Single- and multiparameter magnetic resonance imaging for diagnosing and severity grading of chronic pancreatitis

• Published in: Abdominal imaging 2023-02, Vol.48 (2), p.630-641
• Main Authors: Olesen, Søren Schou, Steinkohl, Emily, Hansen, Tine Maria, Drewes, Asbjørn Mohr, Frøkjær, Jens Brøndum
• Format: Article
• Language: English
• Subjects: Diabetes; Diabetes mellitus; Diffusion Magnetic Resonance Imaging - methods; Exocrine Pancreatic Insufficiency; Gastroenterology; Hepatology; Humans; Imaging; Magnetic induction; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Medical diagnosis; Medical imaging; Medicine; Medicine & Public Health; Pancreas; Pancreas - diagnostic imaging; Pancreatitis; Pancreatitis, Chronic - diagnostic imaging; Parameters; Radiology; Regression analysis; Relaxation time; Resonance; Stiffness; Subgroups
• ISSN: 2366-0058; 2366-004X; 2366-0058
• DOI: 10.1007/s00261-022-03760-6

Purpose The study aimed to determine the performance of advanced magnetic resonance imaging (MRI), including a multiparametric MRI-index, for diagnosing and severity grading of chronic pancreatitis (CP) at various functional stages with focus on detection of CP with preserved pancreatic function. Methods Fifty-four CP patients and 35 healthy controls underwent MRI including assessment of pancreatic volume, main pancreatic duct (MPD) diameter, T1 relaxation time, magnetic resonance elastography (MRE) derived stiffness, and intravoxel incoherent motion (IVIM) diffusion-weighted imaging. Patients were categorized into three subgroups: Preserved pancreatic function ( n  = 14), partial pancreatic insufficiency (exocrine insufficiency or diabetes, n  = 25), and complete pancreatic insufficiency (exocrine insufficiency and diabetes, n  = 15). A multiparametric MRI-index was based on ordinal logistic regression analysis. Diagnostic performances of MRI parameters for diagnosing CP at different functional stages were determined using receiver operating characteristic (ROC) analysis. Results All MRI parameters differed across CP subgroups and healthy controls (all P   0.75), with the MRI-index tending to outperform MRE (ROC-AUC 0.77 vs. 0.63; P  = 0.10). Conclusion Quantitative assessments of T1 relaxation time and MRE-derived stiffness seem promising for diagnosing CP at different functional stages and may together with multiparametric MRI-index be used for early identification, staging and monitoring of CP. Graphical abstract