Matthew Allen Lewis PhD, Presenter: Nothing to Disclose

Beth A. Furey MD, BEng, Abstract Co-Author: Nothing to Disclose

Ananth Madhuranthakam PhD, Abstract Co-Author: Nothing to Disclose

April Alexander Bailey MD, Abstract Co-Author: Nothing to Disclose

Gaurav Khatri MD, Abstract Co-Author: Nothing to Disclose

Diane Mary Twickler MD, Abstract Co-Author: Nothing to Disclose

At our tertiary referral center, identifying the presence and extent of implanted mesh on clinical imaging has become paramount as patients are increasingly referred to us for complications including pain and infection after multiple surgical interventions and partial or incomplete mesh resection. Although surgical mesh is not readily apparent or differentiable from dense scar on clinical MR, it is identifiable in phantoms using US and MRI. We are using phantoms to develop analysis tools to exploit the regular structure of these surgical devices with the goal of improving detection and localization.

Imaging evaluation of surgical mesh implanted for repair of incontinence and pelvic organ prolapse is an important medical and legal problem. To date, the optimal modality for imaging evaluation of mesh in the pelvis has been ultrasound (US), but anatomic location and extent is limited on this modality. We sought to evaluate two approaches to improve post-surgical mesh evaluation including US with co-registered MRI for improved anatomic localization, and the development of US/MRI techniques that incorporate a priori information on the structure of the surgical mesh to improve detection. We describe preliminary experiments in phantoms containing surgical mesh.

Two surgical mesh samples (coarse and fine) were attached to 3D printed polymer supports using sutures and embedded in gelatin US phantoms. Both mesh phantoms were imaged simultaneously in a Philips Achieva 3T MRI using various pulse sequences (T2W_TSE, T1W_FFE 3D, 3D_PD, FFE UTE). Phantoms were imaged serially with the Verasonics US research platform (Redmond, WA) using an L7-4 ultrasound transducer. Volumetric images were acquired through linear robotic motion in 1 mm steps. Visualization and analysis of the co-registered images were performed using OsiriX (Pixmeo, Geneva) and NIH ImageJ.

In these studies, both meshes were readily apparent with both modalities (Figure 1). Only the US images demonstrated the characteristic pearls-on-a-string appearance of mesh as described in the literature. Line profiles demonstrate periodicity associated with mesh and may provide insight for new in vivo detection methods.

Lewis, M, Furey, B, Madhuranthakam, A, Bailey, A, Khatri, G, Twickler, D, Investigations of the Radiographic Appearance of Surgical Mesh in Phantoms with Volumetric Ultrasound and Magnetic Resonance Imaging.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14014166.html