HOSPITAL MORTALITY FOLLOWING INTENSIVE CARE ADMISSION OF ADULTS WITH KIDNEY DISEASE 2018-2020: AN AUSTRALIAN DATA LINKAGE STUDY
Dominic Keuskamp1,2, CHRISTOPHER E DAVIES1,2, PHILIP A CLAYTON1,2,3, DAVID V PILCHER4,5,6, SHAILA CHAVAN5, PAUL J SECOMBE5,6,7, SARAH L JONES8,9, BENJAMIN E REDDI2,10, STEPHEN P MCDONALD1,2,3 1Australia & New Zealand Dialysis & Transplant Registry, South Australian Health & Medical Research Institute, Adelaide, SA, Australia2Faculty of Health & Medical Sciences, The University of Adelaide, Adelaide, SA, Australia3Central & Northern Adelaide Renal & Transplantation Services, Royal Adelaide Hospital, Adelaide, SA, Australia4Department of Intensive Care, Alfred Hospital, Melbourne, VIC, Australia5Centre for Resource Outcome & Evaluation, Australian & New Zealand Intensive Care Society, Camberwell, VIC, Australia6Australian & New Zealand Intensive Care Research Centre, School of Public & Preventive Medicine, Melbourne, VIC, Australia7Intensive Care Unit, Alice Springs Hospital, Alice Springs, NT, Australia8Intensive Care Unit, Northern Health, Epping, VIC, Australia9Department of Nephrology, Northern Health, Epping, VIC, Australia10Intensive Care Unit, Royal Adelaide Hospital, Adelaide, SA, Australia
Abstract
Aim: To describe mortality following intensive care unit (ICU) admission of adults with kidney disease, including those with acute kidney injury (AKI), receiving maintenance dialysis or with a functioning kidney transplant.
Background: Variation in mortality following ICU admission by differing levels of kidney disease is not well quantified.
Methods: Data were linked probabilistically between the Australia & New Zealand Dialysis & Transplant Registry, and years 2018-2020 of the Australian & New Zealand Intensive Care Society Adult Patient Database. Risk of in-hospital mortality, adjusted for diagnosis and a modified APACHE III severity-of-disease score, was compared for cohorts of differing kidney function (defined by pre-operative eGFR / AKI / dialysis / transplant) using multivariable logistic regression of medical, emergency surgical & elective surgical admissions. The study population included all eligible first admissions (n= 442,369), excluding those post-kidney transplant.
Results: Unadjusted mortality for medical admissions increased with increasing level of kidney disease from 9.5% (eGFR 60 to 89 ml/min/1.73 m²) to 35.6% (AKI), compared to 6.5% for adults with eGFR ≥ 90 ml/min/1.73 m²; mortality for emergency surgical admissions overall was lower and elective surgical admissions overall was lowest. Adjusted mortality for medical admissions remained elevated for those with kidney disease, especially AKI (OR 4.1 95% CI 3.8-4.3) and less so dialysis (2.6, 2.2-3.0) and transplant (1.9, 1.4-2.4). Mortality risk exceeded that for adults with normal kidney function following emergency surgical admissions (AKI [5.3, 4.6-6.0] and dialysis [4.3, 3.1-5.8] cohorts respectively) and elective surgical admissions ([7.5, 6.0-9.5] and [8.9, 5.9-13.3]).
Conclusions: Adults with severe kidney disease, AKI and those receiving maintenance dialysis experienced more than twice the adjusted mortality following ICU admission than those with normal kidney function.
Biography
Dr Dominic Keuskamp is a Postdoctoral Research Fellow with the Australia & New Zealand Dialysis & Transplant (ANZDATA) Registry at the South Australian Health & Medical Research Institute. His work is focussed on using data from the registry and from linkage to other sources to answer questions related to the predictors and outcomes of kidney failure treated with kidney replacement therapy. Additional work includes retrospective and prospective analyses of dialysis demand to inform health service planning and advocacy.