Characterization of the mechanisms responsible for Artificial Kidney-Induced Kidney injury after acute kidney failure in intensive care units – MAKIKI

The repetition of acute kidney injury (AKI) episodes is recognized as a major risk factor for chronic kidney disease (CKD). It is therefore crucial for the long-term quality of life of the patients to limit the incidence of AKI and to improve renal function recovery. Renal replacement therapy (RRT) is frequently needed during the management of severe AKI. We published the first large randomized controlled trial on RRT timing (AKIKI, New England Journal of Medicine, 2016). This trial compared a classical 'early' initiation strategy with a delayed one, in the absence of AKI-related severe complication. No difference in mortality was observed between the 2 strategies. However, the delayed strategy (which allowed a substantial number of patients to escape RRT) was associated with faster renal recovery. These two main findings were recently confirmed by the STARRT-AKI, a randomized controlled international trial. Indeed, the number of patients dependent on RRT after 90 days was higher in the early RRT group. Renal replacement therapy could thus impair or delay renal recovery by acting as a “second hit” in patients already suffering from a first one (sepsis and/or shock, a very frequent situation in ICU patients). Applied to a sensitized tubular epithelium, RRT would lead to a maladaptive repair and CKD.
Several mechanisms may constitute this second hit: 1/impairment of hemodynamic; 2/activation of platelets and neutrophils by the contact with the dialysis membrane; 3/removal of plasma molecules involved in tubule regeneration (e.g. vit B3). To test these hypotheses, we will use an experimental rat model of RRT and a human biobank collected during a large multicenter randomized controlled trial (AKIKI2).
We have already set up the rat model using the hemodialysis membrane, which is now well mastered by our team. To explore the concept of RRT as a second hit, we will compare the degree of renal injury and recovery in rats receiving an administration of lipopolysaccharides (LPS) (to mimic a septic AKI) without RRT to rats receiving LPS and undergoing RRT. We will monitor the hemodynamic variations and assess the glomerular filtration rate (GFR) by transcutaneous measurement of the elimination kinetics of fluorescein isothiocyanate (FITC)-sinistrin during the RRT session. Several other parameters will be examined before, just after and 24h after RTT: plasma creatinine and urea concentration (renal function), biomarkers of renal stress such as Kidney Injury Molecule 1 (KIM-1) and renal histology by Masson's Trichrome, Giemsa stainings and immunohistological KIM1/Megalin ratio. We will also quantify platelet activation and platelet-neutrophil aggregation by FACs. Finally, we will measure plasma concentration of vitamin B3 and NAD+ metabolites.
The human biobank comes from the AKIKI 2 study, a large RCT which we performed in the preceding years and was published in 2021 (Lancet). Blood and urine samples were collected at different relevant time-points. A total of 10,715 samples taken from 295 patients are stored in Bichat Hospital (Paris, HUPNV-BRC). We will measure the level of urinary KIM-1 and urinary C–C motif chemokine ligand 14 (CCL14). An increase of these biomarkers in patients of the group that received more frequent RRT will provide supplemental argument for the hypothesis of artificial kidney-induced kidney injury. Plasma concentration of vitamin B3 and NAD+ metabolites will be also measured and correlated with the delay of renal function recovery. We will also complete the AKIKI 2 study with a long-term follow-up. The analysis of this large clinical database in association with the biobank will be a great opportunity to increase the knowledge of renal function recovery after an episode of severe AKI and to identify if RRT strategies may affect long-term renal outcome.