Secondary lymphedema (LE) can develop after treatment for any cancer which involves the removal or irradiation of lymph nodes. LE is chronic swelling which is characterized, at least in its early phase, by an increase in extracellular fluid. As LE progresses, changes occur in cellular architecture: there is an accumulation of adipose tissue and development of tissue fibrosis¹. The standard treatment for LE is limb compression, however the appropriate pressure required for optimum effect and indeed the actual effects of compression on lymphatic drainage and tissue water distribution are unknown. Despite the signature changes in tissue water composition, diffusion weighted MRI (DWI) has not been investigated for assessment of lymphedema. The aim of this study was to examine the effectiveness of DWI in identifying changes to tissue structure in the forearm due to LE.Three patients with confirmed LE were imaged in a GE Discovery MR750W 3.0 T clinical MRI scanner using an 8-channel phased array coil and a single-axis DWI sequence with TE/TR = 85/2000 ms. 0.55 × 0.55 mm isotropic voxels were acquired in three gradient directions with a slice thickness of 10 mm at b-values of 0, 100, 200, 400, 800, and 1600 s/mm­². Standard deviation of noise was determined using voxels outside the imaging forearm with estimated SNR_b=0 = 28. Proton density (PD) scans with same planes as DWI and 0.14 × 0.14 mm voxels were obtained for each for reference. Nonlinear parametric regression analysis was performed on all DWI voxel data using monoexponential (ADC) and kurtosis models as in ².Representative slices from the normal and edematous arms of patient 3 are shown in Fig. 1. The typical honeycomb-like structure of LE can be seen in the PD images in areas of fatty tissue, especially in the regions of interest (ROI) indicated in red. ADC is elevated in the edematous tissue. Kurtosis adjusted diffusivity (D_k) is also elevated in LE, though less obviously than ADC. Kurtosis (K) is low in LE relative to fat and muscle. Similar trends trends were observed in the arms of the other two patients.The increase in ADC in edematous tissue is consistent with an increased partial volume of freely diffusing interstitial water. This also correlates with a shift of the kurtosis toward zero (free water) and is unusual in regard to the majority of tissue pathologies where an increase in kurtosis is generally observed. Additionally, the diffusion parameter changes do not correspond exactly with the honeycomb regions on the PD scans, suggesting that the structural changes revealed on PD imaging do not reliably map the buildup of interstitial water. This suggests that the DWI may be identifying additional changes to the tissue structure currently not identified and potentially useful for understanding and optimizing limb compression treatments for LE.DWI suggests that lymphedema results in tissue changes not previously detected on proton density imaging. DWI may be a useful tool for characterization of lymphedema and optimization of compression based treatments. Future studies will assess tissue changes before and after intervention and treatment.