Importance of Abdominal Ultrasound in 17 Patients with Histologically Confirmed Autoimmune Pancreatitis (AIP)

Abstract

Purpose: Autoimmune pancreatitis (AIP) is an independent, underdiagnosed, rare form of chronic pancreatitis. The goal of this study is to document ultrasound findings in histologically confirmed AIP in order to determine the diagnostic value of ultrasound.

Materials and Methods: 6 of 74 DEGUM instructors for internal medicine (level 3 and 2) provided anonymized clinical and sonographic data from 17 patients with histologically confirmed AIP.

Results: 9/17 patients had diffuse AIP, 8/17 had focal AIP, 14/17 suffered from upper abdominal pain, 9/17 had jaundice, and 3/9 had an elevated IgG4 level. Ultrasound showed diffuse hypoechoic organ enlargement in 9/17 cases and a hypoechoic tumor with an unclear border in 8/17 cases. AIP was verified by ultrasound-guided percutaneous core biopsy in 14 cases, by biopsy of the bile duct in 1 case, and by surgical biopsy in 2 cases. Involvement of the hepatobiliary system was present in 7/17 patients and autoimmune cholangitis was verified in 5 cases. All patients experienced remission after immunosuppressive treatment. The pancreatic duct had a normal width in 11 cases, was dilated in 5 cases, and was stenosed over a long stretch in 3 cases. Contrast-enhanced ultrasound did not show uniform signal increase but also no significantly reduced or absent perfusion.

Conclusion: Ultrasound can be diagnostically useful if the clinical manifestations of AIP are known. While the diffuse form allows an ad-hoc suspected diagnosis, the focal form can only be suspected in the case of additional extrapancreatic involvement. Contrast-enhanced ultrasound (CEUS) contributes greatly to the differentiation from ductal adenocarcinoma in the case of focal AIP.

Zusammenfassung

Ziel: Die Autoimmunpankreatitis (AIP) ist eine eigenständige, unterdiagnostizierte, seltene Form der chronischen Pankreatitis. Ziel dieser Studie ist es die Ultraschallbefunde bei histologisch verifizierter AIP zu erfassen, um den diagnostischen Stellenwert der Sonografie zu ermitteln.

Material und Methoden: Von 74 DEGUM-Ausbildern der Sektion Innere Medizin (Stufe III und II) stellten 6 Kollegen anonymisierte klinische und sonografische Daten von 17 Patienten mit histologisch gesicherter AIP zur Verfügung.

Ergebnisse: Von 17 Patienten wiesen 9 eine diffuse und 8 eine fokale AIP auf, 14/17 litten unter Oberbauchschmerzen und 9/17 waren ikterisch, IgG-4 war bei 3/9 erhöht. Sonografisch zeigten 9/17 eine diffuse echoarme Organschwellung und 8/17 einen unscharf begrenzten echoarmen „Tumor“. 14-mal wurde die AIP durch eine sonografisch gesteuerte perkutane Stanzbiopsie, einmal durch eine Biopsie des Gallengangs und 2 mal durch operative Gewebeentnahme verifiziert. Eine Mitbeteiligung des hepatobiliären Systems lag bei 7/17 Patienten vor, 5-mal wurde eine Autoimmuncholangitis verifiziert. Alle Patienten zeigten nach immunsuppressiver Therapie eine Remission. Der Pankreasgang wurde 11-mal normal weit, 5-mal dilatiert und 3-mal längerstreckig stenosiert beschrieben. Die Kontrastmittelsonografie zeigte keinen einheitlichen Signalanstieg, jedoch in keinem Fall eine stark abgeschwächte oder fehlende Perfusion.

Schlussfolgerung: In Kenntnis der klinischen Manifestationsformen der AIP kann die Sonografie diagnostisch hilfreich sein. Während die diffuse Form eine Ad-hoc-Verdachtsdiagnose erlaubt, ist die fokale Form nur bei zusätzlicher extrapankreatischer Beteiligung zu vermuten. Die Kontrastmittelsonografie (CEUS) trägt bei fokaler AIP wesentlich zur Unterscheidung vom duktalen Adenokarzinom bei.

Key words

autoimmune pancreatitis - pancreas - bile ducts - ultrasound - CEUS

Introduction

The term “autoimmune pancreatitis” (AIP) as coined in 1995 by Yoshida has become established and defines a new syndrome of clinical, laboratory-based, and histological findings. Case reports and large patient series are available primarily from Japan. Up to 10 % of cases of chronic pancreatitis are autoimmune pancreatitis [1] [2] [3]. The clinical picture has been divided into two types that can be clinically and histologically differentiated: type 1 is referred to as lymphoplasmacytic sclerosing pancreatitis (LPSP), type 2 as idiopathic duct-centric pancreatitis (IDCP) [4]. While AIP is usually associated with an increased IgG4 level (corresponding to type 1) in Asia, this value is only increased in a small portion of cases in Central Europe [5].

The diagnosis of AIP is complex and always requires an imaging method in addition to laboratory parameters. Imaging and an increased IgG4 level are typically sufficient for diagnosing type 1 while histological confirmation is usually necessary in the case of type 2.

Although most patients are presumably primarily examined via transabdominal ultrasound, a PubMed search yielded minimal literature regarding this [6]. However, the importance of EUS was shown in a series of studies [7] [8] [9]. Our goal is to examine the importance of abdominal ultrasound for AIP.

Materials and Methods

In 2009, 74 DEGUM instructors (DEGUM level 3 and 2) for internal medicine were asked to document anonymized clinical and sonographic findings including disease course in the case of histologically confirmed AIP on a data collection form.

In addition to anamnestic data, such as previous diseases and disease symptoms, diagnostic measures, laboratory tests, including IgG4, and immunoserological findings were recorded. Concomitant autoimmune diseases, the response of AIP to an immunosuppressive treatment, and the disease course were documented.
The ultrasound findings for the pancreas were documented in detail. Location, extent, echo pattern, duct changes, as well as additional findings in the liver, bile ducts, and gall bladder were documented. The findings of contrast-enhanced ultrasound (CEUS) were described according to the usual criteria (arterial/parenchymatous/venous phase and perfusion pattern). 14 of 74 DEGUM instructors responded to the survey, 6 colleagues stated that they did not have any histologically confirmed patients, one seminar leader did not provide the data because it was included in another study, and 2 instructors declined for other reasons. 5 DEGUM instructors (including 4 DEGUM level III instructors) participated in the study with 14 patients. With 3 own cases, data from 17 patients were available for evaluation. 

The cases were diagnosed between 1999 and 2009.

Results

Demographic data

The average age at the time of diagnosis was 50.9 years (28 – 77 years). On average, women and men were of an equal age. 10 patients were male and 7 were female. A diffuse form of AIP was seen in 9 patients, while 8 patients had a focal “tumor-like” variant, 2 of which were operated on. Patients with focal AIP were on average older (58 years: 29 – 77 years) than those with diffuse AIP (45 years: 28 – 69 years). All patients were undergoing inpatient treatment. The disease courses were followed up for 3 – 36 months (median 14.5 months). Diagnosis was made in 10 patients within 6 months after the occurrence of the first symptoms, in 4 patients after 6 – 12 months, and in 3 cases the first symptoms had occurred more than a year ago.

Symptoms

14 of 17 patients complained of mild upper abdominal pain. 9 of 17 patients had jaundice. 6 of 9 patients with jaundice had a focal form of AIP. Jaundice was seen in 3 of 9 cases of diffuse AIP. Relevant weight loss was seen 6 times, fatigue was seen in 3 cases. Pruritus, fatty stool, acholic stool, vomiting, and lack of appetite were seen in one case each.

Primary suspected clinical diagnosis

The suspected diagnosis of AIP was stated only twice immediately after initial examination and abdominal ultrasound. Five masses were interpreted as pancreatic carcinoma and one as neuroendocrine pancreatic tumor. Three acute and three chronic cases of pancreatitis, one case of primary sclerosing cholangitis (PSC), one case of gastritis, and one case of sarcoidosis were also documented.

Association with autoimmune diseases

In 4 of 17 cases there was a concomitant “extrapancreatic” autoimmune disease (Hashimoto's thyroiditis 2, ulcerative colitis 1, Sjögren Syndrome 1). In five patients autoimmune cholangitis was histologically confirmed (n = 1) or assumed on the basis of immunoserology (n = 4).

Laboratory findings

[Table 1] shows the results of the most important laboratory findings. The basic laboratory values reported incomplete in approximately one-third of the cases. The lipase level was triple the norm in only 5 of 13 patients and the values were between 1000 and 2000 U/I in 2 cases. The IgG4 level was determined in 9 of 17 patients with 6 findings being normal. Among 3 positive findings, the value was less than double the normal value in 1 case, triple the normal value in 1 case, and 10 times the normal value in 1 case. The frequently missing determination of numerous autoantibodies is due to the limited knowledge at the time of the study.

Table 1 Diagnosis-relevant laboratory findings in 17 patients with AIP.

parameter	normal values	result
CRP	< 5 mg/l	< 5	5 – 50	> 50	no data
(n) number of patients		8	5	2	2
lipase	13 – 60 U/l	< 60	60 – 180	> 240
(n)		7	2	5	4
bilirubin	< 1.2 mg/dl	< 1.2	1.2 – 4.9	> 5
(n)		8	2	5	2
IgG4	15 – 128 mg/dl	< 128	129 – 256	> 256
(n)		6	1	2	8

Abdominal ultrasound

The findings of percutaneous abdominal ultrasound are shown separately for diffuse and focal AIP in [Table 2].

Table 2 B-mode ultrasound in 17 patients with autoimmune pancreatitis.

ultrasound		focal form n = 8	diffuse form n = 9
pancreas	location	head/body/tail 6  /  1  /  1	total organ 9
	echo pattern	hypoechoic/isoechogenic/echogenic 8  /  0  /  0	hypoechoic/isoechogenic/echogenic 9  /  0  /  0
	calcifications	0	0
	pancreatic duct	normal/dilated/compressed 5  / 3  /  2	normal/dilated/compressed 6 / 2  /  1
bile duct/gall bladder (wall thickening, +/- dilated bile ducts)	extrahepatic	2	4
	intrahepatic	0	0
	gall bladder	0	1
complications	pseudocysts Thrombosis (portal-, splenic vein)	0 0	0 0

In 9 patients with diffuse AIP the pancreas was described as hypoechoic, diffusely enlarged, and sausage-shaped. The parenchyma was inhomogeneous in two cases. There were no calcifications or cystic lesions. The pancreatic duct was described as dilated in two cases and as compressed in one case. The extrahepatic bile ducts were dilated and/or had focal wall thickening in 4 patients. Consecutive dilated and stenosed bile duct segments were additionally seen in one case. In one case the wall of the stone-free gallbladder had irregular hypoechoic thickening. CEUS was performed in 7 of 9 cases with diffuse AIP ([Table 3]). In all cases 1.0 to 2.2 ml SonoVue^® were administered depending on the equipment. The signal behavior was described as minimally reduced in 4 cases, as enhanced in 2 cases, and as normal with respect to the contrast agent behavior of a healthy pancreas in 1 case. The contrast agent distribution was described as irregular in 3 cases. Spleen or portal vein thrombosis was not present in any case. The contrast agent quantity was administered on an individual basis as a function of the experience of the examiner and the ultrasound equipment being used. Since no healthy parenchyma is available for comparison in diffuse AIP, the evaluation was performed subjectively in comparison to the contrast agent behavior of a normal pancreas. Guidelines for uniform evaluation of CEUS of the pancreas were not available at the time of the study [10].

Table 3 Contrast-enhanced ultrasound of the pancreas in 13 patients with autoimmune pancreatitis: signal acquisition compared to normal finding.

contrast-enhanced ultrasound (CEUS)	focal form n = 6	diffuse form n = 7
hyperperfused	1	2
isoperfused	4	1
moderately hypoperfused	1	4
avascular	0	0

In 8 patients with focal “tumor-like” AIP ([Table 2]), the lesion was detected in the head of the pancreas in 6 cases, at the head-body junction in one case, and in the medial tail in one case. The “tumor” could be effectively delimited from the healthy parenchyma in all cases. The maximum mass diameter ranged from 20 to 55 mm. The parenchyma of the tumor-like lesion was described as hypoechoic in all cases. The neighboring healthy pancreatic parenchyma was echogenic in 4 cases and hypoechoic in one case. The pancreatic duct was prestenotically uniformly slightly dilated in 3 cases and compressed in 2 cases. The bile ducts were dilated in two cases and also had narrowed segments in one case. Multiple enlarged peripancreatic lymph nodes were described in 2 cases. In 6 of 8 cases, CEUS with SonoVue^® (1.0 – 1.8 ml i. v. as bolus) was performed in addition ([Table 3]). In 4 cases contrast agent uptake occurred as in a healthy pancreatic parenchyma. In the tumor-like lesion increased perfusion and reduced perfusion were seen in one case each compared to the neighboring parenchyma. The signal behavior of the neighboring healthy parenchyma was described as normal in 5 cases and there was no data in 3 cases. Portal vein or splenic vein thrombosis was not observed.

[Fig. 1], [2], [3] show examples of sonographic findings in diffuse and focal AIP.

Fig. 1 Diffuse type of autoimmune pancreatitis (from Aug. 2005 to Nov. 2007); 47-year-old, male patient with recurrent moderate upper abdominal pain for one year. a Cross-section of the upper abdomen: hypoechoic pancreas with noticeable general organ enlargement. b Thickened gall bladder wall, the head of the pancreas (P) is visualized mediocaudal to the gall bladder and the common bile duct (CBD) and the portal vein (PV) are seen dorsal to the gall bladder. c hypoechoic thickened CBD wall in terms of autoimmune cholangitis or PSC. d After 4 months of cortisone therapy, normal size of the pancreas and increased echogenicity were seen as a sign of successful treatment.

Abb. 1 Diffuse Form der Autoimmunpankreatitis (von Aug. 2005 bis Nov. 2007); 47-jähriger, männlicher Patient mit rezidivierenden mäßigen Oberbauchschmerzen seit einem Jahr. a Oberbauchquerschnitt: echoarmes Pankreas mit auffallender genereller Organvergrößerung. b Verdickte Gallenblasenwand, mediokaudal der Gallenblase sind der Pankreaskopf (P) und dorsal der D. hepatocholedochus (CBD) und die Portalvene (PV) dargestellt. c echoarm verdickte Wand des Ductus Choledochus passend zu Autoimmuncholangitis oder PSC. d Nach 4-monatiger Cortisonbehandlung normale Größe des Pankreas und Zunahme der Echogenität als Zeichen der erfolgreichen Therapie.

Fig. 2 Focal type of autoimmune pancreatitis (April 2009 to June 2010) 65-year-old male patient with discrete upper abdominal pain. Cross-sections of the upper abdomen: a hypoechoic mass in the body of the pancreas (arrows). b, c CEUS (SonoVue 1.0 ml in bolus) 18 and 21 seconds after injection: slightly reduced and discrete inhomogeneous contrast enhancement (representative oblique section through mass). d After 13 months of treatment: echogenic normal-sized pancreas, focal AIP no longer able to be detected.

Abb. 2 Fokale Form der Autoimmunpankreatitis (Verlauf April 2009 bis Juni 2010) 65-jähriger, männlicher Patient, mit diskreten Oberbauchschmerzen. Oberbauchquerschnitte: a echoarme Raumforderung im Pankreaskorpus (Pfeile). b, c KMUS (SonoVue 1,0 ml im Bolus) 18 und 21 Sekunden nach Injektion: gering verminderte und diskret inhomogene Kontrastmittelanreicherung (repräsentativer Schrägschnitt durch die Raumforderung). d Nach 13 Monaten Therapie: echogenes normal großes Pankreas, keine fokale AIP mehr nachweisbar.

Fig. 3 Focal type of autoimmune pancreatitis with involvement of the bile ducts. 79-year-old male patient with jaundice. a Cross-section of the upper abdomen: hypoechoic, small tumor of the head of the pancreas with unclear border. b Image section of the oblique section of the upper abdomen on the right: irregular, wall-thickened common bile duct (distance markers) with filiform, stenosed lumen and accentuated gall bladder wall. The combination of a circumscribed pancreatic mass with wall-thickened bile ducts is indicative for the diagnosis of AIP. c, d CEUS of the head of the pancreas (SonoVue 0.8 ml as Bolus). c After 10 seconds: still no enhancement of the pancreas. d Isoechogenic enhancement in the arterial phase within 8 seconds (thick arrow: focal AIP, thin arrow: body of the pancreas).

Abb. 3 Fokale Form der Autoimmunpankreatitis mit Gallengangsbeteiligung. 79-jähriger, männlicher Patient mit Ikterus. a Oberbauchquerschnitt: echoarmer, kleiner, unscharf begrenzter, Pankreaskopftumor. b Bildausschnitt Oberbauchschrägschnitt rechts: irregulär, wandverdickter Ductus hepatocholedochus (Distanzmarken) mit filiform, stenosiertem Lumen und akzentuierter Gallenblasenblasenwand. Die Kombination einer zirkumskripten Pankreasraumforderung mit wandverdickten Gallenwegen ist wegweisend für die Diagnose einer AIP. c, d KMUS des Pankreaskopfs (SonoVue 0,8 ml als Bolus). c nach 10 Sekunden: noch kein Enhancement des Pankreas. d innerhalb von 8 Sekunden isoechogenes Enhancement in der arteriellen Phase (dicker Pfeil: fokale AIP, dünner Pfeil: Pankreaskorpus).

ERCP, CT and MRI (MRCP)

The ERCP findings are shown in [Table 4]. ERCP was performed in 14 of 17 patients. The pancreatic duct was changed in 9 patients. In the 8 examined patients with a diffuse form of AIP, the duct was irregularly dilated in 4 cases and uniformly narrowed over a long stretch in 3 cases. The 6 examined focal AIP cases had irregular dilation in 2 cases. The bile ducts showed changes in 12 patients. In the case of the diffuse form, the extrahepatic bile duct was stenosed in 7 of the 8 examined cases, and primary sclerosing cholangitis (PSC) was seen in 4 cases. In the focal AIP patients extrahepatic stenosis was seen in 5 of 6 examined patients and PSC was seen in one case.

Table 4 ERCP in 14 patients with autoimmune pancreatitis.

ERCP	focal form n = 6	diffuse form n = 8
pancreatic duct	normal/dilated/narrowed 3  /  2  /  0	normal/dilated/narrowed 1  /  4  /  3
bile ducts
extrahepatic	normal/stenoses 1  /  5	normal/stenosis 1  /  7
intrahepatic	normal/presentation of PSC 2  /  1	normal/presentation of PSC 2  /  4

Computed tomography was performed in 10 patients, and MRI was performed in only 4 patients, as MRCP in 2 of those cases. Three patients with diffuse AIP were evaluated as normal on CT, one case was interpreted as chronic pancreatitis, and one case as acute necrotizing pancreatitis. Three patients with focal AIP were diagnosed with a pancreatic malignant tumor.

MRI showed one case of a hypervascularized tumor of the head of the pancreas (without definitive diagnosis) and one case of diffuse chronic pancreatitis. On MRCP one case of diffuse AIP was detected and one case of a focal form was characterized as a tumor.

Diagnostic verification

In 14 of 17 patients percutaneous ultrasound-guided fine-needle biopsy of the pancreas resulted in histological confirmation of the diagnosis. Two patients underwent surgery under the assumption of pancreatic carcinoma and in one case histological confirmation of the diagnosis was achieved via an endoscopic bile duct biopsy with verification of autoimmune cholangitis. Periductal plasmacellular infiltrations were visualized in 11 cases and fibrotic changes in 14 cases. Eosiniphilic granulocytic infiltrates were found in one case in the pancreas and once in a bile duct biopsy.

Treatment and course

15 of 17 AIP patients were treated conservatively after standardized steroid treatment for AIP [11] with prednisolone up to a maximum of 1 mg/kg/body weight. In 10 patients (7 cases of diffuse AIP, 3 cases of focal AIP), complete remission with retrogression of the primary sonographic finding occurred a few weeks after the start of treatment. Four patients were additionally treated with azathioprine at the start. One patient received azathioprine and ursodeoxycholic acid (1 g/d) as a result of incomplete remission under monotherapy with cortisone after 3 years and remained symptom-free in the observation period of 2 years. The azathioprine dose was 2 – 3 mg/kg/d. Temporary stents were used in 4 patients for a focal AIP in two cases and a diffuse AIP in two cases.

Two patients with focal AIP underwent surgery due to suspicion of pancreatic carcinoma.

In the case of an average monitoring period of 14.5 months (3 – 36 months), none of the 17 patients developed endocrine or exocrine pancreatic insufficiency. Pseudocysts or pancreatic calcifications were not able to be detected with ultrasound in the course of the disease. One patient developed pancreatic atrophy within 3 years. The initial stenoses in the bile ducts or pancreatic duct receded in all cases.

Discussion

AIP is a rare disease in Europe and is not sufficiently included in differential diagnosis considerations. The clinical picture of AIP has since been well defined and standards for the diagnosis of type 1 and type 2 AIP have been defined. Type 1 AIP represents pancreatic involvement in an IgG4-associated systemic disease. Extrapancreatic organ manifestation is common in this type. Type 2 which is more common in Europe is a pancreatic disease which is associated with a chronic inflammatory intestinal disease in one-third of cases [4]. The most important laboratory parameter, the IgG4 level, is usually missing in type 2. Both AIP types occur with diffuse or focal involvement of the pancreas. Differential diagnosis from pancreatic carcinoma and also from alcohol-induced chronic pancreatitis [12] [13] is difficult.

A PubMed search yielded only one Chinese publication [6] about AIP and abdominal ultrasound although ultrasound is certainly used in most AIP patients as the first imaging method. Our survey was aimed at the possible diagnostic value of abdominal ultrasound in histologically confirmed AIP. The histology of the 17 cases diagnosed between 1999 and 2009 was based primarily on the lymphocytic infiltration and fibrosis of the pancreas. The histological standards for AIP were first defined in 2011 [14]. The collective of 17 patients corresponds well with the literature with respect to age and gender distribution and clinical and laboratory findings [4] and is sufficiently representative with respect to the evaluation of sonographic findings. The immunoserological tests in our collective were inconsistent and incomplete from today's standpoint with respect to the diagnostic period starting in 1999. The IgG4 level was determined only in 9 patients. The IgG4 level was increased in 1 of 5 patients with diffuse AIP and in 2 of 4 with focal AIP.
The operation frequency was very low at 2 of 17 cases or 2 of 8 patients with “tumor-like” AIP. This is probably a consequence of the diagnosis primarily confirmed via percutaneous tru-cut core biopsy (14/17).

Diffuse and focal involvement of the pancreas in AIP can be differentiated on ultrasound. If the non-specific symptoms of AIP are known and clinical and sonographic findings of typical acute pancreatitis are absent, AIP should be assumed in the case of a hypoechoic, diffusely enlarged, sausage-shaped pancreas. This applied to all 9 cases of “diffuse” AIP. The focal form of AIP is usually associated with a hypoechoic “tumor-like” mass in the head of the pancreas and jaundice. The visualization of a long stretch of compressed pancreatic duct – a known sign of AIP – was not queried in a targeted manner but was documented in 3 cases. Pronounced changes in the pancreatic duct (dilation > 3 mm and caliber fluctuations) as well as other signs of chronic pancreatitis were absent throughout.

Thickening of the wall of the extrahepatic and intrahepatic bile ducts possibly with intermediary dilated segments was detected in 7 of 17 patients and indicates concomitant autoimmune cholangitis or primary sclerosing cholangitis (PSC). These findings are inconsistent with a pancreatic carcinoma and serve as an important indication of AIP. Supplementary immunoserological tests are essential and fine-needle biopsy confirms the diagnosis if it is not possible to wait for response to cortisone therapy [14].

The examined patients showed an enhanced, normal or discretely reduced partially inhomogeneous signal increase compared to the surrounding pancreatic parenchyma during the parenchymal phase in the lesion on CEUS. It was not significantly weakened or absent as in ductal carcinoma in any case [15]. However, it must be pointed out that neuroendocrine pancreatic tumors are also characterized by a strong signal. D`Onofrio et al. [16] were able to use CEUS to differentiate a pancreatic tumor in 173 patients and an inflammatory pancreatic tumor in 35 patients (AIP 15, alcoholic chronic pancreatitis 19, hereditary pancreatitis 1) from a ductal adenocarcinoma with a sensitivity of 88.6 % and a specificity of 97.8 %. Even if no definitive statement about the reliability of CEUS for differentiating focal AIP from pancreatic carcinoma is currently possible, the method is already important [15] [16]. [Fig. 4] shows the importance of ultrasound in the diagnosis of AIP in the form of an algorithm.

Fig. 4 Diagnosis of AIP: Importance in sonography.

Abb. 4 Diagnostik der AIP: Stellenwert in der Sonografie.

An endosonographic study with a very small number of cases provides an optimistic report of the diagnostic differentiation between AIP, pancreatic carcinoma, and chronic pancreatitis via elastography [17]. It is too early for a definitive evaluation of this new ultrasound-guided technique in the case of AIP even though this study was included in the elastography guidelines [18].

Since MRI and EUS were not available to all participants, ERCP was preferably used as an additional imaging method in the present study. Of course, MRCP and possibly also EUS have a significantly higher value today. ERCP provides complete information about the ductal system while ultrasound visualizes the incomplete duct system, i. e., preferably the dilated segments. The lack of visualization of the entire length of the pancreatic duct from the head to the tail is a further weakness of conventional ultrasound. According to Kamisawa et al. [19], invasive ERCP is superior to noninvasive MRCP for detecting narrowed segments of the pancreatic duct in AIP. Without a doubt, MRCP and EUS are superior to conventional ultrasound in the visualization of small bile ducts. Fine-needle puncture is the decisive diagnostic measure in the case of uncertain findings [4].

In our patients CT and MRI findings coincided well with ultrasound as a whole. Sahani et al. [20] also saw a broad spectrum of findings with computed tomography similar to ultrasound in 15 AIP patients over a long course. Ductal strictures were first observed late in the course of the disease. In the case of autoimmune pancreatitis, the signal in MRI is reduced in the case of T1-weighted visualization and is slightly increased in the case of T2-weighted visualization. However, it is not possible to differentiate AIP from pancreatitis of another genesis [21].
Ultrasound is very suitable for the follow-up of AIP and additional imaging is usually unnecessary.

Conclusion

The following sonographic constellations are suspicious for AIP:

1. In the case of diffuse, hypoechoic, sausage-like enlargement of the pancreas, it is highly likely that AIP is present if the classic signs of acute pancreatitis are absent.

2. The focal form of AIP is difficult to differentiate from pancreatic carcinoma on B-mode ultrasound. Intrahepatic and extrahepatic bile ducts with wall thickening as also seen in PSC should be cause for suspicion of AIP.

3. The now known involvement of further organs must be taken into consideration [22] [23].

• Literatur

• 1 Tanaka S, Kobayashi T, Nakanishi K et al. Corticosteroid-responsive diabetes mellitus associated with autoimmune pankreatitis. Lancet 2000; 356: 910-911
• 2 Horiuchi A et al. ERCP features in 27 patients with autoimmune pancreatitis. Gastrointest Endosc 2002; 55: 494-499
• 3 Schneider A, Löhr JM. Autoimmune Pankreatitis. Internist 2009; 50: 318-330
• 4 Kamisawa T, Chari ST, Giday SA et al. Clinical profile of autoimmune pancreatitis and its histological subtypes: an international multicenter survey. Pancreas 2011; 40: 809-814
• 5 Ringel J, Lerch MM, Mayerle J. Pankreaserkrankungen – Update 2011. Dtsch Med Wochenschr 2011; 136: 2037-2039
• 6 Su LY, Gao Y, Xu G et al. Retrospective analysis of ultrasonography for autoimmune pancreatitis. Zhonghua Yi Xue Za Zhi 2012 92: 2649-2651
• 7 Hocke M, Dietrich CF. Kontrastverstärkte Endosonographie zur Diagnostik von Pankreaserkrankungen. Dtsch Med Wochenschr 2013; 138: 732-734
• 8 Ignee A, Schüssler G. Elastografie zur Diagnostik der Autoimmunpankreatitis. Ultraschall in Med 2009; 30-V10–03
• 9 De Lisi S et al. Endoscopic Ultrasonography Findings in Autoimmune Pancreatitis: Be Aware oft he Ambiguous Features and Look fort he Pivotal Ones. J Pancreas (Online) 2010; 11: 78-84
• 10 Claudon M, Cosgrove D, Albrecht T et al. Guidelines and Good Clinical Practice Recommendations for Contrast Enhanced Ultrasound (CEUS) – Update 2008. Ultraschall in Med 2008; 29: 28-45
• 11 Kamisawa T, Shimosegawa T, Okazaki K et al. Standardisierte Steroidbehandlung bei Autoimmunpankreatitis. GUT 2009; 58: 154-1507
• 12 Chari ST, Takahashi N, Levy MJ et al. A diagnostic strategy to distinguish autoimmune pancreatitis from pancreatitic cancer. Clin Gastroenterol Hepatol 2009; 7: 1097-1103
• 13 Weber SM, Cubukcu-Dimopulo O et al. Lymphoplasmatic sclerosing pancreatitis: inflammatory mimic of pancreatic carcinoma. J Gastrointest Surg 2003; 7: 129-137
• 14 Shimosegawa T, Chari ST, Frulloni L et al. International consensus diagnostic criteria for autoimmune pancreatitis: guidelines oft he International Association of Pancreatology. Pancreas 2011; 40: 352-358
• 15 Piscaglia F, Nolsoe C, Dietrich CF et al. The EFSUMB Guidelines and Recommendations on the Clinical Practice of Contrast Enhanced Ultrasound (CEUS): Update 2011 on non-hepatic applications. Published online 2011, Ultraschall in Med Georg Thieme Verlag KG Stuttgart. New York. ISSN: 0172-4614
• 16 D’Onofrio M, Zamboni G, Tognolini A et al. Mass-forming pancreatitis: Value of contrast-enhancend ultrasonography. World J Gastroenterol 2006; 12: 4181-4184
• 17 Dietrich CF, Hirche TO, Ott M et al. Real-time tissue elastography in the diagnosis of autoimmune pancreatitis. Endoscopy 2009; 41: 718-720
• 18 Cosgrove D, Piscaglia F, Bamber J et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography. Part 2: Clinical Applications. Ultraschall in Med 2013; 34: 238-253
• 19 Kamisawa T, Tu Y, Egawa N et al. Can MRCP replace ERCP for the diagnosis of autoimmune pancreatitis?. Abdom Imaging 2009; 34: 381-384
• 20 Sahani DV, Sinani N, Deshpande V et al. Autoimmune Pancreatitis: disease evolution, staging, response assessment and CT features that predict response to corticosteroid therapy. Radiology 2009; 250: 118-129
• 21 Irie H, Honda H, Baba S et al. Autoimmune pancreatitis: CT and MR characteristics. Am J Roentgenol 1998; 170: 1323-1327
• 22 Beyer G, Menzel J, Krüger PC et al. Autoimmunpankreatitis. Dtsch Med Wochenschr 2013; 138: 2359-2374
• 23 Kale KD, Takahashi N et al. Autoimmune Pankreatitis: Pancreatic and Extrapancreatic Imaging Findings. Am J Roentgenol 2009; 192: 431-437