Article Text

Download PDFPDF

Liver dysfunction related to hepatitis B and C in patients with inflammatory bowel disease treated with immunosuppressive therapy
  1. C Loras1,
  2. J P Gisbert2,3,
  3. M Mínguez4,
  4. O Merino5,
  5. L Bujanda3,6,
  6. C Saro7,
  7. E Domenech3,8,
  8. J Barrio9,
  9. M Andreu10,
  10. I Ordás3,11,
  11. L Vida12,
  12. G Bastida3,13,
  13. F González-Huix14,
  14. M Piqueras15,
  15. D Ginard16,
  16. X Calvet3,17,
  17. A Gutiérrez3,18,
  18. A Abad19,
  19. M Torres20,
  20. J Panés3,11,
  21. M Chaparro2,3,
  22. I Pascual4,
  23. M Rodriguez-Carballeira21,
  24. F Fernández-Bañares1,
  25. J M Viver1,
  26. M Esteve1,
  27. for the REPENTINA study,
  28. GETECCU (Grupo Español de Enfermedades de Crohn y Colitis Ulcerosa) Group
  1. 1Department of Gastroenterology, Hospital Universitari Mútua de Terrassa, Fundació per la Recerca Mútua de Terrassa, Terrassa, Spain
  2. 2Department of Gastroenterology, Hospital de la Princesa, Madrid, Spain
  3. 3Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD), Madrid, Spain
  4. 4Department of Gastroenterology, Hospital Clínico Universitario de Valencia, Valencia, Spain
  5. 5Department of Gastroenterology, Hospital de Cruces, Barakaldo-Vizcaya, Spain
  6. 6Department of Gastroenterology, Hospital de Donostia, University of the Basque Country, San Sebastian-Guipúzcoa, Spain
  7. 7Department of Gastroenterology, Hospital de Cabueñes, Gijón, Asturias, Spain
  8. 8Department of Gastroenterology, Hospital Universitari Germans Trias i Pujol, Badalona, Spain
  9. 9Department of Gastroenterology, Hospital Universitario Río Hortega, Valladolid, Spain
  10. 10Department of Gastroenterology, Hospital del Mar, Barcelona, Spain
  11. 11Department of Gastroenterology, Hospital Clínic, Barcelona, Spain
  12. 12Department of Gastroenterology, Hospital Reina Sofía, Cordoba, Spain
  13. 13Department of Gastroenterology, Hospital la Fe, Valencia, Spain
  14. 14Department of Gastroenterology, Hospital Josep Trueta, Girona, Spain
  15. 15Department of Gastroenterology, Consorci de Terrassa, Catalonia, Spain
  16. 16Department of Gastroenterology, Hospital de Son Dureta, Palma de Mallorca, Spain
  17. 17Department of Gastroenterology, Hospital Parc Taulí de Sabadell, Sabadell, Spain
  18. 18Department of Gastroenterology, Hospital General de Alicante, Alicante, Spain
  19. 19Department of Gastroenterology, Hospital de St Llorenç, Viladecans, Spain
  20. 20Department of Gastroenterology, Hospital de l'Esperit Sant, Barcelona, Spain
  21. 21Internal Medicine, Hospital Universitari Mútua de Terrassa, Fundació per la Recerca Mútua de Terrassa, Terrassa, Spain
  1. Correspondence to Maria Esteve, Department of Gastroenterology, Hospital Mútua de Terrassa, Universitat de Barcelona, Plaça Dr Robert n 5, 08221 Terrassa, Barcelona, Catalonia, Spain; mestevecomas{at}telefonica.net

Footnotes

  • Linked articles 217331.

  • Funding CIBEREHD is funded by the Instituto de Salud Carlos III.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the ethics committee of the Hospital Universitari Mútua de Terrassa.

  • Provenance and peer review Not commissioned; externally peer reviewed.

View Full Text

Statistics from Altmetric.com

Significance of this study

What is already known about this subject?

  • HBV and HCV may reactivate in patients receiving chemotherapy, being more frequent and severe for HBV than for HCV.

  • Treatment with immunosuppressants in other disease conditions has been associated with a more rapid progression to liver cirrhosis in patients infected with HCV.

  • Case reports of severe HBV reactivation in patients with IBD mainly treated with infliximab have been reported.

  • The true frequency and clinical relevance of liver dysfunction in patients with IBD receiving immunosuppressants is not known.

What are the new findings?

  • The frequency and severity of liver dysfunction in patients with IBD treated with immunosuppressants is described for the first time in a large cohort of HBV- and/or HCV-infected patients.

  • Liver dysfunction in patients with IBD treated with immunosuppressants is more frequent and severe in those with HBV than in HCV carriers and is associated with combined immunosuppression.

  • No definite HBV reactivations were found in anti-HBc positive patients lacking HBsAg.

  • The percentage of liver cirrhosis in IBD-infected patients with HBV and HCV is similar to that reported in natural history studies and does not seem to be influenced by immunosuppressants.

How might it impact on clinical practice in the foreseeable future?

  • It is clear from the present study that, irrespective of the number and type of immunosuppressants administered, all HBsAg positive patients should receive prophylactic antiviral treatment before starting immunosuppressant therapy. In contrast, the systematic use of antiviral prophylactic treatment in anti-HBc positive patients lacking HBsAg is not recommended. Nevertheless, liver function tests and HBV DNA should be periodically monitored in these patients.

Introduction

Data from patients with hepatitis B virus (HBV) and hepatitis C virus (HCV) infection receiving chemotherapy for haematological malignancies show that liver dysfunction, including fulminant hepatitis by either reactivation or enhanced replication, may be a frequent complication.1 The frequency and severity of liver dysfunction seems to be more important for HBV than for HCV.2 However, treatment with immunosuppressants has been associated with a more rapid progression to liver cirrhosis in HCV-infected patients.3 Reactivation of HBV infection related to infliximab in inflammatory bowel disease (IBD) has been reported in a number of cases4–9 and may be severe if prophylactic treatment is not administered.4 9 By contrast, anti-tumour necrosis factor (TNF) therapies seem to have no adverse effect on HCV infection or may even improve the short-term virological response to specific treatments.10 A recent European evidence-based consensus on the prevention, diagnosis and management of opportunistic infections in IBD highlights the lack of information on the potential effect of various immunosuppressants on the course of viral hepatitis.11 Moreover, the potential deleterious effect of different immunosuppressants, given alone or in combination, on the outcome of HBV and HCV infection has never been systematically evaluated in patients with IBD.

This information is highly relevant. Despite the fact that mortality in IBD was reported to be similar or slightly higher to that of the general population, a large proportion of deaths are due to opportunistic infections related to immunosuppression.12

The aim of this study was to evaluate the influence of different immunosuppressants on liver dysfunction in a large cohort of patients with IBD infected with HCV and/or HBV.

Patients and methods

Design and inclusion criteria

This was a multicentre nationwide study performed in 19 Spanish hospitals. A systematic search was performed of all patients with ulcerative colitis (UC) and Crohn's disease (CD) with hepatitis B and C infection markers. In order to be included in the study, all patients had to: (1) be diagnosed with UC or CD; (2) have positive markers of present or past hepatitis B and/or C infection; (3) have been treated with immunosuppressants at some time for IBD; and (4) have reliable clinical and analytical data in the clinical records. Those patients with antibodies to hepatitis B surface antigen (anti-HBs) alone showing effective HBV vaccination were not included. Eligible patients were identified from the previous REPENTINA study,13 from the ENEIDA registry (IBD national registry on genetic and environmental determinants) and from the local database of each participant hospital. Medical records were reviewed to obtain the date of the hepatitis and IBD diagnoses as well as all immunosuppressants received by each patient, treatment duration, liver function tests and viral markers before, during and after each immunosuppressant. Laboratory analytical data obtained several months after immunosuppressant withdrawal were recorded in order to identify late reactivations. In addition, to establish whether immunosuppressants increased the number and severity of reactivations compared with those expected due to the natural history of viral infection, those reactivations occurring far away from immunosuppressant administration were also recorded.

An electronic database for the REPENTINA study (REGISTRO HEPATITIS ENFERMEDAD INFLAMATORIA INTESTINAL) was created to record all the information via the Internet (http://www.repentina.com/). Reliable patient data were available from 1983 to 2009 and the database remained opened from March 2006 to March 2009. The number of patients included per hospital ranged from 1 to 24 and the number of IBD patient files between participant centres ranged from 210 to 1337.

Factors related to hepatic disease

The following analytical laboratory variables assessing liver function were recorded: aminotransferase levels (ALT, AST), alkaline phosphatase (AP), γ-glutamyltranspeptidase (GGT), total bilirubin, albumin, haemogram and prothrombin time, all measured by standard laboratory technique. Clinical data related to hepatic function such as presence of hepatic encephalopathy or ascites were recorded in order to establish the existence of acute hepatic failure (see definitions below).

HBV and HCV infection markers

The following markers were recorded: hepatitis B surface antigen (HBsAg), antibodies to hepatitis B surface antigen (anti-HBs), antibodies to hepatitis B core protein (anti-HBc), hepatitis B e antigen (HBeAg), antibodies to hepatitis B e antigen (anti-HBe) and antibodies to hepatitis C virus (anti-HCV), HCV RNA and HBV DNA. Owing to the long duration of the study period, the standard tests to detect viral antigens and antibodies directed against them and PCR techniques to detect viral load have increased and improved technically over time.14 15

Definitions

Because of differences in the natural history of hepatitis B and C, different terminology is used to define liver dysfunction associated with these infections in the context of immunosuppressant treatment.

Reactivation of HBV is a liver dysfunction defined as an increase of 1.5–2-fold the baseline value of alanine transaminase (ALT) plus an increase of >2000 IU/ml (104 copies/ml) HBV DNA levels or DNA reappearance in a negative patient. Reactivation was considered probable when an increase in ALT was observed in the absence of other potential causative factors but DNA values were not available.

Enhanced HCV replication is defined as appearance of liver dysfunction characterised by a significant increase in RNA-HCV or RNA reappearance in a negative patient plus an increase of 1.5–2-fold the baseline value of ALT. Probable liver dysfunction related to immunosuppressants in an HCV-infected patient was considered when either a significant increase of ALT or RNA-HCV virus load was observed. It should be taken into account that the laboratory analytical profile related to immunosuppressant administration was available in some patients diagnosed with non-A, non-B hepatitis before 1989 who were later diagnosed with hepatitis C. If HBV-DNA or HCV-RNA was not available, particular attention was paid to rule out drug-related toxicity.

Acute liver failure was defined as a sudden and severe impairment of liver function based on worsening in analytical variables such as bilirubin (>2 mg/dl), albumin (<34 g/l) or prothrombin time (<50%). Fulminant liver failure was defined as severe acute failure complicated by hepatic encephalopathy.16

The diagnosis of hepatic cirrhosis was established by clinical, laboratory, analytical and morphological criteria (CT scan/ultrasound/liver biopsy).

Statistical analysis

The results are expressed as mean±SEM and percentages. In univariate analysis the χ2 test and Fisher exact test were used to compare proportions and the Student t test was used to compare quantitative variables. Multivariate analysis was performed using binary logistic regression analysis. Factors considered relevant for liver dysfunction (number of immunosuppressants and prophylactic antiviral treatment) were entered into logistic regression analysis as dichotomous variables. Kaplan–Meier analyses were used to calculate reactivation curves with the log-rank test to compare them. Cox proportional-hazards regression was used to analyse the effect of the number of immunosuppressants (0, 1 or >1) on the risk of reactivation. The odds ratio (OR) and hazard ratio (HR) and its 95% confidence interval (CI) were calculated to assess the strength of each significant association. All statistics were handled with the SPSS 13.0 for Windows statistical package (SPSS Inc).

Results

Demographics of the IBD population and hepatic dysfunction related to immunosuppressants

A total of 162 patients (74 with UC and 88 with CD; 96 men, 66 women; mean age 51.1±1.1 years) with HBV and HCV infection were identified. The median time from IBD diagnosis at inclusion was 144 months (range 12–672). The maximum disease activity during disease evolution was moderate in 50% of the patients and severe in 36.6%, and the majority of them had an episodic evolution (73%). In table 1 the clinical characteristics of the IBD population according to the Montreal classification are shown.

Table 1

Clinical characteristics of IBD population according to the Montreal classification

One hundred and four patients had HBV markers, 74 had HCV markers and 16 patients had markers of both infections. A description of the different viral markers is presented in table 2.

Table 2

Distribution of viral markers in patients with inflammatory bowel disease with hepatitis B and C

The patients received a mean of 2.9 courses of immunosuppressants (range 1–11) and 78 patients (48.1%) received ≥2 simultaneous immunosuppressants. One hundred and twenty-four patients (76.5%) received prolonged immunosuppression defined as immunosuppressive treatment administered for >3 months as maintenance treatment; 84% of the patients received corticosteroids, 69% azathioprine, 24% infliximab, 8% ciclosporin, 7% adalimumab, 5.6% methotrexate, 1.9% tacrolimus and 0.6% certolizumab.

Rates of liver dysfunction and liver failure in HBsAg and HCV-RNA positive patients are shown in table 3.

Table 3

Percentage of liver dysfunction and liver failure in patients with inflammatory bowel disease with hepatitis B and C infection

The frequency of liver dysfunction was significantly higher in HBV-infected patients than in HCV-infected patients (p=0.045). In addition, the appearance of liver failure in HBV-infected patients with liver dysfunction was also significantly more frequent than in HCV-infected patients (p=0.049).

Liver dysfunction related to HBV

Liver dysfunction was observed in 9 of 25 HBsAg positive patients (36%). None of the clinical or therapeutic factors evaluated were significantly related to liver dysfunction in the univariate analysis (table 4), although a trend towards a higher rate of HBV reactivations was observed in HBV-DNA positive patients, in those treated with >1 immunosuppressant and in patients with prolonged immunosuppression.

Table 4

Factors related to liver dysfunction in hepatitis B-infected patients

Conversely, patients treated with prophylactic antiviral treatment showed a trend towards a lower rate of liver dysfunction. Of the variables described in table 4, the number of immunosuppressants and preventive antiviral treatment, considered the most clinically relevant factors, were included in a multivariate analysis. The only factor significantly related to HBV reactivation was treatment with ≥2 concomitant immunosuppressants (OR 8.75; 95% CI 1.16 to 65.66). Moreover, the Cox proportional hazard method using the number of immunosuppressants as a covariate showed that being treated with ≥2 immunosuppressants significantly increased the number of reactivations compared with treatment with only one immunosuppressant or not being treated with immunosuppressants (table 5). In figure 1 the reactivation curves related to exposure to 0, 1 and >1 immunossuppressants are shown (overall p value 0.011).

Table 5

Reactivations related to periods of immunosuppression (Cox proportional hazard analysis)

Figure 1

Reactivation curves during treatment with 0, 1 and >1 immunosuppressants (IMM). HBV, hepatitis B virus.

In table 6 the type of immunosuppressants, prophylactic treatment administered, outcome of patients with reactivation and changes in IBD therapy due to reactivation are detailed.

Table 6

Hepatitis B virus-infected patients with reactivations

In contrast, the majority of the 16 positive HBsAg patients without reactivations were treated with only one immunosuppressant for a short period of time and/or prophylactic antiviral treatment (table 7).

Table 7

Hepatitis B virus-infected patients without reactivations

No definite cases of liver dysfunction related to the immunosuppressants occurred among HBsAg negative patients/anti-HBc positive with or without anti-HBs. However, in one of these cases, positive for anti-HBc and anti-HBs, HBV-DNA was detected twice in peripheral blood while under treatment with infliximab plus azathioprine. The aminotransferase levels remained within the normal range and a loss of anti-HBs was observed during viral load peaks, suggesting the existence of an occult HBV infection (figure 2).

Figure 2

Evolution of hepatitis B virus DNA in a patient with antibodies to hepatitis B core protein (anti-HBc) without hepatitis B surface antigen (HBsAg) treated with infliximab plus azathioprine.

Only in two additional cases was an increase in aminotransferase levels related to corticosteroid treatment documented. Although the association suggests a probable reactivation, HBV DNA was not available for analysis in these cases.

Liver dysfunction related to HCV

Liver dysfunction was observed in 8 of 51 RNA-HCV positive patients (15.7%), and no cases of RNA-HCV reappearance were observed in anti-HCV positive patients with negative RNA-HCV. A clearly enhanced replication related to immunosuppressants was observed in 6 out of 8 patients and was likely in the remaining 2 cases (one with a marked increase in viral load and the other with a marked increase in ALT but lacking HCV-RNA). Owing to the low rate of liver dysfunction in HCV-infected patients, no additional analysis for risk factors was performed. Nevertheless, it is important to note that, of the eight patients with HCV-related liver dysfunction, seven were receiving treatment only with corticosteroids and the remaining patient was being treated with azathioprine. The course of liver dysfunction was very mild in all the cases except in one patient who died of liver failure while receiving steroids. This patient also had anti-HBc positive markers without anti-HBs and tested positive for antibodies against HIV (AIDS).

Hepatic cirrhosis in HBV- and HCV-infected patients with IBD receiving immunosuppressive therapy

Four of the 25 HBsAg positive patients (16%) developed liver cirrhosis after a mean of 13.5±4.2 years from the diagnosis of HBV infection. No cases of cirrhosis were detected in anti-HBc positive patients without HBsAg.

Among the 74 HCV-positive patients, 8 (10.8%) were diagnosed with liver cirrhosis after a mean time of 16.7±4.8 years from the diagnosis of HCV infection. One case developed hepatocellular carcinoma. No relation was found between the diagnosis of liver cirrhosis and the use of an immunosuppressant alone or in combination. There was also no relationship between the development of cirrhosis and the duration of immunosuppressant therapy (table 8).

Table 8

Association between cirrhosis and immunosuppression in hepatitis B and C virus-infected patients

Discussion

To the best of our knowledge, this is the largest series assessing the outcome of HVB and HCV infection in patients with IBD and its relation to immunosuppressive therapy. We have shown that liver dysfunction in patients with IBD treated with immunosuppressants is more frequent and severe in patients with HBV than in those with HCV infection. These data are in keeping with results previously reported in patients receiving chemotherapy for haematological malignancies.1 2 Apart from its novelty, this information is important for its clinical relevance since the type, duration and combination of immunosuppressants differs from that used in oncological therapy and in other disease conditions requiring immunosuppressants.

In this regard, most of the HBV-infected patients who had reactivation were receiving treatment with ≥2 immunosuppressants for a long period of time, were positive for HBV-DNA and/or had not received prophylactic antiviral treatment. However, the only factor significantly related to HBV reactivation in the multivariate analysis was combined treatment with ≥2 immunosuppressants. In fact, in previous studies treatment with a combination of immunosuppressants was identified as the most consistent factor related to the risk of suffering de novo and/or reactivation of opportunistic infections in IBD.17 18 In our study the therapeutic approach adopted for the majority of patients during the follow-up after reactivation was to administer the least possible immunosuppression to maintain IBD in remission together with prophylactic antiviral treatment.

Most cases of liver dysfunction in HBV- and HCV-infected patients related to immunosuppressants reported in the literature in patients with IBD were associated with infliximab administration.4–9 19–21 In contrast, our study shows that no one particular immunosuppressant seems to be more involved in the development of liver dysfunction, which probably develops as a consequence of combined immunosuppression. In this sense, all reported cases of liver dysfunction related to anti-TNF treatment occurred in patients receiving not only anti-TNF agents but combined immunosuppression.

We observed a trend towards a lower percentage of HBV reactivation among patients receiving antiviral prophylactic treatment. These results are in keeping with previous studies demonstrating efficacy of antiviral prophylactic treatment in the prevention of HBV reactivation in immunosuppressed patients.22 Statistical significance in our study may not have been reached due to the small number of patients receiving prophylactic treatment. This was due in part to the inclusion of a historical series of patients diagnosed before the availability and wide use of effective antiviral drugs such as lamivudine. In fact, lamivudine has been used for HBV infection since 199523 and was approved by the European Medicines Agency in 1999. However, until very recently there was a lack of awareness of the risk of HBV reactivation in patients with IBD treated with immunosuppressants which contributes to a low perception of the potential danger. Until 2005 there were no recommendations on HBV and HCV screening in IBD guidelines for patients requiring immunosuppressants.24

It is clear from the present study that, irrespective of the number and type of immunossuppressants administered, all HBsAg positive patients should receive prophylactic antiviral treatment before starting immunosuppressant therapy. In fact, an almost significant difference was observed in the number of reactivations in patients receiving one immunosuppressant compared with that observed in periods without immunosuppression. These results suggest that the risk of HBV reactivation is related to the magnitude of the immunosuppression.

In contrast, it is worth noting the low percentage of probable but not confirmed HBV reactivation among anti-HBc positive patients without HBsAg (2 of 63, 3.13%) with no cases of liver failure, which argues against the systematic use of antiviral prophylactic treatment in this condition. This attitude differs from that recommended for patients undergoing chemotherapy, particularly with rituximab. In these patients, antiviral preventive treatment is recommended if anti-HBc is detected, even in those patients lacking HBsAg, based on reported cases of severe HBV reactivation in this setting.25 To our knowledge, only one case of HBV reactivation in a patient with CD with positive anti-HBc and negative HBsAg treated with corticosteroids and infliximab has previously been reported.8 Nevertheless, liver function tests and HBV DNA should be periodically monitored in anti-HBc positive patients with or without anti-HBs during immunosuppressive therapy since occult HBV carriers are more frequent in this particular group.26 27 In our study, in one case treated with infliximab plus azathioprine occult HBV was confirmed, since HVB-DNA was detected twice in peripheral blood without evidence of reactivation. Owing to the retrospective design of the present study, our recommendation concerning anti-HBc positive patients with IBD receiving immunosuppressive therapy needs to be confirmed in an appropriately designed prospective study such as that which is currently ongoing (REPENTINA III Study).

With regard to HCV, mild liver dysfunction was observed only in 15.7% of HCV-RNA positive patients, most of them related to corticosteroids alone and in one case to azathioprine. The only case showing severe liver impairment and death was a patient with simultaneous anti-HBc positive markers and AIDS. Though not confirmed, this patient had a possible occult HVB infection which may have contributed to liver dysfunction. Occult HBV infection frequently occurs in HIV/HCV co-infected patients, producing a severe course of liver disease.28–30

The distinction between liver dysfunction related to immunosuppressants and liver dysfunction related to HBV or HCV may be difficult in patients with probable but not confirmed reactivations. However, in our study, all cases with probable reactivations (2 HCV, 2 anti-HBc and 1 HBsAg positive patients) were treated with corticosteroids, which are drugs not related to hepatotoxicity.

In the present study, the proportion of patients with liver cirrhosis and the mean time to develop it in both HBV and HCV infection was within the range of that commonly reported, considering the natural history of both infections. In this sense, the progression to liver cirrhosis is similar to that previously observed in immunocompetent infected patients.31 32 Thus, the administration of immunosuppressants, as commonly used in IBD, does not seem to increase progression to end-stage liver disease. This is in contrast with observations in liver transplant recipients in whom increased progression to liver cirrhosis related to immunosuppression has been suggested.33 34

In conclusion, liver dysfunction in patients with IBD treated with immunosuppressants was more frequent and severe in those with HBV than in those with HCV infection, with combined immunosuppression being the most important factor related to HBV reactivation. The type of immunosuppressants used in IBD management does not seem to accelerate the progression to liver cirrhosis in HCV- and HBV-infected patients.

Acknowledgments

The authors are grateful to Amelia Rodríguez and M Pilar Civit for their helpful assistance in collecting data.

References

View Abstract

Footnotes

  • Linked articles 217331.

  • Funding CIBEREHD is funded by the Instituto de Salud Carlos III.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the ethics committee of the Hospital Universitari Mútua de Terrassa.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles