The European Directorate for the Quality of Medicines and HealthCare announced on 4 August 2026 that a revised version of the European Pharmacopoeia monograph Macrogols (1444) has been published in Pharmeuropa 38.3 for comment.
The revision introduces a new gas chromatographic procedure for the determination of ethylene glycol and diethylene glycol, revises the associated limits, and replaces two wet chemistry identification tests with infrared absorption spectrophotometry. The draft carries the reference PA/PH/Exp. 13H/T (25) 59 ANP and covers all twelve grades of macrogol described in the monograph, from macrogol 300 through to macrogol 35 000.
The context for the revision is well known to anyone who has followed excipient safety over the past few years. Poisoning incidents involving glycol-contaminated or deliberately adulterated excipients have kept ethylene glycol and diethylene glycol at the top of the regulatory agenda, and pharmacopoeias have been working through the excipient monographs most likely to carry these impurities.
The EDQM places the macrogols revision alongside the recently adopted revisions of Propylene glycol (0430) and Glycerol (0496), and confirms that dedicated glycol tests are being prepared for liquid sorbitol and maltitol. Both glycols are also genuine process impurities in macrogols, particularly in the lower molecular mass grades, so this is not solely an adulteration control.
Why the Current GC Method for Ethylene Glycol and Diethylene Glycol Is Being Replaced
The current procedure runs on a glass-packed column, 1.8 m by 2 mm, filled with silanised diatomaceous earth impregnated with 5 percent m/m macrogol 20 000. The briefing note gives two reasons for abandoning it. The column is available only through on-demand manufacturing, which makes routine supply difficult and creates a single point of failure for any laboratory relying on it.
Users have also repeatedly reported difficulty meeting the system suitability requirements, which in practice means repeated column conditioning, unstable retention and analytical results that are hard to defend during an inspection or a supplier qualification review.
Anyone who has tried to run the existing method will recognise the picture. Adjusting the initial oven temperature until diethylene glycol elutes somewhere between 14 and 16 minutes is not a robust starting point for a compendial method, and the absence of an internal standard leaves the result exposed to injection variability.
The Proposed Capillary GC Procedure for Macrogols (1444)
The proposed method is a capillary GC procedure with flame ionisation detection, built around a derivatisation step and an internal standard. The sample is prepared by weighing 0.100 g of the substance into a 2 mL glass vial, adding 100 microlitres of the decanol internal standard solution, 0.5 mL of pyridine and 220 microlitres of a derivatisation solution consisting of ten volumes of N,O-bis(trimethylsilyl)trifluoroacetamide and one volume of chlorotrimethylsilane.
The vial is closed and heated at 80 °C for 30 minutes, then allowed to cool. Three reference solutions are prepared, with reference solution (b) used to identify the ethylene glycol and diethylene glycol peaks and reference solution (c) used for system suitability.
Separation is carried out on a 30 m by 0.32 mm fused silica column coated with methylpolysiloxane at a film thickness of 0.25 microns, with HP-1 named in a footnote as suitable. Helium is the carrier gas, with a flow rate of 3 mL/min for the first 11.5 minutes rising to 10 mL/min thereafter, and a split ratio of 1:10.
The oven starts at 60 °C, holds for one minute, ramps to 165 °C by 11.5 minutes, then to 330 °C by 14.8 minutes, and holds until 18.8 minutes. The injection port is set to 250 °C and the detector to 350 °C, with an injection volume of 1 microlitre. Relative to the internal standard, which elutes at about 10 minutes, ethylene glycol appears at about 0.4 and diethylene glycol at about 0.8. The only stated system suitability criterion is a signal-to-noise ratio of at least 10 for both analyte peaks in reference solution (c).
Related Article: Peak Tailing in Chromatography
| Parameter | Current monograph | Proposed draft |
|---|---|---|
| Column | Glass packed column, 1.8 m x 2 mm, silanised diatomaceous earth impregnated with 5 per cent m/m macrogol 20 000 | Fused silica capillary, 30 m x 0.32 mm, methylpolysiloxane, film thickness 0.25 microns (HP-1 cited as suitable) |
| Carrier gas | Nitrogen, 30 mL/min | Helium, 3 mL/min to 11.5 min then 10 mL/min, split 1:10 |
| Sample treatment | Direct dilution in acetone | Silylation with BSTFA and chlorotrimethylsilane in pyridine, 80 °C for 30 min |
| Quantification | External standard | Internal standard (decanol) |
| Oven programme | Isothermal start adjusted to give 14 to 16 min retention for diethylene glycol, then ramp of about 30 °C at 2 °C/min, not above 170 °C | 60 °C for 1 min, to 165 °C by 11.5 min, to 330 °C by 14.8 min, hold to 18.8 min |
| System suitability | Repeatability of response over 5 replicate injections | Signal to noise ratio of at least 10 for both analytes in reference solution (c) |
| Limits | Combined limit of 0.4 per cent for ethylene glycol and diethylene glycol | Ethylene glycol not more than 620 ppm; diethylene glycol not more than 0.10 per cent |
New Limits for Ethylene Glycol and Diethylene Glycol
The current monograph applies a single combined limit of 0.4 per cent for the sum of ethylene glycol and diethylene glycol. The draft replaces this with individual limits: not more than 620 ppm for ethylene glycol and not more than 0.10 per cent for diethylene glycol. The briefing note explains the reasoning.
The ethylene glycol figure is aligned with the ICH Q3C guideline on residual solvents, while the diethylene glycol figure matches the limit already applied in the Ph. Eur. monograph on Propylene glycol (0430).
The practical effect is a substantial tightening. Expressed on a common basis, the existing combined limit corresponds to 4000 ppm, whereas the proposed limits cap ethylene glycol at 620 ppm and diethylene glycol at 1000 ppm. A batch that comfortably passed the old specification could fail the new one. The test remains restricted to macrogols with a relative molecular mass below 1000, so grades 300, 400 and 600 are in scope and the higher grades are not.
Identification by Infrared Absorption Spectrophotometry
The second element of the revision is less commercially sensitive but worth noting. Identification tests B and C, the sulfuric acid and mercuric chloride fume test and the potassium thiocyanate and cobalt nitrate colour test, are deleted and replaced by a single infrared absorption spectrophotometry test under general chapter 2.2.24.
Two reference standards are introduced to cover the molecular mass range, macrogol 400 CRS for grades 300 to 600 and macrogol 4000 CRS for grades 1000 to 35 000. Defined tolerances of plus or minus 10 wavenumbers are permitted on the listed transmission minima or absorption maxima, with a larger tolerance allowed on specific bands to accommodate variation in molecular mass and molecular mass distribution. Viscosity is retained as identification A because it distinguishes Macrogols (1444) from high-molecular-mass macrogols (2444).
The removal of a mercuric chloride-based test is a welcome secondary outcome. Compendial methods that rely on mercury salts create disposal, handling and occupational exposure obligations that most laboratories would rather do without, and eliminating one of them removes a recurring justification burden from the laboratory safety file.
Impact on Excipient Manufacturers and Pharmaceutical Quality Units
For excipient manufacturers and for the quality units that qualify them, the consultation period is the point at which influence is possible. The EDQM has been explicit that it wants analytical data, not only editorial comment. It has asked interested parties to submit results generated with the proposed procedure or with an alternative procedure, and has stressed that actual values from batches of pharmaceutical quality currently on the market are essential to confirm that the proposed limits are fit for purpose.
If routine batch data sits close to 620 ppm for ethylene glycol, that is exactly the information the Group of Experts needs to see before the text is adopted.
Several points deserve attention while the text is still in draft.
- Confirm which macrogol grades your sites use and whether they fall below the relative molecular mass threshold of 1000, since only those grades carry the glycol test.
- Review historical release and incoming data for ethylene glycol and diethylene glycol against the proposed individual limits rather than the combined limit, and identify any grade or supplier that would move close to specification.
- Assess the analytical impact of the silylation step. BSTFA and chlorotrimethylsilane are moisture sensitive, and water content in the sample will consume derivatisation reagent, so sample handling and reagent age become genuine method variables.
- Check laboratory readiness for helium supply and for a 330 °C oven programme, and confirm the availability of decanol of appropriate quality as internal standard.
- Factor the change into supplier agreements, specifications and analytical procedures, and plan for method verification under general chapter 5.26 once the revised monograph is adopted.
Manufacturing authorisation holders should also read this against the existing expectation, set out in the Commission guidelines on the formalised risk assessment for excipients of 19 March 2015 and in EU GMP Part I, that excipient controls are proportionate to the risk the excipient presents. A tightened compendial control does not remove the obligation to assess supply chain risk for glycol bearing excipients, but it does raise the baseline against which that assessment is judged.
How to Comment on the Pharmeuropa 38.3 Draft Monograph
Comments are due by 30 September 2026. Instructions are set out in the EDQM guide on how to comment. Parties in states that are signatories to the European Pharmacopoeia Convention should route comments through their national pharmacopoeia authority, while comments from other countries and from industry associations go directly to the EDQM through the HelpDesk. The draft text itself is available in Pharmeuropa 38.3, which is freely accessible.
