AIP Type 1
AIP Type 2
Zamboni et al. 
van Buuren et al. 
Kleeff et al. 
Church et al. 
The United Kingdom
Detlefsen et al. 
de Castro et al. 
Frulloni et al. 
Maire et al. 
Czakó et al. 
Germany, Italy, Belgium, Denmark, Austria, Switzerland, Croatia
In most series, the diagnosis was based on the Mayo Clinic’s HISORt criteria [15, 16] or histological features of resection specimens. The overall gender distribution shows a male predominance of 65 % (50–100 %), and the median age was 51 years, thus lying within the margins of large international multicenter studies [14, 17]. The average age and percentage of male patients with proven LPSP/AIP type 1, however, were higher in these international cohorts (61 years; 74–77 % male), suggesting that the reported European cases include a significant proportion of patients with AIP type 2 that were not identified as belonging to that subtype. Two more recent reports from France and Hungary consistently report a proportion of AIP type 2 of greater than 33 % among their study population [18, 19], and the percentage of GEL-positive AIP was reported to be as high as 38.5 % in another study using core biopsy specimens . Taken together, these data suggest that the prevalence of AIP type 2 in Europe is higher than the international average, and the published international surveys underestimate the importance of AIP type 2. Although the quality of the summarized European studies is variable and none of them uses the International Consensus Diagnostic Criteria introduced by the IAP in 2011 , the results warrant caution when dealing with suspected cases of AIP in clinical practice as patients from Europe could be younger and the female proportion higher than previously assumed.
Other features of AIP among European cohorts seem to be in keeping with what is published in international surveys. The radiological appearance varies widely with the presence of a pancreatic mass in 27–96 % and diffuses swelling in 23–72 % of the cases. Obstructive jaundice is a common clinical feature that has been reported in 56 % of cases, and an association with inflammatory bowel disease was noted in 25 % of the patients in the studies that recorded this characteristic organ involvement.
The overall prevalence and incidence of AIP in Europe remains elusive. In two single-center retrospective analyses from the Netherlands, a total of 913 patients underwent a pancreaticoduodenectomy for presumed malignancy, of which 30 (3.3 %) turned out to have AIP [22, 23]. On one hand, this indicates that AIP remains a rare diagnosis despite the increasing attention among various clinical specialties but also suggests a rather high prevalence among certain subgroups of patients. In another single center’s experience from Italy, about 5 % of all chronic pancreatitis patients had AIP .
The correct diagnosis of AIP remains a clinical challenge. Especially the high prevalence of type 2 AIP in Europe and North America, with its lack of elevated serum IgG4 concentrations, a broader age spectrum, and a higher proportion of mass-forming lesions that mimic malignant disease, prevented a European adoption of the early diagnostic guidelines, which were based on the experience in Japan .
To our knowledge, the only set of diagnostic criteria for AIP from Europe was proposed by the group from Verona. In contrast to the Japanese and Asian consensus criteria (imaging of pancreatic parenchyma and duct appearance, elevated serum IgG4/autoantibodies, LPSP with IgG4-positive plasma cells) [26, 27] or the American HISORt criteria (histology showing LPSP/IDCP, suggestive imaging, elevated serum IgG4, other organ involvement, response to steroids) , the Italian criteria do not consider serological changes as diagnostic, reflecting the high prevalence of IgG4-negative AIP type 2 among their study cohort. They emphasize the importance of a steroid response as proof of diagnosis, when clinically justified and also suggested that of the following four criteria, three had to be fulfilled to allow the diagnosis of AIP :
Suggestive radiological features on CT or MRI (no ERCP required)
Association with autoimmune disease
Consistent histology (lymphoplasmacytic infiltration and GELs)
Response to steroid therapy (clinically and on imaging)
The diagnostic work-up that used these Italian criteria initially distinguished between diffuse “sausage-shaped” swelling of the pancreas and a focal, hypodense mass that mimics malignancy . In cases of diffuse swelling, acute pancreatitis was considered the principal differential diagnosis, which can be easily ruled out by the absence of elevated serum pancreatic enzyme activity, absence of pancreatic necrosis, and only mild abdominal symptoms. In combination with associated autoimmune disease and a response to steroids, the diagnosis can be made without the need of pancreatic biopsies according to the authors’ experience. In the case of a focal mass, the histological exclusion of malignancy by EUS-guided biopsy would be the first step. If the sample is suggestive for AIP, a steroid trail can complete the diagnosis.
The International Consensus Diagnostic Criteria, which were introduced in 2011, combine the different approaches and allow to establish the diagnosis of AIP in accordance to regional needs and preferences . For example, ductal imaging is no longer required to make the definitive diagnosis for AIP type 1 as diagnostic ERP is not commonly used in Western countries. In equivocal cases, it remains, however, permitted.