The typical features of PSC (with or without concomitant AIH) are intrahepatic and/or extrahepatic large duct involvement with cholangiographically characteristic multifocal strictures and segmental dilatation or beading. Alternatively, PSC may present as histologically characteristic bile duct changes similar to long-standing large duct obstruction and eventually ductopenia, but with negative cholangiogram, thus designated small duct PSC [5, 12, 13]. Among PSC patients with positive cholangiogram, an overlap with definite or probable AIH has in adults (or mainly adults) been diagnosed in 1.4%-35% [4, 6, 14–16] while in children diagnosed in 28%-49% of the cases [5, 7, 9, 17, 18]. However, PSC in children is frequently characterized serologically by florid AIH-like features including ANA and SMA, elevated IgG and histologically by interface hepatitis, wherefore these children may be diagnosed as AIH in the absence of cholangiographic studies or with an initially normal cholangiogram, unless characteristic bile duct changes are identified histologically. Moreover, despite bile duct involvement, cholestatic biochemistry in children is relatively rare, many having normal ALP and in some cases even normal GGT at presentation . Hence, when either AIH or PSC is diagnosed in children, it may take years before an overlap syndrome, i.e. AISC, is diagnosed . To further confuse the picture, “incidental” histologic biliary changes may occur in classic AIH, which does not appear to develop features of PSC. Czaja and Carpenter  presented 84 AIH patients (including a small but unspecified number of children) among which ten appeared histologically to have destructive cholangitis or ductopenia. However, the AIH treatment response was not dependent on bile duct injury wherefore the authors were reluctant to consider the lesions as PSC. They proposed that the bile duct lesions were coincidental findings associated with classic disease, or weak expressions of a variant syndrome. In the study of Gregorio et al.  8/26 children (31%) with AIH (in the absence of radiologic features of cholangiopathy) showed histologic biliary features of which one was classified as (small duct) PSC.
The diagnostic criteria formulated by the IAIHG  do not readily allow for a distinction between AIH (i.e., without PSC) and AISC, and is not directly applicable in children. Thus, the score for our patient could be calculated to 21, based on – among others – an ALP:ALAT ratio <1.5, but ALP is often normal in children with PSC and should probably be substituted with GGT [9, 20].
In the current case, the liver changes are those of a chronic cholangiopathy within the spectrum of PSC. Peculiar to our patient was the severe inflammatory changes in relation to the small intrahepatic bile ducts with heavy neutrophil infiltration and features of destructive injury and marked hyperplasia, resembling the granulocytic epithelial lesion (GEL) in autoimmune pancreatitis (AIP) type 2 , a feature which we have previously seen published only once, in a case study of Grammatikopoulos et al. , who described a 13-year-old boy with concomitant Crohn colitis and AISC with GEL, who showed long-term remission of liver disease after steroid treatment.
Bile duct injuries in PSC and AISC may be considered multifactorial, potentially involving immune-mediated, chemical, genetic, ischaemic, and infectious mechanisms, as reviewed by Krones et al. . In the above mentioned case  as well as ours, the GELs and the remission after steroid treatment supports an immune-mediated disease in line with AIP type 2, which interestingly is associated with IBD . This is in contrast to the IgG4-associated PSC and hepatic inflammatory pseudotumour, which may occur concomitant with AIP type 1 .
The hyperplastic bile duct lesions described in this case bear some resemblance to the so-called hepatitis-associated bile duct lesion type 3 , which in fact appears to be an interface hepatitis related liver-cell lesion. However, in the current biopsy specimen the cytoplasmic cytokeratin 19 as well as mucin and MUC5AC expression in some of the lesions indicates that the cells are more likely of bile ductular origin. Normal small bile ducts and ductules do not secrete mucin or express mucin core proteins (MUCs). Overexpression in the intrahepatic biliary tree and cultured biliary epithelial cells has been shown to be due to e.g., bacterial infection, presumably via lipopolysaccharide induced synthesis of tumour necrosis factor-α and activation of protein kinase C .
Among children with AISC, almost half have concomitant IBD, but CD is relatively rarely specified. In the study of Gregorio et al.  one child with AIH and probable small duct PSC associated with CD is mentioned. Feldstein et al.  described that out of 40 children with PSC, 14 had overlap with AIH, and 8 out of 52 had CD, but did not specify how many had concomitant AISC and CD. Miloh et al. described 12 children with AISC of which 6 had concomitant IBD but it is not clear how many had CD. In none of this cases were GEL like changes described.
No guidelines for the treatment of AISC (isolated or in combination with IBD) have been established, but a combination of UDCA and immunosuppressive treatment has been recommended . The present trend is to target the treatment to one of the diseases and then to adjust it depending on symptoms and side effects . Patients with large duct AISC have significantly worse prognosis than patients with AIH or large duct PSC [2, 5, 9]. Small duct PSC show a more protracted course than large duct PSC [7, 12], but the number of patients with small duct AISC is too small for a prognostic evaluation. However, one may speculate that the AIP type 2-like liver damage signifies a better response to steroids compared to PSC in general.