Start with two dates that sit badly together.
In 2000, China’s national drug regulator began stripping the medicinal standards from a group of herbs that contain aristolochic acids, and the Chinese Pharmacopoeia dropped the entries for a series of related plants. Twenty years later, in December 2020, the U.S. Food and Drug Administration published a standing order telling import staff to hold shipments of those same botanicals at the border without testing them first.
Read those two facts side by side and the usual argument about Chinese medicine falls apart. The country most often described as careless with its traditional remedies removed the most dangerous ones from its own official list a quarter of a century ago. The country with the stricter reputation still needs a border alert, because the plants keep arriving anyway.
That gap is the real subject here. It is not old knowledge against laboratory science. It is a regulated national pharmacopoeia on one side and an export market nobody governs on the other, and almost everything a U.S. reader needs to know sits inside it.
The herbs, the kidneys, and the name that let them in
Aristolochic acids come from plants in the birthwort family. In the early 1990s a group of women in Belgium who had taken weight-loss preparations developed rapid, irreversible kidney failure. The preparations were supposed to contain one plant. They contained another.
FDA’s own account of how this happens is plainer than most summaries of it. The agency says the plant named on the label was mis-identified, and that products were confused with aristolochic-acid-containing material because the traditional Chinese names were misinterpreted or similar. The alert’s botanical table shows how crowded those names are. It lists guan mu tong and guang mu tong as common or other names for Aristolochia species, guang fang ji and fang chi for Aristolochia fangchi, and xi xin for two species of Asarum. Every one of those is a name the table attaches to a plant on the alert’s own list, not to the benign plant a buyer thought they were getting.
That is the mechanism, and it matters because it is not a story about a traditional prescription doing what it was supposed to do. Two plants, one benign and one carcinogenic, travelled under names close enough to be confused, and the supply chain could not tell them apart. Verifying which species is in a package is its own subject, handled in this library’s article on who checks how a botanical ingredient was collected.
The Belgian cases were also not the end of it. The same compounds turn up in products sold as dietary supplements decades later, which is why the border alert exists.
The word nephropathy hides what these products do, so use FDA’s own language instead. The agency states that products containing aristolochic acid cause renal damage and can cause or contribute to renal failure, and that the serious effects are irreversible. The Belgian patients in the published cohorts had already reached kidney transplant. Cancer of the urinary tract lining came after that, on top of the kidney disease.
What the federal carcinogen file records
The National Toxicology Program, part of the National Institute of Environmental Health Sciences, lists aristolochic acids in the Report on Carcinogens as known to be human carcinogens. The copy published as the Fifteenth Edition records that the listing dates to the Twelfth Report in 2011. That is the strongest classification the program issues, and it is reserved for substances with human evidence.
The underlying numbers are stark. Two prevalence studies of Belgian women who developed aristolochic acid nephropathy and went on to receive kidney transplants found urothelial cancer, mostly of the upper urinary tract, in roughly 40 to 46 percent of them. The report notes that neither study had an unexposed comparison group, which is a real limit and belongs in any honest account.
A second line of evidence comes from Balkan endemic nephropathy, a kidney disease in rural Danube communities. Among patients who had both that disease and upper-tract urothelial cancer, 78 percent carried a distinctive mutation pattern in the p53 tumor suppressor gene.
Taiwan supplies the third. A 2012 study in the Proceedings of the National Academy of Sciences examined 151 patients with upper urinary tract cancer and 25 controls with kidney cancer. Taiwan has the highest reported incidence of that cancer anywhere in the world. Among patients carrying the signature mutation, 83 percent had aristolactam DNA adducts in the kidney cortex, a chemical fingerprint of exposure.
A liver-cancer paper, a rebuttal, and a meeting in Beijing
In October 2017 a team publishing in Science Translational Medicine sequenced 98 liver cancers from two Taiwanese hospitals and reported that 78 percent carried the aristolochic acid mutational signature. Searching 1,400 liver cancers from elsewhere, they found the signature in 47 percent of Chinese cases, 29 percent of Southeast Asian, 13 percent of Korean, 2.7 percent of Japanese, 4.8 percent of North American and 1.7 percent of European ones.
Within weeks, two researchers at the Chinese Academy of Sciences published a rebuttal arguing that the same mutation pattern is produced by other carcinogens, naming tobacco-related 4-aminobiphenyl, 1,3-butadiene, vinyl chloride and two chemotherapy drugs. Without knowing what each patient was exposed to, they wrote, the conclusion was unpersuasive. Both papers are in print and neither has retracted.
The most revealing document is the third one. A 2017 paper in the China journal of Chinese materia medica records that the State Administration of Traditional Chinese Medicine convened a conference on aristolochic acid toxicity as soon as the liver-cancer article appeared. Its experts listed objections: no clear exposure history, no established tumor-producing dose or latency, hepatitis B unaccounted for, small samples. The same paper is where the regulatory chronology comes from. It states that China’s drug administration had been abolishing the medicinal standards for aristolochic-acid herbs since 2000 and that the pharmacopoeia had removed the records.
So the evidence on liver cancer sits in the category of measured but contested, where competent people read the same data differently. The evidence on kidney damage and upper urinary tract cancer does not. That one is settled, and the federal carcinogen listing says so.
Notice what the Beijing conference was not. It was not a denial that the compounds are toxic. The objections were about one study’s design, and the same paper that records them also records that the regulator had already removed the herbs. A system can accept that an ingredient is dangerous, delist it, and still argue about a specific paper’s methods. Treating the argument as proof of indifference misses what the document says.
A related figure gets stretched past what it can carry. The U.S. Drug-Induced Liver Injury Network, which enrolled patients at eight referral centers from 2004 to 2013, found that 130 of 839 enrolled subjects had liver injury caused by herbals and dietary supplements, and that the share of cases attributed to those products rose from 7 percent to 20 percent across the study period. Non-bodybuilding products led to death or transplant in 13 percent of cases against 3 percent for conventional medications. Those are real numbers about a regulatory category that also includes bodybuilding supplements, green tea extract and much else. They are not a measurement of Chinese herbal medicine, and writing that Chinese medicine causes a fifth of American liver injury misreports the study.
What a border alert does
Import Alert 54-10 covers bulk and finished dietary supplement products containing aristolochic acid. The mechanism has a dull name and a sharp effect: detention without physical examination.
Normally FDA samples a fraction of incoming shipments and lets the rest through. Under this kind of alert, the burden flips. A covered shipment is held on arrival, and the importer has to produce evidence that it is free of the substance before it moves. Nobody at the port has to prove the shipment is contaminated first.
The scope is broad. It covers any species of Aristolochia, fourteen named Asarum species, Cocculus orbiculatus and Thottea siliquosa. The alert states plainly that products containing aristolochic acid cause renal damage and can cause or contribute to renal failure, and warns that a product with a small amount may be used for years before irreversible harm appears.
The alert also carries a list of firms. The entry on the attachment is a pharmaceutical company in Tainan City, Taiwan, with a product recorded as Xi Xin and classified as a dietary supplement, published in September 2009. FDA maintains a second standing alert under the title Chinese Herbal Medicines, numbered 66-10 and listed in the agency’s own industry index, though the detail page would not load for this article. How recalls and import enforcement work in general is covered elsewhere in this library.
What “regulated in China” looks like from here
It is easy to picture Chinese medicine in China as a folk practice happening around the edges of hospitals. The scale says otherwise. A Xinhua report carried on the English site of China’s National Health Commission in October 2020 put the number of traditional Chinese medicine institutions in the country at 65,809 at the end of 2019, attributing that figure to the National Administration of Traditional Chinese Medicine. The same report says registered clinics had reached 19,000 by September 2020, and names no source at all for that one.
Those are the most recent official English-language figures this article could find, and the honest statement is that there is no current series. Searches of China’s own English-language government sites produced nothing newer for hospitals or licensed practitioners. A number with a 2019 date attached is better reporting than a number with no date at all, and a reader is entitled to know that the trail stops there.
What this matters for is not national pride. It is that the products arriving in American mailboxes were manufactured somewhere, and the regulatory system covering that somewhere decided two decades ago that some of these plants had no place in medicine. The products still sold under the old pharmacy names are, by that system’s own standards, outside it.
What the page most readers find leaves out
Search for official U.S. information on this subject and you land on the National Center for Complementary and Integrative Health, whose consumer page was last updated in April 2019.
The page is not silent about safety. It says some Chinese herbal products have been contaminated with toxic compounds, heavy metals including arsenic, lead and cadmium, pesticides and microorganisms. It names undeclared pharmaceuticals found in such products, specifically the blood thinner warfarin and the anti-inflammatory diclofenac. It says manufacturing errors in which one herb is mistakenly replaced with another have caused serious complications.
What it never says is aristolochic acid. Three separate reads of that page, each asking for different terms, found no mention of aristolochic acid, Aristolochia, nephropathy, the kidney, the liver or hepatotoxicity anywhere on it.
That is the position a reader is in. The single best-documented harm in this entire subject, carrying a federal carcinogen listing and a live import alert from a sister agency, does not appear on the federal consumer page about the subject. Finding it means going past that page to the toxicology program and to FDA.
The system in its own words, and what gets tested
Chinese medicine is not primarily a list of herbs. It is a diagnostic method, and the clearest description of it comes from inside the tradition rather than from outside commentary.
Researchers at the University of Hong Kong’s School of Chinese Medicine, writing in the journal Chinese Medicine in 2023, describe syndrome differentiation as a unique and important component of practice that guides both diagnosis and treatment. In the framework they tested for heart failure, a practitioner sorts a patient into syndrome elements: qi deficiency, yang deficiency, yin deficiency, blood stasis and phlegm retention, assessed partly through examination of the tongue. Qi deficiency was the dominant pattern in their sample. These are the tradition’s own categories, reported here as the tradition’s categories, and they are not claims about anatomy.
That same paper names the difficulty from the inside. Its authors identify the diversity of syndrome differentiation criteria and the breadth of syndrome patterns as the major bottleneck for research, which is why they built and validated a questionnaire across 661 heart failure patients in the first place. When they compared the syndromes their model produced against characteristic tongue images, 91 percent matched. That is internal consistency within one framework, not external proof that the framework tracks disease.
Two things follow. First, the federal consumer page defines Chinese medicine by its practices, naming acupuncture, tai chi and herbal products, and never explains qi or yin and yang at all. Second, almost nothing in the research literature tests the system as a system. Trials test one piece at a time.
Graded piece by piece, the picture is uneven. For Chinese herbal products as a class there is insufficient evidence to assess effectiveness, and the federal page says the reason is study quality rather than negative results. For acupuncture there is positive trial evidence with genuinely contested interpretation, which this library handles in its article on acupuncture. Tai chi has the strongest record of anything in this family and is covered in its own article.
The World Health Organization added an optional traditional medicine chapter to the eleventh revision of its disease classification. In November 2019 the European Academies’ Science Advisory Council and the Federation of European Academies of Medicine published a joint statement calling that inclusion premature, and wrote that “Concepts of body and disease used in TCM have not been substantiated by conventional scientific investigation.” What a disease classification does and does not mean is the subject of the sibling article on institutions, which owns that question.
The animals in the medicine cabinet
Some traditional preparations use animal parts, and the international wildlife treaty has a standing resolution about exactly that. Resolution Conf. 10.19, amended at the twelfth and fourteenth Conferences of the Parties, opens by recognizing that wild fauna and flora are used in many forms of traditional medicine and that uncontrolled use of several endangered species has been a subject of concern.
Its posture is worth noticing. Rather than condemning the practice, the resolution recommends that countries work closely with groups of traditional-medicine practitioners and consumers on public education, investigate alternative ingredients, and consider artificial propagation where that would relieve pressure on wild populations.
The pangolin shows what the stakes look like. At the 2016 Conference of the Parties, delegates voted 114 to 1 with 5 abstentions to move the Chinese and Sunda pangolin to the treaty’s highest protection tier, which prohibits international commercial trade. The U.S. Fish and Wildlife Service reported that roughly 18,500 kilograms of pangolin scales were seized from illegal shipments between 2013 and 2016, representing somewhere between 5,100 and 39,000 individual animals depending on the species. The agency named traditional Asian medicine as the primary driver.
Before you take anything from this cabinet
Nothing here tells you to start or stop any treatment. That decision belongs with a clinician who knows your history. What follows is a set of questions worth asking first, built for this article.
- Can the seller name every plant in the product by botanical name, not just by a Chinese pharmacy name? A single pharmacy name can cover more than one species.
- Does any ingredient belong to Aristolochia, Asarum, Cocculus or Thottea? Those four are the ones FDA holds at the border.
- Was the product bought from a retailer who will still answer the phone if a problem appears? A website that vanishes takes the lot number with it.
- Have you handed the actual package, not a description of it, to the clinician who manages your prescriptions? Undeclared warfarin and diclofenac have both turned up in products of this kind.
- Is the product being sold to treat a diagnosed disease? A disease claim changes what the product legally is, and nothing sold that way has been reviewed for the claim it makes.
None of those questions is answerable from a marketing page, which is the point of asking them. Two of them a careful seller can answer in a sentence. A seller who cannot name the species has told you something useful.
Some symptoms deserve same-week attention rather than watchful waiting, especially in anyone taking imported herbal products. Flank or back pain, blood in the urine, a drop in kidney function on a routine blood panel, yellowing of the eyes or skin, dark urine, pale stools or unexplained exhaustion all belong in front of a physician, with the product in hand. Kidney damage from aristolochic acid is not reversible, and the carcinogen file records that cancer can appear years after exposure stops.
The most useful thing to take from all of this is narrower than the debate that surrounds it. The question is not whether a remedy is traditional. Traditional use is history, and history is not evidence of safety or of benefit. The question is whether anyone in the chain between a hillside and your kitchen can say, on paper, which plant is in the box. In the documented cases that ruined kidneys, nobody could.
Sources
- National Center for Complementary and Integrative Health, National Institutes of Health. “Traditional Chinese Medicine: What You Need To Know (content last updated April 2019; verified in three separate reads that the page never mentions aristolochic acid, Aristolochia, nephropathy, kidney, liver or hepatotoxicity).” nccih.nih.gov
- U.S. Food and Drug Administration. “Import Alert 54-10, Detention Without Physical Examination of Bulk/Finished Dietary Supplements Products Containing Aristolochic Acid (published 12/10/2020; charge statement revised 04/23/2019).” accessdata.fda.gov
- U.S. Food and Drug Administration. “Import Alerts for Industry 66, Drugs (index listing Import Alert 66-10, Chinese Herbal Medicines).” accessdata.fda.gov
- National Toxicology Program, National Institute of Environmental Health Sciences. “Report on Carcinogens, Fifteenth Edition: Aristolochic Acids (listed as known to be human carcinogens; first listed in the Twelfth Report on Carcinogens, 2011).” ntp.niehs.nih.gov
- U.S. Fish and Wildlife Service. “CITES unites to change fate of pangolins (press release, September 28, 2016).” fws.gov
- Convention on International Trade in Endangered Species of Wild Fauna and Flora. “Resolution Conf. 10.19 (Rev. CoP14), Traditional medicines.” cites.unia.es
- World Health Organization. “Frequently Asked Questions: Traditional Medicine (ICD-11 supplementary chapter; optional dual coding, excluded from the mortality use case).” who.int
- European Academies' Science Advisory Council and Federation of European Academies of Medicine. “Traditional Chinese Medicine: A Statement by EASAC and FEAM (November 2019).” ria.ie
- National Health Commission of the People's Republic of China. “TCM medical institutions total 65,809 by the end of 2019 (Xinhua report, updated October 26, 2020, attributing figures to the National Administration of Traditional Chinese Medicine).” en.nhc.gov.cn
- Gao Y, Xiao X-H, Zhu X-X, Liang A-H, Zhang B-L. “Study and opinion on toxicity of aristolochic acid. Zhongguo Zhong Yao Za Zhi, 2017;42(21):4049-4053. PubMed. DOI: 10.19540/j.cnki.cjcmm.2017.0160.” pubmed.ncbi.nlm.nih.gov
- Chen C-H, Dickman KG, Moriya M, et al.. “Aristolochic acid-associated urothelial cancer in Taiwan. Proc Natl Acad Sci U S A, 2012;109(21):8241-6. PubMed. DOI: 10.1073/pnas.1119920109.” pubmed.ncbi.nlm.nih.gov
- Ng AWT, Poon SL, Huang MN, et al.. “Aristolochic acids and their derivatives are widely implicated in liver cancers in Taiwan and throughout Asia. Sci Transl Med, 2017;9(412):eaan6446. PubMed. DOI: 10.1126/scitranslmed.aan6446.” pubmed.ncbi.nlm.nih.gov
- Zhou G, Zhao X. “Carcinogens that induce the A:T > T:A nucleotide substitutions in the genome. Front Med, 2017;12(2):236-238. PubMed. DOI: 10.1007/s11684-017-0611-y.” pubmed.ncbi.nlm.nih.gov
- Navarro VJ, Barnhart H, Bonkovsky HL, et al.. “Liver injury from herbals and dietary supplements in the U.S. Drug-Induced Liver Injury Network. Hepatology, 2014;60(4):1399-408. PubMed. DOI: 10.1002/hep.27317.” pubmed.ncbi.nlm.nih.gov
- Leung AYL, Zhang J, Chan CY, et al.. “Validation of evidence-based questionnaire for TCM syndrome differentiation of heart failure and evaluation of expert consensus. Chin Med, 2023;18(1):70. PubMed. DOI: 10.1186/s13020-023-00757-1.” pubmed.ncbi.nlm.nih.gov