Jonathan Matthew Rubin MD, PhD, Presenter: Equipment support, ZONARE Medical Systems, Inc Research grant, General Electric Company Equipment support, General Electric Company Equipment support, Koninklijke Philips Electronics NV Consultant, Koninklijke Philips Electronics NV

Jingping Xu DPhil, Abstract Co-Author: Nothing to Disclose

Ryan W. Stidham MD, Abstract Co-Author: Nothing to Disclose

Laura Johnson BS, Abstract Co-Author: Nothing to Disclose

Peter Higgins MD, Abstract Co-Author: Nothing to Disclose

Kang Kim PhD, Abstract Co-Author: Nothing to Disclose

To assess the ability of non-linear ultrasound strain imaging to distinguish between edematous and fibrotic bowel wall in a rat model of Crohn's Disease.   This distinction is critical for determining the appropriate therapy.

A simplified tissue model based on a finite element (FE) analysis was integrated with a laboratory developed ultrasound radio frequency (RF) signal synthesis program. 2D speckle tracking was applied to this model to simulate the nonlinear behavior of the strain developed in a target inclusion over the applied average strain to the surrounding tissues. A nonlinear empirical equation was formulated and optimized to best match to the developed strain-to-applied strain relation obtained from the FE simulation.  For an animal model, acute and chronic inflammatory bowel disease was induced in Lewis rats with trinitrobenzene sulfonic acid (TNBS)-ethanol treatments. In vivo ultrasound (US) in-phase/quadrature (I/Q) data were acquired at a frame rate of about 60 frames per second (depending on depth) by using a linear ultrasound probe (L10-5, ZONARE Medical System, Mountains View, CA, USA)  Ultrasound data sets were continually collected while the abdomen was being pushed by the ultrasound probe to develop about 10% average strain over a total of 3 seconds.  The non-linear parameter in the Hill equation was used to assess in the bowel.

The  FE analysis supported the non-linear strain-strain hypothesis for edematous vs. fibrotic tissue.  In the 8 rats, the extracted non-linear parameter differentiated the three different tissue types with 1.96 ± 0.12 for normal, 1.50 ± 0.09 for the acutely inflamed, and 1.03 ± 0.08 for the chronically fibrosed tissue type. Two-tailed t-tests were all significant(normal/ fibrotic (P = 0.0000185), normal/acutely inflamed (P = 0.0013), and fibrotic /acutely inflamed (P = 0.0029)).

Even in this small study, non-linear strain estimates can clearly distinguish edematous from fibrotic bowel wall in a rat model of Crohn's Disease.   In general, non-linear elasticity should greatly increase the dynamic range of ultrasound tissue hardness measures. 

Distinguishing edematous from fibrotic bowel is critical for therapy in Crohn's Disease.  Our preliminary study suggests that non-linear ultrasound strain imaging can accomplish this.

Rubin, J, Xu, J, Stidham, R, Johnson, L, Higgins, P, Kim, K, Nonlinear Strain Elasticity Imaging for Distinguishing Edematous from Fibrotic Bowel Wall in Crohns Disease.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11006229.html