A 33-Year-Old Woman with Jaundice after Azithromycin Use

• CASE REPORT
• DIFFERENTIAL DIAGNOSIS
• PATHOLOGIC FINDINGS
• CLINICAL COURSE
• COMMENTS
• SUMMARY
• DIAGNOSIS
• REFERENCES

CASE REPORT

A 33-year-old woman presented with a history of migraine, and a flu-like syndrome for which she had taken a 5-day course of azithromycin followed by one day of erythromycin. She began to have dark urine and pruritus, and 3 days later, her friend noticed her to be jaundiced. Three additional days later, she presented to her primary care physician with jaundice. Her laboratory results included total and direct bilirubinof 11.3/6.8 mg/dL (normal = 0.1-1.2/0.0-0.8), alanine aminotransferase (ALT) of 583 U/L (normal = 1-53), aspartate aminotransferase (AST) of 208 U/L (normal = 1-50). She had been taking propranolol for 1 year, sumatriptan for 6 months and oral contraceptive steroids on-and-off for 15 years and for about 1 year prior to this presentation. All medications were stopped. Over the next 3 weeks, her pruritus worsened with total bilirubin of 25 mg/dL.

She presented to the Mount Sinai Hospital for further management after trying cholestyramine without any relief. Review of her prior work-ups, which included ultrasound and CT scan, showed no dilatation of bile ducts. There was no history of blood transfusion or intravenous drug use. She had about one drink of alcohol per month. She worked as a computer database person. Her family history was significant for her paternal grandfather with ``liver cancer.''

On examination, she was markedly jaundiced with numerous scratch marks over her entire body, but there were no stigmata of chronic liver disease. There was no hepatosplenomegaly, ascites or pedal edema. Her blood tests at this time showed total and direct bilirubin of 22.8/20.4 mg/dL, ALT of 42 U/L, AST of 45 U/L and alkaline phosphatase of 228 U/L (normal = 20-130). The viral serologies were all negative. These included negative IgM anti-HAV, HBsAg, IgM anti-HBc, and HCV RNA by qualitative PCR. Anti-smooth muscle and anti-LKM antibodies were not detectable. Anti-nuclear antibodies were ``weakly positive,'' and serum protein electrophoresis showed normal level of gamma-globulin. Ceruloplasmin level was normal. A repeat ultrasound showed no bile duct dilatation. A liver biopsy was performed.

DIFFERENTIAL DIAGNOSIS

The patient presented with signs and symptoms of cholestasis with pruritus and jaundice. There was no history of hepatitis virus exposure and chronic alcoholism. Her laboratory work-up showed high serum transaminases without any evidence of hepatitis virus infection, autoantibodies or Wilson's disease. Ultrasound and CT scan showed no dilatation of bile ducts, which ruled out extrahepatic cholestasis or main duct ``obstructive jaundice.'' These findings were supported by serum alkaline phosphatase, which is usually higher than three times the normal value in biliary obstruction.

The cause of her cholestatic hepatitis appeared to be azithromycin, which she stopped taking three days prior to the development of jaundice and pruritus. There were only a handful of cases of azithromycin-related cholestasis reported in the literature,[1] [2] unlike the well-described erythromycin-related cholestasis.[3] [4] [5] Oral contraceptive-related cholestasis was unlikely since the patient had been taking oral contraceptive for several years without any side effects. Symptoms of oral contraceptive cholestasis usually occurred during the first few cycles of ingestion.

PATHOLOGIC FINDINGS

Liver needle biopsy of the patient showed severe cholestasis with bile casts in the canaliculi (Fig. [1]). There was mild proliferation of bile ductules, which was accompanied by polymorphonuclear cell infiltrate and occasional bile duct damage (Fig. [2]). Eosinophils were observed in portal tracts and in the lobules. There was mild chronic inflammation in the lobules and in rare portal tracts.

CLINICAL COURSE

She was treated with cholestyramine and six courses of plasmapheresis, which resulted in significant improvement of the pruritus. She maintained normal coagulation without any sign of encephalopathy as her bilirubin levels gradually came down. Two months after her first intake of azithromycin, her total bilirubin was 1.4 mg/dL with ALT of 60mg/dL. At her 4-year follow-up visit, she continued to maintain entirely normal biochemical liver tests.

COMMENTS

Cholestasis in general is classified as either extrahepatic, where there is a mechanical obstruction to the main bile ducts, or intrahepatic, where such an obstructive lesion cannot be demonstrated and the defect is in the hepatocytes or in microscopic bile ducts within the liver. The distinction may be possible by an accurate history, physical examination, and basic biochemical tests. Visualization of the bile ducts is indicated if the patient is cholestatic. The first procedure in distinguishing intrahepatic cholestasis from extrahepatic, main duct ``obstructive'' cholestasis is ultrasound or CT scanning to show whether or not the intrahepatic bile ducts are dilated.

Drugs in common use can cause toxic effects on the liver, which can mimic almost every naturally occurring liver disease affecting man. In every patient with liver disease, it is essential to record all medications taken within the last three months. Early suspicion of a drug-related hepatic toxicity and, if possible, diagnosis of the cause, are essential. Hepatocellular necrosis, hepatitis, and cholestasis in particular should arouse a greater degree of suspicion, especially if no other cause has been found.

Drug-related liver damage is typically idiosyncratic. Many different mechanisms for idiosyncratic hepatotoxicity have now been elucidated. They include individual genetic variation in the metabolism of drugs, and the development of immune reactions to a drug or its metabolites. With the exception of a few drugs shown to cause liver damage in patients using a particular metabolic pathway, idiosyncratic drug injury is unpredictable in the sense that the susceptibility of individual patients cannot be tested before the drug is given. Timely recognition and drug withdrawal may avoid unnecessary intervention and prevent perpetuation of the hepatic injury. However, often the onset of hepatic injury by drugs may be delayed from days to several months.

In our patient, azithromycin was likely the cause of her cholestatic disease. However, it is possible that the erythromycin that the patient took for one day contributed to the injury. Cholestasis with hepatitis is a feature of macrolide hepatotoxicity.[6] This is best documented with erythromycin estolate.[3] [4] Erythromycin is known to induce its own biotransformation by enhancing microsomal enzymes in the liver and particularly some isoenzymes of the cytochrome P-450 system with high affinity to erythromycin.[7] Onset is 2 to 25 days after exposure. Anorexia, nausea, and vomiting are common. Severe right upper quadrant pain may occur and can closely mimic acute cholecystitis.[8] Recovery occurs within 1 to 3 weeks. Cholestatic hepatitis in newer macrolide antibiotics, such as azithromycin, has also been reported in rare cases.[1] [2] Azithromycin has a long half-life and has been demonstrated to reach high liver concentrations, exceeding the serum levels by 25 to 200 folds.[9] [10] The early liver injury could be hypersensitivity-mediated with subsequent metabolite-dependent lesions of sinusoidal and canalicular plasma membrane leading to a prolonged ductular cholestasis.[2] The high liver concentration and persistence of the drug and the relative late increase of hepatic serum enzymes may support this hypothesis. This raises the possibility of intrinsic effect of azithromycin, in addition to the more common idiosyncratic hepatotoxicity found in other older macrolides. Most intrinsic hepatotoxins produce liver damage within a few hours or days, whereas in the idiosyncratic type of injury there is often a latent period of many days or weeks before liver disease becomes apparent.

Oral contraceptive related hepatic injury was unlikely in this patient. It should be included in the differential diagnosis if jaundice would have appeared in the first few cycles of administration.[11] [12] [13] Estrogen is the important agent, although progestin may augment the effect. This reaction is rare and its incidence is decreasing as the hormone content in the contraceptive preparation is reduced.

Liver biopsy is not performed in every single instance of drug-induced cholestasis. However, histological assessment may help to exclude other causes of liver disease and may show features that suggest a drug origin. Biopsy also may be useful in predicting the prognosis, especially indicating the possible evolution of chronic cholestasis. Biochemical evidence of improvement after drug withdrawal is occasionally supported by a return to normal histology.

Liver biopsies show inflammatory infiltrates in the acini and evidence of hepatocellular damage. This complex was recognized as ``hepatocanalicular.''[14] The portal infiltrate often includes eosinophils and these are occasionally abundant, but their absence does not exclude drug-induced cholestasis. Small interlobular ducts shows minimal abnormalities, such as irregular distribution of epithelial cell nuclei, cytoplasmic vacuolization, variation in nuclear size, and infiltration by lymphocytes to a more profound abnormalities such as bile duct damage and infiltration of neutrophils. In most patients removal of the offending drug leads to rapid improvement. Occasionally, the cholestasis is slow to improve and the liver biopsy specimen shows cholestasis only, with no fibrosis or other evidence of progressive disease. In rare instances, true chronic disease develops on the basis of severe bile duct damage and duct loss with consequent fibrosis and other features of chronic biliary disease.[15] This is called ``vanishing bile duct syndrome'' and has been reported after a number of drugs, among which is erythromycin ethylsuccinate.[16] [17]

The principles of management involve prompt withdrawal of the offending drug and symptomatic therapy to alleviate pruritus. Therapeutic strategies to alleviate pruritus are similar to those for other causes of cholestasis.[18] [19] They include cholestyramine and ursodeoxycholic acid, with rifampicin and opiate antagonists as second line therapies. Other methods that have been employed were plasmapheresis-as was in our patient-phototherapy, and cascade resin perfusion.[8]

SUMMARY

A 33-year-old woman presented with jaundice and pruritus. She had taken a 5-day course of azithromycin followed by one day of erythromycin for a flu-like syndrome. She was previously healthy with no history of hepatitis virus exposure or other liver-related diseases. Total bilirubin, serum transaminases, and alkaline phosphatase were highly elevated. She was diagnosed with azithromycin-induced cholestasis. Other possible causes of cholestasis were ruled out by laboratory tests and imaging studies. Liver biopsy showed changes consistent with drug-induced cholestasis. She was treated with cholestyramine and underwent six courses of plasmapheresis. Two months later, her total bilirubin and serum transaminases were back to normal. The patient maintained entirely normal biochemical liver tests at her 4-year follow-up visit.

DIAGNOSIS

Azithromycin-induced cholestasis

REFERENCES

• 1 Longo G, Valenti C, Gandini G. Azithromycin-induced intrahepatic cholestasis.  Am J Med . 1997;  102 217-218
  PubMedSearch in Google Scholar
• 2 Macaigne G, Mokbel M, Marty O. Acute pseudoangiocholitic hepatitis probably induced by azithromycin.  Gastroenterol Clin Biol . 2000;  24 969-970
  PubMedSearch in Google Scholar
• 3 Diehl A M, Latham P, Boinott J K. Cholestatic hepatitis from erythromycin ethylsuccinate.  Am J Med . 1984;  76 931-934
  CrossrefPubMedSearch in Google Scholar
• 4 Keeffe E B, Reis T C, Berland J E. Hepatotoxicity to both erythromycin estolate and erythromycin ethylsuccinate.  Dig Dis Sci . 1982;  27 701-704
  CrossrefPubMedSearch in Google Scholar
• 5 Zafrani E S, Ishak K G, Rudzki C. Cholestatic and hepatocellular injury associated with erythromycin esters. Report of nine cases.  Dig Dis Sci . 1979;  24 385-396
  CrossrefPubMedSearch in Google Scholar
• 6 Periti P, Mazze T, Mini E. Adverse effects of macrolide antibacterials.  Drug Saf . 1993;  9 346-364
  PubMedSearch in Google Scholar
• 7 Danan G, Descatoire V, Pessayre D. Self-induction by erythromycin of its own transformation into a metabolite forming an inactive complex with reduced cytochrome P-450.  J Pharmacol Exp Ther . 1981;  218 509-514
  PubMedSearch in Google Scholar
• 8 Chitturi S, Farrell G. Drug-induced cholestasis.  Sem Gastrointest Dis . 2001;  12 113-124
  PubMedSearch in Google Scholar
• 9 Lalak N J, Morris D L. Azithromycin clinical pharmacokinetics.  Clin Pharmacokinet . 1993;  25 370-374
  CrossrefPubMedSearch in Google Scholar
• 10 Westphal J F, Vetter D, Brogard J M. Hepatic side-effects of antibiotics.  J Antimicrob Chemother . 1994;  33 387-401
  PubMedSearch in Google Scholar
• 11 Orellana-Alcalde J M, Dominguez J P. Jaundice and oral contraceptive drugs.  Lancet . 1966;  2 1279-1280
  PubMedSearch in Google Scholar
• 12 Metreau J M, Chumeaux D, Berthelot P. Oral contraceptives and the liver.  Digestion . 1972;  7 318-335
  PubMedSearch in Google Scholar
• 13 Lindberg M C. Hepatobiliary complications of oral contraceptives.  J Gen Intern Med . 1992;  7 199-209
  CrossrefPubMedSearch in Google Scholar
• 14 Zimmerman H J, Ishak K G. Hepatic injury due to drugs and toxins. In: MacSween RNM, Anthony PP, Scheuer PJ, et al., eds. Pathology of the Liver, 4th ed Edinburgh: Churchill Livingstone 2002: 621-710
  Search in Google Scholar
• 15 Davies M H, Harrison R F, Elias E. Antibiotic-associated with severe, prolonged, intrahepatic cholestasis.  J Hepatol . 1994;  20 112-116
  CrossrefPubMedSearch in Google Scholar
• 16 Desmet V J. Vanishing bile duct syndrome in drug-induced liver disease.  J Hepatol . 1997;  26(Suppl 1) 31-35
  CrossrefPubMedSearch in Google Scholar
• 17 Geubel A P, Nakad A, Rahier J. Prolonged cholestasis and disappearance of interlobular bile ducts following chlorpropamide and erythromycin ethylsuccinate. Case of drug interaction?.  Liver . 1988;  8 350-353
  PubMedSearch in Google Scholar
• 18 Jones E A, Bergasa N V. The pruritus of cholestasis.  Hepatology . 1999;  29 1003-1006
  CrossrefPubMedSearch in Google Scholar
• 19 Granco J. Pruritus.  Curr Treat Options Gastroenterol . 1999;  2 451-456
  PubMedSearch in Google Scholar

REFERENCES

• 1 Longo G, Valenti C, Gandini G. Azithromycin-induced intrahepatic cholestasis.  Am J Med . 1997;  102 217-218
  PubMedSearch in Google Scholar
• 2 Macaigne G, Mokbel M, Marty O. Acute pseudoangiocholitic hepatitis probably induced by azithromycin.  Gastroenterol Clin Biol . 2000;  24 969-970
  PubMedSearch in Google Scholar
• 3 Diehl A M, Latham P, Boinott J K. Cholestatic hepatitis from erythromycin ethylsuccinate.  Am J Med . 1984;  76 931-934
  CrossrefPubMedSearch in Google Scholar
• 4 Keeffe E B, Reis T C, Berland J E. Hepatotoxicity to both erythromycin estolate and erythromycin ethylsuccinate.  Dig Dis Sci . 1982;  27 701-704
  CrossrefPubMedSearch in Google Scholar
• 5 Zafrani E S, Ishak K G, Rudzki C. Cholestatic and hepatocellular injury associated with erythromycin esters. Report of nine cases.  Dig Dis Sci . 1979;  24 385-396
  CrossrefPubMedSearch in Google Scholar
• 6 Periti P, Mazze T, Mini E. Adverse effects of macrolide antibacterials.  Drug Saf . 1993;  9 346-364
  PubMedSearch in Google Scholar
• 7 Danan G, Descatoire V, Pessayre D. Self-induction by erythromycin of its own transformation into a metabolite forming an inactive complex with reduced cytochrome P-450.  J Pharmacol Exp Ther . 1981;  218 509-514
  PubMedSearch in Google Scholar
• 8 Chitturi S, Farrell G. Drug-induced cholestasis.  Sem Gastrointest Dis . 2001;  12 113-124
  PubMedSearch in Google Scholar
• 9 Lalak N J, Morris D L. Azithromycin clinical pharmacokinetics.  Clin Pharmacokinet . 1993;  25 370-374
  CrossrefPubMedSearch in Google Scholar
• 10 Westphal J F, Vetter D, Brogard J M. Hepatic side-effects of antibiotics.  J Antimicrob Chemother . 1994;  33 387-401
  PubMedSearch in Google Scholar
• 11 Orellana-Alcalde J M, Dominguez J P. Jaundice and oral contraceptive drugs.  Lancet . 1966;  2 1279-1280
  PubMedSearch in Google Scholar
• 12 Metreau J M, Chumeaux D, Berthelot P. Oral contraceptives and the liver.  Digestion . 1972;  7 318-335
  PubMedSearch in Google Scholar
• 13 Lindberg M C. Hepatobiliary complications of oral contraceptives.  J Gen Intern Med . 1992;  7 199-209
  CrossrefPubMedSearch in Google Scholar
• 14 Zimmerman H J, Ishak K G. Hepatic injury due to drugs and toxins. In: MacSween RNM, Anthony PP, Scheuer PJ, et al., eds. Pathology of the Liver, 4th ed Edinburgh: Churchill Livingstone 2002: 621-710
  Search in Google Scholar
• 15 Davies M H, Harrison R F, Elias E. Antibiotic-associated with severe, prolonged, intrahepatic cholestasis.  J Hepatol . 1994;  20 112-116
  CrossrefPubMedSearch in Google Scholar
• 16 Desmet V J. Vanishing bile duct syndrome in drug-induced liver disease.  J Hepatol . 1997;  26(Suppl 1) 31-35
  CrossrefPubMedSearch in Google Scholar
• 17 Geubel A P, Nakad A, Rahier J. Prolonged cholestasis and disappearance of interlobular bile ducts following chlorpropamide and erythromycin ethylsuccinate. Case of drug interaction?.  Liver . 1988;  8 350-353
  PubMedSearch in Google Scholar
• 18 Jones E A, Bergasa N V. The pruritus of cholestasis.  Hepatology . 1999;  29 1003-1006
  CrossrefPubMedSearch in Google Scholar
• 19 Granco J. Pruritus.  Curr Treat Options Gastroenterol . 1999;  2 451-456
  PubMedSearch in Google Scholar