Buying & Verification 9 min read Updated 2026-09-28

How to Read a Bacteriostatic Water COA and Spot One That Proves Nothing

A certificate of analysis is a testable claim, not a trust badge. This guide walks each line of a bacteriostatic water COA, gives the published specification to check it against, and lists what a decorative COA leaves out.

Written by Bacteriostatic Water US. Sourced from 14 primary references including the FDA-approved product label, USP monographs and general chapters, CDC injection-safety guidance, and peer-reviewed literature indexed in PubMed. How we write and check guides.

A small glass vial with a grey stopper and plain silver seal on a white lab bench beside a rack of teal-capped sample vials and a closed white folder
IllustrationA certificate of analysis is only useful when it can be matched to the vial in front of you.

Key Technical Takeaways

  • A COA is only meaningful if it names a lot number, a test method, a specification, and a measured result. A result without its specification proves nothing.
  • Check results against published limits you can verify independently: 0.9% (9 mg/mL) benzyl alcohol and pH 4.5 to 7.0 from the FDA label, and less than 0.5 EU/mL endotoxin from the USP monograph (0.25 EU/mL is the Water for Injection limit).
  • "Passed", "Conforms" and green checkmarks are not data. A real endotoxin line reads like "< 0.05 EU/mL by kinetic chromogenic LAL", not "Passed".
  • The lot number on the COA must match the lot printed on the vial in your hand. A COA for a different lot (or a generic undated PDF) documents nothing about your vial.
  • A QR code is a convenience, not evidence. Scan it before you trust it and confirm it resolves to a lot-specific document rather than a storefront.

1. What a Certificate of Analysis Actually Is

A certificate of analysis (COA) is a test report for one manufactured lot. It is not a badge, a seal, or a statement of intent. It records four things for every test performed: the method used, the specification the result had to meet, the measured result, and the lot it applies to.

Remove any one of those four and the document stops being evidence. A result with no specification cannot be judged. A specification with no result is a promise. A test with no named method cannot be repeated. And a report with no lot number cannot be connected to the vial sitting on your bench.

The one-sentence test. Read any line of a COA and ask: could a second laboratory repeat this and get a comparable number? “Bacterial endotoxins: < 0.05 EU/mL by kinetic chromogenic LAL, specification < 0.5 EU/mL” passes. “Endotoxins: Passed ✓” does not.

2. An Annotated Example COA, Line by Line

Most pages on this topic list the tests a COA should contain. Few show what the finished document looks like, so we drew one. The example below is a generic mock-up, not any company’s real certificate: the lot number is a placeholder and the results are illustrative values inside the published limits. The numbered markers follow the contents FDA’s Q7 guidance recommends for a certificate: each test with its acceptance limit and numerical result, the batch number, release and expiry dates, and a dated signature from the quality unit.1

What a complete bacteriostatic water COA shows
A generic example certificate of analysis with seven numbered checkpoints: manufacturer, lot, dates, quantitative results with limits and methods, the limited use of Conforms, the testing laboratory, and the quality unit sign-off. Limits: FDA-approved label2, USP monograph3, FDA endotoxin guide4. Layout: FDA Q7 guidance, section 11.41. Illustration, not a real document.

Two lines in the example deserve a note. Particulate matter is counted by instrument: USP <788> passes an injection of 100 mL or less when it averages no more than 6,000 particles of 10 µm or larger and no more than 600 of 25 µm or larger per container by light obscuration.5 Visible particulates are different: FDA expects every final container to be inspected and any unit with visible particles rejected, so “Conforms” is a fair way to report that line.6

3. The Specifications You Can Check Independently

The strongest thing about bacteriostatic water as a purchase is that its key numbers are public. You do not have to take a supplier’s word for what the specification should be, because the FDA-approved label and the USP-NF monograph state it.23 Check the seller’s results against these, not against the seller’s own claims:

TestPublished specificationWhere the number comes fromWhat a real result looks like
Benzyl alcohol assay0.9% (9 mg/mL) or 1.1% (11 mg/mL)FDA-approved label2“9.02 mg/mL by HPLC”
pH4.5 to 7.0 (typical 5.7)FDA-approved label2“5.8, potentiometric”
Bacterial endotoxins< 0.5 EU/mL (a house spec such as < 0.25 EU/mL may be tighter)FDA inspection guide, crediting USP Supplement 4a (1984); the 0.25 EU/mL figure is the Water for Injection limit4“< 0.05 EU/mL, kinetic chromogenic LAL”
SterilityNo growthUSP <71> membrane filtration3“No growth at 14 days, USP <71>” (the compendial test incubates for 14 days)7
Particulate matterNo more than 6,000 particles ≥ 10 µm and 600 ≥ 25 µm per containerUSP <788> light obscuration limits for injections of 100 mL or less5“310 and 12 per container, USP <788>”
ContainerType I if borosilicate glassUSP <660> glass classification8“Type I borosilicate, hydrolytic resistance per <660>”

Note the pattern in the right-hand column: every credible result is a number with a unit and a method. That is the difference between a laboratory document and a graphic.

4. Seven Red Flags on a Bacteriostatic Water COA

1. No lot number, or a lot that doesn’t match your vial

This is the single most important check and the one most often skipped. A COA certifies one batch. If the document does not carry a lot number, or carries one that differs from the lot printed on the vial in your hand, it tells you nothing about what you bought. Suppliers who publish a single undated COA for all orders are publishing a sample, not a certificate.

A cardiologist who wrote a public thread on checking vials bought outside pharmacies listed the same check first:

“Match the lot number on the physical vial to the COA. Exactly. No match, stop there.”

Afshine EmraniMD, FACC, cardiologist, Los Angeles Heart Specialists, Tarzana, CAPost on X, August 11, 2026

2. Results stated as “Passed”, “Conforms” or a checkmark

“Conforms” is legitimate for a small number of attribute tests, such as appearance or a net content check against a label claim. It is not legitimate for assay, pH, or endotoxins, which produce numbers. A COA where every line reads “Passed” has removed exactly the information that would let you evaluate it.

3. No analytical method named

“Endotoxin: < 0.5 EU/mL” is weaker than “Endotoxin: < 0.05 EU/mL by kinetic chromogenic LAL.” The method determines what the number means and whether it can be reproduced.4

4. No specification printed beside the result

A result in isolation invites you to assume it is good. Real COAs print the acceptance criterion next to the measurement so the reader can judge rather than trust.

5. No test date or release date

Bacteriostatic water carries an expiration date, and the label directs storage at 20 to 25°C.2 A COA without a date cannot be placed on that timeline.

6. No laboratory identity or signatory

Someone has to be accountable for the numbers. A COA should name the testing laboratory and carry an authorised signatory or release identifier. “Third-party tested” without naming the third party is not a claim that can be checked.

7. A QR code you haven’t actually scanned

QR codes have become a trust signal in this category, which is precisely why they deserve scrutiny. A QR code is a link. It can resolve to a lot-specific PDF, or to a storefront, or to an affiliate tracking URL, and all three look identical printed on a label.

Scan it before you trust it. Point your phone at the code and read the URL bar before the page loads. A verification code should land on a document that names the lot printed on that vial. If it lands on a product listing, a checkout, or a URL carrying referral or campaign parameters, it is a sales link wearing a verification badge. That is worth knowing either way. It does not necessarily mean the product is bad, but it does mean the QR code proved nothing.

5. What FDA Expects of a COA (and Why Manufacturers Don’t Take Them on Trust)

The regulated drug industry already has rules for this problem, and they are stricter than most buyers assume. FDA’s Q7 guidance, written for active pharmaceutical ingredients, is the closest thing to a published template for a COA. It says a certificate should list each test performed with its acceptance limits and the numerical results obtained, be dated and signed by the quality unit, and name the original manufacturer. When a repacker or broker issues its own certificate, it should name the lab that did the testing, reference the original manufacturer, and attach a copy of the original batch certificate.1

Even a certificate that meets that standard is not taken on faith. Under 21 CFR 211.84(d)(2), a drug manufacturer may accept a supplier’s report of analysis in place of its own testing only if it runs at least one specific identity test on the material itself and periodically validates the supplier’s results.9 In other words, the industry that writes COAs treats every COA it receives as a claim to be checked.

FDA warning letters show why. In a 2018 letter to an Indian manufacturer of over-the-counter drugs, FDA reported that the firm’s own personnel told investigators the microbiological results recorded on its certificates of analysis had been falsified and the testing had not been performed.10 A buyer cannot audit a lab, but can do the part that costs nothing: read the numbers, check them against the published limits, and confirm the lab and the lot are real.

6. What to Ask a Supplier, Word for Word

If a COA is not published, or the published one is generic, these four questions separate suppliers who hold documentation from suppliers who do not. Any legitimate distributor can answer all four quickly:

  1. “Can you send the COA for the specific lot you will ship me?” The answer should be yes, and the lot should match what arrives.
  2. “Which laboratory performed the testing, and is it the manufacturer or an independent lab?” Both answers are acceptable. Refusing to say is not.
  3. “What method was used for the benzyl alcohol assay and the endotoxin test?” Expect a chromatographic method for the assay and an LAL method for endotoxins.4
  4. “Are you the manufacturer, the labeler, or a reseller?” This is a fair question and a revealing one. A reseller can be perfectly reputable, but a reseller who describes the manufacturer’s testing as their own is telling you something about how they handle information.

The six questions below turn the sections above into a sequence you can run against any COA in a couple of minutes. A single “no” means the document cannot be verified, whatever else it says.

Six checks before you trust a COA
A flowchart of six yes or no questions: matching lot, numerical results, printed limits, limits that match the published ones, named methods, and a named, dated and signed lab. Any no stops the check. Published limits: FDA-approved label2 and FDA endotoxin guide4.

7. Documentation Is Not the Whole Picture

A clean COA describes the product at the moment it left the manufacturer. Two things happen afterwards that no certificate covers.

Transit and storage. The label specifies 20 to 25°C.2 Nothing on the outside of a parcel records whether that held. This is the argument for buying from a supplier who controls storage, and for not buying injectable diluents from general-merchandise marketplaces where the storage history is unknowable.

Your own handling. Once you puncture the vial, the documentation stops and technique takes over: a new sterile needle and syringe for every entry, a clean preparation area, and discard within 28 days of first puncture.1112 The most thoroughly tested vial in the world becomes unsafe with careless re-entry. See the 28-day puncture rule for the reasoning.

8. If Something Looks Wrong

FDA publishes guidance on evaluating online sellers of drug products, and maintains a public list of warning letters issued to internet pharmacies.1314 Both are worth a search before a first order from an unfamiliar supplier. And regardless of paperwork: inspect the vial itself. Cloudiness, particulates, a compromised seal, or a label that does not match the COA are reasons to stop, independent of what any document says.

For the surrounding purchase decision (retail pharmacy, marketplace, or direct supplier), see where to buy bacteriostatic water, and the Hospira label reference for what the best-known brand’s presentations actually specify.

Frequently Asked Questions

What is a certificate of analysis for bacteriostatic water?

A certificate of analysis (COA) is a lot-specific test report. It records which tests were run on one manufactured batch, the method used, the specification each test had to meet, and the measured result. A COA that does not identify a specific lot is not a certificate of analysis. It is marketing.

How can you tell a bacteriostatic water COA is fake or meaningless?

The clearest signals are: no lot number, no test date, results given as 'Passed' or 'Conforms' with no measured value, no named analytical method, no specification alongside each result, and a lot number that does not match the vial in your hand. Any one of these makes the document unverifiable.

What numbers should a bacteriostatic water COA show?

Benzyl alcohol should assay at 0.9% (9 mg/mL) or 1.1% (11 mg/mL) depending on the presentation, and pH should fall within 4.5 to 7.0, both stated on the FDA-approved label. The USP monograph limits bacterial endotoxins in bacteriostatic water to less than 0.5 EU/mL (0.25 EU/mL is the limit for Water for Injection, not bacteriostatic water). Results should be reported as measured values against those limits.

Can you trust a supplier's COA without testing the product yourself?

FDA does not let drug manufacturers do that. Under 21 CFR 211.84(d)(2), a manufacturer may accept a supplier's report of analysis only if it runs at least one specific identity test itself and periodically confirms that the supplier's results are reliable. For a buyer, the practical version is to treat a COA as a claim to check: match the lot, read the numbers against published limits, and confirm the lab exists.

Does a QR code on a vial prove the product is genuine?

No. A QR code is a link, and a link can point anywhere, including a storefront or an affiliate page. Scan it before you rely on it. A verification QR code should resolve to a document naming the exact lot printed on that vial. If it lands on a product listing, it is a convenience feature, not evidence.

References & Primary Sources

Every technical claim on this page is sourced to a pharmacopeial monograph, a federal regulatory body, or peer-reviewed literature. Links open the primary source directly.

  1. Regulatory / pharmacopeialQ7 Good Manufacturing Practice Guidance for Active Pharmaceutical IngredientsU.S. Food and Drug Administration · Guidance for Industry (ICH), Revision 1, September 2016. Section 11.4, Certificates of Analysis
  2. Regulatory / pharmacopeialBacteriostatic Water for Injection, USP: FDA prescribing information (Hospira, Inc.)DailyMed, U.S. National Library of Medicine (NIH) · FDA-approved product label, SPL set ID 87d6e9dc
  3. Regulatory / pharmacopeialBacteriostatic Water for Injection (official monograph)United States Pharmacopeia (USP-NF) · Official USP monograph · full text requires a USP subscription; the public page shows a preview
  4. Regulatory / pharmacopeialBacterial Endotoxins/Pyrogens: Inspection Technical GuideU.S. Food and Drug Administration · ITG No. 40
  5. Peer-reviewed literatureIssues and Challenges of Subvisible and Submicron Particulate Analysis in Protein SolutionsScherer TM, et al. · The AAPS Journal. 2012;14(2):236-243. doi:10.1208/s12248-012-9335-8
  6. Regulatory / pharmacopeialInspection of Injectable Products for Visible ParticulatesU.S. Food and Drug Administration · Draft Guidance for Industry, December 2021
  7. Peer-reviewed literatureSterility Testing for Cellular Therapies: What Is the Role of the Clinical Microbiology Laboratory?Gebo JET, Lau AF. · Journal of Clinical Microbiology. 2020;58(7):e01492-19. doi:10.1128/JCM.01492-19
  8. Regulatory / pharmacopeialGeneral Chapter ⟨660⟩ Containers: GlassUnited States Pharmacopeia (USP-NF) · Hydrolytic resistance and Type I / II / III glass classification · full text requires a USP subscription; the public page shows a preview
  9. Regulatory / pharmacopeial21 CFR 211.84: Testing and approval or rejection of components, drug product containers, and closuresU.S. Food and Drug Administration, Electronic Code of Federal Regulations · Title 21, Part 211, paragraph (d)(2)
  10. Regulatory / pharmacopeialWarning Letter: JT Cosmetics & Chemicals Pvt Ltd. (MARCS-CMS 554478)U.S. Food and Drug Administration · Warning letter, July 27, 2018
  11. Public health authorityPreventing Unsafe Injection PracticesU.S. Centers for Disease Control and Prevention · Injection Safety: clinical safety guidance for healthcare providers
  12. Regulatory / pharmacopeialGeneral Chapter ⟨797⟩ Pharmaceutical Compounding: Sterile PreparationsUnited States Pharmacopeia · USP-NF general chapter · full text requires a USP subscription; the public page shows a preview
  13. Regulatory / pharmacopeialBeSafeRx: Your Source for Online Pharmacy InformationU.S. Food and Drug Administration
  14. Regulatory / pharmacopeialInternet Pharmacy Warning LettersU.S. Food and Drug Administration · Drug Supply Chain Integrity
Share this guideXLinkedInFacebookReddit
PureBac™ Certified Solvents

Need 7x tested bacteriostatic water for your protocol?

Every PureBac vial contains USP-grade non-pyrogenic water with exact 0.9% benzyl alcohol, verified by 7 analytical lab tests with an on-vial batch QR code. Shipped same day from Tampa, FL.

In stock, tested, and out the door by three.

Order before 3 PM EST and it ships from Tampa today.
Shop vials from $7.99