American Journal of Transplantation 2005; 5: 2796–2798Blackwell Munksgaard

Copyright C© Blackwell Munksgaard 2005

doi: 10.1111/j.1600-6143.2005.01065.x

Case Report

Is Liver Transplantation Advisable for IsoniazidFulminant Hepatitis in Active ExtrapulmonaryTuberculosis?

Rafael Barcenaa,∗, Elena Otona, Maria Angeles

Morenob, Jesus Fortunc, Miguel

Garcia-Gonzaleza, Ana Morenoc

and Emilio de Vicented

aHepato-Gastroenterology Service, bInternal MedicineService, cInfectious diseases Service and dGeneralSurgery Service, Ramon y Cajal Hospital, Madrid, Spain∗Corresponding author: Rafael Barcena,rbarcena.hrc@salud.madrid.org

Antituberculous treatment is a well-known cause offulminant hepatic failure (FHF). This could lead to livertransplantation as the only possible treatment, whichon the other hand could be contraindicated due to ac-tive tuberculosis. The risk of aggressive disseminationof the disease after transplantation is not clearly de-termined by the current second-line antituberculoustherapies. We report a case of vertebral tuberculosistreated with rifampin, isoniazid and pyrazinamide. Hedeveloped an FHF that was treated with urgent livertransplantation. Despite the immunosuppression, thedisease was well controlled with ciprofloxacin, etham-butol and streptomycin and the patient is in goodhealth 23 months after transplantation. In conclusion,active extrapulmonary tuberculosis should perhaps beconsidered for liver transplantation when FHF devel-ops due to anti-tuberculous drugs.

Key words: Infection, liver failure, liver transplanta-tion, toxicity

Received 4 April 2005, revised and accepted for publi-cation 4 July 2005

Introduction

Isoniazid was introduced in 1952 for tuberculosis and ini-tially its hepatic toxicity was underestimated (1). This toxi-city was well described by Garibaldi in 1952 (2). From 10%to 20% of patients present a slight increase in aminotrans-ferase levels around 2–3 times above their normal values,during the first 2 months and frequently it resolves in spiteof going on with the treatment (3,4). Hepatitis, hepaticnecrosis and death have been described with this drug (5)and the only treatment of fulminant hepatic failure (FHF) is

liver transplantation. However, few cases have been pub-lished and tuberculosis evolution after transplantation isnot very well known (6,7).

We report one case of active extrapulmonary tuberculosiswith FHF due to isoniazid plus pyrazinamide, being finallytransplanted.

Case Report

We present the case of a 39-year-old patient whose maincomplaint was back pain, which had started 6 years be-fore his first consultation in our hospital. There were noremarkable features in his family or personal medical his-tory except for one episode of fever 2 years before whichlasted more than 1 month and whose origin was not com-pletely understood. In December 2002, he was admittedto our hospital because the back pain had worsened and hepresented a temperature of 39–40◦C during a long periodof 10 days. All hemocultures and urocultures were neg-ative. On magnetic resonance imaging (MRI), there wasa pathologic fracture of L5 vertebra and two probable ab-scesses, which extended toward the posterior and ante-rior sides, with a partial compromise of vertebral canal anda psoas abscess. The initial treatment was vancomycinand ceftazidime without a clear improvement. An openbiopsy was done showing granulomes, but the culture ofthe tissue was not positive at that moment to mycobac-terium. The clinical worsening of the patient and the pres-ence of positive tuberculin skin test and granulomes onbiopsy were considered enough at that moment to beginantituberculous treatment: isoniazid 300 mg/day, rifampin720 mg/day and pyrazinamide 1800 mg/day. Three days af-ter starting that treatment, fever disappeared and the clini-cal situation improved. Rifampin had to be stopped due toan exanthematic reaction, so it was changed to ethambu-tol plus levofloxacin. The clinical evolution was good andfever completely disappeared.

One month later, the patient was admitted again becauseof jaundice. The initial blood tests showed a bilirubin of12.2 mg/dL, AST 2062 IU/L and ALT 1993 IU/L. The treat-ment was immediately stopped with partial resolution ofthe hepatitis in 20 days. In March 2003, a new antituber-culous treatment was started (pyrazinamide, streptomycin,ciprofloxacin and ethambutol), since fever had increased to

2796

Liver Transplantation in Tuberculosis

39◦C. Five days later, cytolysis worsened again, so pyrazi-namide was definitely stopped.

Fifteen days later, a lumbar tumefaction appeared anda skin fistula was formed. The exudate was examinedand cultured, so Mycobacterium tuberculosis was isolated,confirming the disease.

Bilirubin continued increasing until a maximum of 38 mg/dLand the patient developed an FHF in April 2003. At thatmoment, the patient showed a liver atrophy and ascitis onultrasound. Hepatic encephalopathy grade II started andcoagulopathy reached an INR of 8, so the patient was fi-nally transplanted on April 25, 2003. High fever never dis-appeared until the moment of transplantation, so other in-fections were ruled out.

The explanted liver weighed 500 g and confluent submas-sive necrosis confirmed the diagnosis.

The antituberculous treatment before and after transplan-tation was initially ethambutol 800 mg/day, ciprofloxacin1000 mg/day and intravenous streptomycin 1 g/day.Tacrolimus was employed as immunosuppressant and my-cofenolate was later added due to slight renal failure. Thepatient was discharged on May 21, 2003, with completedisparition of fever.

The triple antituberculous therapy was continued, justchanging intravenous streptomycin to 1 g intramuscularonly three times per week. He has received this treatmentuntil the end of the follow-up (March 2005) and currentlythe therapy has not been stopped because the MRI stillshows a very small collection next to L4–L5 vertebrae.However, a dramatic amelioration has been observed com-pared to the image 1 year before. The disk has been de-stroyed with a fusion of both vertebral bodies. Back painhas also decreased but he still needs tramadol for completecontrol. Except for this complaint, he remains completelyasymptomatic.

The final diagnosis was vertebral tuberculosis and FHF dueto isoniazid plus pyrazinamide successfully treated with or-thotopic liver transplantation.

Discussion

Transplanted patients need immunosuppressive drugs forlife, with all their risks and adverse events. General dis-semination of tuberculosis after transplantation is well de-scribed due to active immunosuppression. Then, antituber-culous drugs are needed from the first moment, with theirevident toxicity.

Nearly, one-third of transplanted patients who are treatedwith antituberculous drugs will develop hepatic toxicity,which will be severe in half of them (8).

In a large study of 55 patients with post-transplant tuber-culosis, 31% died despite of the correct treatment (8). Inmore recent studies, prognosis seems to be better (9).

There is no alternative for FHF in these patients. Few seriesof patients have been published about on this issue. Farrellet al. (6) report two successfully transplanted cases. Oneof them was an FHF due to isoniazid chemoprophylaxis, sopost-transplant treatment was not needed. The other onewas receiving complete treatment with isoniazid, rifampinand pyrazinamide for pulmonary tuberculosis and he de-veloped an FHF. The patient was transplanted despite theisolation of M. tuberculosis in the sputum before trans-plantation. Ethambutol, streptomycin and ofloxacin werestarted. No tuberculosis reactivation was reported in thefollow-up after 4 months (6).

Another article (7) reports the outcome of seven cases ofFHF secondary to tuberculosis prophylaxis or treatment.Five of them were transplanted and only one died due toreasons not related to tuberculosis. The other four caseswere alive at the end of follow-up (mean 455 days, range162–797). Only one patient was being treated for active tu-berculosis, which would have led to post-transplant treat-ment. However, a spontaneous improvement of liver func-tion occurred, so he did not require transplantation. Thefive transplanted patients had received the drugs due totuberculin skin test conversion and there are no data aboutpost-transplant treatment or disease reactivation.

The question of considering active tuberculosis as a con-traindication to liver transplantation is a matter of debate.Based on the scarce information in the literature, universalrecommendations are not available on this issue. There-fore, each case should be individualized.

Our patient suffered a tuberculous espondilodiscitis withsmall nondrainable abscesses around. There was nopulmonary affection or other extrapulmonary localiza-tions. Before transplantation, treatment with ethambutol,ciprofloxacin and streptomycin was started and it was con-tinued later. During the following months, he presentedan evident improvement of his back pain and fever disap-peared.

Based on our case and other reported cases, we do notconsider the FHF due to antituberculous drugs and activeextrapulmonary disease as an absolute contraindication forliver transplantation. After this procedure, isoniazid mustnot be used as severe toxicity could appear again. More-over, transplanted patients seem to be more susceptibleto other antituberculous drugs (8,10). Other drugs suchas pyrazinamide, streptomycin, ethambutol and ofloxacinmust be used without higher risk of toxicity or rejection(9,10).

Our patient has received the described treatment for23 months without showing hepatic toxicity. Definitive

American Journal of Transplantation 2005; 5: 2796–2798 2797

Barcena et al.

randomized or controlled studies have not been performedto establish optimal regimens without isoniazid and ri-fampin. The guidelines from the American Thoracic Soci-ety, the Centers for Disease Control and Prevention andInfectious Diseases Society of America recommend ex-tended treatment during 18–24 months to lessen the riskof relapse in such cases (11). A prolonged course has beenrecommended in HIV and other immunosuppressed pa-tients. Finally, we decided to continue the treatment until24 months, because MRI showed a very small collectionnext to L4–L5 vertebrae during several months.

The management of tuberculosis in liver-transplanted pa-tients differs from general population due to an immuno-compromised status and a higher potential hepatotoxic-ity. The classical Public Health approach of short-coursetherapy with very effective drugs (isoniazid and rifampin)could be considered for those patients who present tu-berculosis some years after transplantation. Liver functionmust be closely monitored due to increased hepatotoxic-ity. In patients who present tuberculosis in the followingweeks or months after transplantation, the toxicity of an-tituberculous drugs could be confounded with other allo-graft disfunctions, but a very effective therapy should beemployed initially. A gentle sustained approach of longertherapies with second-line drugs should be reserved for pa-tients transplanted due to toxicity of antituberculous drugs.

After liver transplantation, immunosuppressants and an-tituberculous treatment must be prescribed togetherwith possible drug interactions. Rifampin increases thecatabolism of corticosteroids and cyclosporine, dramati-cally lowering serum levels even with careful monitoring.Therefore, rejection has shown to be a cause of mortalityin these patients. Tacrolimus may be the alternative (8). Ri-fampin has an evident hepatic toxicity, which could haveled to FHF together with isoniazid, so both drugs shouldbe avoided in that particular case. Streptomycin could po-tentiate renal toxicity of immunosuppressants.

In a patient who develops FHF due to isoniazid, it doesnot seem advisable to reuse this drug after transplanta-tion in spite of being another liver with different suscepti-bility to toxicity. The combination of other available drugsshould be the initial approach. Isoniazid toxicity is relatedto the genotype of cytochrome P450 2E1 and N-acetyltransferase, which are the main components of isoniazidmetabolism and therefore the production of toxic metabo-lites. Recently, homozygous wild genotype CYP 2E1 c1/c1has showed a higher risk of hepatic toxicity (20%, oddsratio 2.52) than the patients with c2 allele (9%, p = 0.009).The risk gets higher in slow acetylators, increasing to 7.43if both conditions present together (12).

A good approach could be to continue the treatment withnonhepatotoxic drugs, excluding the one responsible forprevious FHF. If isoniazid has to be used, genotype of CYP2E1 on a biopsy of the graft could help assess the new

probability of hepatic toxicity of this drug in the transplantedpatient. The phenotype of N-acetyl transferase in thesecases may also be determined to complete the real sus-ceptibility of the patient (13). A wild genotype CYP 2E1c1/c1 in a slow acetylator would lead to definitely avoidingisoniazid. In extreme clinical situations, careful monitoringof aminotransferases could help to detect toxicity at a veryearly stage.

In conclusion, we propose liver transplantation as the onlyalternative for FHF secondary to isoniazid, in active ex-trapulmonary tuberculosis, despite the difficulties of post-transplant anti-tuberculous treatment, as infection controlcan be achieved.

References

1. Dickinson DS, Bailey WC, Hirschowitz BI, Soong SJ, Eidus L,Hodgkin MM. Risk factors for isoniazid (INH)-induced liver dis-function. J Clin Gastroenterol 1981; 3: 271–279.

2. Garibaldi RA, Drusin RE, Ferebee SH, Gregg MB. Isoniazid-associated hepatitis: report of an outbreak. Am Rev Respir Dis1972; 106: 357–365.

3. Mitchell JR, Zimmerman HJ, Ishak KG et al. Isoniazid liver injury:clinical spectrum pathology and probable pathogenesis. Ann In-tern Med 1976; 84: 181–192.

4. Sharer L, Smith JP. Serum transaminase elevations and otherhepatic abnormalities in patients receiving isoniazid. Ann InternMed 1969; 71: 1113–1116.

5. Snider DE, Caras GJ. Isoniazid-associated hepatitis deaths: a re-view of available information. Am Rev Respir Dis 1992; 145: 494–497.

6. Farrell FJ, Keeffe EB, Man KM, Imperial JC, Esquivel CO. Treat-ment of hepatic failure secondary to isoniazid hepatitis with livertransplantation. Dig Dis Sci 1994; 39: 2255–2259.

7. Meyers BR, Halpern M, Sheiner P et al. Acute hepatic failure inseven patients after prophylaxis and therapy with antituberculousagents. Successful treatment with orthotopic liver transplanta-tion. Transplantation 1994; 58: 372–377.

8. Aguado JM, Herrero JA, Gavalda J et al. Clinical presentationand outcome of tuberculosis in kidney, liver and heart transplantrecipients in Spain. Transplantation 1997; 63: 1278–1286.

9. Torre-Cisneros J, Caston JJ, Moreno J et al. Tuberculosis in thetransplant candidate: importance of early diagnosis and treat-ment. Transplantation 2004; 77: 1376–1380.

10. Meyers BR, Papanicolaou GA, Sheiner P, Emre S, Miller C. Tuber-culosis in orthotopic liver transplant patients: increased toxicity ofrecommended agents: cure of disseminated infection with non-conventional regimens. Transplantation 2000; 69: 64–69.

11. Blumberg HM, Burman WJ, Chaisson RE et al. American ThoracicSociety/Centers for Disease Control and Prevention/InfectiousDiseases Society of America: treatment of tuberculosis. Am JRespir Crit Care Med 2003; 167: 603–662.

12. Huang YS, Chern HD, Su WJ et al. Cytochrome P450 2E1 geno-type and the susceptibility to antituberculosis drug-induced hep-atitis. Hepatology 2003; 37: 924–930.

13. Huang YS, Chern HD, Su WJ et al. Polymorphism of the N-acetyltransferase 2 gene as a susceptibility risk factor for anti-tuberculosis drug-induced hepatitis. Hepatology 2002; 35: 883–889.

2798 American Journal of Transplantation 2005; 5: 2796–2798