MO336: Renal Recovery After Acute Kidney Injury Associated to COVID Requiring CRRT During ICU Hospitalization

Abstract BACKGROUND AND AIMS Renal recovery (RR) after AKI is a determinant outcome of future comorbidity and mortality in critical care patients. Related predictive factors remain uncertain. METHOD We retrospectively analyzed patients admitted to ICU between January 2020 and February 2021 from our critical nephrology database. We analyzed adult patients with diagnosis of AKI (KDIGO criteria) treated with renal replacement therapy (RRT) during ICU hospitalization. We excluded patients with dialysis support previous to the admission. The main outcomes we evaluated were (1) RR (successful suspension of RRT without hyperkalemia, increase in serum creatinine (SCr), hypervolemia or acidemia after 1 week without RRT, with urine volume > 500 mL/d without diuretic treatment or > 2000 mL/d with diuretics). (2) Mortality during hospitalization. RESULTS We found 1442 patients were admitted to ICU, 418 presented AKI (29.8%), of them, 178 patients (64% male) required RRT (AKI-RRTd) in ICU during follow-up, with mean age of 66 year old (52.8% >65 year). Main comorbidity and demographic data are in Table 1. Mean time in ICU was 19 days (RIC 11–35). The most frequent admission cause was non-surgical pathologies (93%), 53% of admitted patients had COVID-19 as main diagnosis (95 patients). There was need of vasoactive support in 73.6%, ventilatory support (82.6) and 67.2% of patients had fluid overload. The indication of dialysis was determined by a nephrologist: mainly oliguria, acidosis, hyperkalemia, fluid overload and increase SCr. Mean SCR at admission was 2.5 mg/dL. There were missing data in 48% of basal SCr (known SCr between 1 and 12 months prior to admission). Total mortality in AKI-RRTd was 70.8% (126 patients). In COVID patients, was 77.9% (74 patients). We found renal recovery in 63.4% of total survivors (33/52 patients). When analyzing COVID, there were 21 survivors, and we found renal recovery in 80.9% of patients. Patients who did not achieved renal recovery had longer ICU stay (median: 20 days, RIC: 4–26) and inhospitalization (median: 41 days, RIC: 29–58). Those patients were older, and had higher morbidity (diabetes), higher SCr at ICU admission and lower urine output. Their fluid balance was higher at 48 h after CRRT initiation (OR 3.05, 95% CI 1.39–6.65, P <.01). In COVID population without renal recovery, there were more urgent dialysis onset (OR 8.33, 95% confidence interval (95% CI) 1.04–66.2; P = .04), age > 65 year (OR 6.48, 95% CI 1.94–21.6; P < .01), positive fluid balance at 48 h after RRT (OR 3.25; 95% CI 1.09–9.69; P = .03). The risk factors for mortality, were age > 65 year (OR 4.14, 95% CI 2.05–8.35; P < .01), mechanical ventilation (OR 3.28, 95% CI 1.48–7.30; P < .01), haemodynamic support (OR 4.37, 95% CI 2.14–8.92; P < .01). Otherwise, lower SCr at admission (OR 0.82, 95% CI 0.71–0.93; P < .01) and at instauration of RRT (OR 0.75, 95% CI 0.065–0.88; P < .01) were associated to lower mortality. In COVID patients, fluid overload at RRT initiation (OR 10.83, 95% CI 1.37–85.36; P = .02), age > 65 year old (OR 8.85, 95% CI 2.68–29.1; P < .01) and FiO2 > 50% at RRT start (OR 2.77, 95% CI 1.02–7.50; P = .04) were associated to higher mortality. CONCLUSION In ICU patients with AKI-RRT dependence, negative fluid balance at 48 h after RRT onset and in COVID patients, age < 65 year old, negative fluid balance at 48 h after RRT onset and non-urgent onset of RRT were related with renal recovery.