Gaëlle Diserens1, Waldo Valenzuela2, Maryam Seif1, Laila Mani3, Daniel Fuster3, Christoph Stettler4, Bruno Vogt3, Mauricio Reyes2, Chris Boesch1, and Peter Vermathen1
1Depts Clinical Research and Radiology, University of Bern, Bern, Switzerland, 2Institute for Surgical Technology and Biomechanics, University of Bern, Bern, Switzerland, 3Dept. of Nephrology, Hypertension and Clinical Pharmacology, University of Bern, Bern, Switzerland, 4Dept. of Endocrinology, Diabetes and Clinical Nutrition, University of Bern, Bern, Switzerland
Synopsis
Renal ectopic lipid accumulation may lead to kidney dysfunction. The study purpose was to determine (1) renal ectopic lipid content in overweight type-2
diabetic patients compared to (a) overweight non-diabetic patients and (b) lean
volunteers by 1H-MRS and (2) renal sinus fat content by DIXON-MRI in
the same three patient groups. This study demonstrates that renal ectopic lipids appear to be not
higher in overweight diabetic patients compared to overweight non-diabetic
subjects, while ectopic lipids are higher in both groups compared to healthy
subjects. Significantly higher renal sinus bulk lipids were
detected for overweight diabetic patients compared to BMI-matched
non-diabetics. Purpose
There is increasing evidence that renal ectopic lipid accumulation
leads to kidney dysfunction1,2,3. These ectopic lipids have been associated with type-2
diabetes, obesity-related glomerulopathy and with organ function2,4.
However, to our knowledge, no MRI and no MRS study has been performed in humans
to non-invasively assess renal ectopic lipids or bulk lipids in the renal sinus
in diabetic
patients.
The purpose of this study was to use MR techniques to determine (1) the renal ectopic lipid
content in overweight patients with type-2 diabetes compared to (a) overweight non-diabetic
patients and (b) lean healthy volunteers by optimized 1H-MRS and (2)
the fat content of the renal sinus by DIXON fat-water imaging in the same three
groups of patients. We hypothesized the renal ectopic lipid content and the fat
in renal sinus to be increased in obese diabetic patients.
Subjects and Methods
Study Population
Three groups of 8 patients were measured on a 3T-MR Scanner
(Verio,Siemens): (1) overweight (BMI>28) Type-II diabetic, (2) overweight (BMI>28)
non-diabetic, (3) lean healthy (BMI=18-25) (Table 1).
MR-Imaging and Spectroscopy
A single voxel PRESS sequence with PACE respiratory-triggering
(TR=2000ms, TE=35ms, 16 measurements with 4 acquisitions, 6 saturation bands, voxel size between 10x12x12mm3 and
15x15x15mm3) was placed in visually fat-free
renal parenchyma.
A standard two-point DIXON sequence was performed in breathold for
fat/water imaging of the right kidney in three orthogonal directions, with
TR=5.9ms, TE=2.5/3.7ms, FOV=388x400mm2.
Data Processing
MRS: Fat and water peaks from the sum of
the 16 SVS spectra of each patient were fitted using jMRUI AMARES5, the
lipid ratio was calculated as Fat/(Fat+Water).
MRI: Kidneys from coronal Dixon data
were segmented using FISICO6 (Fig.1) and the renal volume as well as
the bulk renal sinus fat content determined, after establishing a threshold for
supposedly pure lipid content.
Statistical analyses (Student's t-tests, correlations) were performed
using Excel and Matlab. Bonferroni corrections for
multiple comparisons were applied.
Results
Renal ectopic lipids by SVS
The renal ectopic lipid content in the lean healthy group was only 0.21%±0.08%.
One healthy subject had more than 7-times higher content than the average of the
other healthy subjects and was thus excluded from the statistical analysis,
also because retrospective analysis demonstrated partial voxel-placement
outside renal parenchyma.
The average renal ectopic lipid content was significantly higher for
the overweight group (0.75%±0.59%) and for the overweight+diabetic group
(0.67%±0.46%), compared to the lean healthy group (Table 2, Fig. 2a). No significant
difference was found between the overweight group compared to the overweight+diabetic
group.
Renal sinus fat by Dixon
imaging
The amounts of renal sinus fat were significantly different between
all three groups (p<0.02 for all), with highest content for the overweight+diabetic
group (39.2 cm3± 12.0cm3), intermediate for the
overweight group (24.9cm3±7.2cm3) and lowest for the
healthy group (12.3cm3±6.8cm3) (Table 2, Fig.2b).
Kidney Volume without sinus
lipid volume
The renal volume of the lean healthy group after subtraction of the
sinus lipid volume (107cm3±18cm3) was significantly lower
than the volume of the overweight group (155cm3±29cm3)
and also significantly below that of the overweight+diabetic group (181cm3±57cm3)
(Table 2, Fig.2c). Renal volumes of the overweight group and
overweight+diabetic group were not significantly different.
Correlations between the lipid
contents and the BMI
No significant correlation was found between the renal ectopic lipid
measured by SVS and the renal sinus fat (R=0.15, p=0.49) and between ectopic
lipid and BMI (R=0.38, p=0.07).
The renal sinus fat amount of all subjects correlated strongly with
BMI (R=0.70, p<0.001, Fig.3a). Kidney volumes after removal of the sinus
lipid volume were significantly correlated with BMI (R=0.69, p<0.001, Fig.3b).
Discussion and Conclusion
This study demonstrates that renal ectopic lipids appear to be not
higher in overweight diabetic patients compared to overweight non-diabetic
subjects, while the ectopic lipids are higher in both groups compared to
healthy subjects. This result suggests
no increase of renal ectopic lipid content in relation with diabetes.
In contrast, significantly higher bulk lipids in the renal sinus
were detected for overweight diabetic patients compared to BMI-matched
non-diabetics. In addition, our data suggests that fat infiltration into the
renal sinus occurs– though at lower extend – with increased BMI, as also shown
previously7. The kidney volume without the renal sinus fat content
shows a significant correlation with BMI, suggesting that a complex
reorganisation process seems to take place inside the kidney itself7.
While the renal sinus fat content seems to be strongly affected by the diabetes
status, the renal volume is mainly related to the BMI and not to diabetes, as the
kidney volume was not significantly different between the two overweight groups.
Acknowledgements
This work was supported by the Swiss National Science Foundation SNF
grant #320030-138150.References
1. Bobulescu IA. et al., Triglycerides in the Human Kidney Cortex:
Relationship with Body Size. Plos One 2014; 9(8):e101285.
2. de Vries AP. et al., Fatty kidney: emerging role of ectopic lipid in
obesity-related renal disease. Lancet Diabetes Endocrinol. 2014; 2:417.
3. Herman-Edelstein M. et al., Altered renal lipid metabolism and renal
lipid accumulation in human diabetic nephropathy. J Lipid Res. 2014; 55(3).
4. Sieber J. et al., Free fatty
acids and their metabolism affect function and survival of podocytes. Front Endocrinol. 2014; 5:186.
5. Vanhamme L. et al., Improved method for accurate and efficient
quantification of MRS data with use of prior knowledge. Journal of Magnetic Resonance 1997, 129: 35-43.
6. Valenzuela W. et al., Correction Tool for Active Shape Model Based Lumbar
Muscle Segmentation. In IEEE Engineering in Medicine and Biology Society
(pp. 1–4), 2015.
7. Foster MC. et al., Fatty Kidney, Hypertension, and Chronic Kidney
Disease, The Framingham Heart Study. Hypertension 2011; 58:784.