MRI follow‑up for pancreatic intraductal papillary mucinous neoplasm: an ultrashort versus long protocol
Johansson K, Mustonen H, Nieminen H, Haglund C, Lehtimäki TE, Seppänen H.
Abdom Radiol (NY). 2022 Feb;47(2):727-737. doi: 10.1007/s00261-021-03382-4.
Intraductal papillary mucinous neoplasms (IPMNs) are mucin-producing tumors that arise from the pancreatic duct and are considered precursor lesions to pancreatic cancer [1]. There are three types of IPMNs differentiated based on morphologic differences.
with different risk profiles for malignant transformation. The Main duct IPMNs (MD-IPMN) are characterized by involvement with the main pancreatic duct (MPD) and identified by a dilated MPD (5 mm) without an associated cyst or other cause for ductal obstruction. Branch duct IPMNs (BD-IPMN) arise from a branch of the MPD and are identified as unilocular or multilocular pancreatic cysts with communication with the MPD, which measures <5 mm. Mixed-type IPMNs (MT-IPMN) meet the criteria for both MD and BD IPMNs [2].
Given that approximately an average of 8% of all pancreatic cancers arise from these lesions, IPMN follow-up offers an opportunity for early cancer detection [3]. Most IPMNs are accidentally detected when abdominal imaging is performed for another indication, as they are mostly asymptomatic. The presence of some clinical symptoms, such as abdominal pain or bloating, has been associated with malignant transformation in patients with IPMNs [4, 5]. Abdominal MR and MR cholangiopancreatography (MRCP) is the investigation of choice for initial characterization and follow-up of IPMN, thanks to its high contrast and spatial resolution and lack of ionizing radiation. However, MR examinations suffer from long acquisition time and, the spreading of MR short and ultra-short protocols is reaching interest also in IPMNs follow-up; however, the possibility to optimize MR protocols for IPMNs follow-up needs to be investigated [6].
The authors of the present study aimed to assess whether an MR ultrashort magnetic protocol (USP) that includes only axial T2-weighted half-Fourier single-shot turbo spin-echo (HASTE) sequences and 3D MRCP sampling perfection with application-optimized contrasts using different flip angle evolution (SPACE) sequences is comparable in terms of diagnostic accuracy with longer protocols in the identification of worrisome features (WF) and high-risk stigmata (HRS) and, in the measurement of the major diameter of the largest cyst and main pancreatic duct (MPD). Reductions in cost and time related to USP were also assessed.
This retrospective study included a final population of 112 surveillance patients evaluated for pancreatic cysts at Helsinki University Hospital who underwent examination with 1.5 Tesla MRI. Three MRI datasets were analyzed including USP as abovementioned then, long protocol (LP) including axial T2-weighted HASTE with and without fat saturation (fs), thin and thick coronal T2-weighted HASTE fs, axial T1-weighted fast low angle shot (FLASH) fs, axial T1-weighted in/opposed phase, T2-weighted 3D MRCP SPACE and maximum intensity projection (MIP) reconstruction, diffusion-weighted imaging (DWI), and T1-weighted fs volumetric interpolated breath-hold (VIBE) before and after the administration of the contrast agent; and, finally short protocol (SP) including axial T2-weighted HASTE, thin and thick coronal T2-weighted HASTE fs, axial T1-weighted FLASH fs, 3D MRCP SPACE, and MIP.
Two radiologists evaluated the three datasets for each patient in a random order to reduce recall bias. The following parameters were evaluated from the datasets: cyst and MPD size, cystic and MPD mural nodules, abrupt change in the MPD caliber, distal parenchymal atrophy, lymph node enlargement, and a solid pancreatic tumor presence. The WF and HRS analysis included cystic and MPD mural nodules, an MPD size of ≥ 5 mm, an abrupt change in the MPD caliber with distal pancreatic atrophy, a cyst of ≥ 3 cm, and/or lymphadenopathy.
Results showed agreement between longer protocols and USP for cystic mural nodules, MPD mural nodules, solid pancreatic tumor, and WF/HRS (94.9%, 99.1%, 99.1%, 92.4% respectively). Non-significant differences were reported in the largest cyst measurement nor in the mean MPD value. In addition, no significant systematic bias in the difference between methods and the confidence intervals was observed.
Then, all true cases were identified using USP and S-LP, and all three datasets led to identifying true cystic mural nodules. All the measurements performed showed an agreement from moderate to strong between the two readers.
Regarding time and cost analysis total sequence acquisition time was on average 23 minutes for LP, 13 minutes for SP, and 7 minutes for USP, while USP was estimated, based on Hospital costs, as 77% that of SP and 39% of LP.
Despite the interesting results, this study had some limitations such as the retrospective nature, the lack of complete sampling of IPMN with WS/HRS also due to the conservative approach in an elderly population, lack of analysis performed with 3.0T that might differ from 1.5T, and the aspect related to the different costs between three protocols that were estimated at the University Hospital of Helsinki, and, therefore, price variations may exist between hospitals and countries.
The study carried out and the results obtained could be very relevant in clinical practice in terms of time and cost saving, without affecting diagnostic assessment. Enabling the use of USP for surveillance of IPMN without WF/HRS might allow for improving clinical workflow without the risk of missing important diagnostic information.
References:
1. Keane MG, et al. A Review of the Diagnosis and Management of Premalignant Pancreatic Cystic Lesions. J Clin Med. 2021 Mar 19;10(6):1284.
2. Furukawa T, et al. Classification of types of intraductal papillary-mucinous neoplasm of the pancreas: a consensus study. Virchows Arch. 2005 Nov;447(5):794-9.
3. Pergolini I, et al. Long-term Risk of Pancreatic Malignancy in Patients With Branch Duct Intraductal Papillary Mucinous Neoplasm in a Referral Center. Gastroenterology. 2017 Nov;153(5):1284-1294.e1.
4. Fernández-del Castillo C, at al. Incidental pancreatic cysts: clinicopathologic characteristics and comparison with symptomatic patients. Arch Surg. 2003 Apr;138(4):427-3; discussion 433-4.
5. Kim JR, at al. Clinical implication of serum carcinoembryonic antigen and carbohydrate antigen 19-9 for the prediction of malignancy in intraductal papillary mucinous neoplasm of pancreas. J Hepatobiliary Pancreat Sci. 2015 Sep;22(9):699-707.
6. Delaney FT, at al. An abbreviated MRI protocol for surveillance of cystic pancreatic lesions. Abdom Radiol (NY). 2021 Jul;46(7):3253-3259.
Dr. Carlotta Rucci is a radiology resident at "Sapienza" University of Rome - Sant'Andrea University Hospital in Rome, attending the fourth year of residency. Her main fields of interest in the area of diagnostic imaging are Abdominal Oncologic Imaging, Cardiovascular Imaging and, Breast Imaging. She has also been actively involved in clinical imaging research, and she has contributed as a co-author in some publications.
Comments may be sent to