eMedicine Specialties > Radiology > Gastrointestinal

Pancreas, Mucinous Cystic Neoplasm: Follow-up

Author: Ali Nawaz Khan, MBBS, FRCS, FRCP, FRCR, LRCP, Chairman of Medical Imaging, Professor of Radiology, NGHA, King Fahad National Guard Hospital, King Abdulaziz Medical City, Riyadh, Saudi Arabia
Coauthor(s): Aali J Sheen, MD, MBChB, FRCS, Consulting Hepatobiliary Surgeon, HepatoBiliary Unit, Manchester Royal Infirmary; Sumaira MacDonald, MBChB, PhD, MRCP, FRCR, Lecturer, Sheffield University Medical School; Endovascular Fellow, Sheffield Vascular Institute; Haren Varia, MB ChB, FRCR, Consultant, Department of Clinical Radiology, Blackpool, Fylde and Wyre NHS Trust
Contributor Information and Disclosures

Updated: Mar 12, 2009

Intervention

Diagnosis

Despite advances in imaging, the diagnosis of pancreatic masses remains difficult. The differential diagnosis of a cystic pancreatic mass is large. Accurate diagnosis is imperative for appropriate patient management; therefore, occasionally obtaining tissue for histologic diagnosis may be essential.

Although the indications for percutaneous pancreatic biopsy are not standardized, in many circumstances, biopsy is clearly indicated. Such circumstances include cases of inoperable pancreatic tumor, suspected lymphoma, and metastasis or neuroendocrine tumors, as well as those in which cystic tumors must be differentiated.

Pancreatic biopsy is usually performed with fine needles used for cytology. The use of larger needles for core biopsy slightly improves sensitivity, especially in masses due to chronic pancreatitis or neuroendocrine tumors. Complications seldom occur; among them, acute pancreatitis and tumoral seeding are the most important. Percutaneous pancreatic biopsy is a safe, flexible, and reliable method for diagnosing pancreatic masses.¹⁵

The roles of EUS and EUS FNA in the evaluation of cystic pancreatic lesions are evolving. The detailed images provided by EUS and, hence, the ability to target the biopsy at suspicious areas within the pancreatic cystic lesion may prove to be invaluable. Improvements in EUS equipment will further secure the role of this technology in evaluating patients for these lesions.¹⁶

Anderson and Scheiman discuss the use of EUS in serous cystadenoma, which is the most common of the nonmucinous cystic pancreatic neoplasms.¹⁷ These microcystic tumors were previously termed glycogen-rich cystadenomas because of the presence of glycogen in the cyst epithelium. A small number of these lesions are macrocystic, and differentiating them from mucinous lesions may be difficult. However, EUS-guided FNA can provide diagnostic material from the cyst fluid. The second most common nonmucinous cyst is the islet cell tumor. These rare cystic tumors may or may not be accompanied by excess hormone production. EUS-guided aspiration may again be useful in establishing the diagnosis.

Analysis of cyst fluid obtained by image-guided aspiration is a useful adjunct in the evaluation of cystic pancreatic masses. Fluid amylase levels are characteristically increased in over 95% of pseudocysts; therefore, a normal amylase level can be used to exclude a pseudocyst. Fluid CEA levels can be elevated in a variety of cystic lesions, including mucinous cystic pancreatic tumors and some pseudocysts. However, this level is always low or normal in serous cystadenomas. If an accurate diagnosis is not achieved, resection rather than observation is the preferred option.

Resection

If diagnosed correctly, all mucinous cystic tumors of the pancreas should be resected, as a small risk of malignant potential exists. Complete resection has a favorable prognosis, with a low probability of recurrence. The prognosis is better than that for ductal adenocarcinoma of the pancreas, with intraductal papillary tumors having the best prognosis. Aspiration of the cyst is generally avoided due to the risk of a pancreatic fistula formation. The insidious nature of mucinous pancreatic cysts allows many patients to survive for many years symptom-free.¹⁸

Pancreatic resection is the treatment of choice in IPMT, which successfully relieves symptoms and prevents progression to malignant degeneration. Conservative management techniques, such as sphincterotomy, stenting, and clearance of intraductal mucus are temporary measures that may relieve symptoms. However, they do not address the malignant potential of these tumors.

Pancreaticoduodenectomy is generally advocated, given the predominance of IPMT in the head of the pancreas, but distal pancreatectomy is usually performed for tumors in the pancreatic body or tail. When the tumor involves the entire ductal system, total pancreatectomy is the usual option. Intraoperative frozen sections are required to confirm disease-free margins after ductal resection. If the results are positive, extended resection is warranted.

Recurrence

Ikeda and associates clarified the patterns of recurrence and prognosis after resection of IPMTs in 14 patients with histologically proven intraductal papillary adenocarcinoma (12 patients) or intraductal papillary adenoma (2 patients). Six patients were alive with no evidence of disease. Two patients died from other causes. Six patients had recurrences. The median survival time was 46 months. In the 6 recurrent cases, the median postoperative disease-free interval was 38 months. Four patients died from recurrence, and the median survival time after recurrence was 6 months.

The major site of recurrence was the pancreatic remnant. The other sites were 2 liver metastases in 2 cases of peritoneal spread and 1 case of local spread. Because of the low frequency of lymph-node metastases, surgery to preserve pancreatic function may be recommended, especially for localized tumors such as the branch type of tumor. It is important to avoid incomplete resection by using intraoperative pancreatoscopy and ultrasonography. Long-term follow-up after surgery is necessary, even with curative resection. The authors recommended total pancreatectomy for recurrences without distant metastases.

If a pancreatic pseudocyst can be ruled out, evaluation should subsequently focus on identifying those tumors, which require surgical resection because of actual or potential malignancy. As opposed to ductal adenocarcinoma, cystic neoplasms with malignant potential are slow growing, and a relatively favorable prognosis has been reported, even in the setting of malignant degeneration. Tumors with malignant potential include MCNs, IPMTs, papillary cystic neoplasms, and cystic islet-cell tumors. In contrast, serous cystadenomas are almost universally benign. The diagnostic examination of choice is spiral CT with intravenous contrast enhancement, which allows tumor localization and, sometimes, discrimination of serous cystadenomas from other neoplasms.

Survival and prognosis

Kitagawa and associates examined the case records of 63 patients with IPMT treated with resection to determine the predictors of prognosis.¹⁹ IPMT was diagnosed by using the surgical specimen and the WHO definition. Predictors were determined by using univariate and multivariate analyses.

Actuarial 3- and 5-year survival for patients with benign disease was 95% and 83%, and for those with malignant disease, rates were 52% and 44%, respectively (P = .0048). In the 33 patients with malignant disease (mean follow-up, 35 mo), gross mucus was a predictor of prolonged survival after univariate and multivariate analyses (odds ratios, 4.34 and 4.55, respectively), whereas alcohol abuse was a predictor of poor survival (odds ratios, 3.41 and 3.60, respectively). Gross mucus observed during endoscopy was a predictor of benign IPMT, and in the group with malignant IPMT, the presence of gross mucus was associated with better survival. Survival was also strongly associated with either benign IPMT or negative staining for p53 overexpression.

Recurrence in the duct of the pancreatic remnant after surgery in IPMT has not been reported after curative resection, and 5-year survival rates greater than 80% are common for noninvasive tumors.

Maire and associates examined case records of 73 patients with malignant IPMT to assess prognostic factors after surgical resection, and to compare long-term survival with that of patients surgically treated for ductal adenocarcinoma.²⁰ Surgical treatments for IPMT were pancreaticoduodenectomy (n = 46) and distal (n = 14), total (n = 11), or segmentary (n = 2) pancreatectomy. The operative mortality rate was 4%. IPMT corresponded to in situ (n = 22) or invasive carcinoma (n = 51). In the latter group, 17 had lymph-node metastases.

Overall median survival was 47 months. In patients with in situ and invasive carcinoma, 5-year survival rates were 88% and 36%, respectively. On univariate analysis, the following were associated with a fatal outcome: abdominal pain, high serum CA 19-9 concentrations before surgery, caudal localization, invasive carcinoma, lymph-node metastases, peripancreatic extension, and malignant relapse. On multivariate analysis, lymph-node metastasis was the only prognostic factor (odds ratio, 7.5; 95% confidence interval, 3.4, 16.4).

The overall 5-year survival rate was higher in patients with malignant invasive IPMT than in those with pancreatic ductal carcinoma (36% vs 21%, P = .03). However, it was similar in the subset with stage II or stage III tumors. The authors concluded that the prognosis of patients with resected in situ or invasive stage I malignant IPMT is excellent. In contrast, the prognosis of those with locally advanced forms is as poor as that of patients with pancreatic ductal adenocarcinoma.

Nakagohri and associates published their long-term outcomes after surgical resection.²¹ They retrospectively reviewed the clinicopathologic features and outcomes in patients who underwent pancreatic resection for noninvasive and minimally invasive intraductal papillary mucinous adenocarcinoma. Minimally invasive structures were pathologically observed in 5. The mean age of patients with noninvasive (n = 16) or minimally invasive (n = 5) adenocarcinoma was 61 years. Of patients with minimally invasive adenocarcinoma, 4 had abdominal pain. Conversely, 7 patients with noninvasive adenocarcinoma had no complaint. The mean size of noninvasive and minimally invasive tumors was 2.5 cm (range, 0.8 -4.0 cm) and 3.3 cm (range, 2.5-4.5 cm), respectively.

The overall 5-year and 10-year survival rates for all 21 patients were 89% and 47%, respectively. Disease recurred in 3 patients; 2 patients with minimally invasive adenocarcinoma and 1 with noninvasive adenocarcinoma. Recurrence sites were the peritoneum (n = 2) and MPD of the remnant pancreas (n = 1). Five disease-free patients died from unrelated causes. The remaining 13 patients were alive and disease-free 3-12 years after surgery. Noninvasive and minimally invasive intraductal papillary mucinous adenocarcinoma had a favorable prognosis after surgical treatment.

Katoh et al treated 45 patients with cystadenoma and cystadenocarcinoma of the pancreas.²² Their patients were divided into 2 groups: group 1, which included 26 patients treated between 1963 and 1983, and group 2, which included 19 patients treated between 1984 and 1988. The rate of resection for cystadenoma was 67% for group 1 (n = 15) and 100% for group 2 (n = 11). The operative mortality rate was 0% for both groups. Pathologically, 17 patients (69%) had serous cystadenoma, and 9 (31%) had mucinous adenoma. Except for 5 of the 15 patients from group 1 who died from other causes, all patients were healthy.

The resection rate for cystadenocarcinoma was 36% for group 1 (n = 11) and 100% for group 2 (n = 8). In 2 patients from group 1 and in 1 patient from group 2, the tumors had previously been diagnosed as benign by means of operative biopsy of the cyst wall. No operative deaths occurred in either group. Cystic neoplasm was suspected from the sonographic and CT results in 70% patients in group 2. In 2 patients (25%), pseudocyst associated with chronic pancreatitis was preoperatively diagnosed. Adenocarcinoma was diagnosed in 3 patients by means of needle biopsy and cytologic examination of pancreatic juice. In group 1, 82% (9 patients) died from recurrent carcinoma; 2 patients were alive without disease at 5 and 8 years. In group 2, 38% (3 patients) died from recurrent carcinoma, 2 patients died from other causes, and 3 patients were still alive.

Horvath and Chabot reported results of an aggressive approach to cystic pancreatic neoplasms in a retrospective cohort analysis of 25 patients with cystic neoplasms of the pancreas.²³ Data included patient demographics, presenting symptoms, operative procedure, pathologic diagnosis, perioperative morbidity and mortality rates, survival, and symptomatic follow-up data. Twenty-one patients were women, with a mean age of 60 years for the entire cohort. Mean follow-up was 24 months (range, 6 mo to 4.3 y). Complete follow-up was possible in 92%.

In 23 patients, resection was curative, and in 2, it was palliative. One patient with an uncinate mass underwent partial pancreatectomy; 4, distal pancreatectomy; and 9, distal pancreatectomy with splenectomy. Among 11 patients who required pancreatoduodenectomy, 4 had tumors involving the portal vein, which necessitated portal vein resection. Pathologic analysis revealed 12 serous cystadenomas, 4 mucinous cystadenomas, 3 mucinous cystadenocarcinomas, 5 intraductal papillary cystic neoplasms, and 1 serous cystadenocarcinoma.

The overall perioperative complication rate was 40%, with 5 major and 5 minor complications. The 11 patients who underwent pancreatoduodenectomy alone had 1 major and 4 minor complications. There were no pancreatic fistulas or portal vein thromboses and no operative mortalities. Two patients with mucinous cystadenocarcinomas died of their disease at 6 and 16 months postoperatively. All 11 pancreatoduodenectomy patients had only mild pancreatic insufficiency relieved by daily enzyme replacement.

The authors concluded that the good outcomes in this study supported an aggressive surgical approach in all patients with a cystic neoplasm of the pancreas, if they are medically fit to tolerate surgery. This approach was justified for the following reasons: (1) preoperative differentiation of a benign versus malignant tumor is unreliable, and routine testing for this purpose is of questionable utility; (2) potential adverse consequences of nonresectional therapy are significant; (3) perioperative morbidity and mortality rates are low with pancreatic surgery; and (4) the prognosis with curative resection is good.

Medicolegal Pitfalls

• Accurate diagnosis of mucinous pancreatic tumors is important because of their ability to masquerade as pancreatic pseudocysts and their high cure rate following surgical treatment.

Special Concerns

Das and associates retrospectively studied 150 patients with 166 incidentally discovered pancreatic cysts with a view to look at cystic pancreatic neoplasms with respect to the development of significant growth and to identify predictors of such growth.²⁴   The study revealed that most incidentally detected cystic neoplasms of the pancreas did not have significant growth during follow-up. The study also showed that growth is unlikely to occur before 2 years following a baseline evaluation. The authors recommend that the optimal imaging interval during follow-up of these patients should be at 2 years from the baseline evaluation. This criterion particularly applies to cystic lesions 3.0 cm or less in size and without intracystic or mural nodules.²⁴

• Search for CME/CE on This Topic »

• Pancreas Transplantation (Transplantation)
• Neoplasms of the Endocrine Pancreas (Oncology)
• Pancreas, Serous Cystadenoma (Radiology)
• Pancreas Divisum (Radiology)

• Artificial Pancreas Helpful After Pancreatic Resection
• Extended CMV Prophylaxis After Kidney, Pancreas Transplants Lowers Risk, Costs
• ADA 2009: Artificial Pancreas Showing Safety, Therapeutic Efficacy in Clinical Trials

• Early Detection of Pancreatic Cancer: Why, Who, and How to Screen
• Pancreatic Cancer Screening: State of the Art
• Improving Outcomes for Operable Pancreatic Cancer: Is Access to Safer Surgery the Problem?

• Kidney & Pancreas Transplant Resource Center
• Pancreatitis Resource Center
• Cystic Fibrosis Resource Center