Differentiating Minimal Fat Angiomyolipoma from clear cell Renal Cell Carcinoma: comparison of monoexponential, biexponential, and stretched exponential Diffusion-weighted imaging
Haojie Li1, Lili Liang1, Anqin Li1, Qiong Li1, Yao Hu1, Hui Lin2, Daoyu Hu1, and Zhen Li1

1Departments of Radiology, Tongji Hospital, Tongji Medical College, WU HAN, China, People's Republic of, 2GE Healthcare,MR Research China,WU HAN, WU HAN, China, People's Republic of

Synopsis

Since DWI with different models may demonstrate different aspects of tissue properties, it should be valuable to compare and explore their roles in renal tumors. To our knowledge, however, no comparison of these different diffusion imaging approaches for the differential diagnosis in renal tumors has been investigated so far. The purpose of this study was to quantitatively compare and evaluate the potential clinical value of various diffusion parameters obtained from monoexponential, biexponential, and stretched exponential DWI models for differentiation of MFAML from ccRCC.

Purpose

To qualitatively evaluate and compare the potential of various diffusion parameters obtained from monoexponential, biexponential, and stretched exponential Diffusion-weighted imaging models in differentiating between Minimal Fat Angiomyolipoma (MFAML) and clear cell Renal Cell Carcinoma (ccRCC).

Materials and Methods

The Institutional Review Board (IRB) approved this retrospective study and waived the informed consent requirement. Multi-b values DWI(0-1700 sec/mm2)were performed in 126 patients with either pathologic analysis-confirmed MFAML (n=24) or ccRCC (n=102) on a 3-T MRI scanner. For the obtained diffusion weighted images, biexponential model was employed to calculate parameters including isotropic apparent diffusion coefficient (ADC), true ADC (ADCslow) and perfusion fraction (f) while water molecular diffusion heterogeneity index (a) and distributed diffusion coefficient (DDC) was calculated using stretched exponential model. Parameters from all models were compared between MFAML and ccRCC through two-sample t-test. Receivers operating characteristic (ROC) and intra-class correlation coefficient analysis were applied for statistical evaluations

Results

ADC, ADCslow and a values were significantly lower in the MFAML group than that in the ccRCC group (P < 0.01) while f and DDC values were significantly higher in the MFAML group than that in the ccRCC group (P < 0.01). The area under the ROC curve (AUC) of both a (0.991) and ADCslow (0.920) were significantly greater than those of ADC (0.742), f (0.787), and DDC (0.762) in differentiating MFAML from ccRCC.

Conclusion

Multi-b values DWI are feasible and useful in the noninvasive tissue characterization of renal tumors. Water molecular diffusion heterogeneity index (a) and ADCslow may provide additional information and improve the differential diagnosis between MFAML and ccRCC compared with conventional diffusion parameters.

Acknowledgements

No acknowledgement found.

References

[1]Cohen HT, et al. New England Journal of Medicine 2015;353:2477-2490. [2]Rheinheimer S, et al.Eur J Radiol 2012;81:e310-6. [3]Gaing B, et al. Invest Radiol 2015;50:144-52.

Figures

Figure 1: Flowchart of the study population

Note.—Data in parentheses are percentages.

* Date are mean±standard deviation.


Figure.4 Bar graphs of ADC, ADCslow, f, DDC, a values averaged between ccRCC(n=102) and MFAML(n=32). Error bars= stand deviations across subjects. ADC, ADCslow, DDC are in units of Í10-3mm2/s.Parameters marked with asterisk are significant.*=p<0.001.

Note.—AUC = area under the receiver operating characteristic curve, ADCslow=slow ADC, * Values are in units of Í10-3 mm2/s.

Figure.2 Images obtained in a 49-year-old man with clear cell Renal Cell Carcinoma (Furhman II). A. T2-weighted MR image shows that the tumor is hypointense(arrow). B. The ADC map and C, the ADCslow map show decreased ADC values. D. The f map show increased values. E. The DDC map show no increased values. F. The a map show decreased values.

Figure.3 Images obtained in a 42-year-old man with Minimal Fat Angiomyolipoma. A. T2-weighted MR image shows that the tumor is hypointense(arrow). B. The ADC map and C, the ADCslow map show decreased ADC values. D. The f map and E, the DDC map show increased values. F. The a map show decreased values.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
3499