Coeliac disease and the liver: spectrum of liver histology, serology and treatment response at a tertiary referral centre

Transcription

1 JCP Online First, published on September 29, 2017 as /jclinpath Coeliac disease and the liver: spectrum of liver histology, serology and treatment response at a tertiary referral centre Kaushik Majumdar, 1 Puja Sakhuja, 1 Amarender Singh Puri, 2 Kavita Gaur, 1 Aiman Haider, 1 Ranjana Gondal 1 1 Department of Pathology, G B Pant Institute of Postgraduate Medical Education and Research, New Delhi, India 2 Department of Gastroenterology, GB Pant Institute of Postgraduate Medical Education and Research, New Delhi, India Correspondence to Professor Puja Sakhuja, Department of Pathology Academic Block, G B Pant Institute of Postgraduate Medical Education and Research, New Delhi , India; pujasak@ gmail. com Received 28 June 2017 Revised 31 August 2017 Accepted 1 September 2017 To cite: Majumdar K, Sakhuja P, Puri AS, et al. J Clin Pathol Published Online First: [please include Day Month Year]. doi: / jclinpath Abstract Background Coeliac disease (CD) is a gluten-sensitive enteropathy diagnosed on the basis of ESPGHAN criteria and clinical response to gluten-free diet (GFD). Histological abnormalities on liver biopsy have been noted in CD but have seldom been described. Aims To assess the histological spectrum of coeliac hepatitis and possibility of reversal of such features after a GFD. Methods Twenty-five patients with concomitant CD and hepatic derangement were analysed for clinical profile, laboratory investigations and duodenal and liver biopsy. A histological comparison of pre- and post-gfd duodenal and liver biopsies was carried out, wherever possible. Results Fifteen patients presenting with CD subsequently developed abnormal liver function tests; 10 patients presenting with liver disease were found to have tissue positive transglutaminase in 70% and antigliadin antibodies in 60%. Serological markers for autoimmune liver disease (AILD) were positive in eight patients. Liver histology ranged from mild reactive hepatitis, chronic hepatitis, steatosis to cirrhosis. Liver biopsies after a GFD were available in six cases, of which five showed a decrease in steatosis, portal and lobular inflammation and fibrosis score. Conclusion Coeliac hepatitis could be a distinct entity and the patients may present with either CD or secondary hepatic derangement. Evaluation for the presence of CD is recommended for patients presenting with AILD, unexplained transaminasaemia or anaemia. This is one of the very few studies demonstrating the continuum of liver histological changes in coeliac hepatitis. Trial of a GFD may result in clinicopathological improvement of coeliac hepatitis. Introduction Coeliac disease (CD) is characterised by a genetically determined autoimmune enteropathy in response to the gluten content of the diet. Various genetic, immunological and environmental factors have been implicated in the aetiopathogenesis of the disease. In addition to the gastrointestinal manifestations, CD may be associated with liver diseases like cryptogenic hepatitis/cirrhosis, autoimmune liver disease (AILD) including primary biliary cirrhosis (PBC) and primary sclerosing cholangitis (PSC). 1 2 A question which arises in this context is whether hepatic insult in association with CD ( coeliac hepatitis ) has any histological signatures. Compared with AILD, cryptogenic hepatitis or coeliac hepatitis may have different aetiopathogenesis, clinical presentation, histology and hence, differing response to treatment. 1 The occurrence of hepatic derangement in patients with CD was first described in the 1970s in both adults and children, which usually normalised after gluten from the diet. 3 5 The spectrum of liver involvement ranges from mild liver injury, steatosis and limited fibrosis to progressive liver disease, including cirrhosis A PubMed-based search for histopathology-based studies pertaining to CD associated with chronic liver disease (CLD) between 2000 and 2017 showed that original research on liver histology in CD is relatively unexplored territory. Previous articles have primarily dealt with serology alone. This study aimed to examine the histological features in liver biopsy associated with coeliac hepatitis and the possibility of reversal of such features after a gluten-free diet (GFD). In addition, investigation to determine the possible presence of CD has been carried out in patients with unexplained transaminasaemia, to search for potential coeliac hepatitis. Patients and methods Patients who unequivocally showed the presence of concomitant CLD and CD were selected from archival records at the departments of pathology and gastroenterology at our institute from 2004 to The diagnosis of CD was based on the European Society of Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) criteria, which include demonstration of histological small intestinal mucosal abnormality when the patient is consuming a normal diet. 8 9 The second requirement is an unequivocal remission of symptoms on a strict GFD. Histological assessment of the duodenal biopsies for villous architecture and other features of CD was carried out in accordance with the Marsh-Oberhuber classification Diagnostic criteria for CLD included those used for Child-Turcotte-Pugh (CTP) scoring of liver disease, such as hyperbilirubinaemia, hypoalbuminaemia, prolonged prothrombin time, ascites (serum albumin ascitic fluid gradient >1.1), raised liver enzymes, together with ultrasonographic features, endoscopic evidence of portal Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath Copyright Article author (or their employer) Produced by BMJ Publishing Group Ltd under licence.

2 Figure 1 Flow chart outlining the study design and subsequent diagnostic approach. CD, coeliac disease; CLD, chronic liver disease; CTP, Child- Turcotte-Pugh; FU, follow-up; GFD, gluten-free diet; IHC, immunohistochemistry. hypertension (oesophageal varices) or liver histology (where available). Patients were divided into two categories depending on whether CD or CLD was the diagnosis at presentation, based on clinical features and laboratory investigations (figure 1). In all cases relevant clinical details for CLD, such as history of jaundice (past or present), gastrointestinal bleed, hepatic encephalopathy and ascites were reviewed. The patients with CD were subcategorised into classical (diarrhoeal) or non-diarrhoeal presentation. Non-diarrhoeal CD presents with extraintestinal manifestations such as short stature, refractory iron deficiency anaemia or osteoporosis, in the absence of diarrhoea. 13 Laboratory investigations included a complete liver function test and serological workup with antinuclear antibodies (ANA), anti-smooth muscle antibodies (ASMA), anti-tissue transglutaminase (TTG), antigliadin antibodies (AGA) and endomysial 2 Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath

3 antibodies (EMA). Liver biopsies were screened to search for a spectrum of histological abnormalities, including features of mild reactive hepatitis, chronic hepatitis, steatohepatitis, autoimmune liver disease and cirrhosis. Cases having minimal lobular inflammation without significant interface activity with or without minimal (<5%) non-zonal steatosis were defined as mild reactive hepatitis. Chronic hepatitis was characterised by the presence of varying degrees of portal inflammation with interface activity and lobular inflammation with or without necrosis. The inflammatory activity of cases belonging to these two categories was scored using the modified Ishak s Histological Activity Index (necroinflammatory score). 14 Steatohepatitis was defined as the presence of hepatic steatosis and inflammation with hepatocyte injury (ballooning) with or without fibrosis. 15 Histological scoring (NAS score) was assigned to such cases. 16 Fibrosis was scored in all liver biopsies using Ishak s scoring system. 14 Data for liver histology were available in 21 of the 25 cases; follow-up liver biopsies were available in six patients after they had started a GFD. Immunohistochemistry for CD8 on paired pre- and post-treatment liver biopsies could be performed in three cases to look for any post-gfd reduction of CD8 positive cells. Results From the records of the departments of pathology and gastroenterology at our institute, 25 patients were found to have both CD and CLD. There were 8 male and 17 female subjects, with a mean age of 24.3 years (range 10 50). Out of approximately 400 patients diagnosed with CD during the given time period, around 40% had hypertransaminasaemia on follow-up, of whom 10% had hepatosplenomegaly or other evidence of CLD, for which a liver biopsy was performed for further evaluation. On the other hand, approximately 10% of patients with CLD (approximately 1000 cases at our institute in the same time period) on follow-up developed small bowel diarrhoea or unexplained anaemia, of whom 10% had serological or histological evidence of CD (figure 2). CD as the primary manifestation Fifteen patients were initially diagnosed with CD and subsequently found to have CLD. The presenting feature at the time of diagnosis was typical small bowel diarrhoea in 12 (80%), and non-diarrhoeal CD in the remaining three (20%) patients. Eight of these 15 patients had anaemia. Antibody for TTG was positive in all 15 patients (100%), whereas AGA were found in 13 (86.7%) patients and EMA in 3 out of 5 patients. Duodenal biopsy in eight (53.3%) patients showed severe villous abnormality (Marsh-Oberhuber type 3C), while three (20%) showed moderate villous abnormality (type 3B) and another three (20%) showed mild villous abnormality (type 3A) (table 1). Intraepithelial lymphocytes (IELs) were more than 30 per 100 duodenal enterocytes in all cases. Clinical features suggesting the development of liver disease in these 15 patients were ascites in 6 (40%), jaundice in 4 (26.7%), hepatomegaly in 4 (26.7%) and splenomegaly in 5 (33.3%) patients. The laboratory abnormalities seen in these patients were a reduced serum albumin in 9 (60%), raised liver enzymes (ALT/AST/ALP) in 8 (53.3%) and hyperbilirubinaemia in 4 (26.7%) patients (table 1). The CTP score for classification of liver disease ranged from 6 to 11, with a median value of 8. The interval between the two diagnoses ranged from 3 to 24 months. Autoimmune serological markers were available in six patients. ANA positivity was documented in two patients, and antimitochondrial antibody (AMA) and ASMA positivity in one patient each; two of these patients were Figure 2 Out of approximately 400 patients diagnosed with coeliac disease (CD) during the given time period, around 40% had hypertransaminasaemia on follow-up, of whom around 10% had clinicobiochemical evidence of chronic liver disease (CLD), confirmed by liver biopsy; around 10% of patients with CLD on follow-up developed small bowel diarrhoea or unexplained anaemia, of whom 10% had serological or histological evidence of CD (approximate percentages). HSP, hepatosplenomegaly. Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath

4 Table 1 Clinical and laboratory findings in patients with CD and CLD Findings Primary diagnosis CD Liver biopsy indicated Liver biopsy available Mild reactive hepatitis with steatosis 4 (33.3%) 1 (11.1%) 5/21 (23.8%) Chronic hepatitis 6 (50%) 4 (44.4%) 10/21 (47.6%) Steatohepatitis 1 (8.3%) (with cirrhosis) 1 (11.1%) 2/21 (9.5%) AILD 1 (8.3%) (PBC) 3 (33.3%) 4/21 (19%) Patient demographics Age (years), mean (range) 23.3 (10 50) 26.8 (12 50) M:F 4:11 4:6 Signs and symptoms of CD Diarrhoea small intestinal 12 (80%) 9 (90%) Non-diarrhoeal CD 3 (20%) 1 (10%) Anaemia 8 (53.3%) 6 (60%) Positive serology for CD IgA TTG 15 (100%) 7 (70%) IgA AGA 13 (86.7%) 6 (60%) EMA (n=5) 3/5 2/5 Duodenal biopsy (Marsh-Oberhuber classification) Marsh 3C 8 (53.3%) 6 (60%) Marsh 3B 3 (20%) 4 (40%) Marsh 3A 3 (20%) nil Marsh 1 1 (6.7%) nil Signs and symptoms of CLD Jaundice 4 (26.7%) 4 (40%) Ascites 6 (40%) 5 (50%) Gastrointestinal bleed 2 (13.3%) 2 (20%) Hepatomegaly 4 (26.7%) 5 (50%) Splenomegaly 5 (33.3%) 7 (70%) Biochemistry for CLD Reduced albumin 9 (60%) 8 (80%) Raised liver enzymes (ALT/AST/ALP) 8 (53.3%) 7 (70%) Hyperbilirubinaemia 4 (26.7%) 5 (50%) Positive autoimmune serology n=6 n=8 ANA 2 1 AMA 1 0 ASMA 1 3 AGA, antigliadin antibody; AILD, autoimmune liver disease; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AMA, antimitochondrial antibody; ANA, antinuclear antibody; ASMA, anti-smooth muscle antibody; AST, aspartate aminotransferase; CD, coeliac disease; CLD, chronic liver disease; EMA, antiendomysial antigen antibody; PBC, primary biliary cirrhosis; TTG, anti-tissue transglutaminase antibody. CLD Total negative for autoimmune markers. Liver histology was available in 12 patients. Four cases showed mild reactive hepatitis, with maintained lobular architecture, minimal to mild chronic inflammation in the portal tracts and minimal fibrosis. Focal microvescicular fatty change and occasional foci of lobular inflammation were seen in two cases. Chronic hepatitis was seen in six cases, with predominantly maintained lobular architecture and expansion of the portal tracts by lymphocytes and a few plasma cells, polymorphs and eosinophils (figure 3). Foci of lobular inflammation, focal ballooning and apoptosis were also seen. One of these cases showed focal duct injury. However, features suggestive of autoimmune hepatitis, such as interface hepatitis, plasma cell rich infiltrate, bridging necrosis, centrilobular necrosis or hepatocytic rosetting, were not seen. Ishak s modified Histological Activity Index score of mild reactive hepatitis ranged from 1 to 3, while that of chronic hepatitis ranged from 5 to 8 (table 2). Considerable fibrosis with fibrous spurring ( F2) was seen in three of these six cases showing chronic hepatitis (figure 4). The case with AMA positivity showed duct loss, portal lymphocytic infiltrate admixed with a few plasma cells, bile ductular proliferation, focal interface hepatitis and periportal Mallory hyaline, suggestive of PBC (figure 5). One case showed lobular disarray with diffuse severe microvescicular to macrovescicular steatosis, large foci of lobular inflammation and interface activity, consistent with steatohepatitis. Parenchymal nodules separated by fibrous septae bearing mild lymphocytic infiltrate were also seen (figure 6A). The NAS score for this case was 7 (table 2). Follow-up after a GFD showed an improvement in body mass index (BMI), haemoglobin and albumin levels in all cases. The increase in BMI in these patients ranged from 1.4 to 7.3 kg/m 2, the increase in haemoglobin ranged from 1.4 to 5.3 g/ dl and the serum albumin increase ranged from 0.4 to 1.5 g/ dl. Follow-up duodenal biopsies revealed restoration of villous pattern with a decrease in IELs. Follow-up liver biopsies after a 4 Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath

5 Downloaded from on October 3, Published by group.bmj.com Figure 3 (A) Pretreatment duodenal biopsy showing severe villous abnormality, increased intraepithelial lymphocytes (IELs) and crypt hyperplasia (Marsh 3C). (B) Liver biopsy (same patient) showing evidence of chronic hepatitis with expansion of the portal tracts by lymphocytes and few plasma cells, polymorphs and foci of lobular inflammation. (C) Follow-up duodenal biopsy after gluten-free diet (GFD) revealed restoration of villous pattern with decrease in IELs. (D) Liver biopsy after GFD, showing marked resolution of inflammation (haematoxylin & eosin stain, 100x). GFD were available in three cases initially diagnosed as CD, all of which showed histological improvement with reduction of fibrosis (figure 3). The case with steatohepatitis and parenchymal nodule formation showed near complete resolution of steatosis and marked regression of fibrosis (figure 6A, B) (table 3). CLD as the primary manifestation CLD was the initial manifestation in 10/25 (40%) patients. On subsequent investigation they were found to have CD. The clinical features at presentation suggestive of a liver disease in these patients were splenomegaly (70%), hepatomegaly (50%), ascites (50%), jaundice (40%) and gastrointestinal bleeding (20%). Liver function tests showed a reduced serum albumin in 8/10 (80%) patients, raised liver enzymes in 7/10 (70%) and hyperbilirubinaemia in 5/10 (50%) patients. The CTP score ranged from 5 to 9, with a median value of 7. Autoimmune serology was available in 8/10 patients. ASMA positivity was found in three of these eight (37.5%) patients and ANA positivity was seen in one patient (12.5%) (table 1). Viral serology for hepatitis B and hepatitis C were negative in all cases. Liver biopsy findings were available in nine of these 10 patients. Mild reactive hepatitis as described was seen in one Figure 4 (A) Pretreatment duodenal biopsy showing mild villous abnormality, increased intraepithelial lymphocytes and crypt hyperplasia (Marsh 3A) (haematoxylin & eosin stain, 100x). (B) Liver biopsy (same patient) showing evidence of chronic hepatitis with fibrous spurring (Ishak s score: F4) (Masson s trichrome stain, 200x). (C, D). Immunohistochemistry for CD8 in pre- and post-treatment liver biopsies, respectively, showing some reduction of post-gluten-free diet CD8 positive lymphocytes (200x). case, while chronic hepatitis with mild to moderate interface hepatitis and fibrous spurring was seen in four cases (figure 6C). Plasma cells were seen as part of the inflammatory infiltrate but were not conspicuous; mild focal ballooning was also observed in five cases. Bile duct injury was noted in four patients. The Histological Activity Index score of the case of mild reactive hepatitis was 1 and those of chronic hepatitis ranged from 5 to 6 (table 2). One of these cases had shown Table 2 Histological assessment of inflammation (HAI and NAS) in the CLD and CD groups Histological category CD CLD Minimal reactive hepatitis (median HAI, range) 2 (1 3) 1 (Number of cases) (4) (1) Chronic hepatitis (median HAI, range) 6 (5 8) 5 (5 6) (Number of cases) (6) (4) Steatohepatitis (NAS score) 7 4 (Number of cases) (1) (1) CD, coeliac disease; CLD, chronic liver disease; HAI, Histological Activity Index; NAFLD, non-alcoholic fatty liver disease; NAS, NAFLD Activity Score. Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath Figure 5 (A) Case with antimitochondrial antibody (AMA) positivity; duodenal biopsy showing moderate villous abnormality with increased intraepithelial lymphocytes (Marsh 3B) (haematoxylin & eosin stain, 200x). (B, C) Liver biopsy showing duct loss, portal lymphocytic infiltrate admixed with a few plasma cells, bile ductular proliferation and focal interface hepatitis, suggestive of primary biliary cirrhosis (Masson s trichrome stain, 100x and 200x, respectively) (D) Immunohistochemistry for CK7 confirming duct loss and bile ductular proliferation (200x). 5

6 Table 4 Fibrosis stage (Ishak s) in liver biopsies of chronic liver disease (CLD) and coeliac disease (CD) Fibrosis stage (Ishak) CD (n=12) CLD (n=9) F0 4 0 F1 3 2 F2 2 1 F3 1 2 F4 1 3 F5 0 1 F6 1 0 Figure 6 (A) Case of Marsh 3C, on liver biopsy showing lobular disarray with diffuse severe microvescicular to macrovescicular steatosis, lobular inflammation and parenchymal nodules separated by fibrous septae (Masson s trichrome stain, 100x). (B) Marked resolution of steatohepatitis and fibrosis on gluten-free diet (haematoxylin & eosin stain, 100x). (C) Chronic hepatitis with moderate interface hepatitis and fibrous spurring (F4), which subsequently on duodenal biopsy showed changes of Marsh 3C (Masson s trichrome stain, 200x). (D) Case with anti-smooth muscle antibody positivity showing moderate interface hepatitis, focal rosetting of hepatocytes, portocentral bridging necrosis and fibrosis (F5), suggestive of autoimmune hepatitis (subsequently Marsh 3C) (Masson s trichrome stain, 100x). Parenchymal nodules separated by fibrous septae bearing mild lymphocytic infiltrate were also seen. Table 3 Pre- and post-gfd liver biopsies in patients presenting with CD/CLD Case no. Diagnosis at presentation Fibrosis score - first biopsy Interval between biopsies (months) Fibrosis score - second biopsy Remarks 1 CD F2 8 F1 Decrease in fibrosis, portal and lobular inflammation 2 CD F4 12 F2 Decrease in fibrosis, portal and lobular inflammation; no steatosis on follow-up 3 CLD F5 2 F6 Patient died; ASMA+ 4 CLD F1 13 F0 No significant fibrosis or lobular inflammation on follow-up 5 CLD F4 17 F3 Mild decrease in fibrosis and less lobular inflammation; ANA+ 6 CD F6 11 F2 Marked decrease in hepatic injury, no steatosis ANA, antinuclear antibody; ASMA, anti-smooth muscle antibody; CD, coeliac disease; CLD, chronic liver disease; GFD, gluten-free diet. steatosis in a previous liver biopsy performed elsewhere; subsequently, a biopsy done 3 months later at our hospital showed chronic hepatitis. This case had a NAS score of 4 (table 2). According to Ishak s staging, seven cases showed fibrosis stage >2 (table 4). Three out of four cases with positive autoimmune serology showed focal rosetting of hepatocytes, interface hepatitis and porto-central bridging necrosis suggestive of autoimmune hepatitis (figure 6D). Subsequently, CD with small intestinal type of diarrhoea was seen in nine (90%) of these 10 patients, while one case had non-diarrhoeal CD. The interval between the two diagnoses ranged from 2 to 132 months. Six (60%) patients, including the one with non-diarrhoeal CD had associated anaemia. On serological screening antibody for TTG was positive in 7 (70%), AGA in 6 (60%) and EMA in 2 out of 5 patients. Duodenal biopsy showed severe villous abnormality (type 3C) in 6 (60%) and moderate villous abnormality (type 3B) in 4 (40%) cases (table 1). Seven patients had high titre for more than one antibody for CD. The median titre of TTG and AGA was 140 and 41.6 IU/L, respectively. Duodenal biopsies following a GFD in these cases also showed restoration of villous architecture with decrease in IELs. Follow-up liver biopsies were available in three cases initially diagnosed as CLD. One patient with ASMA positivity showed an increase in the degree of fibrosis; this patient subsequently died. The remaining two cases showed a decrease in steatosis, portal and lobular inflammation with mild reduction of fibrosis (table 3). Immunohistochemistry for CD8 was possible in three follow-up liver biopsies, which showed reduction in inflammation and/or fibrosis; these cases also showed reduction of CD8 positive cells in portal tracts with a GFD (figure 4). Discussion The hepatic manifestations that may be associated with CD include asymptomatic rises of liver enzymes, non-specific hepatitis, steatosis and autoimmune liver disease A possible pathogenetic mechanism of CLD in association with CD might be impaired gut mucosal integrity, leading to malabsorption and increased mucosal permeability, triggering an autoantibody or toxin-mediated liver injury. 19 Mucosal inflammation in CD also leads to exposure of tissue transglutaminase in the endomysium, which acts as the autoantigen recognised by endomysial IgA antibody. TTG IgA antibodies may also serve as a risk factor for other autoimmune diseases; hence CD and CLD may have a common autoimmune basis. Other common predisposing factors for both diseases to coexist are the HLA molecules, especially HLA-DR3, known to have an overlapping role in the pathogenesis of CD and autoimmune hepatitis (AIH). 9 6 Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath

7 CD as the primary manifestation An increase in liver enzymes has been previously documented in patients with CD; some larger case series have shown such rises in 40 60% of patients In this study, 15 patients with CD were subsequently found to have CLD. Investigations showed abnormal liver function tests, such as reduced serum albumin in 60%, raised liver enzymes in 53.3% and hyperbilirubinaemia in 26.7% of patients. The spectrum of histological abnormalities in liver biopsies occurring in association with CD ranges from mild reactive hepatitis, steatosis, chronic active hepatitis (CAH), fibrosis, to cirrhosis in varying degrees and percentages In one such study, CAH found in five of seven liver biopsies examined, was related to chronic hepatitis B infection in three patients, hepatitis C in one and autoimmune aetiology in the remaining one patient. 17 Mild hepatic steatosis is a common histological finding in patients with CD who undergo biopsy. In some patients it may be more severe, attributed to the presence of malnutrition, with documentation of response to a GFD. In this study, liver histology was available in 12/15 patients, with mild reactive hepatitis seen in four patients, histological evidence of chronic hepatitis in six patients, steatohepatitis in one patient and AILD suggestive of PBC in one patient. Autoimmune markers were available in six patients; ANA positivity was found in two patients, ASMA in one patient and AMA was positive in the patient with PBC. In a large series including 909 children and young adults with CD, the prevalence of autoimmune disorders was found to be significantly higher in patients with CD than in the control group. AIH was also observed in 1.1% of these patients, with no cases occurring in the healthy control group. It was concluded that the prevalence of autoimmune disorders in CD depends on the duration of exposure to gluten. 22 The prevalence of hypertransaminasaemia in patients with CD ranges from 15% to 61% (in this study: 40%) and conversely, the prevalence of CD in patients with unexplained hypertransaminasaemia is 10%, 1 which also agrees with this study. The question which needs to be answered is whether cryptogenic hepatitis/ coeliac hepatitis and the AILDs that arise in patients with CD, are distinct entities or part of a broad spectrum of coeliac liver disease. 1 Alternatively, are these co-occurrences of CD and hepatic insults merely chance associations? Mounajjed et al 1 mentioned five patients with CD having non-specific findings on liver biopsy, where no obvious cause other than CD could be demonstrated for liver disease. Liver biochemistry improved on a GFD; hence these cases may be examples of potential coeliac hepatitis. Similarly herein, of 15 cases with a primary diagnosis of CD, autoimmune or other causes of liver disease could not be elicited in 11 cases. Liver histology in these cases showed mild hepatitic inflammation, chronic hepatitis, steatohepatitis and cirrhosis. These cases also showed biochemical improvement on a GFD, together with documented histological improvement in all three cases where follow-up biopsies were available, establishing the diagnosis of coeliac hepatitis. The liver histology mentioned in this group is rather non-specific and shows a continuum of changes, depending on the duration of gluten exposure. This might be because the patients had already been following a GFD if CD was the initial diagnosis. Hence, specific groups who should be evaluated for coeliac hepatitis include CLD with alarming features of CD, or CD with unexplained transaminasaemia or organomegaly. CLD as the primary manifestation Studies have shown that in patients with unexplained hypertransaminasaemia, about 10% have positive serological screening for CD. 23 Small intestinal biopsies from these patients also showed changes of CD that responded to a GFD, corroborating the positive serology Liver biopsies done in some of these patients showed a non-specific inflammatory process, and liver chemistry normalised with a GFD In our study CLD was the presenting manifestation in 10 patients, who later were found to have CD. TTG was positive in 70% and AGA in 60% of these patients. Duodenal biopsy was done in all the cases. Severe villous abnormality (type 3C) was seen in 60% (6/10 patients) and moderate villous abnormality (type 3B) in 40% of the biopsies. Markers for AILD were investigated in eight of these 10 patients; ASMA positivity was found in three (37.5%), while ANA positivity was seen in one patient. In a study involving 47 consecutive patients with AIH, three cases were positive for IgA anti TTG and EMA antibodies and were subsequently confirmed to have CD through small bowel histology. 24 Mounajjed et al 1 observed that the histological features of liver biopsy in CD-associated AILD were typical of the respective liver disease, with inconsistent response to a GFD. Hence the authors inferred that the association of AIH, PBC and PSC with CD may represent clustering of diseases in patients prone to autoimmune conditions. In our study, cases of CD demonstrating serological evidence of AIH showed histological features of chronic hepatitis; one case of PBC revealed characteristic histomorphology. Liver biopsy after a GFD was available in two patients, of whom one deteriorated and died, while the other showed marginal biochemical and histological improvement. Studies have shown a high prevalence of CD (3.5%) in autoimmune cholestasis and AILD Hence, based on the present study and other previous studies, the potential candidates who should be evaluated for CD are patients with AILD or those having CLD/ unexplained hypertransaminasaemia, or unexplained anaemia (non-bleeder). Table 5 compares the liver histology of the present study with data from two previous studies. Future work correlating duodenal histological grade (in accordance with the Marsh-Oberhuber classification) with CTP scoring (for severity of hepatic manifestations) involving larger case series may be useful. To conclude, CD may be a potential cause of cryptogenic liver disease or coeliac hepatitis. AILDs that arise in patients with CD and coeliac hepatitis are possibly distinct entities with different immunopathogenesis, clinical presentation, histology and hence, Table 5 Comparison of the present data with prior studies according to the histological diagnosis Author, journal, year Total cases AIH PBC PSC Cryptogenic hepatitis Steatohepatitis Viral hepatitis Miscellaneous Mounajjed et al 1 Am J Clin Pathol Drastich P et al 27 World J Gastroenterol (WD) Present data AIH, autoimmune hepatitis; PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; WD, Wilson s disease. Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath

8 possibly, different response to a GFD. Patients with coeliac hepatitis may be a potentially treatable group responsive to a GFD. The liver histology of this group is rather non-specific and shows a continuum of changes (mild reactive hepatitis, chronic hepatitis, steatohepatitis to cirrhosis) depending on the duration of gluten exposure, because the patients may be already following a GFD if CD is the initial diagnosis. For effective management, evaluation to determine the possible presence of CD should be recommended for patients with AILD, unexplained transaminasaemia, anaemia or organomegaly. Take home messages Liver histology in coeliac disease encompasses a broad spectrum of changes, ranging from mild reactive hepatitis, chronic active hepatitis, to steatohepatitis and even advanced fibrosis. Coeliac hepatitis is a potentially treatable subset of cryptogenic hepatitis responsive to a gluten-free diet. Histological changes may reverse after dietary intervention. In cases where the aetiology of liver disease is unknown, evaluation for coeliac disease may be rewarding. Handling editor Runjan Chetty Contributors KM: Collection and analysis of data, literature review and drafting of the manuscript. PS, ASP: Concept of work, providing intellectual content of critical importance and editing and final approval of the manuscript. KG: Collection and analysis of data, image preparation, literature review and drafting of the manuscript. AH, RG: Performing histopathological analysis, providing intellectual content of critical importance and editing of the manuscript. Competing interests None declared. Provenance and peer review Not commissioned; externally peer reviewed. Article author(s) (or their employer(s) unless otherwise stated in the text of the article) All rights reserved. No commercial use is permitted unless otherwise expressly granted. References 1 Mounajjed T, Oxentenko A, Shmidt E, et al. The liver in celiac disease: clinical manifestations, histologic features, and response to gluten-free diet in 30 patients. Am J Clin Pathol 2011;136: Fasano A, Catassi C. Current approaches to diagnosis and treatment of celiac disease: an evolving spectrum. Gastroenterology 2001;120: Lindberg T, Berg NO, Borulf S, et al. Liver damage in coeliac disease or other food intolerance in childhood. Lancet 1978;1: Pollock DJ. The liver in coeliac disease. Histopathology 1977;1: Hagander B, Berg NO, Brandt L, et al. Hepatic injury in adult coeliac disease. Lancet 1977;2: Gaur K, Sakhuja P, Puri AS, et al. Gluten-Free hepatomiracle in "celiac hepatitis": a case highlighting the rare occurrence of nutrition-induced near total reversal of advanced steatohepatitis and cirrhosis. Saudi J Gastroenterol 2016;22: Kaukinen K, Halme L, Collin P, et al. Celiac disease in patients with severe liver disease: gluten-free diet may reverse hepatic failure. Gastroenterology 2002;122: Walker-Smith JA, Guandalini S, Schmitz J, et al. Revised criteria for diagnosis of coeliac disease. Report of working group of European Society of Paediatric Gastroenterology and Nutrition. Arch Dis Child 1990;65: Davison S. Coeliac disease and liver dysfunction. Arch Dis Child 2002;87: Mulder C, Rostami K, Marsh MN. When is a coeliac a coeliac? Gut 1998;42: Rostami K, Kerckhaert J, Tiemessen R, et al. Sensitivity of antiendomysium and antigliadin antibodies in untreated celiac disease: disappointing in clinical practice. Am J Gastroenterol 1999;94: Green PH, Rostami K, Marsh MN. Diagnosis of coeliac disease. Best Pract Res Clin Gastroenterol 2005;19: Puri AS, Garg S, Monga R, et al. Spectrum of atypical celiac disease in North Indian children. Indian Pediatr 2004;41: Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995;22: Chalasani N, Younossi Z, Lavine JE, et al. The diagnosis and management of nonalcoholic fatty liver disease: practice guideline by the American Association for the Study of liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Am J Gastroenterol 2012;107: Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005;41: Bardella MT, Fraquelli M, Quatrini M, et al. Prevalence of hypertransaminasemia in adult celiac patients and effect of gluten-free diet. Hepatology 1995;22: Jacobsen MB, Fausa O, Elgjo K, et al. Hepatic lesions in adult coeliac disease. Scand J Gastroenterol 1990;25: Rubio-Tapia A. Murray JA: The liver in celiac disease. Hepatology 2007;46: Naschitz JE, Yeshurun D, Zuckerman E, et al. Massive hepatic steatosis complicating adult celiac disease: report of a case and review of the literature. Am J Gastroenterol 1987;82: Franzese A, Iannucci MP, Valerio G, et al. Atypical celiac disease presenting as obesityrelated liver dysfunction. J Pediatr Gastroenterol Nutr 2001;33: Ventura A, Magazzù G, Greco L. Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease. SIGEP Study Group for Autoimmune Disorders in Celiac Disease. Gastroenterology 1999;117: Bardella MT, Vecchi M, Conte D, et al. Chronic unexplained hypertransaminasemia may be caused by occult celiac disease. Hepatology 1999;29: Villalta D, Girolami D, Bidoli E, et al. High prevalence of celiac disease in autoimmune hepatitis detected by anti-tissue tranglutaminase autoantibodies. J Clin Lab Anal 2005;19: Volta U, Rodrigo L, Granito A, et al. Celiac disease in autoimmune cholestatic liver disorders. Am J Gastroenterol 2002;97: Volta U, De Franceschi L, Molinaro N, et al. Frequency and significance of antigliadin and anti-endomysial antibodies in autoimmune hepatitis. Dig Dis Sci 1998;43: Drastich P, Honsová E, Lodererová A, et al. Celiac disease markers in patients with liver diseases: a single center large scale screening study. World J Gastroenterol 2012;18: Majumdar K, et al. J Clin Pathol 2017;0:1 8. doi: /jclinpath

9 Coeliac disease and the liver: spectrum of liver histology, serology and treatment response at a tertiary referral centre Kaushik Majumdar, Puja Sakhuja, Amarender Singh Puri, Kavita Gaur, Aiman Haider and Ranjana Gondal J Clin Pathol published online September 29, 2017 Updated information and services can be found at: References alerting service These include: This article cites 27 articles, 3 of which you can access for free at: #BIBL Receive free alerts when new articles cite this article. Sign up in the box at the top right corner of the online article. Notes To request permissions go to: To order reprints go to: To subscribe to BMJ go to:

10 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具