Investigations on the performance of NGAL as an early marker of renal injury have certainly not stood still since my short introductory article in CLI [1]. Papers on NGAL are appearing at a rate of over 2 a week, and the main emphasis is indeed on its renal role. This is not to say that other aspects of this fascinating molecule are being ignored: its pathophysiological and marker role in adenocarcinomas are receiving attention, as are its roles in inflammation, atheroma and thrombi, not to mention what may be its main function: as a binder and transporter of siderophore iron.

NGAL as a marker of renal injury

Several studies have confirmed and consolidated the usefulness of NGAL as an early marker of renal injury, and more are under way.

Cardiac procedures

Pediatric cardiopulmonary bypass operations

The first important study [2] was in children undergoing cardiopulmonary bypass and showed that urinary NGAL concentrations over 50 ng/ml 2 hours after surgery were highly predictive of a subsequent increase in serum creatinine of 50% or more over basal values. This was a very "clean" study, as these were children undergoing elective cardiac surgery and were as healthy as their cardiac abnormality allowed.

Adult cardiopulmonary bypass operations

A similarly conceived study in adult cardiac surgery patients has recently appeared [3]. This showed a marked rise in urinary NGAL even in the first postoperative sample taken, the interesting finding being that a notable rise was also seen in patients who did not subsequently develop a serum creatinine rise of 50% or more. However, in those patients who did, the urinary NGAL continued to rise, peaking at 3 hours postoperatively, while it fell in the rest of the patients. The urine levels observed were such that any cutoff values diagnostic of subsequent acute renal dysfunction (ARD) at 1 hour or 3 hours after surgery would be many times greater than that observed in the pediatric cases [2].

Many adult patients coming to cardiopulmonary bypass surgery will have accumulated a variety of other disorders, some of them, such as inflammatory disorders, associated with moderate elevations of NGAL. Furthermore, NGAL is also expressed in arterial endothelium and smooth muscle associated with atherosclerotic plaques as well as in macrophages within the plaques [4], so it is possible that some NGAL may be released from these sources during cardiac surgery and may be rapidly excreted by the kidneys when normal circulation is restored. However, it is also possible that most of the "immediate" release of NGAL into the urine is of renal origin. In that case, it could be postulated that the NGAL response of the kidney is so rapid and sensitive that even a minor ischemic renal injury is easily detected, even though it will quickly recover without giving rise to functional impairment leading to a 50% rise in serum creatinine.

A potential cause of renal injury in these operations is the use of large doses of aprotinin as an antifibrinolytic agent to diminish blood loss. It will be interesting to assess to what degree the use of aprotinin correlates with NGAL responses and subsequent ARD.

Percutaneous coronary interventions

NGAL as a marker of renal injury has also been studied in patients subjected to percutaneous coronary interventions with coronary angiography (PCI) [5, 6]. Here the chief renal risk is seen as that due to the radiographic contrast agents used. A significant (but modest) rise in serum NGAL was seen in the first sample at 2 hours after PCI, peaking at 4 hours, while the urine rise in NGAL peaked at 8 hours. Multivariate analysis was performed with serum creatinine, which did not rise significantly during the 48 hours of study, and with serum cystatin C, which peaked at 24 hours. None of the patients developed ARD. This is an interesting study of potential subclinical renal injury from the contrast agent, in which NGAL was clearly the earliest responding marker. However, if the rise in NGAL was really due to nephrotoxic renal injury, it is strange that the serum response anticipated the urinary response. It should be noted that both the mean serum and the mean urinary NGAL were elevated above normal basal levels. NGAL is present in the arteries and atheromatous plaques being visualized in this study, and raised plasma NGAL has been observed in patients with risk factors for atherosclerosis [7]. It is therefore also possible that the early rise in serum NGAL may have been due to a direct effect of the procedure on the cannulated arteries.

Rapidity of the NGAL response

The fact that the postoperative sample in [3] was taken at various times after the start of bypass raises the question of just how rapid the renal NGAL response to injury is. In the published clinical studies, the first sample has been taken either 1 hour or 2 hours after the start or conclusion of the potentially injurious procedure. The fact that very substantial rises are seen after 1 hour suggests that significant rises might also be observed after an even shorter time - 30 minutes? 15 minutes? This question remains to be formally addressed. However, if the kidney is capable of mounting an NGAL response in very much under an hour, it casts doubt on whether the initiation of the response can be entirely due to upregulation of transcription and translation, and would imply that pre-synthesized NGAL is being released by a secretory mechanism. While protein expression of NGAL by the kidney is modest in its normal state, the NGAL that appears in tubule cells after an injury (cisplatin nephrotoxicity) shows a punctate appearance in light-microscopic immunohistochemistry [8], strongly suggestive of localization to specific secretory granules. This is also in line with the presence of a typical signal peptide in the NGAL precursor. These questions and many other details of the secretion of NGAL from the injured tubule cell require elucidation.

Renal transplantation

NGAL determinations will probably also have a part to play in renal transplantation. The correlation of urinary and serum NGAL values with subsequent graft function has been studied in 30 brain-dead kidney donors [9]. A very clear predictive value of the urinary NGAL value was found, such that values below 65 ng/ml were associated with immediate function of both grafts, values from 80-150 ng/ml were associated with immediate function of only one graft, and values above 150 ng/ml with delayed or primary non-function of both grafts. Serum values showed a similar trend. Early measurement of urinary NGAL in the recipient also has predictive value, high levels predicting delayed graft function [10]. Graft biopsy also shows a strong correlation between NGAL immunostaining intensity and cold ischemia time, peak postoperative serum creatinine and delayed graft function [11]. The possible value of NGAL measurements in predicting acute rejection episodes requires further study.

Critically ill patients

Unselected patients

A prime concern in assessing the value of NGAL as a marker of renal injury is that it should also function in patients with a variety of serious concomitant conditions that not only put the patients at high risk of renal injury but also lead to independent elevations of NGAL from non-renal sources. Such conditions include sepsis and serious local infections, in which NGAL is released from neutrophils, serious chest infections, in which there may be further NGAL release from the airway epithelium, and a variety of adenocarcinomas which also express NGAL. In studies of "unselected" critically ill patients, it should always be borne in mind that the composition of the patient population in fact depends on the referral and admissions policy of the health authority in question, so that statistical conclusions from a single such study may not apply in detail to patients at other centers. In one study in which the author participated, both urinary and serum NGAL values were found to have a very good predictive value for the development of ARD, and highly elevated urinary values were strongly indicative of a requirement for hemodialysis or continuous renal replacement therapy [12]. To illustrate the "interference" from NGAL presumed to be extra-renal in origin, the median of the maximal urinary NGAL values in patients with sepsis who did not develop ARD was 111 ng/ml (range 9-2540), while corresponding figures for those who developed ARD were 2670 ng/ml (162-25,300). The corresponding figures for patients with cancer were 111 ng/ml (29-1710) without ARD, and 2210 ng/ml (269-20,900) with ARD. The cutoff values required for the optimal exclusion of NGAL elevations not associated with ARD were in the region of 400 ng/ml for both urine and plasma.

Special cases

Urinary NGAL has also been studied in children with diarrhea-associated hemolytic-uremic syndrome [13]. Here a cutoff value of 200 ng/ml was found, so that children with lesser values within the first 5 days of hospitalization were much less likely to require dialysis. This provides a further illustration of the fact that cutoffs for the optimal diagnostic or prognostic use of NGAL as a marker of kidney damage are very dependent on the patient population under study.

Chronic kidney disease

While the chief diagnostic role of NGAL is seen as a marker for acute renal injury, studies are also appearing on its behavior in chronic renal disorders. These include some very different pathologies, and it is evident both that these will have very different effects on NGAL levels and that the stage of evolution of the particular pathology will also be significant. It is therefore hardly to be expected that NGAL levels will be applicable to any generalized diagnostic or prognostic evaluations of undifferentiated chronic renal failure. Nevertheless, it has been found that urinary NGAL levels were strongly to moderately correlated with renal disease activity and renal damage, but not with extra-renal disease activity or damage, in patients with childhood-onset systemic lupus erythematosus (SLE) [14]. Children with chronic kidney disease in stages 2-4 showed serum NGAL levels that correlated significantly if not supremely well with the glomerular filtration rate (GFR), and at lower values of GFR, serum NGAL correlated rather better than cystatin C [15]. However, whether the hint provided by this correlation in a group of patients with heterogeneous renal conditions can be turned to good use in assessing individual patients will depend on more detailed studies being carried out in chronic kidney diseases of defined etiology.

Renal and extra-renal NGAL: free NGAL and NGAL-MMP-9 complex

It is apparent from some of the above that NGAL is far from exclusively produced by the injured kidney and different levels of extra-renal secretion of NGAL in concomitant conditions will influence the cutoffs that have to be employed to make NGAL levels optimally diagnostic of renal injury. NGAL of extra-renal origin continues to receive attention, both with respect to its role in neutrophils and inflammation and in relation to its expression in certain cancers. In renal tubule cells, neutrophil precursors and other cells including epithelial cells that have undergone malignant transformation, NGAL may be co-expressed with gelatinase B, also called matrix metaloproteinase-9 (MMP-9). Human NGAL has an extra cysteine residue that permits covalent linkage to MMP-9, a feature that has given NGAL its present name, but this residue is absent in rats and mice, so that covalent NGAL-MMP-9 complexes are not seen in these animals [16]. The relationship between free NGAL and NGAL-MMP-9 complex is by no means a regular one, and this relationship requires further study both in relation to renal injury and to extra-renal NGAL release from inflamed or malignantly transformed epithelia. For the time being, the status of NGAL-MMP-9 complex in the diagnosis of acute renal injury is uncertain, and it would be prudent to regard free NGAL and NGAL-MMP-9 complex as two independent markers that are likely to show a different response pattern in different pathologies. It is also very unclear how the linkage of NGAL to MMP-9 in humans relates to NGAL's important function as a binder and carrier of siderophore iron. This linkage has an inhibitory effect on MMP-9 autodegradation and thus tends to enhance the proteolytic effect of the enzyme [17], an effect that has been given importance in studies on the pathology of atheromatous plaques and the effects of mural thrombus on the wall of aortic aneurysms. Whether the effect of NGAL on MMP-9 is in fact quantitatively sufficient to bear interpretations of this type is still a matter of speculation.

Conclusions

In the current phase of investigations, NGAL is still being evaluated as a renal marker, and levels are thus being compared with other renal markers, especially the traditional functional marker, serum creatinine, and the more recent marker of GFR, serum cystatin C. The trouble with these comparisons is that raised NGAL of renal origin is a direct response to tubule cell injury, while the others are functional markers that may, after a period of time, reveal the consequences of such injury on the accumulation of creatinine and cystatin C in the blood. Serum creatinine is the physician's old, cheap and unreliable friend; despite repeated warnings in textbooks and reviews, it is sometimes forgotten just how variable the rate of creatinine release to the blood may be in patients of very different dietary habits and muscle mass, and how rapidly this may change in critical inflammatory and infective illness [18]. So many of the currently accepted definitions of ARD, or the RIFLE criteria for acute renal injury, which are based on absolute and/or relative rises in serum creatinine have little hope of providing an accurate diagnosis of ARD in seriously ill patients where the supply-side of the creatinine is so variable. Cystatin C may be a better indicator of glomerular filtration because its supply is less variable, but still, it takes time for cystatin C to accumulate in the blood after the renal injury. So assessments of NGAL as a marker of acute renal injury by comparison with these markers or clinical criteria based on them, are comparisons of different entities, destined to give muddy answers. Such studies must be performed with due care to follow these markers over an adequate period of time to allow for the slow response of the functional markers.

If it were possible to measure NGAL exclusively of renal origin, NGAL would probably be one of the finest markers of renal tubule cell damage that could be conceived. In urine, NGAL shows a 10,000-fold concentration rise from normal levels to the highest levels due to renal injury; in plasma the maximum rise is about 100-fold. This makes NGAL potentially a very sensitive marker of different degrees of renal injury. However, the lower end of this wide dynamic range is occupied by NGAL rises from extra-renal sources and it will be the diagnostician's privilege and responsibility to take this into account.

References

1. Uttenthal LO. NGAL: a marker molecule for the distressed kidney? Clin Lab Internat November 2005; 29(7): 39-41

2. Mishra J et al. Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renal innjury after cardiac surgery. Lancet 2005; 365(9466): 1231-1238.

3. Wagener G et al. Association between increases in urinary neutrophil gelatinase-associated lipocalin and acute renal dysfunction after adult cardiac surgery. Anesthesiology 2006; 105: 485-491.

4. Hemdahl AL et al. Expression of neutrophil gelatinase-associated lipocalin in atherosclerosis and myocardial infarction. Arterioscler Thromb Vasc Biol 2006; 26: 136-142.

5. Bachorzewska-Gajewska H et al. Neutrophil-gelatinase-associated lipocalin and renal function after percutaneous coronary interventions. Am J Nephrol 2006; 26: 287-292.

6. Bachorzewska-Gajewska H et al. Neutrophil gelatinase-associated lipocalin (NGAL) correlations with cystatin C, serum creatinine and eGFR in patients with normal serum creatinine undergoing coronary angiography. Nephrol Dial Transplant 2007; 22: 295-296.

7. Elneihoum AM et al. Leukocyte activation in atherosclerosis: correlation with risk factors. Atherosclerosis 1997; 131: 79-84.

8. Mishra J et al. Neutrophil gelatinase-associated lipocalin: a novel early urinary biomarker for cisplatin nephrotoxicity. Am J Nephrol 2004; 24: 307-315.

9. Junge G et al. Urinary neutrophil gelatinase-associated lipocalin (NGAL) in brain-dead kidney donors indicates post transplant graft function. Transplantation 2006: 15(1 Suppl 2):787.

10. Parikh CR et al. Urine NGAL and IL-18 are predictive biomarkers for delayed graft function following kidney transplantation. Am J Transplant 2006; 6: 1639-1645.

11. Mishra J et al. Kidney NGAL is a novel early marker of acute injury following transplantation. Pediatr Nephrol 2006; 21: 856-863.

12. Bangert K et al. NGAL is significantly increased in urine and plasma in acute renal failure. Intensive Care Med 2006; 32(Suppl 1): S10.

13. Trachtman H et al. Urinary neutrophil gelatinase-associated lipocalcin in D+HUS: a novel marker of renal injury. Pediatr Nephrol 2006; 21: 989-994.

14. Brunner HI et al.Urinary neutrophil gelatinase-associated lipocalin as a biomarker of nephritis in childhood-onset systemic lupus erythematosus. Arthritis Rheum 2006; 54: 2577-2584.

15. Mitsnefes MM et al. Serum neutrophil gelatinase-associated lipocalin as a marker of renal function in children with chronic kidney disease. Pediatr Nephrol 2007; 22: 101-108.

16. Kjeldsen L et al. Human neutrophil gelatinase-associated lipocalin and homologous proteins in rat and mouse. Biochim Biophys Acta 2000; 1482: 272-283.

17. Yan L et al. The high molecular weight urinary matrix metalloproteinase (MMP) activity is a complex of gelatinase B/MMP-9 and neutrophil gelatinase-associated lipocalin (NGAL). Modulation of MMP-9 activity by NGAL. J Biol Chem 2001; 276:37258-37265.

18. Mehta RL, Chertow GM. Acute renal failure definitions and classification: Time for a change? J Am Soc Nephrol 2003; 14: 2178-2187.

The author

Lars Otto Uttenthal, M.A., D.Phil., B.M., B.Ch., M.R.C.P. (UK)

Scientific Director

AntibodyShop A/S

This article was originally written for Clinical Laboratory International, April 2007 issue, and reproduced here with permission from the publisher.