Mini-Review Traditional Urinary Biomarkers in the ...cjasn.asnjournals.org/content/7/1/167.full.pdf · Mini-Review Traditional Urinary Biomarkers in the Assessment of ... whereas

Mini-Review

Traditional Urinary Biomarkers in the Assessment ofHospital-Acquired AKI

Mark A. Perazella and Steven G. Coca

SummaryTraditional biomarkers, such as urine chemistries and urine microscopic elements, are used in the diagnosis andcare of patients with AKI. Urine chemistries, such as fractional excretion of sodium and fractional excretion ofurea, are useful for differentiating prerenal AKI from acute tubular necrosis only in select patients. Urinemicroscopy using a quantitative evaluation of the urine sediment for renal tubular epithelial cells, renal tubularepithelial cell casts, and granular casts has recently been shown to differentiate prerenal AKI from acute tubularnecrosis and also provide prognostic information. Urine microscopy has also been noted to compare favorablywith new urine biomarkers for diagnosis and prognosis of AKI. Thus, current information on urine diagnosticssuggests that urine chemistries have a limited role in differential diagnosis of AKI, whereas urine microscopy andnew urine biomarkers may be used together to differentiate prerenal AKI from acute tubular necrosis and predictsuch outcomes as worsened AKI, acute dialysis, and death.

Clin J Am Soc Nephrol 7: 167–174, 2012. doi: 10.2215/CJN.09490911

“. . .the ghosts of dead patients that haunt us do notask why we did not employ the latest fad of clinicalinvestigation. They ask us, why did you not test myurine?”

—Sir Robert Grieve Hutchison (1871–1960)

IntroductionAKI occurs commonly in the hospital, and its in-cidence continues to increase (1). Of note, AKI is as-sociated with multiple adverse outcomes, includingincident and progressive CKD, ESRD, and death (2).In the hospital, most AKI cases are due to prerenalAKI or acute tubular necrosis (ATN), with a smallcontribution from acute glomerular, vascular, and in-terstitial diseases and urinary obstruction. We focuson the two most common causes of AKI.

Evaluation of patients with AKI has become morestandardized through the use of such definitions asthe Risk-Injury-Failure-Loss-End Stage (RIFLE) andAcute Kidney Injury Network (AKIN) criteria todiagnose and classify this entity (3,4). These criteria,however, do not permit differentiation of the varioustypes of AKI, including prerenal and ATN, whichrequire inhomogeneous management. This is particu-larly relevant because excessive fluid repletion in pa-tients with ATN may be associated with untowardoutcomes. Thus, clinicians depend on clinical toolssuch as history, examination, ultrasonography, andcertain laboratory data to make this distinction. Forthe most part, the laboratory tests used to differenti-ate these two common causes of hospital-acquiredAKI are urinary tests: urine chemistries and urinalysiswith urine microscopy.

Urine ChemistriesFractional Excretion of SodiumFractional excretion of sodium (FeNa) is calculated

as follows:

ð½U/P�Na/ ½U/P� creatinine3 100Þ

where U indicates urine and P indicates plasma. It wasone of the first urine chemistries used to differentiateprerenal AKI from ATN. It is based on the premisethat intact tubules will reabsorb sodium in the pre-renal setting whereas injured tubules occurring withATN will not. In 1976, Espinel tested this hypothesisin 17 highly selected patients with oliguric AKI (5). Incontrast to urine sodium, FeNa tested well in thisgroup: ,1% for prerenal and .3% for ATN. It is no-table that patients with CKD, glucosuria, bicarbonat-uria, and other comorbid conditions were excluded.A follow-up study in 87 patients with AKI and pre-renal, oliguric ATN, and nonoliguric ATN noted aseparation of prerenal AKI from both types of ATN(6). However, many patients in the ATN groups, inparticular nonoliguric ATN, had FeNa levels of1%–2%. Miller and colleagues evaluated FeNa in 102patients with five subtypes of AKI using fairly strin-gent diagnostic criteria (7). They found that FeNatested well in prerenal and oliguric ATN, but 10% ofpatients with nonoliguric ATN had FeNa levels ,1%.Patients with obstruction had high FeNa, whereasthose with acute GN had a low FeNa level. Sub-sequently, numerous authors described various formsof ATN with FeNa ,1% and prerenal AKI with FeNa.1% (8). One of the major reasons for reduced FeNautility relates to the use of this test in disease states(Table 1) not included in the initial studies. Such

Section ofNephrology, YaleUniversity School ofMedicine, NewHaven, Connecticut

Correspondence:Dr. Mark A. Perazella,Section ofNephrology, YaleUniversity School ofMedicine, BB 114,330 Cedar Street,New Haven,CT 06520-8029.Email: [email protected]

www.cjasn.org Vol 7 January, 2012 Copyright © 2012 by the American Society of Nephrology 167

diseases disturb the typical tubular response to sodium re-absorption and make FeNa poorly reflective of the actualcause of AKI.

Fractional Excretion of UreaFractional excretion of urea (FeUrea) is calculated as

follows:

ð½U/P� urea/ ½U/P� creatinine3 100Þ:

It is based on the same premise as FeNa except that ureareabsorption is less affected by loop and thiazide diuretics,which act distally to the proximal tubule (the major siteof urea reabsorption). FeUrea ,35% reflects prerenal AKI,whereas FeUrea.50% reflects loss of tubular function fromATN. In 1992, Kaplan and Kohn noted a discordance be-tween FeNa .2% and FeUrea ,35% in 6 patients withprerenal AKI (9). Retrospective chart review by theseauthors noted a similar discordance in 20 prerenal patients.Ten years later, Carvounis and colleagues tested the utilityof FeUrea in 102 patients with prerenal AKI (n=50 withoutdiuretic; n=27 with diuretic) and ATN (n=25) of variouscauses (10). Diagnosis was based on clinical measures,urine microscopy, urine chemistries, and “response totherapy.” FeUrea ,35% tested very well, with a sensitivityof 90% and a specificity of 96%, for the diagnosis of pre-renal AKI. Subsequent study of FeUrea had less favorableresults. Pépin and colleagues tested FeUrea in 99 patientswith prerenal AKI (n=23 with no diuretic; n=43 with di-uretic) or ATN (n=12 with no diuretic; n=21 with diuretic)and, in contrast to the previous study, found the test un-helpful and slightly worse than FeNa (11). FeUrea ,35%had a sensitivity of 68% and a specificity of 48% for pre-renal AKI. Results were no better when stratified by di-uretic use. Possible explanations for the difference in testingresults between these two studies exist. For example, dif-ferent definitions of prerenal and ATN were used. More-over, in Pépin and colleagues’ study, patients were olderand sicker, with more comorbid conditions (such as CKDand diabetes mellitus) and need for intensive care unit care.

Pépin and coworkers’ study had more patients who werenonoliguric and had an increased time from AKI diagnosisto specimen collection. This last point is particularly rele-vant because it could have allowed interventions such asintravenous fluids to be administered before measurementof FeUrea.Diskin and colleagues examined the utility of FeUrea

(,40% cutoff) in 100 consecutive hospitalized patientswith oliguric AKI (,600 ml/d) due to prerenal AKI(n=80) or intrinsic AKI (12). FeUrea ,40% tested verywell in detecting prerenal AKI (98% overall accuracy [78of 80 patients]), even in the presence of diuretics (correctin 54 of 56 patients). Although FeUrea performed well inoliguric prerenal AKI, it is unclear whether the studyincluded patients with underlying CKD, diabetes mellituswith glucosuria, or bicarbonaturia, which can reduce testaccuracy.Thus, urine chemistries such as FeNa and FeUrea appear

to have a limited role in differentiating prerenal AKI fromATN in many patients with hospital-acquired AKI. As such,these tests should be used only in specific patients and incertain clinical scenarios.

Urine MicroscopyA time-honored laboratory test used to evaluate AKI is

urine microscopy. For example, visualization of red bloodcell casts in the urine sediment is fairly definitive for GN. Inaddition, numerous textbooks and review articles posit thatthe presence of renal tubular epithelial cells (RTECs), coarsegranular casts, and RTEC casts in the urine is evidence forATN, whereas bland sediment and hyaline casts are con-sistent with prerenal AKI. This is a logical claim becausean ischemic or nephrotoxic insult causes tubular injury,with resulting apoptosis or necrosis of RTECs. The RTECsare shed into tubular lumens, where they are excreted freeor as RTEC casts or granular casts (Figure 1) that can beexamined in a fresh urine sediment. An important conceptis that prerenal AKI and ATN are a spectrum and maysometimes coexist. As such, the urine sediment would beexpected to have more cells and casts with severe ATN as

Table 1. Limitations of fractional excretion of sodium

Scenarios with FeNa , 1% Scenarios with FeNa . 2%normal kidney function with low or moderatesalt intake

normal kidney function with high salt intakeor IV saline

acute GN late urinary obstructionearly AIN late AINacute urinary obstruction glucosuriatransplant rejection bicarbonaturia

FeNa , 1% despite ATN FeNa . 2% despite prerenal AKIAKI with liver failure or CHF use of diureticssepsis-associated AKI CKDradiocontrast nephropathy FeNa after IVF therapynonoliguric ATN glucosuriamyoglobinuric ATN bicarbonaturiahemoglobinuric ATN salt-wasting disorders

FeNA, fractional excretion of sodium; AIN, acute interstitial nephritis; ATN, acute tubular necrosis; CHF, congestive heart failure;IV, intravenous; IVF, intravenous fluid.

168 Clinical Journal of the American Society of Nephrology

compared with prerenal AKI with patchy tubular injury;thus, it seems logical to assess the urine findings quantita-tively. Most nephrologists prescribe to this way of thinking,but until recently the literature confirming this practice wasrelatively limited.

Differential DiagnosisSmall studies have shown the utility of urine microscopy

in differentiating prerenal AKI from ATN. In 1991, Graberand coworkers focused on cells and casts present in theurine of 21 patients with hospital-acquired ATN (13). Urinemicroscopy demonstrated classic RTECs in 76% of pa-tients, granular casts in 62%, and atypical RTECs with cy-toplasmic vesicles in 76%, suggesting that a thoroughurine sediment examination could confirm a diagnosis ofATN in approximately three quarters of patients. Perazellaand colleagues examined the utility of urine microscopyand a urine sediment score (based on RTECs and granularcasts) in 231 patients with hospital-acquired AKI due toprerenal AKI or ATN (14). The likelihood ratios for ATNincreased in a dose-dependent fashion as the number ofgranular casts or RTECs increased and declined for

prerenal AKI (Table 2). In patients with an initial diagnosisof ATN (before urine microscopy), the presence of granu-lar casts or urine sediment score $2 had a positive pre-dictive value of 100% for a final diagnosis of ATN. Inpatients with an initial diagnosis of prerenal AKI, thelack of granular casts or a urine sediment score of 1 had anegative predictive value of 91% for a final diagnosis of pre-renal AKI. Thus, critical performance of urine microscopyappears to be useful to differentiate the most common causesof hospital-acquired AKI.

PrognosisAlthough AKI differentiation is important to guide

appropriate therapy, the current era of assessment of theutility of diagnostic tests demands that important clinicalend points are measured. In AKI, worsening of kidneyfunction, requirement for dialysis, and death are such endpoints. Marcussen and colleagues undertook such a studyin 51 patients with hospital-acquired AKI who had prerenalAKI, “non-ATN” AKI, or ATN (15). They found that thenumber of urinary cells and granular casts had a positivecorrelation with the magnitude of increase in serum creat-inine (i.e., worsening AKI). In addition, dialysis-requiringpatients had more RTECs and granular casts than patientswho did not require dialysis. In 2008, Chawla andcolleagues demonstrated that an “AKI cast scoring index”was useful in predicting severity and nonrecovery from AKI(16). Their score, based on RTEC casts and granular casts,was derived from 30 patients with ATN by three blindedreaders. A higher cast score (2.55 versus 1.57; P=0.04) wasassociated with AKI nonrecovery compared with AKI recov-ery (Table 3).In 2010, Perazella and coworkers used urine microscopy

and a modified urine sediment score (Figure 2) to evaluatethe outcome of worsening AKI in 197 patients with AKIthat was defined by the AKIN criteria and was due toprerenal AKI or ATN (17). The composite outcome of“worsening AKI” consisted of higher AKIN stage, dialysisrequirement, and death. With use of a urine sedimentscore based on granular casts and RTECs, increasing urinescore was associated with an increasing adjusted relativerisk for worsening AKI (score, 0 versus $3; relative risk, 7.3[95% confidence interval, 3.8–9.6]). The urine sedimentscore remained predictive for worsening AKI for all theindividual outcome components (Table 3 and Figure 3).Thus, urine microscopy appears to have utility not onlyin differentiating AKI but also in predicting severity ofAKI and death.

Urine Diagnostics in the 21st CenturyRequiem for Traditional Urine Biomarkers?Urine diagnostics in the 21st century have focused on the

development and evaluation of novel biomarkers for AKI.These new urinary proteins were discovered in animalsexposed to ischemia reperfusion, nephrotoxins, and otherforms of kidney injury (18–20). Multiple studies haveshown them to also increase in various forms of humanAKI (21). Studies of these novel biomarkers have examinedtheir utility not only in the differential diagnosis of AKIbut also in identifying early AKI and predicting clinicaloutcomes.

Figure 1. | Ischemic or nephrotoxic processes cause tubular injurywith apoptosis and necrosis. Tubular cells are shed into tubular lu-mens, where they are released into the urine as free renal tubularepithelial cells, renal tubular epithelial cell casts, or granular casts.These urine elements can be observed under the microscope in pa-tients with acute tubular necrosis.

Clin J Am Soc Nephrol 7: 167–174, January, 2012 Traditional Biomarkers in the Assessment of AKI, Perazella and Coca 169

With the rapid development of novel urinary biomark-ers as a point-of-care test for AKI, one is left to ponderwhether traditional biomarkers such as urine chemistriesand urine microscopy will become outdated. As reviewed,FeNa and FeUrea are easy to obtain but are useful fordifferentiating AKI only in select groups of patients. Urinemicroscopy is inexpensive and readily available andmaintains good performance in differential diagnosisand predicting clinical outcomes, but it requires trainingand experience. In addition, it is time consuming—it

requires obtaining a fresh urine specimen, centrifugingthe urine and preparing the slide, and viewing numerousfields under the microscope. In addition, the Clinical Lab-oratory Improvement Amendment limits the clinician’sability to perform this test outside the central laboratoryand requires other forms of certification, such as forProvider-Performed Microscopy. Thus, it appears thatthe field is ripe for a replacement test, such as the novelurine biomarkers. However, before traditional markersare deemed obsolete, a comparison of their performance

Table 2. Likelihood ratios for prerenal AKI and acute tubular necrosis based on urine microscopy (14)

Urine Findings ATN Prerenal AKILikelihood Ratio

ATN Prerenal AKI

Granular casts/LPF0 23 84 0.23 4.351–5 73 21 2.97 0.346–10 23 2 9.68 0.10.10 8 0 ∞ 0total 125 106

RTE cells/HPF0 75 88 0.72 1.391–5 38 18 1.97 0.516–20 11 0 ∞ 0.20 1 0 ∞ 0total 125 106

ATN, acute tubular necrosis; LPF, low-power field; RTE, renal tubular epithelial; HPF, high-power field.

Table 3. Studies evaluating urine microscopy and prognosis of AKI

Study Year(Reference) Population Patients

(n) Scoring System Outcomes Findings

Chawla et al.,2008 (16)

AKI on renalconsultservice

18 Grade 1–4a Renal nonrecovery AUC, 0.79

Perazella et al.,2010 (17)

AKI on renalconsultservice

197 Score 0 to $3b Worsened AKI(increase in AKIN stage,RRT, or death)

AUC, 0.75Score 1: RR, 3.4Score 2: RR, 6.6Score $3: RR, 7.3

Bagshaw et al.,2011 (23)

ICU patientswith AKI

83 Score 0 to $3c A) Worsened AKIB) RRT/death

AUC, 0.85Score 1–2: OR, 5.6Score $3: OR, 8.0

Hall et al.,2011 (24)

AKI $ stage 1 249 Score 0 to $3b Worsened AKI (increasein AKIN stage, RRT,or death)

AUC, 0.66; NRI, 24%Score 1: RR, 1.6Score 2: RR, 2.3Score $3: RR, 3.5

AUC, area under the curve; AKIN,AcuteKidney InjuryNetwork; RRT, renal replacement therapy; RR, relative risk; ICU, intensive careunit; OR, odds ratio; NRI, net reclassification index.aGrade 1: none (no casts or renal tubule epithelial [RTE] cells); grade 2: at least 1 cast or RTE cell seen but ,10% of low-power field(LPFs); grade 3: many casts and RTE cells seen on.10% but,90% of LPFs; grade 4: sheets of muddy brown casts, casts, and RTE cellsseen on .90% of LPFs.b0 points: 0 casts or 0 RTE cells; 1 point each: 1–5 casts per LPF or 1–5 RTE cells per high-power field (HPF); 2 points each:$6 casts perLPF or $6 RTE cells per HPF.c0 points: 0 casts or 0 RTE cells; 1 point each: 1 cast or 1 RTE cell per HPF 2 points each: 2–4 casts or RTE cells per HPF; 3 points each:$5casts or RTE cells per HPF.

170 Clinical Journal of the American Society of Nephrology

with that of new biomarkers is required to provide a de-finitive answer.

Traditional versus Novel Urine BiomarkersKoyner and colleagues noted the lack of utility of FeNa

and FeUrea in early detection and clinical prognosis ofAKI in patients undergoing cardiac surgery comparedwith several novel biomarkers (22). The first study to com-pare traditional urinary biomarkers (including urine mi-croscopy) with novel biomarkers was undertaken in 83intensive care unit patients with AKI by Bagshaw and col-leagues (23). Their prospective, two-center cohort study ex-amined the utility of a urine microscopy score (UMS), urinechemistries, and plasma/urine neutrophil gelatinase-associated lipocalin (NGAL) in predicting worsening kid-ney function, dialysis need, and death in septic andnonseptic patients with AKI. The UMS was higher in septicpatients with AKI, correlated with urine NGAL (r=0.41;

P=0.012), and predicted worsening AKI (UMS, 0 versus$3; adjusted odds ratio, 8.0) by increased RIFLE criteria.UMS was associated with greater likelihood of dialysis re-quirement and crude hospital death. A UMS $3 had thefollowing characteristics for detecting “worsening AKI”:sensitivity, 0.67; specificity, 0.95; positive predictive value,0.80; and negative predictive value, 0.91. Urine chemistries(urine sodium, FeNa, FeUrea) were not associated withthe clinical outcomes. Thus, urine microscopy andurine NGAL correlate fairly well and are complementaryin predicting worsening AKI in intensive care unit patients(Table 3).Most recently, Hall and coworkers undertook a prospec-

tive cohort study to evaluate traditional urine biomarkersand novel urine biomarkers in hospitalized patients whodeveloped early AKI according to the AKIN criteria (24).They tested the utility of these biomarkers in predictingseveral clinical outcomes as well as differentiating variousforms of AKI. After exclusion, 249 patients were enrolledon the first day of meeting AKI criteria, more than halfwere older than 65 years of age, and nearly 50% werein the intensive care unit; the mean baseline GFR was69630 ml/min per 1.73 m2. The causes of AKI were asfollows: prerenal AKI (66%), ATN (20%), and “other”(14%). Seventy-two patients (29%) met the primary com-posite outcome of “worsened AKI or in-hospital death.”The adjusted risk for the primary outcome was approxi-mately threefold higher in those with upper than in thosewith lower values of urine NGAL, kidney injury molecule-1,IL-18, and microscopy score (Figure 4). Secondary out-comes, such as higher AKI stage, dialysis requirement,nephrology consultation, and death, were associated withhigher urine biomarker quartiles or microscopy score. FeNaand FeUrea were not useful for differential diagnosis orpredicting outcomes. Importantly, risk classification ofAKI (determined by the net reclassification index and in-tegrated discrimination improvement) was significantlyimproved after the novel biomarkers or urine microscopymeasures were added to standard clinical variables. Thesedata suggest that novel urine biomarkers and urine micros-copy are useful to differentiate early hospital-acquired AKIand improve upon the baseline clinical determination ofprognosis (Table 3).

Moving ForwardSo what guidance can we recommend to clinicians who

care for patients who develop AKI while hospitalized? Fordiagnosis, FeNa and FeUrea must be used in an evidence-based fashion—knowing the limitations of the tests andthe clinical scenarios in which the tests operate well andwhere they fail. With regard to urine microscopy, using aquantitative approach to urine sediment findings (num-bers of cells or casts per low- or high-power field) willimprove diagnostic accuracy, not only for glomerulardiseases but also for prerenal AKI and ATN. Sometimescombining urine chemistries and quantitative urine mi-croscopy will provide better diagnosis and distinction ofprerenal AKI from ATN. Moreover, urine microscopyalso adds prognostic information, which can help the cli-nician plan ahead for the patient (dialysis preparation,family counseling). Although these traditional tests are

Figure 2. | A urine sediment score of 0–4 is used to quantitativelyevaluate AKI. A score of 2 is achieved in (A) $6 renal tubular epi-thelial cells/high-power field, and a score of 2 is achieved in (B) $6granular casts/low-power field.

Clin J Am Soc Nephrol 7: 167–174, January, 2012 Traditional Biomarkers in the Assessment of AKI, Perazella and Coca 171

inexpensive and readily available, their limitations mayprovide a niche for novel biomarkers, which may provideuseful point-of-care information about AKI diagnosis andprognosis in the future.Because AKI outcomes have not measurably improved

over the past several years, and the number of CKD casesafter AKI may exceed .100,000 per year (25–28), wemust intensify efforts to improve outcomes. To accom-plish this, data from observational studies should beincorporated into the design of randomized controlledtrials of AKI therapies. Because therapies are more likelyto demonstrate benefit only for patients who experiencepersistent AKI (i.e., ATN), urine microscopy should beused as an inclusion criterion for trial enrollment. Thiswould allow only patients who have a high likelihood of

progressing to more severe AKI, such as those withATN, to enter studies. Perhaps this approach will permitthe nephrology community to identify beneficial thera-pies for AKI.In conclusion, on the basis of the current information

on urine diagnostics, it appears that urine microscopyand new urine biomarkers may be used together to informon early AKI, differentiate prerenal AKI from ATN, andpredict such outcomes as worsened AKI, dialysis re-quirement, and death. Urine chemistries may providesome insight into differential diagnosis when used inselect patients but have little role in most hospitalizedpatients with AKI. Figure 5 shows one possible approachfor urine diagnostics in patients with hospital-acquiredAKI.

Figure 3. | Urine sediment score and individual components of the composite outcome. AKIN, Acute Kidney Injury Network (17).

Figure 4. | Urine biomarker groups (neutrophil gelatinase-associated lipocalin, IL-18, kidney injury molecule (KIM-1) in quartiles and urinemicroscopy score of 0, 1, 2, ‡3) and primary outcome occurrence (worsened AKI or in-hospital death) (24).

172 Clinical Journal of the American Society of Nephrology

References1. HsuCY,McCullochCE, FanD,Ordo~nez JD, ChertowGM,GoAS:

Community-based incidence of acute renal failure. Kidney Int72: 208–212, 2007

2. Okusa MD, Chertow GM, Portilla D Acute Kidney Injury Advi-sory Group of the American Society of Nephrology: The nexus ofacute kidney injury, chronic kidney disease, and World KidneyDay 2009. Clin J Am Soc Nephrol 4: 520–522, 2009

3. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P; AcuteDialysis Quality Initiative Workgroup: Acute renal failure -definition, outcome measures, animal models, fluid therapy andinformation technology needs: the second international Con-sensusConference of the AcuteDialysisQuality Initiative (ADQI)Group. Crit Care 8: R205–R212, 2004

4. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, WarnockDG, Levin A Acute Kidney Injury Network: Acute Kidney InjuryNetwork: Report of an initiative to improve outcomes in acutekidney injury. Crit Care 11: R31, 2007

5. Espinel CH: The FENa test. Use in the differential diagnosis ofacute renal failure. JAMA 236: 579–581, 1976

6. Espinel CH, Gregory AW: Differential diagnosis of acute renalfailure. Clin Nephrol 13: 73–77, 1980

7. Miller TR, Anderson RJ, Linas SL, HenrichWL, Berns AS, GabowPA, Schrier RW: Urinary diagnostic indices in acute renal failure:A prospective study. Ann Intern Med 89: 47–50, 1978

8. Pru C, Kjellstrand C: Urinary indices and chemistries in the dif-ferential diagnosis of prerenal failure and acute tubular necrosis.Semin Nephrol 5: 224–233, 1985

9. Kaplan AA, Kohn OF: Fractional excretion of urea as a guide torenal dysfunction. Am J Nephrol 12: 49–54, 1992

10. Carvounis CP, Nisar S, Guro-Razuman S: Significance of thefractional excretion of urea in the differential diagnosis of acuterenal failure. Kidney Int 62: 2223–2229, 2002

11. Pepin MN, Bouchard J, Legault L, Ethier J: Diagnostic perfor-mance of fractional excretion of urea and fractional excretion ofsodium in the evaluations of patients with acute kidney injurywith or without diuretic treatment.Am J KidneyDis 50: 566–573,2007

12. Diskin CJ, Stokes TJ, Dansby LM, Radcliff L, Carter TB: Towardthe optimal clinical use of the fraction excretion of solutes inoliguric azotemia. Ren Fail 32: 1245–1254, 2010

13. Graber M, Lane B, Lamia R, Pastoriza-Munoz E: Bubble cells:renal tubular cells in the urinary sediment with characteristics ofviability. J Am Soc Nephrol 1: 999–1004, 1991

14. Perazella MA, Coca SG, Kanbay M, Brewster UC, Parikh CR:Diagnostic value of urine microscopy for differential diagnosis ofacute kidney injury in hospitalized patients. Clin J Am SocNephrol 3: 1615–1619, 2008

15. Marcussen N, Schumann J, Campbell P, Kjellstrand C: Cyto-diagnostic urinalysis is very useful in the differential diagnosisof acute renal failure and can predict the severity. Ren Fail 17:721–729, 1995

16. Chawla LS, Dommu A, Berger A, Shih S, Patel SS: Urinary sedi-ment cast scoring index for acute kidney injury: A pilot study.Nephron Clin Pract 110: c145–c150, 2008

17. Perazella MA, Coca SG, Hall IE, Iyanam U, Koraishy M, ParikhCR: Urine microscopy is associated with severity and worseningof acute kidney injury in hospitalized patients. Clin J Am SocNephrol 5: 402–408, 2010

18. Han WK, Bailly V, Abichandani R, Thadhani R, Bonventre JV:Kidney Injury Molecule-1 (KIM-1): A novel biomarker forhuman renal proximal tubule injury. Kidney Int 62: 237–244, 2002

19. Mishra J,MaQ, PradaA,MitsnefesM, Zahedi K, Yang J, Barasch J,Devarajan P: Identification of neutrophil gelatinase-associatedlipocalin as a novel early urinary biomarker for ischemic renalinjury. J Am Soc Nephrol 14: 2534–2543, 2003

20. Mishra J, Mori K, Ma Q, Kelly C, Barasch J, Devarajan P: Neu-trophil gelatinase-associated lipocalin: A novel early urinarybiomarker for cisplatin nephrotoxicity. Am J Nephrol 24: 307–315, 2004

21. Siew ED,Ware LB, Ikizler TA: Biological markers of acute kidneyinjury. J Am Soc Nephrol 22: 810–820, 2011

22. Koyner JL, VaidyaVS, BennettMR,MaQ,Worcester E, Akhter SA,Raman J, Jeevanandam V, O’Connor MF, Devarajan P, BonventreJV, Murray PT: Urinary biomarkers in the clinical prognosis andearly detection of acute kidney injury. Clin J Am Soc Nephrol 5:2154–2165, 2010

23. Bagshaw SM, Haase M, Haase-Fielitz A, Bennett M, DevarajanP,, Bellomo R: A prospective evaluation of urine microscopy inseptic and non-septic acute kidney injury [published onlineahead of print June 9, 2011]. Nephrol Dial Transplant doi:10.1093/ndt/gfr331

24. Hall IE, Coca SG, Perazella MA, Eko UU, Luciano RL, Peter PR,Han WK, Parikh CR: Risk of poor outcomes with novel and tra-ditional biomarkers at clinical AKI diagnosis [published onlineahead of print October 27, 2011]. Clin J Am Soc Nephrol 2011doi:10.2215/CJN.04960511

Figure 5. | Potential approach to AKI using traditional and novel biomarkers. FeNA, fractional excretion of sodium; FeUrea, fractional ex-cretion of urea; RRT, renal replacement therapy.

Clin J Am Soc Nephrol 7: 167–174, January, 2012 Traditional Biomarkers in the Assessment of AKI, Perazella and Coca 173

25. Hsu CY: Linking the population epidemiology of acuterenal failure, chronic kidney disease and end-stage renaldisease. Curr Opin Nephrol Hypertens 16: 221–226,2007

26. Ishani A,NelsonD,Clothier B, Schult T,Nugent S,GreerN, SlininY, Ensrud KE: The magnitude of acute serum creatinine increaseafter cardiac surgery and the risk of chronic kidney disease,progression of kidney disease, and death. Arch Intern Med 171:226–233, 2011

27. Ishani A, Xue JL, Himmelfarb J, Eggers PW, Kimmel PL, MolitorisBA, Collins AJ: Acute kidney injury increases risk of ESRD amongelderly. J Am Soc Nephrol 20: 223–228, 2009

28. Lo LJ, Go AS, Chertow GM, McCulloch CE, Fan D, Ordonez JD,,Hsu CY: Dialysis-requiring acute renal failure increases the risk ofprogressive chronic kidney disease. Kidney Int 76:893–899, 2009.

Published online ahead of print. Publication date available at www.cjasn.org.

174 Clinical Journal of the American Society of Nephrology