Hospitalized cases
Four acute liver-injury cases were investigated in Hong Kong.
Executive summary
A real but bounded safety concern
HemoHIM has a documented 2021 Hong Kong safety signal involving four cases of acute liver injury among users of the implicated product and laboratory detection of undeclared methoxsalen in samples collected during the investigation.
Between April and September 2021, four women aged 42 to 72 presented to Hong Kong public hospitals with acute liver injury after reporting HemoHIM use for approximately two weeks to six months. Reported symptoms included tea-coloured urine, jaundice and vomiting. All four were admitted to hospital. Hong Kong's Department of Health obtained two patient-associated product samples, and the Government Laboratory confirmed methoxsalen in both. Source 1
On November 2, 2021, the Department of Health reported that a HemoHIM sample obtained from an Atomy premises in Cheung Sha Wan also contained methoxsalen. According to information Atomy provided to the Department of Health, the product had been manufactured in Korea and imported for local distribution through the company's website. Source 2
A later clinical report in the Hong Kong Medical Journal stated that liver function improved after HemoHIM was discontinued and that no alternative medical causes were identified. Laboratory testing detected methoxsalen and psoralen, and subsequent analyses of samples from different sources produced consistent findings. The authors noted that the listed HemoHIM ingredients should not contain psoralen or methoxsalen and called for further investigation into why those compounds were present. Source 3
Scientific interpretation:
Taken together, these findings constitute a credible signal of product-associated liver injury in the investigated Hong Kong setting. They do not establish that every HemoHIM formulation, production lot or currently marketed product contains methoxsalen or causes liver injury.
Methoxsalen detected
Government testing identified undeclared methoxsalen in official product samples.
Short-term reassurance
Animal and human studies of defined preparations did not identify major short-term safety signals.
No universal conclusion
Current lots, rare events, long-term use and excluded populations remain separate questions.
Hong Kong investigation
Clinical cases plus product testing
April to September 2021
Four women admitted with acute liver injury
Patients were aged 42 to 72 and had reportedly used HemoHIM for roughly two weeks to six months.
November 1, 2021
Two patient-associated samples tested positive for methoxsalen
The Hong Kong Department of Health reported that Government Laboratory testing confirmed methoxsalen in both samples obtained during the investigation.
November 2, 2021
A further sample from an Atomy premises also tested positive
Hong Kong authorities reported methoxsalen in a HemoHIM sample obtained from an Atomy premises in Cheung Sha Wan.
Subsequent clinical publication
Methoxsalen and psoralen reported, with improvement after withdrawal
The HKMJ report described improvement after HemoHIM cessation, no identified alternative medical causes and consistent findings across additional samples.
Why methoxsalen mattered:
In Hong Kong, methoxsalen is classified as a Part 1 poison and prescription medicine under the Pharmacy and Poisons Ordinance. The Department of Health noted that it is used medically for conditions including psoriasis and vitiligo.
Source 2
What the investigation established
The combination of clinical cases, improvement after stopping the product, lack of an identified alternative medical cause and chemical findings in investigated samples is stronger than an isolated spontaneous adverse-event report.
What it did not establish
The Hong Kong investigation did not demonstrate that every HemoHIM product worldwide contained methoxsalen or psoralen, that current production lots contain the same compounds, that all users face the same risk or that a specific manufacturing or agricultural mechanism explains the unexpected chemicals.
Important boundary:
Explanations such as botanical variation, ingredient substitution, cross-contamination or manufacturing differences remain hypotheses unless supported by direct evidence from the relevant production chain.
Singapore pharmacovigilance
Additional signal, lower evidentiary certainty
The Singapore Health Sciences Authority's May 2023 bulletin reported two adverse-event reports involving raised liver enzymes in users of HemoHIM. Source 6
How to interpret these reports:
They show that raised liver enzymes in HemoHIM users were also reported to Singapore's pharmacovigilance system. They do not establish causality, incidence, relative risk, the affected product lot, product chemistry or the frequency of liver injury among users.
Spontaneous adverse-event reporting systems are designed primarily to detect possible safety signals. They cannot, by themselves, determine how often a product causes a reaction.
The previously cited assertion of a May 2022 HemoHIM liver-enzyme report remains excluded unless the original primary bulletin is located and independently verified.
Controlled research
What animal and human studies contribute
2024 rat toxicology study
A 2024 study evaluated acute toxicity, 28-day repeated exposure, 13-week repeated exposure and genotoxicity using a standardized HemoHIM preparation. Source 4
In the 13-week study, rats received doses up to 2,000 mg/kg/day. The authors reported no test-item-related clinical signs or mortality at the tested doses and established a reported NOAEL of 2,000 mg/kg/day in both sexes.
The bacterial reverse-mutation, chromosome-aberration and mouse-micronucleus assays were reported as negative.
Appropriate conclusion:
The tested standardized HemoHIM preparation produced no adverse findings considered toxicologically significant under the doses and study conditions examined, and the specific genotoxicity assays performed were negative.
2025 randomized human safety study
The study randomized 96 healthy adults aged 30 to 59, with 48 allocated to HemoHIM and 48 to placebo. Participants assigned to HemoHIM consumed 40 g per day for eight weeks using a liquid study formulation containing 50 percent HemoHIM extract. Source 5
96 participants
48 HemoHIM and 48 placebo.
90 participants
47 HemoHIM and 43 placebo.
Intention-to-treat
The authors reported analysis including all 96 randomized participants.
8 weeks
40 g/day of the active liquid preparation.
The researchers reported no adverse events in the HemoHIM group and no clinically significant adverse changes in vital signs, hematology, blood chemistry or urinalysis.
Alkaline phosphatase increased by approximately 4.9 U/L in the HemoHIM group. The change reached statistical significance, but values remained within the stated normal range, and the investigators regarded the change as clinically insignificant.
Appropriate conclusion:
The study provides useful evidence of short-term tolerability for the tested formulation among selected healthy adults.
Why the controlled studies do not settle every safety question
The animal findings apply to the standardized preparation, species, dose range and study durations tested. They do not establish long-term human safety or exclude rare human idiosyncratic reactions.
The human trial also used a selective population. Participants with relevant disease or disease-related medication use, pregnancy or breastfeeding, regular herbal or supplement use and several other characteristics were excluded.
A study with 48 participants initially assigned to HemoHIM and eight weeks of follow-up cannot meaningfully rule out rare adverse reactions or effects that emerge only after longer exposure.
The study was funded by Kolmar BNH, and three authors were affiliated with the company's Food Science R&D Center. This relationship should be disclosed, but it does not by itself invalidate the results.
Evidence hierarchy
What each evidence source can and cannot support
| Evidence source | What it can support | What it cannot establish |
|---|---|---|
| Hong Kong clinical cluster plus chemical testing | A credible historical signal involving acute liver injury and methoxsalen-containing investigated product samples. | Frequency among all users, universal causation or composition of all global lots. |
| Hong Kong distributor-premises sample | Methoxsalen was also found in an official sample obtained from an Atomy premises. | That all Atomy-distributed HemoHIM contained methoxsalen. |
| Singapore spontaneous reports | A separate pharmacovigilance signal involving raised liver enzymes. | Incidence, confirmed causality, relative risk or affected-lot chemistry. |
| 2024 rat toxicology | Safety findings for the tested standardized preparation, doses, species and duration. | Long-term human safety or absence of rare human reactions. |
| 2025 randomized trial | Short-term tolerability of a specified formulation among selected healthy adults. | Safety in excluded populations, rare events, long-term use or all marketed formulations. |
| FDA NDI review | Adequacy of the safety information submitted for the proposed new dietary ingredient. | Proof that all marketed HemoHIM was unsafe, contaminated or responsible for illness. |
Core principle:
A small controlled trial can find good short-term tolerability while post-market surveillance identifies uncommon serious events that the trial is too small or too short to detect. These findings are not necessarily contradictory.
United States regulatory review
What the FDA action actually means
The FDA history requires careful separation of two source types. The FDA New Dietary Ingredient inventory establishes notification numbers and response dates. The actual October 18, 2021 FDA response letter is required to support substantive statements about deficiencies in the HemoHIM submission.
August 17, 2021
HemoHIM Herbal Extract submission date.
October 18, 2021
Response date recorded in the FDA inventory.
December 20, 2022
Subsequent HemoHIM submission date.
February 24, 2023
Response date for the subsequent notification.
The October 2021 response letter raised concerns regarding the adequacy of information supplied to establish identity and a reasonable expectation of safety under the proposed conditions of use. Issues discussed included botanical identity, manufacturing information, controls for unwanted impurities, differences involving Paeonia species cited in supporting evidence and limitations in the toxicological and human safety information. FDA letter
Regulatory interpretation:
FDA did not make a blanket determination that every marketed HemoHIM product was unsafe. The issue was whether the New Dietary Ingredient notification provided an adequate safety basis for the proposed ingredient and conditions of use.
The FDA inventory records the notification numbers and response dates. FDA inventory
Photographed Philippine carton
What the packaging can and cannot establish
The photographed carton appears to display the notation FR. No. 4000001300251 and For Philippines Only.
This observation is limited to what is visibly printed on the packaging. It should not be interpreted as independent confirmation that the registration remains active, that the number applies to the exact formulation currently inside the carton, that the product satisfies all current regulatory requirements or that the individual production lot has undergone testing for methoxsalen, psoralen or other unexpected compounds.
Evidence distinction:
Registration status and lot chemistry are different questions. Even a valid registration does not constitute a chemical analysis certificate for every production lot.
Consumer interpretation
Caution without overstatement
A person using HemoHIM who develops symptoms potentially consistent with liver injury, including jaundice, tea-coloured or dark urine, unusually pale stool, persistent nausea or vomiting, marked fatigue, abdominal discomfort or generalized itching, should stop using the product and seek medical assessment promptly.
The Hong Kong experience shows why supplement use should be disclosed when clinicians investigate unexplained liver injury. Source 1 Source 3
If an adverse reaction is suspected
Sealed sachets
Keep remaining unopened units if product testing becomes necessary.
Carton and lot details
Retain batch, lot and expiry information.
Purchase records
Keep receipts or transaction information where available.
Medical context
Disclose supplement use during clinical assessment.
People who are pregnant or breastfeeding, have liver disease, previously experienced drug or supplement-related liver injury or take multiple prescription medicines were not adequately represented in the controlled HemoHIM safety study. Consultation with a physician or pharmacist before use is therefore reasonable. Source 5
This recommendation reflects limited product-specific evidence in these populations. It should not be interpreted as proof that HemoHIM necessarily harms every person in these groups.
Conclusion
The scientifically defensible position
The evidence supports a real but bounded safety concern. In Hong Kong in 2021, four women developed acute liver injury after using the implicated HemoHIM product. Hong Kong authorities confirmed methoxsalen in two patient-associated product samples and subsequently in another HemoHIM sample obtained from an Atomy premises. A later clinical report documented improvement after HemoHIM withdrawal, found no identified alternative medical causes and reported detection of methoxsalen and psoralen, with consistent findings in additional samples from different sources.
The evidence does not establish that every HemoHIM formulation, current manufacturing lot or product sold in every country contains methoxsalen or psoralen. It also does not show that liver injury is inevitable or common among users. A 13-week rat study of a standardized preparation and an eight-week randomized human trial of a specified liquid formulation found no major toxicological or short-term clinical safety signal under their respective study conditions.
The most accurate consumer message is caution without overstatement. The 2021 Hong Kong event is sufficiently documented to warrant clinical attention and continued product-quality scrutiny. It should not be generalized into a claim that every current HemoHIM product is contaminated or inherently harmful to the liver. Establishing the status of a particular product requires current regulatory information and, where warranted, independent analysis of sealed units from the relevant production lot.
Primary sources
Source file for publication
1. Hong Kong Department of Health
DH investigates suspected poisoning cases relating to oral product, November 1, 2021.
https://www.info.gov.hk/gia/general/202111/01/P2021110100785.htm
2. Hong Kong Department of Health
Public urged not to buy or consume oral product "Hemohim" containing undeclared controlled substance, November 2, 2021.
https://www.info.gov.hk/gia/general/202111/02/P2021110200746.htm
3. Hong Kong Medical Journal
Liver injury associated with the use of health supplement HemoHIM.
https://www.hkmj.org/abstracts/v28n5/413b.htm
4. HemoHIM toxicology study
Evaluation of acute, repeated dose 28-day and 13-week oral toxicity and genotoxicity of HemoHIM.
https://pmc.ncbi.nlm.nih.gov/articles/PMC11717728/
5. HemoHIM randomized controlled safety study
Safety and Tolerability of HemoHIM: A Randomized, Placebo-Controlled Study, 2025.
https://pmc.ncbi.nlm.nih.gov/articles/PMC12324994/
6. Singapore Health Sciences Authority
Health Product Safety Information Summary, May 2023.
HSA safety bulletin PDF
7. U.S. Food and Drug Administration
New Dietary Ingredient Notification inventory.
FDA NDI inventory
8. U.S. Food and Drug Administration
NDI 1215 response letter, HemoHIM Herbal Extract, October 18, 2021.
FDA response letter