So you've discovered a pancreatic cyst …

Internists should follow a stepwise progression of diagnosis and management of pancreatic cysts.


The first step in managing an incidentally discovered pancreatic cyst is to consider how it was found, said Michelle A. Anderson, MD, MSc, at an Internal Medicine Meeting 2019 session titled “Pancreas Problems: Unraveling the Mysteries.”

“If the test that discovered the cyst was not a pancreatic-protocol CT—[for example] your patient goes into the ER because you think they have kidney stones, and you get back the report that they have a cyst—that's almost always going to be a noncontrast study,” she said.

Surveillance of pancreatic cysts is the area that most internists will be dealing with in their practices said Michelle A Anderson MD MSc Photo by Kevin Berne
Surveillance of pancreatic cysts is the area that most internists will be dealing with in their practices, said Michelle A. Anderson, MD, MSc. Photo by Kevin Berne

The next step in this case is to order a pancreatic-protocol CT or MRI/magnetic resonance cholangiopancreatography (MRCP), or refer for endoscopic ultrasound (EUS). MRCP is preferred over CT because the latter is less sensitive for defining and detecting cystic neoplasms, said Dr. Anderson, an associate professor at Michigan Medicine at the University of Michigan in Ann Arbor.

However, these additional tests are not warranted in certain situations, she said. If there is the classic appearance of a serous cyst, which has many microcompartments and resembles a honeycomb, the patient has a serous cystadenoma. “These have no risk of becoming cancerous and don't need further evaluation,” Dr. Anderson said.

Additional testing is also not warranted in patients who aren't able to undergo surgery, Dr. Anderson noted. “Why would we spend more time and money doing MRI or further testing or endoscopic ultrasound in a person who is not a surgical candidate? Just document that that's your decision making in the chart and move on.”

Similarly, a patient with a pseudocyst and a history of acute pancreatitis doesn't need further evaluation, because most cysts in the setting of acute pancreatitis resolve independently, she noted.

However, she cautioned that patients who present with acute-onset pancreatitis, no pancreatitis history, and a newly discovered pancreatic cyst are in a different category altogether.

“Cysts that are formed in the setting of acute pancreatitis take time to wall off. That fluid collection has to get a wall around it. And so you should be thinking ... this is a patient that already had a cyst, and that's a warning sign. That is probably a neoplastic cyst, and it's causing the acute pancreatitis,” Dr. Anderson said.

Keep in mind that while most cystic neoplasms of the pancreas are asymptomatic, patients might begin reporting symptoms after they hear they have a cyst, said Dr. Anderson. “They're going to go online—all of our patients do this—they're going to Google this, and then they're going to come in scared, because they think that they have pancreatic cancer. And then they're going to start to have these nondescript symptoms: ‘I think I have some abdominal pain. It hurts over here,’” she said. “Be careful about ascribing symptoms to [these cysts], and mostly reassure the patient. Most of these cysts have low cancer risks.”

What to look for on imaging

Worrisome imaging findings include cysts 3 cm in diameter or larger, an abrupt change in the main pancreatic duct, main-duct dilation that is 5 mm or more in diameter and focal dilation of the main pancreatic duct, and a mural nodule or a solid component. “Any of the patients that have features like this, either signs or symptoms or imaging characteristics such as this, should be referred to a multidiscipline center and/or undergo EUS, possibly with biopsy,” Dr. Anderson said.

Patients should also be referred to EUS or to a subspecialist if they present with jaundice secondary to a cyst or have elevated serum CA19-9 levels along with a cyst, she noted.

Otherwise, EUS is helpful in evaluating pancreatic cysts if a CT or an MRI was indeterminate or if there is cutoff of the common bile duct or the main pancreatic duct, Dr. Anderson said. “One of the things that … we have seen in patients that have these cystic neoplasms is that they're at an increased risk for a synchronous lesion elsewhere in the pancreas, a solid adenocarcinoma. And we don't understand why, it's probably a field defect, and they just have this increased risk. And so EUS can detect these small tumors,” she said.

EUS can also help identify cysts based on appearance. In contrast to the honeycomb appearance of serous cysts, mucinous cysts are often macrocystic and can be uniocular or multiocular, and intraductal papillary mucosal neoplasms (IPMNs) have mural nodules that can be targeted for fine-needle biopsy, Dr. Anderson said.

She discussed two cases, the first a 55-year-old man who was listed for a kidney/pancreas transplant. He was asymptomatic and had undergone cross-sectional imaging as part of his pretransplant evaluation, which revealed a cystic pancreatic lesion with different compartments. On biopsy, carcinoembryonic antigen (CEA) was measured at 1,400, identifying the cyst as mucinous. “There's no question that's what it was,” Dr. Anderson said. “Because the main duct was normal in diameter … that by definition means that this is a side-branch IPMN.” This type of cyst carries a low risk for transformation to pancreatic cancer, less than 15%, Dr. Anderson said, so the patient was recommended for transplant.

In contrast, an 86-year-old woman who had fallen at home underwent imaging in the ED and was found to have a cyst with a mural nodule measuring almost a centimeter in size. The main pancreatic duct had a diameter of 11.2 mm versus a normal diameter of 3 mm. Endoscopic evaluation revealed a main-duct IPMN and significant amounts of mucus emanating from the pancreas.

“That mucus, that thick tenacious mucus, is actually what we think causes pancreatitis in some of these patients,” Dr. Anderson said. “It plugs up the ducts and leads to an actual episode of pancreatitis.” Patients with these imaging findings have a high lifetime risk of pancreatic cancer, at least 40%, she noted. In this case, no biopsy was necessary, and the patient opted against a pancreatectomy. Given her age and the associated risks, Dr. Anderson said, “I don't know that that was a bad decision.”

What to watch for

Surveillance of pancreatic cysts is the area that most internists will be dealing with in their practices, Dr. Anderson said. She provided a quick overview of patients in whom surveillance is not needed: those with symptomatic lesions (they should be referred for multidisciplinary evaluation immediately), those with high-risk features who are good candidates for surgery, and those who are poor candidates for surgery or decline it.

Management and surveillance of unresected cysts are primarily guided by size, Dr. Anderson said. “A lot of these patients will have more than one cyst, right? They'll have multifocal disease; they'll have a seven-millimeter, a centimeter, a two-centimeter one … you should be guided by the diameter of the largest cyst.”

Cysts that are less than a centimeter in size can be scanned every two years for four years, versus those greater than 3 cm, which should be referred to a multidisciplinary team. If patients aren't referred, however, MRI alternating with EUS should be done every six months for three years.

The next thing to consider is whether the cyst is remaining stable, Dr. Anderson said. Cysts that stay stable in size after three to five years can be imaged every two or three years. Some guidelines, like those from the American Gastroenterological Association (AGA), recommend that follow-up can be discontinued after five years of no change, while the American College of Gastroenterology (ACG) takes a more conservative approach, she said.

“I will tell you that very few gastroenterologists are comfortable with discontinuing, especially, say, in a middle-aged patient that has these cysts. I have seen personally doctors that stopped watching these and patients transformed into cancer, so I don't know that I agree with the AGA guideline,” Dr. Anderson said. “I think the ACG guideline is middle-of-the-road in terms of its aggressiveness and its recommendations, and this is the guideline that I follow.”

Internists following unresected cysts with interval imaging should refer patients who develop new-onset diabetes, especially if they have a normal body mass index, Dr. Anderson said. “That should be ringing a bell in your head. That should be a red flag to you that something's changed,” she said. In addition, many cysts will grow over time, but the cutoff for acceptable growth is less than 3 mm per year. “If they grow three or more millimeters per year, that's a worrisome sign and you should refer that patient,” she said.

Patients who have undergone resection and have been diagnosed with a mucinous cystic neoplasm (MCN) do not need surveillance, Dr. Anderson said. “MCNs don't recur. If they have a lesion resected completely, they don't need any further follow-up. On the other hand, if the patient has IPMN, that's a field defect. It affects the entire pancreas, and those people require surveillance.”

Those with low-grade dysplasia or moderate dysplasia should undergo a history and physical and MRI or EUS every six months. Reoperation should be considered if high-grade dysplasia is noted on the final pathology report, Dr. Anderson said.

“Some of these patients will have cancer in their resection. If that happens, and that's your patient, then you should follow the NCCN [National Comprehensive Cancer Network] guidelines that you would if the patient just had run-of-the-mill pancreatic cancer,” she said. “And it's never wrong to consider referring to a pancreatologist or a multidiscipline team.”