1Acceptance criteria by type of medium.
Test “each batch of ready-prepared medium and each batch of medium, prepared either from dehydrated medium or from the ingredients”.¹ What each kind must show:
| Medium and use | Inoculum | Incubation | Passes when | Source |
|---|---|---|---|---|
| Solid, for counts | Not more than 100 CFU | Table below | Within a factor of 2 of a standardised inoculum; fresh inoculum: comparable to an approved batch | JP 4.05 I 3.3¹ |
| Liquid, for counts | Not more than 100 CFU | Table below | Clearly visible growth comparable to an approved batch | JP 4.05 I 3.3¹ |
| Specified-organism media | Not more than 100 CFU | Not more than the shortest time in the test: 18 h where it says 18 to 24 h | Growth comparable to an approved batch | JP 4.05 II 2.3; USP FAQ¹² |
| Inhibitory properties | At least 100 CFU | Not less than the longest time: 24 h where it says 18 to 24 h | No growth of the test organism | JP 4.05 II 2.3; USP FAQ¹² |
| Indicative properties | Not more than 100 CFU | Within the test’s range | Colonies and reactions comparable to an approved batch | JP 4.05 II 2.3¹ |
| Sterility test | Not more than 100 CFU | Bacteria 3 days at most, fungi 5 | Clearly visible growth | JP 4.06 3.2³ |
| Monitoring and APS | APS: 100 CFU or less (FDA: under 100). Monitoring: the procedure sets it | Set in the procedure | Reference organisms and local isolates grow | Annex 1 9.36, 10.9; PI 058-1; FDA 2004⁴⁵⁶ |
The chapters set the organisms and conditions:
| Organism | Strain, for example | Counting media | Conditions | Sterility medium |
|---|---|---|---|---|
| S. aureus | ATCC 6538 | TSA and TSB | 30 to 35 °C, 3 days at most | Fluid thioglycollate |
| P. aeruginosa | ATCC 9027 | TSA and TSB | 30 to 35 °C, 3 days at most | Fluid thioglycollate |
| B. subtilis | ATCC 6633 | TSA and TSB | 30 to 35 °C, 3 days at most | Soya-bean casein digest |
| C. albicans | ATCC 10231 | TSA; SDA | TSA 30 to 35 °C, SDA 20 to 25 °C; 5 days at most | Soya-bean casein digest |
| A. brasiliensis | ATCC 16404 | TSA; SDA | TSA 30 to 35 °C, SDA 20 to 25 °C; 5 days at most | Soya-bean casein digest |
| C. sporogenes | ATCC 19404 or 11437 | Not used | Not used | Fluid thioglycollate |
The strain numbers come from JP Tables 4.05-I-1 and 4.06-1, which list equivalent collection numbers too.¹³ Media for specified organisms are tested with the organism they are meant to find, and selective media also have inhibition pairs: MacConkey broth must grow E. coli and inhibit S. aureus; cetrimide agar must grow P. aeruginosa and inhibit E. coli.¹
These rows come from texts harmonised across USP (61, 62, 71), the European Pharmacopoeia (2.6.12, 2.6.13, 2.6.1) and JP (4.05, 4.06), which FDA’s Q4B guidances accept as interchangeable; quotes are from USP’s free chapter 61, official from 1 May 2025, and JP XVIII, the latest English edition.⁷⁸⁹¹⁰¹ EU GMP adds that “The performance of all culture media should be verified prior to use”, with an in-use shelf life that is “established, documented and scientifically justified”.¹¹
2How the factor of 2 is calculated.
The rule for solid media is one sentence: “For solid media, growth obtained must not differ by a factor greater than 2 from the calculated value for a standardized inoculum.”¹⁰ USP’s own answer reads it the same way: with 100 CFU, “acceptable counts are: 100/2 = 50 CFU to 100 x 2 = 200 CFU”, a factor “introduced to take account of the variability of the method”.¹² JP XIX, published in Japanese on 10 April 2026, spells it out: “1／2から2倍以内”, within half to twice the suspension’s value.¹³ To apply it:
- Take the calculated value: the CFU the standardised inoculum delivers to each plate, such as a stable suspension’s stated count, at most 100.
- Count the plates after incubation and take the mean.
- Divide the mean count by the calculated value.
- Accept a ratio from 0.5 to 2.0. As a percentage that is 50% to 200% recovery; as counts, half to double the calculated value. A ratio of exactly 2 does not differ by a factor greater than 2, so it passes.
The limit runs both ways: a count above double fails too, as a sign of a wrong inoculum or count. The comparator is the calculated value, not a count on another plate.
Two nearby rules are easy to confuse:
- A freshly prepared inoculum is judged differently: growth must be “comparable to that previously obtained with a previously tested and approved batch of medium”.¹⁰ The chapter gives no number. USP says the approved lot need not be run in parallel if its growth was clearly recorded and, answering the same question for chapter 62, “You can define the comparability criterion yourself”, the factor of 2 being one option.¹²² Plating one suspension on both lots with a stated ratio, as the example does, is easy to defend.
- Method suitability uses the same factor against a control without product: the mean must not differ “by a factor greater than 2 from the value of the control”, and growth reduced by more than that means modifying the procedure. The most-
probable- number method uses the control’s 95% confidence limits instead.¹
In the harmonised chapter these are the only recovery tolerances: a factor of 2 for plate counts and confidence limits for MPN. Its other numbers include counting rules, such as counting plates with fewer than 250 colonies for the total aerobic count and fewer than 50 for yeasts and moulds.¹⁰¹ A 70% or 75% minimum recovery, if a procedure uses one, is the site’s choice, not a pharmacopoeial requirement.
3Example: growth promotion of a TSA lot.
Example, all invented: the laboratory, lots and counts are illustrative.
TSA lot M26-118, prepared from dehydrated medium on 2 March 2026, is quarantined until tested. Each organism is plated in duplicate from a standardised suspension and incubated at 30 to 35 °C: the bacteria for 3 days, the fungi for 5.
| Organism | Calculated CFU | Plate counts | Mean | Ratio | Result |
|---|---|---|---|---|---|
| S. aureus | 58 | 51, 55 | 53 | 0.91 | Pass |
| P. aeruginosa | 64 | 60, 66 | 63 | 0.98 | Pass |
| B. subtilis | 47 | 41, 38 | 39.5 | 0.84 | Pass |
| C. albicans | 52 | 21, 27 | 24 | 0.46 | Fail |
| A. brasiliensis | 39 | 33, 30 | 31.5 | 0.81 | Pass |
C. albicans needed a mean between 26 and 104 (52 ÷ 2 and 52 × 2) and recovered 24. The same suspension plated on the previously approved lot M26-097 gave 49, a ratio of 0.94, so the inoculum was sound and the new lot is the suspect.
What happens next:
- The lot stays in quarantine. No sample was read on it, so nothing else is affected. Had the lot been used in parallel, as JP allows for sterility test media, every test read on it would be in scope.³
- The investigation looks for a cause in the lot. Its preparation record shows the agar was held molten for six hours against the procedure’s limit of three. JP’s guidance on media control has molten agar held at 45 to 50 °C, but “holding for a long-time should be avoided”.¹⁴ The lot is rejected.
- A replacement lot is tested in full, all five organisms, before release.
4Growth promotion record template.
The laboratory that uses the medium normally runs the test; Annex 1 expects the end user to do it for media used in sterile manufacture.⁴ The lot stays in quarantine until read, up to 3 days for bacteria and 5 for fungi, except sterility test media, which may be tested “before or in parallel with the test on the product to be examined”, at the product results’ risk.¹³ A second person reviews the record, as 21 CFR 211.194(a)(8) requires.¹⁵
| Field | Record | Requirement |
|---|---|---|
| Medium and lot | Name and lot, prepared or purchased, in-use expiry, storage. If prepared: date, dehydrated medium lot and mass, water volume, sterilisation conditions, pH after sterilisation, equipment | EU GMP 6.21, 6.23, 6.24; JP G4-2-180 1.1¹¹¹⁴ |
| Test organisms | Species, strain number, passages from the master seed (5 at most) | JP 4.05 I 3.1¹ |
| Suspension | Time prepared; used within 2 hours, or 24 hours at 2 to 8 °C | JP 4.05 I 3.1¹ |
| Inoculum | Calculated CFU per plate or portion, 100 at most | JP 4.05 I 3.3¹ |
| Incubation | Incubator, temperature, start and end times | JP Table 4.05-I-1¹ |
| Results | Counts and ratio, or visible growth; inhibition and indicative reactions | JP 4.05 I 3.3, II 2.3¹ |
| Controls | Negative control; approved-lot comparison for a fresh inoculum | JP 4.05 I 3.2, 3.3¹ |
| Sterility of medium | For sterility test media: 14 days, no growth | JP 4.06 3.1³ |
| Local isolates | For monitoring and APS media | Annex 1 10.9⁴ |
| Decision | Pass or fail, second-person review, date, release of the lot | EU GMP 6.23; 21 CFR 211.194(a)(8)¹¹¹⁵ |
| Failure | Investigation number and its conclusion | 21 CFR 211.192; FDA letter¹⁶ |
5What to do when a lot fails.
- Keep the lot out of use and list everything already read on it. FDA has faulted an investigation that did not review every batch tested on the failing media lots.¹⁶
- Check the test before blaming the medium: inoculum count, passage number, suspension age, incubator records and the storage of the reference organisms. In February 2025 FDA cited ABR Laboratory LLC for not investigating a refrigerator holding growth promotion reference organisms that was out of range for more than 24 hours, reaching 17.4 °C.¹⁷
- Find the cause; do not retest to a pass. FDA’s letter of 2 July 2025 to Exela Pharma Sciences, LLC cited a sterility test medium used “despite growth promotion testing failing twice and passing on a third test”; the investigation had found no product impact “because retesting of the media ultimately achieved passing results”.¹⁶ See OOS investigation.
- For a media fill, repeat the run. FDA’s 2004 guidance says that if growth promotion fails, any contamination found should still be investigated and “the media fill promptly repeated”, and the cause of the failure investigated.⁶ See media fill.
- For a purchased lot, tell the supplier and review its qualification.¹⁸
6Edge cases the sources settle.
- Local isolates. In the EU, monitoring and APS media need them as well as the reference organisms.⁴ FDA’s inspection programme allows environmental isolates for growth promotion; its 2004 guidance says isolates can be substituted or added, as appropriate.¹⁹⁶ PIC/S still wants the pharmacopoeial organisms for APS media.⁵
- Media fill timing. Test the media lot before use, and again after incubation: Annex 1 wants samples of the filled units inoculated “with a suitable range of reference organisms and suitably representative local isolates”.⁴ PIC/S PI 058-1, in force since 28 October 2026, wants that “wherever possible” in the actual APS containers, at 100 CFU or less, and “an inoculum in the lower range is preferable”.⁵ FDA investigators ask when it is done: “before/after filling; after incubation; etc.”.¹⁹ See media fill.
- Monitoring media. FDA asks for growth promotion on “all lots of prepared media” used for environmental monitoring, and its investigators ask whether those media are shown to detect moulds and yeasts as well as bacteria. Neither FDA nor Annex 1 sets an inoculum for them.⁶¹⁹⁴ JP’s guidance adds that Grade A and B media not sterilised after packaging should be incubated before use, “all media (100%)”, so that contamination is not carried in and false positives are avoided.¹⁴ The panels in chapters 61, 62 and 71 are for the media in those tests; USP 1116, often cited for monitoring media, is numbered between 1000 and 1999, a range USP says states no requirements.²⁰ See Annex 1 environmental monitoring.
7What inspectors ask and cite.
FDA’s programme for sterile drug inspections gives investigators these questions:¹⁹
- “Are growth promotion studies performed on each type of medium used?”
- “Are growth promotion studies conducted every time a media fill is done?”
- “What organisms are used to perform the growth promotion tests? Are any environmental organisms used?”
Its warning letter of 10 October 2024 to Analytical Food Laboratories, Inc. cited 21 CFR 211.160 for testing with one organism instead of the full panel:²¹
For example, growth promotion tests for media batches 06-19-23-3 and 06-19-23-5 were performed with only one indicator microorganism, Staphylococcus aureus, and did not include the other challenge microorganisms listed in USP <61>.
Its letter of 20 August 2025 to Chromatography Institute of America found sterility failures in a media lot the firm could not show was growth promotion tested before use, and answered the firm’s reliance on supplier certificates by asking for:¹⁸
A remediation plan for qualification of your suppliers of growth media. Include a commitment to perform growth promotion testing on all incoming media lots used for microbial and sterility testing, to verify their suitability for use.
In August 2024 FDA told Regenerative Processing Plant, LLC that its media fill data “were not reliable” without growth promotion on the media.²² For supplier certificates in general, see relying on a supplier’s CoA.
8How it runs in Seal.
In Seal, each prepared or purchased media lot carries its receipt, release, growth promotion, storage, expiry and use genealogy, and growth promotion stays connected with environmental monitoring, sterility, bioburden and organism identification to the place, activity, batch and decision it concerns (microbiology LIMS). Each monitoring sample carries the media lot it used and the incubation it went through (environmental monitoring).
For media fills, a Growth Promotion Evidence record holds the media lot, organism panel, inoculum, conditions, expected growth, observed growth, controls and approval. Growth promotion is a dependency rather than an attachment: a failed or invalid result holds the conclusion and scopes the affected media and runs (aseptic process simulation). Neil reviews method-suitability recovery evidence organism by organism against paired controls, and an absent control cannot be treated as a passing recovery result (review microbiology recovery).
References
- 1The Japanese Pharmacopoeia, Eighteenth Edition, English version, general test 4.05, Microbiological Examination of Non-sterile Products, parts I and II, Ministry of Health, Labour and Welfare, official from 7 June 2021. https:/
/ www. mhlw. go. jp/ content/ 11120000/ 000945683. pdf (read 8 November 2026) - 2FAQs: general chapter 62, Microbial Enumeration of Nonsterile Products: Tests for Specified Microorganisms, questions 7 and 8, United States Pharmacopeia. https:/
/ www. usp. org/ frequently- asked- questions/ microbial- enumeration- nonsterile- products- tests- specified- microorganisms (read 8 November 2026) - 3The Japanese Pharmacopoeia, Eighteenth Edition, English version, general test 4.06, Sterility Test, sections 2 and 3 and Table 4.06-1, Ministry of Health, Labour and Welfare, official from 7 June 2021. https:/
/ www. mhlw. go. jp/ content/ 11120000/ 000945683. pdf (read 8 November 2026) - 4EudraLex Volume 4, Annex 1, Manufacture of Sterile Medicinal Products, paragraphs 9.36, 9.45 and 10.9, European Commission, 22 August 2022. https:/
/ health. ec. europa. eu/ system/ files/ 2022- 08/ 20220825_ gmp- an1_ en_ 0. pdf (read 8 November 2026) - 5PI 058-1, Interpretation of the most important changes in the Annex 1 on the manufacture of sterile products (2022), question 9.45ii, PIC/S, in force 28 October 2026. https:/
/ picscheme. org/ docview/ 11495 (read 8 November 2026) - 6Sterile Drug Products Produced by Aseptic Processing: Current Good Manufacturing Practice, guidance for industry, sections IX.A and X.B, FDA, September 2004. https:/
/ www. fda. gov/ media/ 71026/ download (read 8 November 2026) - 7Q4B Annex 4A(R1), Microbiological Examination of Nonsterile Products: Microbial Enumeration Tests General Chapter, guidance for industry, FDA, September 2010. https:/
/ www. fda. gov/ media/ 71241/ download (read 8 November 2026) - 8Q4B Annex 4B(R1), Microbiological Examination of Nonsterile Products: Tests for Specified Microorganisms General Chapter, guidance for industry, FDA, September 2017. https:/
/ www. fda. gov/ media/ 71267/ download (read 8 November 2026) - 9Q4B Annex 8(R1), Sterility Test General Chapter, guidance for industry, FDA, September 2017. https:/
/ www. fda. gov/ media/ 73134/ download (read 8 November 2026) - 10Microbiological Examination of Nonsterile Products: Microbial Enumeration Tests, general chapter 61, Stage 4 harmonised text, official 1 May 2025, United States Pharmacopeia. https:/
/ www. usp. org/ sites/ default/ files/ usp/ document/ harmonization/ gen- method/ 20240426HSm98800. pdf (read 8 November 2026) - 11EudraLex Volume 4, Part I, Chapter 6, Quality Control, paragraphs 6.21, 6.23 and 6.24, European Commission, in operation from 1 October 2014. https:/
/ health. ec. europa. eu/ document/ download/ c74c8720- 27bf- 4252- 808f- d65a206a90bb_ en?filename=2014- 11_ vol4_ chapter_ 6. pdf (read 8 November 2026) - 12FAQs: general chapter 61, Microbial Examination of Nonsterile Products: Microbial Enumeration Tests, questions 6, 7 and 10, United States Pharmacopeia. https:/
/ www. usp. org/ frequently- asked- questions/ microbial- examination- nonsterile- products- microbial- enumeration- tests (read 8 November 2026) - 13The Japanese Pharmacopoeia, Nineteenth Edition, Japanese text, general test 4.05, section I 3.3, MHLW Notification No. 193, Ministry of Health, Labour and Welfare, 10 April 2026. https:/
/ www. mhlw. go. jp/ content/ 11120000/ 001689414. pdf (read 8 November 2026) - 14The Japanese Pharmacopoeia, Eighteenth Edition, English version, General Information G4-2-180, Control of Culture Media and Strains of Microorganisms Used for Microbial Tests, sections 1.1 to 1.3, pages 2686 to 2688, Ministry of Health, Labour and Welfare. https:/
/ www. mhlw. go. jp/ content/ 11120000/ 000904454. pdf (read 8 November 2026) - 1521 CFR 211.194(a)(8), Laboratory records, eCFR. https:/
/ www. ecfr. gov/ current/ title- 21/ section- 211. 194 (read 8 November 2026) - 16Exela Pharma Sciences, LLC, warning letter, MARCS-CMS 709859, FDA, 2 July 2025. https:/
/ www. fda. gov/ inspections- compliance- enforcement- and- criminal- investigations/ warning- letters/ exela- pharma- sciences- llc- 709859- 07022025 (read 8 November 2026) - 17ABR Laboratory LLC, warning letter, MARCS-CMS 696872, FDA, 10 February 2025. https:/
/ www. fda. gov/ inspections- compliance- enforcement- and- criminal- investigations/ warning- letters/ abr- laboratory- llc- 696872- 02102025 (read 8 November 2026) - 18Chromatography Institute of America dba Compounder’s International Analytical Laboratory, warning letter, MARCS-CMS 708944, FDA, 20 August 2025. https:/
/ www. fda. gov/ inspections- compliance- enforcement- and- criminal- investigations/ warning- letters/ chromatography- institute- america- dba- compounders- international- analytical- laboratory- 708944- 08202025 (read 8 November 2026) - 19Compliance Program 7356.002A, Sterile Drug Process Inspections, Part III and Attachment A, questions 64 to 67, 93 and 145, FDA, implemented 11 September 2015. https:/
/ www. fda. gov/ media/ 75174/ download (read 8 November 2026) - 20FAQs: identifying official text, question 6, United States Pharmacopeia. https:/
/ www. usp. org/ frequently- asked- questions/ identifying- official- text (read 8 November 2026) - 21Analytical Food Laboratories, Inc., warning letter, MARCS-CMS 687513, FDA, 10 October 2024. https:/
/ www. fda. gov/ inspections- compliance- enforcement- and- criminal- investigations/ warning- letters/ analytical- food- laboratories- inc- 687513- 10102024 (read 8 November 2026) - 22Regenerative Processing Plant, LLC, warning letter, MARCS-CMS 672154, FDA, 16 August 2024. https:/
/ www. fda. gov/ inspections- compliance- enforcement- and- criminal- investigations/ warning- letters/ regenerative- processing- plant- llc- 672154- 08162024 (read 8 November 2026)
AQuestions and answers
How do you perform a growth promotion test?
Inoculate each new batch of medium with not more than 100 CFU of each test organism, from a culture no more than five passages from the master seed. Incubate counting media for no longer than the chapter allows for that organism, 3 days for bacteria and 5 for fungi, and specified-organism media for no longer than the shortest time in the test. With a standardised inoculum on solid media, divide the mean count by the calculated value and accept 0.5 to 2.0; on liquid media, look for clearly visible growth. Record the strains, passages, suspension age and a negative control, and have a second person review it.
Do diluents and buffers need growth promotion testing?
Not as growth media. USP’s answers on chapter 61 say growth promotion is checked on agar media and nutritive broth, not on diluted broth. The requirement for diluents runs the other way: for the harmonised sterility test, FDA’s Q4B guidance says diluting and rinsing fluids should not have antibacterial or antifungal properties. Any neutraliser must be shown to work and not to harm the organisms, with a blank that has neutraliser and no product.
Can a supplier’s growth promotion certificate replace in-house testing?
It depends on the jurisdiction. Japan’s pharmacopoeia, in its general information on media control, allows periodic testing instead of every batch once the supplier’s data are shown reliable, in writing or by a site visit, and the result and expiry can be confirmed. EU Annex 1 says testing of media for sterile manufacture should normally be done by the end user, with any reliance on the supplier justified and shipping conditions considered. FDA asked a laboratory that relied on certificates to commit to testing every incoming lot used for microbial and sterility testing.
How is growth promotion tested on selective media?
Against each property the test calls for. Growth-promoting: not more than 100 CFU of the organism the medium should recover, which must grow comparably to an approved batch. Inhibitory: at least 100 CFU of an organism it should suppress, which must not grow. Indicative: the colonies or reaction it should give. USP explains the timing as the worst case: where a test says 18 to 24 hours, read growth at no more than 18 hours and inhibition at no less than 24.
What is the growth promotion test recovery formula?
Recovery is the mean count on the new medium divided by the reference value, times 100. With a standardised inoculum the reference is its calculated count and the pharmacopoeia accepts 50% to 200%, so a 60 CFU inoculum that gives a mean of 48 recovers 80% and passes. With a freshly prepared inoculum the reference is growth on a previously approved batch, and your procedure sets the range. A one-sided minimum such as 70% is a site rule, not a pharmacopoeial requirement.