Comparison of Regulations for Arsenic and Heavy Metals in Herbal Medicines Using Pharmacopoeias of Nine Counties/Regions

Summary

Inada, Kiuchi, and Urushihara (Keio University, Division of Drug Development and Regulatory Science) reviewed and compared pharmacopoeial regulations for arsenic, lead, cadmium, and mercury in herbal medicines across seven countries and two regions, plus the WHO guidelines and the ISO 18664:2015 standard. The study covered more than 2,000 herbal medicine monographs. The authors found that upper limits and analytical test methods for elemental impurities varied substantially by country and regulatory body, and that full alignment with WHO or ISO recommendations had not been achieved. The paper concludes that “loose harmonization” toward internationally agreed standards represents a feasible option to balance diversity, safety, and international trade in herbal medicines.

Study Design

This is a documentary review, not an analytical measurement study. No laboratory measurements were performed and no samples were collected. The authors compiled the monographs for herbal medicines from nine pharmacopoeias and two international standards, then systematically extracted and compared the stated upper limits for arsenic, lead, cadmium, and mercury and the analytical test methods authorized for each element in each document. Figures were created using JMP Pro 16 (SAS Institute Inc., NC, USA) with bubble plots showing the distribution of limit values across documents.

The nine pharmacopoeias and two international standards covered are listed in Table 1 of the paper:

Selection of these nine sources from an initial pool of 63 candidate countries and regions is described in Figure 1 of the paper: 59 listed in the Index of World Pharmacopoeias and Pharmacopoeial Authorities plus 4 participating in the Forum for the Harmonization of Herbal Medicines (FHH) were considered; 3 had no pharmacopoeia, 45 had no English-language pharmacopoeia, 2 had no criteria for elemental impurities, 3 were difficult to obtain, and 1 referred to another pharmacopoeia; 9 remained.

Samples

No physical samples were analyzed. The units of analysis are pharmacopoeial monographs. The total number of listed herbal medicines assessed across all documents exceeded 2,000. By document, the numbers of listed herbal medicines (animal/botanical and mineral) were as follows (Table 3):

Results

Proportion of monographs with elemental impurity stipulations

For animal/botanical medicines, 100% of monographs in EP10 (254/254) and HKCMMS (322/322) carried elemental impurity stipulations. BP6 covered 95% (79/83), IP2018 covered 80% (72/90), JP18 covered 43% (75/173), KP12 covered 95% (157/165), USP43-NF38 covered 66% (68/103), and VP5 covered 9% (29/328). CP2020 covered only 5% (32/587). WHO covered all 117 items.

For mineral medicines, stipulations were rare. CP2020, JP18, KP12, USP43-NF38, and VP5 had some stipulations; HKCMMS had none; and about 75% of mineral medicines in CP2020 had no stipulations (Figure 3 of the paper). JP18 provided stipulations for about 83% of its six listed mineral medicines.

Upper limits for arsenic (animal/botanical medicines)

WHO and BP6 set a uniform upper limit of 2 ppm for all listed animal/botanical herbal medicines. HKCMMS set 2 ppm for all 322 items. KP12 set 3 ppm for all 157 items. EP10 set a uniform upper limit of 1 ppm for most items (251/254). CP2020, JP18, USP43-NF38, and VP5 set varying upper limits ranging from 0.2 to 20 ppm. EP10 stipulated a specific upper limit of 90 ppm for Kelp, explained by the high proportion of organic arsenic species (arsenobetaine and arsenosugar) in kelp, which are generally considered less toxic than inorganic arsenic (paper p. 971). ICH-Q3D recommended 1.5 ppm for arsenic.

Upper limits for lead (animal/botanical medicines)

WHO recommended 10 ppm as the upper limit for lead. No items in any document exceeded the WHO recommendation. BP6 set 5 ppm for all 83 items. HKCMMS set 5 ppm for all 322 items. KP12 set 5 ppm for all 157 items. EP10, CP2020, and USP43-NF38 also set 5 ppm for most animal/botanical medicines (253/254, 26/28, and 63/68, respectively). VP5 set varying upper limits from 1 to 10 ppm. ICH-Q3D recommended 0.5 ppm.

Upper limits for cadmium (animal/botanical medicines)

WHO recommended 0.3 ppm for cadmium. ICH-Q3D recommended 0.5 ppm. Upper limits for cadmium in most herbal medicines were above the WHO recommendation of 0.3 ppm (Figure 4c). BP6 and HKCMMS set 1 ppm as the upper limit for all 83 and 322 items, respectively. EP10 set 1 ppm for most items (251/254). CP2020 and VP5 set varying upper limits, with the highest limits of 5 and 2 ppm, respectively. KP12 and USP43-NF38 set 0.3 and 0.5 ppm for most items (145/157 and 56/65, respectively). No study documents stipulated upper limits of cadmium for mineral medicines.

Upper limits for mercury (animal/botanical medicines)

All upper limits for mercury in the study documents were below the ICH-Q3D upper limit of 3 ppm (Figure 4d). For animal/botanical medicines, BP6 set 0.1 ppm for all 83 items. EP10 set 0.2 ppm for all 254 items. HKCMMS set 0.2 ppm for all 322 items. KP12 set 0.2 ppm for all 157 items. CP2020, USP43-NF38, and VP5 set varying upper limits from 0.2 to 1 ppm; CP2020 and USP43-NF38 set 0.2 ppm and 1 ppm for most items (24/27 and 59/65, respectively). No study documents stipulated upper limits for mercury in mineral medicines.

Upper limits for total heavy metals

WHO and ICH-Q3D did not provide recommendations for total heavy metals. BP6 and IP2018 set 20 ppm for total heavy metals for all listed herbal medicines. JP18 set varying upper limits from 10 to 40 ppm. CP2020 set varying upper limits for a small fraction of listed herbal medicines, ranging from 10 to 40 ppm. KP12 set 20 ppm for Longgu, the only mineral medicine for which it set an upper limit. Upper limits for mineral medicines varied more widely than those for animal/botanical medicines.

Upper limits for mineral medicines

Most pharmacopoeial documents specified total heavy metals for mineral medicines rather than individual elements. Upper limits for mineral medicines were rarely uniform and included higher values than those for most animal/botanical medicines. USP43-NF38 specified lead upper limits for Kaolin (10 ppm) and Ground Limestone (3 ppm) but not total heavy metals.

Methods (brief)

The analysis method is a systematic documentary review. The authors examined monographs from all study documents (Table 1). For ISO 18664:2015, only the analytical methods section was investigated because the standard provides test method guidance only and does not set upper limits for individual herbal medicines. The number of monographs with elemental impurity stipulations and the test methods authorized were tabulated for each study document.

Analytical methods permitted by each pharmacopoeia were systematically compiled by element and document (Table 4 of the paper):

• ISO 18664:2015 specifies only instrumental methods: Atomic Absorption Spectrometry (AAS), Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES), Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES), and Inductively Coupled Plasma Mass Spectrometry (ICP-MS).
• JP18 and IP2018 specified only chemical methods for all herbal medicines. The chemical methods included the arsenic test, the lead test, and the heavy metal limit test.
• JP18 stipulated that AAS, ICP-AES/ICP-OES, and ICP-MS were applicable when the heavy metal limit test was not applicable due to turbidity or other reasons.
• CP2020, VP5, and BP6 adopted both instrumental methods (AAS or ICP-MS) and chemical methods for arsenic.
• USP43-NF38 adopted chemical methods for arsenic and lead.
• KP12 described use of a mercury analyzer for mercury but provided no specific analytical method.
• KP12, JP18, CP2020, VP5, and BP6 adopted the total heavy metal test method and also used chemical methods.
• In 2020, the United States Pharmacopoeial Convention replaced the heavy metal limit test in USP with the ICH-Q3D (R1) guideline method, which measures each heavy metal individually.

No detection limits (LOD/LOQ) are reported in this paper. No certified reference material (CRM) recovery data are reported. The paper does not report any experimental measurements.

Evidence Fitness

This source provides regulatory context only. It is a documentary comparison of published pharmacopoeial limits; it contains no original measurement data and no occurrence data. The source is appropriate for populating the regulatory limits section of metal pages (arsenic, lead, cadmium, mercury) and for the herbal medicine and botanical supplement pages where pharmacopoeial compliance limits are relevant. It cannot support contamination profile values such as typical concentrations, percentiles, or occurrence rates in herbal medicine matrices. Evidence fitness is EF-4 (context only).

Limitations

The authors identify several limitations (paper p. 973). Several documents containing stipulations for herbal medicines other than major pharmacopoeias were not included. Four separate pharmacopoeias for traditional medicine, Ayurveda, Siddha, Unani, and Homeopathy in India, and the Japanese Standards for Non-Pharmacopoeial Crude Drugs for herbal medicines commercially available in Japan but not listed in JP18, were excluded per the pre-defined selection criteria. The study was restricted to documents with English-language editions, which excluded 45 candidate countries and regions.

The authors also acknowledge the difficulty of establishing “one-size-fits-all” regulations for herbal medicines given regional variability in composition arising from distinct cultivars, soils, climates, and seasonality.

Implications

Certification: The compiled limits are directly relevant to HMT&C international positioning for botanical supplement and herbal medicine categories. The strictest applicable limit across all relevant jurisdictions governs. EP10 and HKCMMS consistently apply uniform limits and are among the more stringent frameworks for most elements, while JP18 and IP2018 regulate only arsenic and total heavy metals without specifying individual heavy metals other than arsenic.

Courses: Core reference for a module on pharmacopoeial regulatory frameworks and global harmonization of botanical supplement standards. Table 4 on analytical method requirements across jurisdictions is a teaching resource for QA professionals.

App: Useful for populating regulatory compliance windows on herbal/botanical ingredient pages. Cadmium limits for herbal medicines broadly exceed the WHO recommendation of 0.3 ppm.

Microbiome: Not applicable.

Provenance notes

Open access under CC BY 4.0. Published in Therapeutic Innovation and Regulatory Science, volume 57 pages 963 to 974. PMCID: PMC10400675. PMID: 37202592. PDF hash verified: c5a0c8836eb665408dff7e0dac79e68bd8a3b8fdd846129a4b8830195f9f2f45.

Authors are affiliated with the Division of Drug Development and Regulatory Science, Faculty of Pharmacy, Keio University, Tokyo, Japan (Inada and Urushihara) and the Division of Natural Medicines, Faculty of Pharmacy, Keio University (Kiuchi). The authors declare no conflicts of interest. The study received no specific funding.

Related evidence

• Arsenic
• Lead
• Cadmium
• Mercury
• Herbal botanicals
• Ingredient class — botanicals/herbs/traditional formulas
• WHO herbal-medicine contaminant guidance (2007)

Update history

No substantive edit history is available in this build. The full commit record is available in git.