GxP Insights

ISO 14644-3 Cleanroom Qualification: Tests, Criteria and Recovery

Operational guide to cleanroom qualification under ISO 14644-3:2019 and EU GMP Annex 1: HVAC tests, acceptance criteria, airflow studies, recovery and requalification.

A Aldo Xhango 6 min read
✓ Official sources and references ✓ Practical approach ✓ For pharmaceutical professionals
GUIDEGXP · PRACTICAL GMP INSIGHTS
Qualifica Cleanroom (ISO 14644-3): Guida ai test essenziali

Short answer: ISO 14644-3:2019 does not itself assign an air-cleanliness class and it is not a complete GMP qualification protocol. It provides test methods for demonstrating how a cleanroom, clean zone or clean-air device performs against its intended use. In pharmaceutical applications, test selection and acceptance criteria should be approved before execution and linked to the URS, design, risk assessment, Contamination Control Strategy (CCS), qualification protocol and applicable requirements.

Qualification should therefore produce more than a particle count or an HVAC certificate. It should create a defensible evidence chain: requirement → risk → test → acceptance criterion → result → deviation/CAPA → qualified state.

Practical rule
Do not ask “which tests does ISO 14644-3 always require?” Ask “which performance characteristics must be demonstrated for this installation, in this occupancy state and for this process?”

ISO 14644-1, -2 and -3: three different purposes

These parts are often conflated, but they answer different questions:

  • ISO 14644-1: how to classify air cleanliness by particle concentration.
  • ISO 14644-2: how to establish monitoring that demonstrates continued performance against the classification.
  • ISO 14644-3:2019: test methods for verifying performance characteristics of cleanrooms and clean air.

ISO 14644-3:2005 is withdrawn. New protocols and document revisions should reference the 2019 edition, corrected in 2020, while controlled copies and applicable contractual requirements should always be checked within the company document system.

How it connects to EU GMP Annex 1 and Annex 15

For sterile manufacture, EU GMP Annex 1 clearly differentiates cleanroom qualification from routine operational environmental monitoring. Qualification should follow Annex 15 principles and, where relevant to the design and operation, can include installed filter-system leakage and integrity, airflow volume and velocity, pressure differences, airflow direction and visualisation, microbial contamination, temperature, relative humidity, recovery and containment leakage.

The ISO 14644 series is therefore an important methodological reference, but it does not replace the GMP assessment. The protocol should explain why a test is included or excluded, which conditions represent worst case, and how the result supports the CCS and intended use.

Before testing: what the protocol should define

A technically correct test can still be weak in an inspection if the acceptance criterion was selected after the result was known. Before execution, define at least:

  1. scope and boundaries: rooms, zones, filters, terminals, barrier systems and exclusions;
  2. occupancy state: as-built, at-rest or operational, including equipment, doors, utilities and personnel configuration;
  3. prerequisites: completed commissioning, HVAC balancing, valid calibration, maintenance and cleaning status;
  4. method and measurement locations: instruments, uncertainty, traceability, sampling plan and raw-data controls;
  5. acceptance criteria: derived from the URS, design, applicable standards, Annex 1, risk assessment and intended use;
  6. event handling: deviations, repeats, invalidations, CAPA and report approval.

Practical qualification-test matrix

TestWhat it demonstratesEvidence to retainSource of the criterion
Installed-filter leakage/integrityNo unacceptable leakage through filter, seals, frame or installationFilter map, aerosol and instruments, scan, results and repairsProtocol, filter grade, applicable method and approved specification
Airflow volume and velocityHVAC capacity and protection at the working positionPoint grid, conditions, values, mean/variation and uncertaintyURS, design and, for UDAF, scientific/CCS justification
Pressure differencesAirflow direction and separation between areasStable values, door/configuration status, setpoints, alarms and delaysCCS, zoning, URS and approved limits
Airflow direction/visualisationNo ingress or contamination path towards critical zonesIdentified videos, scenarios, simulated interventions, observations and conclusionsProcess, CCS, Annex 1 and protocol
RecoveryTime to restore the defined cleanliness stateInitial condition, curve, endpoint, time and repeatabilityProcedure, intended use and qualified value
Temperature, RH and containmentEnvironmental conditions and leakage relevant to product/processMapping, operating conditions, trends and leakage test where applicableURS, process, safety and risk assessment

1. Installed HEPA/ULPA filter integrity

The in-situ test verifies the installed filtration system after transport and installation. It is not the same as the manufacturer’s filter-efficiency certificate. Execution should cover the filter medium, seals, frame and potential bypass paths, using an aerosol, instrument and scan rate consistent with the approved method.

There is no single percentage limit that should be applied indiscriminately to every filter and installation. Define the criterion in the protocol according to filter grade, method, design specification and applicable requirements. Any repair should be traceable, assessed and followed by the required retest.

2. Airflow volume, velocity and air-change rate

For unidirectional systems, measurements should demonstrate adequate protection at the working position, not only close to the filter face. EU GMP Annex 1 gives a homogeneous 0.36–0.54 m/s guidance value at the working position for UDAF systems unless otherwise scientifically justified in the CCS; it should not be turned into a universal specification for every cleanroom.

In non-unidirectional rooms, air changes per hour are a design and balancing parameter, not a fixed GMP acceptance criterion. They should support classification, recovery, heat load, pressure control and contamination control under the defined operational conditions.

3. Pressure differences and alarms

Pressure supports the intended direction of air movement between areas of different criticality. The appropriate value depends on zoning, doors, pass boxes, extraction, containment and process. The current Annex 1 requires setpoints and criticality to be considered in the CCS; differences identified as critical should be continuously monitored and recorded, with warnings and a defined response.

Qualification should therefore challenge more than a closed-door reading: consider stability, direction, alarms, justified delays, loss of air supply and, where relevant, recovery after door opening.

4. Airflow visualisation: the smoke study must represent the process

Visualisation demonstrates that air does not move from less clean to cleaner areas or travel over operators and equipment before reaching exposed product. For aseptic processing, generic or at-rest-only scenarios provide limited evidence.

Studies should cover at-rest and in-operation conditions, including routine activities and relevant interventions. Videos should be identified and retained; the report should link each scenario to the risk and state the release point, configuration, operators, conclusion and corrective actions. The visualisation medium itself should not create an uncontrolled risk to the environment, equipment or product.

5. Recovery: qualify a time, do not chase a number

The recovery test measures the time required to return to a defined cleanliness level after a controlled disturbance. EU GMP Annex 1 gives a guidance value of less than 20 minutes for the clean-up period, but the qualified time should be determined, documented and embedded in site procedures.

The protocol should state the initial condition, challenge method, measurement locations, endpoint, calculation and repetitions. A longer-than-expected result does not automatically prove an HVAC failure: it calls for investigation of balancing, loads, infiltration, dead zones, configuration and test method.

The Pragmatic GMP

Every Monday: a free English briefing on GMP decisions, regulatory change, operational controls and signals from the pharmaceutical sector.

Subscribe free

When to requalify

Calendar frequency should not be the only trigger. In addition to planned requalification, an impact assessment should determine which tests need repeating after events such as filter replacement, HVAC changes, new layouts or equipment, building work, pressure excursions, loss of control, adverse trends or process changes.

Repeating every test is not always necessary, but an overly narrow retest can leave risk unassessed. Document the decision and link it to change control, the CCS and system knowledge.

QA review checklist

  • Does the protocol distinguish classification, qualification and monitoring?
  • Are the standard edition, method, instruments and calibration identified?
  • Are occupancy state and operational configuration reproducible?
  • Were acceptance criteria approved before testing and traced to requirements/risks?
  • Do measurement locations cover critical positions and worst case?
  • Are raw data, video and electronic files attributable and protected?
  • Do repeats and invalidations have a documented rationale?
  • Do deviations assess impact on product, CCS and qualified state?
  • Does the report clearly conclude what is qualified and within which limits?

Frequently asked questions

Does ISO 14644-3 classify a cleanroom?

No. Particle classification is covered by ISO 14644-1. Part 3 provides test methods for performance characteristics.

Is the HEPA leakage limit always 0.01%?

No. It should not be applied as a universal limit without confirming relevance. Criterion, aerosol, instrument and method should match the filter, specification and approved protocol.

Is there a fixed GMP air-change rate?

No. The value should support intended use and demonstrated performance: classification, recovery, pressure cascade, loads and contamination control.

Must airflow visualisation be recorded?

For EU GMP Annex 1 application, video recordings of airflow-pattern studies should be retained. They should allow the scenario, configuration and conclusion to be understood.

Must recovery always be below 20 minutes?

Annex 1 gives less than 20 minutes as a guidance value for the clean-up period. The site should determine, document and follow its qualified time.

Related guidance and tools

Primary sources

Editorial note: this article summarises requirements and operational principles without reproducing normative procedures. Always consult controlled copies of the standards and requirements applicable to your installation.

THE PRAGMATIC GMP · EVERY MONDAY

The GMP topics that matter, in 7 minutes.

One GMP topic, one real-world example and one practical action, based on official sources and inspection trends.
Discover The Pragmatic GMP