Context: A broad range of drugs and chemicals are capable of evoking acute kidney injury, which is conventionally determined by rising serum creatinine concentrations. However there are important limitations to this approach, and there has been interest in alternative biomarkers that might provide a more sensitive and rapid means of detecting acute kidney injury. Most of the available clinical data have thus far been ascertained in patients requiring critical care or with acute sepsis. However, if a sensitive indicator of acute kidney injury were developed, then this could provide a significantly improved means of detecting the effects of acute drug or toxin exposure.
Objective: To review the available data concerning potential biomarkers of acute kidney injury and to assess their relative strengths and weaknesses in comparison to existing methods based on serum creatinine concentrations. A large number of possible biomarkers have been proposed. Evidence for individual biomarkers is reviewed with a particular emphasis on those with potential application in clinical toxicology. Where available, comparative data are presented.
Methods: There were 236 papers identified using Medline, Embase, and Google Scholar databases, of which 52 were considered directly relevant. CREATININE: Creatinine is subject to glomerular filtration and, to a lesser extent tubular secretion. Serum concentrations are an insensitive marker of acute kidney injury, and the speed of an increase from baseline depends on the magnitude of the acute injury and pre-existing kidney functional reserve. A wide range of inter-individual concentrations means that single time-point determinations are difficult to interpret, and acute kidney injury may not manifest as a detectable increase in serum creatinine concentrations until at least 24-48 h after the primary insult. KIDNEY ENZYMES: Enzymes are often localised to specific anatomical locations, and acute injury may cause a detectable increase in urinary activity due to up-regulated activity or leakage due to cell membrane disruption. Key examples include gamma-glutamyl transpeptidase (GGT), glutathione-S-transferase (GST), and N-acetyl-glucosaminidase (NAG), which are found predominantly in the proximal tubule and urinary enzyme activity increases after acute exposure to heavy metals and other nephrotoxins. NEUTROPHIL GELATINASE-ASSOCIATED LIPOCALIN: Neutrophil gelatinase-associated lipocalin (NGAL) is expressed by renal tubular epithelium, and a rise in urinary concentrations may provide an indicator of acute renal injury caused by any one of a broad range of provoking factors that is detectable before a rise in serum creatinine concentrations. CYSTATIN C: Serum and urinary cystatin C concentrations are closely related to kidney function and, for example, in acute tubular necrosis allow better prediction of the need for renal replacement therapy than serum creatinine concentrations. KIDNEY INJURY MOLECULE 1: Kidney injury molecule 1 (KIM-1) is expressed in the proximal tubule in the setting of acute ischaemia. For example, urinary KIM-1 concentrations becomes detectable within 24 h of acute tubular necrosis. Urinary KIM-1 expression may be detected after exposure to a variety of nephrotoxic agents, even when serum creatinine concentrations do not increase, and this has been accepted by regulatory authorities as a sensitive biomarker of acute kidney injury during early drug development.
Conclusions: Novel biomarkers appear capable of offering a more sensitive means of detecting acute kidney injury than existing approaches. Certain of these allow discrimination between the various mechanisms and anatomical site of acute injury. Ultimately, clinical assessment might incorporate a panel of different biomarkers, each informing on the integrated aspects of glomerular, tubular and interstitial function. Presence of biomarkers may in some cases detect mild or transient renal dysfunction that is presently undetected, and the clinical relevance needs further exploration. Whilst many potentially useful biomarkers have been proposed, comparatively few clinical data exist to support their validity in routine practice. Further prospective clinical studies are required to examine the validity of biomarkers after acute drug or toxin exposure, and to establish whether they might offer improved clinical outcomes in the setting of clinical toxicology.