Mania associated with herbal medicines, other than Cannabis: a systematic review and quality assessment of case reports

Background: DSM-5 introduced the diagnostic category of substance/medication-induced bipolar and related disorder. This systematic review examines published reports linking mania with the consumption of herbal medicines (HM), excluding cannabis. Putative pathophysiological mechanisms that may account for the reported HM being associated with mania are discussed.

Methods: A systematic search of EMBASE, CINAHL, Health Source, PsychINFO, and PubMed. The quality of case reports meeting inclusion criteria was assessed using the modified Quality Assessment Scale by Agbabiaka.

Results: Nineteen single and seven multiple-case reports met inclusion criteria. These yielded a study sample of 35 case reports, 28 of herbal medicine associated mania, 5 of hypomania, and two mixed states, in 17 females [age in years M(SD) = 43.1(13.2)] and 18 males [40.7(18.1)]. A total of 11 herbal medicines were implicated. Case reports by herbal medicine (number of reports) comprised: St John's wort (Hypericum perforatum) (14); Ginseng (Panax ginseng) (5); brindleberry (Garcinia cambogia) (4); ma-huang (Ephedra sinica) (3); “herbal slimming pills” (2); Herbalife products (2); Hydroxycut (1); horny goat weed (Epimedium grandiflorum) (1); “herbal body tonic” (1); celery root (Apium graveolans) (1), and a “herbal mixture” (1). All case reports were associated with use rather than withdrawal of herbal medicines. Only one case report was rated for probability of association using a standardized algorithm. Laboratory assays to confirm composition of the herbal preparation were reported in only one article describing two cases and indicating admixture of a likely causal pharmaceutical in the herbal preparation.

Conclusions: Causal attributions are problematic given the limited number of reports, antidepressant co-prescribing in 7 cases, insufficient data regarding pattern and type of herbal medicine use, and lack of a reference frequency for spontaneous mania.The quality assessment scores across the 26 papers (35 case reports) were as follows: low quality (0), lower-medium quality (9), upper-medium quality (10) and high quality (7). Putative pathophysiological mechanisms were postulated for nine of the 11 herbal medicines and centered on HPA-axis activation and increased monoamine activity. Systematic study of the association between herbal medicines and the course of bipolar disorder may contribute to defining targets for pathophysiological research.