[Show abstract][Hide abstract] ABSTRACT: Reviewing the literature in many fields on proposed risk models reveals problems with the way many risk models are developed. Furthermore, papers reporting new risk models do not always provide sufficient information to allow readers to assess the merits of the model. In this review, we discuss sources of bias that can arise in risk model development. We focus on two biases that can be introduced during data analysis. These two sources of bias are sometimes conflated in the literature and we recommend the terms resubstitution bias and model-selection bias to delineate them. We also propose the RiGoR reporting standard to improve transparency and clarity of published papers proposing new risk models.
[Show abstract][Hide abstract] ABSTRACT: Background:
In Central America, an epidemic of chronic kidney disease of unknown cause disproportionately affects young male agricultural workers.
Longitudinal cohort study.
Setting & participants:
284 sugarcane workers in 7 jobs were recruited from one company in northwestern Nicaragua. Blood and urine samples were collected before and near the end of the 6-month harvest season.
Job category (cane cutter, seeder, seed cutter, agrichemical applicator, irrigator, driver, and factory worker); self-reported water and electrolyte solution intake.
Outcomes & measurements:
Changes in levels of urinary kidney injury biomarkers normalized to urine creatinine level, including neutrophil gelatinase-associated lipocalin (NGAL), interleukin 18 (IL-18), N-acetyl-β-d-glucosaminidase (NAG), and albumin; serum creatinine-based estimated glomerular filtration rate (eGFR).
Mean eGFR was 113 mL/min/1.73 m(2) and <5% of workers had albuminuria. Field workers had increases in NGAL and IL-18 levels that were 1.49 (95% CI, 1.06 to 2.09) and 1.61 (95% CI, 1.12 to 2.31) times as high, respectively, as in non-field workers. Cane cutters and irrigators had the greatest increases in NGAL levels during the harvest, whereas cane cutters and seeders had the greatest increases in IL-18 levels. Electrolyte solution consumption was associated with lower mean NGAL and NAG levels among cane cutters and lower mean IL-18 and NAG levels among seed cutters; however, there was no overall effect of hydration among all workers. On average, workers with the largest increases in NGAL and NAG levels during the harvest had declines in eGFRs of 4.6 (95% CI, 1.0 to 8.2) and 3.1 (95% CI, -0.6 to 6.7) mL/min/1.73 m(2), respectively.
Surrogate exposure measure, loss to follow-up.
Results are consistent with the hypothesis that occupational heat stress and volume depletion may be associated with the development of kidney disease, and future studies should directly measure these occupational factors. The presence of urine tubular injury markers supports a tubulointerstitial disease that could occur with repeated tubular injury.
American Journal of Kidney Diseases 10/2015; DOI:10.1053/j.ajkd.2015.08.022 · 5.90 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Individual biomarkers of renal injury are only modestly predictive of acute kidney injury (AKI). Using multiple biomarkers has the potential to improve predictive capacity. In this systematic review, statistical methods of articles developing biomarker combinations to predict AKI were assessed. We identified and described three potential sources of bias (resubstitution bias, model selection bias, and bias due to center differences) that may compromise the development of biomarker combinations. Fifteen studies reported developing kidney injury biomarker combinations for the prediction of AKI after cardiac surgery (8 articles), in the intensive care unit (4 articles), or other settings (3 articles). All studies were susceptible to at least one source of bias and did not account for or acknowledge the bias. Inadequate reporting often hindered our assessment of the articles. We then evaluated, when possible (7 articles), the performance of published biomarker combinations in the TRIBE-AKI cardiac surgery cohort. Predictive performance was markedly attenuated in six out of seven cases. Thus, deficiencies in analysis and reporting are avoidable, and care should be taken to provide accurate estimates of risk prediction model performance. Hence, rigorous design, analysis, and reporting of biomarker combination studies are essential to realizing the promise of biomarkers in clinical practice.Kidney International advance online publication, 23 September 2015; doi:10.1038/ki.2015.283.
Kidney International 09/2015; DOI:10.1038/ki.2015.283 · 8.56 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Background:
The interaction between baseline kidney function and the performance of biomarkers of acute kidney injury (AKI) on the development of AKI is unclear.
Post hoc analysis of prospective cohort study.
Setting & participants:
The 1,219 TRIBE-AKI Consortium adult cardiac surgery cohort participants.
Unadjusted postoperative urinary biomarkers of AKI measured within 6 hours of surgery.
AKI was defined as AKI Network stage 1 (any AKI) or higher, as well as a doubling of serum creatinine level from the preoperative value or the need for post-operative dialysis (severe AKI).
Stratified analyses by preoperative estimated glomerular filtration rate (eGFR) ≤ 60 versus > 60mL/min/1.73m(2).
180 (42%) patients with preoperative eGFRs≤60mL/min/1.73m(2) developed clinical AKI compared with 246 (31%) of those with eGFRs>60mL/min/1.73m(2) (P<0.001). For log2-transformed biomarker concentrations, there was a significant interaction between any AKI and baseline eGFR for interleukin 18 (P=0.007) and borderline significance for liver-type fatty acid binding protein (P=0.06). For all biomarkers, the adjusted relative risk (RR) point estimates for the risk for any AKI were higher in those with elevated baseline eGFRs compared with those with eGFRs≤60mL/min/1.73m(2). However, the difference in magnitude of these risks was low (adjusted RRs were 1.04 [95% CI, 0.99-1.09] and 1.11 [95% CI, 1.07-1.15] for those with preoperative eGFRs≤60mL/min/1.73m(2) and those with higher eGFRs, respectively). Although no biomarker displayed an interaction for baseline eGFR and severe AKI, log2-transformed interleukin 18 and kidney injury molecule 1 had significant adjusted RRs for severe AKI in those with and without baseline eGFRs≤60mL/min/1.73m(2).
Limited numbers of patients with severe AKI and post-operative dialysis.
The association between early postoperative AKI urinary biomarkers and AKI is modified by preoperative eGFR. The degree of this modification and its impact on the biomarker-AKI association is small across biomarkers. Our findings suggest that distinct biomarker cutoffs for those with and without a preoperative eGFR≤60mL/min/1.73m(2) is not necessary.
American Journal of Kidney Diseases 09/2015; DOI:10.1053/j.ajkd.2015.07.027 · 5.90 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Assessment of deceased-donor organ quality is integral to transplant allocation practices, but tools to more precisely measure donor kidney injury and better predict outcomes are needed. In this study, we assessed associations between injury biomarkers in deceased-donor urine and the following outcomes: donor AKI (stage 2 or greater), recipient delayed graft function (defined as dialysis in first week post-transplant), and recipient 6-month eGFR. We measured urinary concentrations of microalbumin, neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), IL-18, and liver-type fatty acid binding protein (L-FABP) from 1304 deceased donors at organ procurement, among whom 112 (9%) had AKI. Each biomarker strongly associated with AKI in adjusted analyses. Among 2441 kidney transplant recipients, 31% experienced delayed graft function, and mean±SD 6-month eGFR was 55.7±23.5 ml/min per 1.73 m(2). In analyses adjusted for donor and recipient characteristics, higher donor urinary NGAL concentrations associated with recipient delayed graft function (highest versus lowest NGAL tertile relative risk, 1.21; 95% confidence interval, 1.02 to 1.43). Linear regression analyses of 6-month recipient renal function demonstrated that higher urinary NGAL and L-FABP concentrations associated with slightly lower 6-month eGFR only among recipients without delayed graft function. In summary, donor urine injury biomarkers strongly associate with donor AKI but provide limited value in predicting delayed graft function or early allograft function after transplant.
Journal of the American Society of Nephrology 09/2015; DOI:10.1681/ASN.2015040345 · 9.34 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Background and aims:
Hepatorenal syndrome is a severe complication of cirrhosis and associates with significant mortality. Vasoconstrictor medications improve renal function in patients with hepatorenal syndrome. However, it is unclear to what extent changes in serum creatinine during treatment may act as a surrogate for changes in mortality. We have performed a meta-analysis of randomized trials of vasoconstrictors assessing the association between changes in serum creatinine, taken as a continuous variable, and mortality, both while on treatment and during the follow-up period for survivors.
The electronic databases of PubMed, Web of Science and Embase were searched for randomized trials evaluating the efficacy of vasoconstrictor therapy for treatment of HRS type 1 or 2. The relative risk (RR) for mortality was calculated against delta creatinine. The proportion of treatment effect explained (PTE) was calculated for delta creatinine.
Seven trials enrolling 345 patients were included. The correlation between delta creatinine and ln (RR) was moderately good (R2 = 0.61). The intercept and parameter estimate indicated a fall in creatinine while on treatment of 1 mg/dL resulted in a 27% reduction in RR for mortality compared to the control arm. In patients surviving the treatment period, a fall in creatinine while on treatment of 1 mg/dL resulted in a 16% reduction in RR for post-treatment mortality during follow-up. The PTE of delta creatinine for overall mortality was 0.91 and 0.26 for post-treatment mortality.
Changes in serum creatinine in response to vasoconstrictor therapy appear to be a valid surrogate for mortality, even in the period following the completion of treatment.
PLoS ONE 08/2015; 10(8):e0135625. DOI:10.1371/journal.pone.0135625 · 3.23 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Since human urine is the most readily available biofluid whose proteome changes in response to disease, it is a logical sample for identifying protein biomarkers for kidney diseases.
Potential biomarkers were identified by using a multi-proteomics workflow to compare urine proteomes of kidney transplant patients with immediate and delayed graft function. Differentially expressed proteins were identified, and corresponding stable isotope-labeled internal peptide standards were synthesized for scheduled multiple reaction monitoring (MRM).
The Targeted Urine Proteome Assay (TUPA) was then developed by identifying those peptides for which there were at least 2 transitions for which interference in a urine matrix across 156 MRM runs was <30%. This resulted in an assay that monitors 224 peptides from 167 quantifiable proteins.
TUPA opens the way for using a robust mass spectrometric technology, MRM, for quantifying and validating biomarkers from amongst 167 urinary proteins. This approach, while developed using differentially expressed urinary proteins from patients with delayed versus immediate graft function after kidney transplant, can be expanded to include differentially expressed urinary proteins in multiple kidney diseases. Thus TUPA could provide a single assay to help diagnose, prognose, and manage many kidney diseases. This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
[Show abstract][Hide abstract] ABSTRACT: Given growth in kidney transplant waitlists and discard rates, donor kidney acceptance is an important problem. We used network analysis to examine whether organ procurement organization (OPO) network centrality affects discard and outcomes.
We identified 106,160 deceased donor kidneys recovered for transplant from 2000 to 2010 in Scientific Registry of Transplant Recipients. We constructed the transplant network by year with each OPO representing a node and each kidney-sharing relationship between OPOs representing a directed tie between nodes. Primary exposures were the number of different OPOs to which an OPO has given a kidney or from which an OPO has received a kidney in year preceding procurement year. Primary outcomes were discard, cold-ischemia time, delayed graft function, and 1-year graft loss. We used multivariable regression, restricting analysis to the 50% of OPOs with highest discard and stratifying remaining OPOs by kidney volume. Models controlled for kidney donor risk index, waitlist time, and kidney pumping.
An increase in one additional OPO to which a kidney was given by a procuring OPO in a year was associated with 1.4% lower likelihood of discard for a given kidney (odds ratio, 0.986; 95% confidence interval, 0.974-0.998) among OPOs procuring high kidney volume, but 2% higher likelihood of discard (odds ratio, 1.021; 95% confidence interval, 1.006-1.037) among OPOs procuring low kidney volume, with mixed associations with recipient outcomes.
Our study highlights the value of network analysis in revealing how broader kidney sharing is associated with levels of organ acceptance. We conclude interventions to promote broader inter-OPO sharing could be developed to reduce discard for a subset of OPOs.
[Show abstract][Hide abstract] ABSTRACT: Higher levels of plasma neutrophil gelatinase-associated lipocalin (pNGAL) are an early marker of acute kidney injury and are associated with increased risk of short-term adverse outcomes. The independent association between pNGAL and long-term mortality is unknown.
In this prospective observational cohort study, we studied 1191 adults who underwent cardiac surgery between 2007 and 2009 at 6 centers in the TRIBE-AKI cohort. We measured the pNGAL on the pre-operative and first 3 post-operative days and assessed the relationship of peri-operative pNGAL concentrations with all-cause mortality.
During a median follow-up of 3.0 years, 139 participants died (50/1000 person-years). Pre-operative levels of pNGAL were associated with 3-year mortality (unadjusted HR 1.96, 95% CI 1.34,2.85) and the association persisted after adjustment for pre-operative variables including estimated glomerular filtration rate (adjusted HR 1.48, 95% CI 1.04-2.12). After adjustment for pre- and intra-operative variables, including pre-operative NGAL levels, the highest tertiles of first post-operative and peak post-operative pNGAL were also independently associated with 3-year mortality risk (adjusted HR 1.31, 95% CI 1.0-1.7 and adjusted HR 1.78, 95% CI 1.2-2.7, respectively). However, after adjustment for peri-operative changes in serum creatinine, there was no longer an independent association between the first post-operative and peak post-operative pNGAL and long-term mortality (adjusted HR 0.98,95% CI 0.79-1.2 for first pNGAL and adjusted HR 1.19, 95% CI 0.87-1.61 for peak pNGAL).
Pre-operative pNGAL levels were independently associated with 3-year mortality after cardiac surgery. While post-operative pNGAL levels were also associated with 3-year mortality, this relationship was not independent of changes in serum creatinine. These findings suggest that while pre-operative pNGAL adds prognostic value for mortality beyond routinely available serum creatinine, post-operative pNGAL measurements may not be as useful for this purpose.
PLoS ONE 06/2015; 10(6):e0129619. DOI:10.1371/journal.pone.0129619 · 3.23 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: APOL1 genotype is associated with advanced kidney disease in African Americans, but the pathogenic mechanisms are unclear. Here, associations of APOL1 genotype with urine biomarkers of glomerular and tubular injury and kidney function decline were evaluated.
431 human immunodeficiency virus (HIV)-infected African American women enrolled in Women's Interagency HIV Study (WIHS).
Albumin-creatinine ratio (ACR), 4 tubular injury biomarkers (interleukin 18 [IL-18], kidney injury molecule 1 [KIM-1], neutrophil gelatinase-associated lipocalin [NGAL], and α1-microglobulin [A1M]), and kidney function estimated using the CKD-EPI cystatin C equation.
Participants were genotyped for APOL1 single-nucleotide polymorphisms rs73885319 (G1 allele) and rs71785313 (G2 allele). Urine biomarkers were measured using stored samples from 1999-2000. Cystatin C was measured using serum collected at baseline and 4- and 8-year follow-ups.
At baseline, ACRs were higher among 47 women with 2 APOL1 risk alleles versus 384 women with 0/1 risk allele (median, 24 vs 11mg/g; P<0.001). Compared with women with 0/1 risk allele, women with 2 risk alleles had 104% higher ACRs (95% CI, 29-223mg/g) and 2-fold greater risk of ACR>30 (95% CI, 1.17-3.44) mg/g after multivariable adjustment. APOL1 genotype showed little association with urine IL-18:Cr ratio, KIM-1:Cr ratio, and NGAL:Cr ratio (estimates of -5% [95% CI, -24% to 18%], -20% [95% CI, -36% to -1%], and 10% [95% CI, -26% to 64%], respectively) or detectable urine A1M (prevalence ratio, 1.13; 95% CI, 0.65-1.97) in adjusted analyses. Compared with women with 0/1 allele, women with 2 risk alleles had faster eGFR decline, by 1.2 (95% CI, 0.2 to 2.2) mL/min/1.73m(2) per year, and 1.7- and 3.4-fold greater rates of incident chronic kidney disease (95% CI, 1.1 to 2.5) and 10% annual eGFR decline (95% CI, 1.7 to 6.7), respectively, with minimal attenuation after adjustment for glomerular and tubular injury biomarker levels.
Results may not be generalizable to men.
Among HIV-infected African American women, APOL1-associated kidney injury appears to localize to the glomerulus, rather than the tubules.
Published by Elsevier Inc.
American Journal of Kidney Diseases 04/2015; 65(6). DOI:10.1053/j.ajkd.2015.02.329 · 5.90 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Children undergoing cardiac surgery may exhibit a pronounced inflammatory response to cardiopulmonary bypass (CPB). Inflammation is recognized as an important pathophysiologic process leading to acute kidney injury (AKI). The aim of this study was to evaluate the association of the inflammatory cytokines interleukin (IL)-6 and IL-10 with AKI and other adverse outcomes in children after CPB surgery.
This is a sub-study of the Translational Research Investigating Biomarker Endpoints in AKI (TRIBE-AKI) cohort, including 106 children ranging in age from 1 month to 18 years undergoing CPB. Plasma IL-6 and IL-10 concentrations were measured preoperatively and postoperatively [day 1 (within 6 h after surgery) and day 3].
Stage 2/3 AKI, defined by at least a doubling of the baseline serum creatinine concentration or dialysis, was diagnosed in 24 (23 %) patients. The preoperative IL-6 concentration was significantly higher in patients with stage 2/3 AKI [median 2.6 pg/mL, interquartile range (IQR) 2.6 0.6-4.9 pg/mL] than in those without stage 2/3 AKI (median 0.6 pg/mL, IQR 0.6-2.2 pg/mL) (p = 0.03). After adjustment for clinical and demographic variables, the highest preoperative IL-6 tertile was associated with a sixfold increased risk for stage 2/3 AKI compared with the lowest tertile (adjusted odds ratio 6.41, 95 % confidence interval 1.16-35.35). IL-6 and IL-10 levels increased significantly after surgery, peaking postoperatively on day 1. First postoperative IL-6 and IL-10 measurements did not significantly differ between patients with stage 2/3 AKI and those without stage 2/3 AKI. The elevated IL-6 level on day 3 was associated with longer hospital stay (p = 0.0001).
Preoperative plasma IL-6 concentration is associated with the development of stage 2/3 AKI and may be prognostic of resource utilization.
[Show abstract][Hide abstract] ABSTRACT: Research has identified improved biomarkers of acute kidney injury (AKI). Cystatin C (CysC) is a better glomerular filtration rate marker than serum creatinine (SCr) and may improve AKI definition.
To determine if defining clinical AKI by increases in CysC vs SCr alters associations with biomarkers and clinical outcomes.
Three-center prospective cohort study of intensive care units in New Haven, Connecticut, Cincinnati, Ohio, and Montreal, Quebec, Canada. Participants were 287 patients 18 years or younger without preoperative AKI or end-stage renal disease who were undergoing cardiac surgery. The study dates were July 1, 2007, through December 31, 2009.
For biomarker vs clinical AKI associations, the exposures were first postoperative (0-6 hours after surgery) urine interleukin 18, neutrophil gelatinase-associated lipocalin, kidney injury molecule 1, and liver fatty acid-binding protein. For clinical AKI outcome associations, the exposure was Kidney Disease: Improving Global Outcomes AKI definition (based on SCr or CysC).
Clinical AKI, length of stay, and length of mechanical ventilation. We determined areas under the receiver operating characteristic curve and odds ratios for first postoperative biomarkers to predict AKI.
The SCr-defined vs CysC-defined AKI incidence differed substantially (43.6% vs 20.6%). Percentage agreement was 71% (κ = 0.38); stage 2 or worse AKI percentage agreement was 95%. Interleukin 18 and kidney injury molecule 1 discriminated for CysC-defined AKI better than for SCr-defined AKI. For interleukin 18 and kidney injury molecule 1, the areas under the receiver operating characteristic curve were 0.74 and 0.65, respectively, for CysC-defined AKI, and 0.66 and 0.58, respectively, for SCr-defined AKI. Fifth (vs first) quintile concentrations of both biomarkers were more strongly associated with CysC-defined AKI. For interleukin 18 and kidney injury molecule 1, the odds ratios were 16.19 (95% CI, 3.55-73.93) and 6.93 (95% CI, 1.88-25.59), respectively, for CysC-defined AKI vs 6.60 (95% CI, 2.76-15.76) and 2.04 (95% CI, 0.94-4.38), respectively, for SCr-defined AKI. Neutrophil gelatinase-associated lipocalin and liver fatty acid-binding protein associations with both definitions were similar. The CysC definitions and SCr definitions were similarly associated with clinical outcomes of resource use.
Compared with the SCr-based definition, the CysC-based definition is more strongly associated with urine interleukin 18 and kidney injury molecule 1 in children undergoing cardiac surgery. Consideration should be made for defining AKI based on CysC in clinical care and future studies.
[Show abstract][Hide abstract] ABSTRACT: Acute kidney injury (AKI) is a common complication after cardiac surgery and is associated with worse outcomes. Since heart fatty acid binding protein (H-FABP) is a myocardial protein that detects cardiac injury, we sought to determine whether plasma H-FABP was associated with AKI in the TRIBE-AKI cohort; a multi-center cohort of 1219 patients at high risk for AKI who underwent cardiac surgery. The primary outcomes of interest were any AKI (Acute Kidney Injury Network (AKIN) stage 1 or higher) and severe AKI (AKIN stage 2 or higher). The secondary outcome was long-term mortality after discharge. Patients who developed AKI had higher levels of H-FABP pre- and postoperatively than patients who did not have AKI. In analyses adjusted for known AKI risk factors, first postoperative log(H-FABP) was associated with severe AKI (adjusted odds ratio (OR) 5.39 (95% confidence interval (CI), 2.87-10.11) per unit increase), while preoperative log(H-FABP) was associated with any AKI (2.07 (1.48-2.89)) and mortality (1.67 (1.17-2.37)). These relationships persisted after adjustment for change in serum creatinine (for first postoperative log(H-FABP)) and biomarkers of cardiac and kidney injury, including brain natriuretic peptide, cardiac troponin-I, interleukin-18, liver fatty acid binding protein, kidney injury molecule-1, and neutrophil gelatinase-associated lipocalin. Thus, perioperative plasma H-FABP levels may be used for risk stratification of AKI and mortality following cardiac surgery.Kidney International advance online publication, 1 April 2015; doi:10.1038/ki.2015.104.
Kidney International 04/2015; 88(3). DOI:10.1038/ki.2015.104 · 8.56 Impact Factor