From the AGA Journals

VIDEO: Newer MRI hardware, software significantly better at detecting pancreatic cysts

View on the News

Newer MRIs much better at detecting pancreatic cysts

The increasing prevalence of pancreatic cystic lesions on MRI scanning may provide an important opportunity for detection of early precursors of pancreatic cancer – or may represent just another insignificant incidental finding. What is the implication of a small asymptomatic cyst?

MRI scanning of the pancreas has revolutionized our ability to detect early cystic neoplasms of the pancreas. Pancreatic cysts appear as well-defined, small, round fluid-filled structures within the pancreas. The inner structures – such as septations, nodules, and adjacent masses – offer clues as to the type of cyst and the risk of malignancy. But the real strength of pancreatic MRI scanning is the ability to detect and portray small cysts and the adjacent main pancreatic duct.

The size, number, and distribution of cysts over time can be tracked with MRI surveillance. By tracking the diameter of cysts and calculating the rate of growth of cysts, clinicians may be able to predict the development of malignancy in intraductal papillary mucinous neoplasms.

How should these patients be managed clinically? Once a cyst has been identified, are clinicians obligated to notify the patient, monitor the cyst with an established surveillance program, or biopsy the cyst? If the cyst is very small and benign appearing, can the clinician ignore the finding and perhaps not notify the patient?

Once again, we are watching dilemmas unfold as technology outstrips our understanding of diseases and their management. We are going to need some good correlations between imaging and tissue of pancreatic cystic lesions. In the meantime, it is important to reserve the use of pancreatic MRI scanning to high-risk patients or patients with CT scan abnormalities.

Dr. William R. Brugge, AGAF, is professor of medicine, Harvard Medical School, and director, Pancreas Biliary Center, Massachusetts General Hospital, both in Boston. He is a consultant with Boston Scientific.


 

FROM CLINICAL GASTROENTEROLOGY AND HEPATOLOGY

As magnetic resonance imaging technology continues to advance year after year, so does MRI’s ability to accurately detect pancreatic cysts, according to a new study published in the April issue of Clinical Gastroenterology and Hepatology (doi: 10.1016/j.cgh.2015.08.038).

“To our knowledge, this is the first study to analyze the relationship between the technical improvements in imaging techniques (specifically, MRI) and the presence of incidentally found PCLs [pancreatic cystic lesions],” said the study authors, led by Dr. Michael B. Wallace of the Mayo Clinic in Jacksonville, Fla.

Dr. Michael B. Wallace

Dr. Michael B. Wallace

Dr. Wallace and his coinvestigators launched this retrospective descriptive study selecting the first 50 consecutive abdominal MRI patients at the Jacksonville Mayo Clinic during January and February of each year from 2005 through 2014, for a total of 500 cases who met inclusion criteria included in the study. Patients were excluded if they had preexisting symptomatic or asymptomatic pancreatitis, either acute or chronic, pancreatic masses, pancreatic cysts, pancreatic surgery, pancreatic symptoms, or any pancreas-related indications found by MRI.

The clinic underwent periodic MRI updates over the course of the 10-year study, along with requisite software updates to “take advantage of the new hardware technology,” the study explains. Major hardware improvements, provided by Siemens Medical Solutions USA, were Symphony/Sonata, Espree/Avanto, and Aera/Skyra, while software updates corresponding to each hardware update were VA, VB, and VD, respectively.

SOURCE: AMERICAN GASTROENTEROLOGICAL ASSOCIATION

Furthermore, each software update had other, smaller upgrades, leading to a total of 20 combinations of MRI hardware and software on which MRIs were performed over the 10 years. Every MRI taken included “an axial and a coronal T2-weighted single-shot (HASTE) pulse sequence [with] TR 1400-1500 ms, TE 82-99 ms, and slice thickness 5-7 mm (gap, 0.5-0.7 mm).” Each MRI was analyzed by a pancreatic MRI specialist to find incidental cysts.

The number of patients found with pancreatic cysts increased incrementally from 2005 to 2014, with 2010 being the year with the highest number. A total of 208 subjects (41.6%) were found to have incidental cysts, but only 44 of these cases were discovered in the original MRI. The presence of cysts was associated with older age in patients who had them; only 20% of all subjects under 50 years of age had cysts, compared to 32.4% of those between 50 and 60 years, 54.9% of those between 60 and 70 years, and 61.5% of those over the age of 70 years (P less than .01).

Additionally, 56.4% of all subjects with diabetes (P less than .01), 59.0% of subjects with nonmelanoma skin cancer (P less than .03), and 58.1% of those with hepatocarcinoma (P less than .02) were also found to have cysts. Most striking, however, is that newer hardware and software permutations were able to detect cysts in 56.3% (Skyra) of patients who had them, compared with only 30.3% (Symphony) of patients who underwent MRI on older technology.

“The variable field strength” (1.5 T vs. 3 T) was not significantly related to the presence of PCLs,” Dr. Wallace and his coauthors concluded. “We believe this may be secondary to the lack of power of the analysis, because only 6% of the examinations were 3-T studies. Therefore, we speculate that this relationship may be confirmed if the number of 3-T studies increased.”

Males and females each made up roughly 50% of the study population, with a median age of 60 years and 85% being white. Additionally, 4% of subjects had a family history of pancreatic cancer, 12% had a personal history of solid organ transplant, and 53% had a personal history of smoking.

This study was funded by the Mayo Clinic. Dr. Wallace disclosed that he has received grant funding from Olympus, Boston Scientific, and Cosmo Pharmaceuticals, and travel support from Olympus. No other authors reported any financial disclosures.

[email protected]

Next Article: