A 36-year-old lady was admitted to hospital on 2 separate occasions with symptoms and signs of cholestasis. Ten weeks prior to each admission the patient had received intravenous methylprednisolone therapy for an acute exacerbation of multiple sclerosis. The liver function tests were significantly deranged, a liver screen was negative, and the liver biopsy showed a mixed picture of cholestatic and hepatocellular injury. On each occasion the symptoms resolved and liver biochemistry returned to normal after 2 months. The temporal relationship to steroid treatment, absence of any other identifiable cause of liver injury, and identical event on re-challenge to the drug, confirm this was an adverse drug reaction to intravenous methylprednisolone.
Multiple sclerosis (MS) is a demyelinating disease of the central nervous system which typically follows a relapsing-remitting course, however, a number of patients will go on to develop secondary progressive disease. An established treatment for acute relapses involves high-dose corticosteroids in the form of intravenous methylprednisolone .
This may reduce the duration of the relapse, but probably has no affect upon consequent deficits acquired or the frequency of future relapses. Although side effects due to high-dose steroids are relatively common and well documented, severe hepatoxicity is extremely rare.
We report a case of severe hepatoxicity due to methylprednisolone.
A 36-year-old lady was referred to our hospital with a 2-month history of anorexia and fatigue, and a 4-day history of progressive jaundice and pruritus associated with the passage of pale stools and dark urine.
Four years earlier the patient was diagnosed with MS. She was on no regular treatment for this and she had received no steroids previously.
Ten weeks before the onset of jaundice she suffered an acute relapse of MS and was treated with a course of pulse-dose intravenous methylprednisolone. This was administered as 1g of Solu-Medrone (Pharmacia Ltd, Milton Keynes, England), dissolved in 250ml of 5% dextrose solution, given intravenously over 1 hour. This was repeated on 3 consecutive days. Her symptoms improved.
However, a few weeks later she started to experience fatigue and 10 weeks later developed cholestatic jaundice. She had been given no other drugs in the months prior to the relapse, during the acute relapse or in the intervening 10 weeks following treatment. There was no past history of jaundice, her alcohol intake was less than 14 units per week and the patient had not traveled abroad recently. There was no history of intravenous drug abuse and she had never received blood transfusions.
On admission the patient was profoundly icteric; there was evidence of ataxia due to MS and the only abnormal clinical finding of relevance was tenderness over the right hypochondrium. There was no hepatosplenomegaly or ascites and the patient was not encephalopathic or pyrexial. There were no signs of chronic liver disease.
The patient's liver function tests were abnormal with an alkaline phosphatase of 232 IU/L, an aspartate transaminase of 2349 IU/L and a bilirubin of 348 Ámol/L. The prothrombin time was normal at 14.2 seconds (control 13.6 seconds). A liver ultrasound scan suggested mild intrahepatic dilatation with a 3mm hyperechoic area in the gall bladder indicating the presence of a gallstone or polyp.
Endoscopic retrograde cholangiopancreatography was performed which was normal - in particular, there was no evidence of stones or biliary dilatation.
A liver screen showed evidence of previous hepatitis A infection (IgG positive, IgM negative), she had been immunized against hepatitis B and as a result the hepatitis B surface antibody was positive (>1000 iu/L) but e antigen and core IgG antibody were negative. The hepatitis C antibody was negative.
The antinuclear, antimitochondrial, smooth muscle, liver-kidney microsomal antibodies were negative. The serum copper, ceruloplasmin, 24-hour urinary copper, serum iron and ferritin were normal and a slit lamp examination for Kayser-Fleischer rings was normal.
During the next 2 weeks the patient's condition and liver function tests improved without specific treatment and she was discharged. The liver function had reverted to normal 2 months after discharge and no cause for this episode, at that time, was identified. Methylprednisolone was not considered as a potential cause at this stage because the jaundice occurred 10 weeks after administration and it is not a recognized cause of such a picture.
The patient experienced no further problems until she was readmitted to our hospital some 18 months later with an identical episode.
She experienced an acute exacerbation of her MS and was given a 3-day course of intravenous methylprednisolone therapy. Ten weeks later she developed cholestatic jaundice and fatigue. No other drugs had been given prior to, during, or following this episode.
On admission she was profoundly jaundiced with visible scratch marks. Her clinical examination was identical to before. Her liver function tests confirmed a raised alkaline phosphatase of 170 IU/L, an aspartate transaminase of 1664 IU/L and a bilirubin of 492 Ámol/L. Her prothrombin time was increased to 15.9 seconds (control 12.7).
A repeat liver screen was again negative so a liver biopsy was performed. There was collapse of the normal reticulin pattern. The hepatocytes showed degeneration, and there was intracanalicular and hepatocyte cholestasis. There were increased numbers of lymphocytes in the portal tracts and the parenchyma. There was no evidence of fibrosis or cirrhosis. The appearances were of intrahepatic cholestasis and parenchymal hepatitis with no specific cause identifiable but changes entirely compatible with a drug related event.
The liver function tests gradually improved and she was discharged after 2 weeks. The liver function tests had again reverted to normal 2 months after discharge.
The temporal relationship to treatment with the onset of liver dysfunction on 2 separate occasions, coupled to the absence of any other identifiable cause, make it most likely that it was methylprednisolone that resulted in hepatocellular damage. There are, as far as we are aware, no published cases of methylprednisolone induced liver injury. Increases in alanine transaminase and alkaline phosphatase have been observed but these are usually mild and not associated with clinical symptoms .
High-dose intravenous methylprednisolone is used to treat many conditions including graft rejection following transplantation, severe erythema multiforme, angioneurotic edema, acute spinal cord injury and relapses of MS. It is a corticosteroid with anti-inflammatory properties 5 times that of hydrocortisone and is metabolized in the liver and kidney. The excipients of the preparation are sodium biphosphate and sodium phosphate, and the active ingredient is methylprednisolone sodium succinate . It is unlikely that the excipients are the cause of hepatoxicity as these are inert compounds present in many other preparations.
The mechanism by which methylprednisolone might produce hepatic injury is unclear. It is most probably an idiosyncratic reaction due to a direct toxic effect of the steroid or its metabolite. In this case the biopsy showed evidence of bile canalicular injury with associated inflammation and there was also some acute hepatocellular injury evident resulting in a mixed cholestatic and hepatocellular hepatitis. An immunogenic reaction is less likely due to the fact there was a delay of some 10 weeks before the onset of jaundice and the exposure to the drug, in addition, there was no associated fever, rash or peripheral eosinophilia.
The liver is a major site of adverse drug reactions (ADRs). A Danish survey reported liver damage to account for 5% of total ADRs, rising to 15% of those that resulted in mortality .
Liver ADRs are the most common reason for withdrawal of drugs from the market. Whilst pre-market surveillance detects common ADRs, those seen in clinical practice are rare (1/10,000 to 1/100,000 prescriptions). Their description depends, in the United Kingdom, on Yellow Card reporting. This is used to report all suspected ADRs with new drugs and serious ADRs with established drugs. Inevitably with this type of system both under and over reporting of events occurs.
In the majority of cases of ADRs an unequivocal diagnosis cannot be made. However, a good history and high index of suspicion can alert one to the likelihood. The onset of liver damage may occur within days of initiating drug treatment, in which case the link may be obvious, or it may occur up to 90 days following continued administration when the link may be missed. There may, as demonstrated in this case, be a delay of weeks following withdrawal of the drug before damage becomes apparent. In this instance the link may be missed on the first presentation and only register on repeated exposure to the drug.
An improvement in symptoms and liver function tests on withdrawal of the drug, coupled to the exclusion of other causes of liver dysfunction, is often sufficient to make the diagnosis. In this case the drug had only been given for 3 days so ongoing administration was not an issue.
The speed of recovery on withdrawal depends on the type of injury. Acute hepatocellular injury may take up to 2 months to recover whilst cholestatic injury may take up to 6 months, and may worsen up to 2 weeks following withdrawal, very occasionally it can be irreversible.
In the case of drugs known to be hepatotoxic the pattern of injury on liver biopsy may be characteristic, for example, cholestatic hepatitis following co-amoxiclav. However, as in this case, where there have been few, if any previous reports of hepatoxicity, the pattern of liver injury on biopsy is less helpful. Re-challenge reactions can support the diagnosis but can be dangerous particularly in acute hepatocellular hypersensitivity type reactions. In this particular case, because the diagnosis was not made on the initial presentation, the patient was inadvertently re-challenged with methylprednisolone which resulted in a second identical episode strongly supporting this being a drug-induced adverse event.
The aim of treatment of drug-induced hepatotoxicity is complete recovery. This involves withdrawal of the drug and subsequent supportive therapy. Awareness should be maintained for the occurrence of hepatotoxicity with the use of different derivatives of a drug. With this patient, further treatment with methylprednisolone would be contraindicated and consequently other options such as beta-interferon therapy will have to be considered.
This article was first published on GastroHep.com on 10 October 2003.
Dr Claire L Bent
Dr Gareth AO Thomas
Department of Gastroenterology, University Hospital of Wales, Heath Park, Cardiff CF14 4XW, UK
Dr Gareth Thomas
Department of Gastroenterology, University Hospital of Wales, Heath Park, Cardiff CF14 4XW, UK
Phone or Fax: +44 (0)2920-743-291
- Milligan NM, Newcombe R, Compston DAS. A double-blind controlled trial of high dose methylprednisolone in patients with multiple sclerosis. I. Clinical effects. J Neurol Neurosurg Psychiatry 1987; 50: 511-16.
- Summary of product characteristics for Solu-Medrone, Pharmacia Ltd. ABPI Compendium of Data Sheets 2003.
- Friis H, Andreasen PB. Drug induced hepatoxicity: an analysis of 1100 cases reported to the Danish Committee on Adverse Drug Reactions between 1978 and 1987. J Intern Med 1992; 232: 133-8.