Microbial limits
USP General Chapter <2021> 'Microbial Enumeration Tests — Nutritional and Dietary Supplements' and <2022> 'Microbiological Procedures for Absence of Specified Microorganisms — Nutritional and Dietary Supplements' define the consensus microbiological specifications and test methods for dietary supplements. The framework covers Total Aerobic Microbial Count (TAMC), Total Yeast and Mold Count (TYMC), and four specified pathogens (E. coli, Salmonella, Staphylococcus aureus, Pseudomonas aeruginosa) with limits that vary by product matrix (botanical, mineral, animal-source, probiotic). Distinct from USP <61>/<62> drug-product methods because supplement matrices include living-organism products (probiotics, enzymes) that fail drug-method assumptions.
01What the supplement microbiology framework actually is
USP created the <2020 series> in the late 2000s to address a recurring problem: the <61>/<62> drug-product microbiological methods made implicit assumptions (homogenous matrix, antimicrobial preservation, no living-organism content) that fail for many supplement matrices. Botanical powders are heterogeneous; probiotics deliberately contain billions of live organisms; herbal tinctures have inherent antimicrobial activity from ethanol or essential oils. The <2021>/<2022>/<2023> chapters provide methods and limits adapted for these realities.
02The two test categories
- Enumeration (USP <2021>) — quantifies the total population of viable microorganisms. Two counts: TAMC (Total Aerobic Microbial Count, includes bacteria) and TYMC (Total Yeasts and Molds Count). Reported as CFU/g or CFU/mL.
- Specified microorganism absence (USP <2022>) — tests for the presence (or absence) of four pathogens of public-health concern: Escherichia coli, Salmonella spp., Staphylococcus aureus, Pseudomonas aeruginosa. Reported as 'present' or 'absent in 10 g' (or 1 g for higher-risk matrices).
A defensible Part 111 microbiological specification includes one TAMC limit + one TYMC limit + at minimum the Salmonella absence test, and for botanical / animal-source materials all four pathogens.
03USP <2023> matrix-specific limits
USP <2023> 'Microbiological Attributes of Nonsterile Nutritional and Dietary Supplements' sets default limits that vary by matrix. The principle: matrices with higher inherent microbial burden (botanicals, animal-source) get more permissive limits than synthetic / mineral products that should be cleaner.
| Matrix | TAMC (CFU/g) | TYMC (CFU/g) | Required absence tests |
|---|---|---|---|
| Synthetic vitamins / minerals | ≤10³ | ≤10² | Salmonella (10 g); E. coli (10 g) |
| Plant-origin botanical, untreated | ≤10⁵ | ≤10⁴ | Salmonella (10 g); E. coli (10 g) |
| Plant-origin botanical, treated (e.g., steam-sterilised) | ≤10⁴ | ≤10³ | Salmonella (10 g); E. coli (10 g) |
| Animal-source (gelatin, glandulars) | ≤10⁴ | ≤10³ | Salmonella (10 g); E. coli (10 g); S. aureus (1 g) |
| Probiotics (excluding the probiotic itself) | n/a (probiotic is intentional) | ≤10² typical | Salmonella (10 g); E. coli (10 g); S. aureus (10 g) |
| Topical / oral-mucosal (e.g., throat sprays) | ≤10² | ≤10¹ | All four specified pathogens |
These are USP defaults. A manufacturer may set tighter internal limits (and many do for product-quality or shelf-life reasons) but cannot release outside the USP limits without justification.
04The probiotic-counting trap
Probiotic products fail standard TAMC testing because the probiotic organism IS what TAMC measures — a 10 billion CFU/serving Lactobacillus product would 'fail' a 10⁵ TAMC limit by a factor of 100,000. The standard solution: TAMC is performed on a selective medium that excludes the deliberately-added probiotic genus (e.g., MRS agar excluded), or the manufacturer specifies and validates a method that quantifies the probiotic separately (the CFU label-claim assay using strain-specific plating or qPCR) and reports TAMC against a selectively-inhibited count of OTHER organisms.
The validation challenge is documenting that the selective inhibition does not also suppress the indicator organisms TAMC is meant to detect. USP <2021> chapter section 3 specifically discusses this neutralisation / validation requirement; FDA inspectors increasingly cite probiotic manufacturers who used a 'TAMC = excluded' workaround without the supporting validation record.
05Method selection — pour plate, MPN, rapid
| Method | Throughput | LOD | Best for | Compendial alignment |
|---|---|---|---|---|
| Pour-plate (classical) | Low — 3 to 5 day incubation | 1 CFU/g | Reference / disputed-result re-test | USP <2021> primary |
| Membrane filtration | Medium | Sub-CFU/g | Liquid matrices (extracts, oils) | USP <2021> primary |
| Most Probable Number (MPN) | Low — 7-day | 0.3 MPN/g | Difficult-to-plate matrices | USP <2021> alternative |
| Rapid (ATP bioluminescence, qPCR, automated immunoassay) | High — 24-48 hour | Variable | Routine release with validated linkage to reference | Per USP <1223> alternative-method validation |
Most modern supplement contract manufacturers run rapid methods (3M Petrifilm, BioMerieux TEMPO / BAX, Charm) for routine release with a periodic pour-plate cross-check, validated per USP <1223> 'Validation of Alternative Microbiological Methods'. The validation evidence is the most-frequent FDA inspection question — 'show me your method validation' is the standard inspector ask.
06Common Warning Letter failure modes
- Microbial spec set per ingredient without a finished-product spec — release decisioning has nothing to check against.
- Specified-pathogen test sample size wrong (1 g instead of 10 g) — does not detect the regulatory limit.
- Probiotic TAMC handled via 'excluded' note in the BPR without validation evidence of selective inhibition.
- Rapid method deployed without USP <1223> alternative-method validation against the reference method.
- Limits set tighter than USP <2023> defaults without internal capability data — chronic OOS rate creates a deviation-investigation backlog.
- Botanical material with no treatment step expected to meet treated-botanical limits — incoming material disposition cascade.
- Stability programme microbiological test points missing — TYMC drift over shelf life invalidates the label expiration.
- Salmonella detection without confirmation by selective enrichment + species ID — false positive shut-down or false negative release.
07How V5 Ultimate handles microbial limits
- Matrix-tagged microbial spec template auto-applied at material setup (synthetic / botanical / animal / probiotic) with USP <2023> default limits.
- Probiotic flag forces selective-method validation evidence linkage before spec is signed.
- Rapid-method library with USP <1223> validation record requirement; method cannot be used in LIMS until validation linkage is present.
- Specified-pathogen result form enforces 10 g sample-size + confirmation step.
- Stability programme automatically schedules TYMC at every stability time point for moisture-sensitive matrices.
- QC unit two-person e-sig required on microbiological disposition before lot moves to RELEASED.
- Botanical material treatment flag (steam / irradiation / ethylene-oxide) drives applicable limit tier automatically.
Frequently asked questions
Q.Is USP <2021>/<2022> mandatory?+
Not by federal mandate, but it is the consensus standard FDA cites and that USP-monograph products must comply with. Most national-brand manufacturers adopt it as the floor.
Q.Can I use USP <61>/<62> instead?+
For drug-product matrices, yes — that's their intended use. For supplements they often work but the <2020 series> was created specifically because drug methods make assumptions that fail for probiotic and complex-botanical matrices.
Q.Why is TAMC harder for probiotics?+
Because the probiotic IS what TAMC measures. The answer is selective inhibition or strain-specific counting, validated per USP <2021>. FDA cites manufacturers who used a workaround without validation evidence.
Q.Can I use a rapid method?+
Yes, validated against the USP reference method per USP <1223> 'Validation of Alternative Microbiological Methods'. Validation evidence is the most-frequent FDA inspection ask.
Q.How often must I test stability for microbials?+
At every stability time point if the matrix is moisture-sensitive (most powders) or has live-organism content (probiotics, enzymes). Once-on-release without stability points cannot defend the labelled shelf life.
Q.What sample size for the pathogen tests?+
USP <2023> default is 10 g for Salmonella and E. coli; 1 g may apply for some matrices but document the justification. Wrong sample size is a frequent Warning Letter observation.
Q.Do I need to test every lot?+
Yes for most matrices, unless you have a documented and statistically justified skip-lot plan per §111.75(c). Most operators run every lot for botanicals and animal-source materials because risk is variable batch-to-batch.
Primary sources
- USP General Chapter <2021> Microbial Enumeration Tests — Nutritional and Dietary Supplements
- USP General Chapter <2022> Microbiological Procedures for Absence of Specified Microorganisms — Nutritional and Dietary Supplements
- USP General Chapter <2023> Microbiological Attributes of Nonsterile Nutritional and Dietary Supplements (limits)
- AOAC INTERNATIONAL — Official Methods 2014 / 991 / 998 (supplement microbiology)
- 21 CFR 111.75(b) — Quality control tests on finished batches
- FDA — Bacteriological Analytical Manual (BAM)
Further reading
- Finished-product specs (111.70(e))The Part 111 rule microbial limits sit inside.
- Scientifically valid method (111.320)Validation / verification framework microbiological methods must pass.
- Heavy metals (supplements)The other major finished-product spec category.
- Component specs (111.70(b))Microbial limits also apply to incoming ingredients.
- Environmental monitoringPlant-side controls upstream of finished-product spec.
V5 Ultimate ships with the Microbial limits controls already wired in — audit trail, e-signatures, validation evidence. Free trial, no credit card, onboard in days, not months.