Letter: retrospective reviews of liver-related case reports allegedly associated with Herbalife present insufficient and inaccurate data – authors' reply



We would like to address several points raised in the letter to the editor by Dr Appelhans et al.[1] relevant to Herbalife, in which their products were mentioned as a potential cause of liver injury in our article published in Alimentary Pharmacology and Therapeutics.[2]

The assessment of incidence and causality of hepatotoxicity from herbal products can be difficult for several reasons, such as under-reporting, unknown active ingredients, lack of specific histopathology and a ‘signature’ hepatotoxicity phenotype for some of the compounds. Yet, despite limitations, the CIOMS scale has been the most widely used causality assessment tool for drug-induced liver injury, as well as herbal hepatotoxicity and bona fide cases of such toxicity have been described in the literature.

Herbalife company has a number of products (each product often has several ingredients) distributed in many countries, with the possibility of regional variations in product quality. Thus, the products and their specific ingredients may have altered over time. Thus, to be able to pinpoint and alleged culprit hepatotoxin in Herbalife products may be very challenging.

Despite aforementioned barriers, several cases of hepatotoxicity from herbal products distributing under the name of Herbalife have been published in peer-reviewed, high-impact journals, mainly from three case series from Switzerland (10 cases), Israel (14 cases) and Spain (20 cases).[3-5] In our opinion, all of these series have adequately reported clinical information and have utilised generally-accepted causality assessment for herbal hepatotoxicity. Causality relatedness was ‘probable’ in most cases, although ‘certain’ (with positive rechallenge test) cases were also reported (two from Switzerland, three from Israel and one from Spain).[3-5] Furthermore, three cases due to possible adulteration with bacteria have also been credibly reported.[6]

We do agree that some limitations exist among these reports, such as variety of the Herbalife product ingested, non-Herbalife product co-ingestion, inadequate ingredient analysis and possibility for other causes of liver disease, but these limitations do not preclude our drawing attention to these reports. These issues have been raised by Herbalife representatives as well.[7-12]

We have searched the literatures regarding Herbalife hepatotoxicity and analysed available data carefully without bias, although some references that are letters to Editor from the manufacturer have not been cited in our review due to publishing space limitation.

In summary, we have a reason to believe that some ingredients in one of the various Herbalife products can be associated with liver injury, unless high-quality data regarding the safety of Herbalife product, other than post-marketing reports, emerges. Although the incidence is presumably very low, the possibility for Herbalife product-induced hepatotoxicity is important for practicing physicians to recognise, and it deserved to be reported in our review. Lastly, our conclusions are along the same lines as other authorities of drug-induced liver injury.[13-16]


The authors' declarations of personal and financial interests are unchanged from those in the original article.2