A need exists to noninvasively assess renal interstitial fibrosis, a common process to all kidney diseases and predictive of renal prognosis. In this translational study, Magnetic Resonance Imaging (MRI) T1 mapping and a new segmented Diffusion-Weighted Imaging (DWI) technique, for Apparent Diffusion Coefficient (ADC), were first compared to renal fibrosis in two well-controlled animal models to assess detection limits. Validation against biopsy was then performed in 33 kidney allograft recipients (KARs). Predictive MRI indices, ΔT1 and ΔADC (defined as the cortico-medullary differences), were compared to histology. In rats, both T1 and ADC correlated well with fibrosis and inflammation showing a difference between normal and diseased kidneys. In KARs, MRI indices were not sensitive to interstitial inflammation. By contrast, ΔADC outperformed ΔT1 with a stronger negative correlation to fibrosis (R2 = 0.64 against R2 = 0.29 p < 0.001). ΔADC tends to negative values in KARs harboring cortical fibrosis of more than 40%. Using a discriminant analysis method, the ΔADC, as a marker to detect such level of fibrosis or higher, led to a specificity and sensitivity of 100% and 71%, respectively. This new index has potential for noninvasive assessment of fibrosis in the clinical setting.

Acute Kidney Injury (AKI) is strongly associated with adverse outcome and mortality independently of the cause of renal damage [1][2][3] . The mechanisms determining the deleterious systemic effects of AKI are poorly understood and specific interventions, including optimization of renal replacement therapy, had a marginal effect on AKI-associated mortality in clinical trials [4][5][6][7][8] . The kidney contributes to up to 40% of systemic glucose production by gluconeogenesis during fasting and stress conditions, mainly synthesized from lactate in the proximal tubule [9][10][11][12] , rendering this organ a major systemic lactate disposal 13 . Whether kidney gluconeogenesis is impaired during AKI and how this might influence systemic metabolism remains unknown. Here we demonstrate that glucose production and lactate clearance are impaired during human AKI using renal arteriovenous catheterization in patients. Using single cell transcriptomics in mice and RNA sequencing in human biopsies, we show that glycolytic and gluconeogenetic pathways are modified during AKI in the proximal tubule specifically, explaining the metabolic alterations. We further demonstrate that impaired renal gluconeogenesis and lactate clearance during AKI are major determinants of systemic glucose and lactate levels in critically ill patients. Most importantly, altered glucose metabolism in AKI emerged as a major determinant of AKIassociated mortality. Thiamine supplementation restored renal glucose metabolism and substantially reduced AKI-associated mortality in intensive care patients. This study highlights an unappreciated systemic role of renal glucose and lactate metabolism in stress conditions, delineates general mechanisms explaining AKI-associated mortality and introduces a potential therapeutic intervention for a highly prevalent clinical condition with limited therapeutic options.To study the impact of AKI on renal glucose and lactate metabolism, we performed renal vein catheterization in patients undergoing cardiac surgery with cardiopulmonary bypass and experiencing (n=18) or not (n=87) post-operative AKI (Supplementary Table 1). We found a switch from net renal lactate uptake to net renal lactate release in patients experiencing AKI (-0.01±0.03 mmol/min to 0.02±0.02 mmol/min, p<0.001). Moreover, AKI patients showed a decrease in the renal net glucose release (0.06±0.07 mmol/min to 0.01±0.04 mmol/min, p=0.016) compared to the control group (Fig. 1 a,b). This suggested a simultaneous reduction of gluconeogenesis and activation of glycolysis in the kidney in response to AKI. Glycosuria was unlikely to be involved since all patients had arterial glucose levels below the glycosuria threshold (5.4 ± 0.94 mmol/L). To further characterize this process, we compared kidney allograft biopsies obtained during transplantation shortly (49±16 minutes) after

Background Kidney cortical interstitial fibrosis (IF) is highly predictive of renal prognosis and is currently assessed by the evaluation of a biopsy. Diffusion magnetic resonance imaging (MRI) is a promising tool to evaluate kidney fibrosis via the apparent diffusion coefficient (ADC), but suffers from inter-individual variability. We recently applied a novel MRI protocol to allow calculation of the corticomedullary ADC difference (ΔADC). We here present the validation of ΔADC for fibrosis assessment in a cohort of 164 patients undergoing biopsy and compare it with estimated glomerular filtration rate (eGFR) and other plasmatic parameters for the detection of fibrosis. Methods This monocentric cross-sectional study included 164 patients undergoing renal biopsy at the Nephrology Department of the University Hospital of Geneva between October 2014 and May 2018. Patients underwent diffusion-weighted imaging, and T1 and T2 mappings, within 1 week after biopsy. MRI results were compared with gold standard histology for fibrosis assessment. Results Absolute cortical ADC or cortical T1 values correlated poorly to IF assessed by the biopsy, whereas ΔADC was highly correlated to IF (r=−0.52, P &lt; 0.001) and eGFR (r = 0.37, P &lt; 0.01), in both native and allograft patients. ΔT1 displayed a lower, but significant, correlation to IF and eGFR, whereas T2 did not correlate to IF nor to eGFR. ΔADC, ΔT1 and eGFR were independently associated with kidney fibrosis, and their combination allowed detection of extensive fibrosis with good specificity. Conclusion ΔADC is better correlated to IF than absolute cortical or medullary ADC values. ΔADC, ΔT1 and eGFR are independently associated to IF and allow the identification of patients with extensive IF.