Characteristics and outcomes of patients with de-novo kidney injury whilst admitted to intensive care

Andonovic, Mark (2022) Characteristics and outcomes of patients with de-novo kidney injury whilst admitted to intensive care. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2022AndonovicPhD.pdf] PDF
Download (13MB)


Kidney injury is a common occurrence amongst patients admitted to the intensive care unit (ICU). Whilst features and short-term outcomes of acute kidney injury (AKI) in ICU have been well documented over the past decade, less research is available regarding longer-term outcomes. A relatively new definition for protracted kidney injury referred to as acute kidney disease (AKD) has been proposed, but minimal data exist as to the characteristics of patients with this condition and their short- and long-term outcomes.

The studies including in this thesis aimed to identify the features of kidney injury suffered whilst admitted to ICU; the short- and long-term survival of patients with kidney injury and the subset of patients who progress to AKD; longterm development of major adverse kidney events and secondary cardiovascular events in these patients; and the features and outcomes of patients with oliguric kidney injury when compared to non-oliguric kidney injury.

This thesis performed retrospective observational cohort studies which identified patients aged 16 or older admitted to the Glasgow Royal Infirmary and Queen Elizabeth University Hospital ICUs in Scotland between 1st July 2015 and 30th June 2018. Patients with known pre-existing established kidney failure (EKF) were identified and classed as their own group. Baseline serum creatinine and subsequent values were used to identify patients with de-novo kidney injury (DNKI) and the remaining patients were classified as having no kidney injury. Patients with DNKI with recovery prior to day 7 were classified as AKI; recovery at day 7 or beyond was classified as AKD. Data extracted from the Scottish Intensive Care Society Audit Group (SICSAG) and Strathclyde Electronic Renal Patient Records (SERPR) databases included patient demographics, in-hospital and long-term mortality, proportion of major adverse kidney events (MAKEs), and cardiovascular events. Data on 24-hour urine output values were extracted from the CarevueTM database to identify and compare oliguric and non-oliguric kidney injury. Multivariable logistic regression was used to identify risk factors for AKD and reported in terms of odds ratios (ORs) and 95% confidence intervals (95% CIs). A Cox proportional hazards model was used to identify factors associated with long-term outcomes and reported as hazard ratios (HRs) and 95% CIs.

Two in every five patients admitted to ICU during the study period went on to suffer from a kidney injury during their admission (40.4%). Approximately one in four of patients who survived to day seven after their injury, progressed to AKD (24.9%). Kidney injury was more common in older, comorbid, male patients admitted from medical specialties with lower baseline estimated glomerular filtration rate (eGFR) admitted as a result of sepsis; progression to AKD was significantly associated with male sex, admission due to sepsis and a lower baseline eGFR. In-hospital mortality was significantly higher in the DNKI group compared to the no injury group (35.9% vs 11.4%); this was also the case for AKD patients compared with AKI patients who survived to day 7 following the initial injury (26.1% vs 11.6%).

In patients who survived to hospital discharge, mortality over the four- and halfyear follow-up period showed a significant reduction in survival in the de-novo injury group when compared with the group without kidney injury with an independently associated 16% increased risk of dying. No significant long-term survival difference was associated with progression to AKD. Development of denovo kidney injury and progression to AKD were both significantly associated with a faster decline in eGFR over time as well as development of MAKEs over the total follow up period (OR = 2.28 for DNKI and OR = 1.25 for AKD).

Presence of DNKI whilst in ICU was significantly associated with a biochemical myocardial injury (HR=1.46); however, progression to AKD did not show any significant association. Neither presence of AKI nor prolonged length of injury had any statistically significant effect on future coronary artery interventions. Whilst DNKI did not show a significant association for future cerebrovascular events, a sub-group analysis on DNKI patients showed progression to AKD was significantly associated with future cerebrovascular events (OR=2.34).

On analysis of DNKI patients with data available on 24-hourly urine output during their admission, 46.4% suffered from oliguric injury. Development of oliguric kidney injury was more commonly seen in older patients, patient admitted from medical specialties, patients with a lower baseline eGFR, and patients admitted due to sepsis. In-hospital mortality was significantly higher in patients with oliguric kidney injury compared to patients with non-oliguric injury (41.9% vs 31.5%). In patients to survive to hospital discharge, oliguric injury was independently associated with increased mortality at 18 months compared with non-oliguric injury, but no significant between group difference was seen in future development of MAKEs. No differences in outcomes were observed between patients with point oliguria compared with persisting oliguria.

The work contained within this thesis thoroughly details the available literature on kidney injury within intensive care, summarises the features of kidney injury in a large cohort of patients admitted to ICU and demonstrates the significant increased risk it confers on in-hospital mortality, long-term survival, future adverse kidney events and future myocardial injury. It also characterises patients with AKD and describes the independently increased risk associated with it and in-hospital mortality and future MAKEs. This information can help clinicians to stratify patients at risk of future adverse outcomes. Future research could be expanded to include a much larger cohort of patients over a wider geographical region thus ensuring these increased risk profiles are not confined to patients solely from the West of Scotland. By helping to identify these high-risk patients, future work could aim to further detail AKD patients at routine ICU follow-up clinics by taking routine blood samples and urinalysis. The data obtained could be used to detect those patients with ongoing kidney dysfunction who may benefit from follow-up with expert nephrologists who could potentially implement therapy tailored to individual patients which may prevent progression to future adverse outcomes.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: R Medicine > R Medicine (General)
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing
Supervisor's Name: Mark, Professor Patrick and Puxty, Dr. Kathryn
Date of Award: 2022
Depositing User: Theses Team
Unique ID: glathesis:2022-83220
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 21 Oct 2022 13:55
Last Modified: 21 Oct 2022 13:58
Thesis DOI: 10.5525/gla.thesis.83220
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year