NGAL in Acute Kidney Injury: From Serendipity to Utility

The incidence of acute kidney injury (AKI) has reached epidemic proportions, affecting an estimated 7% of hospitalized patients, in whom it is an independent predictor of mortality and morbidity.1, 2 In the critical care setting, the prevalence of AKI requiring dialysis is about 6%, with a mortality rate exceeding 60%.3 Once established, the treatment of AKI is largely supportive, at an annual cost surpassing $8 billion in the United States alone.4 The diagnosis currently depends on detection of reduced kidney function by the rise in serum creatinine concentration, which is a woefully inadequate measure in the acute setting for a multitude of reasons.5 Ironically, animal studies have identified several interventions that can prevent AKI if instituted early in the disease process, well before the serum creatinine changes.6 The lack of early predictive biomarkers has been the Achilles' heel in the field, crippling our ability to translate these promising findings to human AKI. Fortunately, studying the early stress response of the kidney itself has uncovered new pathophysiologic insights, and, serendipitously, a number of potential biomarkers.7, 8 The bench-to-bedside story of neutrophil gelatinase-associated lipocalin (NGAL), possibly the most promising novel AKI biomarker, is the subject of this editorial and 2 related articles in this issue of the American Journal of Kidney Diseases.9, 10

High-throughput functional genomic studies identified the NGAL (also known as lipocalin 2 gene [LCN2]), as one of the most upregulated transcripts in the kidney very early after acute injury,11, 12, 13 and NGAL was one of the most rapidly induced proteins in the kidney after experimental AKI.14, 15, 16, 17 The biologic relevance of these findings, including the potential role of NGAL as a therapeutic agent, is discussed by Bolignano et al in this issue.9 However, the observation that, in both experimental and human AKI, NGAL in the tubule cells was distributed in a punctate cytoplasmic localization (reminiscent of a secreted protein) also prompted the search for NGAL in the urine. The early appearance of urinary NGAL in animals with AKI, hours to days before other biomarkers,14, 15, 16, 17 has spawned a number of phase 2 translational studies that are establishing NGAL as an early biomarker for human AKI.

In prospective studies of children undergoing elective cardiac surgery, NGAL measurements revealed a greater than 10-fold increase in the urine and plasma within 2 to 6 hours of bypass initiation in those who developed AKI (defined as a 50% increase in serum creatinine) 2 to 4 days after surgery.18, 19, 20 Both urine and plasma NGAL were excellent independent predictors of AKI, with an area under the curve (AUC) of greater than 0.9 for the measurements at the 2- and 6-hour time points. These findings have now been confirmed in prospective studies of adults who developed AKI after cardiac surgery, in whom urinary NGAL was significantly elevated by 1 to 3 hours after the operation.21, 22 The AUC for the prediction of AKI was in the 0.71 to 0.80 range, a somewhat inferior performance perhaps reflective of confounding variables such as old age, preexisting kidney disease, prolonged bypass times, chronic illness, and diabetes.22 In a prospective multicenter study of adults and children undergoing kidney transplantation, urine and plasma NGAL levels in samples collected on the day of transplantation identified those who subsequently developed delayed graft function (which typically occurred 2 to 4 days later), with an AUC of 0.9 for urine NGAL.23, 24, 25 In prospective studies of adults or children receiving radiocontrast, urine and plasma NGAL predicted radiocontrast-induced nephropathy within 2 to 4 hours after radiocontrast administration, with an AUC of 0.91 to 0.92.26, 27, 28, 29, 30 In the intensive care setting, urine and plasma NGAL measurements predict AKI about 2 days prior to the rise in serum creatinine, with high sensitivity and an AUC of 0.68 to 0.78.31, 32, 33 In a recent study of adults in the emergency department setting, a single measurement of urine NGAL at the time of initial presentation predicted AKI with an outstanding AUC of 0.95, and reliably distinguished AKI from prerenal azotemia versus other causes and from chronic kidney disease.34 Thus, NGAL is emerging as a useful biomarker that predicts development of AKI even in heterogeneous groups of patients with multiple comorbid conditions and with unknown timing of initial kidney injury.

Recent studies have also demonstrated the utility of early NGAL measurements for predicting clinical outcomes of AKI. In children undergoing cardiac surgery, early postoperative urine and plasma NGAL levels correlated with AKI severity, length of hospital stay, dialysis requirement, and death.35, 36 In adults undergoing cardiopulmonary bypass, those who subsequently required renal replacement therapy displayed the highest urine NGAL values upon arrival in the intensive care unit.22 NGAL is also emerging as an early biomarker in interventional trials. For example, a reduction in urine NGAL has been employed as an outcome variable in studies demonstrating the improved efficacy of a hydroxyethylstarch preparation over albumin or gelatin in maintaining renal function in elderly cardiac surgery patients.37, 38 Similarly, urine NGAL was attenuated in adult cardiac surgery patients who experienced a lower incidence of AKI after sodium bicarbonate therapy when compared to sodium chloride.39 Furthermore, adults who developed AKI after aprotinin use during cardiac surgery displayed a dramatic rise in urine NGAL in the immediate postoperative period.40 NGAL measurements are currently included in at least 10 ongoing clinical trials formally listed in the ClinicalTrials.gov registry.

Studies described thus far have utilized research-based assays. A major recent advance has been the development of standardized platforms for the rapid clinical measurement of NGAL. For plasma, an NGAL assay using the Triage device (Biosite Incorporated, San Diego, CA) provides quantitative results available in 15 minutes from just one drop of whole blood or plasma. In children tested 2 hours after undergoing cardiac surgery, plasma NGAL measurement by using the Triage device predicted AKI with an AUC of 0.96.35 In addition, a urine NGAL immunoassay has been developed for a standardized clinical platform (ARCHITECT analyzer, Abbott Diagnostics, Abbott Park, IL). This assay provides results within 35 minutes and requires only 150 μL of urine. For pediatric cardiac surgery, measuring urine NGAL by using the ARCHITECT analyzer at a time point 2 hours following the procedure showed an AUC of 0.95 for prediction of AKI.36 Both assays are currently undergoing multicenter validation in adult populations.

The majority of studies reported thus far have involved small numbers of participants. In this issue, Wagener et al have published their findings in a large (N = 426), heterogeneous cohort of adult cardiac surgical patients from a single center.10 Urine for NGAL measurement by research-based enzyme-linked immunosorbent assay was collected preoperatively, immediately at the end of surgery, and at 3, 18, and 24 hours postoperatively. AKI was defined as an increase in serum creatinine from preoperative values of more than 50% or greater than 0.3 mg/dL during the first 48 hours after surgery. On the whole, their results provide additional support for the use of urinary NGAL as an early AKI biomarker. Patients who subsequently developed AKI demonstrated significantly greater urinary NGAL values immediately at the end of and at 3 hours after surgery. These early urinary NGAL values correlated well with cardiopulmonary bypass and aortic cross-clamp times, whereas neither peak serum creatinine nor relative change in serum creatinine correlated with these known indices of intraoperative renal hypoperfusion.

Are we there yet? Have we found a “renal troponin?” Not quite yet. In the Wagener et al article, the accuracy of urinary NGAL to predict AKI was disappointingly low, with an AUC for the 3-hour postoperative value of only 0.603 and for the 18 hour time point of only 0.611. But before we give up on NGAL as a promising AKI biomarker, we should carefully consider the potential reasons for this poor performance. First, as appropriately acknowledged by the authors, even participants who did not develop AKI had a dramatic increase in early urinary NGAL measurements. Could this be related to their definition of AKI, which required an increase in serum creatinine within 48 hours of the surgery? It is quite common for patients to encounter a substantial increase in serum creatinine even after 48 hours following cardiopulmonary bypass18, 19, 20, 22; these individuals could have been misclassified as “non-AKI controls” in their study, with falsely elevated average urine NGAL values. Second, is it possible that a substantial number of their AKI cases merely had a transient prerenal azotemia? This would falsely decrease the average urine NGAL value in the “AKI group,” because prerenal azotemia is associated with normal urine NGAL concentrations.34 Perhaps reflective of this possibility, the peak urinary NGAL values reported by the same authors in their similar, previously published smaller study21 was nearly 6,000 ng/mL on average in the AKI group, about 4-fold higher than that reported in the present publication. Third, could the authors have missed the true postoperative urine NGAL peak in their AKI cases? Studies to date have suggested that urine NGAL peaks at about 6 hours post–cardiopulmonary bypass, both in adults22 and children,36 a time point that was not investigated in the Wagener et al study. Fourth, could the results have been clouded by technical considerations? The authors examined urine samples that were not frozen at −80°C, which may have resulted in a substantial loss of NGAL immunoreactivity. For example, urine NGAL has been shown to degrade by 21% after 24 hours of storage at 4°C when compared to storage at −80°C.41 In addition, the number of freeze-thaw cycles that their samples were subjected to is not mentioned. Finally, urinary NGAL in this setting might signal kidney damage not severe enough to affect serum creatinine, but severe enough to influence a number of important clinical outcomes.35, 36 In the future, it might be more appropriate to define AKI by either a predictive biomarker of kidney damage or a sensitive measure of decrease in kidney function. Many of these factors could have contributed to the low AUC reported in the present study, in comparison to the AUC of 0.74 to 0.8 previously published by the same authors in a smaller study.21

In summary, NGAL as an AKI biomarker appears to have serendipitously but successfully passed through the preclinical, assay development, and initial clinical testing stages of the biomarker development process.42 It has now entered the prospective screening stage, facilitated by the development of commercial tools for the measurement of NGAL on large populations across different laboratories. The results are awaited with anticipation and will likely determine the ultimate clinical utility. But will any single biomarker such as NGAL fully suffice in AKI? In addition to early diagnosis and prediction, biomarkers should discern AKI subtypes, identify etiologies, predict clinical outcomes, allow for risk stratification, and monitor the response to interventions. A sequential panel of validated biomarkers may ultimately be needed to provide us with all the desired information. Other candidates for inclusion are interleukin 18 (IL-18), kidney injury molecule 1 (KIM-1), cystatin C, and liver-type fatty acid binding protein 1 (L-FABP), to name a few.43, 44 The availability of such personalized and predictive information could revolutionize renal and critical care medicine in the not-too-distant future.