Acute Kidney Injury in Critically-ill Patients With COVID-19: A Single-center Experience of 206 Consecutive Patients.

Abstract

Knowledge about the epidemiology, diagnosis, and management of coronavirus disease 2019 (COVID-19) is rapidly evolving. More evidence is accumulating that COVID-19 is a systemic disease with more extra-pulmonary manifestations than initially thought.1–3 We describe here the incidence, predictors, and prognosis of acute kidney injury (AKI) in critically-ill patients with COVID-19. Data were collected prospectively for all patients with severe COVID-19 admitted to all intensive care units (ICUs) at the Massachusetts General Hospital, Boston, MA, USA between March 13th and April 22nd, 2020. COVID-19 infection was confirmed in all patients via reverse-transcriptase-polymerase-chain-reaction testing. AKI with or without the need for renal replacement therapy (RRT) was defined using the Kidney Disease: Improving Global Outcomes criteria.4 In our institution, the decision to start RRT is made by the nephrology team together with the critical care team. Univariate and multivariable analyses were performed to identify independent predictors for AKI and the need for RRT. A total of 206 consecutive patients were included. The median [interquartile ranges (IQR)] age was 60 years (47, 71), and 134 patients (65.1%) were males; 89 patients (43.2%) had diabetes mellitus, 27 (13.1%) had a history of chronic kidney disease, and 7 (3.4%) were dialysis-dependent before hospital admission. The median (IQR) Sequential Organ Failure Assessment score at the time of ICU admission was 5 (3, 7.5), and 87% of patients were intubated and mechanically ventilated. Of 206 patients, 148 (71.8%) developed AKI. Table S1 in the Supplementary Appendix, https://links.lww.com/SLA/C440 shows the univariate analyses comparing the demographics, comorbidities, and clinical presentation of AKI and non-AKI patients. In the AKI patient population, 62 (41.9%) had significant hyperkalemia (K+ > 5.5 mmol/L), 37 (25.0%) had oliguria/anuria (urine output < 100 mL/d), and 103 (69.6%) developed metabolic acidosis (pH < 7.30). The mean highest creatinine and blood urea nitrogen of patients that developed AKI were 3.80 mg/dL and 77 mg/dL, respectively. Forty-three percent of patients were diagnosed with AKI on hospital admission, whereas 89.2% of the patients developed AKI by hospital day 6 (Fig. 1 in the Supplementary Appendix, https://links.lww.com/SLA/C443). Almost half of the patients (70 patients, 47.3%) with AKI had a severe stage 3 disease, and of those, 46 (65.7%) progressed to require RRT; 24 patients were started and maintained on accelerated veno-venous hemofiltration or continuous veno-venous hemofiltration, 4 patients on hemodialysis, and 18 patients were started on accelerated veno-venous hemofiltration/continuous veno-venous hemofiltration but later transitioned to hemodialysis. The median (IQR) duration of RRT in our patient population was 14.5 (3, 22) days; the median (IQR) duration of RRT in the surviving cohort of patients was 18 (15, 32) days. At the time of this report, out of the 46 patients requiring RRT, 27 (58.7%) are alive; 7 are still in the hospital (5 have recovered renal function and 2 remain RRT dependent) and 20 patients have been discharged (15 have recovered renal function and 5 patients were discharged on RRT). Clotting in the dialysis circuit was encountered in 33 (73.3%) patients undergoing RRT. On multivariable analyses, older age, male gender, and higher body mass index (BMI) were identified as independent predictors for the development of AKI (Table S2 in the Supplementary Appendix, https://links.lww.com/SLA/C441). Similarly, male sex, higher BMI, and hypertension were identified as independent predictors of the need for RRT among AKI patients (Table S3 in the Supplementary Appendix, https://links.lww.com/SLA/C442). To date, both ICU and hospital mortality are significantly higher in the AKI vs. non-AKI patients (27.7% vs 5.2%, P-value < 0.001 and 33.1% vs 8.6%, P-value < 0.001, respectively). At the time of this report, 8 (3.9%) patients are still hospitalized. In this series of critically-ill COVID-19 patients, we report an extremely high incidence of AKI; almost a third of the patients required RRT. This might be secondary to kidney hypoperfusion or medication adverse effects; however, COVID-19 related pathophysiologic mechanisms, including cytokine storm and cell-mediated renal injury, could have played a role and thus warrant further investigation.5,6 Male gender and higher BMI were identified as independent predictors for the development of AKI and the requirement of RRT among critically-ill COVID-19 patients. The rate of AKI reported herein is significantly higher than the one reported by Hirsch et al (71.8% vs 36.6%).7 In their study the identified risk factors for AKI included older age, diabetes mellitus, cardiovascular disease, black race, hypertension, ventilation requirement, and administration of vasopressors.7 The hospital mortality rate of AKI patients was nearly 4 times that of non-AKI patients. Front line clinicians should be aware of these renal morbidities and should monitor closely patients’ renal function with prompt early intervention when possible.