Cardiac Biomarkers and Acute Kidney Injury After Cardiac Surgery
OBJECTIVES: To examine the relationship of cardiac biomarkers with postoperative acute kidney injury (AKI) among pediatric patients undergoing cardiac surgery.
METHODS: Data from TRIBE-AKI, a prospective study of children undergoing cardiac surgery, were used to examine the association of cardiac biomarkers (N-type pro–B-type natriuretic peptide, creatine kinase-MB [CK-MB], heart-type fatty acid binding protein [h-FABP], and troponins I and T) with the development of postoperative AKI. Cardiac biomarkers were collected before and 0 to 6 hours after surgery. AKI was defined as a ≥50% or 0.3 mg/dL increase in serum creatinine, within 7 days of surgery.
RESULTS: Of the 106 patients included in this study, 55 (52%) developed AKI after cardiac surgery. Patients who developed AKI had higher median levels of pre- and postoperative cardiac biomarkers compared with patients without AKI (all P < .01). Preoperatively, higher levels of CK-MB and h-FABP were associated with increased odds of developing AKI (CK-MB: adjusted odds ratio 4.58, 95% confidence interval [CI] 1.56–13.41; h-FABP: adjusted odds ratio 2.76, 95% CI 1.27–6.03). When combined with clinical models, both preoperative CK-MB and h-FABP provided good discrimination (area under the curve 0.77, 95% CI 0.68–0.87, and 0.78, 95% CI 0.68–0.87, respectively) and improved reclassification indices. Cardiac biomarkers collected postoperatively did not significantly improve the prediction of AKI beyond clinical models.
CONCLUSIONS: Preoperative CK-MB and h-FABP are associated with increased risk of postoperative AKI and provide good discrimination of patients who develop AKI. These biomarkers may be useful for risk stratifying patients undergoing cardiac surgery.
- biological markers
- acute kidney injury
- postoperative complications
- thoracic surgery
- creatine kinase
- brain natriuretic peptide
- fatty acid-binding proteins
What’s Known on This Subject:
Acute kidney injury (AKI) occurs in up to 50% of children after cardiopulmonary bypass and is associated with adverse outcomes. Renal biomarkers have been shown to predict postoperative AKI, but few studies have examined cardiac biomarkers for risk classification.
What This Study Adds:
Preoperative levels of creatine kinase-MB and heart-type fatty acid binding protein are strongly associated with the development of postoperative AKI after pediatric cardiac surgery and can be used to improve preoperative clinical risk prediction.
Acute kidney injury (AKI) occurs in up to 50% of children after cardiopulmonary bypass (CPB),1,2 and is independently associated with adverse outcomes, including longer lengths of stay (LOSs), prolonged mechanical ventilation, and higher mortality.3,4 Early identification of patients at risk for developing AKI may allow implementation of strategies to reduce related morbidity and mortality.5,6
Preliminary evidence suggests that biomarkers can help to identify patients at risk for postoperative AKI.2 In both pediatric and adult populations, renal biomarkers, such as serum cystatin C and neutrophil gelatinase-associated lipocalin (NGAL), have been shown to predict postoperative AKI and improve risk classification when combined with other clinical models.7–11 In addition, studies in adults have shown that pre- and postoperative levels of cardiac biomarkers, such as B-type natriuretic peptide (BNP), strongly predict postoperative AKI risk.12 Only 2 studies in children have examined the prognostic value of BNP on postoperative AKI.13,14 Both found that although BNP was not an independent predictor of AKI, it was associated with longer intubation times and higher postoperative mortality. To our knowledge, no studies have examined other cardiac biomarkers, such as troponins, creatine kinase-MB (CK-MB), or heart-type fatty acid binding protein (h-FABP), in children or adults. Preoperative elevations in cardiac biomarkers may reflect the severity of the underlying heart disease,15 whereas postoperative levels of cardiac biomarkers reflect the complexity of the surgery and the degree of intraoperative cardiac damage, which can increase patients’ risk of postoperative complications, such as AKI.
The purpose of this study was twofold. First we evaluated the prognostic utility of 5 cardiac biomarkers (N-type [NT] pro-BNP, CK-MB, h-FABP, high-sensitivity cardiac troponin T [hs-cTnT], and cardiac troponin I [cTnI]), measured preoperatively and immediately after surgery in children, to predict postoperative AKI risk and LOS. We then determined whether these biomarkers provided additional benefit for predicting AKI above that afforded by clinical models.
Data from the Translational Research Investigating Biomarker Endpoints in Acute Kidney Injury consortium study (TRIBE-AKI) were used for these analyses.16,17 In brief, this translational study was designed to validate novel kidney injury biomarkers by using a prospective cohort design with serial sample collections on 3 consecutive days. Participants included children aged 1 month to 18 years undergoing cardiac surgery at 3 academic hospitals. To increase the number of children at risk for AKI, we oversampled children undergoing risk adjustment of congenital heart surgery-1 (RACHS-1) category ≥2 surgeries. Children were recruited preoperatively and followed postoperatively until discharge (n = 319). Institutional review board approval was obtained at each participating center, and all patients provided written informed consent. To capture cardiac biomarker kinetics around the time of cardiac surgery, only patients with a full set of pre- and postoperative samples were included (n = 106). Subject selection was not based on any clinical criteria.
Data on patient demographics and medical history were recorded before surgery. Details about the surgical procedure (eg, type of cardiac abnormality, procedure, bypass time, elective or urgent, and severity) were obtained from the medical record by using standardized definitions of the Society of Thoracic Surgeons data collection tool. Severity of condition and surgical risk were evaluated using the RACHS-1 method.18,19
Venous blood samples were collected preoperatively, within 6 hours after surgery, and on postoperative days 2 and 3. Blood was collected in EDTA tubes and centrifuged to separate plasma, divided into bar-coded 0.5-mL cryovials, and stored at –80°C. One vial from each time-point was used for biomarker measurements with a single freeze-thaw. Biomarkers were measured with a Roche automated analyzer (Roche Elecsys 2010; Roche Diagnostics, Basel, Switzerland) for NT pro-BNP (pmol/L) (coefficient of variation [CV] range 3.6%–7.7%) and hs-cTnT (ng/L) (CV range 2.5%–10.5%), the Beckman Coulter Access II instrument (Beckman Coulter, Brea, CA) for the AccuTnI assay cTnI (µg/L) (CV range 5.4%–20%) and CK-MB (µg/L) (CV range 2.7%–8.2%), and the Evidence Investigator Cytokine Custom Array (Randox Crumlin, United Kingdom) for h-FABP (µg/L) (CV 17%). Preoperative serum creatinine (SCr) was measured as part of routine clinical care using modified Jaffe or enzymatic assays, and preoperative glomerular filtration rates (GFRs) were estimated by using the Schwartz equation.
The primary outcome in this study was development of AKI, which was defined as rise in SCr of ≥50% or 0.3 mg/dL from preoperative baseline within the first 7 days after surgery. Severe AKI was defined as either a doubling of creatinine or dialysis requirement.20–22 Secondary outcomes included in-hospital mortality, hospital and intensive care unit (ICU) LOS, and time to extubation.
Sample characteristics were compared among patients who developed severe AKI, mild AKI, and no AKI by using analysis of variance or Kruskal-Wallis tests for continuous variables and χ2 or Fisher’s exact test for categorical variables. Median biomarker values were plotted and compared across AKI groups by using Kruskal-Wallis tests for NT pro-BNP, cTnI, hs-cTnT, CK-MB, and h-FABP. Colinearity between biomarkers was assessed by using scatterplot and correlation matrices.
We evaluated unadjusted associations between cardiac biomarkers and the development of AKI by using logistic regression. Because 95% of AKI cases occurred in the first 2 postoperative days, we focused on cardiac biomarkers collected preoperatively and immediately postoperatively (within 6 hours of surgery). Biomarker levels were introduced into the models as log transformations to normalize the distributions of these values. Analyses were repeated adjusting for demographic and preoperative characteristics, including patient age, preoperative estimated GFR (eGFR) percentile, hospital site, RACHS-1 category ≥3, and CPB time >120 minutes. Covariates for the models were selected by using a combination of previously reported AKI predictors in this cohort16,23 and significance testing (P < .1).
We calculated the area under the receiver operating characteristic (ROC) curve (AUC) for each biomarker alone to determine its ability to discriminate between patients who did and did not develop AKI. Contingency tables were used to determine the optimal cutpoint for each biomarker. We used 2 approaches to assess the added value of biomarkers in the clinical prediction models. First, we evaluated incremental changes in the AUC when individual biomarkers were added to the clinical model. This approach, however, is often an insensitive measure of the ability of a new marker to add value to a preexisting model because the c-statistic minimally moves after a few powerful risk factors are already in the model. Accordingly, we also calculated the continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indices. The NRI indicates how much more frequently appropriate reclassification of AKI risk occurs than inappropriate reclassification with use of the model containing the biomarker, whereas the IDI measures how far individuals move on average along the continuum of predictive risk. These measures have advantages over the AUC because they directly quantify the appropriateness and amount of reclassification when biomarker values are added to the clinical prediction models.24–26 Both the NRI and IDI have been used extensively in cardiovascular outcomes and nephrology research.27–32
We then evaluated the association between these biomarkers with ICU and hospital LOS by using logistic regression. ICU and hospital stays were dichotomized by using the median value of each (3 days for ICU stay and 7 days for hospital stay) to quantify the relation between elevations in cardiac biomarkers and risk of prolonged hospitalization by using risk measures. Models were adjusted for patient and clinical characteristics as well as AKI status to determine whether AKI explained the longer LOS in patients with high cardiac biomarker values.
Of the 106 pediatric patients enrolled in this study, 23 (21.7%) developed severe AKI and 32 (30.2%) developed mild AKI after cardiac surgery (Table 1). Patients who developed AKI tended to be younger and to have lower preoperative weights. Patients who developed severe AKI had longer hospital and ICU LOS than patients with mild or no AKI. Only 2 deaths occurred in-hospital, both among patients with severe AKI.
The trajectories of the 5 cardiac biomarkers during the first 3 days of hospitalization stratified by AKI status are displayed in Fig 1. Markers of cardiac injury (CK-MB, h-FABP, cTnI, and hs-cTnT) peaked within 6 hours after surgery, whereas markers of cardiac function (NT pro-BNP) peaked on day 2. Across all 4 time-points, median levels of cardiac biomarkers were higher among patients who developed severe AKI compared with mild or no AKI.
Preoperative Cardiac Biomarkers
Fig 2 shows the distributions of the 5 preoperative biomarkers stratified by AKI status. There were significant differences across AKI groups in preoperative CK-MB, hs-cTnT, and h-FABP levels (P = .001, P = .048, and P < .001, respectively), but preoperative levels of NT pro-BNP and cTnI were comparable across groups (P = .2 and P = .3, respectively).
Higher preoperative levels of CK-MB and h-FABP were associated with higher crude odds of developing AKI (Table 2). After adjustment for patient characteristics, the association of preoperative CK-MB and h-FABP with development of postoperative AKI remained significant. A 1-unit increase in log CK-MB was associated with a fivefold increase in the odds of developing AKI after surgery. Similarly, a 1-unit increase in log h-FABP was associated with a threefold increase in the odds of developing AKI. No other preoperative biomarkers were associated with the development of AKI after adjustment for patient characteristics (all P > .1).
ROC analyses revealed preoperative CK-MB and h-FABP to be candidate biomarkers with high discriminatory value for predicting AKI (Fig 3 A and B). When used alone, preoperative CK-MB and h-FABP had relatively high AUCs (CK-MB: AUC 0.70, 95% confidence interval [CI] 0.60–0.81; h-FABP: AUC 0.70, 95% CI 0.60–0.81) (Table 3). Adding preoperative CK-MB or h-FABP to the clinical model did not significantly improve the AUC of the clinical model (P = .60 and P = .51, respectively); however, both biomarkers were associated with significant NRI and IDI values (Table 4). Given that changes in the AUC are insensitive when comparing nested models, it is not surprising that preoperative CK-MB and h-FABP added significant discriminative value to the clinical model when assessed by NRI and IDI, but increased the AUC only slightly in ROC analyses. The optimal cutoff for detecting postoperative AKI was 2.9 µg/L (sensitivity 64.2%, specificity 64.6%) for preoperative CK-MB and 2.6 pg/mL (sensitivity 68%, specificity 68.8%) for preoperative h-FABP (Table 5).
Secondary analyses showed that higher preoperative hs-cTnT, CK-MB, and h-FABP were associated with longer ICU and hospital stays independent of AKI status (Table 6). Preoperative NT pro-BNP and cTnI were associated with longer ICU but not hospital stays. Of the 5 preoperative biomarkers tested, CK-MB and h-FABP exhibited the strongest association with LOS.
Postoperative Day 1 Cardiac Biomarkers
Biomarker distributions on postoperative day 1 are presented in Fig 4. Median levels of biomarkers differed significantly across AKI groups for all biomarkers except CK-MB (P < .05 for NT pro-BNP, cTnI, hs-cTnT, and h-FABP). In unadjusted analyses, only postoperative hs-cTnT was borderline associated with the development of AKI, but adjustment for patient and operative characteristics attenuated this relationships (Table 2). All 5 biomarkers were nonsignificant in multivariable analyses.
Postoperative hs-cTnT yielded the highest AUC for predicting AKI in both unadjusted and adjusted ROC analyses (Fig 3C). When used alone, postoperative hs-cTnT had a moderately high AUC (0.62, 95% CI 0.51–0.73), but addition of hs-cTnT to the clinical model did not significantly improve the AUC of the clinical model (P = .74) (Table 3). Similarly, neither the NRI nor IDI proved significant for hs-cTnT or any other postoperative biomarkers (all P > .05) (Table 4).
In secondary analyses, higher postoperative levels of NT pro-BNP predicted longer ICU and hospital LOS, which persisted even after adjustment for AKI status (Table 6). Postoperative hs-cTnT and h-FABP also were associated with increased ICU but not hospital stay.
In this multicenter, prospective study of pediatric patients undergoing cardiac surgery for congenital conditions, we found that preoperative CK-MB and h-FABP were strong and independent predictors of postoperative AKI. A 1-unit increase in preoperative log CK-MB was associated with a fivefold increase in the odds of AKI, and a 1-unit elevation in preoperative log h-FABP increased the odds of developing AKI by 3 times. We found that a single measurement of each of these biomarkers provided good discrimination between patients who developed AKI and those who did not when used alone and in combination with clinical models. Previous studies examining the association of preoperative renal biomarkers with postoperative AKI have yielded AUC values ranging 0.44 to 0.72 for NGAL, cystatin C, and kidney injury molecule-1.33–36 Additionally, most clinical risk models for the prediction of postoperative AKI have produced AUCs in the range of 0.65 to 0.83, depending on the definition of AKI and the number of clinical parameters included.11,37–40 When placed in the context of previous studies, our findings suggest that CK-MB and h-FABP may be useful biomarkers for assessing preoperative risk of AKI in children undergoing cardiac surgery. However, larger follow-up studies are needed to verify the results of this study and confirm the utility of these biomarkers as risk stratification tools for AKI.
This is the first prospective study analyzing the use of multiple cardiac biomarkers to identify patients at high risk of postoperative AKI in children. Although several studies have evaluated the utility of biomarkers for AKI risk stratification after cardiac surgery in children, most of these studies have focused on renal biomarkers.41–45 To our knowledge, only 1 study in adults and 2 studies in children have examined cardiac biomarkers for the prediction of AKI after cardiac surgery.12–14 However, in all 3 studies, the authors examined only BNP and observed only modest associations between biomarker levels and postoperative AKI. In the pediatric studies, Cantinotti et al13 found that pre- and postoperative BNP were associated with AKI in unadjusted analyses but lost their predictive value once urinary NGAL and other conventional risk factors were added to the model. Similarly, Hornik et al14 found that preoperative BNP was not associated with increased risk of postoperative AKI. Our results for NT pro-BNP are consistent with those reported previously.
Our study expands the current literature by examining additional cardiac biomarkers in the prediction of AKI, including markers of cardiac function (NT pro-BNP) and cardiac injury (cTnI, hs-cTnT, CK-MB, and h-FABP). In our study, preoperative CK-MB and h-FABP outperformed NT pro-BNP in predicting postoperative AKI as assessed by larger odds ratios and higher AUCs. We also found that higher preoperative levels of CK-MB and h-FABP were associated with extended ICU and hospital LOS in addition to AKI, which is consistent with previous studies.46–49 However, addition of AKI to these models did not attenuate the risk ratios, suggesting that the relationship between these biomarkers and extended LOS is mediated by factors other than AKI. These findings suggest that CK-MB and h-FABP may be superior to BNP in evaluating risk of postoperative AKI in children.
Several mechanisms may explain the relationship between preoperative CK-MB and h-FABP with postoperative AKI. In children with congenital heart disease, elevations in CK-MB and h-FABP are most likely released from damaged myocardium, suggesting that patients with higher preoperative levels may have more severe underlying cardiac disease.50 These patients may have reduced cardiac reserve at baseline and may be at higher risk of hemodynamic instability in the perioperative setting, which may place them at greater risk of postsurgical complications, including AKI. Additionally, patients with more severe baseline disease may require longer cross-clamping and CPB times, which can independently increase the risk of postoperative AKI.16
Although the proposed mechanisms are plausible, our data showed mixed results for several of these explanations. For example, neither cross-clamping nor CPB duration was associated with postoperative AKI in our sample. Additionally, it is unclear why only 2 markers of cardiac injury (CK-MB and h-FABP) predicted AKI but not preoperative cardiac troponins. One potential explanation is the large number of children with undetectable levels of preoperative cTnI (n = 55) and hs-cTnT (n = 38). Even with low levels of cardiac damage, the range of cardiac troponin values may not have been wide enough to detect subtle differences between children who developed postoperative AKI and those who did not. In fact, previous studies also have shown that cardiac troponin levels are often undetectable in children with congenital heart abnormalities or other cardiac diseases such as Kawasaki,51–53 but CK-MB levels are frequently elevated and vary by severity and type of cardiac defect.50,54 Thus, it is possible that the kinetics of the cardiac troponins, CK-MB, and h-FABP differ in children.
The high performance of CK-MB and h-FABP in our study may be partly explained by the fact that we limited our sample to only higher-risk patients undergoing complex cardiac surgeries. Larger studies are needed to replicate our findings in patients with varying levels of surgical risk and to determine the cost-effectiveness of monitoring CK-MB and h-FABP preoperatively. Recently, the concept of a renal angina index has been proposed to direct biomarker assessment in noncardiac ICU patients who meet a certain level of risk and demonstrate signs of evolving AKI.55,56 This concept has arisen due to concerns that widespread diagnostic testing may yield high numbers of false-positive results.57 In the cardiac surgery setting, a similar parallel would be clinical risk scores. In TRIBE-AKI, we found that using a simple baseline clinical risk model, including age, gender, RACHS-1 category, preoperative eGFR percentile, CPB time, and study site, predicted postoperative AKI with AUC 0.77 and was improved by adding biomarkers to this model.36 Similar risk models may help to increase the specificity and positive predictive value of biomarker screening tests.58
Before the use of these biomarkers can be incorporated into routine clinical practice, however, AKI prevention strategies must be developed and shown to be efficacious in patients with elevated biomarkers. If replicated, our study has important implications for decisions regarding timing of surgery and for identifying patients at risk for AKI before surgery that may benefit from certain interventions. Although there are no established preventive measures or early treatments for AKI in children undergoing cardiac surgery, risk stratification may help to avoid AKI by minimizing CPB time, avoiding nephrotoxic medications, and optimizing hemodynamics through fluid management and inotropic support. In addition, it may be worth considering postponing the surgery in children at high risk undergoing elective procedures.2
This study has a few limitations. First, despite being a multicenter international study, we examined only patients with RACHS-1 category 2 to 4, which may limit the generalizability of our results. Although our results are compelling, they need to be validated by larger prospective studies that include children with varying levels of surgical risk at baseline. Second, we included only patients for whom both pre- and postoperative samples had been collected. This approach may have been subject to sampling bias if biomarker collection was associated with disease severity; however, the distribution of RACHS-1 scores in our sample was similar to that in the overall pediatric cohort. Third, there is no true gold standard for AKI. Therefore, we based the definition of AKI on elevations in SCr, which may be a potentially flawed outcome variable for evaluating the performance of novel biomarkers. It is possible that this study may have yielded different results if a different definition had been used. Fourth, TRIBE does not contain information on a few potentially important variables, including preoperative mechanical ventilation, inotropic support, and diuretic use, which may have confounded the relationship between cardiac biomarkers and postoperative AKI. Nevertheless, we adjusted for RACHS-1 category as an overall assessment of preoperative illness severity. Fifth, neonates were excluded from this study because previous studies have shown that using SCr to diagnose AKI in neonates can be problematic.59 Neonates have higher baseline levels of SCr at birth, which gradually decline during the first few weeks of life due to ongoing renal functional development.60,61 These changes, combined with the lower muscle mass in neonates, make it difficult to interpret changes in SCr with AKI. Thus, our findings are not generalizable to children <1 year of age, and future multicenter studies should be performed in that age group. Finally, the number of patients experiencing dialysis or death in our study was low, limiting our ability to evaluate the association of biomarkers with these outcomes.
In summary, preoperative levels of CK-MB and h-FABP were highly associated with the development of AKI and prolonged LOS and mechanical ventilation after pediatric cardiac surgery. Both biomarkers improved clinical risk prediction of postoperative AKI; however, additional studies are needed to confirm these findings in patients with varying levels of surgical risk and to evaluate the cost-effectiveness of obtaining these biomarkers for risk-stratification purposes. The ability to predict postoperative sequelae using preoperative clinical data has important implications for identifying patients who may benefit from earlier interventions for AKI. If replicated by future studies, these findings suggest that preoperative measurement of CK-MB and h-FABP may be useful for risk stratifying patients before surgery.
- Accepted January 7, 2015.
- Address correspondence to Chirag R. Parikh, MD, PhD, Department of Internal Medicine, Yale University School of Medicine, 60 Temple Street, Suite 6C, New Haven, CT 06510. E-mail:
Ms Bucholz conceptualized and designed this study, analyzed the data, and drafted the initial manuscript; Drs Whitlow and Eikelboom and Ms Shortt supervised data collection, and reviewed and revised the manuscript; Drs Zappitelli, Devarajan, Garg, Devereaux, Krawczeski, and Kavsak designed the initial Translational Research Investigating Biomarker Endpoints cohort study, reviewed and revised the manuscript, and obtained funding for the study; Ms Thiessen-Philbrook supervised data collection, analyzed the data, and reviewed and revised the manuscript; Dr Parikh conceptualized and designed this study, designed the initial Translational Research Investigating Biomarker Endpoints cohort study, reviewed and revised the manuscript, and obtained funding for the study; and all authors approved the final manuscript as submitted.
FINANCIAL DISCLOSURE: Drs Devarajan and Devereaux are co-inventors on patents submitted for the use of neutrophil gelatinase-associated lipocalin as a biomarker of kidney disease. Dr Devereaux has received research grants from Abbott Diagnostics and Roche Diagnostics. Dr Kavsak has received honorarium and research grants from Abbott Laboratories, Beckman Coulter, Ortho Clinical Diagnostics, Randox Laboratories, and Roche Diagnostics; he has consulted for Abbott Laboratories and Roche Diagnostics. The other authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Supported by the National Institutes of Health (NIH) (R01HL085757 to Dr Parikh) to fund the TRIBE-AKI Consortium to study novel biomarkers of acute kidney injury in cardiac surgery and the O’Brien Center Award from the National Institute of Diabetes and Digestive and Kidney Diseases (P30DK079310). Dr Parikh is also supported by NIH (K24DK090203). Drs Garg and Parikh are also members of the NIH-sponsored Assess, Serial Evaluation, and Subsequent Sequelae in Acute Kidney Injury Consortium (U01DK082185). Biomarker measurements were supported by Roche Diagnostics, Beckman Coulter, and Randox Laboratories. The granting agencies and Roche Diagnostics, Beckman Coulter, and Randox Laboratories did not participate in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript. Funded by the National Institutes of Health (NIH).
POTENTIAL CONFLICT OF INTEREST: Drs Devarajan is a co-inventor on patents submitted for the use of neutrophil gelatinase-associated lipocalin as a biomarker of kidney disease. Dr Devereaux has received research grants from Abbott Diagnostics and Roche Diagnostics. Dr Kavsak has received honorarium and research grants from Abbott Laboratories, Beckman Coulter, Ortho Clinical Diagnostics, Randox Laboratories, and Roche Diagnostics; he has consulted for Abbott Laboratories and Roche Diagnostics. The other authors have indicated they have no potential conflicts of interest to disclose.
- Molitoris BA
- Chertow GM,
- Lazarus JM,
- Christiansen CL,
- et al
- Thakar CV,
- Arrigain S,
- Worley S,
- Yared JP,
- Paganini EP
- Patel UD,
- Garg AX,
- Krumholz HM,
- et al.,
- Translational Research Investigating Biomarker Endpoints in Acute Kidney Injury (TRIBE-AKI) Consortium
- Hernández-Leiva E,
- Dennis R,
- Isaza D,
- Umaña JP
- Parikh CR,
- Devarajan P,
- Zappitelli M,
- et al.,
- TRIBE-AKI Consortium
- Bellomo R,
- Ronco C,
- Kellum JA,
- Mehta RL,
- Palevsky P,
- Acute Dialysis Quality Initiative workgroup
- Cook NR
- Pickering JW,
- Endre ZH
- Fox CS,
- Gona P,
- Larson MG,
- et al
- Hall IE,
- Coca SG,
- Perazella MA,
- et al
- Siew ED,
- Ware LB,
- Gebretsadik T,
- et al
- Tzoulaki I,
- Liberopoulos G,
- Ioannidis JP
- Han WK,
- Wagener G,
- Zhu Y,
- Wang S,
- Lee HT
- Koyner JL,
- Vaidya VS,
- Bennett MR,
- et al
- Kiers HD,
- van den Boogaard M,
- Schoenmakers MC,
- et al
- Mehta RH,
- Grab JD,
- O’Brien SM,
- et al.,
- Society of Thoracic Surgeons National Cardiac Surgery Database Investigators
- Rahmanian PB,
- Kwiecien G,
- Langebartels G,
- Madershahian N,
- Wittwer T,
- Wahlers T
- Krawczeski CD,
- Vandevoorde RG,
- Kathman T,
- et al
- Soler YA,
- Nieves-Plaza M,
- Prieto M,
- García-De Jesús R,
- Suárez-Rivera M
- Lipshultz SE,
- Rifai N,
- Sallan SE,
- et al
- Johnson XD,
- Liu KD
- Copyright © 2015 by the American Academy of Pediatrics