Noninvasive intraductal papillary mucinous neoplasms and mucinous cystic neoplasms: Recurrence rates and postoperative imaging follow-up Dimitrios Xourafas, MD, MPH,a,b Ali Tavakkoli, MD,b Thomas E. Clancy, MD,b and Stanley W. Ashley, MD,b Boston, MA
Background. Although surveillance guidelines for resected invasive mucinous neoplastic cysts are wellestablished, those for noninvasive cysts are not defined. We used our experience with resected noninvasive mucinous neoplastic cysts to define recurrence rates and the optimal frequency of postoperative imaging follow-up. Methods. We reviewed the medical records of 134 patients with resected, pathologically confirmed noninvasive mucinous neoplasms between 2002 and 2012. Demographics, comorbidities, cyst characteristics, and recurrence were evaluated. Survival analysis was used to estimate the distribution of time to recurrence and regression models were used to investigate factors associated with recurrence. Results. Eighty-seven patients with intraductal papillary mucinous neoplasms (IPMNs) were compared with 47 patients with mucinous cystic neoplasms (MCNs). Those with MCNs were more often females (P = .001), significantly younger (P = .001), more symptomatic (P = .009), and had cysts more often located in the tail (P < .001). Median follow-up was 42 months. Recurrence rates for IPMNs were 0%, 5%, and 10% versus 0% for MCNs respectively at postoperative years 1, 2, and 3 (P = .014). On multivariable analysis, size >3 cm (P = .027), higher grade dysplasia (P = .043), and positive resection margins (P < .001) were significantly associated with recurrence. Conclusion. Resected noninvasive IPMNs with moderate- or high-grade dysplasia and negative resection margins require imaging follow-up every 2 years, given the 16% overall recurrence rate. Although the follow-up interval for noninvasive, low-grade, dysplastic IPMNs with negative margins could be lengthened, all noninvasive IPMNs having positive margins require yearly follow-up at the minimum. Resected noninvasive MCNs––irrespective of grade and margin status––do not require surveillance, although the development of branch duct-IPMNs in the remnant pancreas can be investigated in the long term at the discretion of the provider. (Surgery 2015;157:473-83.) From the Harvard School of Public Healtha and the Brigham and Women’s Hospital,b Harvard Medical School, Boston, MA
THE PREVALENCE OF incidentally discovered cystic neoplasms of the pancreas in patients undergoing high-resolution abdominal imaging has increased
Portions of this manuscript were presented at the Massachusetts Chapter of the American College of Surgeons, Boston, MA, December 2011; Academic Surgical Congress, Las Vegas, NV, February 2012; and The Pancreas Club, San Diego, CA, May 2012. Accepted for publication September 19, 2014. Reprint requests: Dimitrios Xourafas, MD, MPH, Harvard School of Public Health, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115. E-mail: dxourafas@partners. org. 0039-6060/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.09.028
to 2.6%.1-4 Pancreatic cysts constitute a broad spectrum of entities, which typically fall into 4 groups: Non-neoplastic cysts, cystic nonepithelial neoplasms, primary cystic epithelial neoplasms, and secondary cystic solid neoplasms.5 Although the first 2 groups often do not require surgical intervention, surgery may be indicated for some nonserous subtypes of neoplastic pancreatic cysts. Mucinous cystic neoplasms (MCNs) and intraductal papillary mucinous neoplasms (IPMNs) are 2 of the most common pancreatic neoplastic cyst types with distinctive histopathologic features and differing malignant potentials.6-9 MCNs have a low prevalence of invasive carcinoma (
Background. Although surveillance guidelines for resected invasive mucinous neoplastic cysts are wellestablished, those for noninvasive cysts are not defined. We used our experience with resected noninvasive mucinous neoplastic cysts to define recurrence rates and the optimal frequency of postoperative imaging follow-up. Methods. We reviewed the medical records of 134 patients with resected, pathologically confirmed noninvasive mucinous neoplasms between 2002 and 2012. Demographics, comorbidities, cyst characteristics, and recurrence were evaluated. Survival analysis was used to estimate the distribution of time to recurrence and regression models were used to investigate factors associated with recurrence. Results. Eighty-seven patients with intraductal papillary mucinous neoplasms (IPMNs) were compared with 47 patients with mucinous cystic neoplasms (MCNs). Those with MCNs were more often females (P = .001), significantly younger (P = .001), more symptomatic (P = .009), and had cysts more often located in the tail (P < .001). Median follow-up was 42 months. Recurrence rates for IPMNs were 0%, 5%, and 10% versus 0% for MCNs respectively at postoperative years 1, 2, and 3 (P = .014). On multivariable analysis, size >3 cm (P = .027), higher grade dysplasia (P = .043), and positive resection margins (P < .001) were significantly associated with recurrence. Conclusion. Resected noninvasive IPMNs with moderate- or high-grade dysplasia and negative resection margins require imaging follow-up every 2 years, given the 16% overall recurrence rate. Although the follow-up interval for noninvasive, low-grade, dysplastic IPMNs with negative margins could be lengthened, all noninvasive IPMNs having positive margins require yearly follow-up at the minimum. Resected noninvasive MCNs––irrespective of grade and margin status––do not require surveillance, although the development of branch duct-IPMNs in the remnant pancreas can be investigated in the long term at the discretion of the provider. (Surgery 2015;157:473-83.) From the Harvard School of Public Healtha and the Brigham and Women’s Hospital,b Harvard Medical School, Boston, MA
THE PREVALENCE OF incidentally discovered cystic neoplasms of the pancreas in patients undergoing high-resolution abdominal imaging has increased
Portions of this manuscript were presented at the Massachusetts Chapter of the American College of Surgeons, Boston, MA, December 2011; Academic Surgical Congress, Las Vegas, NV, February 2012; and The Pancreas Club, San Diego, CA, May 2012. Accepted for publication September 19, 2014. Reprint requests: Dimitrios Xourafas, MD, MPH, Harvard School of Public Health, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115. E-mail: dxourafas@partners. org. 0039-6060/$ - see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2014.09.028
to 2.6%.1-4 Pancreatic cysts constitute a broad spectrum of entities, which typically fall into 4 groups: Non-neoplastic cysts, cystic nonepithelial neoplasms, primary cystic epithelial neoplasms, and secondary cystic solid neoplasms.5 Although the first 2 groups often do not require surgical intervention, surgery may be indicated for some nonserous subtypes of neoplastic pancreatic cysts. Mucinous cystic neoplasms (MCNs) and intraductal papillary mucinous neoplasms (IPMNs) are 2 of the most common pancreatic neoplastic cyst types with distinctive histopathologic features and differing malignant potentials.6-9 MCNs have a low prevalence of invasive carcinoma (
