YEAR IN REVIEW (51 ± 20 ml/min/1.73 m2). Most importantly, the presence of C1q-fixing DSAs after trans plantation was associated with the lowest 5‑year graft survival (54% versus 93% and 94% for patients with non‑C1q fixing DSAs and no DSAs, respectively). Even after adjusting for mean fluorescence intensity, the hazard ratio for graft loss was 4.48 (95% CI 2.69–8.49) if C1q fixation was present. These findings suggest a need to assess C1q fixation in patients with ABMR. In 2013, the limitations of traditional histological approaches and the potential of new molecular techniques for the diag nosis of renal allograft rejection became apparent. It would be premature to replace histological assessments of biopsy samples with molecular techniques. However, the adjuncts reviewed here (molecular scoring, urinary mRNA analysis, C1q fixation assays and assessment of vascular involve ment) can more-finely resolve the diag nostic utility of histological assessment. These adjuncts further illuminate aspects of rejection that warrant future study (such as vascular lesions) and opportunities for noninvasive diagnosis. We hope that the new findings encourage thoughtful, refined clinical studies in which the use of adjunctive testing ultimately improves graft survival and patient outcomes. Transplantation Biology Research Center, Massachusetts General Hospital, Room 5102, Charlestown, MA 02129, USA (N. A. Zwang, L. A. Turka). Correspondence to: N. A. Zwang [email protected] Acknowledgements N. A. Zwang thanks the Brigham and Women’s Hospital/Massachusetts General Hospital Joint Nephrology Fellowship Program. Competing interests L. A. Turka declares an association with the following company: Novartis. See the article online for full details of the relationship. N. A. Zwang declares no competing interests. 1.
2.
3.
4.
Reeve, J. et al. Molecular diagnosis of T cellmediated rejection in human kidney transplant biopsies. Am. J. Transplant. 13, 645–655 (2013). Sellarés, J. et al. Molecular diagnosis of antibody-mediated rejection in human kidney transplants. Am. J. Transplant. 13, 971–983 (2013). Halloran, P. F. et al. Potential impact of microarray diagnosis of T cell-mediated rejection in kidney transplants: the INTERCOM study. Am. J. Transplant. 13, 2352–2363 (2013). Halloran, P. F. et al. Microarray diagnosis of antibody-mediated rejection in kidney transplant biopsies: an international prospective study (INTERCOM). Am. J. Transplant. 13, 2865–2874 (2013).
74 | FEBRUARY 2014 | VOLUME 10
5.
6.
7.
Furness, P. N. et al. International variation in histologic grading is large, and persistent feedback does not improve reproducibility. Am. J. Surg. Path. 27, 805–810 (2003). Mengel, M. et al. Banff initiative for quality assurance in transplantation (BIFQUIT): reproducibility of C4d immunohistochemistry in kidney allografts. Am. J. Transplant. 13, 1235–1245 (2013). Lefaucheur, C. et al. Antibody-mediated vascular rejection of kidney allografts:
a population-based study. Lancet 381, 313–319 (2013). 8. Suthanthiran, M. et al. Urinary-cell mRNA profile and acute cellular rejection in kidney allografts. N. Engl. J. Med. 369, 20–31 (2013). 9. Ingelfinger, J. R. & Alexander, S. I. One step closer to “Rejectostix.” N. Engl. J. Med. 369, 84–85 (2013). 10. Loupy, A. et al. Complement-binding anti-HLA antibodies and kidney-allograft survival. N. Engl. J. Med. 369, 1215–1226 (2013).
ACUTE KIDNEY INJURY IN 2013
Breaking barriers for biomarkers in AKI—progress at last Dinna N. Cruz and Ravindra L. Mehta
In 2013, four important papers were published that provide new insights on biomarkers in acute kidney injury (AKI). These studies demonstrate the potential for biomarkers to aid clinicians in improving the therapeutic management of patients with AKI and potentially improve patient outcomes. Cruz, D. N. & Mehta, R. L. Nat. Rev. Nephrol. 10, 74–76 (2014); published online 24 December 2013; doi:10.1038/nrneph.2013.268
Clinicians who treat patients with acute kidney injury (AKI) are faced with several questions that influence diagnostic and therapeutic decisions: has kidney injury occurred; what is the nature, sever ity and duration of kidney injury; has repair and recovery started; is intervention needed; and what is the prognosis? At present we glean the answers to these ques tions by evaluating sequential changes in the levels of urine output and serum cre atinine, but these evaluations are limited in their scope to guide therapeutic man agement. The emerging field of biomark ers specific to kidney injury holds promise for improving therapeutic management; however, the application of biomarkers in clinical practice has been limited. The past year has seen the publication of four papers that provide new insights in this field and bring us closer to implementing biomarkers in the clinical care of patients with AKI. Change in urine output could be viewed as a biomarker that is already available to the clinician,1 whereas novel biomarkers of AKI await further validation. Hourly urine output is a ‘biomarker’ that is measured 24 times per day in most intensive care units (ICUs), and is generally reported almost in real-time in electronic medical records. Mandelbaum et al. studied the empirical relationships between change in creatinine
concentration, change in urine output, the observation period over which these changes occurred, and clinical outcomes using a detailed ICU database for 14,526 patients. 2 For creatinine, the thresholds examined were an absolute increase of 8.84–88.4 μmol/l, or a relative increase of 125–400% from admission values, and observation periods were 1–7 days. For urine output, the thresholds were 0.1–1.0 ml/kg per h, for observation periods of 2–48 h. Mortality was high when there was a large absolute increase in creatinine levels regardless of the observation period, when the relative increase in creatini ne Key advances ■■ Urine output is emerging as a robust biomarker of acute kidney injury (AKI) associated with distinct outcomes in critically ill patients2 ■■ Combining biomarkers of tissue damage with those reflecting a functional change will enable clinicians to better characterize kidney health and the time points for specific interventions4 ■■ Biomarkers of tissue damage and cellcycle arrest pathways have been shown to predict increasing severity of AKI6 ■■ Identifying patients at high risk of developing ‘renal angina’ could potentially guide biomarker testing in the future9
www.nature.com/nrneph © 2014 Macmillan Publishers Limited. All rights reserved
YEAR IN REVIEW was high and the observation period was longer, and when oliguria was sustained for long periods of time. Overall, mortality increased rapidly as urine output decreased below 0.5 ml/kg per h, which lends support to urine output criteria used in current consensus definitions for AKI.3 The area under the receiver-operating characteristics curve (AUC) for urine output below 0.5 ml/ kg per h was >0.79 for mortality and >0.89 for renal replacement therapy across all the observation periods. The longer this level of urine output persisted the better was the discrimination for both outcomes. Several companies are now developing sensors for electronic monitoring of urine flow that will undoubtedly enhance clinicians’ ability to use this physiological biomarker to improve the management of AKI. However, a key caveat is to distinguish oliguria represent ing the normal autoregulatory response of the kidney formerly termed ‘pre-renal states’ from underlying damage. Different thresholds of change in urine output could improve the discrimination of these con ditions 1 and the Acute Dialysis Quality Initiative group has proposed consensus recommendations that provide a novel approach to biomarker utilization in AKI.4 Biomarkers specific to the kidney can be viewed as belonging to one of two broad classes representing functional changes (for example, serum creatinine, serum cys tatin C and urine output) or kidney damage (for example, proteinuria, urine and serum neutrophil gelatinase-associated lipo calin [NGAL], kidney injury molecule 1 [KIM-1] and liver-type fatty-acid binding protein [LFABP]). Consequently, by com bining biomarkers of functional change and tissue damage one can characterize any patient into one of four dynamic states of kidney health (Figure 1).4 A patient could transition from one state to another and back, depending on the nature, severity and duration of the injury. This approach permits the identification of a novel state of ‘subclinical’ AKI in which kidney tissue damage might occur independently of any functional change as measured by serum creatinine level and urine output. Alternatively, patients might be oliguric and have elevated serum creatinine levels in the absence of any alterations in biomarkers of tissue damage. For the practicing clini cian, this approach permits a more detailed assessment of the underlying condition of the kidney and sequential measurements could guide therap eutic interventions. Various biomarkers of functional change
and tissue damage could be used together for differential diagnosis and progno sis depending on the disease context, for example a combination of serum cystatin C and urine NGAL might be more informa tive in liver disease, whereas KIM-1 and serum creatinine level might be appropriate in patients with sepsis. Emerging evidence supports the concept of using biomarkers in this way; however, further validation is required.5 In the two-part multicenter Sapphire study, over 300 potential biomarkers in urine were examined in 522 critically ill patients during the discovery phase.6 The two top performing markers, insulin-like growth factor-binding protein 7 (IGFBP7) and tissue inhibitor of metalloprotein ases-2 (TIMP-2), were both inducers of G1 cell cycle arrest. After sepsis or ischae mic injury, renal tubular cells enter a brief period of cell cycle arrest, presumably to prevent potentially damaged cells from dividing. 7 In the validation phase of this study (involving 728 patients), these two biomarkers were evaluated for their ability to predict AKI (defined as Kidney Disease Improving Global Outcomes stage 2–3) within 12 h, as well as major adverse kidney events within 30 days (MAKE 30). When used individually, urine IGFBP-7 and TIMP-2 had AUCs of 0.76 and 0.79, respectively.6 As the two markers seemed to have additive predictive value for AKI, the main validation was performed using a combination of the two. The test result is a simple multiplication of the two markers, [TIMP-2] × [IGFBP-7]. The AUC of this dual biomarker was 0.80 for AKI, performing better than previously studied biomarkers. The risk of either AKI or MAKE30 increased sharply when [TIMP-2] × [IGFBP-7] was above 0.3 and doubled when the value was over 2.0. When added to a clinical model, [TIMP2] × [IGFBP-7] significantly improved the prediction value for severe AKI. This study illustrates the utility of combining biomarkers to enhance their predictive and discriminatory capabilities and also highlights the potential of using biomark ers to identify specific molecular pathways contributing to AKI. One of the difficulties in using AKI biomarkers has been to identify which patients would benefit most from their use. Indiscriminate use of biomarker testing in patients at low risk of AKI would render the biomarker nearly useless, as well as unnecessarily increase health-care costs. In
NATURE REVIEWS | NEPHROLOGY
cardiology, troponin is measured in at-risk patients who present with chest pain to rule-in or rule-out acute myocardial infarc tion. The concept of ‘renal angina’ rep resents a combination of risk factors with subtle changes in creatinine, urine output and fluid overload as the equivalent of chest pain, serving as an alert for evolving AKI.8 Hypothetically when biomarkers of AKI are used in patients with renal angina, their pre dictive value will improve. The concept of renal angina could, therefore, potentially guide testing of biomarkers in the future. In 2013, the predictive value of renal angina was evaluated in four paediatric cohorts in ICU (total 584 patients).9 A renal angina index (RAI) was derived as the product of a risk score (1–5) and injury score (1–8) based on a percentage decrease in esti mated creatinine clearance from baseline, or an increase in fluid overload in the first 8 h in ICU. An RAI >8 on day 1 in ICU was considered to be positive for renal angina; the primary outcome was AKI (stage 2–3) on day 3. Sensitivity was 58–93% and speci ficity was 36–90% across all cohorts, and the negative predictive value (NPV) was high at 92–99%, outperforming the Pediatric Risk of Mortality (PRISM-II) score. Recently, the predictive value of the RAI was also vali dated in critically ill adults.10 Similar to the paediatric study, sensitivity was 92%, speci ficity was 62% and NPV was excellent at
No functional change
Functional change
No damage
Damage
No AKI
Subclinical AKI
Functional AKI
Established AKI
Progression Resolution
Figure 1 | Proposed utilization of biomarkers in AKI. The combined use of markers of tissue function and pathology would allow patients to be profiled into one of the four quadrants. AKI is a dynamic process; therefore, sequential testing would permit delineation of the progression from one state to another with possible resolution of disease progression. The choice of markers could be based on their optimal performance in various disease states. Abbreviation: AKI, acute kidney injury. Permission obtained from Nature Publishing Group © Murray, T. P. et al. Kidney Int. 110, 22–26 (2013).
VOLUME 10 | FEBRUARY 2014 | 75 © 2014 Macmillan Publishers Limited. All rights reserved
YEAR IN REVIEW 99%. The high NPV in both of these studies suggest that patients who are negative for renal angina are unlikely to progress to severe AKI; therefore, this is a patient group in which screening for biomarkers of AKI could have low yield. Future studies need to examine the efficacy and cost-effectiveness of selective use of AKI biomarkers using the concept of renal angina. Taken together, these four studies demon strate the rapidly changing landscape of bio markers in AKI. We can now envision how biomarkers might be utilized to improve patient care. The use of biomarkers in patients at high risk of AKI would increase the likelihood of early disease recognition and of interventions. A combination of func tional kidney markers and tissue damage markers should enhance our ability to define appropriate time windows for inter ventions. Sequential testing for biomarkers and clinical evaluation would permit a better delineation of the response to interventions and, therefore, prognosis. In short, we are
76 | FEBRUARY 2014 | VOLUME 10
poised on the verge of a new phase in AKI therapeutic management that will hopefully translate into improved patient outcomes from this devastating disease. University of California, San Diego Medicine 8342, UCSD Medical Center, 200 West Arbor Drive, San Diego, CA 92103, USA (D. N. Cruz, R. L. Mehta). Correspondence to: R. L. Mehta [email protected] Competing interests D. N. Cruz declares associations with the following companies and organizations: Acute Dialysis Quality Initiative, Alere, Toray. R. L. Mehta declares associations with the following companies and organizations: AbbVie, Acute Dialysis Quality Initiative, AlloCure, Astute, Baxter, CSL Behring, Cytopherx, Eli Lilly, Gambro, GlaxoSmithKline, Grifols, Thrasos Therapeutics. See the article online for full details of the relationships. 1.
2.
Mehta, R. L. Acute kidney injury: urine output in AKI—the canary in the coal mine? Nat. Rev. Nephrol. 9, 568–570 (2013). Mandelbaum, T. et al. Empirical relationships among oliguria, creatinine, mortality, and renal replacement therapy in the critically ill. Intensive Care Med. 39, 414–419 (2013).
3.
Kidney Disease Improving Global Outcomes. KDIGO clinical practice guidelines for acute kidney injury. Kidney Int. Suppl. 2, 8–12 (2012). 4. Murray, P. T. et al. Current use of biomarkers in acute kidney injury: report and summary of recommendations from the 10th Acute Dialysis Quality Initiative consensus conference. Kidney Int. http://dx.doi.org/10.1038/ ki.2013.374. 5. Nickolas, T. L. et al. Diagnostic and prognostic stratification in the emergency department using urinary biomarkers of nephron damage: a multicenter prospective cohort study. J. Am. Coll. Cardiol. 59, 246–255 (2012). 6. Kashani, K. et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Crit. Care 17, R25 (2013). 7. Yang, L., Besschetnova, T. Y., Brooks, C. R., Shah, J. V. & Bonventre, J. V. Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat. Med. 16, 535–543 (2010). 8. Goldstein, S. L. & Chawla, L. S. Renal angina. Clin. J. Am. Soc. Nephrol. 5, 943–949 (2010). 9. Basu, R. K. et al. Derivation and validation of the renal angina index to improve the prediction of acute kidney injury in critically ill children. Kidney Int. http://dx.doi.org/10.1038/ ki.2013.349. 10. Cruz, D. N. et al. Renal angina syndrome and risk of severe acute kidney injury in critically ill patients. Clin. J. Am. Soc. Nephrol. (in press).
www.nature.com/nrneph © 2014 Macmillan Publishers Limited. All rights reserved
2.
3.
4.
Reeve, J. et al. Molecular diagnosis of T cellmediated rejection in human kidney transplant biopsies. Am. J. Transplant. 13, 645–655 (2013). Sellarés, J. et al. Molecular diagnosis of antibody-mediated rejection in human kidney transplants. Am. J. Transplant. 13, 971–983 (2013). Halloran, P. F. et al. Potential impact of microarray diagnosis of T cell-mediated rejection in kidney transplants: the INTERCOM study. Am. J. Transplant. 13, 2352–2363 (2013). Halloran, P. F. et al. Microarray diagnosis of antibody-mediated rejection in kidney transplant biopsies: an international prospective study (INTERCOM). Am. J. Transplant. 13, 2865–2874 (2013).
74 | FEBRUARY 2014 | VOLUME 10
5.
6.
7.
Furness, P. N. et al. International variation in histologic grading is large, and persistent feedback does not improve reproducibility. Am. J. Surg. Path. 27, 805–810 (2003). Mengel, M. et al. Banff initiative for quality assurance in transplantation (BIFQUIT): reproducibility of C4d immunohistochemistry in kidney allografts. Am. J. Transplant. 13, 1235–1245 (2013). Lefaucheur, C. et al. Antibody-mediated vascular rejection of kidney allografts:
a population-based study. Lancet 381, 313–319 (2013). 8. Suthanthiran, M. et al. Urinary-cell mRNA profile and acute cellular rejection in kidney allografts. N. Engl. J. Med. 369, 20–31 (2013). 9. Ingelfinger, J. R. & Alexander, S. I. One step closer to “Rejectostix.” N. Engl. J. Med. 369, 84–85 (2013). 10. Loupy, A. et al. Complement-binding anti-HLA antibodies and kidney-allograft survival. N. Engl. J. Med. 369, 1215–1226 (2013).
ACUTE KIDNEY INJURY IN 2013
Breaking barriers for biomarkers in AKI—progress at last Dinna N. Cruz and Ravindra L. Mehta
In 2013, four important papers were published that provide new insights on biomarkers in acute kidney injury (AKI). These studies demonstrate the potential for biomarkers to aid clinicians in improving the therapeutic management of patients with AKI and potentially improve patient outcomes. Cruz, D. N. & Mehta, R. L. Nat. Rev. Nephrol. 10, 74–76 (2014); published online 24 December 2013; doi:10.1038/nrneph.2013.268
Clinicians who treat patients with acute kidney injury (AKI) are faced with several questions that influence diagnostic and therapeutic decisions: has kidney injury occurred; what is the nature, sever ity and duration of kidney injury; has repair and recovery started; is intervention needed; and what is the prognosis? At present we glean the answers to these ques tions by evaluating sequential changes in the levels of urine output and serum cre atinine, but these evaluations are limited in their scope to guide therapeutic man agement. The emerging field of biomark ers specific to kidney injury holds promise for improving therapeutic management; however, the application of biomarkers in clinical practice has been limited. The past year has seen the publication of four papers that provide new insights in this field and bring us closer to implementing biomarkers in the clinical care of patients with AKI. Change in urine output could be viewed as a biomarker that is already available to the clinician,1 whereas novel biomarkers of AKI await further validation. Hourly urine output is a ‘biomarker’ that is measured 24 times per day in most intensive care units (ICUs), and is generally reported almost in real-time in electronic medical records. Mandelbaum et al. studied the empirical relationships between change in creatinine
concentration, change in urine output, the observation period over which these changes occurred, and clinical outcomes using a detailed ICU database for 14,526 patients. 2 For creatinine, the thresholds examined were an absolute increase of 8.84–88.4 μmol/l, or a relative increase of 125–400% from admission values, and observation periods were 1–7 days. For urine output, the thresholds were 0.1–1.0 ml/kg per h, for observation periods of 2–48 h. Mortality was high when there was a large absolute increase in creatinine levels regardless of the observation period, when the relative increase in creatini ne Key advances ■■ Urine output is emerging as a robust biomarker of acute kidney injury (AKI) associated with distinct outcomes in critically ill patients2 ■■ Combining biomarkers of tissue damage with those reflecting a functional change will enable clinicians to better characterize kidney health and the time points for specific interventions4 ■■ Biomarkers of tissue damage and cellcycle arrest pathways have been shown to predict increasing severity of AKI6 ■■ Identifying patients at high risk of developing ‘renal angina’ could potentially guide biomarker testing in the future9
www.nature.com/nrneph © 2014 Macmillan Publishers Limited. All rights reserved
YEAR IN REVIEW was high and the observation period was longer, and when oliguria was sustained for long periods of time. Overall, mortality increased rapidly as urine output decreased below 0.5 ml/kg per h, which lends support to urine output criteria used in current consensus definitions for AKI.3 The area under the receiver-operating characteristics curve (AUC) for urine output below 0.5 ml/ kg per h was >0.79 for mortality and >0.89 for renal replacement therapy across all the observation periods. The longer this level of urine output persisted the better was the discrimination for both outcomes. Several companies are now developing sensors for electronic monitoring of urine flow that will undoubtedly enhance clinicians’ ability to use this physiological biomarker to improve the management of AKI. However, a key caveat is to distinguish oliguria represent ing the normal autoregulatory response of the kidney formerly termed ‘pre-renal states’ from underlying damage. Different thresholds of change in urine output could improve the discrimination of these con ditions 1 and the Acute Dialysis Quality Initiative group has proposed consensus recommendations that provide a novel approach to biomarker utilization in AKI.4 Biomarkers specific to the kidney can be viewed as belonging to one of two broad classes representing functional changes (for example, serum creatinine, serum cys tatin C and urine output) or kidney damage (for example, proteinuria, urine and serum neutrophil gelatinase-associated lipo calin [NGAL], kidney injury molecule 1 [KIM-1] and liver-type fatty-acid binding protein [LFABP]). Consequently, by com bining biomarkers of functional change and tissue damage one can characterize any patient into one of four dynamic states of kidney health (Figure 1).4 A patient could transition from one state to another and back, depending on the nature, severity and duration of the injury. This approach permits the identification of a novel state of ‘subclinical’ AKI in which kidney tissue damage might occur independently of any functional change as measured by serum creatinine level and urine output. Alternatively, patients might be oliguric and have elevated serum creatinine levels in the absence of any alterations in biomarkers of tissue damage. For the practicing clini cian, this approach permits a more detailed assessment of the underlying condition of the kidney and sequential measurements could guide therap eutic interventions. Various biomarkers of functional change
and tissue damage could be used together for differential diagnosis and progno sis depending on the disease context, for example a combination of serum cystatin C and urine NGAL might be more informa tive in liver disease, whereas KIM-1 and serum creatinine level might be appropriate in patients with sepsis. Emerging evidence supports the concept of using biomarkers in this way; however, further validation is required.5 In the two-part multicenter Sapphire study, over 300 potential biomarkers in urine were examined in 522 critically ill patients during the discovery phase.6 The two top performing markers, insulin-like growth factor-binding protein 7 (IGFBP7) and tissue inhibitor of metalloprotein ases-2 (TIMP-2), were both inducers of G1 cell cycle arrest. After sepsis or ischae mic injury, renal tubular cells enter a brief period of cell cycle arrest, presumably to prevent potentially damaged cells from dividing. 7 In the validation phase of this study (involving 728 patients), these two biomarkers were evaluated for their ability to predict AKI (defined as Kidney Disease Improving Global Outcomes stage 2–3) within 12 h, as well as major adverse kidney events within 30 days (MAKE 30). When used individually, urine IGFBP-7 and TIMP-2 had AUCs of 0.76 and 0.79, respectively.6 As the two markers seemed to have additive predictive value for AKI, the main validation was performed using a combination of the two. The test result is a simple multiplication of the two markers, [TIMP-2] × [IGFBP-7]. The AUC of this dual biomarker was 0.80 for AKI, performing better than previously studied biomarkers. The risk of either AKI or MAKE30 increased sharply when [TIMP-2] × [IGFBP-7] was above 0.3 and doubled when the value was over 2.0. When added to a clinical model, [TIMP2] × [IGFBP-7] significantly improved the prediction value for severe AKI. This study illustrates the utility of combining biomarkers to enhance their predictive and discriminatory capabilities and also highlights the potential of using biomark ers to identify specific molecular pathways contributing to AKI. One of the difficulties in using AKI biomarkers has been to identify which patients would benefit most from their use. Indiscriminate use of biomarker testing in patients at low risk of AKI would render the biomarker nearly useless, as well as unnecessarily increase health-care costs. In
NATURE REVIEWS | NEPHROLOGY
cardiology, troponin is measured in at-risk patients who present with chest pain to rule-in or rule-out acute myocardial infarc tion. The concept of ‘renal angina’ rep resents a combination of risk factors with subtle changes in creatinine, urine output and fluid overload as the equivalent of chest pain, serving as an alert for evolving AKI.8 Hypothetically when biomarkers of AKI are used in patients with renal angina, their pre dictive value will improve. The concept of renal angina could, therefore, potentially guide testing of biomarkers in the future. In 2013, the predictive value of renal angina was evaluated in four paediatric cohorts in ICU (total 584 patients).9 A renal angina index (RAI) was derived as the product of a risk score (1–5) and injury score (1–8) based on a percentage decrease in esti mated creatinine clearance from baseline, or an increase in fluid overload in the first 8 h in ICU. An RAI >8 on day 1 in ICU was considered to be positive for renal angina; the primary outcome was AKI (stage 2–3) on day 3. Sensitivity was 58–93% and speci ficity was 36–90% across all cohorts, and the negative predictive value (NPV) was high at 92–99%, outperforming the Pediatric Risk of Mortality (PRISM-II) score. Recently, the predictive value of the RAI was also vali dated in critically ill adults.10 Similar to the paediatric study, sensitivity was 92%, speci ficity was 62% and NPV was excellent at
No functional change
Functional change
No damage
Damage
No AKI
Subclinical AKI
Functional AKI
Established AKI
Progression Resolution
Figure 1 | Proposed utilization of biomarkers in AKI. The combined use of markers of tissue function and pathology would allow patients to be profiled into one of the four quadrants. AKI is a dynamic process; therefore, sequential testing would permit delineation of the progression from one state to another with possible resolution of disease progression. The choice of markers could be based on their optimal performance in various disease states. Abbreviation: AKI, acute kidney injury. Permission obtained from Nature Publishing Group © Murray, T. P. et al. Kidney Int. 110, 22–26 (2013).
VOLUME 10 | FEBRUARY 2014 | 75 © 2014 Macmillan Publishers Limited. All rights reserved
YEAR IN REVIEW 99%. The high NPV in both of these studies suggest that patients who are negative for renal angina are unlikely to progress to severe AKI; therefore, this is a patient group in which screening for biomarkers of AKI could have low yield. Future studies need to examine the efficacy and cost-effectiveness of selective use of AKI biomarkers using the concept of renal angina. Taken together, these four studies demon strate the rapidly changing landscape of bio markers in AKI. We can now envision how biomarkers might be utilized to improve patient care. The use of biomarkers in patients at high risk of AKI would increase the likelihood of early disease recognition and of interventions. A combination of func tional kidney markers and tissue damage markers should enhance our ability to define appropriate time windows for inter ventions. Sequential testing for biomarkers and clinical evaluation would permit a better delineation of the response to interventions and, therefore, prognosis. In short, we are
76 | FEBRUARY 2014 | VOLUME 10
poised on the verge of a new phase in AKI therapeutic management that will hopefully translate into improved patient outcomes from this devastating disease. University of California, San Diego Medicine 8342, UCSD Medical Center, 200 West Arbor Drive, San Diego, CA 92103, USA (D. N. Cruz, R. L. Mehta). Correspondence to: R. L. Mehta [email protected] Competing interests D. N. Cruz declares associations with the following companies and organizations: Acute Dialysis Quality Initiative, Alere, Toray. R. L. Mehta declares associations with the following companies and organizations: AbbVie, Acute Dialysis Quality Initiative, AlloCure, Astute, Baxter, CSL Behring, Cytopherx, Eli Lilly, Gambro, GlaxoSmithKline, Grifols, Thrasos Therapeutics. See the article online for full details of the relationships. 1.
2.
Mehta, R. L. Acute kidney injury: urine output in AKI—the canary in the coal mine? Nat. Rev. Nephrol. 9, 568–570 (2013). Mandelbaum, T. et al. Empirical relationships among oliguria, creatinine, mortality, and renal replacement therapy in the critically ill. Intensive Care Med. 39, 414–419 (2013).
3.
Kidney Disease Improving Global Outcomes. KDIGO clinical practice guidelines for acute kidney injury. Kidney Int. Suppl. 2, 8–12 (2012). 4. Murray, P. T. et al. Current use of biomarkers in acute kidney injury: report and summary of recommendations from the 10th Acute Dialysis Quality Initiative consensus conference. Kidney Int. http://dx.doi.org/10.1038/ ki.2013.374. 5. Nickolas, T. L. et al. Diagnostic and prognostic stratification in the emergency department using urinary biomarkers of nephron damage: a multicenter prospective cohort study. J. Am. Coll. Cardiol. 59, 246–255 (2012). 6. Kashani, K. et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Crit. Care 17, R25 (2013). 7. Yang, L., Besschetnova, T. Y., Brooks, C. R., Shah, J. V. & Bonventre, J. V. Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury. Nat. Med. 16, 535–543 (2010). 8. Goldstein, S. L. & Chawla, L. S. Renal angina. Clin. J. Am. Soc. Nephrol. 5, 943–949 (2010). 9. Basu, R. K. et al. Derivation and validation of the renal angina index to improve the prediction of acute kidney injury in critically ill children. Kidney Int. http://dx.doi.org/10.1038/ ki.2013.349. 10. Cruz, D. N. et al. Renal angina syndrome and risk of severe acute kidney injury in critically ill patients. Clin. J. Am. Soc. Nephrol. (in press).
www.nature.com/nrneph © 2014 Macmillan Publishers Limited. All rights reserved
