Can the Signal-to-Noise Ratio of Choline in Magnetic Resonance Spectroscopy Reflect the Aggressiveness of Endometrial Cancer?

2015 
Rationale and Objectives To differentiate endometrial cancer (ECa) from benign lesions in endometrial or in submucosa (BLs-ESm), and investigate whether the signal-to-noise ratio (SNR) of choline-containing compounds (Cho) obtained from three-dimensional 1 H magnetic resonance spectroscopy (MRS) is associated with the aggressiveness of ECa. Materials and Methods Thirty-three patients with ECa and 15 patients with BLs-ESm underwent preoperative multivoxel 1 H MRS at 3 T MR. The amplitude of Cho peak of each voxel was recorded, and the corresponding SNR of Cho peak (Cho SNR ) was calculated. The maximum Cho SNR (max Cho SNR ) for each lesion was identified. The max Cho SNR of ECa and BLs-ESm, as well as type I ECa and type II ECa, was compared. The relationship between max Cho SNR and pathologic characteristics of tumors, including tumor grade, stage, type, and tumor size, was analyzed. Results The mean max Cho SNR (±standard deviation [SD]) was 30.93 ± 16.89 for ECa and 10.40 ± 3.07 for BLs-ESm ( P SNR for type II ECa (48.54 ± 21.46) was higher than that for type I ECa (26.19 ± 12.02, P  = .001). There were no significant differences among different grades ( P  = .449). The Spearman coefficient between max Cho SNR and stage was 0.423 ( P  = .014); the difference existed only between Ia and III ECa ( P  = .048). The Pearson coefficient between Cho SNR and tumor size was 0.515 ( P  = .002). Conclusions The max Cho SNR obtained from MRS can differentiate ECa from BLs and type I ECa and type II ECa, but cannot differentiate between each grade ECa and each International Federation of Gynecology and Obstetrics stage ECa. However, max Cho SNR increased with the increase in International Federation of Gynecology and Obstetrics stage and size of ECa.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    27
    References
    8
    Citations
    NaN
    KQI
    []