- Email: [email protected]
Endoscopic ultrasonographic findings predict the risk of carcinoma in branch duct intraductal papillary mucinous neoplasms of the pancreas
Pancreatology 12 (2012) 141e145
Contents lists available at SciVerse ScienceDirect
Pancreatology journal homepage: www.elsevier.com/locate/pan
Original article
Endoscopic ultrasonographic findings predict the risk of carcinoma in branch duct intraductal papillary mucinous neoplasms of the pancreas Noritoshi Kobayashi a, *, Kazuya Sugimori b, Takeshi Shimamura a, Kunihiro Hosono a, Seitaro Watanabe a, Shingo Kato a, Michio Ueda c, Itaru Endo c, Yoshiaki Inayama d, Shin Maeda a, Atsushi Nakajima a, Kensuke Kubota a a
Gastroenterology Division, Yokohama City University Graduate School of Medicine, 3-9, Fuku-ura, Kanazawa-ku, Yokohama 236-0004, Japan Gastroenterological Center, Yokohama City University Medical Center, Yokohama, Japan Gastroenterological Surgery Division, Yokohama City University Graduate School of Medicine, Yokohama, Japan d Pathological Division, Yokohama City University Graduate School of Medicine, Yokohama, Japan b c
a b s t r a c t Keywords: Intraductal papillary mucinous neoplasm of pancreas Endoscopic ultrasonography Branch duct type Diagnosis Mural nodule
Background: The preoperative diagnosis of branch duct intraductal papillary mucinous neoplasm (IPMN) of the pancreas can be very difficult, since low-risk and high-risk lesions can be difficult to differentiate even after cytological analysis. The purpose of this study was to evaluate the preoperative diagnostic value of endoscopic ultrasonography (EUS) in differentiating low-risk and high-risk IPMNs. Methods: We retrospectively identified 36 patients who underwent preoperative EUS for branch duct IPMNs. The pathological diagnosis after surgical resection was low-grade dysplasia (n ¼ 26), moderate dysplasia (n ¼ 1), high-grade dysplasia or carcinoma in situ (n ¼ 5), and invasive carcinoma (n ¼ 4). We divided the patients into two groups: low risk (low-grade dysplasia or moderate dysplasia) and high risk (high-grade dysplasia or carcinoma). We focused on the diameter of the cystic dilated branch duct, the main pancreatic duct, and the mural nodule as measured using the EUS findings. Results: The cystic dilated branch duct diameter (31.5 mm vs. 41.9 mm, P ¼ 0.0225) was significantly correlated with low-risk and high-risk IPMNs, but the main pancreatic duct diameter (5.37 mm vs. 5.44 mm, P ¼ 0.9418) was not significantly correlated with the low-risk and high-risk IPMNs. The mural nodule diameter of the papillary protrusions (4.3 mm vs. 16.4 mm, P < 0.0001) and the width diameter of the mural nodule (5.7 mm vs. 23.2 mm, P < 0.0001) were significantly correlated with low-risk and highrisk IPMNs. Conclusions: The mural nodule of papillary protrusions diameter and width diameter observed using EUS was a reliable preoperative diagnostic finding capable of distinguishing low-risk and high-risk IPMNs. Copyright Ó 2012, IAP and EPC. Published by Elsevier India, a division of Reed Elsevier India Pvt. Ltd. All rights reserved.
1. Introduction Intraductal papillary mucinous neoplasms (IPMNs) are characterized by the intraductal proliferation of mucin-producing epithelial cells and the cystic dilatation of pancreatic ducts [1]. IPMNs are thought to progress from low-grade, moderate, and high-grade dysplasia (carcinoma in situ [CIS]) to invasive carcinoma originating as intraductal papillary mucinous carcinoma (IPMC) [2]. Although IPMNs are currently thought to have the potential to become malignant [3], whether all IPMNs have such potential and
* Corresponding author. Tel.: þ81 45 787 2640; fax: þ81 45 784 3546. E-mail address: [email protected] (N. Kobayashi).
the time course for progression remains unknown. Therefore, the indications for surgery based on the biological behavior of this tumor are currently of great concern [4,5]. However, preoperative diagnoses capable of distinguishing benign from malignant IPMNs are challenging to make. IPMNs have been subdivided into main duct IPMNs and branch duct IPMNs based on imaging studies or histological features [1,6]. Main duct IPMNs and branch duct IPMNs are associated with a significantly different prevalence of cancer, ranging from 57% to 92% and from 6% to 46%, respectively. Consequently, this classification has prognostic implications [1,4,6e9]. In recent consensus guidelines for IPMNs, main duct IPMNs have been regarded as having a higher malignant potential, and surgical resection has been recommended [6,8e14]. However, the treatment strategy for
1424-3903/$ e see front matter Copyright Ó 2012, IAP and EPC. Published by Elsevier India, a division of Reed Elsevier India Pvt. Ltd. All rights reserved. doi:10.1016/j.pan.2011.12.008
142
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
branch duct IPMNs has not been determined. The International Consensus Guidelines have put forward an algorithm for surgical management that is based on cyst size, patient symptoms, and a high-risk appearance (mural nodules, thick septum or positive cytology) [14]. Recent advances in imaging modalities have led to the incidental identification of many asymptomatic cystic lesions, and preoperative histological or cytological diagnoses are not always sufficient to determine the surgical indications for this tumor [15,16]. Some previous series have demonstrated a positive correlation between size and malignant IPMN pathology [17e19]. Many imaging modalities for the examination of IPMNs have also been used as preoperative diagnostic tools. In particular, endoscopic ultrasonography (EUS) is regarded as the most reliable modality for examining IPMNs [20,21]. We performed EUS for preoperative diagnosis in 36 cases of branch duct IPMNs. We then compared the EUS findings and the final pathological diagnoses retrospectively. The results of this study suggest that the EUS findings of branch duct IPMNs may enable a more appropriate selection of patients for operative resection. 2. Materials and methods Patients with IPMNs were identified by searching the Yokohama City University Hospital databases for patients treated between 1992 and 2010. There are no Caucasians in my study. Two hundred and thirty-eight patients with imaging and/or histological findings consistent with IPMNs were analyzed in this study. We subclassified the IPMNs as either main duct or branch duct IPMNs according to the preoperative imaging findings and the postoperative histological findings. In this study, we did not include any mixed-type IPMNs because pure branch duct IPMNs are very rare and many branch IPMNs are reclassified as mixed type [14,20]. Consequently, we defined the subclassification according to the dominant lesions of the papillary epithelial cells. Branch duct IPMNs were diagnosed when EUS and imaging findings (CT, MRCP, or ERCP) showed one or more pancreatic cysts communicating with the main pancreatic duct. We diagnosed 41 patients as having main duct IPMNs and 197 patients as having branch duct IPMNs. Pancreatic resection was indicated for main duct IPMNs and for branch duct IPMNs with symptomatic lesions and/or lesions suspected of being malignant. We defined malignant branch duct IPMNs as those in which a cytological analysis of the pancreatic juice collected by the transpapillary route was class IV or V and/ or the cystic dilated branch diameter was over 30 mm with a mural nodule based on EUS findings. These indications were based on the results of a previously reported study and the consensus guidelines [11,14]. We used EUS (UM20, UM2000; Olympus Japan, Tokyo) to obtain a preoperative diagnosis in 36 cases. We measured the height and width of the mural nodules, the main pancreatic duct diameter, and the cystic dilated branch diameter based on EUS findings. We defined the “height of the mural nodules” as the maximum length perpendicular to the cystic dilated branch wall and the “width of the mural nodules” as the maximum length along the walls of the dilated branches and perpendicular to the maximum height of the mural nodules (Fig. 1A). “Cystic dilated branch” and “main pancreatic duct” were defined as the maximum sizes based on the EUS findings. In situations where a multilocular cyst was present, we measured the maximum size of all the cystic components. The EUS findings were checked by two expert endoscopists (K.K, N.K). All the IPMNs were pathologically reexamined, and the diagnosis of IPMN was confirmed. Surgically resected specimens were fixed in 10% formalin and cut into serial 5-mm-thick slices, horizontally in the pancreas head and sagittally in the pancreas body and tail. All the sections were stained with hematoxylin and eosin
Fig. 1. Endoscopic ultrasonography clearly shows the mural nodule in a cystic dilated branch of the pancreas (A). The solid line represents the width of the mural nodule. The dotted line represents the height of the mural nodule. Macroscopic findings of the mural nodule in the cystic dilated branch (B). The macroscopic width and height of the mural nodule were nearly equal to the EUS findings. This tumor was pathologically diagnosed as a high-grade dysplasia (carcinoma in situ).
for pathological examination. We also measured the height and lateral spread of the mural nodules pathologically after they had been fixed in 10% formalin; the same definitions as those described above were used (Fig. 1B). We also examined the pathological grade according to the WHO classification (IPMN with low-grade dysplasia, IPMN-LD; IPMN with moderate dysplasia, IPMN-MD; IPMN with high-grade dysplasia, IPMN-HD; invasive carcinoma originating IPMC, IC-IPMC). The statistical analysis was performed using the Stat View program (SAS Institute, Cary, NC, USA). Continuous data were presented as the mean SD. The size cut-off values for differentiating benign and malignant IPMNs were investigated by constructing receiver operating characteristic (ROC) curves and calculating the area under the curves. Continuous variables were analyzed using the Student t-test. The c2 test was used for categorical variables. P values of less than 0.05 were considered statistically significant. The relations were also analyzed using the correlation coefficients. 3. Results Thirty-six cases (21 men, 15 women) were analyzed in this study. The postoperative pathological examination revealed 26 cases of IPMN-LD, one case of IPMN-MD, five cases of IPMN-HD, and four cases of IC-IPMC. Twenty-four asymptomatic cases were discovered incidentally. Eleven cases initially presented with abdominal pain, and one case presented with back pain. One case presented with a body weight loss of 3 kg during a 6-month period. The median age at the time of the initial diagnosis was 66 years. Malignant diseases of extra-pancreatic organs were present in four cases. Diabetes mellitus (DM) was present in four cases. We defined IPMN-LD and IPMN-MD as low-risk IPMNs and IPMN-HD and IC-IPMC as high-risk IPMNs. First, we compared the low-risk IPMNs and the high-risk IPMNs with regard to the clinical
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
143
Table 1 Clinical characteristics of the patients with branch duct IPMNs.
Total IPMNs Low-risk IPMNs (IPMN-LD/IPMN-MD) High-risk IPMNs (IPMN-HD; CIS/IC-IPMC) P value (Low-risk vs High-risk)
Pathological diagnosis (IPMN-LD/IPMN-MD/IPMN-HD; CIS/IC-IPMC)
Gender (male/female)
Age (median)
Location (h/b/t/m)
Symptom
Extra-pancreatic cancer
DM
36 (26/1/5/4) 27 (26/1) 9 (5/4)
36 (21/15) 27 (14/13) 9 (7/2)
66 (36e78) 66 (36e77) 66 (49e78)
(15/13/2/6) (12/10/2/3) (3/3/0/3)
10 (27.8%) 8 (29.6%) 2 (22.2%)
4 (11.1%) 3 (11.1%) 1 (11.1%)
4 (11.1%) 3 (11.1%) 1 (11.1%)
0.1851
0.9303
0.9741
0.925
0.925
IPMN-LD, IPMN with low-grade dysplasia; IPM-MD, IPMN with moderate dysplasia; IPMN-HD, IPMN with high-grade dysplasia; CIS, carcinoma in situ; IC-IPMC, Invasive carcinoma originating in IPMC; h, head; b, body; t, tail; m, multiple lesions; DM, diabetes mellitus.
Table 2 EUS findings of the branch duct IPMNs. EUS findings
Main pancreatic duct (mm)
Cystic branch duct (mm)
Papillary protrusions of the nodule (mm)
Width of the nodule (mm)
Total IPMNs Low-risk IPMNs (IPMN-LD/IPMN-MD) High-risk IPMNs (IPMN-HD; CIS/IC-IPMC) P value (Low-risk vs High-risk)
5.39 (3e13) 5.37 (3e11) 5.44 (3e13) P ¼ 0.9418
34.1 (10e70) 31.5 (10e53) 41.9 (25e70) P ¼ 0.0225
7.39 (0e35) 4.3 (0e15) 16.4 (10e35) P < 0.0001
10.1 (0e60) 5.7 (0e22) 23.2 (8e60) P < 0.0001
EUS, endoscopic ultrasonography.
background (Table 1). The sex ratio and the median age at the time of the initial diagnosis were not significantly different between the two groups (male/female, 14/13 vs. 7/2, P ¼ 0.1851; 64.8 vs. 65.1 years, P ¼ 0.9303). The incidence of symptomatic episodes was not significantly different between the two groups (P ¼ 0.9741), nor was the rate of extra-pancreatic malignant disease (P ¼ 0.9250). The prevalence of DM was not significantly different between the two groups (P ¼ 0.9250). We next compared the EUS findings for low-risk and high-risk IPMNs. The cystic dilated branch duct diameter (31.5 mm vs. 41.9 mm, P ¼ 0.0225) was significantly correlated with low-risk and high-risk IPMNs (Table 2). However, the main pancreatic duct diameter (5.37 mm vs. 5.44 mm, P ¼ 0.9418) was not significantly correlated with low-risk and high-risk IPMNs. The mural nodule diameter of the papillary protrusions (4.3 mm vs. 16.4 mm, P < 0.0001) and the width diameter of the mural nodule (5.7 mm vs. 23.2 mm, P < 0.0001) were significantly correlated with lowrisk and high-risk IPMNs (Table 2). We next examined the ROC curve and the area under the curve (AUC). The mural nodule diameter of the papillary protrusion (0.957) and the width diameter (0.947) were excellent clinical parameters for distinguishing low-risk from high-risk IPMNs (Fig. 2). If a cut-off point of 10 mm was adopted for papillary protrusion, the sensitivity was 100%, the specificity was 85.2%, and the accuracy was 88.9%. If a cut-off point of 15 mm was adopted for lateral spread, the sensitivity was 88.9%, the specificity was 92.6%, and the accuracy was 91.7% (Table 3). Finally, we compared the macroscopic pathological size of the mural nodule with the EUS findings. The correlation coefficients of the diameter of papillary protrusion (r ¼ 0.840) and the width diameter (r ¼ 0.868) were highly correlated with the EUS findings and the macroscopic pathological findings (Fig. 3A,B).
present study, however, these factors were not correlated with high-risk IPMNs. Previous studies included many symptomatic cases (about 70%). Consequently, these tumors may have been more advanced, and the patients’ backgrounds may have differed from those in the present study. In our study, the final pathological and radiological diagnoses were only branch duct-type IPMNs, and many cases were asymptomatic early-stage tumors (symptomatic cases accounted for only 10% of the series). In practice, many cystic lesions in the pancreas are discovered incidentally, so our data may reflect normal clinical practice better and be more reliable for identifying malignant disease in asymptomatic branch duct IPMNs. The present study focused on the value of preoperative EUS findings, especially for mural nodules. In this study, the accuracy ranged from 88.9% to 91.3%, depending on the mural nodule diameter. In previous studies, the accuracy of EUS for discriminating benign and malignant forms varied from 65% to 94% [15,23e25]. According to many past reports, the accuracy of other modalities was 75%e78% for MDCT, 78%e86% for MRI, 79%e83% for ERP, and an impressive 88% for combined intraductal US and peroral pancreatoscopy [15,23e25]. Thus, EUS may be slightly superior to other modalities for distinguishing low-risk and highrisk IPMNs.
4. Discussion IPMNs of the pancreas are problematic because of the difficulty in preoperative diagnosis. The clinical behavior and operative indications of branch duct IPMNs are very difficult to predict. In the present study, no predictive factors of high-risk IPMNs other than imaging findings were found among the patient-related factors that were examined. According to past reports, a patient age of greater than 50 years, Caucasian ancestry, and body weight loss are independent risk factors for malignant IPMNs [22]. In the
Fig. 2. ROC curve and the area under the curve (AUC) for distinguishing low-risk from high-risk IPMNs according to the size of the mural nodule. The mural nodule diameter of the papillary protrusion (0.957) and the width (0.947) of the mural nodule were excellent clinical parameters for distinguishing low-risk and high-risk IPMNs.
144
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
Table 3 Accuracy of imaging diagnosis for the branch duct IPMNs by diameter of mural nodule as visualized using EUS.
Papillary protrusions of the nodule >10 mm Lateral spread of the nodule >15 mm
Sensitivity
Specificity
Predictive positive value
Predictive Negative value
Accuracy
100.00% 88.90%
85.20% 92.60%
69.30% 80.00%
85.20% 96.20%
88.90% 91.70%
Some previous reports of large series of branch duct IPMNs have focused on the cystic dilated branch diameter, mural nodule, main pancreatic duct diameter, and tumor size based on CT findings. Jang’s large study (n ¼ 138) revealed that a tumor size of 20 mm and/or papillary protrusions were indications for surgical treatment [26], while Rodriguez (n ¼ 145) revealed that surgical treatment should be performed for tumors with a size of greater than 30 mm, symptomatic tumors, or mural nodules [10]. Woo’s study (n ¼ 190) reported that observation is recommended for patients with branch duct IPMNs smaller than 30 mm with no suspicious features visible upon CT imaging [27]. The presence of large tumors on CT findings was regarded as an obvious indication of malignancy. Cases in which the indications for surgical treatment were difficult to decide included equivocal small tumors visible on CT findings. EUS was very useful for identifying these small lesions as branch duct IPMNs. Few studies have examined the preoperative diagnostic value of EUS findings for branch duct IPMNs. Okabayashi reported that branch duct IPMNs with a cystic size of greater than 30 mm and a mural nodule height of greater than 5 mm on EUS findings should be resected [28]. Kubo reported that EUS findings showing large cystic lesions (>40 mm) with irregular thick septa and a large mural nodule (>10 mm) were strongly suggestive of malignancy [21]. And Kanno reported that EUS findings showing large mural nodules (>6.5 mm) were suggestive of invasive IPMN [29]. Kobayashi reported that patients with branch duct-type large cysts with septum-like strictures should undergo surgery, while those with small papillary protrusions have a benign course on EUS
Fig. 3. Comparison of the macroscopic pathological size of the mural nodule with the EUS findings. The correlation coefficients for the diameter of papillary protrusion (A) (r ¼ 0.840) and the width of the mural nodule (B) (r ¼ 0.868) were highly correlated with the EUS findings and the macroscopic pathological findings. High-risk IPMNs are shown as black circles.
[20]. In our study, papillary protrusions (>10 mm), and the width diameter (>15 mm) were highly accurate (91.7%) predictors of high-risk IPMNs. Our study was retrospective and limited by small sample size. Nevertheless, our criteria are simple and available one, so any pancreatic centre should be able to use them. In our study, we defined the width diameter as the measurement perpendicular to the maximum height of the tumor nodule. Of course, in this study, “width” was not equal to the lateral spread of the microscopic pathological findings. Nevertheless, the width diameter of the nodule on the EUS findings was correlated with the macroscopic pathological findings. According to general rules for the study of pancreatic cancer in Japan, the tumor diameter of IPMNs is defined as the longest diameter of the lateral spread of the tumor [30]. So, the width diameter of the mural nodule was a reasonable parameter for defining the size of the mural nodule. In some cases of IC-IPMC, there was a difference between the EUS measurement and the pathological measurement of mural nodule size. The mural nodule and invasive component of IPMCs may be difficult to distinguish using EUS in some cases. In our study, the cystic dilated branch diameter was also useful for distinguishing low-risk and high-risk IPMNs. Some reports, including the International Consensus guidelines, suggest that a diameter of 30 mm or 40 mm was important for discriminating between benign and malignant tumors [14,31], but many recent reports have indicated that the cystic dilated branch diameter does not reflect tumor malignancy [32,33]. The cystic dilated branch diameter is a controversial indicator for surgical resection. In general, the cytological and histological analyses of specimens obtained using EUS-FNA, EUS-FNAB, and via the papilla are also very important for diagnosing IPMN. However, the sensitivity of the cytological evaluation of pancreatic juice for differentiating benign and malignant forms is reported to be as low as 13%e47% [15,16]. Furthermore, the aspiration of fluid directly from the dilated pancreatic ducts using EUS-FNA does not improve the sensitivity (21%) [34]. In our study, the sensitivity of the cytological analysis of pancreatic juice from the transpapilla was only 36.1% (data not shown). Another study has revealed that the histological analysis of EUS-FNA biopsy or transpapillary biopsy specimens had an accuracy of 91% for the positive diagnosis of malignant IPMNs [35]. The authors did not report any severe complications from this procedure. In this previous study, all the patients had symptomatic advanced IPMNs, and most of the cases were radiologically diagnosed as main duct IPMNs and were invasive pancreatic cancers derived from IPMNs. In the present study, most of the branch duct IPMNs contained very small nodules, and the branch duct connecting to the main duct was very narrow. Thus, the target of the EUS-FNA biopsy was a very small lesion of less than 10 mm, and the biopsy cup could not be passed through the main pancreatic duct to the nodule. Consequently, cytological and histological analyses of biopsy specimens are technically very difficult to perform for branch duct IPMNs. Further advanced endoscopic tools are required for the use of these methods to become widely used for the diagnosis of branch duct IPMNs. In our study, the EUS findings were highly accurate (88.8%e 91.7%) for the discrimination of low-risk and high-risk IPMNs. The EUS findings of the mural nodule (height > 10 mm, lateral spread > 15 mm) are reliable predictors of a high risk of malignancy among branch duct IPMNs. This study was limited by relatively
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
small numbers which precluded more detailed analysis and we are planning to perform a larger retrospective study to strengthen the data. This report is none-the-less the first to indicate that the width diameter of the tumor as visualized using EUS is a good predictor of a high risk of malignancy among branch duct IPMNs.
[16]
[17]
Conflicts of interest No conflicts of interest exist. Acknowledgments None of the authors have been funded by any Foundation or other nongovernmental source. There are no financial disclosures from any authors.
[18]
[19]
[20]
[21]
References [1] Hruban RH, Takaori K, Klimstra DS, Adsay N, Albores-Saavedra J, Fukushima N, et al. An illustrated consensus on the classification of pancreatic intraepithelial neoplasia and intraductal papillary mucinous neoplasms. Am J Surg Pathol 2004;28:977e87. [2] Hruban RH, Pitman MB, Klimstra DS. AFIP atlas of tumor pathology fourth series Fascicle 6: tumor of the pancreas. Washington DC: American Registry of Pathology; 2007. 75e110. [3] Tanaka M. Intraductal papillary mucinous neoplasm of the pancreas: diagnosis and treatment. Pancreas 2004;28:282e8. [4] Kobari M, Egawa S, Shibuya K, Shimamura H, Sunamura M, Takeda K, et al. Intraductal papillary mucinous tumors of the pancreas comprise two clinical subtypes. Arch Surg 1999;134:1131e6. [5] Yamaguchi K, Ogawa Y, Chijiiwa K, Tanaka Y. Mucin-hypersecreting tumors of the pancreas: assessing the grade of malignancy preoperatively. Am J Surg 1996;171:427e31. [6] Terri B, Ponsot P, Paye F, Hammel P, Sauvanet A, Molas G, et al. Intraductal papillary mucinous tumors of the pancreas confined to secondary ducts show less aggressive pathologic features as compared with those involving the main. Am J Surg Pathol 2000;24:1372e7. [7] Tanaka M, Kobayashi K, Mizumoto K, Yamaguchi K. Clinical aspects of intraductal papillary mucinous neoplasm of the pancreas. J Gastroenterol 2005;40:669e75. [8] Sohn TA, Yeo CJ, Cameron JL, Hruban RH, Fukushima N, Campbell KA, et al. Intraductal papillary mucinous neoplasms of the pancreas: an updated experience. Am Surg 2004;239:788e97. [9] Doi R, Fujimoto K, Wasa M, Imamura M. Surgical management of intraductal papillary mucinous tumor of the pancreas. Surgery 2002;132:80e5. [10] Rodriguez JR, Salvia R, Crippa S, Warshaw AL, Bassi C, Falconi M, et al. Branchduct intraductal papillary mucinous neoplasms: observations in 145 patients who underwent resection. Gastroenterology 2007;133:72e9. [11] Jang JY, Kim SW, Ahn YJ, Yoon YS, Choi MG, Lee KU, et al. Multicenter analysis of clinicopathologic features of intraductal papillary mucinous tumor of the pancreas: is it possible to predict the malignancy before surgery? Ann Surg Oncol 2005;12:124e32. [12] Salvia R, Crippa S, Falconi M, Bassi C, Guarise A, Scarpa A, et al. Branch-duct intraductal papillary mucinous neoplasms of the pancreas: to operate or not to operate? Gut 2007;56:1086e90. [13] Pelez-Luna M, Chari ST, Smyrk TC, Takahashi N, Clain JE, Levy MJ, et al. Do consensus indications for resection in branch duct intraductal papillary mucinous neoplasm predict malignancy? A study of 147 patients. Am J Gastroenterol 2007;102:1759e64. [14] Tanaka M, Chari S, Adsay V, Femandez-del Castillo C, Falconi M, Shimizu M, et al. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology 2005;6:17e32. [15] Hara T, Yamaguchi T, Ishihara T, Tsuyuguchi T, Kondo F, Kato K, et al. Diagnosis and patient management of intraductal papillary-mucinous tumor of the
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29] [30] [31]
[32]
[33]
[34]
[35]
145
pancreas by using peroral pancreatoscopy and intraductal ultrasonography. Gastroenterology 2002;122:34e43. Inoue H, Tsuchida A, Kawasaki Y, Fujimoto Y, Yamasaki S, Kajiyama G. Preoperative diagnosis of intraductal papillary-mucinous tumors of the pancreas with attention to telomerase activity. Cancer 2001;91: 1429e34. Salvia R, Fernandez-del Castillo C, Bassi C, Thayer SP, Falconi M, Mantovani W, et al. Main-duct intraductal papillary mucinous neoplasms of the pancreas: clinical predictors of malignancy and long-term survival following resection. Ann Surg 2004;239:678e87. Sugiyama M, Izumisato Y, Abe N, Masaki T, Mori T, Atomi Y. Predictive factors for malignancy in intraductal papillary-mucinous tumors of the pancreas. Br J Surg 2003;90:1244e9. Nara S, Onaya H, Hiraoka N, Shimada K, Sano T, Sakamoto Y. Preoperative evaluation of invasive and noninvasive intraductal papillary-mucinous neoplasms of the pancreas: clinical, radiological, and Pathological Analysis of 123 cases. Pancreas 2009;38:8e16. Kobayashi G, Fujita N, Noda Y, Ito K, Horaguchi J, Takasawa O, et al. Mode of progression of intraductal papillary-mucinous tumor of the pancreas: analysis of patients with follow-up by EUS. J Gastroenterol 2005;40:744e51. Kubo H, Chijiiwa Y, Akahoshi K, Hamada S, Sumi T, Takashima M, et al. Intraductal papillary-mucinous tumors of the pancreas: differential diagnosis between benign and malignant tumors by endoscopic ultrasonography. Am J Gastroenterol 2001;96:1429e34. Buscaglia JM, Giday SA, Kantsevoy SV, Jagannath SB, Magno P, Assumpcao L, et al. Patient-and cyst-related factors for improved prediction of malignancy within cystic lesions of the pancreas. Pancreatology 2009;9:631e8. Cellier C, Cuillerier E, Palazzo L, Rickaert F, Flejou JF, Napoleon B, et al. Intraductal papillary and mucinous tumors of the pancreas: accuracy of preoperative computed tomography, endoscopic retrograde pancreatography and endoscopic ultrasonography, and long-term outcome in a large surgical series. Gastrointest Endosc 1998;47:42e9. Barbe L, Ponsot P, Vilgram V, Terris B, Flejou JF, Sauvanet A, et al. Intraductal papillary mucinous tumors of the pancreas. Clinical and morphological aspects in 30 patients. Gastroenterol Clin Biol 1997;21:278e86. Yamao K, Ohashi K, Nakamura T, Suzuki T, Watanabe Y, Shimizu Y, et al. Evaluation of various imaging methods in the differential diagnosis of intraductal papillary-mucinous tumor (IPMT) of the pancreas. Hepatogastroenterology 2001;48:962e6. Jang JY, Kim SW, Lee SE, Yang SH, Lee KU, Kim SC, et al. Treatment guideline for branch duct type intraductal papillary mucinous neoplasms of the pancreas: when can we operate or observe? Ann Surg 2008;15: 199e205. Woo SM, Ryu JK, Lee SH, Yoon WJ, Kim YT, Yoon B. Branch duct intraductal papillary mucinous neoplasms in a retrospective series of 190 patients. Br J Surg 2009;96:405e11. Okabayashi T, Kobayashi M, Nishimori I, Sugimoto T, Namikawa T, Okamoto K, et al. Clinicopathological features and medical management of intraductal papillary mucinous neoplasms. J Gastroenterol Hepatol 2006;21:462e7. Kanno A, Satoh K, Hirota M, Hamada S, Umino J, Itoh H, et al. J Gastroenterol 2010;45:952e9. General rules for the study of pancreatic cancer. Japan Pancreas Society. 6th ed. Tokyo: Kanehara; 2009. Sadakari Y, Ienaga J, Kobayashi K, Miyasaka Y, Takahata S, Nakamura M, et al. Cyst size indicates malignant transformation in branch duct intraductal papillary mucinous neoplasm of the pancreas without mural nodules. Pancreas 2010;39:232e6. Bernard P, Scoazec JY, Joubert M, Kahn X, Le Borger J, Berger F, et al. Intraductal papillary-mucinous pathological examination of 53 cases. Arch Surg 2002;137:1274e8. Schmidt CM, White PB, Waters JA, Yiannoutsos CT, Cummings OW, Baker M, et al. Intraductal papillary mucinous neoplasms: predictors of malignant and invasive pathology. Ann Surg 2007;246:644e51. Brandwein SL, Farrell JJ, Centeno BA, Brugge WR. Detection and tumor staging of malignancy in cystic, intraductal, and solid tumors of the pancreas by EUS. Gastrointest Endosc 2001;53:722e7. Maire F, Couveland A, Hammel P, Ponsot P, Palazzo L, Aubert A, et al. Intraductal papillary mucinous tumors of the pancreas: the preoperative value cytologic and histopathologic diagnosis. Gastrointest Endosc 2003;58: 701e6.
Contents lists available at SciVerse ScienceDirect
Pancreatology journal homepage: www.elsevier.com/locate/pan
Original article
Endoscopic ultrasonographic findings predict the risk of carcinoma in branch duct intraductal papillary mucinous neoplasms of the pancreas Noritoshi Kobayashi a, *, Kazuya Sugimori b, Takeshi Shimamura a, Kunihiro Hosono a, Seitaro Watanabe a, Shingo Kato a, Michio Ueda c, Itaru Endo c, Yoshiaki Inayama d, Shin Maeda a, Atsushi Nakajima a, Kensuke Kubota a a
Gastroenterology Division, Yokohama City University Graduate School of Medicine, 3-9, Fuku-ura, Kanazawa-ku, Yokohama 236-0004, Japan Gastroenterological Center, Yokohama City University Medical Center, Yokohama, Japan Gastroenterological Surgery Division, Yokohama City University Graduate School of Medicine, Yokohama, Japan d Pathological Division, Yokohama City University Graduate School of Medicine, Yokohama, Japan b c
a b s t r a c t Keywords: Intraductal papillary mucinous neoplasm of pancreas Endoscopic ultrasonography Branch duct type Diagnosis Mural nodule
Background: The preoperative diagnosis of branch duct intraductal papillary mucinous neoplasm (IPMN) of the pancreas can be very difficult, since low-risk and high-risk lesions can be difficult to differentiate even after cytological analysis. The purpose of this study was to evaluate the preoperative diagnostic value of endoscopic ultrasonography (EUS) in differentiating low-risk and high-risk IPMNs. Methods: We retrospectively identified 36 patients who underwent preoperative EUS for branch duct IPMNs. The pathological diagnosis after surgical resection was low-grade dysplasia (n ¼ 26), moderate dysplasia (n ¼ 1), high-grade dysplasia or carcinoma in situ (n ¼ 5), and invasive carcinoma (n ¼ 4). We divided the patients into two groups: low risk (low-grade dysplasia or moderate dysplasia) and high risk (high-grade dysplasia or carcinoma). We focused on the diameter of the cystic dilated branch duct, the main pancreatic duct, and the mural nodule as measured using the EUS findings. Results: The cystic dilated branch duct diameter (31.5 mm vs. 41.9 mm, P ¼ 0.0225) was significantly correlated with low-risk and high-risk IPMNs, but the main pancreatic duct diameter (5.37 mm vs. 5.44 mm, P ¼ 0.9418) was not significantly correlated with the low-risk and high-risk IPMNs. The mural nodule diameter of the papillary protrusions (4.3 mm vs. 16.4 mm, P < 0.0001) and the width diameter of the mural nodule (5.7 mm vs. 23.2 mm, P < 0.0001) were significantly correlated with low-risk and highrisk IPMNs. Conclusions: The mural nodule of papillary protrusions diameter and width diameter observed using EUS was a reliable preoperative diagnostic finding capable of distinguishing low-risk and high-risk IPMNs. Copyright Ó 2012, IAP and EPC. Published by Elsevier India, a division of Reed Elsevier India Pvt. Ltd. All rights reserved.
1. Introduction Intraductal papillary mucinous neoplasms (IPMNs) are characterized by the intraductal proliferation of mucin-producing epithelial cells and the cystic dilatation of pancreatic ducts [1]. IPMNs are thought to progress from low-grade, moderate, and high-grade dysplasia (carcinoma in situ [CIS]) to invasive carcinoma originating as intraductal papillary mucinous carcinoma (IPMC) [2]. Although IPMNs are currently thought to have the potential to become malignant [3], whether all IPMNs have such potential and
* Corresponding author. Tel.: þ81 45 787 2640; fax: þ81 45 784 3546. E-mail address: [email protected] (N. Kobayashi).
the time course for progression remains unknown. Therefore, the indications for surgery based on the biological behavior of this tumor are currently of great concern [4,5]. However, preoperative diagnoses capable of distinguishing benign from malignant IPMNs are challenging to make. IPMNs have been subdivided into main duct IPMNs and branch duct IPMNs based on imaging studies or histological features [1,6]. Main duct IPMNs and branch duct IPMNs are associated with a significantly different prevalence of cancer, ranging from 57% to 92% and from 6% to 46%, respectively. Consequently, this classification has prognostic implications [1,4,6e9]. In recent consensus guidelines for IPMNs, main duct IPMNs have been regarded as having a higher malignant potential, and surgical resection has been recommended [6,8e14]. However, the treatment strategy for
1424-3903/$ e see front matter Copyright Ó 2012, IAP and EPC. Published by Elsevier India, a division of Reed Elsevier India Pvt. Ltd. All rights reserved. doi:10.1016/j.pan.2011.12.008
142
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
branch duct IPMNs has not been determined. The International Consensus Guidelines have put forward an algorithm for surgical management that is based on cyst size, patient symptoms, and a high-risk appearance (mural nodules, thick septum or positive cytology) [14]. Recent advances in imaging modalities have led to the incidental identification of many asymptomatic cystic lesions, and preoperative histological or cytological diagnoses are not always sufficient to determine the surgical indications for this tumor [15,16]. Some previous series have demonstrated a positive correlation between size and malignant IPMN pathology [17e19]. Many imaging modalities for the examination of IPMNs have also been used as preoperative diagnostic tools. In particular, endoscopic ultrasonography (EUS) is regarded as the most reliable modality for examining IPMNs [20,21]. We performed EUS for preoperative diagnosis in 36 cases of branch duct IPMNs. We then compared the EUS findings and the final pathological diagnoses retrospectively. The results of this study suggest that the EUS findings of branch duct IPMNs may enable a more appropriate selection of patients for operative resection. 2. Materials and methods Patients with IPMNs were identified by searching the Yokohama City University Hospital databases for patients treated between 1992 and 2010. There are no Caucasians in my study. Two hundred and thirty-eight patients with imaging and/or histological findings consistent with IPMNs were analyzed in this study. We subclassified the IPMNs as either main duct or branch duct IPMNs according to the preoperative imaging findings and the postoperative histological findings. In this study, we did not include any mixed-type IPMNs because pure branch duct IPMNs are very rare and many branch IPMNs are reclassified as mixed type [14,20]. Consequently, we defined the subclassification according to the dominant lesions of the papillary epithelial cells. Branch duct IPMNs were diagnosed when EUS and imaging findings (CT, MRCP, or ERCP) showed one or more pancreatic cysts communicating with the main pancreatic duct. We diagnosed 41 patients as having main duct IPMNs and 197 patients as having branch duct IPMNs. Pancreatic resection was indicated for main duct IPMNs and for branch duct IPMNs with symptomatic lesions and/or lesions suspected of being malignant. We defined malignant branch duct IPMNs as those in which a cytological analysis of the pancreatic juice collected by the transpapillary route was class IV or V and/ or the cystic dilated branch diameter was over 30 mm with a mural nodule based on EUS findings. These indications were based on the results of a previously reported study and the consensus guidelines [11,14]. We used EUS (UM20, UM2000; Olympus Japan, Tokyo) to obtain a preoperative diagnosis in 36 cases. We measured the height and width of the mural nodules, the main pancreatic duct diameter, and the cystic dilated branch diameter based on EUS findings. We defined the “height of the mural nodules” as the maximum length perpendicular to the cystic dilated branch wall and the “width of the mural nodules” as the maximum length along the walls of the dilated branches and perpendicular to the maximum height of the mural nodules (Fig. 1A). “Cystic dilated branch” and “main pancreatic duct” were defined as the maximum sizes based on the EUS findings. In situations where a multilocular cyst was present, we measured the maximum size of all the cystic components. The EUS findings were checked by two expert endoscopists (K.K, N.K). All the IPMNs were pathologically reexamined, and the diagnosis of IPMN was confirmed. Surgically resected specimens were fixed in 10% formalin and cut into serial 5-mm-thick slices, horizontally in the pancreas head and sagittally in the pancreas body and tail. All the sections were stained with hematoxylin and eosin
Fig. 1. Endoscopic ultrasonography clearly shows the mural nodule in a cystic dilated branch of the pancreas (A). The solid line represents the width of the mural nodule. The dotted line represents the height of the mural nodule. Macroscopic findings of the mural nodule in the cystic dilated branch (B). The macroscopic width and height of the mural nodule were nearly equal to the EUS findings. This tumor was pathologically diagnosed as a high-grade dysplasia (carcinoma in situ).
for pathological examination. We also measured the height and lateral spread of the mural nodules pathologically after they had been fixed in 10% formalin; the same definitions as those described above were used (Fig. 1B). We also examined the pathological grade according to the WHO classification (IPMN with low-grade dysplasia, IPMN-LD; IPMN with moderate dysplasia, IPMN-MD; IPMN with high-grade dysplasia, IPMN-HD; invasive carcinoma originating IPMC, IC-IPMC). The statistical analysis was performed using the Stat View program (SAS Institute, Cary, NC, USA). Continuous data were presented as the mean SD. The size cut-off values for differentiating benign and malignant IPMNs were investigated by constructing receiver operating characteristic (ROC) curves and calculating the area under the curves. Continuous variables were analyzed using the Student t-test. The c2 test was used for categorical variables. P values of less than 0.05 were considered statistically significant. The relations were also analyzed using the correlation coefficients. 3. Results Thirty-six cases (21 men, 15 women) were analyzed in this study. The postoperative pathological examination revealed 26 cases of IPMN-LD, one case of IPMN-MD, five cases of IPMN-HD, and four cases of IC-IPMC. Twenty-four asymptomatic cases were discovered incidentally. Eleven cases initially presented with abdominal pain, and one case presented with back pain. One case presented with a body weight loss of 3 kg during a 6-month period. The median age at the time of the initial diagnosis was 66 years. Malignant diseases of extra-pancreatic organs were present in four cases. Diabetes mellitus (DM) was present in four cases. We defined IPMN-LD and IPMN-MD as low-risk IPMNs and IPMN-HD and IC-IPMC as high-risk IPMNs. First, we compared the low-risk IPMNs and the high-risk IPMNs with regard to the clinical
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
143
Table 1 Clinical characteristics of the patients with branch duct IPMNs.
Total IPMNs Low-risk IPMNs (IPMN-LD/IPMN-MD) High-risk IPMNs (IPMN-HD; CIS/IC-IPMC) P value (Low-risk vs High-risk)
Pathological diagnosis (IPMN-LD/IPMN-MD/IPMN-HD; CIS/IC-IPMC)
Gender (male/female)
Age (median)
Location (h/b/t/m)
Symptom
Extra-pancreatic cancer
DM
36 (26/1/5/4) 27 (26/1) 9 (5/4)
36 (21/15) 27 (14/13) 9 (7/2)
66 (36e78) 66 (36e77) 66 (49e78)
(15/13/2/6) (12/10/2/3) (3/3/0/3)
10 (27.8%) 8 (29.6%) 2 (22.2%)
4 (11.1%) 3 (11.1%) 1 (11.1%)
4 (11.1%) 3 (11.1%) 1 (11.1%)
0.1851
0.9303
0.9741
0.925
0.925
IPMN-LD, IPMN with low-grade dysplasia; IPM-MD, IPMN with moderate dysplasia; IPMN-HD, IPMN with high-grade dysplasia; CIS, carcinoma in situ; IC-IPMC, Invasive carcinoma originating in IPMC; h, head; b, body; t, tail; m, multiple lesions; DM, diabetes mellitus.
Table 2 EUS findings of the branch duct IPMNs. EUS findings
Main pancreatic duct (mm)
Cystic branch duct (mm)
Papillary protrusions of the nodule (mm)
Width of the nodule (mm)
Total IPMNs Low-risk IPMNs (IPMN-LD/IPMN-MD) High-risk IPMNs (IPMN-HD; CIS/IC-IPMC) P value (Low-risk vs High-risk)
5.39 (3e13) 5.37 (3e11) 5.44 (3e13) P ¼ 0.9418
34.1 (10e70) 31.5 (10e53) 41.9 (25e70) P ¼ 0.0225
7.39 (0e35) 4.3 (0e15) 16.4 (10e35) P < 0.0001
10.1 (0e60) 5.7 (0e22) 23.2 (8e60) P < 0.0001
EUS, endoscopic ultrasonography.
background (Table 1). The sex ratio and the median age at the time of the initial diagnosis were not significantly different between the two groups (male/female, 14/13 vs. 7/2, P ¼ 0.1851; 64.8 vs. 65.1 years, P ¼ 0.9303). The incidence of symptomatic episodes was not significantly different between the two groups (P ¼ 0.9741), nor was the rate of extra-pancreatic malignant disease (P ¼ 0.9250). The prevalence of DM was not significantly different between the two groups (P ¼ 0.9250). We next compared the EUS findings for low-risk and high-risk IPMNs. The cystic dilated branch duct diameter (31.5 mm vs. 41.9 mm, P ¼ 0.0225) was significantly correlated with low-risk and high-risk IPMNs (Table 2). However, the main pancreatic duct diameter (5.37 mm vs. 5.44 mm, P ¼ 0.9418) was not significantly correlated with low-risk and high-risk IPMNs. The mural nodule diameter of the papillary protrusions (4.3 mm vs. 16.4 mm, P < 0.0001) and the width diameter of the mural nodule (5.7 mm vs. 23.2 mm, P < 0.0001) were significantly correlated with lowrisk and high-risk IPMNs (Table 2). We next examined the ROC curve and the area under the curve (AUC). The mural nodule diameter of the papillary protrusion (0.957) and the width diameter (0.947) were excellent clinical parameters for distinguishing low-risk from high-risk IPMNs (Fig. 2). If a cut-off point of 10 mm was adopted for papillary protrusion, the sensitivity was 100%, the specificity was 85.2%, and the accuracy was 88.9%. If a cut-off point of 15 mm was adopted for lateral spread, the sensitivity was 88.9%, the specificity was 92.6%, and the accuracy was 91.7% (Table 3). Finally, we compared the macroscopic pathological size of the mural nodule with the EUS findings. The correlation coefficients of the diameter of papillary protrusion (r ¼ 0.840) and the width diameter (r ¼ 0.868) were highly correlated with the EUS findings and the macroscopic pathological findings (Fig. 3A,B).
present study, however, these factors were not correlated with high-risk IPMNs. Previous studies included many symptomatic cases (about 70%). Consequently, these tumors may have been more advanced, and the patients’ backgrounds may have differed from those in the present study. In our study, the final pathological and radiological diagnoses were only branch duct-type IPMNs, and many cases were asymptomatic early-stage tumors (symptomatic cases accounted for only 10% of the series). In practice, many cystic lesions in the pancreas are discovered incidentally, so our data may reflect normal clinical practice better and be more reliable for identifying malignant disease in asymptomatic branch duct IPMNs. The present study focused on the value of preoperative EUS findings, especially for mural nodules. In this study, the accuracy ranged from 88.9% to 91.3%, depending on the mural nodule diameter. In previous studies, the accuracy of EUS for discriminating benign and malignant forms varied from 65% to 94% [15,23e25]. According to many past reports, the accuracy of other modalities was 75%e78% for MDCT, 78%e86% for MRI, 79%e83% for ERP, and an impressive 88% for combined intraductal US and peroral pancreatoscopy [15,23e25]. Thus, EUS may be slightly superior to other modalities for distinguishing low-risk and highrisk IPMNs.
4. Discussion IPMNs of the pancreas are problematic because of the difficulty in preoperative diagnosis. The clinical behavior and operative indications of branch duct IPMNs are very difficult to predict. In the present study, no predictive factors of high-risk IPMNs other than imaging findings were found among the patient-related factors that were examined. According to past reports, a patient age of greater than 50 years, Caucasian ancestry, and body weight loss are independent risk factors for malignant IPMNs [22]. In the
Fig. 2. ROC curve and the area under the curve (AUC) for distinguishing low-risk from high-risk IPMNs according to the size of the mural nodule. The mural nodule diameter of the papillary protrusion (0.957) and the width (0.947) of the mural nodule were excellent clinical parameters for distinguishing low-risk and high-risk IPMNs.
144
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
Table 3 Accuracy of imaging diagnosis for the branch duct IPMNs by diameter of mural nodule as visualized using EUS.
Papillary protrusions of the nodule >10 mm Lateral spread of the nodule >15 mm
Sensitivity
Specificity
Predictive positive value
Predictive Negative value
Accuracy
100.00% 88.90%
85.20% 92.60%
69.30% 80.00%
85.20% 96.20%
88.90% 91.70%
Some previous reports of large series of branch duct IPMNs have focused on the cystic dilated branch diameter, mural nodule, main pancreatic duct diameter, and tumor size based on CT findings. Jang’s large study (n ¼ 138) revealed that a tumor size of 20 mm and/or papillary protrusions were indications for surgical treatment [26], while Rodriguez (n ¼ 145) revealed that surgical treatment should be performed for tumors with a size of greater than 30 mm, symptomatic tumors, or mural nodules [10]. Woo’s study (n ¼ 190) reported that observation is recommended for patients with branch duct IPMNs smaller than 30 mm with no suspicious features visible upon CT imaging [27]. The presence of large tumors on CT findings was regarded as an obvious indication of malignancy. Cases in which the indications for surgical treatment were difficult to decide included equivocal small tumors visible on CT findings. EUS was very useful for identifying these small lesions as branch duct IPMNs. Few studies have examined the preoperative diagnostic value of EUS findings for branch duct IPMNs. Okabayashi reported that branch duct IPMNs with a cystic size of greater than 30 mm and a mural nodule height of greater than 5 mm on EUS findings should be resected [28]. Kubo reported that EUS findings showing large cystic lesions (>40 mm) with irregular thick septa and a large mural nodule (>10 mm) were strongly suggestive of malignancy [21]. And Kanno reported that EUS findings showing large mural nodules (>6.5 mm) were suggestive of invasive IPMN [29]. Kobayashi reported that patients with branch duct-type large cysts with septum-like strictures should undergo surgery, while those with small papillary protrusions have a benign course on EUS
Fig. 3. Comparison of the macroscopic pathological size of the mural nodule with the EUS findings. The correlation coefficients for the diameter of papillary protrusion (A) (r ¼ 0.840) and the width of the mural nodule (B) (r ¼ 0.868) were highly correlated with the EUS findings and the macroscopic pathological findings. High-risk IPMNs are shown as black circles.
[20]. In our study, papillary protrusions (>10 mm), and the width diameter (>15 mm) were highly accurate (91.7%) predictors of high-risk IPMNs. Our study was retrospective and limited by small sample size. Nevertheless, our criteria are simple and available one, so any pancreatic centre should be able to use them. In our study, we defined the width diameter as the measurement perpendicular to the maximum height of the tumor nodule. Of course, in this study, “width” was not equal to the lateral spread of the microscopic pathological findings. Nevertheless, the width diameter of the nodule on the EUS findings was correlated with the macroscopic pathological findings. According to general rules for the study of pancreatic cancer in Japan, the tumor diameter of IPMNs is defined as the longest diameter of the lateral spread of the tumor [30]. So, the width diameter of the mural nodule was a reasonable parameter for defining the size of the mural nodule. In some cases of IC-IPMC, there was a difference between the EUS measurement and the pathological measurement of mural nodule size. The mural nodule and invasive component of IPMCs may be difficult to distinguish using EUS in some cases. In our study, the cystic dilated branch diameter was also useful for distinguishing low-risk and high-risk IPMNs. Some reports, including the International Consensus guidelines, suggest that a diameter of 30 mm or 40 mm was important for discriminating between benign and malignant tumors [14,31], but many recent reports have indicated that the cystic dilated branch diameter does not reflect tumor malignancy [32,33]. The cystic dilated branch diameter is a controversial indicator for surgical resection. In general, the cytological and histological analyses of specimens obtained using EUS-FNA, EUS-FNAB, and via the papilla are also very important for diagnosing IPMN. However, the sensitivity of the cytological evaluation of pancreatic juice for differentiating benign and malignant forms is reported to be as low as 13%e47% [15,16]. Furthermore, the aspiration of fluid directly from the dilated pancreatic ducts using EUS-FNA does not improve the sensitivity (21%) [34]. In our study, the sensitivity of the cytological analysis of pancreatic juice from the transpapilla was only 36.1% (data not shown). Another study has revealed that the histological analysis of EUS-FNA biopsy or transpapillary biopsy specimens had an accuracy of 91% for the positive diagnosis of malignant IPMNs [35]. The authors did not report any severe complications from this procedure. In this previous study, all the patients had symptomatic advanced IPMNs, and most of the cases were radiologically diagnosed as main duct IPMNs and were invasive pancreatic cancers derived from IPMNs. In the present study, most of the branch duct IPMNs contained very small nodules, and the branch duct connecting to the main duct was very narrow. Thus, the target of the EUS-FNA biopsy was a very small lesion of less than 10 mm, and the biopsy cup could not be passed through the main pancreatic duct to the nodule. Consequently, cytological and histological analyses of biopsy specimens are technically very difficult to perform for branch duct IPMNs. Further advanced endoscopic tools are required for the use of these methods to become widely used for the diagnosis of branch duct IPMNs. In our study, the EUS findings were highly accurate (88.8%e 91.7%) for the discrimination of low-risk and high-risk IPMNs. The EUS findings of the mural nodule (height > 10 mm, lateral spread > 15 mm) are reliable predictors of a high risk of malignancy among branch duct IPMNs. This study was limited by relatively
N. Kobayashi et al. / Pancreatology 12 (2012) 141e145
small numbers which precluded more detailed analysis and we are planning to perform a larger retrospective study to strengthen the data. This report is none-the-less the first to indicate that the width diameter of the tumor as visualized using EUS is a good predictor of a high risk of malignancy among branch duct IPMNs.
[16]
[17]
Conflicts of interest No conflicts of interest exist. Acknowledgments None of the authors have been funded by any Foundation or other nongovernmental source. There are no financial disclosures from any authors.
[18]
[19]
[20]
[21]
References [1] Hruban RH, Takaori K, Klimstra DS, Adsay N, Albores-Saavedra J, Fukushima N, et al. An illustrated consensus on the classification of pancreatic intraepithelial neoplasia and intraductal papillary mucinous neoplasms. Am J Surg Pathol 2004;28:977e87. [2] Hruban RH, Pitman MB, Klimstra DS. AFIP atlas of tumor pathology fourth series Fascicle 6: tumor of the pancreas. Washington DC: American Registry of Pathology; 2007. 75e110. [3] Tanaka M. Intraductal papillary mucinous neoplasm of the pancreas: diagnosis and treatment. Pancreas 2004;28:282e8. [4] Kobari M, Egawa S, Shibuya K, Shimamura H, Sunamura M, Takeda K, et al. Intraductal papillary mucinous tumors of the pancreas comprise two clinical subtypes. Arch Surg 1999;134:1131e6. [5] Yamaguchi K, Ogawa Y, Chijiiwa K, Tanaka Y. Mucin-hypersecreting tumors of the pancreas: assessing the grade of malignancy preoperatively. Am J Surg 1996;171:427e31. [6] Terri B, Ponsot P, Paye F, Hammel P, Sauvanet A, Molas G, et al. Intraductal papillary mucinous tumors of the pancreas confined to secondary ducts show less aggressive pathologic features as compared with those involving the main. Am J Surg Pathol 2000;24:1372e7. [7] Tanaka M, Kobayashi K, Mizumoto K, Yamaguchi K. Clinical aspects of intraductal papillary mucinous neoplasm of the pancreas. J Gastroenterol 2005;40:669e75. [8] Sohn TA, Yeo CJ, Cameron JL, Hruban RH, Fukushima N, Campbell KA, et al. Intraductal papillary mucinous neoplasms of the pancreas: an updated experience. Am Surg 2004;239:788e97. [9] Doi R, Fujimoto K, Wasa M, Imamura M. Surgical management of intraductal papillary mucinous tumor of the pancreas. Surgery 2002;132:80e5. [10] Rodriguez JR, Salvia R, Crippa S, Warshaw AL, Bassi C, Falconi M, et al. Branchduct intraductal papillary mucinous neoplasms: observations in 145 patients who underwent resection. Gastroenterology 2007;133:72e9. [11] Jang JY, Kim SW, Ahn YJ, Yoon YS, Choi MG, Lee KU, et al. Multicenter analysis of clinicopathologic features of intraductal papillary mucinous tumor of the pancreas: is it possible to predict the malignancy before surgery? Ann Surg Oncol 2005;12:124e32. [12] Salvia R, Crippa S, Falconi M, Bassi C, Guarise A, Scarpa A, et al. Branch-duct intraductal papillary mucinous neoplasms of the pancreas: to operate or not to operate? Gut 2007;56:1086e90. [13] Pelez-Luna M, Chari ST, Smyrk TC, Takahashi N, Clain JE, Levy MJ, et al. Do consensus indications for resection in branch duct intraductal papillary mucinous neoplasm predict malignancy? A study of 147 patients. Am J Gastroenterol 2007;102:1759e64. [14] Tanaka M, Chari S, Adsay V, Femandez-del Castillo C, Falconi M, Shimizu M, et al. International consensus guidelines for management of intraductal papillary mucinous neoplasms and mucinous cystic neoplasms of the pancreas. Pancreatology 2005;6:17e32. [15] Hara T, Yamaguchi T, Ishihara T, Tsuyuguchi T, Kondo F, Kato K, et al. Diagnosis and patient management of intraductal papillary-mucinous tumor of the
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29] [30] [31]
[32]
[33]
[34]
[35]
145
pancreas by using peroral pancreatoscopy and intraductal ultrasonography. Gastroenterology 2002;122:34e43. Inoue H, Tsuchida A, Kawasaki Y, Fujimoto Y, Yamasaki S, Kajiyama G. Preoperative diagnosis of intraductal papillary-mucinous tumors of the pancreas with attention to telomerase activity. Cancer 2001;91: 1429e34. Salvia R, Fernandez-del Castillo C, Bassi C, Thayer SP, Falconi M, Mantovani W, et al. Main-duct intraductal papillary mucinous neoplasms of the pancreas: clinical predictors of malignancy and long-term survival following resection. Ann Surg 2004;239:678e87. Sugiyama M, Izumisato Y, Abe N, Masaki T, Mori T, Atomi Y. Predictive factors for malignancy in intraductal papillary-mucinous tumors of the pancreas. Br J Surg 2003;90:1244e9. Nara S, Onaya H, Hiraoka N, Shimada K, Sano T, Sakamoto Y. Preoperative evaluation of invasive and noninvasive intraductal papillary-mucinous neoplasms of the pancreas: clinical, radiological, and Pathological Analysis of 123 cases. Pancreas 2009;38:8e16. Kobayashi G, Fujita N, Noda Y, Ito K, Horaguchi J, Takasawa O, et al. Mode of progression of intraductal papillary-mucinous tumor of the pancreas: analysis of patients with follow-up by EUS. J Gastroenterol 2005;40:744e51. Kubo H, Chijiiwa Y, Akahoshi K, Hamada S, Sumi T, Takashima M, et al. Intraductal papillary-mucinous tumors of the pancreas: differential diagnosis between benign and malignant tumors by endoscopic ultrasonography. Am J Gastroenterol 2001;96:1429e34. Buscaglia JM, Giday SA, Kantsevoy SV, Jagannath SB, Magno P, Assumpcao L, et al. Patient-and cyst-related factors for improved prediction of malignancy within cystic lesions of the pancreas. Pancreatology 2009;9:631e8. Cellier C, Cuillerier E, Palazzo L, Rickaert F, Flejou JF, Napoleon B, et al. Intraductal papillary and mucinous tumors of the pancreas: accuracy of preoperative computed tomography, endoscopic retrograde pancreatography and endoscopic ultrasonography, and long-term outcome in a large surgical series. Gastrointest Endosc 1998;47:42e9. Barbe L, Ponsot P, Vilgram V, Terris B, Flejou JF, Sauvanet A, et al. Intraductal papillary mucinous tumors of the pancreas. Clinical and morphological aspects in 30 patients. Gastroenterol Clin Biol 1997;21:278e86. Yamao K, Ohashi K, Nakamura T, Suzuki T, Watanabe Y, Shimizu Y, et al. Evaluation of various imaging methods in the differential diagnosis of intraductal papillary-mucinous tumor (IPMT) of the pancreas. Hepatogastroenterology 2001;48:962e6. Jang JY, Kim SW, Lee SE, Yang SH, Lee KU, Kim SC, et al. Treatment guideline for branch duct type intraductal papillary mucinous neoplasms of the pancreas: when can we operate or observe? Ann Surg 2008;15: 199e205. Woo SM, Ryu JK, Lee SH, Yoon WJ, Kim YT, Yoon B. Branch duct intraductal papillary mucinous neoplasms in a retrospective series of 190 patients. Br J Surg 2009;96:405e11. Okabayashi T, Kobayashi M, Nishimori I, Sugimoto T, Namikawa T, Okamoto K, et al. Clinicopathological features and medical management of intraductal papillary mucinous neoplasms. J Gastroenterol Hepatol 2006;21:462e7. Kanno A, Satoh K, Hirota M, Hamada S, Umino J, Itoh H, et al. J Gastroenterol 2010;45:952e9. General rules for the study of pancreatic cancer. Japan Pancreas Society. 6th ed. Tokyo: Kanehara; 2009. Sadakari Y, Ienaga J, Kobayashi K, Miyasaka Y, Takahata S, Nakamura M, et al. Cyst size indicates malignant transformation in branch duct intraductal papillary mucinous neoplasm of the pancreas without mural nodules. Pancreas 2010;39:232e6. Bernard P, Scoazec JY, Joubert M, Kahn X, Le Borger J, Berger F, et al. Intraductal papillary-mucinous pathological examination of 53 cases. Arch Surg 2002;137:1274e8. Schmidt CM, White PB, Waters JA, Yiannoutsos CT, Cummings OW, Baker M, et al. Intraductal papillary mucinous neoplasms: predictors of malignant and invasive pathology. Ann Surg 2007;246:644e51. Brandwein SL, Farrell JJ, Centeno BA, Brugge WR. Detection and tumor staging of malignancy in cystic, intraductal, and solid tumors of the pancreas by EUS. Gastrointest Endosc 2001;53:722e7. Maire F, Couveland A, Hammel P, Ponsot P, Palazzo L, Aubert A, et al. Intraductal papillary mucinous tumors of the pancreas: the preoperative value cytologic and histopathologic diagnosis. Gastrointest Endosc 2003;58: 701e6.