Toxic hepatitis induced by a herbal medicine: Tinospora crispa.
Traditional Asian herbal remedies are becoming increasingly popular in many countries. Tinospora crispa is a climbing plant used as a traditional herbal medicine in South Asia, and is considered to be medicinally interchangeable in India with the related species T. cordifolia (also called T. sinensis and T. malabarica). Use of Tinospora crispa as herbal medicine has increased over recent years, and several studies have assessed its hypoglycemic, antioxidant and analgesic properties (Sangsuwan et al. 2004). However, few data are available concerning its toxicity. We report a case of toxic hepatitis related to overuse of T. crispa pellets, with histological assessment and analytical identification of T. crispa components by reversed-phase thin-layer chromatography (RP-TLC) and reversed-phase high performance liquid chromatography with ultraviolet spectrometry (RP-HPLC-UV) of dichloromethane extracts of the pellets.
A 49-year-old male with chronic low back pain bought a herbal medicine at a market in Vietnam that was supposed to be T. crispa, and started taking 10 pellets per day. About four weeks later, on return to France, he started to experience dark urine and pale stools, asthenia and right hypochondrial pain, but continued to take the herbal remedy. Two months after starting treatment, he was referred to hospital with jaundice. He had no medical history, had not used any other drugs and had not been exposed to any chemicals over recent months. He also had no history of illicit drug use or alcohol abuse. Physical examination on admission to hospital was as follows: body temperature: 36.9[degrees]C; blood pressure: 104/70 mmHg; pulse: 70bpm; respiratory rate: 18/min; jaundice, no flapping tremor, and no hepatosplenomegaly. Additional neurological and general physical examination results were normal.
Blood tests showed aspartate aminotransferase: 1.169 IU/1, alanine aminotransferase: 2.029 IU/1, total bilirubin: 20.47 mg/dl, direct bilirubin: 13.29 mg/dl, [gamma]-glutamyltransferase: 243 IU/1, alkaline phosphatase: 81 UI/1, lactate dehydrogenase: 885 IU/1, albumin: 35.6g/1 and prothrombin time (PT): 63%. Blood glucose was normal. Viral hepatitis including hepatitis A, B, C, E, EBV, CMV, HSV, VZV, HHV-6 and HHV-8 and autoimmune hepatitis were eliminated (virus quantification by PCR). Upper abdominal ultrasound was normal. Liver biopsy showed centrilobular necrosis with inflammatory cell infiltration compatible with a toxic etiology (Fig. 1). A diagnosis of probable T. crispa induced hepatitis was established on the basis of the personal history, physical examination and laboratory findings. The herbal medicine was stopped on admission. Serum aminotransferases peaked on the second day and then started to decrease. The patient was discharged two weeks after admission and serum aminotransferases, PT and bilirubin were normal two months later.
Unused pellets provided by the patient were analyzed in order to identify the incriminated material (Fig. 2A). Semiquantitative inductively coupled plasma mass spectrometry (ICP-MS) analysis of the pellets showed no drugs or toxic metals. Microscopic characteristics of crushed pellets were compared with those of a powdered certified reference sample of T. crispa stems provided by the Hanoi University of Pharmacy, and this analysis revealed the presence of common botanical elements. Further evidence for the identification of T. crispa was provided by RP-TLC and RPHPLC-UV analyses of a dichloromethane extract of the pellets. On the basis of literature data, an improved HPLC method was developed for chemical fingerprinting of T. crispa (Cavin et al. 1998). The chromatograms were obtained for a 100mg/ml [CH.sub.2][Cl.sub.2] solution of extract. Chromatographic conditions: Zorbax[R] SB-C18 (250 mm [+ or -] 4.6 mm, 5 [micro]m) column, mobile phase: a gradient mixture of mobile phase A (MeOH) and B (water-0.05% TFA) pumped at a flow rate of 1.0ml/min (gradient program [time (min)/%B] set as 0/90, 60/20, 65/20, 70/0, 75/0, 80/90, 85/90), injection volume: 20 [micro]l, and UV detection at 254 nm. Chromatograms revealed the presence of specific markers of I. crispa i.e. N-trans--and cisferuloyltyramines, and the furanoditerpenoids borapetosides C and F (Fukuda et al. 1993), identified by comparison of their retention times (HPLC) or Rf values (TLC) with those of standards, respectively run or deposited individually. These standards are not commercially available and were isolated from T. crispa stems by column liquid chromatography (LC) and characterized by nuclear magnetic resonance (NMR) and mass spectroscopy (Fig. 2B) (Fukuda et al. 1993; Chen et al. 2010). Finally, the chromatographic profiles were found to be qualitatively similar, with similar concentrations, in the sample investigated and in the extract from the reference sample; the only difference was a higher relative content of borapetoside F (Fig. 3) in the incriminated sample.
We report a case of Tinospora-induced hepatitis with histological assessment and analytical identification of the herbal compound, with complete recovery after discontinuation of the herbal medicine. No other toxic drug or ongoing medicine was used by the patient, and no infectious or immune disease was diagnosed as a possible cause of liver injury. Histological examination and the kinetics of laboratory markers also strongly supported this causality, as liver injury started to improve on discontinuation of the herbal remedy, and had fully recovered two months later. Liver toxicity of Tinospora crispa is consistent with its composition, as high contents of furanoditerpenoids such as borapetosides have been identified (Fukuda et al. 1993). These compounds are metabolized into reactive species that cause glutathione depletion and may be responsible for hepatotoxicity (Stickel et al. 2005). In rats, repeated high doses of alcoholic extract of Tinospora crispa have been shown to cause hepatotoxicity and potential renal toxic effects (Chavalittumrong et al. 1997). No renal injury was observed in the case reported here. Recently, a randomized controlled trial was conducted to determine the efficacy of Tinospora crispa as additional treatment in patients with type 2 diabetes. In a series of twenty patients who received 1 gram of Tinospora crispa powder twice daily for 6 months, two showed a marked elevation of liver enzymes that returned to normal after discontinuation of treatment. Furthermore, one human case of hepatotoxicity was reported in 2007 with chronic overuse of a herbal preparation of Tinospora crispa stems (Denis et al. 2007).
In conclusion, this report suggests that chronic Tinospora crispa overuse is responsible for hepatotoxicity that may fully recover after discontinuation of treatment. Herbal medicines are commonly considered to be safe because they are natural products, but they are composed of multiple chemicals responsible for both therapeutic and toxic effects. In view of the growing popularity of herbal medicines and traditional remedies in many countries, more case reports and studies are required to assess the safety of these traditional remedies.
Received 24 January 2014
Received in revised form 21 February 2014
Accepted 18 April 2014
Conflict of interest
There is no conflict of interest.
Cavin, A., Hostettmann, K., Dyatmyko, W., Potterat, 0., 1998. Antioxidant and lipophilic constituents of Tinospora crispa. Planta Med. 64,393-396.
Chavalittumrong, P., Attawish, A., Chuthaputti, A., Chuntapet, P., 1997. Toxicological study of crude extract of Tinospora crispa Mierex Hook F.&Thoms. Thai J. Pharm. Sci. 21. 199-210.
Chen, C.Y., Wang, Y.D., Wang, H.M., 2010. Chemical constituents from the roots of Synsepalum dulcificum. Chem. Nat. Compd. 46,448-449.
Denis, G., Gerard, Y., Sahpaz, S., Laporte, R., Viget, N., Ajana, F., Riff, B., Mouton, Y., Bailleul, F., Yazdanpanah, Y., 2007. Prophylaxie antipaludeenne par plantes medicinales: hepatite toxique a Tinospora crispa. Therapie 62, 271-272.
Fukuda, N., Yonemitsu, M., Kimura, T., 1993. Isolation and structure elucidation of the five furanoid diterpene glycosides borapetoside C-G. Liebigs Ann. Chem. 12, 491-495.
Sangsuwan, C., Udompanthurak, S., Vannasaeng, S., Thamlikitkul, V., 2004. Randomized controlled trial of Tinospora crispa for additional therapy in patients with type 2 diabetes mellitus. J. Med. Assoc. Thai. 87,543-546.
Stickel, F., Patsenker, E., Schuppan, D., 2005. Herbal hepatotoxicity. J. Hepatol. 43, 901-910.
J. Langrand (a,b), *, H. Regnault (d), X. Cachets C. Bouzidi (c) A.F. Villa (a,b), L. Serfaty (d), R. Garnier (a,b), S. Michel (c)
(a) Centre Antipoison de Pahs, AP-HP Hopital Lariboisiere, Pahs, France
(b) Universite Paris Diderot, Sorbonne Paris Cite, France
(c) Laboratoire de Pharmacognosie, UMR 8638 CNRS, Faculte de Pharmacie, Universite Paris Descartes, Sorbonne Paris Cite, Paris, France
(d) Service d'hepatologie, AP-HP Hopital Saint Antoine, Paris, France
* Corresponding author at: Universite Paris Diderot, Sorbonne Paris Cite, France. Tel.: +33 1400544328.
E-mail address: firstname.lastname@example.org (J. Langrand).