The diagnostic efficacy of diffusion tensor imaging generated by gadolinium-based magnetic resonance imaging for patients with chronic kidney disease.

BACKGROUND: Chronic kidney disease (CKD) can lead to systemic inflammatory responses and other cardiovascular disease. Diffusion tensor imaging findings generated by gadolinium-based MRI (DTI-GBMRI) is regarded as a standard method for assessing the pathology of CKD. To evaluate the diagnostic value of DTI-GBMRI for renal histopathology and renal efficiency, renal fibrosis and damage, noninvasive quantification of renal blood flow (RBF) were investigated in patients with CKD.
METHODS: CKD patients (n = 186) were recruited and underwent diagnosis of renal diffusion tensor imaging findings generated by MRI (DTI-MRI) or DTI-GBMRI to identify the pathological characteristics and depict renal efficiency. The cortical RBFs and estimated glomerular filtration rate were compared in CKD patients undergone DTI-GBMRI (n = 92) or DTI-MRI (n = 94).
RESULTS: Gadolinium enhanced the diagnosis generated by DTI-MRI in renal fibrosis, renal damage, and estimated glomerular filtration rate. The superiority in sensitivity and accuracy of the DTI-GBMRI method in assessing renal function and evaluating renal impairment was observed in CKD patients compared with DTI-MRI. Outcomes demonstrated that DTI-GBMRI had higher accuracy, sensitivity, and specificity than DTI-MRI in diagnosing patients with CKD.
CONCLUSION: In conclusion, DTI-GBMRI is a potential noninvasive method for measuring renal function, which can provide valuable information for clinical CKD diagnosis.

1. Introduction

Chronic kidney disease (CKD) is a public health problem worldwide, which is a gradual impairment of renal function.[ Patients with CKD are at an increased risk of cardiovascular disease and other chronic conditions with their daily lives.[ Clinically, CKD is characterized by a low estimated glomerular filtration rate (eGFR, <60 mL/min/1.73 m2), which may progress at varying rates depending on blood pressure management, history of decreased GFR, level of proteinuria.[ CKD may result in abnormalities of multiple physiological processes including removal of waste products of metabolism, heterogeneous disorders on kidney structure and function, electrolyte balance that substantially increases mortality risk due to atherosclerotic cardiovascular disease (ASCVD).[ Therefore, CKD patients require imaging monitoring to accurately predict the risk of declining renal function and guide therapeutic schedules.

CKD is generally diagnosed by imaging of the kidney using ultrasonography (US), contrast-enhanced ultrasound (CEUS), computed tomography (CT), or magnetic resonance imaging (MRI).[ Cardiovascular magnetic resonance (CMR) is a standard method for imaging impairment in kidney function, renal injury, and fibrosis.[ MRI technique in particular has the potential to present both structural and functional parameters in the kidney by using targeted magnetic nanoparticles.[ Gadolinium (GB)-based are an important aid in MRI diagnostics for improving the detection and characterization of pathologic processes.[ Gadolinium GB-based contrast agents are frequently used to enhance the diagnostic efficacy of MRI in patients with kidney diseases.[ In addition, GB is widely employed as a contrast agent for MRI and has generally been considered to be safe in patients with kidney disease.[ Furthermore, a potential alternative imaging modality for evaluating GFR and renal fibrosis is diffusion tensor imaging findings generated by Gadolinium-based MRI (DTI-GBMRI).[ Thus, DTI-GBMRI may be ideal for renal function assessment in patients with CKD.

The purpose of this study was to comprehensively assess renal histopathology and renal efficiency, renal fibrosis and damage, noninvasive quantification of renal blood flow (RBF) in patients with CKD using DTI-GBMRI. The sensitivity and accuracy between DTI-GBMRI and DTI-MRI method in assessing renal function and evaluating renal impairment was compared in CKD patients. The association between the RBF value and eGFR was analyzed in patients with CKD.

2. Materials and methods

2.1. Patients

A total of 186 CKD patients were recruited in Hongqi Hospital Affiliated To Mudanjiang Medical University between January 2017 and May 2019. The inclusion criteria were as follows: age > 18 years; and Stage 2 of CKD (60 ml/min/1.73 m2 ≤ eGFR <90 mL/min/1.73 m2); Stage 3a of CKD (30 mL/min/1.73 m2 ≤ eGFR <60 mL/min/1.73 m2); The exclusion criteria were as follows: patients with HIV infection, polycystic kidney disease, cancer; transplant recipients; pregnant and breastfeeding women; and history of adverse reaction to gadolinium. The protocol was approved by the ethics committee of Hongqi Hospital Affiliated To Mudanjiang Medical University. CKD patients received DTI-GBMRI (n = 92) or DTI-MRI (n = 94) diagnosis. All participants signed written informed consent.

2.2. Magnetic resonance imaging

MRI examinations were performed using a 3 T unit (Canon Medical Systems, Tustin, CA). Diffusion-weighted imaging/diffusion tensor imaging scans were obtained by using the following parameters: 256 diffusion directions, TR: 8000 ms, TE: 60 ms, 8 diffusion-weighted b-values in steps of 200 s/mm2 ranging from b: 0 to 14,000 s/mm2, flip angle: 90°; bw: 1860 Hz/px, transversal base resolution matrix: 128 × 128. For DTI-GBMRI, a bolus injection of 0.2 mL/kg body weight gadolinium (Omniscan; Bracco, Daiichi-San-kyo Co., Ltd, Tokyo, Japan) was intravenously administered, followed by a 20-mL saline flush at 2 mL/s. The location of impairment of renal structure can be observed in MRI images, where exposure to GB-based contrast agent in CKD patients. MR renography was obtained from all CKD patients to analyze renal histopathology and renal efficiency, renal fibrosis and damage, and noninvasive quantification of RBF.

2.3. Outcomes

Kidney volume, the number, diameter, and volume of glomeruli in CKD patients were automatically analyzed using MRI image data. The eGFR was calculated using the estimation equation for CKD patients determined by MRI image data. The GFR was calculated for each kidney by 3 radiologists by using Mirage software.[ The mean transit time (MTT) was used to evaluate function of kidney in CKD patients as described previously.[ RBF was determined by MRI image data.[ The procedure includes preprocessing of image data, segmentation of the kidney region, segmentation of the glomeruli, and quantification of the segmented regions as described previously.[ All parameters were automatically analyzed by Syngo software (Siemens, Erlangen, Germany).

2.4. Statistical analyses

Data are expressed means ± SD. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Inc., Tokyo, Japan). The P values were calculated via independent sample t test for continuous variables and chi-square test for categorical variables. Receiver operating characteristic (ROC) analyses were used to analyze the diagnostic performance of the DTI-GBMRI diagnosis. The cutoff was determined according to the ROC curve, and then the specificity and sensitivity of various scoring systems were calculated separately. Statistical significance was defined as a P value <.05.

3. Results

3.1. Characteristic of patients with CKD

A total of 186 CKD patients were enrolled in this study. All CKD patients in stage 1 or 2 were enrolled between May 2017 and June 2019. A flowchart of CKD patient recruitment is shown in Figure 1. The age of CKD patients was 46.5 years old (range, 30–65 years). CKD patients received diagnosis of DTI-GBMRI (n = 92) or DTI-MRI (n = 94) to identify the pathological characteristics and depict renal efficiency. Table 1 showed the demographics and characteristics of patients with CKD. There were no significant differences in BMI, age, sex, blood pressure, cerebrovascular disease, and intraocular pressure between 2 groups. Signalment and renal biomarkers in 2 groups were not significant difference.

Figure 1.

Flow diagram showing CKD patients in this study. CKD = chronic kidney disease.

Table 1

Characteristic of CKD patients.

	DTI-MRI	DTI-GBMRI
Number	94	92
Male/female	54/40	53/39
Age (y)	52 ± 10	52 ± 10
BMI (kg/m2)	25.2 ± 2.6	25.8 ± 3.1
Blood pressure (mm Hg)
Systolic	132.3 ± 8.5	130.8 ± 9.0
Diastolic	82.6 ± 5.8	84.2 ± 6.5
Cerebrovascular disease	10 (10.6%)	9 (9.8%)
Intraocular pressure (mm Hg)	14.5 ± 2.6	14.8 ± 3.1
Smoking, n (%)	10 (10.6%)	10 (10.9%)
CKD stages
2 (mild)	50 (53.2%)	50 (54.3%)
3a (mild to moderate)	44 (46.8%)	42 (45.7%)

Data are expressed as mean ± or n (%). The P values were calculated via independent sample t test for continuous variables and chi-square test for categorical variables.

BMI = body mass index, CKD = chronic kidney disease, DTI-GBMRI = diffusion tensor imaging findings generated by gadolinium-based MRI, DTI-MRI = diffusion tensor imaging findings generated by MRI.

3.2. The pathological characteristics and depict renal efficiency diagnosed by DTI-GBMRI

We compared the pathological characteristics and depict renal efficiency in CKD patients diagnosed by DTI-GBMRI or DTI-MRI. CKD patients diagnosed by DTI-GBMRI showed more renal vascular lesions and bigger diameters of lesions than those patients in DTI-MRI group. The average diameter diagnosed by DTI-GBMRI was 3.28 cm, while was in 3.02 cm in CKD patients diagnosed by DTI-MRI. DTI-GBMRI had significantly better performance than DTI-MRI in measuring lumen depiction scores (4.8 ± 0.2 vs 3.2 ± 0.2 for arterial inflow, 4.2 ± 0.1 vs 1.8 ± 0.2 for arterial outflow, 3.8 ± 0.3 vs 2.0 ± 0.3 for anastomosis, and 4.4 ± 0.2 vs 2.2 ± 0.2 for venous outflow; P < .01), intraluminal signal homogeneity (0.24 ± 0.03 vs 0.40 ± 0.05, P < .01), and contrast-to-noise ratio efficiency in the venous outflow (5.2 ± 0.3 vs 2.6 ± 0.3, P < .01) (Table 2).

Table 2

The pathological characteristics and depict renal efficiency diagnosed by DTI-GBMRI.

	DTI-MRI	DTI-GBMRI	p value
Diameters of lesions (cm)	3.02 ± 0.32	3.28 ± 0.24	.0045
Depiction scores	3.2 ± 0.2	4.8 ± 0.2	.0032
Arterial inflow	1.8 ± 0.2	4.2 ± 0.1	.0010
Arterial outflow	2.0 ± 0.3	3.8 ± 0.3	.0041
Venous outflow	2.2 ± 0.2	4.4 ± 0.2	.0036
Intraluminal signal homogeneity	0.40 ± 0.05	0.24 ± 0.03	.0057
CNR efficiency	2.6 ± 0.3	5.2 ± 0.3	.0028

Data are reported as mean ± SD. The P values were analyzed using independent sample t test.

CNR = contrast-to-noise ratio, DTI-GBMRI = diffusion tensor imaging findings generated by gadolinium-based MRI, DTI-MRI = diffusion tensor imaging findings generated by MRI.

3.3. Kidney volume and functional parameters estimates

Kidney volumes and functional parameters estimates were compared in CKD patients between DTI-GBMRI and DTI-MRI groups (Table 3). The mean kidney volumes of CKD patients in DTI-GBMRI and DTI-MRI were 2046 ± 214 and 1868 ± 224 mL, respectively (P < .01). Creatinine levels were not significantly different between DTI-GBMRI and DTI-MRI groups. The total glomerular count in CKD patients diagnosed by DTI-GBMRI and DTI-MRI was around 16,320 ± 12,350 and 14,560 ± 10,180, respectively (P < .01). MTTK was significantly higher in the DTI-MRI group (186.2 ± 40.5 seconds) than in the DTI-GBMRI group (137.6 ± 30.6 seconds). DTI-GBMRI revealed stronger impairment of renal perfusion (156 ± 7 vs 293 ± 44 mL/[min × 100 g]; P < .01) and more pronounced increases in T2 (60.1 ± 2.0 vs 45.7 ± 1.2 ms, P < .01) and T1 relaxation times (1938 ± 53 vs 1350 ± 27 ms, P < .01) than DTI-MRI. Apparent diffusion coefficient was 1.39 ± 0.14 × 10−3 and 1.83 ± 0.05 × 10−3 mm2/s in kidneys in the DTI-GBMRI group and DTI-MRI group, respectively (P < .05).

Table 3

Kidney volume and functional parameters estimates diagnosed by DTI-GBMRI.

	DTI-MRI	DTI-GBMRI	P value
Kidney volumes (mL)	1868 ± 224	2046 ± 214	.0008
Creatinine (mg/dL)	1.68 ± 0.32	1.75 ± 0.40	.0091
MTTK	186.2 ± 40.5	137.6 ± 30.6	.0025
Impairment of renal perfusion (mL/min)	293 ± 44	156 ± 7	.0010
T2 pronounce (ms)	45.7 ± 1.2	60.1 ± 2.0	.0040
T1 relaxation times (ms)	1938 ± 53	1350 ± 27	.0012
CNR efficiency (mm2/s)	1.83 ± 0.05 × 10−3	1.39 ± 0.14 × 10−3	.0085

Data are reported as mean ± SD. The P values were analyzed using independent sample t test.

CNR = contrast-to-noise ratio, DTI-GBMRI = diffusion tensor imaging findings generated by gadolinium-based MRI, DTI-MRI = diffusion tensor imaging findings generated by MRI.

3.4. Cortical RBFs and eGFR

The cortical RBFs and eGFR were compared in CKD patients undergone DTI-GBMRI (n = 92) or DTI-MRI (n = 94). The cortical RBF values of CKD patients in the DTI-GBMRI group were lower than patients in DTI-MRI group (315.8 ± 23.6 vs 503.5 ± 32.4, P < .01). Outcomes demonstrated that eGFR was significantly lower in CKD patients who undergone DTI-GBMRI than those undergone DTI-MRI (36.4 ± 20.8 vs 44.7 ± 16.5 mL/min) (Table 4). Results showed that gadolinium enhanced the diagnosis generated by DTI-MRI in renal fibrosis, renal damage, and eGFR.

Table 4

Analysis of Cortical RBFs and eGFR in CDK patients.

	DTI-MRI	DTI-GBMRI	P value
Serum creatinine (mg/dL)	1.68 ± 0.32	1.75 ± 0.40	.0058
eGFR (mL/min/1.73 m2)	36.4 ± 20.8	44.7 ± 16.5	.0072
BRF(ml/min)	503.5 ± 32.4	315.8 ± 23.6	.0020

Data are reported as mean ± SD. The P values were analyzed using independent sample t test.

CKD = chronic kidney disease, DTI-GBMRI = diffusion tensor imaging findings generated by gadolinium-based MRI, DTI-MRI = diffusion tensor imaging findings generated by MRI, eGFR = estimated glomerular filtration rate, RBF = renal blood flow.

3.5. Diagnostic efficacy of DTI-GBMRI

The ROC curve was used to determine the diagnostic efficacy between DTI-MRI and DTI-GBMRI groups in CKD patients. The DTI-GBMRI method showed the superior diagnostic accuracy compared to DTI-MRI. Outcomes demonstrated that DTI-GBMRI had higher sensitivity and specificity than DTI-MRI in diagnosing patients with CKD (Table 5). From the curve, the cutoff value for and DTI-GBMRI were determined with maximum sensitivity and specificity to be 0.92 and 0.85, respectively (Fig. 2).

Table 5

Diagnostic efficacy of DTI-GBMRI in CKD patients.

	DTI-MRI	DTI-GBMRI	P value
Sensitivity (%)	86.8	98.5	.0048
Specificity (%)	88.3	98.2	.0070

Sensitivity and specificity differences between DTI-MRI and DTI-GBMRI were analyzed using Pearson nonparameter correlation analysis.

CKD = chronic kidney disease, DTI-GBMRI = diffusion tensor imaging findings generated by gadolinium-based MRI, DTI-MRI = diffusion tensor imaging findings generated by MRI.

Figure 2.

Receiver operating characteristic curve of DTI-GBMRI and DTI-MRI for the diagnosis of CKD patients. CKD = chronic kidney disease, DTI-GBMRI = diffusion tensor imaging findings generated by gadolinium-based MRI, DTI-MRI = diffusion tensor imaging findings generated by MRI.

4. Discussion

The assessment of early CKD damage is of crucial importance in preventing CKD-induced diseases.[ In this study, we investigated the diagnostic efficacy of DTI-GBMRI in evaluating histopathology and renal efficiency in patients with CKD. Functional parameters associated with renal impairment in kidney function were analyzed in DTI-GBMRI-diagnosed CKD patients with CKD. Findings in this study demonstrated that DTI-GBMRI clearly demonstrated the pathological characteristics and depict renal efficiency compared to MRI in CKD patients. Thus, DTI-GBMRI may be a potential noninvasive method for measuring renal function for CKD patients.

CKD has been associated with increased visual impairment and cardiovascular disease.[ Inflammation and dysfunction of glomerular cells contributes to the cardiovascular disease burden associated with CKD, which is one of the most important risk factors for mortality and morbidity in CKD patients.[ Thus, it is very important to evaluate the degree of CKD and analyze abnormalities of kidney structure or function.[ MRI methods have become available for renal functional evaluation for patients with CKD, which can access renal function combined with high-resolution morphological evaluation of the kidneys and the entire urinary tract.[ In this study, we found that DTI-GBMRI presented higher efficacy in evaluating renal histopathology and renal efficiency, renal fibrosis and damage, and noninvasive quantification of RBF than DTI-MRI (Table 2). Data reported that DTI-GBMRI clearly showed the diameters of lesions, good performance in measuring lumen depiction scores, intraluminal signal homogeneity, and contrast-to-noise ratio efficiency in CKD patients compared to DTI-MRI, suggesting DTI-GBMRI provided the possible origins of differences in estimates of CKD prevalence, and presented possible solutions for tackling the factors responsible for the reported variations in renal injury measurements (Table 3).

Previously, MRI allows the assessment of markers of perfusion without the use of contrast media.[ A study highlighted the potential clinical benefits of early detection in patients predisposed to CKD by using MRI techniques, which provided structural and functional information in the kidney.[ Zhao et al[ found that diffusion-weighted MRI can be used to assess renal fibrosis in CKD patients. Data in this study observed that DTI-GBMRI-diagnosed CKD patients had higher MTTK, stronger impairment of renal perfusion and more pronounced increases in T2 and T1 relaxation times than DTI-MRI in CKD patients. However, there was no significant difference in evaluating kidney volumes between DTI-GBMRI and DTI-MRI groups. A previous study demonstrated that MRI could classify renal function, identify eGFR and functional renal parenchyma RBF in CKD patients.[ Findings in the current study demonstrated that DTI-GBMRI-diagnosed CKD patients showed lower cortical RBF values and eGFR than those patients diagnosed by DTI-MRI.

The MRI method has been used in diagnosing CKD because it is a noninvasive and accessible method.[ However, its operator dependency and low sensitivity reduce its utility in research. This study introduced gadolinium-based MRI to improve operator dependency and low sensitivity in diagnosing cortical RBFs, eGFR, renal fibrosis, renal damage, and eGFR in CKD patients. The strategies of this study compared the diagnostic efficacy between DTI-GBMRI and DTI-MRI, and outcomes found that the DTI-GBMRI method showed the superior diagnostic accuracy, sensitivity and specificity compared to DTI-MRI in patients with CKD.

Limitations of our study included the absence of pathologic correlation with MRI findings. In addition, as with most published articles investigating GB deposition, any GB contrast agent injections in addition to the ones reported in clinical records could not be excluded in the study population. Furthermore, this study did not investigate the potential safety/toxicity issues of DTI-GBMRI. Moreover, the sample size is small.

In conclusion, this study demonstrates the benefits of DTI-GBMRI in measuring renal histopathology and renal efficiency, renal fibrosis and damage, and noninvasive quantification of RBF in CKD patients. Outcomes find that DTI-GBMRI improves testing methodologies for more accurate assessment of cortical RBFs, GFR, pathological characteristics, and depict renal efficiency than DTI-MRI, which further contributes to high sensitivity and specificity. These data suggest that DTI-GBMRI may be a reliable assessment of renal function combined with high-resolution morphological evaluation of the kidneys, as well as accurately identify stage CKD in certain clinical patients.

Author Contributions

Conceptualization: Liu He.

Data curation: Gao Dan.

Formal analysis: Sun Yuanbo.

Investigation: Tang Fengqiong.

Methodology: Liu He, Gao Dan, Li Hongyi.

Project administration: Li Hongyi.

Software: Liu He, Suan Yuanbo.

Supervision: Hu Mingcheng.

Validation: Hu Mingcheng.

Visualization: Hu Mingcheng.

Writing – original draft: Li Hongyi.

Writing – review & editing: Li Hongyi.