The role of kidney diffusion tensor magnetic resonance imaging in children

The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

The Egyptian Journal of Radiology and Nuclear Medicine journal homepage: www.sciencedirect.com/locate/ejrnm

Original Article

The role of kidney diffusion tensor magnetic resonance imaging in children Mehmet Burak Özkan a,⇑, Robert Marterer b, Sebastian Tscheuner b, Utku Mahir Yildirim c, Elif Ozkan d a

Department of Pediatric Radiology, Dr. Sami Ulus Research and Training Hospital, Altindag, Ankara, Turkey Department of Pediatric Radiology, Graz, Austria c _ Department of Radiology, Medikal Park Hospital, Izmir, Turkey d Department of Pediatrics, Sami Ulus Hospital, Ankara, Turkey b

a r t i c l e

i n f o

Article history: Received 4 April 2016 Accepted 19 August 2016 Available online xxxx Keywords: Renal DTI Pediatric eGFR Renal insufficiency Radiology

a b s t r a c t Background: Diffusion tensor imaging (DTI) in magnetic resonance imaging (MRI) provides information about the microstructure of renal tissue and is becoming increasingly useful in the evaluation of relationship between renal structure and function. Objectives: To investigate, whether DTI allows assessment of renal impairment and pathology in pediatric patients with decreased renal functions. Materials and methods: Thirty-two pediatric patients and seventeen healthy children were included in this prospective study. For DTI, a respiratory-triggered coronal echo planar imaging (EPI) sequence was performed. Cortical and medullary mean axial and radial diffusivity and fractional anisotropy (FA) were analyzed. Results: In healthy subjects, the cortical FA values were significantly lower than the medullary FA values (p < 0.001). Cortical and medullary ADC values showed positive correlations (r = 0.499, p = 0.041) and a negative correlation with cortical FA values(r = 0.533, p = 0.028). The eGFR values were negatively correlated with the medullary ADC values (r = 0.484, p = 0.049) in healthy subjects and positively correlated with the k medullary values (r = 0.385, p = 0.027). Additionally in the patient group, the age was positively correlated with FA cortex values (r = 0.411, p = 0.018) and with the medullary ADC values (r = 0.461, p = 0.007). However, the medullary FA values were negatively correlated with the medullary ADC values (r = 0.363, p = 0.038). Tractography of healthy volunteers showed a radial arrangement which converged into the pyramids, whereas renal insufficiency patients had irregular arrangement patterns and architectural distortions in the observed areas. Conclusions: Renal DTI is a promising diagnostic tool in the assessment of microstructural renal changes and correlates with the eGFR. Therefore, it is possible to estimate the arrangement of the tracks which emanate from the renal medulla. Furthermore, diffusion of the water molecules could be carried out. This study demonstrates the usage of DTI in renal pediatric kidneys. Validations in larger cohort groups with histopathological biopsies are needed. Ó 2016 The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer review under responsibility of The Egyptian Society of Radiology and Nuclear Medicine. ⇑ Corresponding author. E-mail addresses: [email protected] (M.B. Özkan), [email protected] (R. Marterer), [email protected] (S. Tscheuner), oebler@gmail. com (U.M. Yildirim), [email protected] (E. Ozkan). http://dx.doi.org/10.1016/j.ejrnm.2016.08.015 0378-603X/Ó 2016 The Egyptian Society of Radiology and Nuclear Medicine. Production and hosting by Elsevier. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

2

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

1. Introduction The major function of the kidneys is to balance the water and solute concentration. Kidneys do this physiologically by the help of glomeruli and tubules, which enable filtration and reabsorption of water and solutes [1–4]. Understanding the basic mechanisms of water diffusion in the kidneys could help to improve the understanding of kidney pathologies. Laboratory parameters, such as serum creatinine or estimated glomerular filtration rate (eGFR), are commonly used for the evaluation of renal tubular function [5–7]. Diffusion weighted imaging (DWI) visualizes water motion at the molecular level and provides useful information on parenchyma microstructure and function, basically depending on the Brownian motion [5–7]. The apparent diffusion coefficient (ADC) is a quantitative parameter calculated from DWI [1]. It was shown that ADC values tend to decrease in chronic kidney diseases, depending upon the degree of renal impairment [5,8]. Kidneys have a well-defined anatomical structure with tubules, collecting ducts and vessels radially oriented toward the pelvis and in which molecules move in a preferential direction [9,10]. To investigate molecular diffusion in kidneys, the direction should be evaluated. In healthy kidney vessels, tubules and collecting ducts are involved in water reabsorption and urine formation. These structures have a highly perfectly organized radial arrangement. This kind of arrangement enables motion of water molecules to move in different directions, such as right to left or superior to inferior. Therefore, this is the reason for the anisotropic manner of the molecular diffusion in renal parenchyma [9]. This highly perfectly designed structural organization gets damaged in pathological processes. This relationship was demonstrated by previous studies [5,6,9,11]. DTI enables to measure the diffusion of the water molecules in at least six vectoral ways. The diffusion anisotropy could be used to understand the microstructure of the renal parenchyma. The fractional anisotropy (FA) describes the degree of anisotropy of a diffusion process whereas ADC describes only the averaged diffusivity in the directional way. Therefore, DTI enables to estimate and to measure the vectoral way of least restricted diffusion [3,12,13]. In the literature, the correlation between agedependent changes and the ability of DTI to estimate renal tissue injuries due to tumors, pyelonephritis and renal artery stenosis was demonstrated. Additionally, it is known that the renal medulla has higher FA values than the cortex [5,9,14]. Studies in adults have demonstrated the ability to show microstructural changes correlating with functional parameters by utilizing DTI [3,4,9,15–16]. This prospective study’s purpose was to examine the value of renal DTI in children and adolescents.

2. Material and methods Divided into two cohorts (normal eGFR of P60 ml/ min/1.73 m2, and reduced eGFR between P25 ml/

min/1.73 m2 and <60 ml/min/1.73 m2), a total of 50 children were included in this prospective single center study. A healthy control group consisted of 17 volunteers, the study group of 33 children. The control group consisted of nine males and eight females with a mean age of 8 ± 0.9 years (3.09–14.7 years). The mean age in the study group was 8.5 years ± 0.7 years (3.02–15.3 years), consisting of 20 male and 12 female children. One patient had to be excluded from this study group, as corticomedullary differentiation (CMD) had not been reported. All of them underwent renal DTI, as further described in the renal MRI acquisition section. The duration of the study was one and half year. To be certain about the absence of renal parenchymal disease, volunteer patients additionally underwent a renal ultrasound examination. Exclusion criteria were MRI contraindications such as ferromagnetic implants or claustrophobia. Because of a possible risk of nephrogenic systemic fibrosis [17], patients with GFRs lower than 25 were not included in the study. The dimensions and corticomedullar differentiation were controlled before the renal MRI examination. All the values regarding these parameters were within normal values. Regarding the b:0 image values differentiation between the cortex and the medulla could be easily evaluated and able to be understood the object regions and edges. A study group sub-group, representing patients with renal impairment (eGFR P 25 and equal or lower 40), was defined. eGFRs were calculated from serum creatinine measurements for all subjects based on the revised Schwarz formula for children and adolescent [18]. The patients were sedated under the age of eight. They were sedated by local anesthesiological products. The study was approved by the institutional review board and written consent was obtained from the parents of all children. 3. Renal MRI acquisition All subjects were scanned in supine position in an Avanto 1.5 T (T) MRI scanner (Siemens Medical Systems, Erlangen, Germany). Spine array (posterior) and body array (anterior) receiver coils were used to maximize image uniformity. The imaging protocol consisted of a respiratorygated (axial 2D, 3D) echo-planar fast spin echo sequence (HASTE) and a coronal T2-weighted sequence. After the HASTE sequence, a respiratory-gated, single-shot diffusion tensor imaging – echo planar imaging (DTI-EPI) acquisition was performed on one kidney (b = 0 and 400 s/mm2, 6 directions + null, TR/TE = 2000 ms/75 ms, imaging slice thickness = 3 mm, 10 imaging slices/subject). In order to achieve a better resolution and higher SNR values, either the left or right kidney was evaluated. Only single renal evaluation was enough for the determination. In all children diffusion MR images were acquired prior to the administration of gadolinium. As a routine part of this renal MRI imaging, we had to use the gadolinium administration in order to get the sufficient renal diagnosis except those under eGFR under 25. Although we had to use gadolinium as a part of imaging technique, the renal DTI was evaluated separately in another work station and the

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

renal MRI determination did not affect the renal DTI results. 4. Imaging post-processing Image post-processing was performed with Diffusion Toolkit software package (www.trackvis.org, version 0.5).

3

Tractography was reconstructed using a deterministic streamline algorithm. This algorithm was based on a user-defined region of interest. To be continuous and homogenous within the previous studies, thresholds were set for minimum fractional anisotropy (FA). A region of interest (ROI) analysis of the ADC and FA maps was performed to obtain measures of mean cortical and medullary

Fig. 1a. Shows the descriptive characteristics of healthy volunteers.

Fig. 1b. Demonstrates the correlations of parameters that are statically significant.

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

4

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Fig. 2. (a) Shows the correlations of parameters in the study group patients.

Fig. 2 (continued)

ADC and FA for each subject. At least three coronal image slices in proximity to the renal hilum were selected for the ROI analysis. Medullary and cortical ROI selections were obtained in the HASTE images with high corticomedullary contrast and high anatomical detail. The ROIs selected in the HASTE images were directly applied to the cor-

responding (and co-registered) ADC and FA maps to obtain cortical and medullary ADC and FA values. Afterward, the mean of all resulting ADC and FA values was calculated. A Diffusion Ellipsoid is completely represented by the Diffusion Tensor, D. FA is calculated from the eigenvalues (k k1, k 2, k k3) of the diffusion tensors [1] The eigenvectors () give the

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

5

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Fig. 2 (continued)

Table 1a The DTI results of t-test for the healthy group. Paired samples test Paired differences Mean

Std. deviation

t Std. error mean

95% confidence interval of the difference Lower

Upper

Pair 1

medullarFA - adcmedulla

1.646212

0.045.584

0.007935

1.662375

1.630049

Pair 2

facortex - adccortex

1.993091

0.094447

0.016441

2.026580

1.959602

Pair 3

k1korteks - k1medulla

0.137862

0.023999

0.823939

0.775056

0.872823

207.460 121.226 34.333

Pair 4

k2korteks - k2medulla

0.197576

0.182962

0.031850

0.132700

0.262451

6.203

Pair 5

k3korteks - k3medulla

0.21393939

0.13665512

0.02378860

0.16548360

0.26239519

8.993

Pair 6

facortex - medullarFA

0.235364

0.023817

0.004146

0.243809

0.226919

Pair 1

medullarFA - adcmedulla

df 32

Sig. (2-tailed) 0.000

Pair 2

facortex - adccortex

32

0.000

Pair 3

k1korteks - k1medulla

32

0.000

Pair 4

k2korteks - k2medulla

32

0.000

Pair 5

k3korteks - k3medulla

32

0.000

Pair 6

facortex - medullarFA

32

0.000

directions in which the ellipsoid has major axes, and the corresponding eigenvalues give the magnitude of the peak in that direction. 5. Statistical analysis Among both cohorts, mean medullary and cortical diffusion parameters (ADC, FA) as well as eGFRs were compared using two-tailed Student’s t tests. Mean cortical

56.769

and medullary ADC and FA were also plotted as a function of eGFR for all subjects and Pearson’s correlation coefficients (R) were determined. Multivariable regression was used to analyze the dependence of whole-kidney regionof-interest tractography volume on FA, eigenvectors and renal function diagnosis. As statistical software (SPSS Statistics 21, IBM, USA) was used, P-values lower than 0.05 were assumed to be statistically significant.

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

6

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Table 1b The DTI results of t-test for the patient group. Paired samples test Paired differences Mean

t

Std. deviation

Std. error mean

df

Sig. (2-tailed)

95% confidence interval of the difference Lower

Upper

Pair 1

medullarFA - adcmedulla

1.646212

0.045584

0.007935

1.662375

1.630049

207.460

32

0.000

Pair 2

facortex - adccortex

1.993091

0.094447

0.016441

2.026580

1.959602

121.226

32

0.000

Pair 3

k1korteks - k1medulla

0.823939

0.137862

0.023999

0.775056

0.872823

34.333

32

0.000

Pair 4

k2korteks - k2medulla

0.197576

0.182962

0.031850

0.132700

0.262451

6.203

32

0.000

Pair 5

k3korteks - k3medulla

0.21393939

0.13665512

0.02378860

0.16548360

0.26239519

8.993

32

0.000

Pair 6

facortex - medullarFA

0.235364

0.023817

0.004146

0.243809

0.226919

32

0.000

56.769

Table 1c The DTI correlation values of t-test for the patient group. Correlations adccortex

medullatractlength

adcmedulla

facortex

k1korteks

k1medulla

aage

Pearson’s Correlation Sig. (2-tailed)

0.461 0.007

medullatractlength

Pearson’s Correlation Sig. (2-tailed)

0.527 0.002

adcmedulla

Pearson’s Correlation Sig. (2-tailed)

k2korteks

Pearson’s Correlation Sig. (2-tailed)

k2medulla

Pearson’s Correlation Sig. (2-tailed)

k3korteks

Pearson’s Correlation Sig. (2-tailed)

k3medulla

Pearson’s Correlation Sig. (2-tailed)

0.440 0.010

eGFR

Pearson’s Correlation Sig. (2-tailed)

0.385 0.027

0.527 0.002

1

k2korteks

0.495 0.003 0.345 0.049

0.461 0.007

0.461 0.007 0.392 0.024 0.476 0.005

6. Results Comparison between the two groups showed that the FA values in the medulla were significantly lower (p < 0.001) in the study group than in the control group. Additionally, the ADC cortex, ADC medulla, and FA cortex were statistically significant lower values (p = 0.049, 0.001, and 0.001 respectively) in the study group compared to the control group. However the k values did not show statistically significant differences between the cortical values in both groups (p = 0.121). (Compare Table 2a.) Within the stepwise regression analysis result medullar tract length showed linear regression with ADC of the medulla (p = 0.015) in study group population. Additionally medullar FA results showed partial regression with ADC medulla when eGFR was less than 60 in the study group (p = 0.029). 7. Control group The mean value of the eGFR in the control group was measured to be 84 mL/min (range 72–119 mL/min) (Fig. 1a).

0.346 0.048

0.644 0.000

The ADC value was significantly (p < 0.001) higher in the cortex than in the medulla; the FA value was significantly (p < 0.001) higher in the medulla than in the cortex. Additionally, the k1, 2, 3 medulla values were higher than in the cortex (p < 0.001, 0.0013, and 0.0002 respectively). In healthy subjects there was negative correlation of eGFR with ADC medulla values (r = 0.484, p = 0.049). For the cortex evaluation, there was negative correlation between ADC cortex values and FA values (r = 0.533, p = 0.028) (Fig. 1b).

8. Study group In the cortical regions, ADC values were significantly (p < 0.001) higher than in the medulla. In opposite, the FA values were significantly (p < 0.001) higher in the medulla than in the cortex. (Fig. 2a and b) (Tables 1a–1c). Multivariate stepwise regressions in the impaired renal function subgroup (eGFR 6 40) showed statistically differing FA values (p = 0.04), when compared to the control group. (The graphics of the detailed analysis are shown in Fig. 3.)

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

0.101544 0.091274

0.114768 0.117331

0.028243 0.029593

0.00698 0.00466

0.000123 0.010393

0.057232 0.054669

0.016303 0.014952

0.05777 0.05545

0.16878 0.17020

0.073572 0.076752 0.040424 0.037244

0.02279 0.02136

Upper Lower

0.03624 0.03629 0.09578 0.09578 0.011 0.014 45 27.378 2.643 2.639 0.600 Equal variances assumed Equal variances not assumed k1medulla

0.279

0.01607 0.01479 0.02540 0.02540 0.121 0.095 45 34.013 1.580 1.717 0.155 Equal variances assumed Equal variances not assumed k1korteks

2.089

0.002964 0.003500 0.022273 0.022273 0.000 0.000 45 19.225 7.515 6.363 0.015 Equal variances assumed Equal variances not assumed facortex

6.365

0.014283 0.015174 0.086000 0.086000 0.000 0.000 45 23.811 6.021 5.668 0.571 Equal variances assumed Equal variances not assumed adcmedulla

0.325

0.025178 0.020076 0.050833 0.050833 0.049 0.015 45 44.771 2.019 2.532 0.043 Equal variances assumed Equal variances not assumed adccortex

4.324

0.008229 0.009474 0.056998 0.056998 0.000 0.000 45 20.150 6.927 6.016 0.092 2.967 Equal variances assumed Equal variances not assumed

Sig.

7

In contrast to the control group, tractography in the study group did not demonstrate astrict radial arrangement of the fibers extending from the proximal portion to the Henle loop. The tracks were loosely arranged along the displaced parenchyma in the study group. This resulted in hollow spaces without tracts in the location of the dilated renal calyces or parts of the renal pelvices, especially in the patients with diagnosed ureteropelvic junction (UPJ) obstruction (Figs. 4a and 4b).

9. Discussion

medullarFA

t F

df

t-test for equality of means Levene’s test for equality of variances

Independent samples test

Table 2a The DTI results of t-test between the groups including cortex and medulla.

Sig. (2-tailed)

Mean difference

Std. error difference

95% confidence interval of the difference

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

The main objective of the study was to assess the feasibility and potential benefits of diffusion tensor imaging in pediatric kidney diseases. We acquired diffusion MR images of sufficient quality to perform DTI tractography in a cohort of children with normal and abnormal renal functional parameters based on the eGFR values. The renal parenchyma is divided into cortex and medulla which is totally different in a histological and morphological point of view. Mainly the cortex part is responsible for the ultrafiltration of plasma. This ultrafiltrate moves through the medullary pyramids. Between the medullary pyramids there are also vessels which pass transverse and parallel to the tubules, enabling active secretion and reabsorption of solutes. In the renal pelvic region, the ultrafiltration moves to the renal papillas ending into minor and major calyx forming the urine [1,2,9]. The formation of urine and ultrafiltration of the kidney needs a nearly perfectly designed microstructural arrangement, in both the cortex and the medulla. In the cortex, there are randomly arranged fibers extending from proximal portions of the glomerule to the distal parts of the convoluted tubules. The fiber network system in the medulla consisted of a parallel vascular network design. The tubulous structures, including the ascending and descending portions of the Henle loop are also found in this parallel designed portion system. In this volumetric design within the comparison of medulla with the cortical region, the fibers found in the medulla were thinner than the cortical tubules [1,9,13,16]. In a previous study with a small population, fibers extending throughout the whole kidney did not show significant differences between kidneys with normal or abnormal function [9]. Our study’s results support this finding, although whole renal fiber tracts were significantly higher in the study group than in the control group. This is most likely due to increasing number of fibers with patient age and was not statistically significant in the multivariate regression analysis. Similar findings have been reported in other previous studies [5,6,8,9]. On the other hand, based on the regression analysis, the medullar FA could have a critical role in the determination and estimation of the renal parenchymal injury. In the evaluation of the fractional anisotropy maps, corticomedullary differentiation detected in the study group was fairly poor when compared with the healthy volunteers. The insufficient resolution of the FA maps between the regions could be as a result of the injury to the loss of fiber tract organization. This could also be due to the

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

8

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Fig. 3. (a–f) Figures are enabled from the regression plot of the test of multiwise step logistic regression test. In this graphics the selected cases represent the study group who is less than or equal to eGFR of 40. The number of selected cases is seventeen kidneys.The unselected cases are the children who have eGFR greater than 40. (a) Shows the relationship of medullar trace length with age. (b) Shows the relationship of ADC medulla with medullar trace length. (c) Shows thepartialregressionplot of medullar FA and age. (d) Shows the scatter plot of medullar FA. (e) Shows the partial regression plot of medullar FA and ADC medulla values. (f) Shows the histogram of selected histogram in the relationship of frequency.

capacity of detecting DTI in the thinner fibers, which could be related to the Tesla magnitude. On the other hand, Jaimes et al. did a study with a 3T machine reporting similar results [9]. Regarding the patient group, there is a decrease in diffusion anisotropy obtained. This could be caused within the structural parenchymal alterations, such as interstitial fibrosis, tubular atrophy and cellular infiltration, basically involving pathologies the medullar part of the renal system. Previously particularly including the small group studies presenting the reduction in the medullary fractional anisotropy values was reported [4–6,8,9]. In our study group, there were seventeen patients with eGFR less than 40 ml/min/1.73 m2. In this group we identified that only medullary FA could be a promising diagnostic tool in the assessment of radial fiber arrangement and its correlation with eGFR. Based on our study report, we could use the medullary FA as a marker of renal structural integrity alteration, potentially useful index for the diagnosis and outcome of kidney disease. Hueper et al. [11], hypothesize that the majority of chronic diseases affecting the renal parenchyma probably alter directed diffusion (FA) before free diffusion (ADC). Previous studies mention that ADC and FA are also influenced by different several factors as in the renal medulla. In think age of the water molecules and physiology there

could be high alterations during the movement of the transportation [6,9]. ADC is mainly influenced by perfusion, whereas the FA is related to the water molecule transport in the collecting tubules. These comments also suggested that DTI is the most appropriate technique for functionally evaluating the kidney as compared to DWI, and the medullary FA represents the most sensitive parameter in detecting renal damage in different diseases. The limitations of the study include a small study population of 49 children. There were no patients with an eGFR lower than 25 ml/min/1.73 m2 and only seventeen with an eGFR between 25 ml/min/1.73 m2 and 40 ml/ min/1.73 m2. We avoided inclusion of patients with a eGFR lower than 25 ml/min/1.73 m2 because of potentially harmful MRI contrast agent effects [17]. Due to the restricted study population, age-dependencies in renal diffusion parameters were no assessed explicitly. Regarding this, we noticed similar results as in the literature [9]. Additionally, we did not consider or compare different b values in the evaluation of DTI imaging, which possibly could help to optimize acquisition technique. Previous studies demonstrated the effects of different b values [19–21], that influence the FA values regarding the limitations for the specific chronic kidney diseases should be evaluated.

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

9

Fig. 3b (continued)

Fig. 3c (continued)

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

10

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Fig. 3d (continued)

Fig. 3e (continued)

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

11

Fig. 3f (continued)

Fig. 4a. Shows the DTI evaluation of a seven years old boy with eGFR diagnosed with 30 after TrackVis program reformat technique: The long green renal fibers show the extension of from proximal tubules to the Henle Loop. The proximal portion of the fibers getting thinner represents the tubule pathology in medulla region.This image was obtained with a view of 55°.

10. Conclusions Our prospective study showed the capability of renal DTI in the usage of alterations induced by various chronic kidney diseases. DTI enables to understand the

renal anisotropy based on physiology without contrast media administration. Medullary FA values seem to be the best promising tool in the association within the complementary of laboratory and morphological parameters.

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

12

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx

Fig. 4b. Shows the anterior-posterior view of the 3-D tractography. The upperportion of the green fibers get thinner and shows loss of connection of network in the radial arrangement at the medullar region. The red fibers shows the crossing of left to right direction. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Conflict of interest The authors declare that there are no conflict of interests.

References [1] Zheng Z, Shi H, Zhang J, Zhang Y. Renal water molecular diffusion characteristics in healthy native kidneys: assessment with diffusion tensor MR imaging. PLoS One 2014;9:e113469. [2] Wang W, Pui MH, Guo Y, et al. MR diffusion tensor imaging of normal kidneys. J Magn Reson Imaging 2014;40:1099–102. [3] Heusch P, Wittsack HJ, Kröpil P, et al. Impact of blood flow on diffusion coefficients of the human kidney: a time-resolved ECGtriggered diffusion-tensor imaging (DTI) study at 3T. J Magn Reson Imaging 2013;37:233–6. [4] Lu L, Sedor JR, Gulani V, et al. Use of diffusion tensor MRI to identify early changes in diabetic nephropathy. Am J Nephrol 2011;34:476–82. [5] Feng Q, Ma Z, Wu J, Fang W. DTI for the assessment of disease stage in patients with glomerulonephritis–correlation with renal histology. Eur Radiol 2015;25:92–8. [6] Liu Z, Xu Y, Zhang J, et al. Chronic kidney disease: pathological and functional assessment with diffusion tensor imaging at 3 T MR. Eur Radiol 2014;25:652–60. [7] Gaudiano C, Clementi V, Busato F, et al. Renal diffusion tensor imaging: is it possible to define the tubular pathway? A case report. Magn Reson Imaging 2011;29:1030–3.

[8] Gaudiano C, Clementi V, Busato F, et al. Diffusion tensor imaging and tractography of the kidneys: assessment of chronic parenchymal diseases. Eur Radiol 2013;23:1678–85. [9] Jaimes C, Darge K, Khrichenko D, et al. Diffusion tensor imaging and tractography of the kidney in children: feasibility and preliminary experience. Pediatr Radiol 2014;44:30–41. [10] Notohamiprodjo M, Dietrich O, Horger W, et al. Diffusion tensor imaging (DTI) of the kidney at 3 T–feasibility, protocol evaluation and comparison to 1.5 T. Invest Radiol 2010;45:245–54. [11] Hueper K, Gutberlet M, Rodt T, et al. Diffusion tensor imaging and tractography for assessment of renal allograft dysfunction-initial results. Eur Radiol 2011;21:2427–33. [12] Fan W, Ren T, Li Q, et al. Assessment of renal allograft function early after transplantation with isotropic resolution diffusion tensor imaging. Eur Radiol 2015;80(3). [13] Cutajar M, Clayden JD, Clark CA, Gordon I. Test-retest reliability and repeatability of renal diffusion tensor MRI in healthy subjects. Eur J Radiol 2011;80:e263–8. [14] Seif M, Lu H, Boesch C, et al. Image registration for triggered and non-triggered DTI of the human kidney: reduced variability of diffusion parameter estimation. J Magn Reson Imaging 2014;00:1–8. [15] Feng Q, Ma Z, Wu J, Fang W. DTI for the assessment of disease stage in patients with glomerulonephritis - correlation with renal histology. Eur Radiol 2014;25:92–8. [16] Notohamiprodjo M, Reiser MF, Sourbron SP. Diffusion and perfusion of the kidney. Eur J Radiol 2010;76:337–47. [17] Idée JM, Fretellier N, Robic C, Corot C. The role of gadolinium chelates in the mechanism of nephrogenic systemic fibrosis: a critical update. Crit Rev Toxicol 2014;44(10):895–913. Nov. [18] Schwartz GJ, Muñoz A, Schneider MF, Mak RH, Kaskel F, Warady BA, Furth SL. J Am Soc Nephrol 2009;20(3):629–37. Mar.

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015

M.B. Özkan et al. / The Egyptian Journal of Radiology and Nuclear Medicine xxx (2016) xxx–xxx [19] Hui ES, Cheung MM, Chan KC, Wu EX. B-value dependence of DTI quantitation and sensitivity in detecting neural tissue changes. Neuroimage 2010;49(3):2366–74. [20] Chuck NC, Steidle G, Blume I, et al. Diffusion tensor imaging of the kidneys: influence of b-value and number of encoding directions on image quality and diffusion tensor parameters. J Clin Imaging Sci 2013;3:53.

13

[21] Lee, Vivian S, Manmeen Kaur, Louisa Bokacheva, Qun Chen, Henry Rusinek, Ravi Thakur, Daniel Moses, Carol Nazzaro, Kramer Elissa L. What causes diminished corticomedullary differentiation in renal insufficiency? J Magn Reson Imaging: JMRI 2007;25(4). http://dx. doi.org/10.1002/jmri.2087.

Please cite this article in press as: Özkan MB et al. The role of kidney diffusion tensor magnetic resonance imaging in children. Egypt J Radiol Nucl Med (2016), http://dx.doi.org/10.1016/j.ejrnm.2016.08.015