#2997 Characterising the natural history of renal recovery after AKI in a prospective cohort—does timing matter?

Abstract Background and Aims Acute kidney injury (AKI) is associated with an increased risk of chronic kidney disease (CKD). Failure of recovery of serum creatinine by 90 days after AKI strongly associates with subsequent long-term reductions in renal function, however there is no consensus agreement on the definition of renal recovery. Further, detailed prospective descriptions of the ‘renal recovery phase’ between AKI and 90 days are lacking. Here we describe outcomes at serial timepoints (days 30, 60 and 90) for a prospective cohort of hospitalised patients with AKI. Method Single centre, prospective cohort study of participants with all stages of AKI. Each participant was characterised in detail, including AKI aetiology, co-morbidities and frailty. Baseline blood and urine samples were obtained at the time of consent. Participants were then followed up at days 30, 60 and 90 with clinical measurements and blood and urine sample collection. Blood tests included serum creatinine and GFR measured by the CKD-EPI equation. The proportion of participants with renal recovery at different creatinine above baseline thresholds as well as major adverse kidney events (MAKE) a composite outcome of death, new dialysis, and persistent renal dysfunction, defined as a 25% or greater decline in GFR was determined at day 30, 60 and 90. Where data was missing, last creatinine / GFR was carried forward for day 60 and 90, or discharge creatinine for day 30. Results A total of 124 participants were recruited. 63% of patients had AKI on hospital admission. The median baseline creatinine was 88 (IQR 70–117) μmol/L and median GFR was 68 (46–87) mL/min/1.73 m2. Median peak creatinine of 306 (IQR 177–458) μmol/L. The majority of AKI was severe (stage 1 (24%) vs stages 2 & 3 (76%)). The most common primary aetiology of AKI was dehydration (46%) followed by sepsis (22%). Table 1 demonstrates the number of participants with ‘recovered’ renal function as defined by a creatinine above baseline of 10%, 20% and 50% from baseline (the latter being reversal of KDIGO AKI stage 1) across the timepoints. Non-recovery of renal function was common across all time points, ranging from 22% to 59% at day 90 depending on which definition of recovery was applied. Table 2 summarises the rate of major adverse kidney events (MAKE) for day 30, 60 and 90, that was predominantly driven by persistent renal dysfunction. Conclusion This study is one of the few prospective descriptions of the AKI recovery from time of AKI to day 90. Whilst it is known that failure to recover creatinine by day 90 is predictive of future renal dysfunction, these data suggest that there is little dynamic change after day 30 and suggests that interventions to prevent the AKI to CKD transition should be targeted in the first 30 days after AKI in order to have the biggest impact. These data also highlight the need for a consensus definition of non-recovery of renal function. Future work on this cohort will therefore include relating different thresholds of renal recovery to outcomes.