Community-acquired versus hospital-acquired acute kidney injury at a large Australian metropolitan quaternary referral centre: incidence, associations and outcomes.

Abstract

There is increasing global incidence of acute kidney injury (AKI) and significant short- and long-term impacts on patients. To determine incidence and outcomes of community-acquired AKI (CA-AKI) and hospital-acquired AKI (HA-AKI) among inpatients in the Australian healthcare setting using modern health information systems. A retrospective cohort study of adult patients admitted to a quaternary hospital in Melbourne, Australia, between 1 January 2018 and 31 December 2019 utilising an electronic data warehouse. Participants included adult patients admitted for >24 h who had more than one serum creatinine level recorded during admission. Kidney transplant and maintenance dialysis patients were excluded. Main outcomes measured included AKI, as classified by the Kidney Disease Improving Global Outcomes (KDIGO) criteria, hospital length of stay and 30-day mortality. A total of 6477 AKI episodes was identified across 43 791 admissions. Of all AKI episodes, 77% (n = 5011), 15% (n = 947) and 8% (n = 519) were KDIGO stage 1, 2 and 3 respectively. HA-AKI accounted for 55.9% episodes. Patients required intensive care unit admission in 22.7% (n = 1100) of CA-AKI and 19.3% (n = 935) of HA-AKI, compared with 7.5% (n = 2815) of patients with no AKI (P = 0.001). Patients with AKI were older with more co-morbidities, particularly chronic kidney disease (CKD). Length of stay was longer in CA-AKI (8.8 days) and HA-AKI (11.8 days) compared with admissions without AKI (4.9 days; P < 0.001). Thirty-day mortality was increased with CA-AKI (10.2%) and HA-AKI (12.8%) compared with no AKI (3.7%; P < 0.001). The incidence of AKI detected by the electronic data warehouse was higher than previously reported. Patients who experienced AKI had greater morbidity and mortality. CKD was an important risk factor for AKI in hospitalised patients.