| Part one Chronic kidney disease:functional assessment with diffusion tensor imaging at 3.0 T MRObjective Our objective was to evaluate the renal functional changes in chronic kidney disease (CKD) using diffusion tensor imaging (DTI) and diffusion tensor tractography (DTT) at 3.0 T MRI.Methods There were fifty-one patients with CKD and 19 healthy volunteers who were examined using DTI at 3.0 T MR(Magnetom Vero; Siemens, Erlangen, Germany). DTI was performed using a respiratory triggering, coronal echo-planar imaging sequence with the following parameters:repetition time/echo time,2000 ms/90 ms; section thickness,4 mm; 11 sections; field of view,400 mm×400 mm; b=0,400, and 600 s/mm2; 20 diffusion directions; two signals acquired; partial Fourier acquisitions,6/8; matrix,192×192; echo spacing,0.77 msec; parallel imaging factor,2; and bandwidth,2170 Hz/pixel.Morphological and DTI images were reviewed by one radiologist (more than five years of experience with abdominal MRI) blinded to all clinical data. In all cases, cortico-medullary differentiation (CMD) was evaluated on fat-saturated T1-weighted images arid graded as follows:1, absent (not visible); 2, reduced (CMD was visible but not clear); and 3, normal (clear definition of CMD). The mean values of fractional anisotropy (FA) and the apparent diffusion coefficient (ADC) were obtained from the renal parenchyma (cortex and medulla). The mean ADC and FA values in the renal parenchyma of each kidney were calculated from the separate averages of the cortical and medullary regions of interest (ROIs). Because the eGFR represented the function of two kidneys, the mean values of the diffusion parameters in both kidneys were calculated for comparison.(1) The differences in the FA and ADC between the two groups were assessed; the mean FA and ADC values were compared between two kidneys. (2) Correlations between imaging results and the estimated glomerular filtration rate (eGFR) were evaluated. (3) FA values in patients with various stages of CKD were compared. Using the K/DOQI CKD classification, patients were divided into subgroups (CKD stages) on the basis of eGFR. Because FA showed the closest correlation to eGFR, FA values in patients with various stages of CKD were compared. (4) Evalute the differeces of the tractography representation patterns showed in different CKD stages.Results1. In all controls, the CMD was confirmed to be normal (grade 3), and no other abnormalities were found except for two small cysts. Further, in all patients, the CMD was normal or reduced (grade 2 or 3) on DTI FA maps or b0 images except one(CMD, grade 1); small renal cysts were found in ten cases and ascites in five cases.2. There were no significant differences in the parenchymal FA and ADC between the left and right kidneys (p>0.05). The renal cortical FA was significantly lower than the medullary in both normal and affected kidneys (p<0.001). However, there was no significant difference between ADC values in the cortex and in the medulla(p>0.05).3. There were positive correlations between eGFR and FA (cortex, r=0.689, p=0.000; and medulla, r=0.696, p=0.000), and between eGFR and ADC (cortex, r=0.310, p=0.017; and medulla, r=0.356, p=0.010).4. The FA values differed significantly among CKD stages and controls (cortex, F=21.007, p= 0.000, and medulla, F=25.388, p=0.000). Moreover, the FA values of kidneys at all stages of CKD, including stage I (patients with normal or high eGFR), were significantly lower than normal.The parenchymal FA was significantly lower in patients than healthy controls, regardless of whether eGFR was reduced. A negative correlation was found between the renal FA and the stage of CKD (cortex, r=-0.572, p=0.000; and medulla, r=-0.589, p=0.000)5. The tractography calculated starting from the FA map showed a regular arrangement of the lines in the medulla, which had a radial orientation. Tractography representation showed variable patterns; in general, patients with stage I and II presented a similar orientation of diffusion when compared to the control subjects, with a regular arrangement of the tracts. In patients with stages Ⅲ,Ⅳor Ⅴ, tractography showed various patterns of disruption of this organisation, supposedly as a result of reduced directed diffusion:an apparently reduced number, premature interruption and disorientation of the tracts.Conclusion DTI is valuable for noninvasive assessment of renal function, the renal parenchymal FA values may be a sensitive diffusion parameter correlated with renal function. Even in the case of mild renal damage (CKD stage I, eGFR>90 ml/min), the decrease in FA could be measured. This imaging technology could therefore be valuable for diagnosis, plan therapy and follow-up in patients with CKD.Part two Diffusion tensor imaging and tractography in assessing renal pathology of chronic kidney diseaseObjective To evaluate the potential of renal diffusion tensor imaging (DTI) and tractography for assessment of renal pathologies in chronic kidney disease (CKD).Methods 69 CKD patients and 19 healthy volunteers were examed with a respiratory-triggered coronal echo-planar DTI-sequence at 3T MR (20 diffusion directions, b=0,400,600 s/mm2). The mean values of fractional anisotropy (FA) and the apparent diffusion coefficient (ADC) in the right kidney were obtained from all the renal parenchyma (cortex and medulla). Renal biopsy was performed in all patients followed in our hospital for accurate diagnosis, and the specimens were scored according to the severity of renal pathology. Based on pathology scores, patients were stratified into four subgroups(group 1,<5 scores, group 2,5-9 scores, group 3,10～14 scores, group 4, no less than 15 scores).(1) Correlations between renal diffusion parameters(FA and ADC values) and pathology scores (including glomerular lesion and tubulointerstitial injurie scores), pathology types were evaluated respectively. (2) Moreover, compare FA values and ages among five subgroups with different pathology scores or with healthy kidneys. (3) Then, associations between fractional anisotropy (FA) and pathology scores, serum creatinine, estimated glomerular filtration rate (eGFR), and age were explored.Results1. The renal CMD was judged to be normal (grade 3) in all controls and to be normal or reduced (grade 2 or 3) in all patients on FA or b0 maps. Two small cysts was found (subtype 1 according to the Bosniak classification) in controls and 9 cysts (subtypes 1-2) in patients. And ascites in five patients.2. Negative correlations were found between FA and the glomerular lesion (cortex, r=-0.499, p=0.000; and medulla, r=-0.530, p=0.000), an between FA and tubulointerstitial injury (cortex, r=-0.631, p=0.000; and medulla, r=-0.724, p=0.000). A significant negative correlation was also observed between renal FA values and pathology scores (cortex, r=-0.717 and medulla, r=-0.792, respectively).3. Significant differences were found among pathological severity subgroups and controls in FA values (cortex, F=43.875, p=0.000, and medulla, F=60.320, p=0.000). FA values differed significantly among the subgroups regard to severity of renal pathology (cortex, F=43.875, p= 0.000, and medulla, F=60.320, p= 0.000, respectively). Moreover, the FA values decreased gradually with an increased severity of renal pathology, but without statistical difference between group II and III in the cortical FA (p=0.163).4. Differences in directional diffusion between controls and patients could be visualized by the renal tractography. Using the same parameters (minimum FA=0.200, maximum angle=30°) in patients of group 3 or 4 patients visible:an apparently reduced number, premature interruption and disorientation of the tracts, representing the destruction of renal microstructure.5. The renal medullary FA was not significantly different among subgroups based on pathology types (F=2.252, p=0.091). Although the cortical FA has a little different among subgroups (F=4.852, p=0.004), which was similar to the changing trend of pathology score (F=7.817, p=0.000).6. The FA values were as the dependent variables, and with the pathology score, serum creatinine, eGFR and age as independent variables. The established regression model got statistical significances (cortex, F= 22.224, p=0.000 and medulla, F=30.549, p=0.000, respectively). There was a significant linear regression association between FA values and pathology score, but none between FA values and serum creatinine, or age. Moreover, a positive correlation between cortical FA and eGFR was found.Conclusion DTI is valuable for noninvasive assessment of pathology by reduced FA and ADC values, as well as the visualized tractography. It could be used for detecting the severity of renal pathology. A decrease in FA could identify the glomerular lesions, tubulointerstitial injuries and could identify different stages of kidney diseases based on the severity of renal pathology. So, DTI has the clinical potential for non-invasive diagnosing, planning therapy and monitoring of CKD. |
PDF Full Text Request