Misconceptions regarding the association between Herbalife products and liver-related case reports in Spain.

摘要

In the July 2011 issue of Pharmacoepidemiology and Drag Safety, the editorial by Manso et al discussed historical reports of consumers from Spain who had reportedly taken Herbalife products and experienced liver-related injury of varying types and severity. Herbalife would like to offer an alternative perspective which severely weakens the authors' conclusions. First, the authors state that Spanish State regulatory authorities published a report informing consumers of the risk of hepatotoxicity associated with Herbalife products. In an egregious omission, the authors failed to state that this communicado was removed by the Spanish authorities on April 15, 2009 after an extensive review of the facts and discussion involving most of these case reports with additional information that Herbalife provided to aid in the investigation. Furthermore, most of these same cases were shared by Spanish authorities with regulatory agencies globally through the World Health Organization's (WHO) UPSALA database and have not led to regulatory action by any governments. Another glaring omission by Manso et al was the fact that the WHO causality classifications were previously applied to most of these same cases, yet were not included in the authors' paper. The failure to acknowledge these assessments and instead applying the Karch/Lasagna criteria, is a major deficiency in this publication, and raises questions about the authors' scientific balance given the serious and unsubstantiated accusations about a company's whole product line. This is especially disconcerting since applying different causality assessment criteria leads to very different causality scores for a number of the cases. In fact, since the authors could not identify a causative agent with a 'known drug response' in any of these cases, none of the cases in this article would have met the Karch/Lasagna criteria for a definite, probable, or possible classification since a 'known drug response' is required for such a classification.