Share this article on:

Celiac Disease–Associated Autoimmune Hepatitis in Childhood: Long-Term Response to Treatment

Nastasio, Silvia*; Sciveres, Marco; Riva, Silvia; Filippeschi, Irene P.*; Vajro, Pietro; Maggiore, Giuseppe*

Journal of Pediatric Gastroenterology and Nutrition: June 2013 - Volume 56 - Issue 6 - p 671–674
doi: 10.1097/MPG.0b013e31828b1dfa
Original Articles: Hepatology and Nutrition

Objectives: Celiac disease (CD) is common in patients with autoimmune liver disease (AILD); however, the long-term response to treatment of patients with AILDs coexistent with CD has not been explored in detail. The aim of the present study was to analyze the features and the long-term response to immunosuppressive treatment in children with autoimmune hepatitis (AIH) associated with CD.

Methods: Retrospective and prospective evaluation of patients followed at a single center.

Results: Among 79 patients with AIH, 15 (19%) had CD (9 type 1, 3 type 2, 3 seronegative). In the group of patients with AIH and CD, female sex was significantly more represented than in the group of patients with AIH alone; also, in the former group, diagnosis was made significantly earlier (P < 0.05). All of the 15 patients on a gluten-free diet achieved sustained remission when treated with prednisone and azathioprine or cyclosporine. The mean period of follow-up was 73 months; discontinuation of therapy was attempted in 9 patients while in remission: 4 patients relapsed, 5 (33%) could definitively stop immunosuppressive treatment with a mean period of treatment-free sustained remission of 89 months (range 26–174). In the same period, treatment discontinuation, attempted in 24 of 64 patients with AIH without CD, was successful in 5 patients (8%; P < 0.05).

Conclusions: Patients with AIH coexisting with CD achieve treatment-free sustained remission in a significantly higher proportion, when compared with patients with AIH without CD, suggesting a possible long-term adjuvant effect of a gluten-free diet.

*Department of Clinical and Experimental Medicine, Division of Pediatric Gastroenterology, University of Pisa, Pisa

Liver Transplantation and Pediatric Hepatology Unit, Ismett UPMC, Palermo

Department of Medicine and Surgery, University of Salerno, Salerno, Italy.

Address correspondence and reprint requests to Giuseppe Maggiore, Division of Pediatric Gastroenterology, University of Pisa, Azienda Ospedaliera Universitaria Pisana, Via Roma 67, Pisa 56123, Italy (e-mail:

Received 27 September, 2012

Accepted 1 February, 2013

The authors report no conflicts of interest.

Celiac disease (CD) is an immune-mediated disorder that may be associated with liver morbidity (1). Reversible liver damage manifesting as hypertransaminasemia (2) may be directly related to gluten consumption as in the so-called celiac hepatitis (3), or may coexist as chronic liver disease, especially of autoimmune etiology (4–6). CD is common in patients with autoimmune liver diseases (AILDs) (7), particularly in children. In fact, it was shown in a multicenter study that 16.4% of pediatric patients with AILD had coexistent CD (8). The pathogenetic role of gluten in triggering AILD is uncertain, and although a possible favorable effect of gluten withdrawal in patients with autoimmune hepatitis (AIH) coexistent with CD has been suggested (9,10), the long-term response to treatment of this subgroup of patients has not been explored in detail.

The aim of the present study was to analyze the long-term response to immunosuppressive treatment in children and adolescents with AIH and CD. The present study was approved by the ethics commission of the Department of Medicine of the University of Pisa. Informed consent was obtained from all of the parents or guardians of the children or, where appropriate, directly from the adolescents.

Back to Top | Article Outline


The study was retrospective for patients observed between January 1995 and June 2000, whereas all consecutive patients who received a diagnosis of AIH were later included. Patients were enrolled in the study until January 2012.

Diagnosis of AIH implied the exclusion of all known causes of liver damage of infectious, toxic, and genetic origin, including Wilson disease and α1-antitrypsin deficiency, and the presence of inflammatory liver damage on liver biopsy (11).

All of the patients included in the study were serologically screened for CD by anti-endomysium antibody and anti-transglutaminase antibody of immunoglobulin A (IgA), or IgG class in case of IgA deficiency. In case of seropositivity, a small bowel histological evaluation was performed, and the diagnosis of CD was confirmed on the basis of the presence of a modified Marsh grade ≥2 (12,13).

All of the patients with AIH were administered an immunosuppressive treatment (14). Patients with a coexistent CD also started a gluten-free diet (GFD). Treatment endpoints, including the timing of attempted treatment discontinuation, as well as postimmunosuppressive drug withdrawal remission and relapse criteria, were in agreement with the recent guidelines (15). Patients regularly underwent clinical and biochemical follow-up evaluations, in which compliance with the treatment and GFD, as well as the occurrence of adverse effects, were assessed.

Differences in the distribution across groups were assessed with the χ 2 test. All of the analyses were conducted using SAS 9.0 (SAS Institute, Cary, NC).

Back to Top | Article Outline


Among the 108 children and adolescents admitted with a suspected AILD, AIH was diagnosed in 79, autoimmune cholangitis in 14, and overlap syndrome hepatitis/cholangitis in 15. Fifteen children and adolescents (M:F 1:7) with AIH (13.8% of AILD and 19% of AIH) were also diagnosed as having CD (Table 1). The mean age at clinical onset was 91.8 months. Of these 15 patients, 7 were already included in a previous multicenter study (8). Family history of an immune-mediated disease was present in 7 of the 15 patients with AIH and CD: CD in 3, inflammatory bowel disease in 2, type 1 diabetes mellitus in 2, and autoimmune thyroiditis in 1. Patients with AIH coexistent with CD were significantly more likely to be girls than those with AIH alone (P < 0.05, χ 2 test). Median age at onset for patients with AIH was significantly lower in those with additional CD than in those without CD (59.6 vs 91.8 months; P < 0.05, χ 2 test). Of the 15 patients with AIH and CD, 2 subgroups were identified: 1 group consisting of 8 children presenting with gastrointestinal symptoms and 1 group of 7 children presenting with signs of liver disease.



Patients presenting with gastrointestinal symptoms, such as chronic diarrhea and clinical evidence of malabsorption, had a mean age of 43.2 months (range 9–123 months); at clinical examination, 3 had hepatomegaly and all except 1 had biochemical evidence of liver damage with serum alanine aminotransferase ranging from 1.5 to 50 times the upper limit of normal. AIH was diagnosed from 2 to 14 months (median 7 months) after the diagnosis of CD: in 7 because aminotransferase activity was persistently abnormal on GFD and, in the patient with normal aminotransferase activity on presentation, because of a later occurrence of acute hepatitis. Three of these 8 patients had an elevated IgG level for age, 6 had autoantibodies (anti-nuclear antibody in 3, LKM1 in 2, and anti-smooth muscle antibody in 1), and 2 were seronegative for all AILD-related autoantibodies.

The 7 patients presenting with liver disease signs had a mean age of 78.3 months (range 30–126). Liver disease was symptomatic in 2 presenting with jaundice and asymptomatic in 5 with the fortuitous discovery of persistent elevation of aminotransferase activity (alanine aminotransferase ranging from 6–65 × upper limit of normal). Six had elevated IgG level for age and 1 had a complete IgA deficiency. Three showed anti-nuclear antibody and/or anti-smooth muscle antibody reactivity, 1 LKM1 reactivity, and 3 were seronegative. Diagnosis of CD was performed in all within 32 months (median 5 months) from the diagnosis of AIH following a systematic serologic screening. International AIH group score (11) and pediatric simplified score (16) were compatible with probable or definite AIH in the 5 seronegative patients belonging to both of the subgroups.

In all 15 patients, liver histology was consistent with the diagnosis of AIH: all had interface hepatitis with various degrees of portal fibrosis; 2 patients with acute onset had severe lobular inflammatory activity; 2 had portal eosinophilic infiltrate; and 2 had evidence of bile duct lesions. Portal fibrosis was severe in 3, and 1 patient had cirrhosis (Table 1). In the 2 patients with bile duct lesions, histological evidence of bile duct damage was mild, and magnetic resonance biliary tract imaging showed absence of medium-to-large size bile duct gross abnormalities.

Associated immune-mediated disorders were diagnosed in 7 children, which are inflammatory bowel disease, cretaceous vasculitis, type 1 diabetes mellitus, vitiligo, autoimmune thyroiditis, psoriasis, bone marrow aplasia, and transient erythroblastopenia.

All 15 patients were treated with immunosuppressive therapy: 14 were treated with prednisone and azathioprine, and 1 patient, affected with type 1 diabetes mellitus, was treated with cyclosporine. Ursodeoxycholic acid, at a dose of 30 mg · kg−1 · day−1, was also administered in the 2 patients with bile duct damage. All of the patients achieved clinical and biochemical remission after a median interval of 8.2 weeks (range 3–20). No significant adverse effects of the treatment were recorded.

When the data were collected, the mean period of follow-up was 73 months. Discontinuation of therapy was attempted in 9 patients while in remission: 4 experienced relapses, and treatment was maintained according to the latest dose associated with remission. In 1 case, cyclosporine was also added to prednisone and azathioprine to maintain remission. It was possible to cease treatment in 5 patients (4 AIH type 1 and 1 seronegative) with a mean follow-up period of sustained, treatment-free remission of 89 months. Of the 64 patients with AIH without coexisting CD followed up in the same center (Table 1), discontinuation of the immunosuppressive treatment was attempted in 24, 5 of which achieved sustained, treatment-free remission (P < 0.05; χ 2 test).

Back to Top | Article Outline


The present study confirms the previous findings from an Italian multicenter survey (8), and recent small pediatric series (17,18), of an elevated prevalence of CD in children with AIH. Liver damage, as evidenced by elevated aminotransferase activity, was present from the first observation in all except 1 celiac patient presenting with gastrointestinal symptoms. Patients presenting because of liver disease were in most cases asymptomatic; however, a hidden CD was diagnosed through a systematic serological screening. Among AILDs reported to coexist with CD (5,7), AIH is the most frequent form observed (8), as was previously shown (8), and in those patients with histological evidence of sporadic, small intrahepatic bile duct damage, medium-to-large bile ducts were unaffected, confirming the diagnosis of AIH (19).

Female sex had a more significant representation in patients with AIH and CD than in patients with AIH not associated with CD (Table 1). This could be related to the higher prevalence of female sex in CD (20). Moreover, diagnosis was made significantly earlier in patients with AIH and CD than in patients with AIH alone, suggesting that patients with known CD are more carefully supervised undergoing routine clinical and laboratory evaluation.

Serum autoantibodies are a useful tool in supporting the diagnosis of AIH in patients with cryptogenic inflammatory liver damage (14); however, as observed (8,21), some children in the present study were seronegative for all autoantibodies including anti-liver cytosol (22) associated with AIH. Seronegative AIH is a well-recognized phenomenon in adults (23), whereas it is not as well defined in children. This suggests that excluding an AILD in patients with CD on the basis of the absence of autoantibodies should only be done with caution. Moreover, even if the difference was not statistically significant, one-third of patients with AIH and CD were seronegative versus approximately 10% of patients with AIH without CD, suggesting, as in a previously reported case (21), that this serological feature may be more common in patients with AIH and CD. All of the patients with AIH associated with CD achieved first remission with conventional immunosuppressive therapy, and even if few patients relapsed during tapering, all achieved sustained remission without significant adverse effects.

With regard to long-term follow-up, it is reported that only approximately 20% of patients with AIH type 1 successfully and permanently stop therapy, and that this endpoint can rarely be obtained in AIH type 2 (24). In our study, we observed that a significantly higher number of children with both types of AIH coexistent with CD could stop the immunosuppressive therapy without relapse when compared with patients without CD. This observation is in agreement with the study of Di Biase et al (10) carried out in a small celiac pediatric population with AIH that showed how GFD associated with immunosuppressive treatment determines a high remission rate. These findings support a possible favorable effect of gluten withdrawal in modifying the natural history of AIH coexistent with CD. Gluten withdrawal may enhance the immunosuppressive effect of the drugs used, and help maintain remission once the treatment has been withdrawn. This is similar to a previous observation that patients with type 1 diabetes mellitus and CD receiving GFD see an improvement in metabolic glycemic control, probably secondary to an improvement of gut permeability (25).

The substantial limitations of the present study are the relatively small number of patients involved owing to the small number of patients fulfilling inclusion criteria and the mixed prospective and retrospective data collection. This methodological bias may influence the results of the present study. A prospective study would clarify our preliminary observation.

In conclusion, our experience confirms the need for serological screening for CD of all patients with AILD. Moreover, it supports the need for a close follow-up of all patients with clinical and/or biochemical signs of liver damage at the time of diagnosis of CD to exclude the coexistence of AIH, particularly in the case of lack of normalization of liver enzymes on GFD. The absence of serum autoantibodies should not lead to the exclusion of AILD.

Although long-term treatment is usually required for the majority of patients with AIH (26), the apparently more favorable response to treatment in the subgroup of children with coexistent CD suggests that in such cases, ceasing therapy earlier may be effective.

Back to Top | Article Outline

Dreckapotheke Redux

Dreckapotheke (DA) (from the German dreck meaning “dirt” or “filth” and apotheke meaning “pharmacy” or “medical recipe”) is scatological pharmacy that incorporates the medicinal compounding of dung, urine, menstrual flow, and other excreta. It can be animal or human; when it is human, it is more often referred to as zebethum occidentale.

DA has ancient roots. The earliest mention is found in the Egyptian pharmacopeia of the Middle Kingdom (1990–1570 BCE), with a focus on crocodile dung as a contraceptive (mixed with honey and natron in a tampon). Hippocrates (?460–?377 BCE) used pigeon, poultry, goat, and donkey dung in various compounds. He recommended a slurry of cow dung for intravaginal use against “inflammation.” Alexandrian Herophilus (335–280 BCE) advocated DA both topically and internally. Ge Hong (283–343 CE) used an oral suspension of human feces for treating food poisoning and severe diarrhea. Paracelsus (1493–1541 CE) compounded zebethum occidentale—a civet of human stools—and his contemporary Martin Luther (1483–1546 CE) in the last months of his life was treated with a soup of garlic and fresh manure. Li Shizhen (1518–1593 CE) formulated his “yellow soup” (feces from various sources) for diarrhea, abdominal pain, vomiting and constipation.

For a span of some 5000 years DA was used medicinally. The era ended in the 19th century, when germ theory evolved the concept of “good” and “bad” bacteria. Today we have changed the name to a more acceptable and sanitized clinical term: fecal microbiome transplant.

“What is past, is prologue.”

The Tempest II: I

Contributed by Angel Colón, MD

Back to Top | Article Outline


1. Ludvigsson JF, Elfstrom P, Broome U, et al. Celiac disease and risk of liver disease: a general population-based study. Clin Gastroenterol Hepatol 2007; 5:63–69.
2. Sainsbury Sanders DS, Ford AC. Meta-analysis: coeliac disease and hypertransaminasaemia. Aliment Pharmacol Ther 2011; 34:33–40.
3. Maggiore G, Caprai S. The liver in celiac disease. J Pediatr Gastroenterol Nutr 2003; 37:117–119.
4. Rubio Tapia A, Murray JA. The liver in celiac disease. Hepatology 2007; 46:1650–1658.
5. Lawson A, West J, Aithal GP, et al. Autoimmune cholestatic liver disease in people with celiac disease: a population-based study of their association. Aliment Pharmacol Ther 2005; 21:401–405.
6. Vajro P, Paolella G, Pisano P, et al. Hypertransaminasemia and coeliac disease. Aliment Pharmacol Ther 2012; 35:203–204.
7. Mirzaagha F, Azali SH, Islami F, et al. Coeliac disease in autoimmune liver disease: a cross-sectional study and a systematic review. Dig Liver Dis 2010; 42:620–623.
8. Caprai S, Vajro P, Ventura A, et al. Autoimmune liver disease associated with celiac disease in childhood: a multicenter study. Clin Gastroenterol Hepatol 2008; 6:803–806.
9. Rubio Tapia A, Murray JA. Celiac disease beyond the gut. Clin Gastroenterol Hepatol 2008; 6:722–723.
10. Di Biase AR, Colecchia A, Scaioli E, et al. Autoimmune liver disease in a paediatric population with coeliac disease—a 10 year-single center experience. Aliment Pharmacol Ther 2010; 31:253–260.
11. Alvarez F, Berg PA, Bianchi FB, et al. International Autoimmune Hepatitis Group report: review of criteria for diagnosis of autoimmune hepatitis. J Hepatol 1999; 31:929–938.
12. Revised criteria for diagnosis of celiac diseaseReport of working group of European Society for Pediatric Gastroenterology and Nutrition. Arch Dis Child 1990; 65:909–911.
13. Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol 1999; 11:1185–1194.
14. Maggiore G, Riva S, Sciveres M. Autoimmune diseases of the liver and biliary tract and overlap syndromes in childhood. Minerva Gastroenterol Dietol 2009; 55:53–70.
15. Manns MP, Czaja AJ, Gorham JD, et al. Diagnosis and management of autoimmune hepatitis. Hepatology 2010; 51:2193–2213.
16. Hiejima E, Komatsu H, Sogo T, et al. Utility of simplified criteria for the diagnosis of autoimmune hepatitis in children. J Pediatr Gastroenterol Nutr 2011; 52:470–473.
17. Tosun MS, Ertekin V, Selimoğlu MA. Autoimmune hepatitis associated with celiac disease in childhood. Eur J Gastroenterol Hepatol 2010; 22:898–899.
18. El-Shabrawi M, El-Karaksy H, Mohsen N, et al. Celiac disease in children and adolescents with autoimmune hepatitis: a single-centre experience. J Trop Pediatr 2011; 57:104–108.
19. Czaja AJ, Carpenter HA. Autoimmune hepatitis with incidental histologic features of bile duct injury. Hepatology 2001; 34:659–665.
20. Megiorni F, Mora B, Bonamico M, et al. HLA-DQ and susceptibility to celiac disease: evidence for gender differences and parent-of-origin effects. Am J Gastroenterol 2008; 103:997–1003.
21. Quail MA, Russel RK, Bellamy C, et al. Seronegative autoimmune hepatitis presenting after diagnosis of coeliac disease: a case report. Eur J Gastroenterol Hepatol 2009; 21:576–579.
22. Bridoux-Henno L, Maggiore G, Johanet C, et al. Features and outcome of autoimmune hepatitis type 2 presenting with isolated positivity for anti-liver cytosol antibody. Clin Gastroenterol Hepatol 2004; 2:825–830.
23. Gassert DJ, Garcia H, Tanaka K, et al. Corticosteroid-responsive cryptogenic chronic hepatitis: evidence for seronegative autoimmune hepatitis. Dig Dis Sci 2007; 52:2433–2437.
24. Mieli-Vergani G, Vergani D. Autoimmune hepatitis in children: what is different from adult AIH? Semin Liver Dis 2009; 29:297–306.
25. Uibo R, Panarina M, Teesalu K, et al. Celiac disease in patients with type 1 diabetes: a condition with distinct changes in intestinal immunity? Cell Mol Immunol 2011; 8:150–156.
26. Mieli-Vergani G, Heller S, Jara P, et al. Autoimmune hepatitis. J Pediatr Gastroenterol Nutr 2009; 49:158–164.

autoimmune hepatitis; celiac disease; immunosuppressive treatment

Copyright 2013 by ESPGHAN and NASPGHAN