Urinary biomarkers in acute kidney injury


Published on

Acute kidney injury (AKI) is common in critically ill patients and is associated with high mortality and cost of hospitalization.

Published in: Business, Technology
1 Like
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Urinary biomarkers in acute kidney injury

  1. 1. Review ArticleUrinary biomarkers in acute kidney injuryRajeev A. Annigeri*Consultant Nephrologist, Apollo Hospitals, Chennai, Tamil Nadu 600 006, Indiaa r t i c l e i n f oArticle history:Received 24 December 2012Accepted 5 February 2013Available online 13 February 2013Keywords:Acute kidney injuryBiomarkerNeutrophil gelatinase-associatedlipocalin (NGAL)a b s t r a c tAcute kidney injury (AKI) is common in critically ill patients and is associated with highmortality and cost of hospitalization. The standard clinical diagnosis of AKI is based on theestimation of serum creatinine, which is a late and unsatisfactory marker of AKI. Anintense search for an early biomarker has yielded considerable success in recent years andseveral urinary biomarkers to predict AKI early and reliably have been identified. Amongthese, urinary concentration of neutrophil gelatinase-associated lipocalin (NGAL) holdsmuch promise and has been validated in several population of AKI in intensive care unit(ICU). Concentrations of other urinary biomarkers such as kidney injury molecute-1 (KIM-1), interleukin-18 (IL-18) and Liver fatty acid binding protein (L-FABP) are promising, butawait validation. Sensitivity and specificity of urine NGAL (uNGAL) to predict AKI is best inhomogeneous population and pediatric patients, but somewhat less in heterogeneousadult population in ICU. In addition, uNGAL predicts severe AKI, renal replacement therapyinitiation and mortality in ICU. In future, further studies are likely to clarify and broadenthe application of urine biomarkers in research and clinical practice in AKI. Future holdsmuch promise in terms of therapeutic interventions based on biomarkers for primary andsecondary prevention in AKI.Copyright ª 2013, Indraprastha Medical Corporation Ltd. All rights reserved.1. IntroductionAcute kidney injury (AKI) is common in hospitalized patients,occurring in up to 40% of patients admitted to intensive careunit (ICU) and is associated with high mortality and cost oftherapy.1Recent studies indicate that even a small incrementin serum creatinine (SCr) of 0.3 mg/dl is associated with sig-nificant increment in mortality and cost of hospitalization.2Till recently there were no uniform criteria to define AKI.Over the last decade, a collaborative effort between nephrol-ogists and intensivists has lead to new criteria to define of AKIbased on SCr and urine output. These diagnostic criterianamely risk, injury, failure, loss, and end-stage kidney (RIFLE)and acute kidney injury network (AKIN) have been validatedand accepted as standard definitions worldwide.3,4However,the new “gold standard” criteria to define AKI appear to beflawed, since they take into consideration only the functionalaspect of AKI, but not the structural injury.SCr is an unsatisfactory criteria to define AKI for followingreasons5: 1) rise in SCr occurs when glomerular filtration rate(GFR) has already declined by 30e40%, implicating that it islate marker of AKI, 2) creatinine generation is reduced inseptic AKI compared to normal population, thereby over-estimating GFR in them, 3) fluid overload, often seen in crit-ically ill patients dilutes SCr concentration, therebyunderestimating its true concentration in serum, and 4) SCrlevel is affected by age, gender, muscle mass and some med-ications that affect the generation and excretion of creatinine.* Fax: þ91 44 28294429.E-mail addresses: drrajeevannigeri@hotmail.com, r_annigeri@yahoo.com.Available online at www.sciencedirect.comjournal homepage: www.elsevier.com/locate/apmea p o l l o m e d i c i n e 1 0 ( 2 0 1 3 ) 3 6 e4 00976-0016/$ e see front matter Copyright ª 2013, Indraprastha Medical Corporation Ltd. All rights reserved.http://dx.doi.org/10.1016/j.apme.2013.02.002
  2. 2. All interventional studies in AKI in humans have eitherfailed or are equivocal, in contrast to animal studies wherebenefit is often seen.6It is widely believed that, late inter-vention based on SCr has been the major stumbling block forthe failed trials in human AKI. In this respect, an urgent needfor a biomarker of AKI that could accurately detect AKI muchbefore a rise in SCr has been widely felt both by clinicians andresearchers. It is expected that interventional trials in AKIapplied very early in the course of AKI detected by early bio-markers could yield positive results.2. Biomarker in AKI: definition andcharacteristicsA biomarker is an indicator of a biological state. A biomarker isa characteristic that is objectively measured and evaluated asan indicator of normal biological processes, pathogenic pro-cesses, or pharmacologic responses to a therapeutic inter-vention.7In the context of AKI, a biomarker is an objectivemeasure of renal tubular injury. This biomarker would bedetected in urine or in the blood following AKI and would beexpected to quantify the degree of tubular injury therebypredicting the severity of AKI. Also, a biomarker of tubularinjury would be useful in distinguishing acute tubular necro-sis from pre-renal azotemia, which is easily reversible.An ideal biomarker of AKI should have all or most of thefollowing characteristics5: 1) basal levels should allow riskstratification, 2) remains elevated for a significantly longerperiod after injury, 3) appear very early, even before a clinicaldiagnosis of AKI can be made, 4) reliably make a clinicaldiagnosis of AKI with high sensitivity and specificity, 5) havewide spectrum of values correlating to degree of injury, 6)unaffected by demographics and clinical factors, 7) present insignificant concentration in urine or blood, 8) easy to measurein the laboratory, 9) the measurement should be inexpensiveand results available rapidly, and 10) able to predict accuratelythe severity of AKI and need for renal replacement therapy.3. Early biomarker in AKI: suitablecandidatesAn intense search for an early biomarker of AKI has resultedin substantial progress in this area. Advances in the molecularbiology in the last decade have enabled scientists to identifyand test a number of biomarkers in AKI. Several candidateshave been identified, but only few have withstood the vigo-rous test in laboratory and clinical setting.The biomarkers of AKI can be classified as8: 1) Functionalmarkers: SCr, serum cystatin-C (CyC), 2) Upregulated proteins:neutrophil gelatinase-associated lipocalin (NGAL), kidneyinjury molecute-1 (KIM-1), interleukin-18 (IL-18), Liver fattyacid binding protein (L-FABP), heat shock protein-72 (Hsp-72),3) Low molecular weight proteins: urine CyC and 4) Enzymes:N-acetyl- b-D-glucosaminidase (NAG), alpha-glutathione s-transferase (a-GST) and p-glutathione s-transferase (pi-GST).Among these, urine NGAL (uNGAL) and plasma NGAL (pNGAL)concentrations are by far the most widely studied in clinicalsetting. My discussion in this paper will mainly focus onuNGAL, with a brief account of other clinically relevantbiomarkers.4. Biology of biomarkers of AKIThe temporal pattern of increase in urinary biomarkers in AKIfollowing cardiac-bypass surgery is shown in Fig. 1.94.1. Neutrophil gelatinase-associated lipocalin (NGAL)Among all the biomarkers, the measurement uNGAL is themost promising in predicting AKI. NGAL exists as a 25-kDmonomer protein, belongs to the lipocalin superfamily. It wasinitially found in activated neutrophils, and has a role as aninnate antibacterial factor. Subsequently, it was shown thatseveral other types of cells, including those in the renal tubuleproduce NGAL in response to various injuries. Under physio-logic conditions, NGAL is expressed in low concentrations inkidney, lung, and gastrointestinal tissue. Circulating NGAL isfiltered by the glomerulus, and reabsorbed entirely in theproximal tubule. It is secreted in low concentrations by thethick ascending limb of the renal tubule and the normal uri-nary concentration of NGAL is very low (<5 ng/mL). In proxi-mal tubular injury, NGAL synthesis is increased andreabsorption may decrease, resulting in increased urinarylevels. In distal tubular injury there is increased distal renaltubular expression and synthesis of NGAL and increased uri-nary levels of NGAL. NGAL is protease resistant and does notseem to be degraded or metabolized after synthesis andsecretion into the tubular lumen.10e12The normal plasma NGAL (pNGAL) level in healthy adults is50e90 ng/mL. The concentration of pNGAL increases in AKI,though the major source of pNGAL in AKI appears to be non-renal. Elevated pNGAL in AKI appears to be due to con-comitant hepatic, pulmonary, intestinal tissue injury orrelease from immune cells such as neutrophils, macrophages,coupled with decreased glomerular filtration of NGAL andreduced proximal tubular absorption of filtered NGAL. How-ever, the relative contribution of these factors causing ele-vation of plasma NGAL in AKI remains to be determined.10e12Measurement of uNGAL and pNGAL are currently availablecommercially13using several methods such as radio-immunoassay, western blot and enzyme-linkedFig. 1 e Temporal pattern of increase in urine biomarkersin AKI following cardiac-bypass surgery.a p o l l o m e d i c i n e 1 0 ( 2 0 1 3 ) 3 6 e4 0 37
  3. 3. immunosorbent assay (ELISA). Unfortunately, at the presenttime there are no available international recommendations orguidelines concerning the quality specifications for NGALassay.12Currently there are no standard reference values ofuNGAL to diagnose AKI are not available. Normally uNGAL isbelow 25 ng/mL and levels above 150 ng/mL are stronglysuggestive of AKI. Age and gender appear to have an impact onthe uNGAL levels but their clinical implications remain to beclarified.124.2. Kidney injury molecule-1 (KIM-1)KIM-1 is a phosphatidylserine receptor, a type I membranereceptor that recognizes apoptotic cells and direct them tolysosomes. KIM-1 is a scavenger receptor on renal epithelialcells that converts the normal proximal tubule cell intoa phagocyte. It has a role in clearing the cellular debris in therenal tubular lumen thereby keeping the tubular lumen pat-ent, which in addition also helps to reduce inflammation.KIM-1 is not expressed in normal kidney, but is expressed inhigh concentration in proximal renal tubules in AKI. Its per-sistence in the epithelial cell until the cell has completelyrecovered, the rapid and robust cleavage of the ectodomainand the ex vivo room temperature stability of the ectodomainmake it a suitable biomarker in AKI. Its presence in the urine ishighly specific for kidney injury.5,144.3. Interleukin-18 (IL-18)Interleukin (IL)-18 is a pro-inflammatory cytokine that isinduced in proximal tubular epithelial cells in response toinjury. After renal injury, IL-18 is secreted in the urine beforea significant decrease in renal function. IL-18 is a novel bio-marker that has been studied in detail in preclinical ischemia-reperfusion models followed by successful translation of thefindings in human kidney diseases. In human studies, urinarylevels of IL-18 are markedly elevated in AKI compared tochronic kidney disease, nephrotic syndrome or urinary tractinfection.5,154.4. Liver fatty acid binding proteins (L-FABP)FABP are small (15 kDa) cytoplasmic proteins abundantlyexpressed in tissues with active fatty acid metabolism. Theirprimary function is the facilitation of long-chain fatty acidtransport, the regulation of gene expression and the reductionof oxidative stress. Urinary L-FABP is undetectable in healthycontrol urine, which is explained by efficient proximal tubularinternalization via megalin-mediated endocytosis. Underischemic conditions, tubular L-FABP gene expression isinduced and in renal disease, the proximal tubular reabsorp-tion is reduced, resulting in increased level of urine L-FABPin AKI.5,165. Biomarkers in AKI: clinical validationSince the availability for clinical use, there has been increas-ing number of prospective observational studies assessing theaccuracy of several biomarkers in several clinical settings. Ofthem, the most widely studied and published are that byNGAL concentration in urine and plasma.5.1. Prediction of AKIThe first landmark study to report predictive ability of uNGALin AKI was by Mishra et al in 2005.17Theymeasuredplasma andurine NGAL serially in 71 children undergoing cardiopulmo-nary bypass and reported excellent predictive ability of uNGALand pNGAL in predicting AKI in this homogeneous pediatricpopulation, with area under the curve for the receiver-operating characteristic (AUC-ROC) of 0.998 for a cutoff valueof uNGAL 50 mcg/L. Since then several studies in children aswell as in adults, both in homogeneous population such as postcardiac surgery,17post contrast administration,18renal trans-plant recipients,19cardio-renal syndrome20and heteroge-neous population of critically ill patients have beenreported.21e24In a recent meta-analysis, Haase et al analyzed19 studies involving 2538 patients in whom 19% developedAKI.25The diagnostic odds ratio (DOR) for elevated uNGAL (fora cut off value 150 mcg/L) in developing AKI was 25.5 (95% CI,8.9e72.8) and AUC-ROC was 0.830 (95% CI, 0.741e0.918). Incardiac surgery patients, the DOR/AUC-ROC of NGAL was 13.1(95% CI, 5.7e34.8)/0.775 (95% CI, 0.669e0.867); in heterogeneouspopulation of critically ill patients was 10.0 (95% CI, 3.0e33.1)/0.728 (95% CI, 0.615e0.834); and after contrast infusion was 92.0(95% CI, 10.7e794.1)/0.894 (95% CI, 0.826e0.950). The predictiveability of uNGAL was similar compared to pNGAL.The prediction of AKI by NGAL was superior to SCr andNGAL was able to predict AKI 24e48 h prior to a clinicaldiagnosis based on RIFLE criteria. However, these results wereobserved in homogeneous group of patients where the timingof renal insult was apparent, such as contrast administrationand post cardiac surgery patients. A few studies have shownthat a single measurement of NGAL in urine or plasma done atthe time of admission was also a good predictor ofAKI.10,12,25,265.2. Prediction of renal replacement therapy (RRT)initiation and mortalitySeveral studies indicate that in addition to predicting AKI,NGAL also predicts severe AKI requiring RRT and mortality incritically ill patients.22Siew et al23reported in 451 critically illpatients significantly elevated uNGAL (ng/mg creatinine)levels in those who required RRT (mean: 548, range: 156e4665)compared with who did not (mean: 61, range: 17e232);(P < 0.001). Also uNGAL levels were significantly higher inthose who died (mean: 223, range: 36e1074) compared tothose survived 56 (mean: range: 16e195); (P < 0.001). Bagshawet al also found a marked elevation of uNGAL in patientsrequiring RRT, compared to those who did not.27Haase et al20analyzed pooled data from 2322 critically ill patients withpredominantly cardio-renal syndrome from 10 prospectiveobservational studies of NGAL. They used the terms NGAL(þ)or NGAL(À) based on whether NGAL was elevated or not andthe terms Cr(À) or Cr(þ) according to according to whether SCrwas normal or elevated. The outcomes studied were renalreplacement therapy (RRT) initiation and hospital mortality.The results showed that NGAL was a better predictor of RRTa p o l l o m e d i c i n e 1 0 ( 2 0 1 3 ) 3 6 e4 038
  4. 4. initiation and mortality than SCr and the results are shown inFig. 2.5.3. Limitations of NGALThere are several limitations to the measurement of NGAL asa biomarker in AKI, which are as follows: 1) NGAL concentra-tion in urine or plasma is most sensitive and specific in pre-dicting AKI in studies of homogeneous population derivedfrom a single center. These studies included patients withsingle, acute, easily identifiable, and predictable nephrotoxicinsults, such as cardiopulmonary bypass or intravenous con-trast. NGAL appears to be less sensitive and specific in moreheterogeneous cohorts with multifactorial causes for AKI, inwhom the time of onset of renal insult cannot be pinpointedaccurately.11,12,252) Several confounding factors may affectaccuracy of NGAL in predicting AKI. Plasma NGAL may befalsely elevated in sepsis without AKI, malignancy and hemo-lysis.12,13Also, uNGAL may be falsely elevated in urinary tractinfection.123) Currently there are no available internationalrecommendations or guidelines concerning the quality speci-fications for NGAL assay and there is no consensus on thecutoff values to predict AKI accurately.124) Accuracy of uNGALappears to vary with age and gender and its predictive ability ismuch better in pediatric population compared to adult popu-lation.11,125) Measurement of plasma and urine NGAL isexpensive, though not prohibitively. The utility of a singlevalue is less accurate that serial values, but serial measure-ment would add to the cost substantially. The cost benefit ofNGAL measurement in clinical practice remains to be assessedespecially in heterogeneous population of critically ill pa-tients.126) Currently, the practical utility of NGAL measure-ment remains speculative, since there are no interventions inAKI of proven clinical benefit. However, this skepticismis likelyto change when interventional studies in early AKI based onbiomarkers that are currently ongoing become available. 7)Most studies excluded CKD population, which may forma substantial population especially the elderly, in ICU.115.4. Clinical validation of other biomarkers in AKIOther biomarkers are not vigorously tested as NGAL. Amongthem, urine KIM-1 holds much promise. Several studies haveshown good, but not excellent predictive ability of urine KIM-1with AUC-ROC of 0.55e0.78.8A few human studies show thatuIL-18 predicts AKI, but with low accuracy (ACU-ROC variedbetween 0.55 and 0.73).8,15It appears that addition of one ofthese biomarkers to NGAL measurement increases the sen-sitivity and specificity of prediction model in AKI, but thisneeds validation.6. AKI biomarkers: implications for thefutureIn future, biomarkers are likely to be incorporated in theexisting diagnostic criteria of AKI, to improve its accuracy. It islikely that more than one biomarker may be simultaneouslyused to increase the sensitivity and specificity to predict AKIand use of a panel of biomarker replacing a single marker isvery likely.26The interventional studies in very early stages ofAKI based on elevated biomarkers are likely to be performed inthe near future and are expected to yield positive results. AKIbiomarkers are useful in determining severity of AKI andhence may be useful to guide to initiate RRT early based ondegree of biomarker elevation.7. ConclusionsEmergence of new early urinary biomarkers to predict AKI hassignificantly advanced our understanding of AKI. Biomarkerssuch as NGAL predict AKI accurately and early, before thedecline in GFR is apparent clinically, more so in homogeneouspopulation. In addition, they also predict severity of AKI, RRTinitiation and hospital mortality. Future holds much promisein terms of therapeutic interventions based on biomarkers forprimary and secondary prevention in AKI.Conflicts of interestThe author has none to declare.r e f e r e n c e s1. Bagshaw SM, George C, Dinu I, et al. A multi-centre evaluationof the RIFLE criteria for early acute kidney injury in criticallyill patients. Nephrol Dial Transplant. 2008;23:1203e1210.2. Chertow GM, Burdick E, Honour M, et al. Acute kidney injury,mortality, length of stay, and costs in hospitalized patients. JAm Soc Nephrol. 2005;16(11):3365e3370.3. Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure-definition, outcome measures, animal models, fluid therapyand information technology needs: the Second InternationalConsensus Conference of the Acute Dialysis Quality Initiative(ADQI) Group. Crit Care. 2004;8:R204eR212.4. Mehta RL, Kellum JA, Shah SV, et al. Acute kidney injurynetwork: report of an initiative to improve outcomes in acutekidney injury. Crit Care. 2007;11:R31.5. Slocum JL, Heung M, Pennathur S. Marking renal injury: canwe move beyond serum creatinine? Translational Res.2012;159:277e289.Fig. 2 e Comparison of NGAL and serum creatinine (SCr) inpredicting renal replacement therapy (RRT) initiation andmortality in cardio-renal syndrome.a p o l l o m e d i c i n e 1 0 ( 2 0 1 3 ) 3 6 e4 0 39
  5. 5. 6. Jo SK, Rosner MH, Okusa MD. Pharmacologic treatment ofacute kidney injury: why drugs haven’t worked and what ison the horizon. Clin J Am Soc Nephrol. 2007;2:356e365.7. Biomarkers Definitions Working Group. Biomarkers andsurrogate endpoints: preferred definitions andconceptual framework. Clin Pharmacol Ther. 2001;69:89e95.8. de Geus HR, Betjes MG, Bakker J. Biomarkers for the predictionof acute kidney injury: a narrative review on current statusand future challenges. Clin Kidney J. 2012;5(2):102e108.9. Parikh CH, Lu JC, Coca SG, et al. Tubular proteinuria in acutekidney injury: a critical evaluation of current status andfuture promise. Ann Clin Biochem J. 2010;47:301e312.10. Shamin D, Dworkin LD. Neutrophil gelatinase-associatedlipocalin (NGAL) as a biomarker for early acute kidney injury.Crit Care Clin. 2011;27:379e389.11. Makris K, Kafkas N. Neutrophil gelatinase-associatedlipocalin in acute kidney injury. Adv Clin Chem.2012;58:141e191.12. Clerico A, Galli C, Fortunato A, Ronco C. Neutrophilgelatinase-associated lipocalin (NGAL) as biomarker of acutekidney injury: a review of the laboratory characteristics andclinical evidences. Clin Chem Lab Med. 2012;50(9):1505e1517.13. Pedersen KR, Ravn HB, Hjortdal VE, et al. Neutrophilgelatinase-associated lipocalin (NGAL): validation ofcommercially available ELISA. Scand J Clin Lab Invest.2010;70:374e382.14. Bonaventura JV. Kidney injury molecule-1 (KIM-1): a urinarybiomarker and much more. Nephrol Dial Transplant.2009;24(11):3265e3268.15. Parikh CR, Mishra J, Thiessen-Philbrook H, et al. Urinary IL-18is an early predictive biomarker of acute kidney injury aftercardiac surgery. Kidney Int. 2006;70:199e203.16. Noiri E, Doi K, Negishi K, et al. Urinary fatty acid-bindingprotein 1: an early predictive biomarker of kidney injury. Am JPhysiol Ren Physiol. 2009;296:F669eF679.17. Mishra J, Dent C, Tarabishi R, et al. Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renalinjury after cardiac surgery. Lancet. 2005;365(9466):1231e1238.18. Hirsch R, Dent C, Pfriem H, et al. NGAL is an early predictivebiomarker of contrast-induced nephropathy in children.Pediatr Nephrol. 2007;22(12):2089e2095.19. Parikh CR, Jani A, Mishra J, et al. Urine NGAL and IL-18 arepredictive biomarkers for delayed graft function followingkidney transplantation. Am J Transplant. 2006;6:1639e1645.20. Haase M, Devarajan P, Haase-Fielitz A, et al. The outcome ofneutrophil gelatinase-associated lipocalin-positivesubclinical acute kidney injury. J Am Coll Cardiol.2011;57(17):1752e1761.21. NickolasTL,O’RourkeMJ,YangJ,etal.Sensitivityandspecificityof a single emergency department measurement of urinaryneutrophil gelatinase-associated lipocalin for diagnosing acutekidney injury. Ann Intern Med. 2008;148:810e819.22. Cruz DN, de Cal M, Garzotto F, et al. Plasma neutrophilgelatinase-associated lipocalin is an early biomarker for acutekidney injury in an adult ICU population. Intensive Care Med.2010;36(3):444e451.23. Siew ED, Ware LB, Gebretsadik T, et al. Urine neutrophilgelatinase associated lipocalin moderately predicts acutekidney injury in critically ill adults. J Am Soc Nephrol.2009;20:1823e1832.24. Constantin JM, Futier E, Perbet S, et al. Plasma neutrophilgelatinase associated lipocalin is an early marker of acutekidney injury in adult critically ill patients: a prospectivestudy. J Crit Care. 2009;25(176):1e6.25. Haase M, Rinaldo Bellomo R, Devarajan P, et al. Accuracy ofneutrophil gelatinase-associated lipocalin (NGAL) indiagnosis and prognosis in acute kidney injury: a systematicreview and meta-analysis. Am J Kidney Dis.2009;54(6):1012e1024.26. Di Grande A, Giuffrida C, Carpinteri G, et al. Neutrophilgelatinase-associated lipocalin: a novel biomarker for theearly diagnosis of acute kidney injury in the emergencydepartment. Eur Rev Med Pharmacol Sci. 2009;13(3):197e200.27. Bagshaw SM, Bennett M, Haase M, et al. Plasma and urineneutrophil gelatinase associated lipocalin in septic versusnon-septic acute kidney injury in critical illness. Intensive CareMed. 2010;36:452e461.a p o l l o m e d i c i n e 1 0 ( 2 0 1 3 ) 3 6 e4 040