Acute kidney injury clinical trial design: old problems, new strategies

Pragmatic platform trials to improve the outcome of patients with acute kidney injury

PURPOSE OF REVIEW

There is an important need for improved diagnostic strategies and treatment among patients with acute kidney injury (AKI). Classical randomized clinical trials have generated relevant results in AKI but are associated with shortcomings, such as high costs and sometimes lack of generalizability. In this minireview, we discuss the value and limits of pragmatic trials and platform trials for AKI research.

RECENT FINDINGS

The implementation of pragmatic and platform trials in critical care settings has generated relevant clinical evidence impacting clinical practice. Pragmatic and platform designs have recently been applied to patients at risk of AKI and represent a crucial opportunity to advance our understanding of optimized treatment and strategies in patients at risk of AKI or presenting with AKI. Trials embedded in electronic health records can facilitate patient enrollment and data collection. Platform trials have allowed for a more efficient study design. Although both pragmatic and platform trials have several advantages, they also come with the challenges and shortcomings discussed in this review.

SUMMARY

Pragmatic and platform trials can provide clinical answers in 'real-life' settings, facilitate a significant sample size enrollment at a limited cost, and provide results that can have a faster implementation in clinical practice.

Convolutional Neural Network Model for Intensive Care Unit Acute Kidney Injury Prediction

INTRODUCTION

Acute kidney injury (AKI) is common among hospitalized patients and has a significant impact on morbidity and mortality. Although early prediction of AKI has the potential to reduce adverse patient outcomes, it remains a difficult condition to predict and diagnose. The purpose of this study was to evaluate the ability of a machine learning algorithm to predict for AKI as defined by Kidney Disease: Improving Global Outcomes (KDIGO) stage 2 or 3 up to 48 hours in advance of onset using convolutional neural networks (CNNs) and patient electronic health record (EHR) data.

METHODS

A CNN prediction system was developed to use EHR data gathered during patients' stays to predict AKI up to 48 hours before onset. A total of 12,347 patient encounters were retrospectively analyzed from the Medical Information Mart for Intensive Care III (MIMIC-III) database. An XGBoost AKI prediction model and the sequential organ failure assessment (SOFA) scoring system were used as comparators. The outcome was AKI onset. The model was trained on routinely collected patient EHR data. Measurements included area under the receiver operating characteristic (AUROC) curve, positive predictive value (PPV), and a battery of additional performance metrics for advance prediction of AKI onset.

RESULTS

On a hold-out test set, the algorithm attained an AUROC of 0.86 and PPV of 0.24, relative to a cohort AKI prevalence of 7.62%, for long-horizon AKI prediction at a 48-hour window before onset.

CONCLUSION

A CNN machine learning-based AKI prediction model outperforms XGBoost and the SOFA scoring system, revealing superior performance in predicting AKI 48 hours before onset, without reliance on serum creatinine (SCr) measurements.

Clusterin in kidney transplantation: novel biomarkers versus serum creatinine for early prediction of delayed graft function

BACKGROUND AND OBJECTIVES

Current methods for rapid detection of delayed graft function (DGF) after kidney transplantation are unreliable. Urinary clusterin is a biomarker of kidney injury but its utility for prediction of graft dysfunction is unknown.

METHODS

In a single-center, prospective cohort study of renal transplant recipients (N=81), urinary clusterin was measured serially between 4 hr and 7 days after transplantation. The utility of clusterin for prediction of DGF (hemodialysis within 7 days of transplantation) was compared with urinary interleukin (IL)-18, neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1, serum creatinine, and clinical variables.

RESULTS

At 4 hr after reperfusion, anuria was highly specific, but of low sensitivity for detection of DGF. At 4 hr, receiver operating characteristic analysis suggested that urinary clusterin, IL-18, kidney injury molecule-1, and NGAL concentration were predictive of DGF. After adjusting for preoperative clinical variables and anuria, clusterin and IL-18 independently enhanced the clinical model for prediction of DGF. Kidney injury molecule-1 only modestly improved the prediction of DGF, whereas NGAL, serum creatinine, and the creatinine reduction ratio did not improve on the clinical model. At 12 hr, the creatinine reduction ratio independently predicted DGF.

CONCLUSION

Both urinary clusterin and IL-18 are useful biomarkers and may allow triaging of patients with DGF within 4 hr of transplantation. Relative performance of biomarkers for prediction of graft function is time-dependant. Early and frequent measurements of serum creatinine and calculation of the creatinine reduction ratio also predict DGF within 12 hr of reperfusion.

Towards a biomarker panel for the assessment of AKI in children receiving intensive care

BACKGROUND

Critically ill children and neonates are at high risk of developing acute kidney injury (AKI). AKI is associated with short- and long-term renal impairment and increased mortality. Current methods of diagnosing AKI rely on measurements of serum creatinine, which is a late and insensitive marker. Few studies to date have assessed AKI biomarkers in a heterogeneous patient cohort.

METHODS

We conducted a prospective feasibility study in a paediatric intensive care setting over a 6-month period to describe the relationship between AKI (defined according to pRIFLE criteria) and new AKI biomarkers.

RESULTS

In total, 49 patients between the ages of 16 days and 15 years were recruited for measurement of plasma cystatin C (Cys-C) and neutrophil gelatinase-associated lipocalin (pNGAL) concentrations, as well as for urinary kidney injury molecule-1 (KIM-1) and urinary NGAL (uNGAL) concentrations. Almost one-half (49 %) of the patient cohort experienced an AKI episode, and Cys-C and pNGAL were the strongest candidates for the detection of AKI. Our data suggest that the widely used estimated baseline creatinine clearance value of 120 mL/min/1.73 m(2) underestimates actual baseline function in patients admitted to paediatric intensive care units.

CONCLUSIONS

This investigation demonstrates the feasibility of new AKI biomarker testing in a mixed patient cohort and provides novel biomarker profiling for further evaluation.

A comparison of the ability of levels of urinary biomarker proteins and exosomal mRNA to predict outcomes after renal transplantation

Background

mRNA for biomarkers of kidney injury extracted from urinary exosomes may reflect or predict levels of the corresponding protein after transplantation and clinical outcomes.

Methods

Urinary exosomes were isolated from patients following renal transplantation, from healthy controls, and patients with CKD. Expression of exosomal mRNA for the injury biomarkers neutrophil gelatinase-associated lipocalin (NGAL), interleukin-18 (IL-18), kidney injury molecule-1 (KIM-1), and cystatin C was compared with the concentrations of corresponding urinary proteins, 18S RNA and serum creatinine.

Results

All biomarker protein concentrations increased after transplantation, and urinary NGAL and IL-18 at 24 and 168 h correlated with the day 7 creatinine reduction ratio (CRR). Exosomal18S RNA increased after transplantation, but exosomal mRNA for NGAL, IL-18 and cystatin C did not correlate with the day 7 CRR, or urinary biomarker concentrations at any time after transplantation. Exosomal NGAL mRNA was lower 4 h after transplantation than in control exosomes. In contrast, exosomal mRNA for cystatin C was unchanged after transplantation and in CKD, although urinary cystatin C temporarily increased following transplantation. Urinary KIM-1 increased after transplantation, but exosomal mRNA for KIM-1 remained undetectable. In CKD 18S RNA was raised, and exosomal mRNA for NGAL, IL-18 and cystatin C was detected in all patients. While urinary NGAL was greater in CKD than control subjects, exosomal NGAL mRNA was unchanged. Exosomal IL-18 mRNA was increased in CKD, but not IL-18 protein.

Conclusions

After renal transplantation, urinary NGAL and IL-18 levels reflect the day 7 CRR. However, while mRNA for these biomarkers is present in exosomes, their levels do not reflect or predict urinary biomarker levels or the CRR. This likely reflects the fact that packaging of mRNA in exosomes is selective, and is not necessarily representative of mRNA in the parent cells responsible for biomarker production.

Challenges and advances in the treatment of AKI

Treating or preventing AKI requires treating or preventing a rise in serum creatinine as well as the immediate and remote clinical consequences associated with AKI. Because a substantial number of patients with AKI progress to ESRD, identifying patients likely to progress and halting progression are important goals for treating AKI. Many therapies for AKI are being developed, including RenalGuard Therapy, which aims to maintain high urine output; α-melanocyte-stimulating hormone, with anti-inflammatory and antiapoptotic activities; alkaline phosphatase, which detoxifies proinflammatory substances; novel, small interfering RNA, directed at p53 activation; THR-184, a peptide agonist of bone morphogenetic proteins; removal of catalytic iron, important in free-radical formation; and cell-based therapies, including mesenchymal stem cells in vivo and renal cell therapy in situ. In this review, we explore what treatment of AKI really means, discuss the emerging therapies, and examine the windows of opportunity for treating AKI. Finally, we provide suggestions for accelerating the pathways toward preventing and treating AKI, such as establishing an AKI network, implementing models of catalytic philanthropy, and directing a small percentage of the Medicare ESRD budget for developing therapies to prevent and treat AKI and halt progression of CKD.

Derivation and validation of the renal angina index to improve the prediction of acute kidney injury in critically ill children

Reliable prediction of severe acute kidney injury (AKI) has the potential to optimize treatment. Here we operationalized the empiric concept of renal angina with a renal angina index (RAI) and determined the predictive performance of RAI. This was assessed on admission to the pediatric intensive care unit, for subsequent severe AKI (over 200% rise in serum creatinine) 72 h later (Day-3 AKI). In a multicenter four cohort appraisal (one derivation and three validation), incidence rates for a Day 0 RAI of 8 or more were 15-68% and Day-3 AKI was 13-21%. In all cohorts, Day-3 AKI rates were higher in patients with an RAI of 8 or more with the area under the curve of RAI for predicting Day-3 AKI of 0.74-0.81. An RAI under 8 had high negative predictive values (92-99%) for Day-3 AKI. RAI outperformed traditional markers of pediatric severity of illness (Pediatric Risk of Mortality-II) and AKI risk factors alone for prediction of Day-3 AKI. Additionally, the RAI outperformed all KDIGO stages for prediction of Day-3 AKI. Thus, we operationalized the renal angina concept by deriving and validating the RAI for prediction of subsequent severe AKI. The RAI provides a clinically feasible and applicable methodology to identify critically ill children at risk of severe AKI lasting beyond functional injury. The RAI may potentially reduce capricious AKI biomarker use by identifying patients in whom further testing would be most beneficial.

Linking injury to outcome in acute kidney injury: a matter of sensitivity

Current consensus definitions of Acute Kidney Injury (AKI) utilise thresholds of change in serum or plasma creatinine and urine output. Biomarkers of renal injury have been validated against these definitions. These biomarkers have also been shown to be independently associated with mortality and need for dialysis. For AKI definitions to include these structural biomarkers, there is a need for an independent outcome against which to judge both markers of functional change and structural markers of injury. We illustrate how sensitivity to need for dialysis and death can be used to link functional and structural (biomarker) based definitions of AKI. We demonstrated the methodology in a representative cohort of critically ill patients, in which an increase of plasma creatinine of >26.4 µmol/L in 48 hours or >50% in 7 days (Functional-AKI) had a sensitivity of 62% for death or dialysis within 30 days. In a development sub-cohort the urinary neutrophil-gelatinase-associated-lipocalin threshold with a 62% sensitivity for death or dialysis was 140 ng/ml (Structural-AKI). Using these thresholds in a validation sub-cohort, the risk of death or dialysis relative to those with no AKI by either definition was, for combined Structural-AKI and Functional-AKI 3.11 (95% Confidence interval: 2.53 to 3.55), for those with Structural-AKI but not Functional-AKI 1.51 (1.26 to 1.62), and for those with Functional-AKI but not Structural-AKI 1.34 (1.16 to 1.42). Linking functional and structural biomarkers via sensitivity for death and dialysis is a viable method by which to define thresholds for novel biomarkers of AKI.