Ursodeoxycholic Acid EP Impurities: Complete A–I Reference

Ursodeoxycholic acid (UDCA) has nine named impurities in the European Pharmacopoeia, lettered A to I. Four of them are not folded into a single catch-all figure — they are controlled individually — and one, lithocholic acid, is tested by a method of its own.

Most specifications you will be sent do not show that structure. They print a total impurities number and stop. That is usually where a buying decision starts to drift.

This reference to ursodeoxycholic acid EP impurities maps every lettered impurity to its chemical name and CAS number, gives the limit that applies to each, and flags three figures that circulate incorrectly online. Every value below comes from the Ph. Eur. monograph for ursodeoxycholic acid (monograph 1275) and the tests it defines.

The nine EP impurities, letter by letter

Two things about that table are worth pausing on. First, the letters are assigned relative to the parent monograph — they are not fixed names. In the CDCA monograph, UDCA is itself impurity A. A specification that says “impurity A” without naming the substance is ambiguous, and this is one of the most common misreadings in bile-acid documentation.

Second, the EP issues reference standards (CRS) for only some of these. The monograph’s system suitability standard contains impurities A and H, and the lithocholic acid test uses a dedicated CRS for impurity C. Those are the pairs the monograph treats as critical — for good reason, as the sections below explain.

What “impurity” actually means in a pharmacopoeia

The related-substances section of a monograph recognises three categories, and mixing them up is the root of most specification arguments.

• Specified (lettered) impurities. Named individually, each with its own limit. For UDCA these are A to I. A result for one of them cannot be “averaged away” against the others.
• Unspecified impurities. Not named in the monograph. They are controlled by a single ceiling that applies to any one of them.
• Total impurities. The sum, controlled by one overall ceiling.

There is also a disregard limit: a threshold below which a peak is simply ignored and not counted at all. It exists because demanding a number for every trace peak would be unworkable. It is not a permitted level of contamination — it is the point below which the monograph stops asking.

Impurity A — chenodeoxycholic acid: the one you cannot switch off

Chenodeoxycholic acid differs from UDCA by a single stereocentre: the hydroxyl at C-7 points α instead of β. That one bond angle is the whole difference, and it is why the two are so easy to confuse and so hard to separate.

CDCA also happens to be the molecule UDCA is usually made from. The 7α substrate is oxidised to a 7-oxo intermediate and then reduced back to the 7β configuration — with the unreacted starting material, and any material reduced back to the α form, arriving at the end of the chain as impurity A. The substrate and the principal impurity are the same molecule, which is exactly why the limit is set where it is.

That is also why a 1.0% ceiling is not a sign of loose control. It is an acknowledgement that driving CDCA to zero would cost more yield than the residual amount is worth. If you want the wider context on UDCA limits, a full UDCA specification sets impurity A alongside the assay, rotation and solvent parameters.

Impurity C — lithocholic acid: the only one with its own test

Lithocholic acid is the exception in the monograph: it is the only impurity given a dedicated test, run by thin-layer chromatography, with a limit of 0.1%. Everything else falls under the liquid-chromatography related-substances test.

The reason is toxicological. Lithocholic acid is the most hydrophobic of the common bile acids, and it is the one regulators single out for separate control. A dedicated, more sensitive test is how the monograph makes sure it cannot hide inside the total-impurities figure.

For a buyer this has one practical consequence: a UDCA certificate of analysis that reports only a total impurities number, with no separate line for lithocholic acid, is not a complete document. The monograph requires the separate result, so its absence is a documentation gap, not a formatting choice.

Impurities F and H — the separation problem

Impurity H is UDCA with the C-3 hydroxyl flipped from β to α. Impurity F is UDCA’s 7-oxo relative. Neither is far from UDCA in polarity, and both sit close to it on a chromatogram, which is why the monograph leans on resolution requirements rather than on generous limits.

The system suitability standard supplied for the assay deliberately contains impurities A and H, because those are the pair a routine method most easily fails to resolve. Impurity H elutes at a relative retention of roughly 0.9 and the method is expected to separate it from the main peak cleanly. If your supplier’s method cannot demonstrate that resolution, the numbers on the certificate are less informative than they look.

Impurity F has a second life outside quality control: the same molecule, 7-ketolithocholic acid, is the key intermediate on the route from CDCA to UDCA, and it trades in its own right. We cover that identity, and why it is worth knowing under three different names, in 7-KLCA in detail.

The three numbers that get quoted wrong

Three figures from this monograph circulate online in versions that are simply not what the pharmacopoeia says. They are worth memorising in the correct form.

The pattern is worth noting: the incorrect figures are not all in one direction. Some relax a limit, some tighten one. That combination is the signature of a table assembled from several sources rather than read from a single monograph — which is why the number alone tells you nothing until you know where it came from.

How to read the related-substances block on a COA

1. Check the method is named. The monograph specifies liquid chromatography for related substances and thin-layer chromatography for impurity C. A certificate that does not say which method produced the numbers is incomplete.
2. Look for a line per specified impurity. A, B, C and the others should each be capable of appearing separately. Their results should be reported against their own limits, not blended.
3. Confirm the lithocholic acid line exists. It is required. If it is missing, ask for it.
4. Read the disregard limit stated on the certificate. If it is not 0.05%, ask why.
5. Sanity-check the totals. Unspecified ≤0.10% and total ≤1.5% are the ceilings. A total well above 1.5%, or a single unspecified peak above 0.10%, is a release failure, not a grey area.
6. Cross-check the assay. Content on the dried basis should sit between 99.0% and 101.0%. An assay reported as “≥98.5%” is using a different definition of the acceptance range.

The rest of the certificate: the parameters that are not impurities

Impurity letters are only part of what a certificate has to show. The same monograph that defines A to I also sets the identity and purity parameters that sit beside them, and a certificate is incomplete without them.

• Content, dried basis: 99.0% to 101.0%
• Specific optical rotation: +58.0° to +62.0° (dried substance) — the identity check that most quickly separates UDCA from its 7α relative
• Loss on drying: not more than 1.0%
• Sulphated ash: not more than 0.1%

The European Pharmacopoeia publishes this monograph, and the reference standards that support it, through the EDQM. General guidance on how impurities are classified, reported and qualified sits with the ICH quality guidelines, which national pharmacopoeias build on rather than duplicate.

It is also worth knowing that the EP is not the only specification a customer may hold you to. The United States Pharmacopeia, published by the USP, sets its own values for several of these parameters, and the two do not always agree. When a customer specifies “pharmacopoeial grade” without naming the pharmacopoeia, that ambiguity is worth resolving before the order rather than after it.

Questions buyers ask

How many impurities does the EP define for UDCA?
Nine are given letters, A to I. Impurities A, B and C are the most frequently discussed, and impurity C is the only one with a test of its own, run by TLC at a limit of 0.1%.

What is the EP limit for chenodeoxycholic acid in UDCA?
Chenodeoxycholic acid is impurity A, limited to 1.0%. It is a relatively generous ceiling because CDCA is also the starting material for UDCA production, and clearing it completely is not economic.

What is the limit for lithocholic acid?
0.1%, under a dedicated thin-layer chromatography test. It is separate from the liquid-chromatography related-substances test that covers the other impurities, because of its toxicological profile.

What is the disregard limit in the UDCA test?
0.05%. Peaks at or below this level are not counted. It is a reporting threshold, not an allowance — a common misreading that turns it into an unofficial specification.

What is UDCA EP Impurity F?
Impurity F is 7-ketolithocholic acid, CAS 4651-67-6. It is the 7-oxo intermediate on the CDCA-to-UDCA route and also trades as a product in its own right.

What is the total impurities limit?
1.5%, covering all impurities other than the individually specified ones. Unspecified impurities are additionally capped at 0.10% each.

If you buy UDCA, ask for these three things

The lettered impurity structure is not something you can change as a buyer — it is set by the monograph. What you can control is whether the documentation in front of you actually reflects it.

Ask for the related-substances block with the method named, a separate lithocholic acid result, and an assay stated on the dried basis. Those three checks separate a certificate produced against the monograph from one assembled to look like it. If you want to see how a full specification line sits together, our ursodeoxycholic acid grade lists the parameters customers ask about most, and you can always request a quotation against your own acceptance criteria.

This article describes pharmacopoeial requirements and analytical interpretation only. It is not a statement of therapeutic use, and it does not set out commercial guarantees. Always work from the pharmacopoeia edition in force at the time of your release.

 

Frequently asked questions

What is the difference between named and unspecified impurities?

Named impurities carry individual limits and their own methods; unspecified ones fall under a total and a general reporting threshold. Both sections of the table do work. The named lines show what the process produces, and the total catches what nobody named yet. A certificate missing either half is half a certificate.

Does meeting the named-impurity limits prove pharmacopoeial compliance?

It is one element of it. Compliance is a judgment across the whole monograph: identity, assay, water, residue and the impurity table together. A certificate that wins on impurities while staying silent on water or identity has not settled anything. Read the table, then read what the table does not contain.

Which impurity lines say the most about a supplier's process?

The epimeric and oxidation-related ones. Chenodeoxycholic acid as impurity A and the keto acids track isomerisation and oxidation control directly. A supplier who holds those lines consistently, batch after batch, has a controlled route. The named table is a process history wearing a specification's clothes.

Can a supplier set tighter limits than the monograph?

Yes, and many do for good commercial reasons. The check is not whether the numbers are tighter but whether limits and methods are both stated. Tight numbers without named methods are decoration, and comparing two tight tables without methods is comparing nothing.

What is the practical value of the total impurities line?

It catches what the named list does not. A process change that introduces a new compound first shows up as a total that creeps while every named line stays green. Buyers who watch only the named columns see process drift about a year after buyers who also watch the total.

Two certificates both say "conforms" with different limits. Which is stricter?

Compare line by line. A result at the ceiling of a tight table is a stronger statement than a comfortable result against a loose one, even though the second looks calmer. "Conforms" is a claim about a relationship between a result and a limit; without seeing both numbers, the word carries no information at all.