THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE Volume 4, Number 1, 1998, pp. 9-13 Mary Ann Liebert, Inc.

Misuse of Herbal Remedies: The Case of an Outbreak of Terminal Renal Failure in Belgium

(Chinese Herbs Nephropathy)

JEAN-LOUIS VANHERWEGHEM, M.D., Ph.D.

ABSTRACT

At least 100 cases of extensive interstitial fibrosis of the kidneys were observed in Belgium in women who had followed a weight-loss regimen that included the use of Chinese herbs. The possible relation between the renal disease and these Chinese herbs was investigated. It was shown that the prescribed Chinese herb called Stephania tetrandra was, in fact, inad­ vertently replaced by another Chinese herb, namely Aristolochia fangchi in the powdered ex­ tracts delivered in Belgium and in France. The development of renal disease in about 100 pa­ tients exposed to the so-called Stephania tetrandra stresses the need for more stringent control of herbal medicine.

INTRODUCTION tween the renal disease and these Chinese herbs was thus investigated and established In early 1992, two women, younger than 50(But , 1993; Van Haelen et al., 1994; Schmeiser years old, presented to us with extensive in­ et al., 1996). The development of the renal dis­ terstitial fibrosis of the kidney without ease in about 100 patients exposed to the so- glomerular lesions that rapidly progressed to called Stephania tetrandra stresses the need for terminal renal failure (Vanherweghem et al., better control of herbal medicine. 1993). These women, who had no previous his­ tory of renal disease, had followed, just before the renal disease, the same weight-loss regimen DESCRIPTION OF CHINESE in the same medical clinic. This clinic special­ HERBS NEPHROPATHY ized in weight-loss regimens for more than 15 years and no renal problems had been previ­ Epidemiology ously encountered. Interestingly, this diet reg­ imen consisted of a preparation containing a The exact number of patients exposed to Chi­ mixture of acetazolamide, fenfluramine, and nese herbs in Belgium is not known exactly. As­ various animal and vegetal extracts was suming that the incriminated herbs have been changed in mid-1990 by introducing powdered distributed in Belgium from May 1990 to Oc­ extracts of Chinese herbs, nominally Stephania tober 1992 (Vanherweghem et al., 1993), that tetrandra and Magnolia officinalis (Vanher­ the yearly amount of distribution was 185 to weghem et al., 1993). The possible relation be­ 200 kg of powders for under the name of

Department of Nephrology, Hopital Erasme, Universite Libre de Bruxelles, Brussels, Belgium.

9 10 VANHERWEGHEM

Stepiiania tetrandra (Vanherweghem et al., 1993) in approximately one-third of the patients (Van and that the mean exposure per patient aver­ Ypersele and Vanherweghem, 1995). aged 900 mg of powder per day for 6 to 12 In most of the cases, renal failure progressed months (Depierreux et al., 1994), about 1500 to despite the fact that the Chinese herbs had been 2000 persons may be assumed to have absorbed withdrawn, plasma creatinine raising from 2.5 such an amount of so-called Stephania tetrandra mg/dL to 8.5 mg/dL for a period of time rang­ powder. About 100 patients with renal disease ing from 6 to 24 months. Interestingly, steroid were recorded, thus representing about 5% of therapy was shown to slow the progression of the exposed population. The first patient the disease (Vanherweghem et al., 1996). recorded by an a posteriori epidemiological sur­ The major pathological lesion consisted of an vey was already on dialysis in April 1991. One extensive interstitial fibrosis with atrophy and patient was recorded in August 1997, despite loss of the tubules, predominantly located in the fact that the incriminated herbs had been the superficial cortex. The glomeruli were rel­ withdrawn from the market in October 1992. atively spaced. They nevertheless showed mild All patients except one, recorded to now were collapse of the capillaries and wrinkling of the women, aged from 18 to 65 years old. Thirty basement membrane. Interlobular and afferent percent of the patients apparently had stable, arterioles showed thickening of their walls due moderate renal failure with a plasma creatinine to a swelling of the endothelial cells (Depier­ below 3.0 mg/dL, and 70% of patients were in reux et al. 1994; Cosijns et al., 1994a). Mild to preterminal or terminal renal failure, treated by moderate atypia and atypical hyperplasia of maintenance dialysis or renal grafting. the urothelium was also observed in pieces of nephrectomies (Cosijns et al., 1994a). Fibrosis could also extend to the pelvis and the ureters Clinical and pathological aspects (Cosijns et al., 1994a; Jadoul et al., 1993). Renal failure was usually not suspected and Two published cases (Cosijns et al., 1994b; was discovered in most cases by routine blood Vanherweghem et al., 1995) were also compli­ testing. Dipstick analysis for proteinuria was cated by microinvasive urothelial carcinomas. negative and urinary sediment was unremark­ able (Vanherweghem et al., 1993; Van Ypersele Relations between the renal disease and and Vanherweghem, 1995). However, micro- Chinese herhs proteinuria was an hallmark of the disease: uri­ nary excretion of five low-molecular weight The time between the introduction of Chi­ proteins (beta2-microglobulin, cystatin C, Clara nese herbs in the weight-loss regimen and the cell protein, retinal binding protein) was in­ outbreak of renal disease circumscribed the creased in all Chinese herbs nephropathy search for the culprit to the Chinese herbs (CHN) patients with renal failure in relation to (Vanherweghem et al., 1993). Further epi­ the degree of renal failure (Kabanda et al. 1995). demiological survey demonstrated that only Urinary excretion of neutral endopeptidase the so-called Stephania tetrandra was associ­ used as a marker of the proximal tubular mass ated with all cases (Depierreux et al., 1994). was markedly decreased, also in relation to the The fact that tetrandrine, the alkaloid derived degree of renal failure (Nortier et al. 1997). from Stephania tetrandra could not be identi­ Taken together, these markers indicate that fied in the capsules taken by the patients sug­ proximal tubular cells were a primary target in gested that this herb might have been replaced CHN. by another toxic vegetal (Vanherweghem et Blood pressure was initially normal in half al., 1993). Inadvertent replacement of Stepha­ of the patients (Van Ypersele and Vanher­ nia tetrandra was suspected because: (1) weghem, 1995). Anemia was present and usu­ Stephania tetrandra belongs the family of Fang- ally more severe than might be anticipated ji besides Aristolochia Fang-ji (Vanherweghem from the degree of renal failure (Van Ypersele et al., 1993). Moreover, the Chinese characters and Vanherweghem, 1995). Curiously, asymp­ and pinyin name of Stephania tetrandra (Fang tomatic aortic insufficiency was demonstrated ji) are identical to that of Aristolochia fangchi. MISUSE OF HERBAL REMEDIES 11

In China, the differentiation is made by pre­ renal tissues of control cases (Schmeiser et al., fixing the name Fang ji either with the epithet 1996). Guang for Aristolochia fangchi or Han for Stephania tetrandra. Obviously, the risk of con­ fusion is considerable in Western markets; and DISCUSSION (2) pathological aspects of CHN is very simi­ lar to that of Balkan endemic nephropathy An extensive interstitial renal fibrosis re­ (BEN) (Vanherweghem et al., 1993; Depier­ sponsible for terminal renal failure, compli­ reux et al., 1994; Cosijns et al., 1994a) the cause cated in some cases by urotheliomas may thus of which is still under controversies, but some be related to the use of Chinese herbs for suggested causes include fungal and plant weight loss. It was shown that the prescribed toxins such as ochratoxin A and aristolochic Chinese herb called Stephania tetrandra was, in acids (Stephanovic and Polenakovic, 1991). fact, inadvertantly replaced by another Chinese Ochratoxin A was not found in the capsules herb, namely Aristolochia fangchi, in the pow­ taken by the patients (Vanherweghem et al., dered extracts delivered in Belgium and 1993). A first attempt to identify aristolochic France. Aristolochia species indeed contain aris­ acids in the capsules taken by the patients tolochic acids that are nephrotoxics (De Smet, failed (Vanherweghem et al., 1993). Later, 1992; Jackson et al., 1964) and carcinogens (De the prepurification steps required by the Smet, 1992; Mengs, 1988). Recently, a rapidly complex formulation of the capsules were progressive interstitial renal fibrosis was ob­ shown to partially destroy aristolochic acids. served in Spain secondary to a chronic intake Further analyses conducted on different of a Aristolochia pistolochia infusion (Pena et al., batches of powders delivered in Belgium un­ 1996). In the cases observed in Belgium, it is der the name of Stephania tetrandra showed possible that the nephrotoxic effects of Aris­ that in fact, most of them did not contain tolochia were increased by the combination of tetrandrine but aristolochic acids (But, 1993; other drugs such as acetazolamide or fenflu­ Van Haelen et al., 1994). The aristolochic acid ramine. The puzzling association of valvular content of the so-called Stephania powder var­ heart disease in about one-third of the cases ied from undetectable to 2.9 mg/g (Van Hae­ may more likely be linked to the concomitant len et al., 1994). Investigations conducted on use of fenfluramine than to the Chinese herbs a larger scale of 46 batches of "Stephania" (Connolly et al. 1997; Vanherweghem, 1997). powders distributed in Belgium show that 30 The mistake observed in Belgium was or will of them contained aristolochic acids and not be responsible for terminal renal failure in at tetrandrine; 7 of them contained tetrandrine least 70 w^omen who simply wanted to lose and not aristolochic acids; 5 of them contained weight. This disaster emphasizes that the med­ both tetrandrine and aristolochic acids; and 4 ical professions has the responsibility to keep of them did not contain either tetrandrine or in mind that the cornerstone of medicine is the aristolochic acid (P. Daenens, Department of critical evaluation of every therapeutic modal­ Toxicology, Katholieke Universiteit Leuven, ity. On one hand, the use of Han Fang Ji as part Belgium, Report of expertise, 1996). Two cases of a weight-loss regimen does not correspond of chronic interstitial nephritis requiring to its traditional use by qualified practitioners maintenance dialysis were described in France of Chinese herbal medicine. On the other hand, and were related to the use of pills containing despite the widespread belief that natural Stephania tetrandra (Pourrat et al, 1994). Later remedies are devoid of side effects, many these pills were shown to contain in fact Aris­ plants contain toxins for humans (Huxtable, tolochia species (B. Stengel, personal commu­ 1990; De Smet, 1992). nication). The origin of herbal poisoning may indeed Finally, aristolochic acid-DNA-adducts, a range in three categories. First, the toxicity of a described biomarker of aristolochic acid expo­ correctly identified plant may be unknown, ig­ sure, were found in all five pieces of nephrec­ nored, or underestimated. For example, veno- tomies of CHN patients, but were absent in the occlusive disease of the liver was related to a 12 VANHERWEGHEM

herbal tea made from Senecio longilobus (Still- herweghem IL. Pathologic aspects of newly described man, 1977); Teucrium chamaedrys (germander) nephropathy related to prolonged use of Chinese herbs. that was prescribed in France for weight loss Am 1 Kidney Dis 1994;24:172-180. Gold CH. Acute renal failure from herbal and patent has led to hepatic necrosis (Larrey et al., 1992); remedies in Blacks. Clin Nephrol 1980;14:128-134. acute renal failure is a frequent and lifethreat- Huxtable Rl. The myth of beneficient nature: The risks of ening complication resulting from African tra­ herbal preparation. Ann Intern Med 1992;117:165-66. ditional herbal remedies (Gold, 1980; Seedat Huxtable Rl. The harmful potential of herbal and other and Nathoo, 1993). Fatal cardiac arrhythmias plant products. Drug Safety 1990;(SuppI 1):126-136. lackson L, Kofman S, Weiss A, Brodowsky H. Aristolochic occurred in Hong Kong after the use of Chi­ acid (NSC-50413): Phase I clinical study. Cancer nese traditional herbal remedies containing Chemother Rep 1964;42:35-37. aconite (Tai et al. 1992). Second, poisoning with ladoul M, De Plaen IF, Cosijns IP, Van Ypersele De Stri­ herbal remedies may result from a misidentifi- hou C. Adverse effects from traditional Chinese medi­ cation of plant species. This possibility is cine. Lancet 1993;341:892-93. Kabanda A, ladoul M, Lauwerijs S, Bernard A, Van Yper­ clearly illustrated by the CHN occurring in Bel­ sele de Strihou C. Low molecular weight proteinuria gium. Third, powdered plant extracts may be in Chinese herbs nephropathy. Kidney Int 1995;48: intentionally adultered by other drugs. As an 1571-1576. example, a case of analgesic nephropathy was Larrey D, Vial T, Pauwels A, Castot A, Biour M, David observed in Malaysia after the long-term use of M, Michel H. Hepatitis after germander (Teucrium traditional Chinese herbal preparations, the chamaedrys) administration: Another instance of herbal medicine hepatotoxicity. Ann Intern Med 1992; analysis of which had revealed the presence of 117:129-32. 120 mg of phenylbutazone per tablet (Sega- Mengs U. Tumor induction in mice following exposure sothy and Samad, 1991). to aristolochic acid. Arch Toxicol 1988;61:504-505. In conclusion, the CHN story draws atten­ Nortier IL, Deschodt-Lanckman MM, Simon S, Thiele- tion to the necessity to submit these sub­ mans NO, De Prez EG, Depierreux M, Tielemans CL, stances to the same quality control as common Richard C, Lauwerys RR, Bernard AM, Vanherweghem IL. Proximal tubular injury in Chinese herbs nephropa­ drugs prior to their release on the market. thy: Monitoring by neutral endopeptidase enzymuria. Moreover, it adds support to previous warn­ Kidney Int 1997;51:288-293. ings about the possible toxicity of certain Pena IM, Borras M, Ramos ], Montoliu ]. Rapidly pro­ herbal remedies (Huxtable, 1992), stressing gressive interstitial renal fibrosis due to a chronic in­ the fact that herbal medicine also requires skill take of a herb (Aristolochia pistolochia) infusion. Nephrol Dial Transplant 1996;11:1359-1360. and experience. Pourrat IL, Montastruc IL, Lecombe IM, Van Ypersele de Strihou C, Cosyns IP. Nephropathie associee a des herbes chinoises: 2 cas. La presse Medicale 1994;23:1669. REFERENCES Schmeiser HH, Bieler CA, Wiessler M, Cisterne IM, Ras- col O, Dumazer P. Detection of DNA adducts formed But PPH. Need for correct identification of herbs inb yherbal aristolochi c acid in renal tissue from patients with poisoning. Lancet 1993;341:637. Chinese herbs nephropathy. Cancer Res 1996;56: Connolly HM, Crary IL, McGoon MD, Hensrud DD, Ed­ 2025-2028 wards BS, Edwards WD, and Schaff HV. Valvular heart Seedat YK, Nathoo BC. Acute renal failure in Blacks and disease associated with fenfluramine-phentermine. N Indians in South Africa: Comparison after 10 years. Engl 1 Med 1997;337:581-588. Nephron 1993;64:198-201. Cosyns IP, ladoul M, Squifflet IP, De Plaen IF, Ferluga D, Segasothy M, Samad S. Illicit herbal preparation contain­ Van Ypersele de Strihou C. Chinese herbs nephropa­ ing phenyl-butazone causing analgesic nephropathy. thy: A clue to Balkan endemic nephropathy? Kidney Nephron 1991;59:166-167 Int 1994;45:1680-1688. Stephanovic V, Polenakovic MH. Balkan nephropathy: Cosyns IP, ladoul M, Squifflet IP, Van Cangh PI, Van Yper­ Kidney disease beyond the Balkans. Am ] Nephrol sele de Strihou C. Urothelial maUgnancy in nephropathy 1991;ll:l-n. due to Chinese herbs. Lancet 1994;344:118. Stillman AS, Huxtable R, Consroe P, Kohnen P, Smith S. De Smet PAG 1992 Aristolochia species. In De Smet PAG, Hepato veno-occlusive disease due to pyrrohzidine Keller K, Hansel R, Chandler RF. Adverse Effects of (Senecio) poisoning in Arizona. Gastroenterology 1977; Herbal Drug. Springer Verlag, Berlin, pp. 79-89. 73:349-52. De Smet PAGM 1992 Drug's used in nonorthodox medi­ Tai YT, But PPH, Young K, Lau CP. Cardiotoxicity after cine. In Dukes MNG (ed) Meyler's side effects of drugs accidental herb-induced aconite poisoning. Lancet Elsevier Science Publishers B.V. pp. 1209-1232. 1992;340:1254-1256. Depierreux M, Van Damme B, Van Den Houte K, Van­ Van Haelen M, Van Haelen-Fastre R, But PPH, Vanher- MISUSE OF HERBAL REMEDIES 13

weghem ML. Identification of aristolochic acids in Chi­ Vanherweghem IL, Tielemans C, Simon ], Depierreux M. nese herbs. Lancet 1994;343:174 Chinese herbs nephropathy and renal pelvic carci­ Van Ypersele de Strihou C, Vanherweghem ML. The noma. Nephrol Dial Transplant 1995;10:270-273. tragic paradigm of Chinese herbs nephropathy. Vanherweghem IL. Association of valvular heart disease Nephrol Dial Transplant 1995;10:157-160. with Chinese herb nephropathy. Lancet 1997;350:1858. Vanherweghem IL, Depierreux M, Tielemans C, Abramowicz D, Dratwa M, ladoul M, Richard C, Van- dervelde D, Verbeelen D, Van Haelen-Fastre R, Van Haelen M. Rapidly progressive interstitial renal fibrosis in young women: Association with slimming Address reprint requests to: regimen including Chinese herbs. Lancet 1993;341: J.L. Vanherweghem, M.D. 387-91. Department of Nephrology Vanherweghem ML, Abramowicz D, Tielemans C, De­ Hopital Erasme pierreux M. Effects of steroids on the progression of re­ Universite Libre de Bruxelles nal failure in chronic interstitial renal fibrosis: A pilot Route de Lennik 808 study in Chinese herbs nephropathy. Am ] Kidney Dis 1996;27:209-215. 1070 Brussels, Belgium