Outcomes In Acute Kidney Injury Requiring Haemodialysis - A Retrospective Cohort Study

Acute kidney injury (AKI) requiring intermittent haemodialysis (AKI-IHD) is associated with significant morbidity and high mortality. There is limited data regarding clinical outcomes in individuals with AKI-IHD in Northern Ireland. The aim of this study was to explore clinical outcomes in a cohort of individuals with AKI-IHD, including rates of recovery to self-sustaining kidney function, mortality rates at 30 days and 2 years from start of haemodialysis, and to investigate potential predictors of these key outcomes.

The Acute Haemodialysis Unit in the Royal Victoria Hospital, Belfast, Northern Ireland, was established in 2011 to provide onsite inpatient intermittent haemodialysis (IHD) to individuals requiring this supportive treatment. A retrospective review of 188 incident IHD patients in the Royal Victoria Hospital from January 2018-December 2022 was undertaken. Demographic and clinical outcome information on 12th May 2023 was obtained from the nephrology electronic database eMed (Mediqal) and the Northern Ireland Electronic Care Record.

188 individuals commenced IHD for the first time as a consequence of life-threatening complications of AKI during the 5-year period (January 2018-December 2022).75% of these patients were not previously known to the nephrology service, (GROUP A, n=142, mean age 63 years, mean baseline serum creatinine 99 μmol/L) while 25% (GROUP B, n=46, mean age 67 years, mean baseline creatinine 278 μmol/L) had been attending a Nephrology Clinic for at least 12 months.A significant proportion of AKI developed during the inpatient admission rather than at initial presentation (GROUP A 47%, GROUP B 50%).92% of GROUP A recovered self-sustaining kidney function before discharge, compared to 59% of GROUP B. A lower baseline serum creatinine was the only predictor of kidney recovery in GROUP B, p value=0.02. No predictors for kidney recovery were identified in GROUP A.The diagnosis of either AKI and/or dialysis was documented in 80% of electronic discharge letters for patients in GROUP A but only 54% of letters for patients in GROUP B.The 30-day mortality (from IHD start) in GROUPA was 14% compared to 9% in GROUP B. Individuals with a diagnosis of heart failure were four times more likely to die before discharge (p value=0.02) and those aged ≥ 70 years twice as likely to die before discharge (p value=0.049). The two-year mortality rate in the two groups was similar (GROUP A 35% vs. GROUP B 37%) despite GROUP B being significantly older.

In this cohort of individuals with AKI-IHD, managed in the Royal Victoria Hospital, Belfast, the majority recovered self-sustaining kidney function.The mortality rates at 30 days were lower than reported in the literature and may be due to careful patient selection. The poorer outcomes associated with AKI-IHD support and a concomitant diagnosis of heart failure or age ≥ 70 years (or both) are useful in guiding clinical and patient expectations and decision making.