Andrew B. Rosenkrantz MD, Presenter: Nothing to Disclose

Laura Heacock MS, MD, Abstract Co-Author: Nothing to Disclose

James S. Babb PhD, Abstract Co-Author: Nothing to Disclose

To evaluate factors associated with the likelihood that abdominopelvic MRI examinations performed for characterization of lesions identified on other imaging modalities will provide information with potential to add value to patient management.

1,132 abdominopelvic lesions in 863 patients in which MRI was performed for further characterization following detection by an alternate imaging modality were included in this retrospective study. Reports of the MRI examinations and of the prior studies were reviewed to classify cases in terms of patient, examination, and lesion related factors. The MRI reports were also classified in terms of various measures reflecting inclusion of content with potential to add value to patient management. Data was analyzed using logistic regression for correlated data.

MRI provided a definitive diagnosis (DD) for 79.2% (897/1132) of lesions, upgraded the severity of the favored diagnosis in 6.2% (70/1132) of lesions, downgraded the severity of the favored diagnosis in 34.5% (390/1132) of lesions, and showed an absence of the suspected lesion in 12.0% (136/1132) of lesions. Provision of a DD was significantly associated with the organ containing the lesion (p<0.001); lesions in the liver, kidney, ovary, gallbladder, bowel, myometrium, and adrenal gland had a DD provided in 73.0%-87.8% of instances, while lesions in the pancreas, spleen, endometrial or cervical canal, or retroperitoneum/mesentery had a DD provided in only 30.0%-51.7% of instances. Change in severity of the favored diagnosis was significantly associated with the prior imaging modality (p≤0.001); this occurred in 47.0-47.5% of instances following CT or PET/CT, compared with 35.5% of instances following ultrasound. Absence of the suspected lesion was significantly associated with the prior imaging modality (p≤0.001); this occurred in 19.3% of instances following PET/CT, compared with 14.8% of instances following ultrasound and 6.2% of instances following CT.

Abdominopelvic MRI examinations performed for further lesion characterization may add value to clinical management in a high fraction of cases, the likelihood of which is influenced by factors related to the given examination.

Policy decisions that impact MRI utilization should recognize factors impacting likelihood of added value, rather than the historical approach of treating all utilization in a homogeneous fashion.

Rosenkrantz, A, Heacock, L, Babb, J, Abdominopelvic MRI for Lesion Characterization: Factors Associated with Likelihood of Added Value.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13011186.html