Preliminary study of multiparametric MRI in diagnosing acute kidney injury and chronic kidney disease without creatinine baseline in 3 months

Acute kidney injury (AKI) is a syndrome characterized by a sudden worsening in kidney function, defined as the quick loss of renal function in the previous 7 days, based on creatinine alterations[1]. Early detection for AKI patients and at risk of long-term kidney injury after AKI are fundamental for timely therapeutic interventions. The chronic phase of AKI and long-term kidney injuries can be diagnosed into different diseases: acute kidney diseases (AKD) if glomerular filtration rate (GFR) < 60 mL/min/1.73 m or creatinine ≥ 150% of baseline level in 3 months, chronic kidney disease (CKD) if there are visible anatomical abnormalities or GFR < 60 mL/min/1.73m for 3 months. Unfortunately, the majority of CKD patients and community-acquired AKI patients with abnormal creatinine levels are confused or cannot be distinguished timely at initial visit, due to lack of creatinine baseline within 3 months. Recently, the clinical application of multiparametric MRI for the early detection of AKI and CKD and prediction of long-term kidney injury is promising[2-4]. Therefore, the objectives of this study was to using DWI and T1 mapping to describe and understand the pathophysiology of AKI and CKD, which is helpful for diagnosing them timely, and try to predict the 3 months’ prognosis of patients with AKI.

We prospectively recruited 20 CKD patients, 16 AKI patients and 20 healthy volunteers as control. All subjects underwent MRI examination at Siemens PRISMA 3.0T. The sequence included T1 mapping (TR/TE: 5.01 ms/2.3 ms; voxel size: 0.8 × 0.8 × 4.0 mm; matrix: 135 × 224; slice thickness: 4 mm; slice number: 48; FOV: 305 × 380 mm; flip angle: 3° and 15°) and DWI (TR/TE: 4500ms/61ms; Voxel size: 1.6×1.6 × 5.0mm; matrix: 104×128; slice-thickness: 5mm; FOV: 325×400mm; b value: 0, 800). Two experienced radiologists used the sygno.via workstation of Siemens and measured T1 and ADC value of the renal cortex and medulla, along with the total kidney volume (TKV) of all subjects in T1WI and T2WI anatomical images. 15 CKD patients and 10 AKI patients obtained the pathological results of kidney biopsy within 2 days after MRI. The serum creatinine (Scr) and estimated glomerular rate filtration (eGFR) value of all subjects at initial visit were recorded as baseline, as well as the 1-3 months’ follow up of all the CKD and AKI patients. One-way analysis of variance was used to compare the MRI biomarkers among groups. Receiver operating characteristic (ROC) curves was used to evaluate the efficacy of MRI biomarkers for diagnosing AKI and CKD and predicting the prognosis of patients with AKI. Statistical analysis was performed using SPSS 26.0, and P<0.05 was considered statistically significant.

The baseline of Scr, eGFR, TKV, cortical and medullary values of ADC and T1 (cADC, mADC, cT1 and mT1) showed significant difference among AKI, CKD and control groups (p<0.001). In pairwise comparisons among the three groups, TKV showed no significant difference between CKD and control groups (p>0.05); cADC and mADC showed no significant difference between AKI and control groups (p>0.05); cT1 and mT1 showed no significant difference between AKI and CKD groups (p>0.05). The AUC values of baseline Scr, baseline eGFR, TKV, cADC, mADC, cT1 and mT1 for distinguishing AKI group from CKD group were 0.80, 0.82, 0.87, 0.82, 0.76, 0.58 and 0.70, respectively. Efficacies of AUCs were further enhanced by combining eGFR and TKV or combining cADC, eGFR and TKV(AUC= 0.95 and 0.98, respectively). The AUC values of baseline Scr, baseline eGFR, TKV, cADC, mADC, cT1 and mT1 for predicting the prognosis of patients with AKI were 0.64, 0.72, 0.86, 0.84, 0.80, 0.72 and 0.92, respectively.

Compared with control group, the ADC values of CKD patients were significantly lower and the T1 values were significantly higher, reflecting that the deposition of fibrous matrix in the interstitial tissue affected the diffusion of water molecules, and the atrophy of renal tubules and the expansion of interstitial space led to the increase of interstitial water content. There was no significant difference in ADC values between AKI and control groups, while T1 values of AKI patients were significantly increased, reflecting slight or no deposition of fibrous matrix in the renal interstitial tissue, mainly manifested as interstitial edema caused by renal interstitial and tubular damage[5]. Decreased ADC values in AKI patients may indicate the presence of chronic changes in the kidney represented by fibrosis, which often predicts poor prognosis. At the same time, medullary T1 value may be a sensitive indicator for predicting the prognosis of AKI patients.

Combination of DWI and T1 mapping has clinical applicability in diagnosing AKI and CKD.