FO068KIDNEY FAILURE AFTER AKI AMONG PEOPLE UNDER NEPHROLOGY CLINIC CARE: A PROVINCEWIDE COHORT STUDY

INTRODUCTION AND AIMS: Poor outcomes after acute kidney injury (AKI) have been well described for hospital cohorts, but not for those already receiving nephrology clinic care. Nevertheless, this is where discussions about future kidney failure risk commonly occur. We assessed whether AKI was independently associated with kidney failure and death (competing risk) in this setting. We then evaluated the widely used five year Kidney Failure Risk Equation (KFRE5yr), and developed refitted models with and without AKI as an additional predictor. METHODS: The study involved people receiving nephrology clinic care in British Columbia 2003-2009 (population 4.5 million), followed to 2016. Predictors were KDIGO AKI during the previous two years, age, sex, proteinuria, eGFR and renal diagnosis. Cox models estimated cause-specific hazard ratios (HR) (AKI vs no AKI) for kidney failure and death, stratified by eGFR (</≥30 ml/min/1.73m2). The published KFRE5yr was compared with refitted prediction models without and with AKI. A KFRE5yr risk threshold ≥3% has been suggested to triage entry into nephrology clinics. We used decision curve analysis to evaluate this threshold among people in clinics with eGFR ≥30. RESULTS: Of 7491 people, 995 had AKI and 6496 had no AKI in the previous two years. Those with AKI had increased subsequent kidney failure (33.1% vs 26.3%) but also increased death (23.8% vs 16.8%) (p-values <0.001), irrespective of age (figure 1), eGFR and renal diagnosis. After adjusting for confounders, AKI was still associated with kidney failure for those with eGFR ≥30 (HR 1.43 [1.13-1.81]), but no longer for those with eGFR <30 (HR 1.05 [0.91-1.21]) (table). For those with eGFR ≥30, KFRE5yr gave inferior predictions of 5 year kidney failure risk than refitted models without or with AKI (respective C statistics 0.701, 0.715 and 0.716). Using decision curve analysis, KFRE5yr at any risk threshold <10% was inferior to “treating all” patients in the clinic (figure 2). Refitted models were superior but AKI made no difference. CONCLUSIONS: AKI is associated with increased kidney failure among people under nephrology clinic care, but does not incrementally improve risk predictions because of a substantial competing risk of death. While KFRE5yr may help triage referrals to nephrology clinics, for people already in clinic with eGFR ≥30, it may not help guide safe discharge from follow-up. FO068 Table Independent association of AKI with subsequent kidney failure and death (cause-specific Cox models)

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.