Acute kidney injury (AKI) is a common and serious condition, the diagnosis of which depends on serum creatinine measurements. Unfortunately, creatinine is a delayed and unreliable biomarker of AKI. The lack of reliable early biomarkers has crippled our ability to translate promising experimental therapies to human AKI. Fortunately, understanding the early stress response of the kidney to acute injuries has serendipitously revealed a number of potential non-invasive biomarkers. The discovery, translation and validation of neutrophil gelatinase-associated lipocalin (NGAL), the most widely studied novel AKI biomarker, is reviewed. NGAL is emerging as an excellent stand-alone troponin-like biomarker in the plasma and urine, as a diagnostic as well as prognostic tool in several common clinical scenarios. The current status of other promising AKI biomarkers, including kidney injury molecule-1 (KIM-1), liver-type fatty acid binding protein (L-FABP), and interleukin-18 (IL-18) is also reviewed. The potential role of biomarker combinations, the use of novel biomarkers independent of creatinine, and the limitations of AKI biomarker studies are also explored.
biomarker, acute kidney injury, neutrophil gelatinase-associated lipocalin, kidney injury molecule-1, interleukin-18, liver-type fatty acid binding protein
Biomarkers of Acute Kidney Injury—an Un-Met Need
The incidence of acute kidney injury (AKI) is rising globally, and so are the associated morbidity and mortality rates. AKI afflicts 5–7% of all hospitalized patients. In critically ill patients, the overall prevalence of severe AKI requiring renal replacement therapy (RRT) is about 6%, with a mortality rate of 60%. Mortality and morbidity from AKI has not substantially improved in the past few decades despite technological advances in supportive care. AKI is largely asymptomatic, and establishing the diagnosis currently hinges on functional biomarkers such as serial serum creatinine measurements. Unfortunately, serum creatinine is a delayed and unreliable indicator of AKI for a variety of reasons. First, normal serum creatinine is influenced by several non-renal factors such as age, gender, muscle mass, muscle metabolism, medications, hydration status, nutrition status, and tubular secretion. Second, a number of acute and chronic kidney conditions can exist with no increase in serum creatinine due to the concept of renal reserve—it is estimated that greater than 50% of kidney function must be lost before serum creatinine rises. Third, serum creatinine concentrations do not reflect the true decrease in glomerular filtration rate in the acute setting, since several hours to days must elapse before a new equilibrium between the presumably steady state of creatinine production and the decreased excretion of creatinine is established. Fourth, serum creatinine production is diminished in critical illnesses such as sepsis, and measured serum creatinine is often further reduced by hemodilution resulting from standard goal-directed fluid therapies. Fifth, an increase in serum creatinine represents a late indication of a functional change in glomerular filtration rate, which lags behind important structural changes that occur in the kidney during the early damage stage of AKI. The delay in AKI diagnosis imposed by our dependence on serum creatinine changes is a problem, since animal studies have identified several interventions that can prevent and/or treat AKI if instituted early in the disease course, well before the serum creatinine begins to rise. The lack of early biomarkers has hitherto hampered our ability to translate these promising therapies to human AKI. Also lacking are reliable methods to assess efficacy of preventive or therapeutic interventions, and early predictive biomarkers of drug toxicity.
Desirable Characteristics of aki Biomarkers
First, with respect to assay characteristics, AKI biomarkers should be non-invasive and easy to perform at the bedside or in a standard clinical laboratory, using easily accessible samples such as blood or urine, with quick turn-around times. The majority of AKI biomarkers described thus far have been measured in the urine. Urinary diagnostics have several advantages, including the non-invasive nature of sample collection, the reduced number of interfering proteins, and the potential for the development of patient self-testing kits. However, disadvantages also exist, including the lack of sample from patients with severe oliguria, and potential changes in urinary biomarker concentration induced by hydration status and diuretic therapy. Plasma-based diagnostics have revolutionized many facets of medicine, as exemplified by the use of troponins for the early diagnosis of acute myocardial infarction. On the other hand, plasma biomarkers may be confounded by extra-renal sources as well as by subclinical changes in renal elimination. Thus, in the case of AKI, it is ideal to develop both urinary and plasma biomarkers.
Second, with respect to diagnostic properties, AKI biomarkers should be sensitive to facilitate early detection, with a wide dynamic range that allow for risk stratification. They should also be highly specific for AKI, enable the identification of AKI sub-types and differentiate AKI from chronic kidney disease (CKD). Ideally, biomarkers are also needed to identify the primary location of injury (proximal tubule, distal tubule, interstitium, or vasculature), and discern AKI etiologies (ischemia, toxins, sepsis, or a combination).
Third, with respect to prognostic abilities, AKI biomarkers should allow for risk stratification (duration and severity of AKI), prediction of hard clinical outcomes (need for renal replacement therapy, length of hospital stay, mortality) and monitoring the response to AKI interventions. Biomarkers associated with clear biologic plausibility and known pathophysiologic mechanisms in AKI are most likely to satisfy the desired diagnostic and prognostic characteristics.
Given the limitations of serum creatinine, the search for improved biomarkers of AKI is of intense contemporary interest. During the past decade, an improved understanding of the early pathophysiologic response of the kidney to stress has uncovered a number of genes and proteins that are rapidly induced in the kidney. They have been implicated in the regulation of novel pathways and mechanisms that modulate the kidney injury. Serendipitously, some of these kidney proteins are also detected in the urine and/or plasma, and are emerging as early non-invasive biomarkers of AKI and its clinical outcomes. Table 75.1 lists the desirable characteristics of AKI biomarkers in general, and illustrates the current status of the four most promising novel AKI biomarkers whose bench-to-bedside translation is chronicled in this chapter. Table 75.2 summarizes the biological characteristics of these and other proposed AKI biomarkers. Since NGAL represents the most extensively studied of the novel biomarkers, it will be the primary focus of this chapter.
|Noninvasive (measured in urine or blood)||Yes||Yes||Yes||Yes|
|Rapid, standardized clinical platforms available||Yes||Yes||No||No|
|Sensitive to establish an early diagnosis of AKI||Yes||Yes||Yes||Yes|
|Results available while damage is limitable||Yes||Yes||Yes||Yes|
|High gradient to allow severity prediction||Yes||Yes||Yes||Yes|
|Specific to intrinsic AKI (versus pre-renal AKI)||Yes||Unknown||Unknown||Unknown|
|Discerns AKI from chronic kidney disease||No||No||No||No|
|Predicts hard clinical outcomes||Yes||Yes||Yes||Yes|
|Predicts response to therapies||Yes||Unknown||Unknown||Unknown|
|Associated with a known mechanism||Yes||Yes||Yes||Yes|
|NGAL||Urine||Distal tubule, collecting duct||Regulates iron trafficking, promotes tubule cell survival and proliferation, limits tubule cell apoptosis|
|NGAL||Blood||Liver, lung, neutrophils||Acute phase reactant, marker of organ cross-talk following acute kidney injury|
|KIM-1||Urine||Proximal tubule||Promotes epithelial regeneration, regulates tubule cell apoptosis|
|IL-18||Urine||Proximal tubule||Initiates and promotes tubule cell apoptosis and necrosis|
|L-FABP||Urine||Proximal tubule||Endogenous antioxidant, suppresses tubulointerstitial damage|
|NAG||Urine||Proximal tubule||Marker of proximal tubule lysosomal enzyme release as a result of damage to proximal tubule|
|β2-MG||Urine||Systemic and Proximal tubule||Marker of altered glomerular permeability and/or decreased proximal tubular reabsorption due to damage|
|Albumin||Urine||Systemic and Proximal Tubule||Marker of altered glomerular permeability and/or decreased proximal tubular reabsorption due to damage|
Neutrophil Gelatinase-Associated Lipocalin (NGAL) as an AKI Biomarker
NGAL Physiology and Pathophysiology
Human NGAL was originally identified as a novel protein isolated from secondary granules of human neutrophils, and subsequently shown to be a 25-kDa protein covalently bound to neutrophil gelatinase. Mature peripheral neutrophils lack NGAL mRNA expression, and NGAL protein is synthesized at the early myelocyte stage of granulopoiesis during formation of secondary granules. NGAL mRNA is normally expressed in a variety of adult human tissues, including bone marrow, prostate, salivary gland, stomach, colon, trachea, lung, liver, and kidney. Several of these tissues are prone to exposure to microorganisms, and constitutively express NGAL protein at low levels. The promoter region of the NGAL gene contains binding sites for a number of transcription factors, including NF-κB. This could explain the constitutive as well as inducible expression of NGAL in several of the non-hematopoietic tissues. Like other lipocalins, NGAL forms a barrel-shaped tertiary structure with a hydrophobic calyx that binds small lipophilic molecules. The major ligands for NGAL are siderophores, small iron-binding molecules. Teleologically, NGAL comprises a critical component of innate immunity to bacterial infection. Siderophores are synthesized by bacteria to scavenge iron from the surroundings, and use specific transporters to recover the siderophore-iron complex, ensuring their iron supply. The siderophore-chelating property of NGAL therefore renders it as a bacteriostatic agent. Experimental evidence for this role is derived from mice genetically modified to lack the NGAL gene, which renders them more susceptible to Gram-negative bacterial infections and death from sepsis.
On the other hand, siderophores produced by eukaryotes participate in NGAL-mediated iron shuttling that is critical to various cellular responses such as proliferation and differentiation. This property provides a molecular mechanism for the documented role of NGAL in enhancing the epithelial phenotype. During kidney development, NGAL promotes epithelial differentiation of the mesenchymal progenitors, leading to the generation of glomeruli, proximal tubules, Henle’s loop, and distal tubules. However, NGAL expression is also markedly induced in injured epithelial cells, including the kidney, colon, liver and lung. This is likely mediated via NF-κB, which is known to be rapidly activated in epithelial cells after acute injuries, and plays a central role in controlling cell survival and proliferation. In the context of an injured mature organ such as the kidney, the biological role of NGAL induction is one of marked preservation of function, attenuation of apoptosis, and an enhanced proliferative response. This protective effect is dependent on the chelation of toxic iron from extracellular environments, and the regulated delivery of siderophore and iron to intracellular sites. Not surprisingly, gene expression studies in AKI have demonstrated a rapid and massive upregulation of NGAL mRNA in the distal nephron segments—specifically in the thick ascending limb of Henle’s loop and the collecting ducts. The resultant synthesis of NGAL protein in the distal nephron and secretion into the urine comprises the major fraction of urinary NGAL. Although plasma NGAL is freely filtered by the glomerulus, it is largely reabsorbed in the proximal tubules. Thus, any urinary excretion of NGAL is likely only when a kidney disease precludes proximal tubular NGAL reabsorption, and/or induces distal tubular de novo NGAL synthesis. With respect to plasma NGAL, the kidney itself does not appear to be a major source. NGAL protein released into the circulation from distant organs such as the liver and lung constitute a distinct systemic pool. Additional contributions to the systemic pool may derive from activated neutrophils, macrophages, and other immune cells. Furthermore, any decrease in GFR would decrease the renal clearance of NGAL, with subsequent accumulation in the systemic circulation in patients with CKD.
Preclinical transcriptome profiling studies identified Ngal (also known as lipocalin 2 or lcn2 ) to be one of the most upregulated genes in the kidney very early after acute injury in animal models. Downstream proteomic analyses also revealed NGAL to be one of the most highly induced proteins in the kidney after ischemic or nephrotoxic AKI in animal models. The serendipitous finding that NGAL protein was easily detected in the urine soon after AKI in animal studies has inspired a number of translational human studies, and NGAL has emerged as an excellent biomarker in the urine and plasma for early diagnosis, therapeutic monitoring, and prediction of prognosis in common clinical AKI scenarios. The deployment of standardized clinical platforms for the rapid and accurate measurement of NGAL in urine and plasma has further facilitated the widespread use and validation of NGAL as a biomarker.
NGAL for the Prediction of AKI and its Severity
In a cross-sectional study of adults with established AKI (doubling of serum creatinine) from varying etiologies, a marked increase in urine and serum NGAL was documented by Western blotting when compared to normal controls. Urine and serum NGAL levels correlated with serum creatinine, and kidney biopsies in subjects with AKI showed intense accumulation of immunoreactive NGAL in cortical tubules, confirming NGAL as a sensitive index of established AKI in humans. An explosion of subsequent studies has now implicated NGAL as an early diagnostic biomarker for AKI in several common clinical situations.
Operations involving cardiopulmonary bypass comprise the most frequent major surgical procedure performed in hospitals worldwide. Even a minor degree of post-operative AKI as manifest by only a 0.2–0.3 mg/dl rise in serum creatinine from baseline (which occurs in up to 30% of cardiac surgeries) is associated with a significant increase in mortality and other adverse outcomes. The pathogenesis of cardiac surgery-associated AKI includes ischemia-reperfusion injury, exogenous toxins (contrast media, non-steroidal anti-inflammatory drugs), endogenous toxins (iron released from hemolysis), and inflammation and oxidative stress (from contact with bypass circuit, surgical trauma, and intra-renal inflammatory responses). The predictive value of NGAL has been most extensively studied in this setting ( Table 75.3 ). In single center prospective studies of children who underwent elective cardiac surgery, AKI (defined as a 50% increase in serum creatinine) occurred one to three days after surgery. In contrast, NGAL measurements by ELISA revealed a 10-fold or more increase in the urine and plasma, within two to six hours of the surgery in those who subsequently developed AKI. Both urine and plasma NGAL were excellent independent predictors of AKI, with an area under the receiver-operating characteristic curve (AUC-ROC) of >0.9 for the two to six hour urine and plasma NGAL measurements. A recent prospective multicenter study of children undergoing cardiac surgery has confirmed the early peak (within six hours of initiating cardiopulmonary bypass) of urine and plasma NGAL associated with higher odds of developing AKI, but the diagnostic accuracy by AUC-ROC analysis was lower (0.71). These findings have also been confirmed in several prospective single center studies of adults who developed AKI after cardiac surgery, in whom urinary and/or plasma NGAL was significantly elevated by one to six hours after the operation. The AUC-ROCs for the prediction of AKI have ranged widely from 0.61 to 0.96. A prospective multicenter study of adults undergoing cardiac surgery has confirmed the early peak of urine and plasma NGAL associated with higher odds of developing AKI, but the diagnostic accuracy by AUC-ROC analysis was only 0.67 to 0.70. The somewhat inferior performance in adult populations may be reflective of confounding variables such as older age groups, pre-existing kidney disease, prolonged bypass times, chronic illness, and diabetes. The predictive performance of NGAL also depends on the definition of AKI employed, as well as on the severity of AKI. For example, the predictive value of plasma NGAL post cardiac surgery was higher for more severe AKI (increase in serum creatinine >50%; mean AUC-ROC 0.79) compared to less severe AKI (increase in serum creatinine >25%; mean AUC-ROC 0.65). Similarly, the discriminatory ability of NGAL for AKI increased with increasing severity as classified by RIFLE criteria. Thus, the AUC-ROC improved progressively for discrimination of R (0.72), I (0.79) and F (0.80) category of AKI. Furthermore, the predictive power of urinary NGAL for AKI after cardiac surgery varied with baseline renal function, with optimal discriminatory performance in patients with normal preoperative renal function. Despite these numerous potential variables, a meta-analysis of published studies in all patients after cardiac surgery revealed an overall AUC-ROC of 0.78 for prediction of AKI, when NGAL was measured within 6 hours of initiation of cardiopulmonary bypass and AKI was defined as a >50% increase in serum creatinine. A current analysis of 25 published studies strongly supports the use of NGAL for AKI prediction after cardiac surgery, with an overall sensitivity of 71%, specificity of 78%, and an average AUC of 0.80 ( Table 75.3 ).
|Reference||Patients||AKI Events||Source||AUC||Sensitivity||Specificity||AKI Definition|
|71||20||Urine||0.99||100||98||RIFLE R or greater|
|81||16||Urine||0.80||73||78||RIFLE R or greater|
|196||99||Urine||0.95||82||90||RIFLE R or greater|
|72||34||Urine||0.71||67||58||RIFLE R or greater|
|426||85||Urine||0.61||39||78||RIFLE R or greater|
|50||9||Urine||0.96||90||78||>0.5 mg/dl Creatinine rise|
|90||36||Urine||0.65||71||39||>0.3 mg/dl Creatinine rise|
|103||13||Urine||0.50||67||11||RIFLE R or greater|
|30||15||Urine||0.85||84||80||RIFLE R or greater|
|374||112||Urine||0.92||85||86||RIFLE R or greater|
|220||60||Urine||0.90||88||83||RIFLE R or greater|
|311||53||Urine||0.71||42||85||Doubling of Creatinine|
|1219||60||Urine||0.67||46||81||Doubling of Creatinine|
|71||20||Plasma||0.90||50||100||RIFLE R or greater|
|120||45||Plasma||0.96||84||94||RIFLE R or greater|
|100||23||Plasma||0.80||79||78||RIFLE R or greater|
|50||9||Plasma||0.80||90||78||>0.5 mg/dl Creatinine rise|
|30||8||Plasma||0.98||100||91||RIFLE R or greater|
|879||75||Plasma||0.64||39||82||RIFLE R or greater|
|374||112||Plasma||0.94||90||88||RIFLE R or greater|
|311||53||Plasma||0.56||27||81||Doubling of Creatinine|
|1219||60||Plasma||0.70||50||82||Doubling of Creatinine|
AKI is a frequent complication in critically ill patients, and results in a 40–60% mortality rate. This patient population is extremely heterogeneous, and the etiology and timing of AKI is often unclear. Etiologies include sepsis, nephrotoxins, hypotension, kidney ischemia, mechanical ventilation, and multi-organ disease. Even in such heterogeneous settings, urine and plasma NGAL measurements have been shown to represent early biomarkers of AKI ( Table 75.4 ). Initial studies in the pediatric intensive care setting demonstrated that NGAL predicted AKI about two days prior to the rise in serum creatinine, with high sensitivity and AUC-ROCs of 0.68–0.78. Several studies have also examined plasma and urine NGAL levels in critically ill adult populations. Urine NGAL obtained on admission predicted subsequent AKI in multi-trauma patients with an outstanding AUC-ROC of 0.98. However, in more mixed populations of all critical care admissions, the performance of urine NGAL on admission was more variable, with an AUC-ROC ranging from 0.71 to 0.89 ( Table 75.4 ). In studies of adult intensive care patients, plasma NGAL concentrations on admission constituted a very good to outstanding biomarker for development of AKI within the next two days, with AUC-ROC ranges of 0.79–0.92. In the closely related setting of the emergency department, where again the causes of AKI are myriad and timing of the initial insult uncertain, a single measurement of urine NGAL at the time of initial presentation predicted AKI with an outstanding AUC-ROC of 0.95 in a single center study of adult patients. This has now been confirmed in children, as well as in a large multicenter prospective study of adults presenting to the emergency department, in which urine NGAL was predictive of subsequent AKI with an excellent AUC of 0.81. However, it should be noted that patients with septic AKI display the highest concentrations of both plasma and urine NGAL when compared to those with non-septic AKI, a confounding factor that may add to the heterogeneity of the results in the critical care setting. In addition, the predictive power of urinary NGAL for AKI in the critically ill was optimal in patients with normal preoperative renal function. A meta-analysis revealed an overall AUC-ROC of 0.73 for prediction of AKI, when NGAL was measured within six hours of clinical contact with critically ill subjects and AKI was defined as a >50% increase in serum creatinine. A current analysis of 18 published studies strongly supports the use of NGAL for AKI prediction in the critically ill population, with an overall sensitivity of 77%, specificity of 81%, and an average AUC of 0.83 ( Table 75.4 ).
|Reference||Patients||AKI Events||Source||AUC||Sensitivity||Specificity||AKI Definition|
|140||106||Urine||0.78||54||97||RIFLE R or greater|
|204||102||Urine||0.89||80||96||RIFLE R or greater|
|635||30||Urine||0.95||90||100||RIFLE R or greater|
|31||11||Urine||0.98||91||95||RIFLE R or greater|
|44||18||Urine||0.86||71||100||RIFLE R or greater|
|632||171||Urine||0.88||89||70||RIFLE R or greater|
|529||147||Urine||0.66||40||80||RIFLE R or greater|
|252||18||Urine||0.80||NR||NR||RIFLE I or greater|
|145||75||Urine||0.87||88||89||RIFLE R or greater|
|1635||96||Urine||0.81||68||81||RIFLE R or greater|
|88||42||Plasma||0.92||82||97||RIFLE R or greater|
|45||24||Plasma||0.79||68||82||RIFLE R or greater|
|307||133||Plasma||0.78||73||81||RIFLE R or greater|
|661||24||Plasma||0.82||96||51||Creatinine rise>0.5 mg/dl|
|44||18||Plasma||0.85||83||86||RIFLE R or greater|
|632||171||Plasma||0.86||82||70||RIFLE R or greater|