#3372 HOST RESPONSE CHANGES ASSOCIATED WITH PERSISTENT ACUTE KIDNEY INJURY IN CRITICALLY ILL PATIENTS WITH COVID-19

Abstract

Abstract Background and Aims In sepsis, the dysregulated host response has been, at least in part, linked to development of acute kidney injury (AKI) and its duration. Data on the host response to COVID-19 and development of persistent-AKI (duration > 48 h) in critically ill patients are scarce. We hypothesized that COVID-19 infected patients who develop persistent-AKI have different host response aberrations. To address this hypothesis, we sequentially measured host immune response biomarkers in critically ill patients admitted to the intensive care unit (ICU). Method We included patients admitted to the ICU from two tertiary hospitals and one primary care hospital between March and July 2020 (first wave). All patients had PCR-confirmed COVID-19. Excluded were readmitted patients, transfers from another ICU, and patients with chronic kidney disease (CKD stage 3 or higher). The presence of AKI was assessed by using hourly urine output and daily serum creatinine levels, and classified into 3 stages according to Kidney Disease: Improving Global Outcomes (KDIGO) criteria. In a subset of patients admitted to the two tertiary hospitals (n=113), 41 plasma protein biomarkers were sequentially measured using a Luminex platform or ELISA, and categorized into five pathophysiological domains: systemic inflammation, endothelial cell activation and dysfunction, coagulation activation, kidney dysfunction, and lung dysfunction. Mixed effects models were used to analyse longitudinal data. Results Of 185 ICU admissions enrolled, 8 were excluded because of existing CKD. Of the remaining 177 patients, 106 patients (59.9%) had AKI within the first 48 h of admission; of these, 76 patients (71.7%) had persistent AKI and 30 patients (28.3%) transient AKI. Patients with persistent AKI more often were obese and had a history of hypertension. A higher disease severity – as determined by the sequential organ failure assessment (SOFA) score - was observed in the persistent AKI group, which was driven by the renal component as the non-renal SOFA score was similar between no-AKI, transient-, and persistent AKI. Sequential protein biomarker analyses revealed that five biomarkers were significantly elevated in persistent-AKI compared to transient-AKI in the first four days, mainly related to inflammation (triggering receptor expressed on myeloid cells-1 [TREM-1], p=.003; tumour necrosis factor receptor 1 [TNF-RI], p < 0.001); lung dysfunction (clara cell secretory protein [CC16], p < 0.001), and kidney dysfunction (Cystatin C [p < 0.001]; neutrophil gelatinase-associated lipocalin [NGAL], p < 0.001). Plasma protein biomarkers reflective of endothelial cell- and coagulation activation were not different in patients with persistent-AKI as compared to patients with transient-AKI. No differences in plasma biomarkers were observed between transient AKI and no-AKI. Conclusion Transient AKI and no-AKI revealed little to no differences, while the persistent AKI group demonstrated stronger host response anomalies across the pathophysiological domains of inflammation and lung dysfunction. In contrast to what has been observed in non-COVID-19-related sepsis, endothelial injury and coagulation activation markers were not associated with AKI trajectories. These findings suggest that COVID-19 can induce a robust immune response that contributes to persistence of AKI, largely mediated through inflammatory responses.