International Journal of Surgery 13 (2015) 49e53
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
Surgical treatment and prognosis of 96 cases of intraductal papillary mucinous neoplasms of the pancreas: A retrospective cohort study Xiaodong Tian, Hongqiao Gao, Yongsu Ma, Yan Zhuang, Yinmo Yang* Department of General Surgery, Peking University First Hospital, 8th Xishiku Street, Xicheng, Beijing 100034, People's Republic of China
h i g h l i g h t s Evaluate our strategy of surgical treatments for different types of IPMN. IPMN has a relative good prognosis. Main duct type and mixed type IPMN should receive surgical treatment. Small branch duct type IPMN without clinical manifestations need no surgery.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 September 2014 Received in revised form 5 November 2014 Accepted 22 November 2014 Available online 25 November 2014
Introduction: The indications, the extent and type of surgery for intraductal papillary mucinous neoplasm (IPMN) are still controversial. This study aimed to investigate clinical manifestation, individualized surgical treatment, and prognosis of IPMN of pancreas. Methods: The clinical data of 96 IPMN cases treated in our hospital between January 2006 and December 2013 were retrospectively analyzed. Among the 96 patients (58 male and 38 female), 46 were main-duct type, 29 were branch-duct type, 21 were mixed type. Pancreatectomy was performed on 78 cases, including pancreaticoduodenectomy on 43 patients, distal pancreatectomy on 25, segmental pancreatectomy on 6, and total pancreatectomy on 4. A regular follow-up without surgical treatment was performed on 18 cases with asymptomatic side branch IPMN less than 3 cm in diameter. Results: The overall postoperative morbidity rate was 33.3%, and there was no perioperative mortality. 46 cases were non-invasive IPMN, 32 cases were invasive IPMN including 14 cases with lymph node metastasis. The five-year survival rate for patients with non-invasive and invasive carcinomas was 96.2% and 35.2%, respectively. The prognosis of invasive cases with lymph node metastasis was significantly worse than those without lymph node metastasis. No progression was found during the follow-up in 18 asymptomatic small branch duct type IPMN patients. Conclusion: IPMN has a relative good prognosis. Main duct type and mixed type IPMN have a higher malignant potential, and should receive surgical treatment. Patients of branch duct type IPMN with lesion 5 mm in diameter, without any other evident cause of obstruction. Cystic pancreatic lesions >5 mm in diameter that communicate with a
non-dilated main pancreatic duct were considered to be BD-IPMN. Mixed type IPMN displayed the aspects of both MD- and BD-IPMN features [6,7]. According to these criteria, 46 patients were diagnosed with MD-IPMN, 29 with BD-IPMN, and 21 with mixed type. The typical CT findings of each type of IPMN were shown in Fig. 1.
2.2. Surgical treatment All the patients received either a surgical procedure or a close surveillance, according to a comprehensive assessment of clinical symptoms, imaging findings, and serum tumor markers (CEA and CA19-9). Indications for surgical treatment were as follows: (1) lesions involving the main duct of pancreas (MDeor mixed type IPMN); (2) BD-IPMN either with lesions >3 cm in diameter, or accompanied by clinical symptoms (abdominal pain or discomfort, jaundice, or apparent weight loss), or with elevated serum level of either CEA or CA19-9. For BD-IPMN patients with lesions 5 ng/mL) and 35.4% (34/96) had an elevated CA19-9 level (>37 U/ mL). We found the differences in clinical characteristics between patients with MDeor mixed type IPMN and those with BD-IPMN (Table 1). Patients with tumor involving the main pancreatic duct had more chances to develop clinical symptoms as well as pancreatic endocrine dysfunction. An elevated serum CA19-9 level was more common in patients with MDeor mixed type IPMN. 3.2. Pathological features All the 78 patients who underwent surgical resections had been diagnosed with IPMN by postoperative pathological analyses, including 46 cases of non-invasive tumor (28 cases of low-grade dysplasia, 11 cases of moderate-grade dysplasia, and 7 cases of high-grade dysplasia) and 32 cases of invasive carcinoma. All resection margins were clear of tumor. There were no cases of lymph node metastasis among patients with non-invasive tumor.
Table 1 Comparison of clinical characteristics between patients with MD or mixed type IPMN and BD-IPMN. Clinical features
MD/mixed type (n ¼ 67)
Among the 32 patients with invasive carcinoma, 14 cases (43.8%) had lymph node metastasis. The comparison of clinical features between non-invasive tumors and invasive carcinomas was shown in Table 2. No significant difference was shown in gender distribution and appearance of symptoms between patients with non-invasive tumor and those with invasive carcinoma. There were significant differences between the two groups in mean age (P ¼ 0.000), tumor size (P ¼ 0.000), as well as CEA level (P ¼ 0.002) or CA19-9 level (P ¼ 0.000). The risk of invasive carcinoma for IPMN patients was significantly high in MD- (45.7%, 21/46) and mixed type cases (33.3%, 7/21), compared with BD-IPMN cases (13.8%, 4/29) (P ¼ 0.017). 3.3. Postoperative complications
2.4. Statistical analysis
Gender (M:F) 35:20 Age 58.6 ± 7.1 Diameter (cm) 2.60 ± 0.85 Clinical symptoms 61 Diabetes 52 Elevated tumor markers CEA＞5 ng/mL 8 CA19-9＞37 U/mL 31
51
BD type (n ¼ 18)
P value
23:18 58.3 ± 5.5 2.42 ± 0.88 11 6
0.455 0.822 0.349 0.000 0.000
2 3
0.458 0.001
A total of 32 postoperative complications occurred in 26 patients (complication rate 33.3%). There was no perioperative death. The postoperative complications included pancreatic fistula in 13 cases (16.7%), delayed gastric emptying in 11 cases (14.1%), biliary fistula in 5 cases (6.4%), intra-peritoneal bleeding in 2 cases (2.56%), and intra-abdominal infection in 1 case (1.28%). All complications were successfully managed with conservative treatments. The mean postoperative hospital stay was 14.4 days. Of the 43 patients who underwent pancreaticoduodenectomy, 27 postoperative complications were found, including pancreatic fistula in 10 patients, delayed gastric emptying in 10 patients, biliary fistula in 4 patients, intra-abdominal bleeding in 2 patients, and intra-abdominal infection in 1 patient. Two of the 25 patients who underwent pancreatic body and tail resection developed pancreatic fistula. Postoperative pancreatic fistula was occurred in 1 of the 6 patients who underwent segmental pancreatectomy. Of the 4 patients undergoing total pancreatectomy, 1 case of delayed gastric emptying and 1 case of biliary fistula were diagnosed. 3.4. Follow-up and survival analysis All the 96 IPMN patients were followed up regularly at an interval of 6 months, with a mean follow-up of 37 months (range: 6e72 months). For the asymptomatic BD-IPMN patients with lesion ＜3 cm in diameter, surgical treatments were introduced into 4 of them during the surveillance, in case of the development of one or more indications of surgery. The mean follow-up for the other 18 asymptomatic small BD-IPMN cases was 34 months (range: 6e65 months), without significant change in the diameter of the lesions. For the 46 postoperative patients with non-invasive tumor, the mean follow-up was 39 months (range: 8e72 months), and the
Table 2 Comparison of clinical features between non-invasive and invasive tumors. Clinical features
Non-invasive tumors (n ¼ 64a)
Gender (M:F) 39:25 Age 56.6 ± 6.2 Diameter (cm) 2.29 ± 0.7 Clinical 46 symptoms IPMN type MD type 25 21 Mixed type 14 BD type 25 Elevated tumor markers CEA＞5 ng/mL 2 CA19-9＞37 U/ 7 mL a
Invasive carcinoma (n ¼ 32)
P value
20:12 62.3 ± 5.9 3.04 ± 0.9 26
0.882 0.000 0.000 0.454 0.017
7 4 8 27
Included the 18 asymptomatic small BD-IPMN patients.
0.002 0.000
52
X. Tian et al. / International Journal of Surgery 13 (2015) 49e53
overall 5-year survival rate was 97.1%. For the 32 patients with invasive carcinoma, the mean postoperative follow-up was 46 months (range: 10e64 months), and the overall 1-, 3-, and 5-year survival rates were 93.7%, 74.7%, and 35.2%, respectively. Among these patients, 1-, 3-, and 5-year survival rates for 18 node-negative cases were 100%, 93.1%, and 56.3%, respectively. Meanwhile, 1-, 3-, and 5-year survival rates for the other 14 node-positive patients were 85.4%, 37.3%, and 0%, respectively. By KaplaneMeier survival curve analysis we observed significant differences in survival between non-invasive cases and invasive cases (P ¼ 0.0001), and between node-negative and node-positive patients (P ¼ 0.0104) (Fig. 2).
Table 3 “High-risk stigmata” and “worrisome features” in BD-IPMN.
4. Discussion
carcinoma, compare with PDACs [11]. Comparison of the prognosis of invasive IPMN patients with that of PDAC patients after radical resection showed that the 5-year survival rates in node-negative patients were 69% for invasive IPMN and 13% for ductal adenocarcinoma, but there was no significant difference in survival rates between groups in node-positive patients [12]. In this study, the 5year survival rate of the 32 patients with invasive cancer was much lower than that of the 46 patients with non-invasive tumor (35.2% vs. 97.1%, P ¼ 0.0001). Meanwhile, the prognosis of node-negative invasive IPMN cases was significantly better than that of nodepositive patients (P ¼ 0.0104). These data indicate that IPMN has a relatively low malignant degree, with a good prognosis in early stage patients, and lymph node metastasis could be a risk factor for worse prognosis in patients with invasive IPMN. IPMN is classified into MD-IPMN, BD-IPMN, and mixed type IPMN, according to the involvement of the pancreatic duct. MDIPMN and mixed type IPMN have significantly higher malignant potential than BD-IPMN [6,7]. Preoperative imaging techniques such as contrast-enhanced CT, MRI with magnetic resonance cholangiopancreatography (MRCP), EUS, and ERCP are sufficient to confirm the diagnosis, classify the IPMN type, and assess the respectability [8]. In the current study, all the 78 resected IPMN patients had been correctly diagnosed before operation. The risk of invasive carcinoma associated with different types of IPMN was 45.7% (21/46) in MD IPMN, 33.3% (7/21) in mixed type IPMN, and 13.8% (4/29) in BD-IPMN. In patients with IPMN involving the main pancreatic duct (MDeor mixed type IPMN), the risk of invasive carcinoma was significantly higher than that in BD-IPMN. In addition to the involvement of main pancreatic duct, the high risk of invasive carcinoma seemed to be associated with higher age, larger size of the lesion, as well as the elevation of serum CEA and CA19-9 levels. The gender of patients and manifestation of the
IPMN belongs to a heterogenous group of pancreatic cystic tumors characterized by the papillary proliferation of mucinproducing cells in the main or branch duct of the pancreas. IPMN occurs typically in elderly people, and is more commonly in men than in women. As relatively slow-growing tumors, more asymptomatic IPMN cases are diagnosed incidentally because of the improvements in imaging techniques. 3.8 If the main pancreatic duct is involved, recurrent chronic pancreatitis could be caused by an obstructed pancreatic duct secondary to increased mucin production, which often manifests as upper abdominal pain, weight loss, diarrhea, or diabetes [8]. Jaundice may occur if the bile duct is invaded or compressed by the tumors. These symptoms have been considered to be associated with a risk of malignancy [5,9]. In the current study, 75% of the 96 IPMN patients had one or more symptoms, which were significantly associated with the involvement of main pancreatic duct (Table 1). However, we observed no correlation between these clinic symptoms and the risk of invasive carcinoma (Table 2). According to the malignant potential, the WHO classified these tumors in categories, ranging from low-grade, intermediate-grade, and high-grade dysplasia (also called “carcinoma in situ”), to invasive carcinoma [8,9]. Surgical resection is the only definitive treatment for IPMN, and complete resection of non-invasive IPMN (low-, intermediate-, and high-grade dysplasia) could achieve an excellent survival rate. On the other hand, the survival rate of invasive IPMN patients is significantly lower than those with noninvasive tumors [10]. There is still no consensus on whether the prognosis of invasive IPMN cases is similar to that of pancreatic ductal carcinomas (PDAC). It is generally concluded that IPMN patients are more often diagnosed at an early stage, even for invasive
High risk stigmata
Worrisome features
Obstructive jaundice in a patient with a Clinical, pancreatitis cystic lesion of the pancreatic head Enhanced solid component Cyst of >3 cm MPD size of >10 mm Thickened enhanced cyst walls MPD size of 5e9 mm Nonenhanced mural nodules Abrupt change in the MPD caliber with distal pancreatic atrophy and lymphadenopathy
Fig. 2. KaplaneMeier survival curves for invasive and non-invasive IPMN patients (left, P ¼ 0.0001) and for invasive carcinoma IPMN with and without lymph node metastasis (right, P ¼ 0.0104).
X. Tian et al. / International Journal of Surgery 13 (2015) 49e53
clinical symptoms indicated no correlation with the risk of invasive carcinoma (Table 2). There is no agreement on the risk factors for the accurate prediction of development of invasive carcinoma from IPMN, except that it is generally agreed that MD- and mixed type IPMN should be resected if the patient is a candidate for surgery [5]. For BD-IPMN patients, the recently updated international consensus guidelines introduced worrisome features in imaging criteria and high-risk stigmata including obstructive jaundice, enhanced solid component, and main pancreatic duct size >10 mm (Table 3). The guidelines recommended that for BD-IPMN with high-risk stigmata, or worrisome features which are confirmed by EUS, resection should be considered in patients fit for surgery [7]. In this series of patients, all MDeor mixed type IPMN and BD-IPMN patients either with lesions >3 cm in diameter, or accompanied by clinical symptoms, or with elevated serum tumor markers, were treated with resections. For the 29 BD-IPMN patients, 7 of them received surgical treatment at the time of diagnosis. Resections were recommended for another 4 patients during the surveillance because of either the enlargement of the lesions or the onset of associated symptoms. Four of the 11 resected BD-IPMN cases were confirmed as invasive carcinomas, and the other 18 asymptomatic small BD-IPMN cases showed no change in the lesions after followup of 6e65 months, indicating that our strategy of surgical treatment is rational for this type of IPMN. However, recent studies reported a high rate of malignancy in “Sendai-negative” BD-IPMN or BD-IPMN smaller than 3 cm, and recommended more aggressive surgical treatment for all IPMN cases, including asymptomatic small BD-IPMN [13,14]. Therefore, the safety of our management for the 18 asymptomatic small BD-IPMN cases needs a long-term inspection. In this study, we retrospectively reviewed the clinical data of 96 IPMN cases during the last 8 years, and evaluated our strategy of surgical treatments for different types of IPMN. Our data confirmed the fact that IPMN has a relatively low malignant potential and a better prognosis, especially in early stage cases, compared with PDAC. Main duct type and mixed type IPMN have higher malignant potential, and should receive active surgical treatment. For BDIPMN patients with lesions >3 cm in diameter, or accompanied by clinical symptoms, or with elevated serum tumor markers, a resection should be recommended because of the high risk of invasive carcinomas. However, patients of BD-IPMN with a
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
Surgical treatment and prognosis of 96 cases of intraductal papillary mucinous neoplasms of the pancreas: A retrospective cohort study Xiaodong Tian, Hongqiao Gao, Yongsu Ma, Yan Zhuang, Yinmo Yang* Department of General Surgery, Peking University First Hospital, 8th Xishiku Street, Xicheng, Beijing 100034, People's Republic of China
h i g h l i g h t s Evaluate our strategy of surgical treatments for different types of IPMN. IPMN has a relative good prognosis. Main duct type and mixed type IPMN should receive surgical treatment. Small branch duct type IPMN without clinical manifestations need no surgery.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 September 2014 Received in revised form 5 November 2014 Accepted 22 November 2014 Available online 25 November 2014
Introduction: The indications, the extent and type of surgery for intraductal papillary mucinous neoplasm (IPMN) are still controversial. This study aimed to investigate clinical manifestation, individualized surgical treatment, and prognosis of IPMN of pancreas. Methods: The clinical data of 96 IPMN cases treated in our hospital between January 2006 and December 2013 were retrospectively analyzed. Among the 96 patients (58 male and 38 female), 46 were main-duct type, 29 were branch-duct type, 21 were mixed type. Pancreatectomy was performed on 78 cases, including pancreaticoduodenectomy on 43 patients, distal pancreatectomy on 25, segmental pancreatectomy on 6, and total pancreatectomy on 4. A regular follow-up without surgical treatment was performed on 18 cases with asymptomatic side branch IPMN less than 3 cm in diameter. Results: The overall postoperative morbidity rate was 33.3%, and there was no perioperative mortality. 46 cases were non-invasive IPMN, 32 cases were invasive IPMN including 14 cases with lymph node metastasis. The five-year survival rate for patients with non-invasive and invasive carcinomas was 96.2% and 35.2%, respectively. The prognosis of invasive cases with lymph node metastasis was significantly worse than those without lymph node metastasis. No progression was found during the follow-up in 18 asymptomatic small branch duct type IPMN patients. Conclusion: IPMN has a relative good prognosis. Main duct type and mixed type IPMN have a higher malignant potential, and should receive surgical treatment. Patients of branch duct type IPMN with lesion 5 mm in diameter, without any other evident cause of obstruction. Cystic pancreatic lesions >5 mm in diameter that communicate with a
non-dilated main pancreatic duct were considered to be BD-IPMN. Mixed type IPMN displayed the aspects of both MD- and BD-IPMN features [6,7]. According to these criteria, 46 patients were diagnosed with MD-IPMN, 29 with BD-IPMN, and 21 with mixed type. The typical CT findings of each type of IPMN were shown in Fig. 1.
2.2. Surgical treatment All the patients received either a surgical procedure or a close surveillance, according to a comprehensive assessment of clinical symptoms, imaging findings, and serum tumor markers (CEA and CA19-9). Indications for surgical treatment were as follows: (1) lesions involving the main duct of pancreas (MDeor mixed type IPMN); (2) BD-IPMN either with lesions >3 cm in diameter, or accompanied by clinical symptoms (abdominal pain or discomfort, jaundice, or apparent weight loss), or with elevated serum level of either CEA or CA19-9. For BD-IPMN patients with lesions 5 ng/mL) and 35.4% (34/96) had an elevated CA19-9 level (>37 U/ mL). We found the differences in clinical characteristics between patients with MDeor mixed type IPMN and those with BD-IPMN (Table 1). Patients with tumor involving the main pancreatic duct had more chances to develop clinical symptoms as well as pancreatic endocrine dysfunction. An elevated serum CA19-9 level was more common in patients with MDeor mixed type IPMN. 3.2. Pathological features All the 78 patients who underwent surgical resections had been diagnosed with IPMN by postoperative pathological analyses, including 46 cases of non-invasive tumor (28 cases of low-grade dysplasia, 11 cases of moderate-grade dysplasia, and 7 cases of high-grade dysplasia) and 32 cases of invasive carcinoma. All resection margins were clear of tumor. There were no cases of lymph node metastasis among patients with non-invasive tumor.
Table 1 Comparison of clinical characteristics between patients with MD or mixed type IPMN and BD-IPMN. Clinical features
MD/mixed type (n ¼ 67)
Among the 32 patients with invasive carcinoma, 14 cases (43.8%) had lymph node metastasis. The comparison of clinical features between non-invasive tumors and invasive carcinomas was shown in Table 2. No significant difference was shown in gender distribution and appearance of symptoms between patients with non-invasive tumor and those with invasive carcinoma. There were significant differences between the two groups in mean age (P ¼ 0.000), tumor size (P ¼ 0.000), as well as CEA level (P ¼ 0.002) or CA19-9 level (P ¼ 0.000). The risk of invasive carcinoma for IPMN patients was significantly high in MD- (45.7%, 21/46) and mixed type cases (33.3%, 7/21), compared with BD-IPMN cases (13.8%, 4/29) (P ¼ 0.017). 3.3. Postoperative complications
2.4. Statistical analysis
Gender (M:F) 35:20 Age 58.6 ± 7.1 Diameter (cm) 2.60 ± 0.85 Clinical symptoms 61 Diabetes 52 Elevated tumor markers CEA＞5 ng/mL 8 CA19-9＞37 U/mL 31
51
BD type (n ¼ 18)
P value
23:18 58.3 ± 5.5 2.42 ± 0.88 11 6
0.455 0.822 0.349 0.000 0.000
2 3
0.458 0.001
A total of 32 postoperative complications occurred in 26 patients (complication rate 33.3%). There was no perioperative death. The postoperative complications included pancreatic fistula in 13 cases (16.7%), delayed gastric emptying in 11 cases (14.1%), biliary fistula in 5 cases (6.4%), intra-peritoneal bleeding in 2 cases (2.56%), and intra-abdominal infection in 1 case (1.28%). All complications were successfully managed with conservative treatments. The mean postoperative hospital stay was 14.4 days. Of the 43 patients who underwent pancreaticoduodenectomy, 27 postoperative complications were found, including pancreatic fistula in 10 patients, delayed gastric emptying in 10 patients, biliary fistula in 4 patients, intra-abdominal bleeding in 2 patients, and intra-abdominal infection in 1 patient. Two of the 25 patients who underwent pancreatic body and tail resection developed pancreatic fistula. Postoperative pancreatic fistula was occurred in 1 of the 6 patients who underwent segmental pancreatectomy. Of the 4 patients undergoing total pancreatectomy, 1 case of delayed gastric emptying and 1 case of biliary fistula were diagnosed. 3.4. Follow-up and survival analysis All the 96 IPMN patients were followed up regularly at an interval of 6 months, with a mean follow-up of 37 months (range: 6e72 months). For the asymptomatic BD-IPMN patients with lesion ＜3 cm in diameter, surgical treatments were introduced into 4 of them during the surveillance, in case of the development of one or more indications of surgery. The mean follow-up for the other 18 asymptomatic small BD-IPMN cases was 34 months (range: 6e65 months), without significant change in the diameter of the lesions. For the 46 postoperative patients with non-invasive tumor, the mean follow-up was 39 months (range: 8e72 months), and the
Table 2 Comparison of clinical features between non-invasive and invasive tumors. Clinical features
Non-invasive tumors (n ¼ 64a)
Gender (M:F) 39:25 Age 56.6 ± 6.2 Diameter (cm) 2.29 ± 0.7 Clinical 46 symptoms IPMN type MD type 25 21 Mixed type 14 BD type 25 Elevated tumor markers CEA＞5 ng/mL 2 CA19-9＞37 U/ 7 mL a
Invasive carcinoma (n ¼ 32)
P value
20:12 62.3 ± 5.9 3.04 ± 0.9 26
0.882 0.000 0.000 0.454 0.017
7 4 8 27
Included the 18 asymptomatic small BD-IPMN patients.
0.002 0.000
52
X. Tian et al. / International Journal of Surgery 13 (2015) 49e53
overall 5-year survival rate was 97.1%. For the 32 patients with invasive carcinoma, the mean postoperative follow-up was 46 months (range: 10e64 months), and the overall 1-, 3-, and 5-year survival rates were 93.7%, 74.7%, and 35.2%, respectively. Among these patients, 1-, 3-, and 5-year survival rates for 18 node-negative cases were 100%, 93.1%, and 56.3%, respectively. Meanwhile, 1-, 3-, and 5-year survival rates for the other 14 node-positive patients were 85.4%, 37.3%, and 0%, respectively. By KaplaneMeier survival curve analysis we observed significant differences in survival between non-invasive cases and invasive cases (P ¼ 0.0001), and between node-negative and node-positive patients (P ¼ 0.0104) (Fig. 2).
Table 3 “High-risk stigmata” and “worrisome features” in BD-IPMN.
4. Discussion
carcinoma, compare with PDACs [11]. Comparison of the prognosis of invasive IPMN patients with that of PDAC patients after radical resection showed that the 5-year survival rates in node-negative patients were 69% for invasive IPMN and 13% for ductal adenocarcinoma, but there was no significant difference in survival rates between groups in node-positive patients [12]. In this study, the 5year survival rate of the 32 patients with invasive cancer was much lower than that of the 46 patients with non-invasive tumor (35.2% vs. 97.1%, P ¼ 0.0001). Meanwhile, the prognosis of node-negative invasive IPMN cases was significantly better than that of nodepositive patients (P ¼ 0.0104). These data indicate that IPMN has a relatively low malignant degree, with a good prognosis in early stage patients, and lymph node metastasis could be a risk factor for worse prognosis in patients with invasive IPMN. IPMN is classified into MD-IPMN, BD-IPMN, and mixed type IPMN, according to the involvement of the pancreatic duct. MDIPMN and mixed type IPMN have significantly higher malignant potential than BD-IPMN [6,7]. Preoperative imaging techniques such as contrast-enhanced CT, MRI with magnetic resonance cholangiopancreatography (MRCP), EUS, and ERCP are sufficient to confirm the diagnosis, classify the IPMN type, and assess the respectability [8]. In the current study, all the 78 resected IPMN patients had been correctly diagnosed before operation. The risk of invasive carcinoma associated with different types of IPMN was 45.7% (21/46) in MD IPMN, 33.3% (7/21) in mixed type IPMN, and 13.8% (4/29) in BD-IPMN. In patients with IPMN involving the main pancreatic duct (MDeor mixed type IPMN), the risk of invasive carcinoma was significantly higher than that in BD-IPMN. In addition to the involvement of main pancreatic duct, the high risk of invasive carcinoma seemed to be associated with higher age, larger size of the lesion, as well as the elevation of serum CEA and CA19-9 levels. The gender of patients and manifestation of the
IPMN belongs to a heterogenous group of pancreatic cystic tumors characterized by the papillary proliferation of mucinproducing cells in the main or branch duct of the pancreas. IPMN occurs typically in elderly people, and is more commonly in men than in women. As relatively slow-growing tumors, more asymptomatic IPMN cases are diagnosed incidentally because of the improvements in imaging techniques. 3.8 If the main pancreatic duct is involved, recurrent chronic pancreatitis could be caused by an obstructed pancreatic duct secondary to increased mucin production, which often manifests as upper abdominal pain, weight loss, diarrhea, or diabetes [8]. Jaundice may occur if the bile duct is invaded or compressed by the tumors. These symptoms have been considered to be associated with a risk of malignancy [5,9]. In the current study, 75% of the 96 IPMN patients had one or more symptoms, which were significantly associated with the involvement of main pancreatic duct (Table 1). However, we observed no correlation between these clinic symptoms and the risk of invasive carcinoma (Table 2). According to the malignant potential, the WHO classified these tumors in categories, ranging from low-grade, intermediate-grade, and high-grade dysplasia (also called “carcinoma in situ”), to invasive carcinoma [8,9]. Surgical resection is the only definitive treatment for IPMN, and complete resection of non-invasive IPMN (low-, intermediate-, and high-grade dysplasia) could achieve an excellent survival rate. On the other hand, the survival rate of invasive IPMN patients is significantly lower than those with noninvasive tumors [10]. There is still no consensus on whether the prognosis of invasive IPMN cases is similar to that of pancreatic ductal carcinomas (PDAC). It is generally concluded that IPMN patients are more often diagnosed at an early stage, even for invasive
High risk stigmata
Worrisome features
Obstructive jaundice in a patient with a Clinical, pancreatitis cystic lesion of the pancreatic head Enhanced solid component Cyst of >3 cm MPD size of >10 mm Thickened enhanced cyst walls MPD size of 5e9 mm Nonenhanced mural nodules Abrupt change in the MPD caliber with distal pancreatic atrophy and lymphadenopathy
Fig. 2. KaplaneMeier survival curves for invasive and non-invasive IPMN patients (left, P ¼ 0.0001) and for invasive carcinoma IPMN with and without lymph node metastasis (right, P ¼ 0.0104).
X. Tian et al. / International Journal of Surgery 13 (2015) 49e53
clinical symptoms indicated no correlation with the risk of invasive carcinoma (Table 2). There is no agreement on the risk factors for the accurate prediction of development of invasive carcinoma from IPMN, except that it is generally agreed that MD- and mixed type IPMN should be resected if the patient is a candidate for surgery [5]. For BD-IPMN patients, the recently updated international consensus guidelines introduced worrisome features in imaging criteria and high-risk stigmata including obstructive jaundice, enhanced solid component, and main pancreatic duct size >10 mm (Table 3). The guidelines recommended that for BD-IPMN with high-risk stigmata, or worrisome features which are confirmed by EUS, resection should be considered in patients fit for surgery [7]. In this series of patients, all MDeor mixed type IPMN and BD-IPMN patients either with lesions >3 cm in diameter, or accompanied by clinical symptoms, or with elevated serum tumor markers, were treated with resections. For the 29 BD-IPMN patients, 7 of them received surgical treatment at the time of diagnosis. Resections were recommended for another 4 patients during the surveillance because of either the enlargement of the lesions or the onset of associated symptoms. Four of the 11 resected BD-IPMN cases were confirmed as invasive carcinomas, and the other 18 asymptomatic small BD-IPMN cases showed no change in the lesions after followup of 6e65 months, indicating that our strategy of surgical treatment is rational for this type of IPMN. However, recent studies reported a high rate of malignancy in “Sendai-negative” BD-IPMN or BD-IPMN smaller than 3 cm, and recommended more aggressive surgical treatment for all IPMN cases, including asymptomatic small BD-IPMN [13,14]. Therefore, the safety of our management for the 18 asymptomatic small BD-IPMN cases needs a long-term inspection. In this study, we retrospectively reviewed the clinical data of 96 IPMN cases during the last 8 years, and evaluated our strategy of surgical treatments for different types of IPMN. Our data confirmed the fact that IPMN has a relatively low malignant potential and a better prognosis, especially in early stage cases, compared with PDAC. Main duct type and mixed type IPMN have higher malignant potential, and should receive active surgical treatment. For BDIPMN patients with lesions >3 cm in diameter, or accompanied by clinical symptoms, or with elevated serum tumor markers, a resection should be recommended because of the high risk of invasive carcinomas. However, patients of BD-IPMN with a
