Skip to main content

Risk Prediction for Acute Kidney Injury in Acute Medical Admissions in the UK

Blackburn, Anna, Gunda, Smita, Lopez, Berenice, Edwards, James, Spittle, Nick, Preston, Rob, Baines, Richard, Little, Jane, Falayajo, Akin, Mahmoud, Huda, and others. (2018) Risk Prediction for Acute Kidney Injury in Acute Medical Admissions in the UK. QJM: An International Journal of Medicine, 112 (3). pp. 197-205. ISSN 1460-2725. E-ISSN 1460-2393. (doi:10.1093/qjmed/hcy277) (KAR id:70916)

PDF Author's Accepted Manuscript
Language: English
Download (255kB) Preview
[thumbnail of Risk Prediction for Acute Kidney Injury.pdf]
This file may not be suitable for users of assistive technology.
Request an accessible format
PDF Pre-print
Language: English

Restricted to Repository staff only
Contact us about this Publication
[thumbnail of RISK-Final-July2018.pdf]
Official URL


Background Acute Kidney Injury (AKI) is associated with adverse outcomes; identifying patients who are at risk of developing AKI in hospital may lead to targeted prevention. This approach is advocated in national guidelines but is not well studied in acutely unwell medical patients. We therefore aimed to undertake a UK-wide study in acute medical units (AMUs) with the following aims: to define the proportion of acutely unwell medical patients who develop hospital-acquired AKI (hAKI); to determine risk factors associated with the development of hAKI; and to assess the feasibility of using these risk factors to develop an AKI risk prediction score. Methods In September 2016, a prospective multicentre cohort study across 72 UK AMUs was undertaken. Data were collected from all patients who presented over a 24-hour period. Chronic dialysis, community-acquired AKI (cAKI) and those with fewer than two creatinine measurements were subsequently excluded. The primary outcome was the development of h-AKI. Results 2,446 individuals were admitted to the AMUs of the 72 participating centres. 384 patients (16%) sustained AKI of whom 287 (75%) were cAKI and 97 (25%) were hAKI. After exclusions, 1,235 participants remained in whom chronic kidney disease (OR 3.08, 95% CI 1.96-4.83), diuretic prescription (OR 2.33, 95% CI 1.5-3.65), a lower haemoglobin concentration and an elevated serum bilirubin were independently associated with development of hAKI. Multivariable model discrimination was moderate (c-statistic 0.75), and this did not support the development of a robust clinical risk prediction score. Mortality was higher in those with hAKI (adjusted OR 5.22; 95% CI 2.23-12.20). Conclusion AKI in AMUs is common and associated with worse outcomes, with the majority of cases community acquired. The smaller proportion of hAKI cases, only moderate discrimination of prognostic risk factor modelling and the resource implications of widespread application of an AKI clinical risk score across all AMU admissions suggests that this approach is not currently justified. More targeted risk assessment or automated methods of calculating individual risk may be more appropriate alternatives.

Item Type: Article
DOI/Identification number: 10.1093/qjmed/hcy277
Uncontrolled keywords: Acute kidney injury, AKI, clinical prediction score, acute medicine
Subjects: H Social Sciences
R Medicine
Divisions: Divisions > Division for the Study of Law, Society and Social Justice > School of Social Policy, Sociology and Social Research > Centre for Health Services Studies
Depositing User: Christopher Farmer
Date Deposited: 10 Dec 2018 16:38 UTC
Last Modified: 29 Oct 2021 14:36 UTC
Resource URI: (The current URI for this page, for reference purposes)
Farmer, Chris:
  • Depositors only (login required):