| Chronic kidney disease (CKD) is a prevalence disease and characterizedby progressive loss of kidney function due to chronic glomerular and/ortubulointerstitial injury. Renal fibrosis and hypoxia is considered to beimportant factors leading to the development of CKD. Clinical indicatorscurrently used for early CKD lack of sensitivity and specificity for thedetection of changes in renal function and renal pathology. To explore anoninvasive, sensitive and repeatable inspection method to evaluate the degreeof renal damage, is the current clinical workers eager concern. Diffusionweighted imaging (DWI), diffusion tensor imaging (DTI), and bloodoxygenation level dependent (BOLD) MRI were used to evaluate CKD patientsin our study. Apparent diffusion coefficient (ADC) value, FA value, and R2*value were measured on both renal cortex and medulla in patients with CKDusing3.0T MR scanner. The difference between healthy volunteers and CKDpatients were analyzed according to renal cortex and medulla. The aim of thisstudy was to explore the clinical application value of functional MR imaging inpatients with CKD. The correlation between MR data and clinical indexes,renal histopathological scores and immunohistochemical markers wereanalyzed.Part1Assessment of renal fibrosis in chronic kidney disease usingdiffusion weighted magnetic resonance imagingObjective: To assess the performance of diffusion weighted imaging(DWI) for the assessment of renal fibrosis in chronic kidney disease (CKD),with histopathology as a reference standard.Methods: From May2013to June2012, forty patients (25male,15female, mean age41.6±17.1years, range16-71years) with a clinicaldiagnosis of CKD were recruited for participation in this study in nephrology department of the third hospital of Hebei medical university. Thirty healthyvolunteers (20male and10women; average age38.3±14.1years, range24-60years) were also recruited and served as controls. All subjects underwent bothstandard MRI and DW MRI. To measure ADC values of both kidneys, ROIswere placed in the medulla and cortex on the ADC map. For CKD patients,serum creatinine (SCr) and24-h-urinary protein(24-h-UPRO) were obtainedwithin one week before or after MR scan. eGFRs were calculated from SCrmeasurements for all subjects. Renal biopsy was performed in25patients withCKD within2weeks following the MRI scans. The atoxvlin and eosin (HE)stain, periodic acid shift’s reaction (PAS) stain, Masson stain, and periodicacid-silver methenamine (PASM) stain were used for pathologic score of renalfibrosis. The sclerotic/fibrotic lesions of glomerulus and renal tubuleinterstitial were evaluated and scored. For the immunohistochemistry test ofTGF-β1and α-SMA, the images were obtained and analzed by the imageanalysis system under light microscope. Ten observation fields were selectedrandomly for each case. After scaled the image, the integrated optical densitywas calculated for each view. Finally, average of the integrated optical densitywas served as data for statistic. Mean medullary and cortical ADC valuesbetween CKD patients and healthy control subjects were compared using twoindependent samples t test, respectively. The difference between renal corticalADC and medullary ADC were analyzed using two independent sample t testfor healthy control subjects and CKD patients respectively. Bivariatecorrelation coefficients were used to investigate the relationship between ADCvalue and SCr/24-h-UPRO/eGFR. Bivariate correlation coefficients were usedto evaluate the relationship between ADC value and renalglomeruli/tubulo-interstial histopathological scores and immunohistochemicalmarkers for TGF-β1and α-SMA respectively.Results: Data from35out of40patients could be successfully analyzed.ICC of renal cortical ADC values and medullary ADC values were0.79and0.82respectively. Cortical and medullary ADC values in patients with CKDwere significantly decreased when compared to those from healthy controls. In CKD group, a significant negative correlation was found between corticalADC values and SCr/24-h-UPRO, and significant positive correlation wasfound between cortical ADC values and eGFR. There was also significantnegative correlation between medullary ADC values and SCr. No significantlycorrelation was found between medullary ADC values and eGFR/24-h-UPRO.There was a significant inverse correlation between cortical ADC andglomeruli lesions histopathologic score, with r=-0.610, p=0.001. There wasa significant inverse correlation between cortical ADC and tubulo-interstiallesions histopathologic score, with r=-0.778, p=0.000. Significant inversecorrelation was found between medullary ADC and glomeruli lesionshistopathologic score, with r=-0.398, p=0.049. Significant inversecorrelation was found between medullary ADC and tubulo-interstial lesionshistopathologic score, with r=-0.640, p=0.001. There was a significantinverse correlation between cortical ADC and theintegrated optical density ofα-SMA, with r=-0.441，p=0.027. No significant correlation was foundbetween medullary ADC and theintegrated optical density of α-SMA, with r=-0.354, p=0.083. No significant correlation was found between cortical ADCand theintegrated optical density of TGF-β1, with r=-0.362, p=0.075. Nosignificant correlation was found between medullary ADC and theintegratedoptical density of TGF-β1, with r=-0.289, p=0.162.Conclusion:3.0T MR diffusion weighted imaging in detection of renalADC in patients with CKD has certain correlation with eGFR and expressionofα-SMA. This can be used to assess the severity of renal fibrosis in patientswith CKD.Part2Assessment of renal lesions in patients with chronic kidneydisease by diffusion tensor magnetic resonance imagingObjective: To evaluate the clinical value of3.0T diffusion tensor MRimaging in detecting renal lesions in patients with chronic kidney disease.Methods: From May2013to June2012, forty patients (25male,15female, mean age41.6±17.1years, range16-71years) with a clinicaldiagnosis of CKD were recruited for participation in this study in nephrology department of the third hospital of Hebei medical university. Sixty-fourhealthy volunteers (39male and25women; average age38.5±13.3years,range21-60years) were also recruited and served as controls. All of thesubjects underwent both standard MRI and diffusion tensor MRI. To measureFA values of both kidneys, ROIs were placed in the medulla and cortex on theADC map. The coronal slices closest to the renal hilum of the left and rightkidneys were selected for ROI analysis. ROIs were placed at the upper, middleand lower pole of each kidney in the medulla and cortex on the FA map. Foreach subject, mean cortical ADC values were obtained by averaging6corticalROIs obtained from both kidneys. Similarly, mean medullary ADC valueswere obtained by averaging6medullary ROIs obtained from both kidneys.For CKD patients, serum creatinine (SCr) and24-h-urinary protein(24-h-UPRO) were obtained within one week before or after MR scan. eGFRswere calculated from SCr measurements for all subjects.Renal biopsy was performed in25patients with CKD within2weeksfollowing the MRI scans. The HE stain, PAS stain, Masson stain, and PASMstain were used for pathologic score of renal fibrosis. The sclerotic/fibroticlesions of glomerulus and renal tubule interstitial were evaluated and scored.For the immunohistochemistry test of TGF-β1and α-SMA, the images wereobtained and analzed by the image analysis system under light microscope.Ten observation fields were selected randomly for each case. After scaled theimage, theintegrated optical density was calculated for each view. Finally,average of theintegrated optical density was served as data for statistic. Dataon intra-observer agreement was determined by intraclass correlationcoefficients (ICC). Mean medullary and cortical FA values between CKDpatients and healthy control subjects were compared using two independentsamples t test, respectively. The difference between renal cortical FA andmedullary FA were analyzed using two independent sample t test for healthycontrol subjects and CKD patients respectively. Bivariate correlationcoefficients were used to investigate the relationship between FA value andSCr/24-h-UPRO/eGFR. Bivariate correlation coefficients were used to evaluate the relationship between FA values and renal histopathological scoresand immunohistochemical markers for TGF-β1and α-SMA respectively.Results: Data from33out of40patients could be successfully analyzed.ICC of renal cortical FA values and medullary FA values were0.77and0.56respectively. Cortical and medullary FA values in patients with CKD weresignificantly decreased when compared to those from healthy controls. InCKD group, cortical FA was significantly lower than medullary FA, witht=-3.598，p=0.001. In control group, cortical FA was significantly lower thanmedullary FA, with t=-15.703，p=0.000. Significant negative correlation wasfound between cortical ADC values and SCr for CKD group. Significantpositive correlation was found between cortical FA values and eGFR. Therewas significant positive correlation between medullary FA values and eGFR.No significantly correlation was found between cortical/medullary ADCvalues and eGFR/24-h-UPRO. There was a significant inverse correlationbetween cortical FA and glomeruli lesions histopathologic score, with r=-0.448, p=0.028. There was a significant inverse correlation between corticalFA and tubulo-interstial lesions histopathologic score with r=-0.487, p=0.016.Significant inverse correlation was found between medullary FA andglomeruli lesions histopathologic score, with r=-0.463, p=0.023. Significantinverse correlation was found between medullaryFA and tubulo-interstiallesions histopathologic score, with r=-0.430, p=0.036.There was a significant inverse correlation between cortical FA and theintegrated optical density of α-SMA, with r=-0.636, p=0.001. There was asignificant inverse correlation between medullary FA and the integratedoptical density of α-SMA, with r=-0.432, p=0.035. No significantcorrelation was found between cortical FA and theintegrated optical density ofTGF-β1, with r=-0.175, p=0.401。No significant correlation was foundbetween medullary FA and the integrated optical density of TGF-β1, with r=-0.189, p=0.357.Conclusion: Renal ADC values of3.0T MR DTI could quantitivelyevaluate the renal pathology changes in patients with CKD. Part3Assessment of renal oxygenation in patients with chronic kidneydisease by blood oxygenation level dependent MR imagingObjective: To evaluate the value of clinical application of3.0T blood oxygenlevel-dependent (BOLD)MR imaging in detecting renal oxygenation inpatients with chronic kidney disease.Methods: From May2013to June2012, forty patients (25male,15female, mean age41.6±17.1years, range16-71years) with a clinicaldiagnosis of CKD were recruited for participation in this study in nephrologydepartment of the third hospital of Hebei medical university. Sixty-fivehealthy volunteers (39male and26women; average age38.3±13.3years,range21-60years) were also recruited and served as controls. All of thesubjects underwent both standard MRI and BOLD MRI. To measure T2*values of both kidneys, ROIs were placed in the medulla and cortex on theT2*map. The coronal slices closest to the renal hilum of the left and rightkidneys were selected for ROI analysis. Then cortical R2*values, medullaryR2*values and medulla/cortex R2*ratio could be calculated. For CKDpatients, serum creatinine (SCr) and24-h-urinary protein (24-h-UPRO) wereobtained within one week before or after MR scan. eGFRs were calculatedfrom SCr measurements for all subjects. Renal biopsy was performed in25patients with CKD within2weeks following the MRI scans. The HE stain,PAS stain, Masson stain, and PASM stain were used for pathologic score ofrenal fibrosis. The sclerotic/fibrotic lesions of glomerulus and renal tubuleinterstitial were evaluated and scored. For the immunohistochemistry test ofTGF-β1and α-SMA, the images were obtained and analzed by the imageanalysis system under light microscope. Ten observation fields were selectedrandomly for each case. After scaled the image, the integrated optical densitywas calculated for each view. Finally, average of the integrated optical densitywas served as data for statistic. Data on intra-observer agreement wasdetermined by intraclass correlation coefficients (ICC). Mean cortical R2*values, medullary R2*values and medulla/cortex R2*ratio between CKDpatients and healthy control subjects were compared using two independent samples t test, respectively. The difference between renal cortical R2*andmedullary R2*were analyzed using two independent sample t test for healthycontrol subjects and CKD patients respectively. Bivariate correlationcoefficients were used to investigate the relationship between R2*value andSCr/24-h-UPRO/eGFR. Bivariate correlation coefficients were used toevaluate the relationship between cortical R2*values, medullary R2*values,medulla/cortex R2*ratio, and glomeruli lesions histopathologic score,tubulo-interstial lesions histopathologic score, the integrated optical density ofα-SMA, the integrated optical density of TGF-β1.Results: Data from35out of40patients could be successfully analyzed.ICC of renal cortical R2*values and medullary R2*values were0.81and0.72respectively. cortical R2*was significantly lower than medullary FA, with t=-7.576, p=0.000. In control group, cortical R2*was significantly lower thanmedullary FA, with t=-13.467, p=0.000. Medullary R2*values in patientswith CKD were significantly increased when compared to those from healthycontrols. Medulla/cortex R2*ratio in patients with CKD were significantlyincreased when compared to those from healthy controls. No significantlydifference of cortical R2*between CKD and controls. No significantlycorrelation was found between cortical R2*values, medullary R2*values,medulla/cortex R2*ratio, and SCr,24-h-UPRO, eGFR, glomeruli lesionshistopathologic score, tubulo-interstial lesions histopathologic score.Significant correlation was found between medullary R2*values andglomeruli lesions histopathologic score, tubulo-interstial lesionshistopathologic score. Significant correlation was found betweenmedulla/cortex R2*ratio and glomeruli lesions histopathologic score,tubulo-interstial lesions histopathologic score. There was a significantcorrelation between corticalR2*and the integrated optical density of TGF-β1.No significant correlation was found between medullary R2*and theintegrated optical density of α-SMA and TGF-β1. No significant correlationwas found between medulla/cortex R2*ratio and the integrated optical densityof α-SMA or TGF-β1. Conclusion:3.0T BOLD MR could quantitively evaluate theoxygenation changes in patients with CKD. |
PDF Full Text Request