RSNA 2004 

Abstract Archives of the RSNA, 2004


SSG24-03

Cartilage Imaging at 3-Tesla versus 1.5 Tesla: Preliminary Results in an Animal Model

Scientific Papers

Presented on November 30, 2004
Presented as part of SSG24: Musculoskeletal (Cartilage Imaging)

Participants

Jeffrey Neil Masi, Presenter: Nothing to Disclose
Christian Allen Sell BS, Abstract Co-Author: Nothing to Disclose
Catherine Phan MD, Abstract Co-Author: Nothing to Disclose
Lynne S. Steinbach MD, Abstract Co-Author: Nothing to Disclose
Sharmila Majumdar PhD, Abstract Co-Author: Nothing to Disclose
Thomas Marc Link MD, Abstract Co-Author: Nothing to Disclose

PURPOSE

To compare the diagnostic performance of MR imaging at 3 Tesla (T) and 1.5T in detecting cartilage lesions created in an animal model using sequences dedicated for cartilage imaging.

METHOD AND MATERIALS

Initially, cartilage sequences (fat saturated proton density weighted fast spin echo [FS PD-FSE] and fat saturated spoiled gradient echo [FS SPGR]) were optimized for imaging at 3T in two human volunteers. The FS PD-FSE and FS SPGR sequences with the best image quality were used to image ten porcine knees in which 29 artificial cartilage lesions had been created with a ceramic scalpel. Corresponding sequences were used at 1.5T for all specimens. The images were assessed by two radiologists in consensus and the diagnostic performance in detecting the lesions was determined. Lesion grade and diameter were obtained in all sequences at 1.5T and3T and compared with the macroscopic findings in specimen sections. Additionally, signal-to-noise ratios (SNR) were calculated.

RESULTS

The highest diagnostic performance was obtained at 3T for both FS PD-FSE and FS SPGR sequences. Using FS PD-FSE sequences with identical acquisition times, 90% of the lesions were detected at 3T while 62% were detected at 1.5T. With the FS SPGR sequences, 83% of the lesions were detected at 3T and 79% at 1.5T, however, the 1.5T sequence had a longer acquisition time. Correct lesion grade assessment was 65% at 3T and 61% at 1.5T with the FS PD-FSE sequences and 75% and 70% with the fs SPGR sequences. Subjective evaluation of image quality was significantly higher at 3T for both FS PD-FSE and FS SPGR sequences (P<0.05). SNR values were significantly higher at 3T (P<0.05).

CONCLUSIONS

Imaging at 3T increased the sensitivity of the detection of artificial cartilage lesions and the accuracy of lesion assessment when compared to imaging at 1.5T in this animal model. In addition image quality and SNR were significantly higher at3T.

Cite This Abstract

Masi, J, Sell, C, Phan, C, Steinbach, L, Majumdar, S, Link, T, Cartilage Imaging at 3-Tesla versus 1.5 Tesla: Preliminary Results in an Animal Model.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4415845.html