We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Diffusion kurtosis imaging for differentiating borderline from malignant epithelial ovarian tumors: A correlation with Ki-67 expression.
- Authors
Li, Hai Ming; Zhao, Shu Hui; Qiang, Jin Wei; Zhang, Guo Fu; Feng, Feng; Ma, Feng Hua; Li, Yong Ai; Gu, Wei Yong
- Abstract
<bold>Purpose: </bold>To investigate the use of diffusion kurtosis imaging (DKI) in differentiating borderline from malignant epithelial ovarian tumors (MEOTs) and to correlate DKI parameters with Ki-67 expression.<bold>Materials and Methods: </bold>Fifty-two consecutive patients with epithelial ovarian tumors (17 borderline epithelial ovarian tumors, BEOTs; 35 MEOTs) were prospectively evaluated using DKI with b values of 0, 500, 1000, 1500, 2000, and 2500 s/mm2 and standard diffusion-weighted imaging (DWI) with b values of 0 and 1000 s/mm2 using a 1.5T magnetic resonance imaging (MRI) unit. The kurtosis (K) and diffusion coefficient (D) from DKI and apparent diffusion coefficient (ADC) from standard DWI were measured, compared, and correlated with Ki-67 expression between the two groups. Statistical analyses were performed using the Mann-Whitney U-test, receiver operating characteristic (ROC) curves, and Spearman's correlation.<bold>Results: </bold>The K value was significantly lower in BEOTs than in MEOTs (0.55 ± 0.09 vs. 0.9 ± 0.2), while the D and ADC values were significantly higher in BEOTs than in MEOTs (2.27 ± 0.35 vs. 1.39 ± 0.37 and 1.72 ± 0.36 vs. 1.1 ± 0.25, respectively) (P < 0.001). For differentiating between BEOTs and MEOTs, the sensitivity, specificity, and accuracy were 88.2%, 94.3%, and 92.3% for K value; 88.2%, 91.4%, and 90.4% for D value; and 88.2%, 88.6%, and 88.5% for ADC value, respectively. However, there were no differences in the diagnostic performances among the three parameters above (K vs. ADC, P = 0.203; D vs. ADC, P = 0.148; K vs. D, P = 0.904). The K value was positively correlated with Ki-67 expression (r = 0.699), while the D and ADC values were negatively correlated with Ki-67 expression (r = -0.680, -0.665, respectively).<bold>Conclusion: </bold>Preliminary findings demonstrate that DKI is an alternative tool for differentiating BEOTs from MEOTs, and is correlated with Ki-67 expression. However, no added value is found for DKI compared with standard DWI.<bold>Level Of Evidence: </bold>1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2017;46:1499-1506.
- Publication
Journal of Magnetic Resonance Imaging, 2017, Vol 46, Issue 5, p1499
- ISSN
1053-1807
- Publication type
journal article
- DOI
10.1002/jmri.25696