IDUS IN THE DIAGNOSIS OF IPMT

Intraductal papillary mucinous tumor of the pancreas (IPMT) is considered as having low‐grade malignant potential, and malignancy is not infrequent. Therefore, accurate diagnosis is indispensable for appropriate patient care. IPMT is classified mainly into two categories based on the distribution of tumor, the main duct type and the branch duct type. In the diagnosis of IPMT, there are four important points: differentiation of IPMT from other pancreatic pathologies; differentiation of malignancy from benign lesions; evaluation of tumor extent along the MPD; and investigation of duct cell carcinoma coexistent with or derived from IPMT. IPMT should be distinguished from chronic pancreatitis, ductal adenocarcinoma, or cystic neoplasms. IPMT often requires pancreaticoduodenectomy for cure, but such invasive surgery should be avoided in patients with benign hyperplasia. Mural nodules in the MPD can be clearly visualized by intraductal ultrasonography (IDUS). Measurement of the height of the tumor mass by IDUS is useful in distinguishing benign from malignant IPMT. Intraductal spread of IPMT along the MPD is demonstrated by IDUS as irregular thickening of the MPD wall. To achieve a tumor‐free margin in surgery, evaluation by IDUS of the extent of the tumor along the MPD is important. The expected accuracy of IDUS in the diagnosis of invasive IPMT is reportedly over 90%.     

CLINICAL SIGNIFICANCE OF INTRADUCTAL ULTRASONOGRAPHY FOR DIAGNOSIS OF INTRADUCTAL PAPILLARY MUCINOUS NEOPLASM OF THE PANCREAS

Intraductal ultrasonography (IDUS) is useful for evaluating the horizontal spread along the main pancreatic duct in cases of intraductal papillary mucinous neoplasm (IPMN) of the pancreas, thus providing valuable information for the determination of the resection line at surgery. Differentiation between benign and malignant IPMN is also indispensable for management decisions. Measurement of the height of mural nodules by IDUS is expected to be useful for differential diagnosis of IPMN. Because IDUS cannot always demonstrate whole lesions, especially in branch duct IPMN, endoscopic ultrasonography plays a complementary role in such cases.     

MANAGEMENT OF INTRADUCTAL PAPILLARY MUCINOUS NEOPLASM OF THE PANCREAS

Intraductal papillary mucinous neoplasms of the pancreas (IPMNs) consist of main duct (MD) type and branch duct (BD) type. The authors describe their way of thinking regarding diagnostic modalities and management for BD type IPMNs. Endoscopic ultrasonography (EUS) and intraductal ultrasonography (IDUS) provide high resolution images of main and branch pancreatic ducts. The cases with nodules demonstrated by EUS and/or IDUS regardless of the size are the indication of operation. There were 235 cases with BD type IPMN who underwent EUS and IDUS between April 1991 and June 2005. A total of 94 patients underwent surgical resection and were histopathologically diagnosed (carcinoma, 10 cases; adenoma, 64 cases; hyperplasia, 20 cases). Diagnoses of 79 cases with nodules detected by EUS or IDUS preoperatively were 10 carcinomas, 61 adenomas and eight hyperplasias. Diagnoses of 15 cases without nodules but with symptoms were three adenomas, 12 hyperplasias and no carcinoma. The authors think that the combination of EUS and IDUS is the best way for diagnosing BD type IPMNs in the present state.     

LATERAL SPREAD ALONG THE MAIN PANCREATIC DUCT IN BRANCH‐DUCT INTRADUCTAL PAPILLARY‐MUCINOUS NEOPLASMS OF THE PANCREAS: USEFULNESS OF INTRADUCTAL ULTRASONOGRAPHY FOR ITS EVALUATION

Background: Intraductal papillary mucinous neoplasms (IPMN) of the pancreas tend to spread intraepithelially along the pancreatic duct wall. We evaluated histopathological intraductal lateral spread (LS) along the main pancreatic duct (MPD) from branch‐duct IPMN and investigated the usefulness of intraductal ultrasonography (IDUS) for its preoperative diagnosis. Patients and Methods: Twenty‐four patients with branch‐duct IPMN who had undergone preoperative IDUS and surgery were reviewed clinicopathologically. The prevalence and histological length of LS along the MPD from branch‐duct IPMN, characteristics of the patients with LS, and efficacy of LS assessment by IDUS were examined. Results: LS along the MPD was observed in 54% of the subjects. In the group of patients with LS, its mean length was 25.2 ± 16.8 mm (5–50 mm) and the diameter of the MPD was 6 mm or greater. Of the patients with LS, those in whom the length of LS along the MPD was longer than the diameter of the cystically dilated branch accounted for 30%. The diameter of the MPD in the group with LS was significantly greater than that in the group without LS (P = 0.03). The sensitivity, specificity, and overall accuracy of IDUS in the detection of LS were 92%, 91%, and 92%, respectively. Conclusion: LS along the MPD was detected in about half of the resected cases of branch‐duct IPMN. Preoperative transpapillary IDUS may be beneficial for the determination of the resection line, especially in those branch‐duct IPMN patients in whom the MPD is 6 mm or greater in diameter.     

WHAT KIND OF BENEFIT DO WE EXPECT FOR PERORAL PANCREATOSCOPY IN THE DIAGNOSIS OF INTRADUCTAL PAPILLARY MUCINOUS TUMOR OF THE PANCREAS?

Intraductal papillary mucinous tumor (IPMT) of the pancreas is characterized by slow growth and a relatively favorable prognosis, however, invasive cancer originating in an IPMT is associated with a poor prognosis. Although various parameters in imaging modalities have been advocated to differentiate between benign IPMN and malignant ones, it is not easy to obtain definite diagnosis based on these parameters. Peroral pancreatoscopy (POPS) allows a clear and direct visualization of the pancreatic duct, providing useful information regarding tumor nature in IPMT. The authors have studied the usefulness of POPS in the diagnosis of IPMT. Nevertheless, its usefulness is not necessarily widely accepted and the significance of POPS is still controversial. In this review, the authors intended to address the diagnostic value of POPS and to clarify its role in the diagnosis of IPMT. The authors think treatment of IPMT can be improved by introducing POPS because the determination of surgical procedure as well as the area of resection based on the preoperative diagnosis of the involvement of the main pancreatic duct and branch duct is inevitable.     

COMPARISON OF THE HEIGHT OF PAPILLARY TUMOR IN INTRADUCTAL PAPILLARY MUCINOUS NEOPLASM OF THE PANCREAS BETWEEN MEASURED PREOPERATIVE IMAGES AND RESECTED MATERIAL

The height of the mural nodules and papillary tumors in main pancreatic duct or dilated branch duct is the most important factor for diagnosis of intraductal papillary mucinous neoplasm (IPMN). In this study, the authors compared the height of the papillary lesions and mural nodules between the height of resected tissues and the height detected by the preoperative imaging tools (endoscopic ultrasonography [EUS] and intraductal ultrasonography [IDUS]) in 38 patients with IPMN. In 21 out of 23 cases of adenoma, and in cases with the non‐invasive cancer, the difference of the height of operative and preoperative analysis measured by EUS and IDUS was within 1–2 mm. EUS and IDUS are useful for diagnosis of degree of malignancy in IPMN.     

DIAGNOSIS AND MANAGEMENT OF INTRADUCTAL PAPILLARY MUCINOUS TUMORS OF THE PANCREAS

Intraductal papillary mucinous tumors (IPMT) of the pancreas are histopathologically adenocarcinoma, adenoma, or even hyperplasia, and, therefore, selection of the treatment strategy is very important. While the International Consensus Guidelines for Management of Intraductal Papillary Mucinous Neoplasms and Mucinous Cystic Neoplasms of the Pancreas have been published, the authors discussed the diagnosis and management of IPMT in the session of Endoscopy Forum Japan 2005. All the discussants agreed to the guideline’s recommendation to resect all main duct IPMT. However, they reserved their consensus on the recommendation of resection for branch duct IPMT which are 5 mm or higher in the mural nodule height. General opinion is that treatment should be considered for lesion where the main duct diameter is over 7 mm, and that dilated branch duct diameter is not an important factor, while it is defined as an important factor in the guidelines.     

Diagnosis and patient management of intraductal papillary-mucinous tumor of the pancreas by using peroral pancreatoscopy and intraductal ultrasonography.

BACKGROUND & AIMS: Intraductal papillary-mucinous tumor (IPMT) of the pancreas has attracted increasing interest because of its unique presentation. The differential diagnosis between malignant and benign tumors is extremely important in the determination of the therapy for IPMT. The aims of this study are to determine the usefulness of peroral pancreatoscopy (POPS) and intraductal ultrasonography (IDUS) in IPMT for the differentiation of malignant from benign disease, and to evaluate the significance of these techniques as new preoperative examinations. METHODS: Sixty histopathologically confirmed patients with IPMT underwent POPS and/or IDUS preoperatively. POPS was perfomed in all patients, and IDUS in 40. Findings of POPS and IDUS were compared with histopathology of resected specimens. The postoperative follow-up data were analyzed. RESULTS: Protruding lesions were detected by POPS in 40 patients. They were classified into 5 groups. Fish-egg-like type with vascular images, villous type, and vegetative type were considered to be malignant. By IDUS, lesions protruding 1 mm or more were observed in 36 patients. Of the lesions protruding 4 mm or more, 88% were malignant. Combination of POPS and IDUS improved the differential diagnosis between benign and malignant IPMT. The 3-year cumulative survival rate and disease-free survival rate were extremely high at 95% and 93%, respectively. CONCLUSIONS: The combination of POPS and IDUS results in a considerably improved differential diagnosis between malignant and benign IPMT and is useful for determining an effective therapeutic approach. These new techniques can contribute to improvements in postoperative results.     

INTRADUCTAL ULTRASONOGRAPHY BEFORE BILIARY DRAINAGE AND TRANSPAPILLARY BIOPSY IN ASSESSMENT OF THE LONGITUDINAL EXTENT OF BILE DUCT CANCER

Aim: We evaluated the diagnostic efficacy of transpapillary intraductal ultrasonography before biliary drainage (IDUS‐BD) and transpapillary biopsy (TPB) for the assessment of the longitudinal extent of bile duct cancer. Methods: Between November 1999 and January 2005, we performed IDUS‐BD and TPB preoperatively in 27 patients with carcinoma of the extrahepatic bile duct. Following IDUS‐BD, TPB was performed under fluoroscopic guidance immediately after endoscopic sphincterotomy. The diagnostic efficacy of IDUS‐BD and TPB for the longitudinal extent of the cancer and the complications which accompanied the procedure were evaluated. Results: The overall success rate of sampling and the diagnostic accuracy of bile duct cancer by TPB were 85.3% (192/225) and 85% (23/27), respectively. The sensitivity, specificity and accuracy of the assessment of the longitudinal extent of cancer on the hepatic and duodenal sides by IDUS‐BD were 82%, 70%, 78% and 85%, 43%, 70%, respectively. Those by a combination of IDUS‐BD and TPB were 88%, 80%, 85% and 77%, 86%, 80%, respectively. Overestimation of the longitudinal extent of BD cancer by IDUS‐BD was mainly due to inflammation and obscure images, especially resulting from collapse of the bile duct on the duodenal side of the tumor, and was corrected by TPB in four of five patients. No serious complications occurred following the combination of IDUS‐BD and TPB. Conclusions: TPB is useful for preoperative histological diagnosis of bile duct cancer. The combination of IDUS‐BD and TPB is practical for evaluation of its longitudinal extent; basically, IDUS‐BD is sufficient on the hepatic side of the tumor, but concomitant TPB is recommended on the duodenal side.     

Differential diagnosis of intraductal papillary‐mucinous tumor of the pancreas by endoscopic ultrasonography and intraductal ultrasonography

Background: Intraductal papillary‐mucinous tumor (IPMT) of the pancreas has a broad spectrum of histology ranging from hyperplasia to adenocarcinoma. Therefore, it is important to differentiate between the malignant and benign lesions to determine the therapeutic strategy for IPMT. Patients and Methods: Thirty‐nine patients with IPMT (27 men and 12 women, mean age: 63.3 years) underwent surgery between January 1985 and March 2002. The size of the cystic lesion, the maximum diameter of the main pancreatic duct (MPD), and the height of the papillary tumor inside the cyst were investigated by endoscopic ultrasonography (EUS) and/or intraductal ultrasonography (IDUS) before operation. These preoperative clinical findings were compared with the pathological findings of the resected specimen. Results: The size of the cystic lesion, the diameter of MPD, and the height of the papillary tumor in cases with malignant IPMT (invasive and non‐invasive carcinoma) were larger than those in cases with benign IPMT (adenoma and hyperplasia). Analysis of the images of the lesions revealed that the following three factors are important for diagnosing IPMT: (i) the size of the cystic lesion is ≥ 30 mm; (ii) the diameter of MPD is ≥ 8 mm; (iii) the height of the papillary tumor inside the cyst is ≥ 3 mm. It was not significant to differentiate between benign and malignant IPMT based on factor (i), but statistically significant (P < 0.001) based on factors (ii) and (iii). Conclusions: EUS and IDUS are useful in the differential diagnosis of IPMT, especially in the differentiation between malignant and benign IPMT.     

DIAGNOSTIC UTILITY OF PERORAL CHOLANGIOSCOPY AND TRANSPAPILLARY INTRADUCTAL ULTRASONOGRAPHY FOR BILE DUCT CARCINOMA

Recently, new diagnostic procedures such as video peroral cholangioscopy (POCS) and transpapillary intraductal ultrasonography (IDUS) have been available for diagnosis of biliary diseases. These new modalities are especially useful for diagnosis of minute bile duct lesions on cholangiogram and correct diagnosis of lateral extension of bile duct carcinoma. In this paper, showing some effective cases of POCS and IDUS, we present our diagnostic approach for bile duct carcinoma and discuss the future prospects of POCS and IDUS.     

INTRADUCTAL ULTRASONOGRAPHY OF THE GALLBLADDER IN APPLICATION OF THE ENDOSCOPIC NASO‐GALLBLADDER DRAINAGE

Background: Although endoscopic naso‐gallbladder drainage (ENGBD) for gallbladder disease is useful, the procedure is difficult and investigations involving many cases are lacking. Furthermore, reports on transpapillary intraductal ultrasonography (IDUS) of the gallbladder using a miniature probe are rare. Methods: A total of 150 patients (119 suspected of having gallbladder carcinoma, 24 with acute cholecystitis (AC), and seven with Mirizzi’s syndrome (MS)) were the subject. (i) ENGBD: We attempted to put ENGBD tube into the GB. (ii) IDUS of the gallbladder: Using the previous ENGBD tube, we attempted to insert the miniature probe into the gallbladder and perform transpapillary IDUS of the gallbladder. In five patients, we attempted three‐dimensional intraductal ultrasonography (3D‐IDUS). Results: (i) ENGBD: Overall success rate was 74.7% (112/150); the rate for the patients suspected of having gallbladder carcinoma was 75.6% (90/119), and was 71.0% (22/31) for the AC and MS patients. Inflammation and jaundice improved in 20/22 successful patients with AC and MS. Success rate was higher when cystic duct branching was from the lower and middle parts of the common bile duct than from the upper part, and was higher when branching was upwards than downwards. (ii) IDUS of the gallbladder: Success rate for miniature probe insertion into the gallbladder was 96.4% (54/56). Lesions could be visualized in 50/54 patients (92.6%). Of these, detailed evaluation of the locus could be performed in 41. In five patients attempted 3D‐IDUS, the relationship between the lesion and its location was readily grasped. Conclusion: IDUS of the gallbladder is superior for diagnosing minute images. Improvement on the device will further increase its usefulness.     

Intraductal papillary-mucinous tumors of the pancreas: differential diagnosis between benign and malignant tumors by endoscopic ultrasonography

OBJECTIVES: Recently, intraductal papillary-mucinous tumor (IPMT) of the pancreas has increasingly been recognized. However, differential diagnosis between benign and malignant IPMT is often difficult using conventional imaging modalities. The purpose of this study was to retrospectively investigate the value of endoscopic ultrasonography (EUS) for differentiating malignant from benign IPMT. METHODS: A total of 51 patients with IPMT were preoperatively examined by EUS. The endosonograhic findings were compared with histopathological findings of the resected specimens. RESULTS: In main duct type IPMT, the diameter of the main pancreatic duct (MPD) was > or =10 mm in seven of the eight malignant tumors, compared with two of the seven benign tumors (p < 0.05). In branch duct type IPMT, three of the four large tumors (>40 mm) with irregular thick septa were malignant lesions. In both main duct type IPMT and branch duct IPMT, eight patients had large mural nodules (>10 mm); seven of the eight tumors were malignant and one of the eight tumors was benign. When the tumor was diagnosed as malignant according to above three findings, EUS was able to differentiate between malignant and benign IPMT with an accuracy of 86%. CONCLUSIONS: Main duct type tumors with > or =10 mm dilated MPD, branch duct type tumors (>40 mm) with irregular septa, and large mural nodules (>10 mm) strongly suggest malignancy on EUS. EUS would be a useful modality for differentiating between benign and malignant IPMT.     

Surgical treatment of intraductal papillary-mucinous tumor (IPMT) of the pancreas: operative indications based on surgico-pathologic study focusing on invasive carcinoma derived from IPMT

Background/Purpose. Between 1979 and 2000, 51 patients with intraductal papillary-mucinous tumor (IPMT) of the pancreas underwent surgical resection. Methods. The patients were reviewed to disclose the surgical pathology of invasive carcinoma derived from IPMT and to determine the surgical indications for IPMT on the basis of the pathologic findings. Results. The incidence of invasive carcinoma derived from IPMT according to the localization of the tumor was as follows: 4/9 (44%) in the main pancreatic duct (MPD type), 4/9 (44%) showing ductal spread from the MPD to branch ducts (mixed type), and 2/33 (6%) in the 2 branch duct (branch type). The maximal size of the intraductal spread of invasive carcinomas (8 of 18 cases in the MPD and mixed type together and 2 of 33 cases in the branch type) was as follows: 6/8 (75%) in the MPD and mixed type were over 6?cm in size, and the 2-branch-type invasive carcinomas were within the 3-cm size range. Conclusions. We concluded that for both invasive and noninvasive IPMTs, surgical resection was necessary for any MPD or mixed-type IPMTs, and that surgical resection was appropriate for branch-type lesions larger than or equal to 3?cm in diameter, or for lesions smaller than 3?cm showing rapid growth on clinical images.     

LIMITATIONS OF PERORAL CHOLANGIOSCOPY IN THE DIAGNOSIS OF BILIARY TUMORS

In patients with carcinoma of the bile duct, a primary tumor with tumor spread along the bile duct is frequently shown as a stricture of the bile duct with ductal irregularity and rigidity on ERCP. In patients in whom histological diagnosis is necessary, larger caliber peroral cholangioscopes with a larger working channel should be chosen because they have the advantage of simultaneous biopsy diagnosis. However, peroral cholangioscopy (POCS) has limitations to inspect mucosal tumor spread along the upstream ducts due to the tight stricture, to obtain sufficient biopsy material for confirming the diagnosis, and to identify submucosal tumor spread. With further improvement of techniques, the combined use of POCS with biopsy under direct vision and intraductal ultrasonography (IDUS) is expected to be the best way in the diagnosis of biliary tumors.     

INTRADUCTAL ULTRASONOGRAPHY IN THE DIAGNOSIS OF BILE DUCT STONES: WHEN AND WHOM?

Background and Aim: Biliary intraductal ultrasonography (IDUS) is highly sensitive in visualizing bile duct stones (BDS). Indications for IDUS, however, in cases of suspected BDS have not yet been established. The aim of the present study was to elucidate adequate indications for IDUS in cases that undergo endoscopic retrograde cholangiopancreatography (ERCP) due to suspected BDS. Methods: A total of 213 patients who were suspected of having BDS were included in this retrospective study. The patients were divided into two groups: Group A in which BDS was visualized by ERCP; and Group B in which BDS was demonstrated only by IDUS. Comparison between the groups was carried out. Results: ERCP successfully visualized BDS in 166 patients. Forty‐seven patients underwent IDUS, which revealed BDS and biliary sludge in 12 and eight patients, respectively. The diameter of the largest stone was 13 ± 6 mm in Group A and 5 ± 1 mm in Group B (P < 0.001). The sensitivity, specificity, and accuracy of ERCP in the diagnosis of BDS were 93%, 100%, and 94%, respectively. The sensitivity was influenced by the size of BDS: 100% in cases of stones ≥8 mm in size, but 74% in those with stones <8 mm. In cases with stones <8 mm, the sensitivity was significantly affected by the bile duct diameter (≥12 mm vs <12 mm, P < 0.05). Conclusion: When ERCP fails to visualize stones in patients with suspected BDS, IDUS is recommended, especially in those with a bile duct ≥12 mm in diameter.     

ENDOSCOPIC DIAGNOSIS OF INTRADUCTAL PAPILLARY MUCINOUS NEOPLASM USING PERORAL PANCREATOSCOPY WITH NARROW BAND IMAGING

A 77‐year‐old man was diagnosed with a pancreas cyst at another hospital. Abdominal ultrasonography revealed a cyst in the head of the pancreas and a small protrusion. These findings suggested intraductal papillary mucinous neoplasm. Cytologic finding of the pancreatic juice revealed a Class III lesion, and intraductal ultrasonography and peroral pancreatoscopy (PPS) were performed. An abnormal course of main pancreatic duct (MPD) prevented the insertion of an ultrasonography probe into the MPD, and PPS was performed. The mucosal surface of the MPD near the papillary area was normal, and narrow band imaging (NBI) clearly showed the vascular structure. When the PPS was inserted more deeply, many small protrusions were observed and NBI delineated the protrusions more clearly. Papillary protrusions were observed in the cyst, but NBI did not reveal any tumor vessels. These findings led to a diagnosis of benign intraductal papillary mucinous neoplasm lesion. Since then, follow‐up examinations have been made. Changes in the cyst and protrusion have not been observed.     

MANAGEMENT STRATEGIES FOR BRANCH DUCT INTRADUCTAL PAPILLARY‐MUCINOUS NEOPLASMS

Management strategies for branch duct intraductal papillary‐mucinous neoplasms (IPMN) have been discussed. The authors’ clinical criteria with special attention to the size of cystic lesion, the diameter of main pancreatic duct, and the size of mural nodule established in 2001 is useful for managing branch duct IPMN. A total of 55 of 60 cases with branch duct IPMN had no radiologic progression during an average follow up of 701 days. In addition, no case out of 27 cases with branch duct IPMN followed up for 3 years changed to positive in cytological examination using pancreatic juice. There were four cases of branch type IPMN with some changes in their image findings. There were two cases of branch type IPMN with pancreatic symptoms. A total of three out of these five cases were surgically resected. The ordinary‐type ductal carcinoma was detected in two cases with branch duct IPMN at 3 or 4 years later. These results suggest that a long‐term careful follow‐up study by computed tomography or ultrasonography at every 6 months would be needed in the management of branch duct IPMN. Further studies will be needed to dissolve this problem in the future.     

Intraductal papillary mucinous tumor of the pancreas: computerized tomography and magnetic resonance imaging features

The aim of this study was to analyze the computerized tomography (CT) and magnetic resonance imaging (MRI) features of intraductal papillary mucinous tumor (IPMT) of the pancreas. The cases of eight patients with pathologically proven IPMT (1 papillary hyperplasia, 7 adenocarcinoma) of the pancreas were retrospectively reviewed. There were five men and three women with ages ranging from 42 to 82 years. Imaging studies included six thin-section dynamic CT scans, seven MRI scans, one MR cholangiopancreatography scan, and two endoscopic retrograde cholangiopancreatography scans. There was only one benign IPMT, which presented as a unilocular cyst in the pancreatic body with no mural nodules and no dilatation of the main pancreatic duct (MPD). All seven patients with malignant IPMT had multilocular cysts with papillary projections in the pancreatic head and/or uncinate process accompanied by dilated MPD (5 diffuse, 2 segmental). Communication between the cystic lesions and the MPD were evident in all seven patients. One patient had small mural nodules in the branch ducts of the pancreatic body and five had a bulging papilla with a patulous orifice. A mass effect resulting in biliary obstruction was shown in one patient. One patient had a ruptured cyst with mucin leakage into the right anterior pararenal space following sono-guided aspiration. In conclusion, the main imaging feature of IPMT in our patients was a multilocular cyst with papillary projections located in the pancreatic head and uncinate process. Although CT and MRI cannot differentiate mucin content from pancreatic juice, communication between the cystic lesion and the dilated MPD and a bulging papilla with a patulous orifice are characteristics of IPMT.     

Endosonographic diagnosis of intraductal papillary‐mucinous pancreatic tumors by intraductal ultrasonography

Background: Intraductal papillary‐mucinous pancreatic tumors (IPMT) are intraductal lesions formed by mucin‐producing epithelium, which proliferates in a papillary pattern, and presents a spectrum from hyperplasia to adenocarcinoma. The value of intraductal ultrasonography (IDUS) for excluding malignancy has not been assessed in a case series previously. Methods: Intraductal ultrasonography was performed in 17 patients with IPMT (12 with adenocarcinoma and five with adenoma) between November 1993 and June 2002. Intraductal ultrasonography was used to determine the maximum height and maximum cross‐sectional area of protruding lesions. Results were compared after dividing the tumors into three groups: a benign lesion group, a non‐invasive cancer group, and an invasive cancer group. The resection line was located over 10 mm from the edge of the protruding lesion visualized by intraductal ultrasonography. Results: All adenocarcinomas had a height ≥ 5 mm and all benign lesions had a height ≤ 3 mm, with this difference being significant (P = 0.0034). The height of non‐invasive and invasive cancer was similar. The maximum cross‐sectional area of the protrusion was smaller for benign lesions (≤ 15 mm²) than for non‐invasive cancer (≥ 34 mm², P = 0.0034). The cross‐sectional area of the protrusion was greater in patients with invasive cancer than in those with non‐invasive cancer (P = 0.0367). All surgical margins have remained clear and no patient has suffered from a recurrence during 1 to 8 years of follow‐up computed tomography and ultrasonography. Conclusions: Intraductal ultrasonography can distinguish benign from malignant IPMT based on the height and maximum cross‐sectional area of the protruding tumor.