Complete resection of SCNs ensures cure (SCN/7, 9, 24). SCNs do not recur either locally or distally after appropriate operative treatment with an R0 resection; therefore, a regular follow-up program with surveillance after complete resection is not necessary and would not be cost effective [1–3]. No adjuvant therapy is needed. For patients with SCNs treated conservatively by surveillance without resection, there are no firm guidelines about the optimal method or timing of surveillance. The interval between serial cross-sectional imaging remains controversial, but most groups recommend surveillance on an annual basis [1, 4–6], while others recommend every 2 years [6].

Complete operative resection of MCNs lacking an invasive component (i.e., benign MCNs and, more importantly, noninvasive proliferative MCNs) is a curative procedure [3, 7]). These neoplasms are solitary and do not recur either locally or distally after complete operative resection [MCN/7]. The survival of these patients is excellent, with 100 % 5-year disease-specific survival [8]. Therefore, for patients without tissue invasion, a regular follow-up program with surveillance (using cross-sectional imaging tests) is also not necessary, thereby saving money and eliminating patient worry [4, 7–10]. One, however, must be confident in the diagnosis of MCN, being certain that the cystic lesion is not an IPMN; the presence of ovarian stroma in the pathology specimen is mandatory to exclude IPMN.

More controversial is the question of long-term survival and surveillance of the more unusual patients with MCNs with tissue invasion who undergo “curative” resection. In the past, numerous articles have claimed survival rates greater than 50 % and up to 70 % for these Grade 1 mucinous “cystadenocarcinomas” [7, 8, 11]; however, these series lumped together MCNs containing a proliferative epithelium (but without any tissue invasion) with the true cystadenocarcinomas which had tissue invasion. Occasionally, unrecognized foci of invasive carcinoma may exist within a presumably noninvasive proliferative MCN [12]; in this case, recurrence and metastases can be observed, which contrasts the absence of recurrences for “true” noninvasive MCNs after complete resection and stresses the importance of a complete histopathologic analysis of all the surface areas of a MCN [12–14]. After a careful evaluation of MCNs containing true invasive carcinoma, 5-year survival rates appear quite poor (15–35 %), albeit still somewhat better than those for typical ductal cancer of the pancreas but consistent with an aggressive malignancy [8]. The extent of invasion is the most important prognostic factor in malignant MCNs [15]. Patients with invasive MCNs require postoperative follow-up imaging, as for patients with pancreatic adenocarcinoma [4]. Some groups suggest that patients with resected malignant MCNs should be followed every 6 months regarding local recurrence and distant metastases (mainly hematogenous) using either CT or MRI [10]. The prognosis for unresectable invasive MCNs is as poor as that for unresectable pancreatic adenocarcinoma [2, 8]. The dramatic difference in prognosis for patients with noninvasive and invasive MCNs highlights the enormous importance of diagnosing and resecting noninvasive MCNs before they progress to an invasive carcinoma whenever possible [16]. Whether resection of recurrent MCN, either locally or distantly (liver, lung), is worthwhile is completely unknown.

In IPMN, the dysplastic component may remain in situ for many years. For a single branch-duct IPMN, several studies suggest strongly that a local anatomic resection is essentially curative in most patients. In contrast, for noninvasive, main-duct IPMNs, occurrence in the remnant gland has been found with variable rates (0–20 %) [2, 9, 14], provided that the frozen-section margin is negative for adenomatous changes and the resected specimen lacks invasive IPMN. One study of patients from two large referral centers (Verona, Italy, and Boston, MA USA) maintained that after a curative resection with negative margins, recurrence is extremely rare [17]; in contrast, another large study showed an 8 % incidence of recurrence in the remnant [10]. Under these conditions, recurrence in the remnant may be due to the presence of multifocal disease [18] or to overlooked residual disease at the ductal margin. In distinct contrast, when the resection specimen contains invasive disease, even if the margins are negative, recurrent IPMN, either in the pancreatic remnant or more commonly in peripancreatic and extrapancreatic sites, occurs in 50–90 % of patients, dramatically altering prognosis and reinforcing the concept that invasive IPMN is a serious, aggressive malignancy [19, 20]. Interestingly, in the Mayo Clinic series [21] of invasive IPMN, recurrence after partial pancreatectomy (18/27; 67 %) was similar to that observed after total pancreatectomy (8/13; 62 %), suggesting no oncologic advantage to total pancreatectomy, similar in principle to ductal adenocarcinoma of the pancreas [22].

The 5-year survival after curative resection of IPMN without invasive cancer is >70 % in most series [21, 23]. Some reports have even suggested a 5-year survival in excess of 90 % after resection. In contrast, after resection of invasive IPMN, even with negative margins, the 5-year survival ranges from only 30 to 50 % [19–21]. Features predicting decreased survival when invasive cancer is present include lymph node metastases, vascular invasion, positive resection margins, and an invasive component measuring >2 cm [20, 24]. Invasive IPMN should be managed as an aggressive malignancy that behaves, in many respects, similar to ductal cancer of the pancreas. Overall survival, however, appears somewhat better with invasive IPMN compared to pancreatic ductal adenocarcinoma, but whether this observation is due to a stage shift with earlier diagnosis of IPMN, as shown in some studies [20], or due to a truly less aggressive tumor biology of invasive IPMN remains controversial.

Because adjuvant therapy for pancreatic ductal adenocarcinoma has been shown to improve survival, it is often recommended in the management of patients with invasive IPMN, especially in the presence of negative prognostic factors (i.e., positive lymph nodes and/or positive margins). The role of adjuvant chemotherapy or chemoradiation therapy in invasive IPMN, however, remains controversial. Interestingly, in a recent study published by Turrini et al., a notable decrease in 5-year overall survival was noted in those patients who received any adjuvant therapy [24]. In this study, re-operation for recurrence in the remnant pancreas was associated with substantial long-term survival; this aggressive surgical strategy should be considered for patients with local recurrence amenable to re-resection. In contrast, a study from Johns Hopkins suggested a survival advantage to the use of adjuvant chemoradiation therapy [25]. Obviously, this topic requires formal study (see above – ADJUVANT/NEOADJUVANT THERAPY).

Routine follow-up surveillance with noninvasive imaging appears to be relevant clinically, potentially therapeutic, and indicated in all patients with IPMNs. If recurrence occurs, selected patients may benefit from further treatment [10]. There are no established guidelines regarding the frequency or type of surveillance imaging to detect potential recurrence. A reasonable strategy for noninvasive IPMNs would be to obtain yearly follow-up with CT or MRI and then increase the interval between imaging if no changes have occurred over several years [9, 10]. For patients with noninvasive IPMN but with low- or moderate-grade dysplasia at the margin, clinical review and MRI/MRCP twice a year (if asymptomatic) have been proposed [4]. Because patients with invasive IPMN have a greater risk of recurrence, this population probably should be evaluated every 6 months with abdominal CT or MRI/MRCP [9, 10]. As noted previously, observation may well be indicated for patients with branch-duct IPMN who are asymptomatic without mural nodules in whom the main duct is not dilated (<6 mm) and the cyst size is <3 cm. The frequency of follow-up can be based on the size of the side branch IPMN: 0–1 cm, yearly; 1–2 cm, every 6–12 months; and 2–3 cm, every 3–6 months. We stress, however, that these are suggested guidelines based on the best interpretation of our current understanding. Abdominal CT, MRI/MRCP, and EUS are utilized for assessing cyst size, presence of mural nodules, and any changes in the diameter of the main duct. The interval of follow-up may be increased in duration if no change is observed during the first 2 years postoperatively [26]. The known increased risk of development of typical pancreatic ductal adenocarcinoma in patients with branch-duct IPMN should also be considered when planning follow-up in patients treated conservatively [22]. Biochemical follow-up surveillance is of little value. Tumor markers, such as serum levels of CEA and CA 19-9, have no value in the follow-up of these patients. The discovery of a pancreatic cyst/mass lesion after operative resection may be related to the presence of a postoperative contained leak, recurrence of IPMN linked to incomplete resection, a new site of IPMN, or rarely a recurrence of cystadenocarcinoma after inadequate histopathologic examination [2]. Similar considerations are pertinent after resection of main-duct IPMN. One other consideration is that dilation of the main pancreatic duct after a pancreatoduodenectomy may also be secondary to stricture at the pancreaticoenterostomy and not always from recurrent IPMN.