An acute deterioration of kidney function, currently known as acute kidney injury (AKI), frequently develops in ICU patients and is always bad news. In this overview we will highlight recent findings related to its diagnosis, epidemiology, and extracorporeal treatment.
Epidemiology of AKIThe AKI-EPI study, a worldwide cross-sectional study on the occurrence of acute kidney injury (AKI) in the first week in intensive care unit (ICU), found AKI in 57 % of the 1800 included patients. AKI severity was associated with adverse patient and kidney outcome at hospital discharge [1]. Analysis of a large cohort (32,045) in a single medical center confirmed the high prevalence (74.5 %) and further illustrated that both AKI severity and duration and the criteria used to define it (creatinine, urine output, or both) had significant impact on short-and long-term outcomes [2]. Another large cohort study in 580 patients with out-of-hospital cardiac arrest found severe AKI in 43 % of the patients and the association with mortality but not with neurological outcome [3]. Last, it must be noted that first validation of the Kidney Disease Improving Global Outcomes (KDIGO) criteria was performed in a large pediatric population [4].
Assessment of renal prognosis and of renal functionSeveral studies demonstrated limits of usual renal dysfunction criteria and of biomarkers in assessing renal function or predicting renal prognosis. First, short episodes of oliguria in ICU patients are frequent and do not always predict subsequent AKI by creatinine criteria. Legrand et al. explored whether AKI biomarkers could predict worsening of kidney function (WKF) in 111 critically ill patients with first episode of oliguria. Interestingly, although plasma neutrophil gelatinase-associated lipocalin (NGAL) might improve the prediction of WKF, the biomarker had a similar performance to serum creatinine [5]. Furthermore, the follow-up of the furosemide stress test study found limited performance of biomarkers to predict WKF, which were largely outperformed by the furosemide stress test [6].The gold standard for glomerular filtration rate (GFR) determination is inulin clearance, but this cannot be measured easily. Carlier and co-workers compared inulin clearance with creatinine clearance and several creatinine and/or cystatin C-based GFR equations in 68 critically ill patients with stable kidney function. As a result of tubular secretion of creatinine, measured creatinine clearance resulted in a slight overestimation of GFR while creatinine-based equations had the worst performance with overestimation of the true GFR; the overestimation increased with hospital stay [7]. This confirms previous findings showing increasing differences between discharge estimated GFR (eGFR) and creatinine clearance with increasing ICU stay in both AKI and non-AKI patients. Reduced creatinine excretion resulting from muscle loss was the main determinant of this difference [8]. Since cystatin is not influenced by muscle mass, the cystatin-based equations did not share the hospi...