Abbreviated non-enhanced magnetic resonance imaging in patients with acute necrotizing pancreatitis
Akash Bansal, Rajath Ramegowda, Pankaj Gupta, Jimil Shah, Jayanta Samanta, Harshal Mandavdhare, Vishal Sharma, Rakesh Kochhar, Manavjit Singh Sandhu
Abdominal Radiology (2022) doi: 10.1007/s00261-022-03531-3
Acute pancreatitis is one of the commonest differentials of acute abdomen which presents to accident and emergency (1). It can be divided into interstitial oedematous pancreatitis and acute necrotizing pancreatitis (1). The Revised Atlanta Classification classifies pancreatic fluid collections according to the presence of necrosis and the interval from pain of onset (2).
CT is the commonest imaging modality used to determine the morphology of pancreatitis and its complications, however the use of MRI is inevitable in some cases. MRI is used in complicated cases as a supplement to CT, and in cases where CT is contra-indicated (such as contrast allergy or acute kidney injury) or when it is advisable to reduce radiation dose (such as in pregnancy or the need for multiple scans) (3). The patients requiring MRI are generally unwell and therefore the long acquisition time often leads to unreliable images of reduced quality.
Bansal et al carried out a retrospective study to assess an abbreviated MRI protocol based on a single axial T2 weighted sequence. They conducted a retrospective analysis of patients with acute necrotizing pancreatitis who underwent MRI between January 2017 and November 2020. Patients with moderately severe or severe acute pancreatitis who underwent MRI within two weeks of admission were included. The exclusion criteria were those with mild or chronic pancreatitis, and those whose MRI was performed later than 2 weeks following admission.
The complete MRI (cMRI) protocol included the following sequences: T2 weighted half Fournier acquisition single shot turbo spin echo (HASTE) axial and coronal, true fast imaging with steady state free precision axial and coronal, T1 weighted gradient echo axial sequence, coronal 2D and 3D MRCP sequences, and DWI with ADC maps. Contrast enhanced axial and coronal T1 weighted sequences using Magnevist were also acquired in some cases. For the abbreviated MRI (AMRI) protocol, only the T2 HASTE axial sequence was evaluated.
Two separate sets of data (AMRI and cMRI) were randomly presented to and independently analyzed by two radiologists who were blinded to the clinical details and to any previous imaging.
The following parameters were assessed: intrapancreatic and extrapancreatic necrosis, percentage of intrapancreatic necrosis (<30% or >30%), presence, site and size of collection, presence of wall, presence and percentage of solid components in the collection, venous thrombosis, presence of ascites, pleural effusion, gallstones, common bile duct stones, main pancreatic duct dilatation and disconnected pancreatic duct.
28 patients were included in the study. Overall, there was no statistical difference in the identification of pancreatic necrosis between AMRI and cMRI, however AMRI under-estimated pancreatic necrosis in one patient. The collections were accurately detected, however the mean size was larger on cMRI compared to AMRI. Disconnected pancreatic duct was seen on cMRI more than AMRI, however this difference was not statistically significant.
Overall acquisition time was 25 seconds for AMRI and 27-32 minutes for cMRI with contrast and 23-27 minutes without contrast. cMRI had some sequences (particularly the MRCP and post-contrast) which were not diagnostic due to respiratory motion artifact.
In conclusion, the authors found that AMRI may be a suitable alternative to cMRI in patients with necrotizing pancreatitis, especially in those with severe disease, as it can be acquired rapidly without the need for contrast injection. The study was limited by the small cohort, therefore further prospective studies with a larger patient cohort should be conducted to support their preliminary observations.
References:
- Banks, P. A., Bollen, T. L., Dervenis, C., Gooszen, H. G., John-son, C. D., Sarr, M. G., Tsiotos, G. G., Vege, S. S., & Acute Pancreatitis Classifcation Working Group (2013). Classifcation of acute pancreatitis--2012: revision of the Atlanta classification and defnitions by international consensus. Gut, 62(1), 102–111.
- Foster BR, Jensen KK, Bakis G, Shaaban AM, Coakley FV (2016) Revised Atlanta Classifcation for Acute Pancreatitis: A Pictorial Essay. Radiographics 36(3):675-87.
- Shyu, J. Y., Sainani, N. I., Sahni, V. A., Chick, J. F., Chauhan, N. R., Conwell, D. L., Clancy, T. E., Banks, P. A., & Silverman, S. G. (2014). Necrotizing pancreatitis: diagnosis, imaging, and intervention. Radiographics, 34(5), 1218–1239.
Dr Stephanie Arrigo is a higher specialist trainee in the Medical Imaging Department of Mater Dei Hospital (Malta) currently preparing for her final exams. She has a special interest in cross sectional imaging.
Comments may be sent to: