| |
Autoimmune Hepatitis Triggered by Statins
|
| |
| |
Journal of Clinical Gastroenterology: Volume 40(8) September 2006 pp 757-761
Alla, Vamsee MD* ; Abraham, Joseph MD*; Siddiqui, Junaid MD* ; Raina, Dimple MD* ; Wu, George Y. MD, PhD* ; Chalasani, Naga P. MD ; Bonkovsky, Herbert L. MD*
Departments of *Medicine
Molecular, Microbial and Structural Biology
The Liver-Biliary-Pancreatic Center
University of Connecticut Health Center, Farmington, CT 06030
The Department of Medicine, Indiana University School of Medicine, Indianapolis, IN 46202
Abstract
Although the cause of autoimmune hepatitis (AIH) is unknown, drugs are believed to be potential triggers in some patients. In isolated case reports, statins have been considered such triggers. Here we describe 3 patients in whom it is probable that statins initiated the development of AIH. Two men (aged 47 and 51) and one woman (aged 57) developed AIH after the initiation of statin therapy. They developed positive titers of antinuclear antibodies, antismooth muscle antibodies (1/40 to 1/160), and hypergammaglobulinemia. Features of all 3 patients met the criteria for AIH according to the International Autoimmune Hepatitis Panel. Liver biopsies in all 3 showed varying stages of fibrosis and plasma cell infiltration, compatible with AIH. The woman developed hepatitis due to statins on 2 separate occasions: the first in 1999, due to simvastatin, and the second in 2001 to 2002, due to atorvastatin, which was severe and persisted even after discontinuing medication. Similarly, in the 2 other cases, exposure to statins preceded development of AIH, which persisted despite discontinuing medications. All 3 patients responded well to prednisone and azathioprine or mycophenolate therapy. 3 similar previously reported cases are reviewed. We conclude that the 3 cases reported here and 3 similar previously reported cases, indicate that severe, ongoing AIH on rare occasions can be triggered by statins.
Although the cause of autoimmune hepatitis (AIH) is unknown, drugs and chemicals are believed to be potential triggers in some patients. Statins are very widely prescribed medications. In isolated case reports, statins have been considered such triggers. Here we describe three patients in whom it is probable that statins triggered the development of AIH.
DISCUSSION (see 3 Case Summaries below)
Autoimmune hepatitis is a chronic necroinflammatory disease of unknown etiology, characterized by the presence of circulating autoantibodies, elevated serum globulins and histologic features of interface hepatitis with lympho-plasmacytic infiltration. Autoimmune hepatitis may be triggered by various environmental factors, including viruses (eg, measles, hepatitis C and HSV), and drugs (eg, minocycline, hydralazine, procainamide) in genetically predisposed patients. It is important to diagnose AIH early and differentiate it from other liver diseases because appropriate immunosuppression will usually improve this condition and delay or prevent development of hepatic decompensation or the need for liver transplantation. There may be overlap features with other autoimmune liver diseases like primary biliary cirrhosis and primary sclerosing cholangitis.1
HMG-CoA reductase inhibitors (statins) are among the most widely used drugs.2 Although they are generally considered safe,3 they may rarely produce serious adverse reactions. Asymptomatic elevations in serum aminotransferases are common in patients taking statins, but severe hepatic toxicity is rare. Nevertheless, since the early 1990s there have appeared sporadic reports of statins triggering autoimmune diseases. These diseases include lupus erythematosus, dermatomyositis, lichen planus, and others.4-9 Statin-induced autoimmune hepatitis previously was reported by Graziadei et al10 Pelli et al11 and Wolters and van Buuren.12 Key features of our 3 cases and those reported by others are summarized in the Table 1. Among the 6 reported cases of AIH triggered by statins, 4 were women and 2 were men; all were aged >45 years. All developed a predominantly hepatocellular pattern of injury. HLA typing was available on 4 patients and all 4 were positive for DR 3, 4, or 7, types known to be associated with increased risk of AIH.1
A striking feature in our case series and in the 3 previously described cases is the apparent class effect of statins to unmask or trigger an autoimmune process resulting in clinically significant hepatitis. In B.B., relatively brief exposure to fluvastatin led to marked elevation in serum ALT, which improved spontaneously (Fig. 1A). Later exposure to simvastatin led to a prompt and dramatic recurrence of severe AIH, with persistence of active hepatitis for nearly a year and marked progression of hepatic fibrosis and inflammation on liver biopsy (Fig. 1A). More likely than not, the initial exposure to fluvastatin led to initial immuno-allergic response, and the subsequent brief exposure to simvastatin led to a more rapid and severe, self-sustaining anamnestic response.
Our 3 patients and the 2 previously reported with data sufficient for analysis had definite or likely AIH, according to the International Autoimmune Hepatitis Group scoring system (Table 1).13
Although statins exhibit immunosuppressive actions, including down-regulation of MHC class 2 antigen expression in response to interferon γ,14-16 in rare instances, perhaps, in particularly susceptible patients, statins may still trigger autoimmune responses (Refs. 5 to 14 and this paper). Perhaps, absent the immunosuppressive properties of statins, such responses would be more frequent.
In summary, although statins are generally safe medications, the 3 cases reported here, and 3 similar previously reported cases (Table 1), indicate that severe, persistent AIH can be triggered by statins.
CASE SUMMARIES
Case 1 (B.B.)
A 51-year-old woman accountant with familial hypercholesterolemia, strong family history of coronary artery disease, hypothyroidism, and hyperlipidemia, was first treated in 1999 with fluvastatin (Lescol) for 12 weeks. She developed marked elevations of serum alanine aminotransferase (ALT), and aspartate aminotransferase (AST). Even after discontinuation of fluvastatin, serum aminotransferases remained elevated for about 4 months. Viral hepatitis serologies, antinuclear antibodies (ANA), antimitochondrial antibodies (AMA), antismooth muscle antibodies (ASMA), antiliver kidney-microsomal-1 antibodies (anti-LKM-1) were negative. Serum globulins were mildly elevated (3.6 g/dL, reference range, 2.3 to 3.5 g/dL). A liver biopsy (performed September 29, 1999) showed lobular hepatitis with infiltration of lymphocytes and plasma cells, spotty hepatocytic oncosis and apoptosis, and mild (Ishak stage 1) portal fibrosis.
In September 2002, due to elevated serum cholesterol, and, after a discussion with her primary care physician, the patient decided to start simvastatin (Zocor), 20 mg daily. Within 2 weeks of starting simvastatin, she felt extremely fatigued with generalized itching, and in 6 weeks noticed discomfort in the right upper quadrant of her abdomen. By December 14, 2002, she was extremely fatigued and hypersomnic. Serum liver enzymes were markedly elevated [ALT 1749 U/L (reference range 10 to 39), AST 1459 U/L (reference range 10 to 36)], although serum albumin [3.8 g/dL (reference range 3.5 to 5.5)], and total bilirubin [1.0 mg/dL (reference range 0.3 to 1.2)] were normal. Simvastatin was discontinued. On December thirtieth repeat liver chemistries showed persistent elevations of ALT and AST, and hyperbilirubinemia (serum total bilirubin, 7.3 mg/dL, direct-reacting bilirubin, 6.0 mg/dL) (Fig. 1A). She was referred to the Liver Center at The University of Connecticut Health Center (UCHC) for further evaluation. In January 2003, her serum AST was 1748 U/L (10 to 42 U/L), ALT was 1420 U/L (10 to 40 U/L), total bilirubin was 11.3 mg/dL (0.2 to 1.2 mg/dL), direct bilirubin was 6.5 mg/dL (0 to 0.4 mg/dL), and albumin was 3.3 g/dL (3.5 to 5.5 g/dL) (reference ranges in parentheses).
An abdominal ultrasound showed mild dilatation of the common bile duct (8 mm). An endoscopic retrograde cholangiopancreatogram was normal, showing no intrahepatic or extrahepatic obstruction. It was felt that drug-induced liver injury due to an idiosyncratic reaction to simvastatin was the most likely diagnosis, and the patient was followed closely as an outpatient. Serum AST and total bilirubin remained high, 1760 U/L and 11 mg/dL, respectively in mid-January 2003 (Fig. 1A). Due to persistent elevations of serum aminotransferases and bilirubin, with further rise on July 20, 2003, a second liver biopsy was done on July 24, 2003. The architecture of liver was distorted due to formation of frequent fibrosis septa and regenerative nodules. The fibrosis score (Ishak scoring system) was 6. Marked piecemeal necrosis was noted around all fibrous septae. Severe lobular hepatitis was also noted. Minimal fatty metamorphosis was present. The inflammatory cells were chiefly composed of plasma cells, lymphocytes, and histiocytes/proliferating Kupffer cells. Occasional apoptotic (acidophilic) bodies were present. Bile ductules showed proliferation and reactive changes. There was no lymphoid nodule formation. Central veins were obliterated and incorporated into regenerative nodules. The overall features were consistent with markedly active postnecrotic cirrhosis, due to AIH. She again had negative viral serologies (hepatitis A, B, and C), but now serum ANA was positive at a titer of 1:80, ASMA was 1:40, and serum globulins were increased 1.5 times the upper limit of normal value. On the basis of the history, clinical picture, liver biopsy, and laboratory testing a diagnosis of AIH, triggered by simvastatin, was made and she was treated with prednisone (60 mg/d) and azathioprine (50 mg/d). She improved both symptomatically and biochemically. After 3 months on these drugs, serum aminotransferases and total bilirubin were normal. The dose of prednisone was gradually reduced and then stopped while she remained on azathioprine alone. In March 2005, azathioprine was stopped as she had had normal serum aminotransferases and bilirubin for an extended period of time. However, soon thereafter, her serum aminotransferases rose again, and she was retreated with combination prednisone and azathioprine therapy. Within 1 month, her serum aminotransferases had again fallen into normal range, and prednisone again was tapered. The patient remains on azathioprine 50 mg/d. Her HLA type was found to be positive for HLA-DR4, but negative for HLA-DR3 and DR7 (HLA typing was done by polymerase chain reaction-based methods at ARUP laboratories, Salt Lake City, UT).
Case 2 (J.K.)
A 57-year-old white, man with primary hypercholesterolemia was started on fluvastatin (Lescol) 40 mg/d, in October 2000, because serum total cholesterol was 326 mg/dL and LDL cholesterol was 198 mg/dL. After 14 months of fluvastatin therapy, at the end of 2001, the patient for an unknown reason was switched to atorvastatin (Lipitor), 40 mg/d. Atorvastatin was continued for 5 months, but was discontinued in May 2002, because the serum ALT and AST were noted to climb from 31 U/L and 26 U/L (in December 2001), to 424 U/L and 109 U/L, respectively (Fig. 1B). The patient denied any significant alcohol intake. Despite cessation of atorvastatin, liver enzymes remained elevated prompting a referral to the Liver Center of UCHC in December 2002. At that time serum ALT and AST were still increased at 109 U/L and 50 U/L, respectively. Appropriate studies excluded viral hepatitis, α-1-antitrypsin deficiency, hemochromatosis, and Wilson disease. The ANA and AMA were negative, but ASMA was noted to be positive with a titer of 1:40. Abdominal ultrasound was normal. In January 2003, the patient received a trial of gemfibrozil because of a persistently high serum cholesterol. Gemfibrozil was discontinued in April, 2003, because the patient's serum ALT and AST had risen further to 940 U/L and 355 U/L, and alkaline phosphatase (ALP) had climbed to 143 U/L (reference range 25 to 125 U/L). During the follow-up period of more than a year, although the patient received no lipid lowering drugs, or other drugs, liver enzymes remained at least 3 to 4 times above the upper limit of normal (ULN) with peak ALT, AST, and ALP 1170 U/L, 786 U/L, and 160 U/L, respectively. In April 2004, the serum total bilirubin, which had been normal previously, was elevated at 3.4 mg/dL. The serum globulin level, which had been 2.0 g/dL (reference range 2.3 to 4.5 g/dL) in 2002, showed an increase to 4.1 g/dL around the same time.
A liver biopsy was performed on June 28, 2004. It showed advanced bridging fibrosis (stage 4/6 Ishak scoring system) with portal, interface, and lobular hepatitis. The inflammatory cells were chiefly lymphocytes and plasma cells, with occasional polymorphonuclear leukocytes and eosinophils. There was evidence of lobular unrest, cell swelling, and pseudo-rosette formation (Figs. 2C, D). On the basis of the patient's clinical course, laboratory results, and liver biopsy findings, a diagnosis of autoimmune hepatitis was made; and the patient was started on prednisone 30 mg/d and azathioprine 100 mg/d in July, 2004. The patient responded well to this treatment with the normalization of the liver tests by September 2004. The dose of prednisone was tapered and discontinued in December 2005. Liver tests have remained normal. The plan is to continue azathioprine for a duration of 2 years. HLA typing, done at ARUP Laboratories as in the case of B.B., showed that the patient is positive for both HLA-DR4 and DR7.
Case 3 (B.J.)
A 51-year-old White man with history of hyperlipidemia was noted to have mildly elevated serum aminotransferases in 1997 and an evaluation at that time (including viral serologies and autoimmune markers) was unremarkable. An abdominal ultrasound showed increased echogenicity of the liver and no other abnormal findings. A liver biopsy, performed in 1997, showed moderate steatosis (-50%) with some ballooning of hepatocytes. From 1997 until the spring of 2003, B.J. was treated with gemfibrozil. Serum ALT and AST, although mildly increased, remained <3 times the upper limit of normal. In May 2003, gemfibrozil was stopped and atorvastatin (Lipitor), 20 mg/d was started. In September 2003, serum ALT was found to be 527 U/L with serum total bilirubin 0.8 mg/dL. A repeat liver biopsy in September 2003, showed moderate steatosis with lobular inflammation, focal necrosis, and mild lymphocytic and eosinophilic infiltration. This was felt to be most consistent with drug-induced liver injury; atorvastatin was discontinued on September 20, 2003.
Subsequently, serum liver enzymes remained elevated in November 2003 (ALT 498 U/L), and were higher in February 2004 (ALT 662 U/L). Serum ALT and total bilirubin continued to rise reaching 1555 U/L, and 2.2 mg/dL, respectively, in August, 2004. At this time, B.J. was referred to the Liver Center at Indiana University. In September 2004, at this Center he appeared icteric and serum AST was 1111 U/L, ALT 1550 U/L, total bilirubin 5.5 mg/dL (direct 2.2 mg/dL), and alkaline phosphatase 203 U/L. His medications included paroxetine, pantoprazole, zolpidem, iron, and insulin, all of which he had taken for 2 or more years. Viral hepatitis serologies were negative. Serum IgA and IgM were normal, but IgG was increased (2.7 g/dL). His ANA and ASMA titers were positive at 1:160 dilution. Serum anti-LKM-1 antibody was negative. Serum ceruloplasmin and α-1 antitrypsin levels had been within the normal range in 1997. A liver biopsy, performed in August 2004, showed marked chronic inflammation within the portal tracts including numerous plasma cells associated with extensive piecemeal necrosis. There was moderate steatosis. There was marked inflammatory activity within the lobule, associated with patchy foci of bridging necrosis. The trichrome stain revealed bridging fibrosis (Ishak fibrosis score 3).
A diagnosis of AIH, precipitated by atorvastatin was made, and immunosuppressive therapy, consisting of prednisone 40 mg/d, and mycophenolate mofetil, 500 mg twice daily was started, on October 6, 2004. Within 2 weeks, serum AST fell to 301 U/L, ALT to 762 U/L, and total bilirubin to 2.1 mg/dL (direct 0.9). Serum liver tests became normal in 7 months, and have remained normal since. In June 2005, serum AST was 36 U/L, ALT 20 U/L, and total bilirubin 0.6 mg/dL. A posttreatment liver biopsy, done in February 2005, showed much improved inflammation and hepatocytic oncosis and apoptosis, but septal and focal bridging fibrosis (Ishak fibrosis score=3). The patient has been maintained on mycophenolate mofetil 500 mg twice daily alone. HLA DR typing was attempted, but due to laboratory error was not completed successfully, and the patient has declined further testing.
|
|
| |
| |
|
|
|
