PURPOSE: To determine multi-parametric MRI features that can help differentiate malignant omental caking from benign omental thickening in the setting of portal hypertension. METHODS: We identified 19 patients with an abnormal omentum on MRI and an available reference standard: 11 patients with portal hypertension and benign omental thickening (9 male, 2 female, mean age 58 ± 6 years) and 8 patients with metastatic omental caking (4 male, 4 female, mean age 61 ± 13 years). Criteria for benign omental thickening were no evidence of malignancy for at least 24 months of follow-up (n = 7), negative ascites cytology (n = 2), or absence of malignancy on pathologic analysis of liver explant (n = 2). Criteria for omental malignancy were positive omental biopsy (n = 6) or ascites cytology (n = 2). Two radiologists (R1 and R2) evaluated characteristics of the thickened omentum on MRI. RESULTS: Findings occurring with significantly higher frequency in malignant omental caking were hyperintensity on high b-value diffusion-weighted imaging (DWI) (R1 88% vs. 0%, R2 88% vs. 0%), hyperenhancement (R1 75% vs. 0%, R2 75% vs. 0%), and convex outer omental contour (R1 88% vs. 0%, R2 75% vs. 9%) (all p ≤ 0.001); discrete omental nodules were significantly more frequent in malignant omental thickening for R1 (63% vs. 0%, p = 0.005). Features not significantly different between groups included decreased ADC, T2 hyperintensity, vessels coursing through the omentum, moderate/large volume ascites, splenomegaly, and mesenteric edema (all p ≥ 0.058). CONCLUSION: Abnormal signal on DWI, hyperenhancement, and convex outer contour are helpful MRI features to differentiate malignant from benign omental thickening.
PURPOSE: To determine multi-parametric MRI features that can help differentiate malignant omental caking from benign omental thickening in the setting of portal hypertension. METHODS: We identified 19 patients with an abnormal omentum on MRI and an available reference standard: 11 patients with portal hypertension and benign omental thickening (9 male, 2 female, mean age 58 ± 6 years) and 8 patients with metastatic omental caking (4 male, 4 female, mean age 61 ± 13 years). Criteria for benign omental thickening were no evidence of malignancy for at least 24 months of follow-up (n = 7), negative ascites cytology (n = 2), or absence of malignancy on pathologic analysis of liver explant (n = 2). Criteria for omental malignancy were positive omental biopsy (n = 6) or ascites cytology (n = 2). Two radiologists (R1 and R2) evaluated characteristics of the thickened omentum on MRI. RESULTS: Findings occurring with significantly higher frequency in malignant omental caking were hyperintensity on high b-value diffusion-weighted imaging (DWI) (R1 88% vs. 0%, R2 88% vs. 0%), hyperenhancement (R1 75% vs. 0%, R2 75% vs. 0%), and convex outer omental contour (R1 88% vs. 0%, R2 75% vs. 9%) (all p ≤ 0.001); discrete omental nodules were significantly more frequent in malignant omental thickening for R1 (63% vs. 0%, p = 0.005). Features not significantly different between groups included decreased ADC, T2 hyperintensity, vessels coursing through the omentum, moderate/large volume ascites, splenomegaly, and mesenteric edema (all p ≥ 0.058). CONCLUSION: Abnormal signal on DWI, hyperenhancement, and convex outer contour are helpful MRI features to differentiate malignant from benign omental thickening.
Authors: Amjad Alhyari; Christian Görg; Christoph Frank Dietrich; Corrina Trenker; Lena Strauch; Ehsan Safai Zadeh Journal: BMJ Open Gastroenterol Date: 2022-05