ORIGINAL ARTICLE
Parenchyma-Sparing Pancreatectomy for Presumed Noninvasive Intraductal Papillary Mucinous Neoplasms of the Pancreas Alain Sauvanet, MD,∗ †‡ S´ebastien Gaujoux, MD, PhD,∗ †‡ Benjamin Blanc, MD,∗ Anne Couvelard, MD, PhD,†‡§ Safi Dokmak, MD,∗ ‡ Marie-Pierre Vullierme, MD,‡¶ Philippe Ruszniewski, MD,†‡ Jacques Belghiti, MD,∗ †‡ and Philippe L´evy, MD†‡ Objective: To assess the feasibility and outcomes of parenchyma-sparing pancreatectomy (PSP), including enucleation (EN), resection of uncinate process (RUP), and central pancreatectomy (CP), as an alternative to standard pancreatectomy for presumed noninvasive intraductal papillary and mucinous neoplasms (IPMNs). Background: Pancreaticoduodenectomy and distal pancreatectomy are associated with significant perioperative morbidity, a substantial risk of pancreatic insufficiency, and may overtreat noninvasive IPMNs. Methods: From 1999 to 2011, PSP was attempted in 91 patients with presumed noninvasive IPMNs, after complete preoperative work-up including computed tomography, magnetic resonance imaging, and endoscopic ultrasonography. Intraoperative frozen section examination was routinely performed to assess surgical margins and rule out invasive malignancy. Follow-up included clinical, biochemical, and radiological assessments. Results: Overall PSP was achieved with a feasibility rate of 89% (n = 81), including 44 ENs, 5 RUPs, and 32 CPs. Postoperative mortality rate was 1.3% (n = 1), and overall morbidity was noteworthy (61%; n = 47). Definitive pathological examination confirmed IPMN diagnosis in 95% of patients (n = 77), all except 2 (3%), without invasive component. After a median follow-up of 50 months, both pancreatic endocrine/exocrine functions were preserved in 92% of patients. Ten-year progression-free survival was 76%, and reoperation for recurrence was required in 4% of patients (n = 3). Conclusions: In selected patients, PSP for presumed noninvasive IPMN in experienced hands is highly feasible and avoids inappropriate standard resections for IPMN-mimicking lesions. Early morbidity is greater than that after standard resections but counterbalanced by preservation of pancreatic endocrine/exocrine functions and a low rate of reoperation for tumor recurrence. Keywords: central pancreatectomy, enucleation, intraductal papillary mucinous neoplasms, parenchyma-sparing pancreatectomy, resection of uncinate process (Ann Surg 2014;260:364–371)
W
ith the widespread use of high-quality cross-sectional imaging, intraductal papillary and mucinous neoplasms (IPMNs) of the pancreas are identified with increasing frequency.1 Routine From the ∗ Department of Hepato-Pancreato-Biliary Surgery–Pˆole des Maladies de l’Appareil Digestif (PMAD), AP-HP, Beaujon Hospital, Clichy, France; †University Paris 7 Denis Diderot, Paris, France; ‡Department HospitaloUniversitaire UNITY (Addressing Unmet Needs for Innovation in HepaTology and GastroenterologY); §Department of Pathology; ¶Department of Radiology; and Department of Gastroenterology, Pˆole des Maladies de l’Appareil Digestif, AP-HP, Beaujon Hospital, Clichy, France. Disclosure: None of the authors have any financial or any other kind of personal conflicts of interest in relation with this study and no funding was received in support of the manuscript. Reprints: Alain Sauvanet, MD, Department of Digestive Surgery, AP-HP, Beaujon Hospital, 100, Bd du G´en´eral Leclerc, 92118 Clichy-Cedex, France. E-mail: [email protected]. C 2014 by Lippincott Williams & Wilkins Copyright ISSN: 0003-4932/14/26002-0364 DOI: 10.1097/SLA.0000000000000601
364 | www.annalsofsurgery.com
resection of all lesions is no longer advocated, surgical indication being presently based on symptoms and risk factors of malignant transformation.1–6 Despite advances reported in recent years, standard pancreatectomies, including pancreaticoduodenectomy and distal or total pancreatectomy, still carry a significant postoperative mortality ranging from 1% to 4% in high-volume centers.7–9 These procedures are also associated with high postoperative morbidity and long-term disorders including exocrine insufficiency10,11 and diabetes mellitus.12–14 Because the natural history of IPMNs is difficult to predict,15,16 the benefit-risk balance of pancreatectomy for noninvasive lesions is questionable, especially in high operative risk patients or in case of branch-duct (BD) disease associated with a low risk of malignant transformation.17 Parenchyma-sparing pancreatectomies (PSPs), including enucleation (EN),6,18 resection of uncinate process (RUP), also called inferior pancreatic head resection,19,20 and central pancreatectomy (CP),21–23 have been proposed for low-grade pancreatic tumors. Regarding IPMNs, these PSPs are theoretically suitable for treating noninvasive lesions but have been scarcely reported for this indication.21,24–26 The aim of this study was to report a single-center experience of PSPs for presumed noninvasive IPMNs, with a specific attention to preoperative patient selection, intraoperative management, postoperative complications, and long-term outcomes including recurrence and pancreatic function.
PATIENTS AND METHODS Data Collection Of the 413 patients operated on with a preoperative diagnosis of IPMN between January 1999 and June 2011, a total of 81 underwent PSP. All surgical indications were discussed in a multidisciplinary pancreatic tumor board including surgeons, pathologists, radiologists, and gastroenterologists. Before surgery, all patients underwent dedicated pancreatic multidetector computed tomography (CT), endoscopic ultrasonography (EUS), and gadolinium-enhanced magnetic resonance imaging (MRI) with cholangiopancreatography using a standardized protocol.
Inclusion and Exclusion Criteria Indications of surgical resection, independently of the planed procedure (PSP vs standard resection), were established according to previously reported criteria17,27 : 1. All patients with main-duct IPMN 2. Patients with branch-duct IPMN and a) diameter above 3 cm b) and/or increasing size at 2 follow-up imaging procedures c) and/or with mural nodules d) and/or presenting with symptoms formally attributable to IPMN (acute pancreatitis or recurrent epigastric pain in the absence of other cause). Annals of Surgery r Volume 260, Number 2, August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 260, Number 2, August 2014
Once surgery was indicated, a PSP (ie, EN, RUP, or CP) was systematically considered for all patients with favorable anatomical localization and no signs suggestive of malignancy such as mural nodules above 5-mm diameter28 or solid mass.4,29 For IPMNs limited to a single branch duct, EN was performed when the dilated duct was superficial with a narrow communicating duct (Fig. 1A) whereas RUP was performed when the dilated branch duct was located deeply in the uncinate process (Fig. 1B). CP was indicated for branch-duct lesions deeply located in the parenchyma of the body of the pancreas, or when the communicating duct was enlarged, or centrally localized with main-duct involvement (dilatation >5 mm) (Fig. 1C). In case of multifocal IPMNs, small branch-duct IPMNs (ie,
Parenchyma-Sparing Pancreatectomy for Presumed Noninvasive Intraductal Papillary Mucinous Neoplasms of the Pancreas Alain Sauvanet, MD,∗ †‡ S´ebastien Gaujoux, MD, PhD,∗ †‡ Benjamin Blanc, MD,∗ Anne Couvelard, MD, PhD,†‡§ Safi Dokmak, MD,∗ ‡ Marie-Pierre Vullierme, MD,‡¶ Philippe Ruszniewski, MD,†‡ Jacques Belghiti, MD,∗ †‡ and Philippe L´evy, MD†‡ Objective: To assess the feasibility and outcomes of parenchyma-sparing pancreatectomy (PSP), including enucleation (EN), resection of uncinate process (RUP), and central pancreatectomy (CP), as an alternative to standard pancreatectomy for presumed noninvasive intraductal papillary and mucinous neoplasms (IPMNs). Background: Pancreaticoduodenectomy and distal pancreatectomy are associated with significant perioperative morbidity, a substantial risk of pancreatic insufficiency, and may overtreat noninvasive IPMNs. Methods: From 1999 to 2011, PSP was attempted in 91 patients with presumed noninvasive IPMNs, after complete preoperative work-up including computed tomography, magnetic resonance imaging, and endoscopic ultrasonography. Intraoperative frozen section examination was routinely performed to assess surgical margins and rule out invasive malignancy. Follow-up included clinical, biochemical, and radiological assessments. Results: Overall PSP was achieved with a feasibility rate of 89% (n = 81), including 44 ENs, 5 RUPs, and 32 CPs. Postoperative mortality rate was 1.3% (n = 1), and overall morbidity was noteworthy (61%; n = 47). Definitive pathological examination confirmed IPMN diagnosis in 95% of patients (n = 77), all except 2 (3%), without invasive component. After a median follow-up of 50 months, both pancreatic endocrine/exocrine functions were preserved in 92% of patients. Ten-year progression-free survival was 76%, and reoperation for recurrence was required in 4% of patients (n = 3). Conclusions: In selected patients, PSP for presumed noninvasive IPMN in experienced hands is highly feasible and avoids inappropriate standard resections for IPMN-mimicking lesions. Early morbidity is greater than that after standard resections but counterbalanced by preservation of pancreatic endocrine/exocrine functions and a low rate of reoperation for tumor recurrence. Keywords: central pancreatectomy, enucleation, intraductal papillary mucinous neoplasms, parenchyma-sparing pancreatectomy, resection of uncinate process (Ann Surg 2014;260:364–371)
W
ith the widespread use of high-quality cross-sectional imaging, intraductal papillary and mucinous neoplasms (IPMNs) of the pancreas are identified with increasing frequency.1 Routine From the ∗ Department of Hepato-Pancreato-Biliary Surgery–Pˆole des Maladies de l’Appareil Digestif (PMAD), AP-HP, Beaujon Hospital, Clichy, France; †University Paris 7 Denis Diderot, Paris, France; ‡Department HospitaloUniversitaire UNITY (Addressing Unmet Needs for Innovation in HepaTology and GastroenterologY); §Department of Pathology; ¶Department of Radiology; and Department of Gastroenterology, Pˆole des Maladies de l’Appareil Digestif, AP-HP, Beaujon Hospital, Clichy, France. Disclosure: None of the authors have any financial or any other kind of personal conflicts of interest in relation with this study and no funding was received in support of the manuscript. Reprints: Alain Sauvanet, MD, Department of Digestive Surgery, AP-HP, Beaujon Hospital, 100, Bd du G´en´eral Leclerc, 92118 Clichy-Cedex, France. E-mail: [email protected]. C 2014 by Lippincott Williams & Wilkins Copyright ISSN: 0003-4932/14/26002-0364 DOI: 10.1097/SLA.0000000000000601
364 | www.annalsofsurgery.com
resection of all lesions is no longer advocated, surgical indication being presently based on symptoms and risk factors of malignant transformation.1–6 Despite advances reported in recent years, standard pancreatectomies, including pancreaticoduodenectomy and distal or total pancreatectomy, still carry a significant postoperative mortality ranging from 1% to 4% in high-volume centers.7–9 These procedures are also associated with high postoperative morbidity and long-term disorders including exocrine insufficiency10,11 and diabetes mellitus.12–14 Because the natural history of IPMNs is difficult to predict,15,16 the benefit-risk balance of pancreatectomy for noninvasive lesions is questionable, especially in high operative risk patients or in case of branch-duct (BD) disease associated with a low risk of malignant transformation.17 Parenchyma-sparing pancreatectomies (PSPs), including enucleation (EN),6,18 resection of uncinate process (RUP), also called inferior pancreatic head resection,19,20 and central pancreatectomy (CP),21–23 have been proposed for low-grade pancreatic tumors. Regarding IPMNs, these PSPs are theoretically suitable for treating noninvasive lesions but have been scarcely reported for this indication.21,24–26 The aim of this study was to report a single-center experience of PSPs for presumed noninvasive IPMNs, with a specific attention to preoperative patient selection, intraoperative management, postoperative complications, and long-term outcomes including recurrence and pancreatic function.
PATIENTS AND METHODS Data Collection Of the 413 patients operated on with a preoperative diagnosis of IPMN between January 1999 and June 2011, a total of 81 underwent PSP. All surgical indications were discussed in a multidisciplinary pancreatic tumor board including surgeons, pathologists, radiologists, and gastroenterologists. Before surgery, all patients underwent dedicated pancreatic multidetector computed tomography (CT), endoscopic ultrasonography (EUS), and gadolinium-enhanced magnetic resonance imaging (MRI) with cholangiopancreatography using a standardized protocol.
Inclusion and Exclusion Criteria Indications of surgical resection, independently of the planed procedure (PSP vs standard resection), were established according to previously reported criteria17,27 : 1. All patients with main-duct IPMN 2. Patients with branch-duct IPMN and a) diameter above 3 cm b) and/or increasing size at 2 follow-up imaging procedures c) and/or with mural nodules d) and/or presenting with symptoms formally attributable to IPMN (acute pancreatitis or recurrent epigastric pain in the absence of other cause). Annals of Surgery r Volume 260, Number 2, August 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 260, Number 2, August 2014
Once surgery was indicated, a PSP (ie, EN, RUP, or CP) was systematically considered for all patients with favorable anatomical localization and no signs suggestive of malignancy such as mural nodules above 5-mm diameter28 or solid mass.4,29 For IPMNs limited to a single branch duct, EN was performed when the dilated duct was superficial with a narrow communicating duct (Fig. 1A) whereas RUP was performed when the dilated branch duct was located deeply in the uncinate process (Fig. 1B). CP was indicated for branch-duct lesions deeply located in the parenchyma of the body of the pancreas, or when the communicating duct was enlarged, or centrally localized with main-duct involvement (dilatation >5 mm) (Fig. 1C). In case of multifocal IPMNs, small branch-duct IPMNs (ie,
