Harriet Carolina Thoeny MD, Presenter: Nothing to Disclose

Maria Triantafyllou, Abstract Co-Author: Nothing to Disclose

Giuseppe Petralia MD, Abstract Co-Author: Nothing to Disclose

Johannes M. Froehlich PhD, Abstract Co-Author: Consultant, Guerbet SA

To prospectively compare the diagnostic performance of 3 functional MRI methods (ultrasmall superparamagnetic particles of iron oxide-enhanced MRI(USPIO); diffusion-weighted MRI(DWI); combined USPIO / DWI (USPIO-DWI)) to detect pelvic lymph node (LN) metastases in normal-sized nodes.

A consecutive series of 75 pts with bladder (n=19), prostate cancer (n=48) or both (n=8) and normal-sized LNs on crossectional imaging underwent two separate MRI exams of the pelvis at 3T after written informed consent. Imaging included T1-and T2-w 3D-sequences with isotropic voxels of 0.75/1 mm3, respectively and transverse DWI of the entire pelvis ( 3b-values: 0,500,1000sec/mm2). After the first MRI USPIO (2.6mg Fe/kg bw) was administered i.v. followed by a second MRI with the identical protocol 24-36 hrs thereafter. All pts underwent template lymphadenectomy. Image analysis was performed by 3 different independent readers separately for all 3 functional methods. Results of  the mean diagnostic performance were reported on a per patient basis for each method separately and correlated to histopathology.

A total of 2993 LNs were resected and 54 metastatic LNs were detected in 20/75 pts (26.7%) on histopathology. On a per patient level the readings of each of the 3 methods yielded the following mean diagnostic results: sensitivity: USPIO= 58.5%, DWI=78.9%, USPIO-DWI= 70.0%; specificity: USPIO=83%, DWI=81.5%, USPIO-DWI=94%; PPV: USPIO=58%, DWI=58.9%, USPIO-DWI=80.8%; NPV: USPIO=84.4%, DWI=92.1%, USPIO-DWI=89.6%; diagnostic accuracies: USPIO=76.4%, DWI=80.9%, USPIO-DWI=87.6%, respectively. The vast majority of the missed metastases had a short axis diameter < 5mm on histopathology. The false positive LNs attributed on USPIO and DWI were mainly due to  fibrosis, lipomatosis or histiocytosis.

DWI allows detection of LN metastases in normal-sized LNs without contrast medium administration in a high number of pts. USPIO-DWI further decreases the number of false positive LNs and facilitates reading, however needs contrast medium at the expense of higher costs and invasiveness. USPIO alone is inferior and is therefore not recommended.

Improvement of LN staging in normal sized LNs would allow to omit extended pelvic LN dissection in case of negative findings. It  would allow to guide the surgeon in case of suspicious LNs or change treatment to adjuvant chemotherapy or hormonal therapy instead of surgery alone.

Thoeny, H, Triantafyllou, M, Petralia, G, Froehlich, J, Functional MRI to Detect Metastases in Normal-sized Pelvic Lymph Nodes: Which Technique Is the Best?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14013423.html