H Zhu1, H Chen, S Zhang, W Peng. 1. Department of Radiology, Fudan University Shanghai Cancer Center, China; Department of Oncology, Shanghai Medical College, Fudan University, China.
Abstract
AIM: To identify the computed tomography (CT) criteria that differentiate intra-abdominal fibromatosis (IAF) from gastrointestinal stromal tumour (GIST). MATERIALS AND METHODS: CT images of 34 pathologically proven cases of IAF (n = 15) and GIST (n = 19) were retrospectively reviewed. Location, contour, border, enhancement pattern, presence of necrosis, vessels, and air within the lesion were analysed. Long diameter (LD), short diameter (SD), LD/SD ratio, degree of enhancement, and lesion/aorta (L/A) CT attenuation ratio were measured and calculated. Significant CT criteria were identified using Fisher's exact test, grouped t-test, and receiver operating characteristic (ROC) curve. Sensitivity and specificity values were calculated when single or multiple CT criteria were used. RESULTS: Extra-gastrointestinal location, ovoid or irregular contour, homogeneous enhancement, absence of intra-lesional necrosis, lower degree of enhancement, and L/A CT attenuation ratio differentiated IAF from GIST (p < 0.05). When any three of these eight criteria were combined, the sensitivity and specificity for diagnosing IAF were 100% (15 of 15) and 89.5% (17 of 19), respectively. CONCLUSION: The following eight CT criteria are helpful to differentiate IAF from GIST: extra-gastrointestinal location, ovoid or irregular contour, homogeneous enhancement, absence of intra-lesional necrosis, a degree of enhancement of less than 40.5 HU in the arterial phase versus 46.5 HU in the portal venous phase, and an L/A CT attenuation ratio <0.315 in the arterial phase versus 0.525 in the portal phase.
AIM: To identify the computed tomography (CT) criteria that differentiate intra-abdominal fibromatosis (IAF) from gastrointestinal stromal tumour (GIST). MATERIALS AND METHODS: CT images of 34 pathologically proven cases of IAF (n = 15) and GIST (n = 19) were retrospectively reviewed. Location, contour, border, enhancement pattern, presence of necrosis, vessels, and air within the lesion were analysed. Long diameter (LD), short diameter (SD), LD/SD ratio, degree of enhancement, and lesion/aorta (L/A) CT attenuation ratio were measured and calculated. Significant CT criteria were identified using Fisher's exact test, grouped t-test, and receiver operating characteristic (ROC) curve. Sensitivity and specificity values were calculated when single or multiple CT criteria were used. RESULTS: Extra-gastrointestinal location, ovoid or irregular contour, homogeneous enhancement, absence of intra-lesional necrosis, lower degree of enhancement, and L/A CT attenuation ratio differentiated IAF from GIST (p < 0.05). When any three of these eight criteria were combined, the sensitivity and specificity for diagnosing IAF were 100% (15 of 15) and 89.5% (17 of 19), respectively. CONCLUSION: The following eight CT criteria are helpful to differentiate IAF from GIST: extra-gastrointestinal location, ovoid or irregular contour, homogeneous enhancement, absence of intra-lesional necrosis, a degree of enhancement of less than 40.5 HU in the arterial phase versus 46.5 HU in the portal venous phase, and an L/A CT attenuation ratio <0.315 in the arterial phase versus 0.525 in the portal phase.