Acute Hepatitis Associated with the Chinese Herbal Product Jin Bu Huan

Methods

Patients

All seven patients were white and had no history of hepatic disease, obesity, diabetes mellitus, or atopy. Six of seven patients were women. All denied a history of excessive alcohol or hepatotoxic drug intake. Risk factors for viral hepatitis were absent in all patients. Five patients resided in Los Angeles, California; three patients had purchased Jin Bu Huan Anodyne Tablets (Kwangsi Pai Se Pharmaceutical/Bose Drug Manufactory, Kwangsi, China) at the same drug store. Two other patients resided in Hawaii and Toronto, Canada, respectively. All patients developed symptoms between March 1993 and March 1994 with the exception of patient 7, whose symptoms began in 1991. Ultrasound examinations of the liver and biliary tract were normal in each patient. Serologic test results in all seven patients were negative for antinuclear, anti-smooth muscle, and antimitochondrial antibodies. Prothrombin times were normal throughout the course of illness of each patient.

Case Reports

Patient 1

A 66-year-old woman presented to her physician with symptoms of fever, nausea, and fatigue for 5 weeks. She was anicteric with a palpable, nontender liver. Stigmata indicating chronic liver disease were absent. She had taken 2 tablets of Jin Bu Huan at night, 2 to 3 times a week for the previous 3 months, for back pain and insomnia. Her medical history included osteoarthritis. She had used nonsteroidal anti-inflammatory drugs during the previous 2 years without adverse effects. Results of serologic tests showed convalescent antibodies for viral hepatitis A and B. A second-generation enzyme-linked immunosorbent assay for anti-hepatitis C virus was negative.

Two weeks after she stopped taking Jin Bu Huan, maximal levels of serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were 9.4 µkat/L and 5.2 µkat/L, respectively, and were near normal 3 weeks later Table 1 and Figure 1 A. At this time, the patient resumed use of Jin Bu Huan for insomnia (2 tablets each night for 7 days). Two weeks after she resumed taking Jin Bu Huan, symptoms returned and enzyme levels were again increased (ALT, 16.0 µkat/L; AST, 9.9 µkat/L). Liver test results returned to normal 3 weeks later. A second-generation enzyme-linked immunosorbent assay for anti-hepatitis C virus was repeated, and the result was again negative. Twelve weeks later, she was asymptomatic and liver test results were normal.

Patient 2

A 24-year-old woman presented to her physician with symptoms of fever, nausea, vomiting, fatigue, and pruritus for 3 weeks. She had deep jaundice, excoriations of her extremities, and tender hepatomegaly. Stigmata indicating chronic liver disease were absent. She was hospitalized for 5 days. She had taken 4 tablets of Jin Bu Huan at night for insomnia, 4 times a week for the previous 2 months. She stopped taking Jin Bu Huan 1 week after the onset of symptoms. She had used an oral contraceptive pill daily for the previous 8 years. No other medical illness was present.

In-hospital results of liver tests showed the following maximal levels: ALT, 24.5 µkat/L; AST, 14.9 µkat/L; alkaline phosphatase, 2.2 µkat/L; and total bilirubin, 479 µmol/L. Results from ultrasound examination of the liver and biliary tract were normal. Results of serologic tests for viral hepatitis A, B, and C; cytomegalovirus; and Epstein-Barr virus were negative. The level of serum ceruloplasmin was normal. The peripheral leukocyte count increased to 10.8 × 10⁹/L with 7% eosinophils (normal, <3% eosinophils). The liver biopsy specimen showed acute hepatitis with focal necrosis, cholestasis, and inflammation with numerous eosinophils in the portal tracts, results consistent with a drug reaction (Figure 2). Pruritus improved with cholestyramine treatment. Nine weeks after discontinuing Jin Bu Huan use, she was asymptomatic and anicteric, showed resolution of eosinophilia, and had normal liver test results (Table 1).

Patient 3

A 45-year-old woman, a friend of patient 2, presented to her physician with symptoms of nausea, anorexia, fatigue, pruritus, and right upper quadrant abdominal pain. Physical examination showed tender hepatomegaly. Stigmata indicating chronic liver disease were absent. She had taken 4 tablets of Jin Bu Huan at night for insomnia, 3 to 4 times a week for the previous 12 weeks. During the previous 6 months, she had intermittently used another Chinese herbal product, Ma Huang (active ingredients are ephedrine and pseudoephedrine), without adverse effects. She had used no other medications and had no medical illness. She stopped taking both herbal products because of her illness.

Two weeks later, she noted jaundice. Results from liver tests showed the following maximal levels: ALT, 21.8 µkat/L; AST, 16.7 µkat/L; alkaline phosphatase, 3.8 µkat/L; and total bilirubin, 58 µmol/L. Results of serologic tests for viral hepatitis A, B, and C and Epstein-Barr virus were negative. During the next 4 weeks, symptoms resolved and liver tests showed decreased levels of enzymes (Table 1). However, 12 weeks after she stopped taking the herbal product, she developed an unidentified illness and increased aminotransferase levels. She denied using Jin Bu Huan, alcohol, or other hepatotoxic drugs. Tests for viral hepatitis A, B, and C and Epstein-Barr virus were repeated, and the results were negative for these viruses. Results from liver tests were maximal 7 weeks later (ALT, 10.2 µkat/L; AST, 6.5 µkat/L) and were normal by 19 weeks (a total of 30 weeks after cessation of Jin Bu Huan).

Patient 4

A 48-year-old woman residing in Hawaii presented to her physician with fatigue and a temperature of 40.6 °C. The physical examination was normal, and stigmata indicating chronic liver disease were absent. She had taken 3 tablets of Jin Bu Huan nightly for 7 weeks for insomnia. She had used no other medications and had no other medical illness. Liver test results were abnormal (ALT, 3.2 µkat/L; AST, 1.2 µkat/L; and alkaline phosphatase, 9.2 µkat/L). Results of serologic tests for viral hepatitis A, B, and C and Epstein-Barr virus were negative, and the serum ceruloplasmin level was normal. The peripheral leukocyte count was 9.3 × 10⁹/L with 6% eosinophils. After cessation of Jin Bu Huan, she became asymptomatic, her eosinophilia resolved, and liver test results returned to normal within 4 weeks (Table 1).

Patient 5

A 46-year-old man presented to his physician with a temperature of 40.6 °C, headaches, fatigue, and tender hepatomegaly. Stigmata indicating chronic liver disease were absent. He had taken 3 tablets of Jin Bu Huan for insomnia, 3 times a week intermittently for 6 months. He had used no other medications and had no other medical illness. Aminotransferase levels were abnormal 2 weeks after stopping Jin Bu Huan (ALT, 6.7 µkat/L; AST, 4.7 µkat/L). Results of serologic tests for viral hepatitis A, B, and C were negative, and the serum ceruloplasmin level was normal. Five weeks after he stopped taking Jin Bu Huan, symptoms improved and liver test results were normal Table 1 and Figure 1 B. After reuse of Jin Bu Huan for 4 weeks, the ALT level increased to 1.7 µkat/L. A second-generation enzyme-linked immunosorbent assay for anti-hepatitis C virus was negative. After Jin Bu Huan use was discontinued, liver test results returned to normal in 4 weeks.

Patient 6

A 31-year-old woman presented to her physician with nausea, vomiting, malaise, and deep jaundice. The physical examination was normal, and stigmata indicating chronic liver disease were absent. She had taken 6 tablets of Jin Bu Huan for insomnia, 4 to 6 times a week intermittently for 10 months and then nightly for 8 weeks. She had used no other medications and had no other medical illness. Liver test results were abnormal (ALT, 35.8 µkat/L; AST, 17.4 µkat/L; alkaline phosphatase, 2.8 µkat/L; and total bilirubin, 262 µmol/L). Results of serologic tests for viral hepatitis A, B, and C were negative, and the serum ceruloplasmin level was normal. Two weeks after she stopped taking Jin Bu Huan, liver test results showed that the enzyme levels were decreasing (Table 1). She is presently asymptomatic.

Patient 7

A 53-year old woman residing in Toronto, Canada, presented to her physician with symptoms of fatigue and headaches. Physical examination showed hepatomegaly, but stigmata indicating chronic liver disease were absent. For the previous 3 years, she had taken 1 to 3 tablets of Jin Bu Huan nightly for arthralgias and insomnia. She had previously used other Chinese herbal products and vitamin A (5000 IU/d) without adverse effects. She had used no other medications and had no other medical illness. Liver test results had been normal 6 months before Jin Bu Huan use. The dose of Jin Bu Huan for the first 12 months was 3 tablets nightly Figure 1 C. At this time, aminotransferase levels increased (ALT, from 2.3 µkat/L to 9.3 µkat/L; AST, from 4.8 µkat/L to 4.9 µkat/L). Results of serologic tests for Epstein-Barr virus and viral hepatitis A, B, and C were negative.

A liver biopsy specimen showed mild-to-moderate fibrous expansion of the portal tracts, mild portal and parenchymal lymphocytic inflammation, and scattered pigmented macrophages. A few neutrophils were present, but eosinophils were absent. Focal hepatocellular necrosis was noted with mild microvesicular steatosis (Figure 3). During the next 18 months of Jin Bu Huan ingestion, she decreased the dose to 1 tablet nightly. Aminotransferase levels decreased but did not return to normal (ALT, 1.6 µkat/L; AST, 0.9 µkat/L). Two weeks after she stopped taking Jin Bu Huan, aminotransferase levels were normal. She continued taking other Chinese herbal products and vitamin A. Aminotransferase levels remained normal when tested 8 weeks later (Table 1).

Jin Bu Huan Dosage and Composition

The package insert for Jin Bu Huan recommended a dosage of 2 to 4 tablets 1 to 3 times a day for pain relief or 1 to 3 tablets at night for insomnia. Thus, all patients were taking appropriate doses. The insert also indicated that tablets contained 70% starch and 30% levo-alkaloid from the plant Polygala chinensis. Tablets from patients 1, 2, and 7 and additional samples from the drug store visited by patients 1, 2, and 3 were analyzed at Colorado State University. Tablets were crushed and extracted with ethyl acetate and yielded a pure residue of levo-tetrahydropalmatine (36% by weight) detected using gas chromatography-mass spectroscopy and nuclear magnetic resonance. This alkaloid is present in the plant genera Stephania and Corydalis but not in the genus Polygala (the plant indicated on the package insert). The remaining constituents were inert (consistent with starch); no other plant materials or chemicals were identified. These ingredients were identical to those extracted from Jin Bu Huan tablets retrieved in previously reported cases of overdose in children [5]. Thus, it is unlikely that the hepatic injury in the three adult patients whose tablets were analyzed was caused by contaminants.

Discussion

These seven patients represent the first reported cases of hepatotoxicity associated with the Chinese herbal product Jin Bu Huan. Serologic testing for autoantibodies; cytomegalovirus; Epstein-Barr virus; and hepatitis viruses A, B, and C was uniformly negative. Other hepatotoxic and metabolic causes for the symptoms were also excluded. Abdominal ultrasound examinations were normal. All patients had variable signs and symptoms including fatigue (100%), hepatomegaly (71%), fever (57%), nausea (57%), abdominal pain (43%), jaundice (43%), vomiting (28%), peripheral eosinophilia (28%), and pruritus (28%). All patients had increased levels (mean, range) of ALT (14.8 µkat/L, 3.2 to 35.8 µkat/L) and AST (9.3 µkat/L, 1.2 to 17.4 µkat/L); five patients had increased levels of alkaline phosphatase (4.1 µkat/L, 2.2 to 9.2 µkat/L); and three had increased total bilirubin levels (266 µmol/L, 58 to 479 µmol/L).

Liver disease developed after ingestion of Jin Bu Huan for a mean of 20 weeks (range, 7 to 52 weeks). The duration of use before manifestations of clinical hepatitis was the same for Jin Bu Huan as for another hepatotoxic herb, germander [7]. The symptoms of six patients resolved within a mean of 8 weeks (range, 2 to 30 weeks), and all patients have completely recovered without developing evidence of chronic liver disease. One patient is asymptomatic, and liver test results show decreasing levels of enzymes.

Results from liver biopsy specimens in patients 2 and 7 were consistent with a drug-induced hepatic injury. In patient 2, the liver biopsy specimen showed focal hepatocellular necrosis with numerous eosinophils in the portal tracts. In contrast, the liver biopsy specimen in patient 7 showed mild nonspecific inflammation with focal hepatocellular necrosis and steatosis. Although portal fibrosis was generally mild, bridging fibrosis, indicating possible chronic liver disease, was present in other areas. Because this patient had ingested Jin Bu Huan for 2.5 years, it is plausible that this herbal product may have caused chronic liver disease.

Patients 1 and 5 reused Jin Bu Huan and promptly became ill again with increased aminotransferase levels. Jin Bu Huan was the only identifiable cause for this second increase in aminotransferase levels, further strengthening the association between Jin Bu Huan ingestion and hepatotoxicity. A relatively short duration of reuse with other herbal products, such as germander [7] and chaparral [8], has also resulted in an abrupt recrudescence of hepatitis. Patient 3 was unusual because decreasing levels of ALT and AST for 6 weeks were followed by increases associated with an undiagnosed illness. Repeat testing for viral infections was negative, and she denied reusing Jin Bu Huan during this period.

The constituents of many herbal formulations consist of ground-up plant products and, thus, contain numerous active ingredients [9]. In contrast, Jin Bu Huan tablets contained a single active ingredient, levo-tetrahydropalmatine, which is present in the plant genera Stephania and Corydalis. The hepatotoxic mechanism of levo-tetrahydropalmatine is unknown. Any mechanism proposed for its toxicity must explain why only a few users develop hepatitis. First, a hypersensitivity reaction may explain the abrupt recrudescence of hepatitis in patients 1 and 5 when Jin Bu Huan was reused. The eosinophilia seen in two patients indicates a possible allergic component. A similar immunoallergic mechanism was proposed to explain the rapid recurrence of hepatitis with use of germander [7]. Second, a direct hepatotoxic effect is suggested by the observation that aminotransferase levels decreased by 83% in patient 7, when the Jin Bu Huan dosage was reduced. A direct hepatotoxic effect may be anticipated on the basis of the structural similarity of levo-tetrahydropalmatine, the active ingredient in Jin Bu Huan and the hepatotoxic pyrrolizidine alkaloids [10, 11]. The direct hepatotoxic effect of pyrrolizidine alkaloids results from binding of active metabolites to hepatocytes [11]. Comfrey, another herbal remedy, also contains pyrrolizidine alkaloids and causes sinusoidal occlusion [12] and veno-occlusive disease of the liver [10, 11]. Recently, furano neo-clerodane, an active metabolite of germander, was identified as the cause of its hepatotoxicity [13]. Finally, an idiosyncratic reaction to Jin Bu Huan also could be responsible for the hepatitis. As many as 60% of patients with an idiosyncratic drug-induced hepatic injury recover from the acute episode within 2 months, and remaining patients recover within 1 year [14].

Because Chinese herbal products are marketed as dietary supplements rather than drugs, they are not subjected to rigorous testing for safety and efficacy. Misidentification of the plant source and percentage of the active ingredient on the Jin Bu Huan package insert suggests improper manufacturing or inaccurate labeling. Further, the term “natural” on a label does not ensure product safety [15]. Hepatotoxicity has been reported with many “natural” herbal products, including germander [7, 16], chaparral [8, 17, 18], senna [19], mistletoe [20], skullcap [21], comfrey [22], Crotolaria [10, 23-25], and herbal teas [10].

In 1991, the State of California, Department of Health Services, issued a bulletin indicating that the sale of certain Asian herbal products was illegal [26]. These formulations are known to be toxic when taken internally. Of interest, berberine is listed in this catalog of toxic herbal products and its toxic ingredient is an alkaloid derivative of levo-tetrahydropalmatine. Surveillance systems or guidelines have not been instituted to ensure quality control of herbal products. Recently, the Food and Drug Administration [27] implemented the MEDWATCH program so that adverse reactions to dietary supplements can be reported (telephone [800] 332-1088). Despite reports of adverse reactions to these products, package inserts seldom warn about possible side effects [1, 2]. Inserts rarely mention compounds present in the product or that contamination with pharmaceutical agents [28] or toxins [29] has occurred. Subjecting herbal formulations to Food and Drug Administration rules and regulations may result in safer and more effective products.