Abstracts

RETURN TO ABSTRACT LISTING


2392. Evaluation of Programmable Shunt Setting on Computed Tomography
Authors * Denotes Presenting Author
  1. Einat Slonimsky *; Milton S. Hershey Medical Center
  2. Alex Mamourian ; Milton S. Hershey Medical Center
Objective:
Programmable ventriculo-peritoneal shunts are routinely imaged with CT and additional X-ray imaging for optimal visualization of the valve settings. Plain films of programmable shunts however, require specific positioning of the head and additional imaging. Historically, metal-related artifacts have limited the diagnostic value of CT when imaging metallic devices. The purpose of our study was to evaluate whether the shunt setting could be determined ex vivo using only a diagnostic non-contrast head CT .

Materials and Methods:
A dried human skull was used as the phantom for the CT testing of these programable shunts. Various CT techniques were used while scanning the phantom with four programmable shunts (Certas TM Plus (Codman, Raynham, Massachusetts), Polaris® (Sophysa, Orsay, France), proGAV 2.0 ® (Braun, Bethlehem, Pennsylvania), and Hakim® (Codman, Raynham, Massachusetts)) attached to its outer table in two different positions with the standard and dual energy non-contrast head CT and iMAR iterative reconstruction. Post processing was used to create 3D volume rendered (VR) reformations of the valve, and shunt setting were evaluated by two readers. For validation purposes, ten observers reviewed the same VR images generated from the CT scan data of each shunt at two settings and were asked to predict the shunt setting using only the 3D volume rendered images and rate the best reconstructions.

Results:
Evaluation of shunt setting was feasible using CT scan with 3D reformats for shunt with favorable design, even using reconstructions created from standard non contrast CT without the application of iMAR. Volume rendered images, created on a 3D workstation in less than 3 minutes, allowed correct prediction of the Polaris® shunt setting by all observers(100%) and 90% of the time with the Certas TM Plus, but no of the Hakim ® and ProGAV 2.0® shunts. iMAR reconstruction of CT data was preferred based on perceived quality of the volume rendered images for both CertasTM Plus and the Polaris® shunt.

Conclusion:
The valve setting of some programmable shunts can be determined using only CT with 3D reconstruction at standard CT doses. Our testing also indicates that the one metal suppression technique available to us, iMAR, improves visualization of the two programmable valves that have a favorable design for CT visualization.