Utility of the combination of IVIM-DWI MRI and baseline eGFR for identifying a high risk of chronic kidney disease progression.

Abstract

Background: Currently, the baseline estimated glomerular filtration rate (eGFR), the urine albumin level and renal fibrosis are the common risk and prognostic factors for chronic kidney disease (CKD). Intravoxel incoherent motion (IVIM) diffusion-weighted imaging (DWI) is a proven noninvasive tool for assessing renal fibrosis. The aim of this study was to evaluate whether IVIM-DWI could be used to identify high-risk patients with CKD during long-term follow-up.

Methods: In this exploratory study, 62 CKD patients who were followed for 5 years and who underwent renal biopsy and IVIM-DWI magnetic resonance imaging (MRI) at the National Clinical Research Center of Kidney Disease in China were enrolled. We recorded baseline data, including clinical, pathology and MRI parameters, and evaluated the associations between baseline parameters and renal outcomes. The value of DWI parameters in predicting end-stage kidney disease (ESKD) was compared with that of clinical and pathological data.

Results: The mean baseline eGFR was 78.1 ± 28.05 ml/min/1.73 m², and the median eGFR slope was -0.07 (-0.43-0.06) ml/min/1.73 m²/yr. Sixteen patients eventually developed ESKD. The values of perfusion fraction (f) were positively correlated with the eGFR slope (r_s = 0.54, p = 0.028). The results of the receiver operating characteristic (ROC) analysis demonstrated that the areas under the curve (AUCs) of total apparent diffusion coefficient (ADC_T), true diffusion coefficient (D) and f in distinguishing ESKD were 0.778 (95% confidence interval [95% CI] 0.65-0.906; p = 0.001), 0.893 (95% CI 0.816-0.97; p <0.001), and 0.823 (95% CI 0.706-0.939; p < 0.001), respectively. For the combination of baseline eGFR with both D and f, the AUC was significantly greater than that for the combination of baseline eGFR and interstitial extracellular matrix volume [AUC 0.955 (95% CI, 0.909 to 1.000) vs. AUC 0.886 (95% CI, 0.803 to 0.969), p = 0.049]. Cox proportional hazard regression revealed that f was a risk and prognostic factor for ESKD after adjustment for baseline variables (p = 0.006).

Conclusion: The combination of baseline eGFR and IVIM-DWI outperforms pathological factors alone in the diagnosis of long-term kidney dysfunction. This study indicated that IVIM-DWI could be a promising tool for identifying patients at high risk of CKD progression.