The association of duration of AKI with subsequent need of chronic dialysis or death

Principal investigator: 
Wynn Cheung
London School of Hygiene and Tropical Medicine

What is already known about this topic and why is it important?

Acute kidney injury (AKI) is an abrupt impairment of kidney function over a very short period of time. Though once thought as fully reversible, vast literature has shown that AKI episodes are associated with worsened prognosis. Moreover, more severe AKI is associated with higher risk of short-term adverse outcomes such as mortality as well subsequent need of dialysis, the latter of which is an indicator of progression from reversible to irreversible injury as well as chronicity of renal dysfunction. In addition, longer AKI episodes and recurrent AKI are both associated with poorer clinical outcomes, though there is less research into this. Nonetheless, these results are sobering and also imply significant socioeconomic costs. Learning more about AKI-associated outcomes and their predictors is crucial to understanding their pathological mechanisms. Equally importantly, this understanding helps better-inform policy-making decisions such that they are able to provide better infrastructure to support patient care.

How will you carry out your study?

Main statistical analyses will include:

  • Multilevel Poisson regression to quantify association between AKI duration / recurrent AKI with risk of 30-day mortality and chronic dialysis within a certain timeframe respectively
  • Multilevel survival analysis (including semi-parametric Cox proportional hazards model for competing risks) to quantify association between AKI duration / recurrent AKI with time to chronic dialysis
  • How will you decide which patients are included in your study?

All patients (paediatric and adult) will be included 

How many patients do you anticipate including?

488,856 individuals (from UKKR lab AKI dataset)

For how long will you follow up these patients?


What value will UKRR data add to the project?

The lab AKI dataset on nation-wide AKI episodes will provide a large and representative sample to investigate AKI and associated outcomes. 

What new information will your study generate and how will this benefit patients?

This will be the first project which utilises the nation-wide dataset obtained via the AKI alert system in place in most hospitals in England. While vast literature shows that more severe AKI is associated with worse prognosis, there is little research into the independent association between AKI duration and short-term mortality as well as long-term dialysis. This project aims to investigate this association and whether or not it varies by laboratories, seasonality, and age (paediatric/adult).

Unlike previous studies which focus on the association between AKI and short-/long-term adverse events in patients with specific comorbidities (e.g. concurrent cardiovascular diseases), results from this project will be generalisable on a national scale. They will provide insight to guide better patient care and resource allocation to different hospitals.


NICE accredited clinical practice guidelines 

Available here

23rd Annual Report

Analyses about the care provided to patients at UK renal centres.

Read the report


A report on the nationwide collection of AKI warning test scores. 

Read the report