Olivio Donati MD, Presenter: Nothing to Disclose

Marco Zanetti MD, Abstract Co-Author: Nothing to Disclose

Beata Bode MD, Abstract Co-Author: Nothing to Disclose

Ladislav Nagy, Abstract Co-Author: Nothing to Disclose

Andreas Schweizer MD, Abstract Co-Author: Nothing to Disclose

Christian W. A. Pfirrmann MD, Abstract Co-Author: Research Consultant, Medtronic, Inc

To evaluate the utility of dynamic gadolinium enhanced MRI (dyMRI) to assess viability of the proximal pole of the scaphoid in patients with nonunion of a scaphoid fracture.

Twenty-eight patients (mean age, 24.3±6.4 years) with nonunion of the scaphoid underwent dyMRI of the wrist 28±19 days before surgery. dyMRI consisted of 40 consecutively acquired coronal T1w images over a time of 1 minute. Two readers (R1, R2) evaluated the standard MRI (T1w, intermediate weighted images with fat saturation, T1w with fat saturation after Gadolinium enhancement) as well as the dyMRI and rated the viability of the proximal scaphoid pole on a 5-point scale. A third reader calculated the steepest upslope (linear fit of three adjacent points on the signal intensity vs. time curve) of gadolinium uptake in a ROI placed in the proximal scaphoid pole. Receiver operating characteristics curves (ROC) were calculated for the ratings of the two readers and for values of steepest upslope. Areas under the curve (AUC) were compared. Diagnostic performance in determining viability of the proximal pole was calculated for R1 and R2. Histology in 11 Patients and surgical findings served as the standard of reference.

Intraoperatively 15 (54%) scaphoid poles were viable and 13 (46%) were necrotic. Sensitivity, specificity, and accuracy in detection of scaphoid necrosis were 53.8%, 93.3% and 75% for R1 and 61.5%, 93.3% and 78.5% for R2. Interreader reliability was excellent (κ=0.92). AUC for R1 and R2 were 0.82 and 0.87, respectively. Diagnostic performance of dyMRI using the value of steepest upslope was inferior to the conventional MRI with an AUC of 0.57. Viable proximal poles enhanced in 14/15 (93.3%) cases. There was no correlation between the results of histology pattern (viable bone, necrotic bone, callus formation) and dyMRI.

Dynamic gadolinium enhanced MRI does not improve diagnostic performance in the evaluation of scaphoidal viability in patients with scaphoid nonunion.

Dynamic acquired gadolinium enhanced MRI of the wrist in patients with scaphoidal nonunion may therefore be omitted.

Donati, O, Zanetti, M, Bode, B, Nagy, L, Schweizer, A, Pfirrmann, C, Dynamic Gadolinium-enhanced MRI for Preoperative Assessment of Scaphoidal Viability in Patients with Scaphoid Nonunion.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9010856.html