Jose T. Mora MD, Presenter: Nothing to Disclose

Pari Pandharipande MD, Abstract Co-Author: Nothing to Disclose

Raul Nirmal Uppot MD, Abstract Co-Author: Nothing to Disclose

Alexander Goehler MD, PhD, Abstract Co-Author: Nothing to Disclose

G. Scott Gazelle MD, PhD, Abstract Co-Author: Consultant, Elbit Medical Imaging Ltd (InSightec - Image Guided Treatment Ltd)

Mukesh Gobind Harisinghani MD, Abstract Co-Author: Nothing to Disclose

Marta Braschi MD, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

et al, Abstract Co-Author: Nothing to Disclose

To determine whether lack of lymphotrophic nanoparticle uptake at MR lymphangiography predicts lymph node malignancy independent of short-axis length using a logistic regression modeling approach.

Forty-four patients (36 men, 8 women; mean age = 71) with known primary malignancy (genitourinary (n=42) and breast (n=2)) who underwent both lymphotrophic nanoparticle-enhanced MRI (LNMRI) and CT-guided biopsy of a lymph node were retrospectively identified. Standard LNMRI protocol was performed prior to and 24-hours following IV administration of an iron-oxide-based lymphotropic nanoparticle. MRI features of the biopsied lymph node were re-interpreted by radiologists blinded to biopsy results. Consensus short-axis measurements were made on unenhanced MRI by two readers blinded to LNMRI; length >=1 cm was categorized as malignant. LNMRI interpretation was performed by an independent reader and solely based upon lack of nanoparticle uptake. LNMRI and short-axis sensitivity and specificity were computed with histology as the standard of reference. A logistic regression model was constructed to determine the association (odds ratio (OR)) between LNMRI and lymph node malignancy when controlling for short-axis length.

Forty-four lymph nodes were evaluated. LNMRI sensitivity and specificity were 100% (24/24) and 45% (9/20), respectively. Mean short-axis was 1.4 cm (range=0.6-2.6 cm). Short-axis sensitivity and specificity for malignancy were 57.1% (20/35) and 55.6% (5/9), respectively. In univariate analysis, malignancy at LNMRI was a statistically significant predictor of histologic malignancy (OR=40.5 (95% CI=2.2-767)), whereas prediction of malignancy by short-axis criteria did not achieve statistical significance (OR=1.7 (95% CI=0.4-7.3)). In multivariate analysis, malignancy at LNMRI remained a strong, independent, statistically significant predictor of histologic malignancy, even when controlling for short-axis length (OR=34.3 (p<.0001)).

Lymphotrophic nanoparticle-enhanced MRI strongly predicts lymph node malignancy independent of short-axis length.

MR lymphangiography can contribute to oncologic lymph node staging beyond use of lymph node morphology alone.

Mora, J, Pandharipande, P, Uppot, R, Goehler, A, Gazelle, G, Harisinghani, M, Braschi, M, et al, , et al, , Lack of Lymphotrophic Nanoparticle Uptake as an Independent Predictor of Nodal Malignancy at MR Lymphangiography: A Logistic Regression Modeling Approach.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5007239.html