Arindam Rano Chatterjee MD, Presenter: Nothing to Disclose

Seth-Emil T. Bartel MD, Abstract Co-Author: Nothing to Disclose

Manu Shri Goyal MD, MSc, Abstract Co-Author: Nothing to Disclose

Matthew Shawn Parsons MD, Abstract Co-Author: Nothing to Disclose

Aseem Sharma MBBS, Abstract Co-Author: Stockholder, General Electric Company

Time taken to obtain MR scans has direct implications for imaging costs, but has not been studied methodically. We sought to quantify potential redundancy in our MR brain sequence protocol by identifying the location of core diagnostic information within the scan.

In this retrospective review, we aggregated consecutive 207 individual adult MRI brain studies. The senior author, a board certified neuroradiologist reviewed the medical and imaging records to document the clinical indication, core diagnostic information (CDI) provided by the MR imaging, and its clinical impact. Imaging findings useful for reaching the diagnosis constituted positive CDI while absence of imaging abnormalities constituted negative CDI. The senior author then selected the Core Sequences (CS) felt to allow for comfortable extraction of the CDI. The selection of CS was validated by presenting the CS to four readers who assessed the relative ease of identification of the CDI within the CS. Potential redundancy was calculated by comparing the number of CS to the number of total sequences obtained.

MR imaging had been performed utilizing an average of 9.4 ± 2.8 sequences obtained over an average of 37.3 ± 12.3 minutes. In comparison, CDI was thought to be easily extractable from an average of 2.1 ± 1.1 core sequences, with an assumed corresponding average scan time of 8.6 ± 4.8 minutes, reflecting a potential redundancy of 74.5 ± 19.1 %. Potential redundancy was least in scans obtained for treatment planning (14.9 ± 25.7%) and highest in scans obtained for follow-up of benign disease processes (81.4 ± 12.6). The selection of CS was considered to be valid with the ease of extracting CDI from CS considered by all four readers to be either easy or equivalent to that from entire scan in 97.4% cases. With a lack of clinical impact on only one case (0.48%), overutilization of imaging was not thought to contribute potential redundancy.

There is high potential redundancy in brain MR imaging protocols that can be targeted for a more efficient utilization of MR scanners.

More indication directed sequence protocoling can help improve efficiency of MR scanner utilization with associated potential for reducing imaging related healthcare costs.

Chatterjee, A, Bartel, S, Goyal, M, Parsons, M, Sharma, A, Potential Redundancy in MR Imaging of Brain: Quantification by Studying the Location of Core Diagnostic Information Across Various Sequences.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006938.html