This is a ready-to-use checklist to work through immediately when a sterilizing-grade filter fails its pre-use or post-use integrity test. A failed integrity test is a sterility-assurance event, and the sequence matters: preserve evidence first, rule out test-system causes with documented evidence rather than assumption, and never let a clean retest alone erase a failing result. Work each item, mark it Done, Gap, or N/A, and do not proceed to disposition with an open Gap in sections 1 through 3. Replace every <<FILL: ...>> placeholder and route it through document control. A filled specimen follows. This is general guidance to adapt and verify, not legal or regulatory advice.
Document control header
| Field | Entry |
|---|---|
| Checklist title | Sterilizing Filter Integrity Test Failure Investigation |
| Document number | <<FILL: CHK-ID>> |
| Batch / lot number | <<FILL>> |
| Filter position (pre-use / post-use, single / redundant configuration) | <<FILL>> |
| Deviation record reference | <<FILL: DEV-ID>> |
| Investigation start date/time | <<FILL>> |
1. Immediate actions and evidence preservation
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 1 | Batch quarantined; no further processing, filling, or release action taken | ||
| 2 | Filter assembly NOT broken down; physical filter preserved for possible retest or autopsy | ||
| 3 | Original failing result captured and retained (printout or electronic record), not deleted or overwritten | ||
| 4 | Instrument, method, and recipe used for the failing test identified and recorded | ||
| 5 | System owner and Quality Assurance notified the same working day | ||
| 6 | Deviation record opened per <<FILL: deviation SOP ID>> |
2. Test-system cause investigation (before any retest)
Rule out a false failure with evidence, not assertion. Every “yes” needs a cited record.
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 7 | Wetting confirmed complete and with the correct fluid (water vs product-wetted recipe) | ||
| 8 | Correct integrity-test limit and recipe confirmed for this filter, wetting fluid, and configuration | ||
| 9 | Test pressure, temperature, and hold time confirmed within the validated method | ||
| 10 | Test rig checked for leaks, loose fittings, or incomplete valve closure independent of the filter | ||
| 11 | Residual product or foam in the test path ruled out as a cause | ||
| 12 | Instrument calibration status confirmed current at time of test | ||
| 13 | A plausible, evidenced test-system cause identified, or none found |
3. Retest determination
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 14 | Retest justification is documented, specific, and tied to an item in section 2, not a general “assumed wetting issue” | ||
| 15 | Retest performed under a procedure, with the original failing result still visible and retained in the record | ||
| 16 | Retest result recorded regardless of outcome (pass or fail) | ||
| 17 | If retest passes: conclusion states the evidenced cause, not “retested, now passes” alone | ||
| 18 | If retest fails again: proceed to section 4 as a genuine failure |
4. Genuine failure impact assessment
Complete this section if no evidenced test-system cause was found, or the retest also failed.
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 19 | Failure timing identified: pre-use (before batch contact) or post-use (after batch filtered) | ||
| 20 | If pre-use: batch confirmed not yet filtered through the failed filter | ||
| 21 | If post-use on a single, non-redundant filter: sterility of the affected batch portion cannot be assured | ||
| 22 | If redundant configuration: status of the other filter (pass/fail) confirmed and validated decision logic applied | ||
| 23 | Extent of affected material determined (whole batch, or a defined portion) | ||
| 24 | Any related batches using the same filter lot, assembly, or test rig identified for scope assessment |
5. Batch disposition
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 25 | Disposition recommendation documented: release, reject, or reprocess (if a validated path exists) | ||
| 26 | Disposition basis ties directly to the impact assessment in section 4, not to the retest alone | ||
| 27 | Quality Assurance disposition decision recorded and signed | ||
| 28 | Qualified Person / release authority sign-off obtained where applicable |
6. Root cause and CAPA
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 29 | Root cause investigated for a genuine filter failure (manufacturing defect, sterilization stress, over-pressure, chemical compatibility degradation, installation damage) | ||
| 30 | Failed filter retained or sent for autopsy where root cause is not otherwise determinable | ||
| 31 | CAPA opened if root cause points to a systemic issue (sizing, handling, sterilization cycle, supplier) | ||
| 32 | Recurring-failure trend checked against prior investigations for this filter type or line |
7. Documentation and closure
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 33 | Full investigation timeline documented: failure, evidence review, retest (if any), disposition, CAPA | ||
| 34 | Deviation record closed with disposition, root cause, and CAPA references | ||
| 35 | Validated state confirmed as maintained, or a revalidation/requalification trigger raised | ||
| 36 | Investigation reviewed and approved by Quality Assurance |
8. References
EU GMP Annex 1, Manufacture of Sterile Medicinal Products (2022), section 8, filtration and integrity testing. FDA Guidance for Industry, Sterile Drug Products Produced by Aseptic Processing, Current Good Manufacturing Practice (2004). 21 CFR 211.192 (production record review) and 211.22 (quality control unit responsibilities).
Confirm the current version and clause numbers of each reference before issue.
9. Signoff
| Role | Name | Signature | Date |
|---|---|---|---|
| Investigation lead | <<FILL>> | ||
| Filtration / Validation SME | <<FILL>> | ||
| Quality Assurance disposition | <<FILL>> |
Filled specimen
Illustrative excerpt for a post-use diffusive flow failure on the upstream filter of a redundant pair.
| # | Item | Done / Gap / N/A | Note |
|---|---|---|---|
| 1 | Batch quarantined | Done | Batch <<FILL>> held, no further movement |
| 2 | Filter assembly preserved | Done | Assembly left intact pending investigation team review |
| 7 | Wetting confirmed | Gap found | Upstream filter wetting time was 3 minutes against a validated 10-minute minimum |
| 13 | Evidenced test-system cause identified | Done | Incomplete wetting, tied to item 7 |
| 14 | Retest justification documented | Done | Retest authorized citing item 7, re-wetted per validated procedure |
| 16 | Retest result recorded | Done | Retest diffusive flow 9 mL/min, within the 15 mL/min limit; original 21 mL/min result retained in the record |
| 22 | Redundant filter status confirmed | Done | Downstream filter passed post-use at 7 mL/min in both original and retest cycles |
| 25 | Disposition recommendation | Done | Release recommended: downstream filter (last barrier) passed throughout; upstream failure traced to an evidenced test-system cause and confirmed by a compliant retest |
| 27 | QA disposition | Done | Approved with full investigation attached |
The batch was supported because the validated redundant-filter decision logic keyed release to the downstream filter passing, and the upstream result was traced to a specific, evidenced wetting-time deviation rather than assumed. Had the downstream filter also failed, or had no evidenced cause been found for the upstream failure, this would have been a genuine-failure path through section 4 rather than a test-system retest.
Common inspection findings this checklist prevents
- Filter assembly broken down before the investigation began, destroying evidence.
- A retest performed and passed with no documented justification for why the original result was invalid.
- The original failing result missing from the record, replaced only by the passing retest.
- A post-use failure on a single, non-redundant filter released with no impact assessment addressing sterility assurance.
- Redundant filtration treated as automatic batch support without confirming the other filter actually passed and the validated logic actually applies.
- No root cause investigation following a genuine filter failure, so the same failure mode recurs.
How to adapt this checklist
- Set your document number and link the deviation record.
- Adjust section 2’s test-system checks to match your actual integrity test method and instrument.
- If your site does not use redundant filtration, remove items 22 and the redundant-logic language from section 4 and section 5, and treat any genuine post-use failure on the single filter as grounds to reject per your quality system.
- Wire section 6 to your existing root cause and CAPA procedures rather than duplicating them here.
- Confirm every reference in section 8 against the current published version before issue.