over-the-counter products, and clinical supplies, in line with Good Manufacturing Practice (GMP) requirements.
• Medical Device Production: Ensures surface cleanliness of devices after all manufacturing and cleaning steps, contributing to product safety and regulatory compliance.
• Cleaning Validation: Supports cleaning process validation activities by setting clear criteria for when visual inspection is an appropriate and sufficient residue detection method.
• Inspector Qualification: Guides the training and competency assessment of personnel who perform visual inspections, ensuring consistent and reliable detections.
• Risk-Based Approaches: Aligns with international guidance on quality risk management (such as ICH Q9), advocating that the extent of qualification and documentation matches the level of risk identified.

Related Standards

ASTM E3263-22e1 references and aligns with several key international standards and regulations:

• ASTM E3106: Guide for science-based and risk-based cleaning process development and validation.
• ASTM E3219: Guide for derivation of health-based exposure limits (HBELs).
• ASTM E2782: Guide for measurement systems analysis (MSA).
• ICH Q7, Q9, Q10: International Conference on Harmonisation Good Manufacturing Practice, Quality Risk Management, and Pharmaceutical Quality System.
• ISO EN 12464: International standard for lighting of indoor workplaces, relevant for visual inspection conditions.
• FDA 21 CFR 211.67(b)(6): U.S. current good manufacturing practice regulation for cleaning and maintenance of equipment.
• EMA Annex 15: European Medicines Agency guidance on qualification and validation.
• FDA Process Validation Guidance: Principles for process validation life cycle, including cleaning validation.

Practical Value

ASTM E3263-22e1 adds significant practical value by:

• Standardizing the qualification of visual inspection, reducing variability and subjectivity in cleanliness assessments.
• Supporting global regulatory compliance and audit readiness by aligning inspection practices with FDA and EMA expectations.
• Facilitating scientifically justified decisions regarding when visual inspection is an appropriate sole or supplementary method for residue detection.
• Enhancing patient safety and product quality through a documented, risk-based approach to cleaning validation and equipment release.

This standard empowers companies to implement sound visual inspection programs that withstand regulatory scrutiny and help maintain high standards of product safety, quality, and integrity.

Standards Content (Sample)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
´1
Designation: E3263 − 22
Standard Practice for
Qualification of Visual Inspection of Pharmaceutical
Manufacturing Equipment and Medical Devices for
Residues
This standard is issued under the fixed designation E3263; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
ε NOTE—Editorial changes were made to Table 1 (Note 1) in February 2023.
1. Scope 1.7 This international standard was developed in accor-
dance with internationally recognized principles on standard-
1.1 This practice provides statistically valid procedures for
ization established in the Decision on Principles for the
determining the visual detection limit of residues and the
Development of International Standards, Guides and Recom-
qualification of inspectors to perform the visual inspection of
mendations issued by the World Trade Organization Technical
pharmaceutical manufacturing equipment surfaces and medical
Barriers to Trade (TBT) Committee.
devices for residues.
1.2 This practice applies to pharmaceuticals (including ac-
2. Referenced Documents
tive pharmaceutical ingredients (APIs); dosage forms; and
2.1 ASTM Standards:
over-the-counter, veterinary, biologics, and clinical supplies)
E2782 Guide for Measurement Systems Analysis (MSA)
and medical devices following all manufacturing and cleaning.
E3106 Guide for Science-Based and Risk-Based Cleaning
This practice is also applicable to other health, cosmetics, and
Process Development and Validation
consumer products.
E3219 Guide for Derivation of Health-Based Exposure Lim-
1.3 This practice applies to many types of chemical residues
its (HBELs)
(including APIs, intermediates, cleaning agents, processing
G121 Practice for Preparation of Contaminated Test Cou-
aids, machining oils, and so forth) that could remain on
pons for the Evaluation of Cleaning Agents
manufacturing equipment surfaces or medical devices that
2.2 ICH Guidance:
have undergone all manufacturing steps including cleaning.
Q7 Good Manufacturing Practice Guidance for Active Phar-
maceutical Ingredients
1.4 This practice applies only to equipment or devices that
have been justified through a Quality Risk Management Q9 Quality Risk Management
Q10 Pharmaceutical Quality System
program to have an acceptable hazard analysis, have cleaning
processes that are repeatable and validated and where Visual
2.3 ISO Standard:
Inspection can be relied upon to determine the cleanliness of
EN 12464 Light and lighting—Lighting of workplaces—
the equipment at the residue limit justified by the HBEL.
Indoor workplaces
2.4 Federal Regulation:
1.5 The values stated in International System of Units (SI)
21 CFR 211 Current Good Manufacturing Practice for Fin-
units are to be regarded as standard. No other units of
ished Pharmaceuticals
measurement are included in this standard.
1.6 This standard does not purport to address all of the
safety concerns, if any, associated with its use. It is the
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
responsibility of the user of this standard to establish appro-
contact ASTM Customer Service at [e-mail]. For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on
priate safety, health, and environmental practices and deter-
the ASTM website.
mine the applicability of regulatory limitations prior to use.
Available from International Conference on Harmonisation of Technical
Requirements for Registration of Pharmaceuticals for Human Use (ICH), ICH
Secretariat, 9, chemin des Mines, P.O. Box 195, 1211 Geneva 20, Switzerland,
This practice is under the jurisdiction of ASTM Committee E55 on Manufac- http://www.ich.org.
ture of Pharmaceutical and Biopharmaceutical Products and is the direct responsi- Available from American National Standards Institute (ANSI), 25 W. 43rd St.,
bility of Subcommittee E55.14 on Measurement Systems and Analysis. 4th Floor, New York, NY 10036, http://www.ansi.org.
Current edition approved May 1, 2022. Published September 2022. Originally Available from U.S. Government Printing Office, Superintendent of
approved in 2020. Last previous edition approved in 2020 as E3263 – 20. DOI: Documents, 732 N. Capitol St., NW, Washington, DC 20401-0001, http://
10.1520/E3263-22E01. www.access.gpo.gov.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
´1
E3263 − 22
2.5 European Guidance: administration and the result of a structured scientific evalua-
EudraLex Volume 4 Guidelines for Good Manufacturing tion of all available pharmacological and toxicological data
Practices for Medicinal Products for Human and Veteri- including both nonclinical and clinical data. E3219
nary Use, Annex 15: Qualification and Validation
3.1.8 lumen, n—SI unit of luminous flux and is the luminous
2.6 U.S. FDA Guidance: flux emitted within a solid angle of 1 steradian by a point
Guide to Inspections Validation of Cleaning Processes source having a uniform intensity of 1 cd.
Guidance for Industry Process Validation: General Prin-
3.1.8.1 Discussion—As the lumen is a measure of energy
ciples and Practices
per unit time, it shall also be related to the watt.
Guidance for Industry PAT A Framework for Innovative
3.1.9 lux, lx, n—unit of illuminance is equal to the illumi-
Pharmaceutical Development, Manufacturing, and Qual-
nation produced by a luminous flux of 1 lumen distributed
ity Assurance
uniformly over an area of 1 m .
Guidance for Industry Data Integrity and Compliant with
Drug CGMP Questions and Answers
3.1.9.1 Discussion—It can also be described as the illumi-
nation on a surface, all points of which are at a distance of 1 m
3. Terminology
from a point source of 1 candela (cd).
3.1 Definitions:
3.1.10 margin of safety, n—difference between the cleaning
3.1.1 ALCOA, n—an acronym referring to data, whether
acceptance limit (based on an HBEL) and the process residue
paper or electronic, requiring data to be Attributable, Legible,
data.
Contemporaneous, Original and Accurate, as defined in U.S.
FDA guidance. 3.1.10.1 Discussion—This value can be used as a measure
of the overall risk to patient safety presented by the cleaning
3.1.2 cleaning process residue, n—any residue, including,
process. The margin of safety can be measured by a number of
but not limited to, active pharmaceutical ingredients (APIs),
ways, including the process capability index (Cpk) and the
cleaning agents, degradation products, intermediates,
process performance index (Ppk).
excipients, and microbes remaining after a cleaning process.
3.1.11 maximum safe carryover, MSC, n—maximum
3.1.3 cleaning qualification, n—the risk evaluation activities
amount of carryover of a residual process residue (for example,
of the cleaning process during Stage 2 of the Cleaning
API, cleaning agent, degradant) into the next product manu-
Validation Lifecycle. Lifecycle verifications provide assurance
factured without presenting an appreciable health risk to
that during routine production the cleaning process is, or
patients.
remains, in a state of control.
3.1.11.1 Discussion—The MSC is calculated from the
3.1.4 cleaning validation, n—collection and evaluation of
HBEL and the total number of doses in a subsequent batch or
data from the cleaning process design stage through cleaning at
into the next manufacturing step, including the final step.
commercial scale that establishes scientific evidence that a
cleaning process is capable of consistently delivering clean 3.1.12 maximum safe surface residue, MSSR, n—maximum
equipment; lifecycle verifications provide assurance that dur- amount of residual process residue (API, cleaning agent,
ing routine production the cleaning process is, or remains, in a degradant, and so forth) that may remain on manufacturing
state of control. equipment or medical device surfaces without presenting an
appreciable health risk to patients.
3.1.5 data integrity, n—data integrity refers to the
completeness, consistency, and accuracy of data.
3.1.12.1 Discussion—The MSSR is calculated from the
3.1.5.1 Discussion—Complete, consistent, and accurate data
MSC and the total surface area of the equipment or device that
should be attributable, legible, contemporaneously recorded,
may result in patient exposure and is expressed in μg/cm ,
original or a true copy, and accurate. 2
mg/in. , and so forth. The MSSR is widely used in cleaning
3.1.6 exposure, n—process by which a human or animal can
validation programs, such as cleaning process development
come into contact with a hazard.
studies, cleaning qualification studies, analytical method vali-
dation recovery studies, as well as for qualification of visual
3.1.6.1 Discussion—Exposure may occur through any route
inspection.
(oral, inhalational, dermal, and so forth). Exposure may be
3.1.13 probability, n—likelihood of occurrence of harm.
short term (acute exposure), of intermediate duration, or long
term (chronic exposure).
3.1.14 qualified expert, n—individual with specific educa-
3.1.7 health-based exposure limit, HBEL, n—dose that is
tion and training in toxicology/pharmacology/
unlikely to cause an adverse effect if an individual is exposed, pharmacotherapy and risk assessment methods that can apply
by any route, at or below this dose every day for a lifetime.
the principles of toxicology to deriving an HBEL. E3219,
21 CFR 211.25(a), and 21 CFR 211.34
3.1.7.1 Discussion—The HBEL, being based on the critical
3.1.15 qualified statistician, n—individual with a working
effect, should be protective of all populations by all routes of
knowledge and education, training, or background in statistics
who can apply statistical analysis to data from cleaning and
Available from https://ec.europa.eu/health/documents/eudralex/vol-4_en.
cleaning validation studies. E3106
Available from U.S. Food and Drug Administration (FDA), 10903 New
Hampshire Ave., Silver Spring, MD 20993, http://www.fda.gov. 3.2 Definitions of Terms Specific to This Standard:
´1
E3263 − 22
3.2.1 attribute agreement analysis, n—assessment of the 3.2.10 visual inspection, VI, n—process of using the human
agreement between the ratings made by inspectors and the eye, alone or in conjunction with various aids, as the sensing
known standards. mechanism from which judgments may be made about the
condition of the surface to be inspected.
3.2.1.1 Discussion—Attribute agreement analysis can be
used to determine the accuracy of the assessments made by 3.2.10.1 Discussion—Supplementary aids, such as a
inspectors and identify which items have the highest misclas- boroscope, enable inspection for residues in hard-to-reach
sification rates. areas (for example, piping) may be included as part of the
3.2.2 compound, n—in this practice, this term may be either visual inspection.
the active pharmaceutical ingredient (API) that is used in the 3.2.11 visual residue limit, VRL, n—lowest level of a residue
2 2
formulation of a pharmaceutical product or a cleaning agent on a surface (in μg/cm or mg/m ) that is visible to a qualified
used to remove residues from equipment or devices. inspector under defined viewing conditions.
3.2.3 degradant, n—product of the breakdown of a mol-
4. Significance and Use
ecule through a degradation process.
4.1 Application of the approach described within this prac-
3.2.4 degradation, n—gradual decomposition of a molecule
tice applies the science-based, risk-based, and statistics-based
in which it is reduced in molecular size in small steps.
concepts and principles introduced in Guides E3106 and
Encyclopedia of Chemistry (1)
E3219.
3.2.5 product, n—in this practice, this term includes phar-
4.2 Application of the approach described within this prac-
maceutical formulations or medical devices used for the
tice provides a science-, risk-, and statistical-based approach
qualification of visual inspection.
for qualifying the inspection of equipment for cleanliness in
3.2.6 qualification, n—operation aimed at proving with
accordance with 21 CFR 211.67(b)(6) and is in accordance
regard to equipment, material, or personnel that the required
with FDA Process Validation Guidance Life Cycle approach.
conditions actually provide the expected results.
4.3 Application of the approach described within this prac-
3.2.7 spike, n—known amount of a solution of a compound/
tice provides a science-, risk-, and statistical-based approach
product/residue that is applied to a surrogate surface or device
for qualifying the visual inspection of equipment for cleanli-
for use in a qualification study.
ness in accordance with European Medicines Agency (EMA)
Annex 15.
3.2.7.1 Discussion—The act of applying these solutions is
termed “spiking” and the surrogate surface or device that the
4.4 Application of the approach described within this prac-
solution is applied to is referred to a “spiked” surrogate surface
tice provides a science-, risk-, and statistical-based approach
or device.
for qualifying the visual inspection of equipment for cleanli-
3.2.8 surrogate surface, n—part that is used as a substitute
ness in accordance with the EMA’s Q&A Guidance (Q&A’s #7
for a piece of manufacturing equipment or a medical device
and #8) (2).
surface.
4.5 Visual Inspection used as described in 4.4 should only
3.2.8.1 Discussion—These are fabricated parts made of the be used in situations where there is a suitable safety margin
same material of construction (MOC) and surface finish as the between the VRL and MSSR and robust detectability at the
manufacturing equipment or the medical device surface. Some VRL.
commonly used surrogate surfaces are called “coupons,” which
4.6 Application of the approach described within this prac-
are square or rectangular pieces (for example, 5 × 5 cm, 10 ×
tice applies the risk-based concepts and principles introduced
10 cm, 4 × 4 in., and so forth) of the manufacturing equipment
in ICH Q9. As stated in ICH Q9, the level of effort, formality,
or medical device MOC. Some surrogate surfaces are actual
and documentation for validation (including cleaning valida-
samples of the medical devices themselves or smaller pieces of
tion) should also be commensurate with the level of risk.
the manufacturing equipment used to represent larger pieces of
4.7 Application of the approach described within this prac-
the manufacturing equipment or medical device.
tice provides a science-, risk-, and statistical-based approach
3.2.9 visual detection index, VDI, n—logarithm of the ratio
for releasing manufacturing equipment and manufactured
on the visual residue limit divided by the maximum safe
medical devices or cleanliness that is compatible with the U.S.
surface residue.
FDA Guidance for Industry, PAT – A Framework for Innova-
3.2.9.1 Discussion—The log of this ratio obtains a logarith-
tive Pharmaceutical Development, Manufacturing, and Quality
mic scale that equals “0” when the values of the MSSR and
Assurance.
visual residue limit (VRL) are equal and becomes negative
4.8 Key Concepts—This practice applies the following key
when the VRL is lower than the MSSR and positive when it is
concepts: (1) visual inspection, (2) quality risk management,
higher. This scale provides a simple and visual means of
(3) science-based approach, (4) statistics-based approach, and
evaluating whether a VRL is low enough to be justified for
(5) process knowledge and understanding.
visual inspection.
5. Procedure
5.1 U.S. Regulation 21 CFR 211.67 (b) has required the
The boldface numbers in parentheses refer to a list of references at the end of
this standard. “inspection of manufacturing equipment immediately before
´1
E3263 − 22
use” since 1979. In general, pharmaceutical manufacturers 5.2.1 The following criteria for the release of equipment
have been releasing some equipment and some compounds without analytical testing (after cleaning qualification is com-
based on a “visual” inspection for many years and the industry pleted) are derived from EMA regulation/guidance and apply
has come to see this “inspection” as a “visual inspection” to the use of this practice (2). These same criteria are
requirement. However, based on the science, visual inspection appropriate for qualifying VI for the validation of cleaning
may not be appropriate in all circumstances. PIC/S (3) states processes for pharmaceuticals and medical devices after ap-
that “spiking studies should determine the concentration at propriate justification (11). See Fig. 1 for an example flow
which most active ingredients are visible,” but there have been diagram of this process.
only a few studies on VI performed in the past with varying
5.2.1.1 The compounds/products/residues selected for
results. In 1993, an article was published that mentioned that
evaluation of VI shall have acceptable hazard levels based on
spiking studies indicated many compounds were visible at
their HBELs derived by a qualified expert. See 9.3.
2 2
approximately 100 μg/4 in. (or 4 μg/cm ) (4). Another article
5.2.1.2 The cleaning processes of the compounds/products/
claimed that residues can be seen down to 1 μg/cm by using
residues selected should be validated and not present any
an additional light source (5). Another article claimed to see
significant concerns for patient safety if Visual Inspection will
residues of several compounds down to approximately
be used for release of equipment with no additional analytical
0.4 μg ⁄cm (6). A series of studies found a range of 0.4 to >10
testing.
μg ⁄cm for several different compounds (7, 8). Studies using a
5.2.1.3 The hazard level of a compound/product and the
different spiking technique calculated detection limits for one
acceptability of the cleaning process should be evaluated to
residue at levels of 3, 5, and 7 μg/cm depending on training
determine acceptability using a risk management tool such as
(9). A logistic-regression-based approach has also been pro-
the Shirokizawa Matrix (14).
posed to derive the limit of visible residue from spiking studies
5.2.1.4 The VI data collected for these compounds/
(10).
products/residues shall demonstrate that VI can be relied on for
5.2 Initial Criteria for Establishing Qualification Programs determining the cleanliness of the equipment at the residue
for VI: limit(s) justified by the HBEL.
NOTE 1—① HBEL Toxicity Scores are calculated as in Reference (12). ② Cpu Scores are calculated as in Reference (13). Companies should determine
what Toxicity Scores and Cpu Scores are acceptable in their organizations. ③ Cleaning process reliability should be demonstrated through review and
analysis of swab/rinse data (e.g., Statistical Process Control) and the history of cleaning including any cleaning failures.
FIG. 1 Example Flow Diagram for Initial Screening of Candidates for Visual Inspection
´1
E3263 − 22
5.2.2 The design of the equipment/device has an impact on 5.3.4.3 The VDI can be used to screen residues for potential
its inspection. Equipment/device design should be considered candidates for VI. Companies should consider a VDI that is
as one element of the Risk Assessment (Hazard Identification) valid and provides assurance that residues can be reliably and
taking into consideration the ability of the inspector to inspect consistently detected. An minimum limit for the VDI of –1 is
the equipment/device easily and adequately. suggested. Values closer to 0 may be acceptable if statistically
justified. If all the residues have a VDI of less than –1, then all
5.2.2.1 Based on their design, some equipment may not be
of them are appropriate candidates for VI. If all the residues
appropriate for VI.
have a VDI of greater than –1 then none of them would be
5.2.2.2 When satisfactory cleaning results cannot be repro-
appropriate for VI. If the VRLs of the residues are unknown, a
ducibly achieved because of limitations in the equipment/
simple screening of the calculated MSSRs using a very
device, the design of the equipment/device may need to be
conservative, "worst case" VRL of 10 μg/cm could be used. 10
modified or replaced before VI can be considered. If the
μg/cm is one of the highest VRLs reported in the literature (7)
equipment/device cannot be modified or replaced, then VI is
and residues that had a VDI less than –1 would be very strong
inappropriate.
candidates for VI. Fig. 2 shows an evaluation of eleven (11)
5.2.3 The history of cleanings with an evaluation of the
drugs using the VDI for determining whether any of them are
historical cleaning data (along with any deviations,
suitable for visual inspection.
investigations, and corrective actions) should be reviewed and
5.3.4.4 It should be understood that 10 μg/cm is a “worst
the products selected for using VI only justified if the risk
case” value for the VRL and most residues should have VRLs
assessment provides a valid basis.
that are much lower. This exercise can be repeated using other
5.2.4 If the initial criteria in 5.2 have been met and
2 2
VRLs (e.g., 5 μg/cm , 1 μg/cm , etc.) to determine what VRL
documented as part of the risk assessment, then the following
levels would be required to support VI for the residues. Such
steps are required next to demonstrate that VI can be relied on
screenings can provide guidance on which product are accept-
for determining the cleanliness of the equipment at the residue
able candidates for visual inspection. (See Appendix X1 for a
limit justified by the HBEL as required in 5.2.1.3.
flow diagram of this screening process.)
5.3 Calculation of MSSR:
5.4 Viewing (Lighting) Conditions:
5.3.1 The MSSR for each product shall be calculated and is
5.4.1 The effect of light and lighting levels on the visual
compared with the VRL. The VRL shall be below the MSSR
inspection should be known and understood from the qualifi-
for visual inspection to be acceptable for that product.
cation studies throughout the lifecycle process of the cleaning
5.3.1.1 The MSSR is derived from the HBEL, which is
validation program.
substance-specific dose that is unlikely to cause an adverse
5.4.2 VI shall be performed under specified conditions. See
effect if an individual is exposed at or below this dose every
Note 1.
day for a lifetime. Therefore, the MSSR is a residue level that
is safe.
NOTE 1—Experiments have shown that light levels, viewing angles, and
distances are not necessarily critical parameters (16). The human eye is
5.3.2 The MSSR, expressed in mass units per surface area
2 capable of rapid adaptation to changing light levels over a very wide range
(for example, μg/cm ), is calculated using (Guide E3106):
of intensities, and the eye adapts to minor differences in light levels almost
instantaneously and unnoticeably (17). Therefore, minor changes in light
MSC
MSSR 5 (1)
levels, distance, or the angle of viewing during inspection may have little
TSA
impact on the ability to inspect successfully. Some studies have been
performed showing no differences in inspection when light levels are
where:
between 200–1400 lux (8). These levels are typical of standard indoor
MSSR = maximum safe surface residue (on shared equip-
lighting of 500–1000 lux (EN-12464).
ment surfaces or the medical device),
5.4.2.1 Examples of inspection conditions may be between
MSC = maximum safe carryover, and
light level of >X, viewing angles of between A and B, and
TSA = total surface area (of shared equipment surfaces or
distances of the medical device).
5.4.3 Qualification studies are best performed in the manu-
5.3.3 The MSSRs for all process residues identified as
facturing or inspection areas under the actual conditions of use
hazards (Guide E3106) should be determined.
if the manufacturing situation as long as there would be no
5.3.4 The acceptability of residues for VI can be measured
impact on the production environment or product quality.
by using the Visual Detection Index (VDI) (15).
Other areas used for the qualification study shall have the
5.3.4.1 The VDI is determined from the ratio of the MSSR
lighting and light levels representative of the manufacturing or
of the residue and the VRL of that residue. By taking the log of
inspection areas where the VI is performed.
this ratio, a value is obtained that equals 0 when the values of
5.4.4 Light levels should be determined for the areas of
the MSSR and VRL are equal, becomes negative when the
operation and the area where the qualification is performed to
VRL is lower than the MSSR, and positive when it is higher
confirm they are equivalent using a light meter capable of
(15).
measuring between 200 to 1400 lux.
5.3.4.2 The VDI is calculated as shown:
5.4.5 The use of ultraviolet (UV) light in the qualification
VRL
studies to enhance the visibility of residues may be of benefit
VDI 5 log (2)
MSSR as many compounds fluoresce under UV light and this should
´1
E3263 − 22
NOTE 1—In this example, the MSSRs for eleven (11) drugs have been calculated based on their HBELs and compared to a VRL of 10 μg/cm . Based
on this conservative VRL, Drugs 1 to 4 would not be considered candidates for visual inspection studies. Drug 5 to 7 have VDIs of 0 to –0.778 and would
also not be considered. Drugs 8 to 11 have VDIs greater than –1 and would be considered appropriate candidates. (Note: if actual VRLs are lower (e.g.,
1 mcg/cm ) then more drugs may be candidates).
FIG. 2 Using the VDI as a Screening Tool for Candidates for Visual Inspection
be explored where possible. If UV is used in the qualification section provides an example of an approach that can be used to
study it must become part of the inspection procedure.
select surfaces. Other robust approaches may be acceptable if
justified.
5.5 Selection of Surfaces for the Qualification Study:
5.5.1 Surfaces used for qualification studies should be 5.5.2.1 A solution of a compound/product/residue is spiked
actual equipment or devices or where this is not possible onto multiple surrogate surfaces (for example, different mate-
surrogates representative of the actual equipment or devices
rials of construction), which are then put in order by multiple
may be used. There are equipment or devices that are not
experienced inspectors from the “hardest-to-see surfaces” to
appropriate for visual inspection based on their design. Also
the “easiest-to-see surfaces.” The spiked surrogate surface that
where the condition of equipment in use for production may
has the highest probability of being the “hardest-to-see sur-
affect the visual detection of residues (e.g., stains, scratches)
face” is then chosen for the qualification of VI studies. Any
either the surrogate should be of an equivalent quality or that
compound can be used for this study (Fig. 3 and Table 1 for an
piece of equipment should be repaired or replaced. In general,
example).
such equipment may not be appropriate for visual inspection.
5.5.2.2 If no one surrogate surface has a higher probability
5.5.2 Spiking studies can be used to screen materials of
than the other surrogate surfaces, then any surrogate surface
construction for the “hardest-to-see surfaces” to determine the
may be chosen for the qualification of VI studies, and in these
appropriate number of qualifications of operators/inspectors
cases, the most common surrogate surface may be chosen (e.g.,
that need to be performed where multiple compound/products/
316L Stainless Steel with a #4 Finish).
devices are being manufactured on common equipment. This
´1
E3263 − 22
NOTE 1—Coupons of seven (7) different MoCs are spiked with a compound/product. Coupons are presented to inspectors in a random order and the
inspectors rank them from “easiest to see” (1) to “hardest-to-see” (7).
FIG. 3 Selection of Hardest-to-See Material of Construction
TABLE 1 Inspector Rankings of MoCs for the Difficulty of Seeing
5.6 Selection of Products for the Qualification Study:
Residues (Ranking: 7 = Hardest / 1 = Easiest)
5.6.1 Spiking and visual ranging studies are used to screen
compounds/products/residues for the “hardest-to-see
NOTE 1—In this example, nine out of ten inspectors scored Coupon D
(probability = 0.9) and only one out of ten inspectors scored Coupon E
compounds/products” to determine the appropriate number of
(probability = 0.1) as the hardest-to-see of the seven coupon MoCs. None
qualifications of operators/inspectors that need to be per-
of the other coupons were ranked as the hardest-to-see (probability = 0.0).
formed.
The residue on Coupon D is, therefore, selected for the VRL determination
5.6.1.1 For low HBEL products where their MSSRs are also
as the hardest-to-see MoC.
low, there should also be a sufficient margin of safety to allow
Insp Insp Insp Insp Insp Insp Insp Insp Insp Insp
MoC Rank Prob
their exclusion from qualification studies otherwise these
1 2 3 4 5 6 7 8 9 10
A 5 6 6 5 6 6 6 5 6 6 6 0.0
compounds would still require qualification studies. Where
B 2 2 2 2 2 1 2 2 2 2 2 0.0
multiple low HBEL products are not visibly different from
C 1 1 1 1 1 2 1 1 1 1 1 0.0
D 7 7 7 6 7 7 7 7 7 7 7 0.9 each other and the data from visual ranging studies indicate
E 6 5 5 7 5 5 5 6 5 5 5 0.1
that their VRLs are likely to be near their MSSRs then these
F 4 3 3 4 3 3 3 3 3 3 3 0.0
products should have the appropriate number of qualification
G 5 4 4 3 4 4 4 4 4 4 4 0.0
studies performed based on the sound risk management and
knowledge management.
5.6.2 Solutions of the compounds/products/residues at the
5.5.2.3 When there are many different materials of construc-
same concentration are spiked onto the “hardest-to-see sur-
tion because of minor parts (for example, gasket materials and
face” surrogate surfaces/devices and presented to the inspec-
so forth), these may be eliminated from these studies if a risk
tors in a random order. The surrogate surfaces are then put in
assessment shows that their surfaces do not pose a significant
order by multiple trained inspectors from the “hardest-to-see
risk for VI and demonstrates that they do not make a significant
compound(s)/product(s)” to the “easiest-to-see compound(s)/
contribution to residue levels in a batch. Materials contacting
product(s)” and ranked as a part of the ongoing qualification
unit doses may not be suitable for such an exception.
program. The spiked surrogate surface(s)/device(s) that has the
5.5.2.4 Materials of construction with known surface prop-
highest probability of being the “hardest-to-see compound(s)/
erties in which the contrast between the surfaces and the
product(s)” is then chosen for the qualification of VI studies
residues make them easy to see (e.g., during screening studies)
(Fig. 4 and Table 2 for an example).
may also be excluded from these studies if documented in the
5.6.3 If no one compound/product/residue has a higher
risk assessment.
probability than the other compounds/products/residues, then
5.5.2.5 Materials of construction with known surface prop-
any compounds/products/residues may be chosen for the quali-
erties in which the contrast between the surfaces and the
fication of VI studies.
residues make them difficult to see (for example, a white
5.6.4 The selection of “hardest-to-see compound(s)/
residue on a white matte surface) may not be appropriate for
product(s)” may be performed before the selection of “hardest-
qualification studies of VI. In these circumstances reliance on
to-see surface(s)” depending on the risk assessment.
visual inspection as the sole means of determining cleanliness
may not be appropriate. 5.7 Preparation of Surrogate Surfaces or Devices:
5.5.2.6 The selection of “hardest-to-see surface(s)” may be 5.7.1 Surrogate surfaces (for example, coupons, devices)
performed before the selection of the “hardest-to-see shall be prepared from the same materials of construction with
compound(s)/product(s)” depending on the risk assessment. similar finishes, coatings, and so forth as the equipment or
NOTE 1—Coupons of the hardest-to-see MoC are spiked with the different compounds/products and then ranked from "easiest to see" to
"hardest-to-see" compound/product.
FIG. 4 Example for Selection of Hardest-to-See Residues
´1
E3263 − 22
TABLE 2 Inspector Rankings of Drugs for the Difficulty of Seeing
with the solvent (for example, methanol) that is used for
Residues (Ranking: 7 = Hardest / 1 = Easiest)
cleaning would be appropriate. See Note 2.
NOTE 1—In this example, 5 out of 10 inspectors scored Coupon C
NOTE 2—Evaporative drying has been studied for many solvents,
(probability = 0.5) and 5 out of 10 inspectors scored Coupon E
including water, and there are significant differences in the deposition
(probability = 0.5) as the hardest-to-see of the seven drugs. None of the
patterns of residues depending on the solvent (18). Consequently, the
other drugs were ranked as the hardest-to-see (probability = 0.0). As the
improper preparation of surrogate surfaces may lead to erroneous conclu-
results were equivocal, the residues of both Drug C and Drug E were both
sions. The use of solvents (for example, methanol) to deposit the
selected for the VRL determination as the hardest-to-see residues.
compounds that are cleaned under aqueous condition or drying them or
Insp Insp Insp Insp Insp Insp Insp Insp Insp Insp
both under conditions not encountered in operations (for example, under
Drug Rank Prob
1 2 3 4 5 6 7 8 9 10
a nitrogen stream) are not recommended.
A 6 6 5 5 5 5 6 6 6 5 5 0.0
B 2 2 2 2 2 2 2 2 2 2 2 0.0 5.7.7.2 Preparation of Surrogate Surfaces Specific to Medi-
C 7 5 7 6 7 6 5 6 7 7 7 0.5
cal Devices—Medical devices are very diverse regarding
D 1 1 1 1 1 1 1 1 1 1 1 0.0
materials, surface finish, design, construction and manufactur-
E 5 7 6 7 6 7 7 7 5 6 7 0.5
F 4 3 3 4 3 3 3 3 3 3 3 0.0
ing processes. The surrogate surface shall be representative of
G 5 4 4 3 4 4 4 4 4 4 4 0.0
the key features of the finished device for VI. Special consid-
eration shall be given to (1) material composition; (2) surface
finish based on manufacturing processes; and (3) design
features such as back tapers or bore holes. Therefore, actual
parts or three-dimensional surrogate parts are often used.
device surfaces the VI qualification is being performed on
Surrogate parts are typically used when actual parts are too
(Practice G121). The type of surface finish or coating or both
expensive or not readily available for use.
shall be identified by the user company of the equipment/
(1) The process residue (e.g. a metal working fluid, a
device.
polishing abrasive, or a cleaning agent) shall be applied in
5.7.2 Surrogate surfaces shall be thoroughly cleaned and
well-defined amounts to the surrogate surface and the surface
examined before preparation to ensure the surrogate surfaces
area covered by the applied process residue shall be determined
are representative of the worst case for the manufacturing
in order to calculate the applied amount per surface area
equipment or device surfaces and relevant for the qualification.
(μg/cm ). Dependent on the device design, it may be important
5.7.3 Clean gloves should be worn when handling surrogate
to determine the VRL for several design features and/or on
surfaces to protect from contamination from fingerprints.
various surface finishes to establish the VRL for the device.
5.7.4 For VI qualification studies to be valid, the surrogate
Each selection of the various points mentioned above has to be
surfaces shall be prepared in a manner that leaves a consistent
justified in a risk assessment.
level of the target residue (μg/cm ) on the surrogate surfaces.
5.7.8 Surrogate surfaces shall be individually marked so
The actual level of μg/cm of residue on the surrogate surface
inspectors may easily identify them during the qualification
should be determined.
studies. If numbering is used to mark, random numbers should
5.7.5 Surrogate surfaces should be prepared in a manner that
be assigned to minimize the likelihood that inspectors may
simulates the conditions the residues will encounter after
remember prior evaluations.
cleaning. For example, residues may have obvious edges or
5.7.9 Surrogate surfaces should be uniquely marked (such
appear as a continuous layer which can affect the determination
as labeled as to the material of construction, for example, 316L
of the Visual Residue Limit. The effects of cleaning agents,
SS/#4 Finish or with the date of manufacture or both) to
drying conditions, etc. and historical information on residues
provide traceability, avoid mix-ups, and avoid invalidating the
should be understood and considered.
qualification studies.
5.7.6 Ensure there is a balanced mix of clean and dirty/
5.7.10 After preparation, all surrogate surfaces should be
soiled surrogate surfaces.
examined to ensure they have been prepared correctly, includ-
5.7.6.1 Use a random number generator to select one half of
ing verifying that the blank surrogate surfaces do not have
the surrogate surfaces for use as standards for clean surfaces
unintended stains, scratches, or fingerprints that may mislead
and the remaining surrogate surfaces for use as standards for
the inspectors and invalidate the qualification study.
dirty or soiled surfaces.
5.7.7 The dirty/soiled coupons or devices of the Material of 5.7.11 Digital images of the spiked surrogate surfaces after
preparation and before use shall be taken and stored for
Construction determined from 5.5 should be spiked with a
solution of the product or residue determined from 5.6 and reference as a baseline condition of the surrogate surfaces for
comparison and evaluation after a period of use. Such images
allowed to dry.
5.7.7.1 Preparation of Surrogate Surfaces Specific to Phar- should be representative of the appearance of the residue on the
surface. Before performing a study, the surrogate surfaces
maceutical Products—For example, an API is dissolved in
purified water, spiked onto the surrogate surface, and then should be examined. If a surrogate surface’s appearance is no
longer representative in comparison to the original
dried in an oven at 90 °C. This procedure would simulate the
actual conditions in an operation involving a final purified photographs, it is not appropriate to use them in the study.
Alternatively, standards may be used for reference.
water rinse on hot equipment surfaces in which API residue
may dry quickly on the equipment. If the equipment is 5.7.11.1 It should be noted that digital data and the compa-
manually cleaned at room temperature, then spiking should rability of digital photography against physical specimens has
simulate this condition. For API manufacturers, deposition limitations due to various factors such as lighting type,
´1
E3263 − 22
configuration, and overall level of lighting in the inspection dilutions) are prepared and a fixed volume of each dilution
area and the settings on the monitor used to display the image. (e.g., 1 mL) is spiked on individual surrogates and allowed to
dry.
5.7.12 If one product is used as representative of a group of
products in a qualification study, the residues of the other
7.2 This method is performed on the selected surrogate
compounds/products shall be equivalent in appearance (for
surfaces or devices spiked with known amounts of the selected
example, a white residue would not be equivalent to a blue
compounds/products/residues spiked at a number of concen-
residue).
trations approximately in the expected range of the VRL.
5.8 Surrogate Surface Storage and Handling: Trained inspectors examine the surfaces under controlled
viewing conditions (for example, light, viewing angle, and
5.8.1 Surrogate surfaces can be easily damaged or contami-
viewing distance) for the presence of residue. The lowest level
nated and this could affect the results of the study so storage,
of residue that is detected by all inspectors is then considered
handling, and maintenance of surrogate surfaces are important.
the VRL for that particular product/compound residue (EN-
5.8.2 Clean gloves should be worn when handling surrogate
12464).
surfaces to protect them from external contamination during
handling.
7.3 Statistically Derived VRLs:
5.8.3 When not in use, surrogate surfaces should be kept in
7.3.1 The objective of this VRL determination is to derive
a protective enclosure to protect from contamination or altera-
the lowest residue level that can be seen by all trained
tion of the clean or spiked surfaces during storage.
inspectors for the product/compound using statistical analysis
5.8.4 Surrogate surfaces should be examined before, and
of the spiked coupon study.
following, any qualification studies to ensure that they are free
7.3.1.1 The approach described in 7.2 results in a rough
from any residues from extraneous sources (for example, dust,
approximation of the VRL and may set the VRL significantly
fingerprints, and so forth) that might interfere with the study
higher than it should be and may not be statistically valid if the
and impact the qualification process.
numbers of inspectors are too low (10).
7.3.2 The visual residue data collected during VRL deter-
6. Inspector Training
minations are binary (clean/dirty, yes/no) and the most suitable
6.1 SOPs shall be written on how VI should be performed. statistical technique that can be applied to binary data is binary
regression, for example, using logistic or probit models. A
6.2 Inspectors performing VIs should be trained to ensure
logistic-regression-based approach has been proposed for VRL
that an appropriate inspection is performed under appropriate
determination in the literature (10).
conditions.
7.3.2.1 These techniques involve fitting a relationship be-
6.3 Inspectors need to demonstrate their ability to perform
tween the binary response and explanatory variables such as
these inspections after training. Statistical techniques, such as
spiked concentration, viewing distance, viewing angle, and
measurement systems analysis, may be used to determine the
light intensity. For modelling, a link function (for example,
effectiveness of the training. Proficiency of inspection can be
logit or probit) that transforms the expected values of the
demonstrated through attribute agreement analysis (Section 8).
response variable to values that can be modeled using linear
regression is used. See Note 3.
6.4 Critical parameters and risks determined during the
qualification of VI should be included in the SOP and training.
NOTE 3—Because of this generalization of linear models, these models
are referred to as generalized linear models.
6.5 It is suggested that simulated residues should also be
compared against appropriate controls for studying the ability
7.3.2.2 The regression parameters for the fitted model are
of inspectors to differentiate between process residues and
estimated using maximum likelihood method. See Note 4.
“false positives” such as those caused by watermarks, surface
NOTE 4—Maximum likelihood estimation is a technique used for
defects, or uneven surface finishing, and so forth, which are
estimating the parameters of a statistical model. In this technique, the
irrelevant for the inspection.
model parameters (namely, maximum likelihood estimates) are obtained
by maximizing the likelihood or log-likelihood functions (see equations).
6.6 Inspectors should have eye exams on a defined schedule
The parameter estimates are computed iteratively using algorithms such as
based on the level of risk. This requirement should be part of
Newton-Raphson or Fisher-scoring. For simple logistic regression, the
the risk assessment. ((19), USP Chapter <1790>).
likelihood function is given by:
6.7 If supplemental tools (such as boroscopes, UV lights, n
y 12y
i i
and so forth) for performing VI are used, inspectors shall be L~β , β ! 5 p~x ! @1 2 p ~x !#
0 1 ) i i
i51
trained on their use.
and the log-likelihood is given by:
n
7. Determination of Visual Residue Limits
LL β , β 5 y log p x 1 1 2 y log 1 2 p x
~ ! @ ~ ~ !! ~ ! ~ ~ !!#
0 1 ) i i i i
i51
7.1 This section describes statistically valid methods for where:
x x , and y y = values of independent variable and binary response
determining the lowest spiked residue level for establishing 1– n 1– n
variable, respectively,
Vis
...