BACKGROUND: The long-term effects of different initiation strategies of continuous renal replacement therapy (CRRT) treatment in intensive care unit (ICU) patients with acute kidney injury (AKI) is unknown.
OBJECTIVES: Examine the impact of early RRT initiation on the long-term risk of chronic kidney disease (CKD), end-stage renal disease (ESRD), and death in separate analyses.
METHODS: All adult patients who required CRRT in the ICU at Aarhus University Hospital, Skejby, Denmark in the period 2005-2015 were identified. Data were obtained from a clinical information system and population-based registries. Patients with ESRD before ICU admission and residency outside Denmark were excluded. Of 1373 identified patients, 1213 were eligible. Early initiation was defined as AKI stage 2 or below at CRRT initiation and late by AKI stage 3 at CRRT initiation. AKI was defined by change in creatinine and urine output. Inverse probability of treatment (IPT) weights were computed from propensity scores. After a 5th percentile trim, the cumulative risk of CKD (eGFR < 60 ml/min/1.73 m2), ESRD, and death was estimated in IPT-weighted cohorts and compared using a Cox regression. With CKD and ESRD as outcome of interest, we accounted for death as a competing risk and only included patients who survived beyond day 90. Furthermore, with CKD as outcome of interest, we only included patients with residency in regions covered by a laboratory database.
RESULTS: The CKD, ESRD and mortality analyses included 203, 401, and 845 patients after trimming, respectively. The 5-year risk of CKD was 39% in the early group and 45% in the late group, corresponding to a hazard ratio (HR) of 0.81 (95% CI, 0.41-1.21) in early compared to late. The 5-year risk of ESRD was 15% in the early group and 16% in the late group, corresponding to a HR of 0.94 (95% CI, 0.39-1.50). The 90-day mortality in the early group was 52% compared to 47% in the late group, corresponding to a HR of 1.18 (95% CI, 0.93-1.43). The 90-day to 5-year mortality was 39% and 42% in the early and late, respectively, with a 90-day to 5-year HR of 0.99 (95% CI, 0.65-1.33).
CONCLUSIONS: Early RRT was be associated with a reduced 5-year risk of CKD, but confidence intervals were wide and included the null. While 90-day mortality may be increased in early CRRT, we found no difference in mortality beyond 90 days or risk of ESRD.
