In IgG4-SC, fibroinflammatory involvement is observed mainly in the submucosa of the bile duct wall, whereas the epithelium of the bile duct is intact [5]. PSC should be ruled out if inflammation is observed, particularly in the epithelium of the bile duct wall. The characteristic pathological findings of IgG4-SC were first reported under the name “lymphoplasmacytic sclerosing pancreatitis with cholangitis” [8]. Abraham et al. [9] reported frequent fibroinflammatory involvement of the gallbladder and common bile duct in patients with lymphoplasmacytic sclerosing pancreatitis. Zen et al. [5] revealed that the bile duct wall in IgG4-SC had pathological features similar to those of AIP, displaying dense infiltrations of lymphocytes and IgG4-positive plasma cells, with extensive fibrosis and obliterative phlebitis. They classified IgG4-SC into 6 categories according to the extent of inflammation and association with an inflammatory pseudotumor. IgG4-positive plasma cells are sparse in the affected bile ducts in PSC, and the luminal side of the bile ducts, including the lining biliary epithelial cells, is preferentially affected compared with IgG4-SC. In PSC, the fibrosis is denser and older, whereas in IgG4-SC, the entire bile duct wall and periductal tissue are affected. However, a recent study by Zhang et al. [10] revealed that 23 (23 %) of 98 liver explants with PSC had periductal infiltration with abundant IgG4-positive plasma cells [10/high-power field (HPF)] in the hilar area.

Serum IgG4 level has been reported to be a useful marker for discriminating AIP from other pancreatic diseases [11]. A cutoff IgG4 level of 135 mg/dL has been widely used as part of the diagnostic criteria for AIP. However, twice the upper limit of the normal value has also been recommended to discriminate AIP from PCa. In the International Consensus Diagnostic Criteria for AIP, once or twice the upper limit of the normal value is included in levels 1 and 2 diagnoses, respectively [12].

Only a few reports have been published concerning the cutoff IgG4 level in the diagnosis of IgG4-SC. We published for the first time, the diagnostic criteria for IgG4-SC based on a comparative study [2]. The cutoff IgG4 level of 135 mg/dL is useful in discriminating IgG4-SC from PCa and PSC. However, the level displayed lower specificity in discriminating IgG4-SC from CC. Oseini et al. [13] evaluated the utility of serum IgG4 level in distinguishing IgG4-SC from CC. They reported that among their 126 patients with CC, 17 (13.5 %) had elevated (>140 mg/dL) and 4 (3.2 %) had a >2-fold increase (>280 mg/dL) in serum IgG4 levels. PSC was present in 31/126 CC patients, of whom 7 (22.6 %) had an elevated serum IgG4 level. The authors concluded that some patients with CC, particularly PSC, had elevated serum IgG4 levels and diagnosis using a twofold higher cutoff serum IgG4 level may not reliably distinguish IgG4-SC from CC. However, a cutoff level fourfold higher than the upper limit of the normal value had 100 % specificity for IgG4-SC.

We recently established a cutoff serum IgG4 level to differentiate IgG4-SC from the 3 other diseases (type 1 IgG4-SC vs PCa, type 2 IgG4-SC vs PSC, and type 3 IgG4-SC vs CC) using serum IgG4 levels measured in 9 Japanese high-volume centers [7]. The cutoff obtained from receiver operating characteristic (ROC) curves displayed similar sensitivity and specificity to those of 135 mg/dL when all the IgG4-SC cases and controls were compared. However, a new cutoff value was established when IgG4-SC subgroups and controls were compared. A cutoff level of 182 mg/dL can increase the specificity to 96.6 % (a 4.7 % increase) for distinguishing types 3 and 4 IgG4-SC from CC. A cutoff level of 207 mg/dL might be useful for completely distinguishing types 3 and 4 IgG4-SC from CC.

Mendes et al. [14] measured the serum IgG4 level in 127 patients with PSC and found that it was elevated in 12 patients (9 %). The patients with elevated IgG4 levels had higher levels of total bilirubin and alkaline phosphatase, higher PSC Mayo risk scores, and lower incidence of inflammatory bowel disease. It is important to note that the time to liver transplantation was shorter in patients with elevated IgG4 levels (1.7 vs 6.5 years, P = 0.0009). As only one of the patients in their series had an abnormal pancreatogram, the documented cases appeared to conform to the diagnosis of IgG4-SC. Therefore, clinical trials in which patients with PSC are evaluated for IgG4 and patients presenting elevated levels are treated with corticosteroids should be considered.

Only 3 sets of diagnostic criteria for IgG4-SC have been proposed [2, 15, 16]. AIP should be clinically discriminated only from PCa. However, IgG4-SC should be discriminated from all of the 3 intractable diseases (PCa, PSC, and CC). Therefore, we should take into account these 3 intractable diseases for differential diagnosis in the diagnostic criteria for IgG4-SC.

We summarize three diagnostic criteria for the abovementioned IgG4-SC. The Research Committee of IgG4-related Diseases and the Research Committee of Intractable Diseases of Liver and Biliary Tract, in association with the Ministry of Health, Labour, and Welfare, Japan, and the Japan Biliary Association, proposed clinical diagnostic criteria for IgG4-SC in 2012 [15] (Table 15.1). The diagnosis of IgG4-SC is based on the combination of the following 4 criteria: (1) characteristic biliary imaging findings, (2) elevation of serum IgG4 concentrations, (3) the coexistence of IgG4-related diseases except those of the biliary tract, and (4) characteristic histopathological features. Furthermore, the effectiveness of steroid therapy is an optional extra diagnostic criterion to confirm accurate diagnosis of IgG4-SC.

The HISORt criteria for the diagnosis of IgG4-SC [3] are based on the characteristic features of IgG4-SC on histological, imaging, and serological examination; other organ involvement; and response to steroid therapy, which parallel the HISORt criteria established for AIP [16].

Our original diagnostic criteria provide a concrete diagnostic algorithm for IgG4-SC (Fig. 15.2) [2]. Association with AIP and other organ involvements are common useful diagnostic parameters in all three IgG4-SC types. Characteristic cholangiogram, liver biopsy, and exclusion of IBD are useful parameters in type 2 IgG4-SC. IDUS findings, exclusion of malignancy by bile duct biopsy, and a serum IgG4 cutoff level of 207 mg/dL were useful parameters in type 3 and 4 IgG4-SC. Although, generally, diagnosis should be made before starting therapy, a steroid trial is needed in some cases.

IgG4-SC displays various cholangiographic features similar to those of PCa, PSC, and CC. The differential diagnosis based on cholangiographic classification is sufficient in clinical practice because the useful modalities depend on the cholangiographic types (Fig. 15.3) [18].

Type 1 IgG4-SC should be differentiated from chronic pancreatitis, PCa, and CC. The modalities useful for differential diagnosis are IDUS [6], endoscopic ultrasound-guided fine-needle aspiration for the pancreas [19], and cytological examination and/or biopsy of the bile duct [6].

Type 2 IgG4-SC should be differentiated from PSC and secondary sclerosing cholangitis (SSC). After ruling out SSC, differential diagnosis between Type 2 IgG4-SC and PSC is necessary [20]. The modalities useful for differential diagnosis are cholangiography [17], evaluations for associated IBD [21, 22], and liver biopsy [23, 24]. Our discriminant analysis formula for cholangiographic features, including age, was able to efficiently discriminate type 2 IgG4-SC from PSC [17]. Band-like strictures, a beaded or “pruned tree” appearance, and diverticulum-like outpouching are significantly more frequent in PSC cases. In contrast, segmental strictures, long strictures with prestenotic dilation, and strictures of the lower common bile duct are significantly more common in IgG4-SC. These differences are illustrated in Fig. 15.4. The characteristic cholangiographic features reflect the underlying pathological processes involved in each condition. In PSC, obliterative fibrosis is the main cause of biliary stenosis, creating short strictures. In contrast, in IgG4-SC, severe lymphoplasmacyte infiltration into bile ducts in the long region is the main cause of biliary stenosis, resulting in long strictures.

Associated ulcerative colitis is suggestive of PSC. IBD is present in only 0–6 % of patients with IgG4-SC [3, 25, 26]. IBD is usually not a feature associated with type 1 AIP, unlike the frequent association of IBD with type 2 AIP [12]. IBD associated with PSC showed characteristic findings different from ordinary IBD [21, 22]. Backwash ileitis, rectal sparing, and low disease activity appear to be features that characterize IBD when it is associated with PSC [21, 22].

The histological features of IgG4-SC on liver biopsy are distinctive and, in conjunction with IgG4 immunohistochemical staining, help distinguish IgG4-SC from PSC [23, 24]. We have already reported that liver needle biopsy is especially useful for distinguishing IgG4-SC from PSC in patients with cholangiographically evident intrahepatic biliary strictures [24]. Four (57 %) of 7 patients with type 2 IgG4-SC presented infiltration ≥10 IgG4 + plasma cells per HPF in liver biopsy samples, whereas none of the PSC patients presented this feature.