Acute Kidney Injury in Liver Disease: Role of Biomarkers

Acute kidney injury (AKI) is a common complication in patients with advanced cirrhosis and is associated with significant mortality. The most common etiologies of AKI in this setting are prerenal azotemia, acute tubular necrosis, and hepatorenal syndrome. Despite the overall poor outcomes of patients with cirrhosis and AKI, potentially efficacious therapies exist but must be tailored to the specific AKI etiology. Unfortunately, determining the etiology of AKI in the setting of cirrhosis is notoriously difficult. Many of the standard diagnostic tools, such as urine microscopy and the fractional excretion of sodium, have traditionally been ineffective. Novel biomarkers of kidney tubular injury may be able to assist with differential diagnosis and the appropriate targeting of treatments by distinguishing structural from functional causes of AKI. In recent studies, both urinary neutrophil gelatinase-associated lipocalin and interleukin-18 have shown the ability to distinguish hepatorenal syndrome from prerenal azotemia and acute tubular necrosis. In addition, multiple biomarkers, including neutrophil gelatinase-associated lipocalin and interleukin-18, have demonstrated the ability to independently predict both progression of AKI and mortality. Critically, recent research also indicated that commonly available tests, fractional excretion of sodium and proteinuria, may also be able to distinguish etiologies of AKI in cirrhosis, but diagnostic cutoffs must be re-conceptualized specifically to this unique AKI setting. Acute kidney injury (AKI) is a common complication in patients with advanced cirrhosis and is associated with significant mortality. The most common etiologies of AKI in this setting are prerenal azotemia, acute tubular necrosis, and hepatorenal syndrome. Despite the overall poor outcomes of patients with cirrhosis and AKI, potentially efficacious therapies exist but must be tailored to the specific AKI etiology. Unfortunately, determining the etiology of AKI in the setting of cirrhosis is notoriously difficult. Many of the standard diagnostic tools, such as urine microscopy and the fractional excretion of sodium, have traditionally been ineffective. Novel biomarkers of kidney tubular injury may be able to assist with differential diagnosis and the appropriate targeting of treatments by distinguishing structural from functional causes of AKI. In recent studies, both urinary neutrophil gelatinase-associated lipocalin and interleukin-18 have shown the ability to distinguish hepatorenal syndrome from prerenal azotemia and acute tubular necrosis. In addition, multiple biomarkers, including neutrophil gelatinase-associated lipocalin and interleukin-18, have demonstrated the ability to independently predict both progression of AKI and mortality. Critically, recent research also indicated that commonly available tests, fractional excretion of sodium and proteinuria, may also be able to distinguish etiologies of AKI in cirrhosis, but diagnostic cutoffs must be re-conceptualized specifically to this unique AKI setting. Clinical Summary•Differential diagnosis and prognosis are extremely challenging in patients with cirrhosis and acute kidney injury (AKI).•Biomarkers reflecting structural kidney injury have shown the ability to differentiate acute tubular necrosis and hepatorenal syndrome.•Fractional excretion of sodium may also be able to distinguish acute tubular necrosis and hepatorenal syndrome but requires a re-conceptualizing of diagnostic cutoffs.•Kidney biomarkers may be able to predict progression and death in patients with AKI and may be able to diagnose AKI earlier than change in creatinine or estimated glomerular filtration rate. •Differential diagnosis and prognosis are extremely challenging in patients with cirrhosis and acute kidney injury (AKI).•Biomarkers reflecting structural kidney injury have shown the ability to differentiate acute tubular necrosis and hepatorenal syndrome.•Fractional excretion of sodium may also be able to distinguish acute tubular necrosis and hepatorenal syndrome but requires a re-conceptualizing of diagnostic cutoffs.•Kidney biomarkers may be able to predict progression and death in patients with AKI and may be able to diagnose AKI earlier than change in creatinine or estimated glomerular filtration rate. Acute kidney injury (AKI) is a common and devastating complication in patients with cirrhosis, occurring in an estimated 19% of hospitalizations,1Garcia-Tsao G. Parikh C.R. Viola A. Acute kidney injury in cirrhosis.Hepatology. 2008; 48: 2064-2077Crossref PubMed Scopus (474) Google Scholar and is associated with significant mortality, 55% to 91%.2Du Cheyron D. Bouchet B. Parienti J.J. Ramakers M. Carbonneau P. The attributable mortality of acute renal failure in critically ill patients with liver cirrhosis.Intensive Care Med. 2005; 31: 1693-1699Crossref PubMed Scopus (149) Google Scholar, 3Fang J.T. Tsai M.H. Tian Y.C. et al.Outcome predictors and new score of critically ill cirrhotic patients with acute renal failure.Nephrol Dial Transplant. 2008; 23: 1961-1969Crossref PubMed Scopus (57) Google Scholar, 4Cholongitas E. Senzolo M. Patch D. Shaw S. O'Beirne J. Burroughs A.K. Cirrhotics admitted to intensive care unit: the impact of acute renal failure on mortality.Eur J Gastroenterol Hepatol. 2009; 21: 744-750Crossref PubMed Scopus (81) Google Scholar The clinical impact of this grave confluence of illnesses will continue to worsen as the incidence of both AKI and cirrhosis are increasing.5Hsu C.Y. McCulloch C.E. Fan D. Ordonez J.D. Chertow G.M. Go A.S. Community-based incidence of acute renal failure.Kidney Int. 2007; 72: 208-212Crossref PubMed Scopus (450) Google Scholar The impact of AKI on mortality is not homogeneous but instead contingent on the etiology of AKI.6Martin-Llahi M. Guevara M. Torre A. et al.Prognostic importance of the cause of renal failure in patients with cirrhosis.Gastroenterology. 2010; 140: 488-496Abstract Full Text Full Text PDF PubMed Scopus (273) Google Scholar The primary causes of AKI in patients with cirrhosis are prerenal azotemia (PRA), acute tubular necrosis (ATN), and hepatorenal syndrome (HRS). The prevalence of these diagnoses among patients with cirrhosis and AKI is shown in Figure 1.3Fang J.T. Tsai M.H. Tian Y.C. et al.Outcome predictors and new score of critically ill cirrhotic patients with acute renal failure.Nephrol Dial Transplant. 2008; 23: 1961-1969Crossref PubMed Scopus (57) Google Scholar, 7Hampel H. Bynum G.D. Zamora E. El-Serag H.B. Risk factors for the development of renal dysfunction in hospitalized patients with cirrhosis.Am J Gastroenterol. 2001; 96: 2206-2210Crossref PubMed Google Scholar, 8Peron J.M. Bureau C. Gonzalez L. et al.Treatment of hepatorenal syndrome as defined by the international ascites club by albumin and furosemide infusion according to the central venous pressure: a prospective pilot study.Am J Gastroenterol. 2005; 100: 2702-2707Crossref PubMed Scopus (51) Google Scholar, 9Terra C. Guevara M. Torre A. et al.Renal failure in patients with cirrhosis and sepsis unrelated to spontaneous bacterial peritonitis: value of MELD score.Gastroenterology. 2005; 129: 1944-1953Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar, 10Du Cheyron D. Bouchet B. Parienti J.J. Ramakers M. Charbonneau P. The attributable mortality of acute renal failure in critically ill patients with liver cirrhosis.Intensive Care Med. 2005; 31: 1693-1699Crossref PubMed Scopus (75) Google Scholar, 11Wu C.C. Yeung L.K. Tsai W.S. et al.Incidence and factors predictive of acute renal failure in patients with advanced liver cirrhosis.Clin Nephrol. 2006; 65: 28-33Crossref PubMed Scopus (89) Google Scholar, 12Moreau R. Durand F. Poynard T. et al.Terlipressin in patients with cirrhosis and type 1 hepatorenal syndrome: a retrospective multicenter study.Gastroenterology. 2002; 122: 923-930Abstract Full Text Full Text PDF PubMed Scopus (419) Google Scholar Of these, HRS portends the worst prognosis, with type 1 associated with a median untreated survival of 2 weeks. HRS develops in patients with cirrhosis when portal hypertension stimulates an abundance of vasodilatory factors, leading to increased splanchnic vasodilatation. As systemic vascular resistance begins to markedly decline, various pathways are activated including the renin-angiotensin system, sympathetic nervous system, and arginine vasopressin, resulting in profound kidney vasoconstriction and hypoperfusion. When vasoconstriction is sufficiently advanced, kidney hypoperfusion is no longer reversible with volume resuscitation and patients experience the progressive and unrelenting decline in kidney function characteristic of HRS. AKI in such patients is, therefore, primarily functional in nature. Despite this grim outlook, the potential exists in for the treatment of AKI in cirrhosis if efficacious interventions are applied to correctly phenotyped patients. Arriving at an accurate diagnosis is imperative because these treatments vary greatly, entail significant expense, use scarce resources, and have potentially significant toxicity. Patients with cirrhosis and PRA require fluids, but the deleterious consequences of overzealous fluid administration, as occurs when ATN is misdiagnosed, are increasingly recognized.13Schrier R.W. Fluid administration in critically ill patients with acute kidney injury.Clin J Am Soc Nephrol. 2010; 5: 733-739Crossref PubMed Scopus (78) Google Scholar In spite of the severity of kidney dysfunction, kidneys in patients with HRS are primarily structurally intact. Kidney function in this setting, therefore, can be markedly improved if kidney blood flow is restored. Terlipressin, a V1-vasoconstrictor that acts to augment systemic circulating volume and, thereby, improve kidney perfusion has shown great promise for the treatment of HRS.14Dobre M. Demirjian S. Sehgal A.R. Navaneethan S.D. Terlipressin in hepatorenal syndrome: a systematic review and meta-analysis.Int Urol Nephrol. 2011; 43: 175-184Crossref PubMed Scopus (54) Google Scholar In addition, in patients with advanced cirrhosis, liver transplantation can restore systemic vascular resistance, mitigate systemic and kidney vasoconstriction, and restore normal kidney hemodynamics. Patients with HRS at the time of liver transplantation can, thus, experience rapid improvement in kidney function post-transplant.15Nadim M.K. Genyk Y.S. Tokin C. et al.Impact of etiology of acute kidney injury on outcomes following liver transplantation: acute tubular necrosis versus hepatorenal syndrome.Liver Transpl. 2012; 18: 539-548Crossref PubMed Scopus (138) Google Scholar Patients with ATN should be dialyzed if clinically indicated, but in such patients with frank structural injury, interventions to restore kidney perfusion do not result in resolution of AKI and application of vasoconstrictors or liver transplantation is, therefore, inappropriate. Finally, patients with ATN must be differentiated from patients with HRS when considering a combined liver/kidney transplant. Unfortunately, current diagnostic strategies are often unable to distinguish between functional and structural injury. As a result, the potential exists for misallocation of scarce resources and potentially harmful unnecessary treatments. Presently, attempts to distinguish PRA, HRS, and ATN begin (outside assessing the clinical context) with withholding diuretics and volume resuscitating the patient with albumin at 1 g/kg/d for 2 days. If the AKI resolves, they are considered to have had PRA. The chief difficulty is in distinguishing HRS from ATN. The primary indicator of AKI, creatinine, is a marker of filtration and, therefore, detects declines in kidney function but cannot determine whether such a decline is because of hypoperfusion or to structural injury. Many tests typically used to identify ATN in the general population are ineffective in the setting of cirrhosis. The fractional excretion of sodium (FENa), although ubiquitously applied by nephrologists evaluating AKI, has historically been difficult to interpret in patients with cirrhosis. Cirrhotic patients frequently present with low urine sodium irrespective of AKI16Magri P. Auletta M. Andreucci M. et al.Sodium retention in preascitic stage of cirrhosis.Semin Nephrol. 2001; 21: 317-322Abstract Full Text PDF PubMed Google Scholar because of extreme renal sodium avidity and even ATN can present with an FENa less than 1%.17Diamond J.R. Yoburn D.C. Nonoligouric acute renal failure associated with a low fractional excretion of sodium.Ann Intern Med. 1982; 96: 597-600Crossref PubMed Scopus (75) Google Scholar The traditional dichotomy where an FENa less than 1% indicates hypoperfusion and more than 1% to 2% signifies tubular dysfunction, and ATN is, therefore, inapplicable, and the test is historically not typically used in cirrhotic patients. Similarly, urine microscopy is potentially helpful in the differential diagnosis of AKI but can be complicated in cirrhosis by biliary staining of sediment and has not been rigorously evaluated in this setting. The gold standard for diagnosing AKI, kidney biopsy, is rarely performed in patients with advanced cirrhosis for fear of bleeding complications. In lieu of these traditional tests, the international ascites club (IAC) attempted to standardize the diagnosis of HRS by establishing 6 clinical criteria.18Salerno F. Gerbes A. Gines P. et al.Diagnosis, prevention and treatment of hepatorenal syndrome in cirrhosis.Gut. 2007; 56: 1310-1318Crossref PubMed Scopus (84) Google Scholar Once PRA has been ruled out by failure to improve withholding of diuretics and albumin resuscitation, those patients meeting 6/6 IAC criteria have been considered to have HRS, whereas those who do not are assumed to have ATN (barring signs consistent with a glomerulonephritis). Unfortunately, these criteria lack specificity as patients with ATN often (1) present with ascites, (2) have creatinine more than 1.5 mg/dL, (3) do not respond to volume resuscitation, (4) lack significant proteinuria or hematuria, and (5) have no gross structural changes to the kidney. Although ATN can certainly be associated with shock, (6) ischemic ATN can develop in the absence of shock and, indeed, frequently occurs in the setting of ostensibly normal blood pressure.19Abuelo J.G. Normotensive ischemic acute renal failure.N Engl J Med. 2007; 357: 797-805Crossref PubMed Google Scholar In addition, the degree of creatinine elevation does not distinguish ATN from HRS.20Verna E.C. Brown R.S. Farrand E. et al.Urinary neutrophil gelatinase-associated lipocalin predicts mortality and identifies acute kidney injury in cirrhosis.Dig Dis Sci. 2012; 57: 2362-2370Crossref PubMed Scopus (126) Google Scholar Very recently, the IAC have proposed a new definition of AKI in cirrhosis based on the adaptation of a modification of the Kidney Disease Improving Global Outcome criteria.21Kidney Disease Improving Global outcomes (KDIGO) acute kidney injury work group. KDIGO clinical practice guideline for acute kidney injury.Kidney Int. 2012; 2: 1-138Google Scholar As long as they meet these criteria (an increase in creatinine of ≥0.3 mg/dL within 48 hours or an increase of ≥50% from baseline within 7 days) for having AKI and fulfill the other 5 traditional IAC criteria for HRS, there is no longer a fixed creatinine threshold that patients much cross before being diagnosed with HRS.22Angeli P. Ginés P. Wong F. et al.Diagnosis and management of acute kidney injury in patients with cirrhosis: revised consensus recommendations of the International Ascites Club.J Hepatol. 2015; 62: 968-974Abstract Full Text Full Text PDF PubMed Scopus (524) Google Scholar In addition, as long as patients present with Stage 2 or 3 AKI or progress from one stage to a higher stage, the creatinine no longer must be more than 2.5 mg/dL before treatment with vasoconstrictive agents is indicated. These extremely welcome changes will enhance sensitivity and facilitate timely treatment of HRS. However, the lack of specificity inherent to a diagnosis centered on creatinine will remain problematic. The critical diagnostic shortcoming is that serum creatinine is a marker of kidney filtration, not injury, and, thus, cannot distinguish functional from structural etiologies of AKI. More discriminating tests are urgently needed to make this distinction to guide the allocation of potent and scarce treatments and to help predict progression of AKI and mortality. Nearly 30 biomarkers of kidney tubular injury have recently been investigated for early detection, differential diagnosis, and prognosis of AKI.23Belcher J.M. Edelstein C.L. Parikh C.R. Clinical applications of biomarkers for acute kidney injury.Am J Kidney Dis. 2011; 57: 930-940Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar Such biomarkers reflect frank structural injury and, thus, should appear in consort with an acute drop in glomerular filtration rate (GFR) attributable to structural damage. Among the most promising are interleukin-18 (IL-18), kidney injury molecule-1 (KIM-1), liver-type fatty acid-binding protein (L-FABP), and neutrophil gelatinase-associated lipocalin (NGAL). These biomarkers appear and peak in the urine at different times after injury ranging from NGAL at 2 hours to KIM-1 at 12 hours.24Krawczeski C.D. Goldstein S.L. Woo J. et al.Temporal relationship and predictive value of urinary acute kidney injury biomarkers after pediatric cardiopulmonary bypass.J Am Coll Cardiol. 2011; 58: 2301-2309Abstract Full Text Full Text PDF PubMed Scopus (277) Google Scholar Such differences in expression patterns not only suggest the potential utility in a biomarker panel but also must be kept in mind as a possible limitation when interpreting experimental results. Injury biomarkers have recently been successfully investigated in multiple clinical settings including cardiac surgery,25Parikh C.R. Coca S.G. Thiessen-Philbrook H. et al.Postoperative biomarkers predict acute kidney injury and poor outcomes after adult cardiac surgery.J Am Soc Nephrol. 2011; 22: 1748-1757Crossref PubMed Scopus (438) Google Scholar intensive care units,26Nisula S. Yang R. Poukkanen M. et al.Predictive value of urine interleukin-18 in the evaluation and outcome of acute kidney injury in critically ill adult patients.Br J Anaesth. 2015; 114: 460-468Crossref PubMed Scopus (52) Google Scholar contrast administration,27Tasanarong A. Hutayanon P. Piyayotai D. Urinary neutrophil gelatinase-associated lipocalin predicts the severity of contrast-induced acute kidney injury in chronic kidney disease patients undergoing elective coronary procedures.BMC Nephrol. 2013; 5: 270Crossref Scopus (36) Google Scholar kidney transplant,28Hall I.E. Yarlagadda S.G. Coca S.G. et al.IL-18 and urinary NGAL predict dialysis and graft recovery after kidney transplantation.J Am Soc Nephrol. 2010; 21: 189-197Crossref PubMed Scopus (269) Google Scholar and general hospital wards.29Hall I.E. Coca S.G. Perazella M.A. et al.Risk of poor outcomes with novel and traditional biomarkers at clinical diagnosis.Clin J Am Soc Nephrol. 2011; 6: 2740-2749Crossref PubMed Scopus (93) Google Scholar Given the tremendous physiological difference between functional and structural AKI in cirrhosis and the impact this distinction has on the potential for successful treatment, injury biomarkers would seem particularly well suited for use in this setting. The critical need for research in this area was recognized by the study group on HRS at the Eighth International Consensus Conference of the Acute Dialysis Quality Initiative Group that listed the evaluation of injury biomarkers for the study of AKI in cirrhosis as one of their primary targeted areas for further research.30Nadim M.K. Kellum J.A. Davenport A. et al.Hepatorenal syndrome: the 8th international consensus conference of the Acute Dialysis Quality Initiative (ADQI) group.Crit Care. 2012; 16: R23Crossref PubMed Scopus (129) Google Scholar In light of this need, several authors have begun to investigate the role of biomarkers both for differential diagnosis and prognosis of AKI in cirrhosis. In one of the first studies investigating the potential for biomarkers for differential diagnosis in cirrhosis, Verna and colleagues20Verna E.C. Brown R.S. Farrand E. et al.Urinary neutrophil gelatinase-associated lipocalin predicts mortality and identifies acute kidney injury in cirrhosis.Dig Dis Sci. 2012; 57: 2362-2370Crossref PubMed Scopus (126) Google Scholar assessed 118 cirrhotic patients with urine samples collected in the emergency room and evaluated the utility of NGAL for the differential diagnosis of their kidney status and risk stratification for mortality. Based on prespecified clinical criteria, 52 (44%) patients had normal kidney function, 14 (12%) stable chronic kidney disease, 17 (14%) PRA, 20 (17%) HRS, and 15 (13%) "intrinsic acute kidney injury" which referred specifically to ATN. Patients with HRS had urine NGAL levels intermediate between PRA (median [interquartile range specified] 105 ng/mL [27.5-387.5] vs 20 ng/mL [15-45], P = .004) and ATN (325 ng/mL [100-700], P < .001). Serum creatinine, however, did not differ between ATN and HRS. Fifteen (13%) patients died. In adjusted analysis, NGAL at a cutoff of 110 ng/mL (odds ratio, 6.05; 95% confidence interval, 1.35-27.2) and diagnosis of HRS (odds ratio, 6.71; 95% confidence interval, 1.76-25.5) independently predicted mortality. Fagundes and others31Fagundes C. Pépin M.N. Guevara M. et al.Urinary neutrophil gelatinase-associated lipocalin as biomarker in the differential diagnosis of impairment of kidney function in cirrhosis.J Hepatol. 2012; 57: 267-273Abstract Full Text Full Text PDF PubMed Scopus (166) Google Scholar studied urinary NGAL in 84 patients with kidney dysfunction, defined as creatinine greater than 1.5 mg/dL. PRA was diagnosed in 16 (19%), CKD in 14 (17%), HRS in 33 (39%), and ATN in 11 (13%). Patients with ATN had NGAL levels significantly higher than seen in PRA, 417 ng/mL (239-2242) vs 30 ng/mL (20-59). HRS was once again intermediate, 76 ng/mL (43-263), and differed significantly from both ATN and PRA. NGAL levels in patients with PRA were similar to cirrhotic patients without kidney dysfunction (n = 157), showing biomarker specificity for structural AKI. Although patients with any form of kidney impairment had higher plasma NGAL than those without, plasma values could not be distinguished among PRA, HRS, ATN, and CKD. Qasem and colleagues32Qasem A.A. Farag S.E. Hamed E. et al.Urinary biomarkers of acute kidney injury in patients with cirrhosis.ISRN Nephrol. 2014; : 376795Google Scholar evaluated the utility of both NGAL and IL-18 in 150 patients with cirrhosis admitted to ICU. Values of both biomarkers in patients with HRS (n = 14) were found to be significantly higher than in those with PRA (n = 17) and significantly lower than in those with ATN (n = 22). NGAL and IL-18 demonstrated an excellent ability to discriminate ATN from HRS with area under the curves (AUCs) of 0.91 and 0.98, respectively. However, although FENa differed significantly among PRA, HRS, and ATN (0.54%, 0.15%, and 4.05%, respectively), an AUC for FENa was not calculated. Patients in the study were diagnosed with PRA and ATN through prespecified clinical criteria and HRS through IAC criteria. Moving beyond the more commonly studied biomarkers, Shah and others33Shah N. Mohamed F.E. Jover-Cobos M. et al.Increased renal expression and urinary excretion of TLR4 in acute kidney injury associated with cirrhosis.Liver Int. 2013; 33: 398-409Crossref PubMed Scopus (86) Google Scholar evaluated KIM-1, α-glutathione S-transferase (αGST), π-glutathione S-transferase (πGST), and toll-like receptor 4 (TLR4) for their ability to identify cirrhotic patients with AKI and to distinguish etiologies. TLR4 can be activated by pathogens to increase nuclear factor kappa-β and cytokines. In patients with cirrhosis, infection and systemic inflammation play a critical role in the development of AKI. In addition to expression in immune cells, TLR4 is found in kidney parenchymal cells and its activation can lead to cellular apoptosis. Urinary TLR4 was significantly higher in patients determined to have kidney dysfunction from infection/inflammation than in those with HRS or in cirrhotics without kidney dysfunction. However, patient numbers were very small, and there was significant overlap in levels between patients with kidney dysfunction and those without. In the same study, urinary KIM-1, αGST, and πGST were significantly higher in patients with decompensated relative to compensated cirrhosis but did not differ between those with and without kidney dysfunction. The driving reason behind the need for new biomarkers is the lack of a safe diagnostic gold standard. Unfortunately, this lack of a gold standard is a significant challenge in the development of new tests as the gold standard against which they will be validated is necessarily suboptimal. Attempting to establish as valid as possible a gold standard without resorting to biopsy, Belcher and others34Belcher J.M. Sanyal A.J. Peixoto A.J. et al.Kidney biomarkers and differential diagnosis of patients with cirrhosis and acute kidney injury.Hepatology. 2014; 60: 622-632Crossref PubMed Scopus (238) Google Scholar used retrospective adjudication by a panel of nephrologists and hepatologists after the patients' death or discharge in study of 102 patients with cirrhosis and AKI whose creatinine either returned to within 25% of baseline within 48 hours of meeting Acute Kidney Injury Network (AKIN) criteria (n = 36) or whose AKI progressed to a higher stage after initially meeting criteria. Patients were adjudicated as PRA (n = 55), ATN (n = 19), and HRS (n = 16). Median values for NGAL, 565 (76-1000) vs 59 ng/mL (22-203); IL-18, 124 (15-325) vs 15 ng/mL (15-65); KIM-1, 8.4 (4.1-18.3) vs 5.1 ng/mL (2.1-10.7); L-FABP, 27 (8-103) vs 10 ng/mL (4-19); and albumin, 92 (44-253) vs 21 mg/dL (4-70) were significantly higher in patients with ATN compared with those without ATN (Fig 2). Comparing the 3 distinct diagnoses, all biomarkers were significantly elevated in ATN relative to PRA, but only NGAL, IL-18, and albumin were statistically higher in ATN compared with HRS and no injury markers distinguished PRA from HRS. However, FENa was lowest in patients with HRS, 0.10% (0.02-0.23), and differed significantly from both PRA, 0.27% (0.13-0.58), and ATN, 0.31% (0.12-0.65). The significantly lower FENa for ATN patients than seen in the studies of Qasem and Tsai (mentioned subsequently) may relate to the means of adjudication or to the inclusion of patients from throughout the hospital rather than only ICU (Qasem) or those with severe sepsis (Tsai). The AUCs and optimal cutoffs for the diagnosis of ATN for each biomarker are listed in Table 1. Using a panel of the top 4 biomarkers, relative risks (RRs) were calculated for the diagnosis of ATN based on number of biomarkers more than their optimal cutoffs (Table 2). Values of the most studied biomarker, NGAL, for patients with PRA, HRS, and ATN across studies are shown in Figure 3.Table 1Measures of Test Performance CharacteristicsReprinted with permission from Belcher et al.34Belcher J.M. Sanyal A.J. Peixoto A.J. et al.Kidney biomarkers and differential diagnosis of patients with cirrhosis and acute kidney injury.Hepatology. 2014; 60: 622-632Crossref PubMed Scopus (238) Google ScholarBiomarkerOptimal CutpointProportion Over Cutpoint With ATNAUC (95% CI)Validation AUC∗Validation AUCs derived from leave-10-out cross-validation performed with SAS Proc Surveyselect.Tubular injury markers NGAL (ng/mL)36525/35 (71%)0.78 (0.69-0.88)0.787 IL-18 (pg/mL)8521/33 (64%)0.71 (0.61-0.81)0.711 KIM-1 (ng/mL)15.415/24 (63%)0.64 (0.53-0.75)0.639 L-FABP (ng/mL)2521/30 (70%)0.69 (0.57-0.80)0.688Tubular function marker FENa (%)0.122/62 (35%)0.56 (0.45-0.68)0.563Glomerular injury marker Albumin (mg/dL)4429/52 (56%)0.73 (0.64-0.83)0.734Abbreviations: ATN, acute tubular necrosis; AUC, area under the curve; CI, confidence interval; NGAL, neutrophil gelatinase-associated lipocalin; IL-18, interleukin-18; KIM-1, kidney injury molecule-1; L-FABP, liver-type fatty acid-binding protein; FENa, fractional excretion of sodium.∗ Validation AUCs derived from leave-10-out cross-validation performed with SAS Proc Surveyselect. Open table in a new tab Table 2Association Between Biomarker Panel and the Diagnosis of ATNReprinted with permission from Belcher et al.34Belcher J.M. Sanyal A.J. Peixoto A.J. et al.Kidney biomarkers and differential diagnosis of patients with cirrhosis and acute kidney injury.Hepatology. 2014; 60: 622-632Crossref PubMed Scopus (238) Google ScholarNumber PositiveRelative Risk∗Unadjusted0 Markers Positive1.001 Marker Positive4.63 (1.29-16.61)2 Markers Positive6.98 (2.14-22.75)3 Markers Positive9.78 (3.10-30.86)4 Markers Positive13.33 (4.40-40.39)Abbreviations: ATN, acute tubular necrosis.Biomarker cutoffs: NGAL, 365 ng/mL; IL-18, 85 pg/mL; L-FABP, 25 ng/mL; Albumin 44 mg/dL.∗ Unadjusted Open table in a new tab Figure 3NGAL values by etiology of acute kidney injury. NGAL values for patients diagnosed with PRA, HRS, and ATN across studies are shown. Values are in microgram per gram creatinine with the exception of the data from Belcher and colleagues, which is in microgram per milliliter as indicated by an asterisk. Abbreviations: ATN, acute tubular necrosis; HRS, hepatorenal syndrome; NGAL, neutrophil gelatinase-associated lipocalin; PRA, prerenal azotemia.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Abbreviations: ATN