Background: There are no suitably powered epidemiological studies of 'transient azotaemia' (TA). The objective of this study was to describe the epidemiology of TA and its independent association with hospital mortality. We hypothesized that TA would be associated with an independent increase in the risk of death.
Methods: We retrospectively studied all patients admitted to a university-affiliated hospital in Australia between January 2000 and December 2002. Patients were excluded if they were <15 years old, were on chronic dialysis, had kidney transplant or if their length of hospital stay was <24 hours. We defined TA as rapidly recovering acute kidney injury (AKI) (return to no-AKI risk, injury, failure, loss, end stage (RIFLE) class within 72 hours of onset). We performed descriptive and comparative statistical analysis of data. The primary outcome of the study was the association between TA and hospital mortality in multivariate logistic regression analysis.
Results: Among 20 126 study patients, 3641 (18.1%) had AKI according to the RIFLE criteria and 1600 had AKI, which recovered during their hospital stay. Recovery of AKI occurred most commonly within 1 day after diagnosis (37.7%, n = 603). Furthermore, 1172 patients (73.3%) who recovered from AKI did so within 3 days (TA). After correcting for confounding factors, compared with patients without AKI, patients with TA had a significantly higher odds ratio for hospital mortality (2.26; 95% confidence interval: 1.85-2.76).
Conclusions: Transient azotaemia is common in hospital patients, represents close to a third of all cases of AKI and is independently associated with a significantly higher risk of death.