Risk factors for RRT and mortality in patients with COVID-19-related AKI in São Paulo, Brazil

Background: In COVID-19, as in SARS, the degree of kidney injury can have major implications for the clinical outcomes. Early reports indicate that, among patients with COVID-19, AKI is common and is associated with worse outcomes. However, COVID-19-related AKI among ICU patients in Brazil has not been well described. Methods: This was a retrospective observational study of the electronic health records of patients with COVID-19-related AKI admitted to the Hospital das Clínicas in the city of São Paulo, Brazil, between March and August of 2020. We applied only KDIGO criteria 2 and 3. We used logistic regression to analyze risk factors for the composite outcome of mortality or RRT. Results: Among the 694 patients with COVID-19-related AKI, the mean age was 63 years and mortality was 66.4%;41% needed vasoactive drugs, 66% needed mechanical ventilation, and 72% needed dialysis. Univariate analysis showed the following risk factors for mortality and RRT at admission: male sex;diabetes;CKD;vasoactive drug use;mechanical ventilation;acidemia;elevated lactate, magnesium, potassium, creatinine, C-reactive protein, creatine phosphokinase, total bilirrubin;proteinuria;hematuria;and increased fractional excretion of potassium (n=98) and sodium (n=110). The factors that remained significant in the multivariate analysis were male sex, vasoactive drug use, serum magnesium >2.5 mg/dL and oliguria (24-h urine output <500 mL). Conclusions: In ICU patients with COVID-19-related AKI, in Brazil and elsewhere, in-hospital mortality is high. The exact mechanism by which hypermagnesemia increases mortality in such patients merits further study.

Adding heparin to citrate in continuous renal replacement therapy may extend filter lifespan in COVID-related AKI

Background: COVID may predispose patients to thrombosis and lower filter lifespan. Association between D-dimer level (DD) and filter clotting in Continuous Renal Replacement Therapy (CRRT) has not been described. Methods: All patients who needed CRRT in Hospital das Clínicas (Brazil) during March to May 2020 (COVID related-AKI (COV+), n=37) and August to September 2019 (COVID unrelated-AKI (COV-), n=18) were studied. Anticoagulation in CRRT in COV+ was done with citrate 3mmol/L (ACD, n=19), or citrate 4mmol/L plus non-fractioned heparin 10U/Kg/h (ACD/Hep, n=18), while in COV- with citrate 3mmol/L only. Data are expressed in median [IQR]. We performed Spearman's correlation between DD and time-free of filter clotting (TFC), and Kaplan-Meier curve to study filter survival by anticoagulation method and DD. Results: ACD/Hep group presented lower filter clotting in 72h when compared to other groups (ACD/Hep: 35% vs ACD: 100% vs COV-: 80%, p< 0.05). Analyzingfilter clotting per patient-day, ACD/Hep also presented less clotting than ACD group (ACD/Hep: 41% vs ACD: 100%, p< 0.05). In COVID patients, median TFC was 33.5 h [17.0;72.0] (ACD: 29.0 h [13.0;68.5], ACD/Hep: 40.0 h [17.0;62.0], p: NS). Clotting time from obese patients did not differ from non obese patients (obese: 31.0 h [18.5;57.2] vs non-obese: 56.0 h [16.8;72.0], p: ns). Median DD in all COVID patients was 3,519 [1420- 13,883]. Patients with DD below median (<3,500) had higher TFC (ACD high DD: 19.0 h [9.00;27.5], ACD/Hep high DD: 34.0 [17.0;62.0], ACD low DD: 57.0 h [27.2;66.8], ACD/Hep low DD: 67.0 h [26.0;72.0];Figure 1). There was statistically significance in correlation between DD and TFC in ACD patients, but not in ACD/Hep group. Conclusions: Heparin may extend filter lifespan in CRRT, and this benefit seem to be greater in high DD patients.