Derivation and validation of a prediction score for acute kidney injury in patients hospitalized with acute heart failure in a Chinese cohort

Acute Kidney Disease After Acute Decompensated Heart Failure

Introduction

Acute kidney disease (AKD) represents a continuum of kidney injury for 7 to 90 days after acute kidney injury (AKI). The incidence and prognosis of AKD after acute decompensated heart failure (ADHF) are currently unclear. The aims of this study were to explore the incidence of AKD and the transition from AKI to AKD, to identify risk factors for AKD and develop a prediction model for any-stage AKD, and to evaluate the prognosis of AKD.

Methods

A total of 7519 patients admitted for ADHF between January 1, 2008, and December 31, 2018, from a multi-institutional database were identified. The composite outcomes after ADHF were stage 3 AKD and all-cause death. The prognosis impact of AKD, including major adverse kidney events (MAKEs), all-cause death, and heart failure hospitalization (HFH), during 5 years of follow-up was analyzed.

Results

The overall incidence of AKI and AKD after ADHF was 9% and 21.2%, respectively; 39.4% of the patients diagnosed with having AKI during ADHF subsequently developed AKD whereas 19.4% of the patients without an identified AKI episode subsequently developed AKD. The predictive scoring models revealed C-statistics of 0.726 (95% CI: 0.712–0.740) for any-stage AKD and 0.807 (95% CI: 0.793–0.821) for the composite of stage 3 AKD and death. Finally, AKD was associated with higher risks of all-cause death, MAKE, and HFH during the 5 years of follow-up (P < 0.001).

Conclusion

AKD after ADHF are associated with adverse outcomes. Our model could help in identification of patients at risk for AKD development, especially in those who did not have an index AKI episode.

A validation study comparing existing prediction models of acute kidney injury in patients with acute heart failure

Acute kidney injury (AKI) is a common complication in acute heart failure (AHF) and is associated with prolonged hospitalization and increased mortality. The aim of this study was to externally validate existing prediction models of AKI in patients with AHF. Data for 10,364 patients hospitalized for acute heart failure between 2008 and 2018 were extracted from the Chang Gung Research Database and analysed. The primary outcome of interest was AKI, defined according to the KDIGO definition. The area under the receiver operating characteristic (AUC) curve was used to assess the discrimination performance of each prediction model. Five existing prediction models were externally validated, and the Forman risk score and the prediction model reported by Wang et al. showed the most favourable discrimination and calibration performance. The Forman risk score had AUCs for discriminating AKI, AKI stage 3, and dialysis within 7 days of 0.696, 0.829, and 0.817, respectively. The Wang et al. model had AUCs for discriminating AKI, AKI stage 3, and dialysis within 7 days of 0.73, 0.858, and 0.845, respectively. The Forman risk score and the Wang et al. prediction model are simple and accurate tools for predicting AKI in patients with AHF.

Renin-Angiotensin-Aldosterone System Optimization for Acute Decompensated Heart Failure Patients (ROAD-HF): Rationale and Design

INTRODUCTION

The long-term benefits of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) on outcomes in patients with chronic congestive heart failure are well-known, making them one of the most widely prescribed medications. However, the administration of ACEIs/ARBs in acute decompensated heart failure (ADHF) can increase the risk of morbidity and mortality secondary to worsening renal function (WRF). A decrease in estimated glomerular filtration rate (eGFR) during the treatment of ADHF has been associated with an increase in mortality proportional to the degree of WRF.

AIM

The aim of our study is to determine whether withholding ACEIs/ARBs during the initial 72 h of admission in patients with ADHF will prevent WRF and allow more effective diuresis.

METHODS

Four hundred and thirty patients will be randomized to the intervention (withholding ACEIs/ARBs) or control (continue/start ACEIs/ARBs) arms for 72 h. Primary outcomes include rates of acute kidney injury (AKI), patient global assessment, and change in kinetic eGFR over 72 h, while secondary outcomes include change in weight, fluid balance, change in signs and symptoms of congestion, change in renal function, change in urinary biomarkers (tissue inhibitor of metalloproteinases 2 [TIMP-2] × insulin-like growth factor-binding protein 7 [IGFBP7]), patients experiencing treatment failure, hospital length of stay (LOS), cost analysis, mortality within 30 days, and hospital readmissions over 30 days and 1 year.

CONCLUSION

This prospective clinical trial will prove if withholding ACEIs/ARBs will prevent AKI in ADHF. It will help us understand the complex interactions between the heart and kidney, and delineate the best treatment strategy for ADHF. Holding ACEIs/ARBs might help preserve renal function, and decrease hospital LOS, readmission rates, and cost of care in ADHF.

REGISTRATION

ClinicalTrials.gov identifier: NCT03695120.

Impact of pre-hospital renal function on the detection of acute kidney injury in acute decompensated heart failure

BACKGROUND

Acute kidney injury (AKI) is a serious complication in patients hospitalized for decompensated heart failure (HF). Currently, AKI definitions consider creatinine levels at admission as reference of baseline renal function (RF). However, renal impairment may already be present at admission. We aimed to study the impact on AKI detection of considering outpatient RF as reference.

METHODS

In a cohort of 458 patients hospitalized for decompensated HF, we studied the occurrence of AKI using the standardized KDIGO criteria and grading (stages: 1, 2, 3), and considering two different definitions according to the RF used as reference or baseline: the latest outpatient measurement prior to admission vs. the first measurement at admission. We compared the prevalence, timing and prognostic value for both AKI definitions.

RESULTS

The definition based on outpatient RF was associated with an increase in overall AKI detection from 20.1% to 33.8% (p < 0.001), and from 3.1% to 5.0% for advanced stages (2-3) (p < 0.001); additionally, 12.5% of patients already had criteria of AKI at admission (36.8% of AKI cases). Both definitions were associated with longer hospital stay. However, only AKI already present at admission, as based on pre-hospital creatinine, was independently associated with all-cause death, in-hospital and after discharge, and death or HF readmission in the follow-up: 1 stage (HR 2.72, 95%CI 1.83-4.06, p < 0.001) and 2-3 stage (HR 7.29, 95%CI, 3.02-17.64, p < 0.001).

CONCLUSIONS

Evaluation of AKI in patients admitted with HF should consider pre-hospital RF, since it improves early identification of AKI and has implications for risk assessment.

Derivation and validation of a prediction score for acute kidney injury secondary to acute myocardial infarction in Chinese patients

Background

Acute kidney injury (AKI) is a major complication of acute myocardial infarction(AMI), which can significantly increase mortality. This study is to analyze the related risk factors and establish a prediction score of acute kidney injury in order to take early measurement for prevention.

Methods

The medical records of 6014 hospitalized patients with AMI in Beijing Anzhen Hospital from January 2010 to December 2016 were retrospectively analyzed. These patients were randomly assigned into two cohorts: one was for the derivation of prediction score (n = 4252) and another for validation (n = 1762). The criterion for AKI was defined as an increase in serum creatinine of ≥ 0.3 mg/dL or ≥ 50% from baseline within 48 h. On the basis of odds ratio obtained from multivariate logistic regression analysis, a prediction score of acute kidney injury after AMI was built up.

Results

In this prediction score, risk score 1 point included hypertension history, heart rate > 100 bpm on admission, peak serum troponin I ≥ 100 μg/L, and time from admission to coronary reperfusion > 120 min; risks score 2 points included Killip classification ≥ class 3 on admission; and maximum dosage of intravenous furosemide ≥ 60 mg/d; risks score 3 points only included shock during hospitalization. In addition, when baseline estimated glomerular filtration rate (eGFR) was less than 90 ml/min·1.73 m², every 10 ml/min·1.73 m² reduction of eGFR increased risk score 1 point. Youden index showed that the best cut-off value for prediction of AKI was 3 points with a sensitivity of 71.1% and specificity 74.2%. The datasets of derivation and validation both displayed adequate discrimination (an area under the ROC curve, 0.79 and 0.81, respectively) and satisfactory calibration (Hosmer–Lemeshow statistic test, P = 0.63 and P = 0.60, respectively).

Conclusions

In conclusion, a prediction score for AKI secondary to AMI in Chinese patients was established, which may help to prevent AKI early.

Multi-perspective predictive modeling for acute kidney injury in general hospital populations using electronic medical records

Objectives

Acute kidney injury (AKI) in hospitalized patients puts them at much higher risk for developing future health problems such as chronic kidney disease, stroke, and heart disease. Accurate AKI prediction would allow timely prevention and intervention. However, current AKI prediction researches pay less attention to model building strategies that meet complex clinical application scenario. This study aims to build and evaluate AKI prediction models from multiple perspectives that reflect different clinical applications.

Materials and Methods

A retrospective cohort of 76 957 encounters and relevant clinical variables were extracted from a tertiary care, academic hospital electronic medical record (EMR) system between November 2007 and December 2016. Five machine learning methods were used to build prediction models. Prediction tasks from 4 clinical perspectives with different modeling and evaluation strategies were designed to build and evaluate the models.

Results

Experimental analysis of the AKI prediction models built from 4 different clinical perspectives suggest a realistic prediction performance in cross-validated area under the curve ranging from 0.720 to 0.764.

Discussion

Results show that models built at admission is effective for predicting AKI events in the next day; models built using data with a fixed lead time to AKI onset is still effective in the dynamic clinical application scenario in which each patient's lead time to AKI onset is different.

Conclusion

To our best knowledge, this is the first systematic study to explore multiple clinical perspectives in building predictive models for AKI in the general inpatient population to reflect real performance in clinical application.

The Japanese clinical practice guideline for acute kidney injury 2016

Acute kidney injury (AKI) is a syndrome which has a broad range of etiologic factors depending on different clinical settings. Because AKI has significant impacts on prognosis in any clinical settings, early detection and intervention is necessary to improve the outcomes of AKI patients. This clinical guideline for AKI was developed by a multidisciplinary approach with nephrology, intensive care medicine, blood purification, and pediatrics. Of note, clinical practice for AKI management which was widely performed in Japan was also evaluated with comprehensive literature search.

The Japanese Clinical Practice Guideline for acute kidney injury 2016

Acute kidney injury (AKI) is a syndrome which has a broad range of etiologic factors depending on different clinical settings. Because AKI has significant impacts on prognosis in any clinical settings, early detection and intervention are necessary to improve the outcomes of AKI patients. This clinical guideline for AKI was developed by a multidisciplinary approach with nephrology, intensive care medicine, blood purification, and pediatrics. Of note, clinical practice for AKI management which was widely performed in Japan was also evaluated with comprehensive literature search.

Renal function and acute heart failure outcome

The interaction between acute heart failure (AHF) and renal dysfunction is complex. Several studies have evaluated the prognostic value of this syndrome. The aim of this systematic review, which includes non-selected samples, was to investigate the impact of different renal function variables on the AHF prognosis. The categories included in the studies reviewed included: creatinine, blood urea nitrogen (BUN), the BUN/creatinine quotient, chronic kidney disease, the formula to estimate the glomerular filtration rate, criteria of acute renal injury and new biomarkers of renal damage such as neutrophil gelatinase-associated lipocalin (NGAL and cystatin c). The basal alterations of the renal function, as well as the acute alterations, transient or not, are related to a worse prognosis in AHF, it is therefore necessary to always have baseline, acute and evolutive renal function parameters.

Calibration drift in regression and machine learning models for acute kidney injury

Objective

Predictive analytics create opportunities to incorporate personalized risk estimates into clinical decision support. Models must be well calibrated to support decision-making, yet calibration deteriorates over time. This study explored the influence of modeling methods on performance drift and connected observed drift with data shifts in the patient population.

Materials and Methods

Using 2003 admissions to Department of Veterans Affairs hospitals nationwide, we developed 7 parallel models for hospital-acquired acute kidney injury using common regression and machine learning methods, validating each over 9 subsequent years.

Results

Discrimination was maintained for all models. Calibration declined as all models increasingly overpredicted risk. However, the random forest and neural network models maintained calibration across ranges of probability, capturing more admissions than did the regression models. The magnitude of overprediction increased over time for the regression models while remaining stable and small for the machine learning models. Changes in the rate of acute kidney injury were strongly linked to increasing overprediction, while changes in predictor-outcome associations corresponded with diverging patterns of calibration drift across methods.

Conclusions

Efficient and effective updating protocols will be essential for maintaining accuracy of, user confidence in, and safety of personalized risk predictions to support decision-making. Model updating protocols should be tailored to account for variations in calibration drift across methods and respond to periods of rapid performance drift rather than be limited to regularly scheduled annual or biannual intervals.

Systematic review of prognostic prediction models for acute kidney injury (AKI) in general hospital populations

OBJECTIVE

Critically appraise prediction models for hospital-acquired acute kidney injury (HA-AKI) in general populations.

DESIGN

Systematic review.

DATA SOURCES

Medline, Embase and Web of Science until November 2016.

ELIGIBILITY

Studies describing development of a multivariable model for predicting HA-AKI in non-specialised adult hospital populations. Published guidance followed for data extraction reporting and appraisal.

RESULTS

14 046 references were screened. Of 53 HA-AKI prediction models, 11 met inclusion criteria (general medicine and/or surgery populations, 474 478 patient episodes) and five externally validated. The most common predictors were age (n=9 models), diabetes (5), admission serum creatinine (SCr) (5), chronic kidney disease (CKD) (4), drugs (diuretics (4) and/or ACE inhibitors/angiotensin-receptor blockers (3)), bicarbonate and heart failure (4 models each). Heterogeneity was identified for outcome definition. Deficiencies in reporting included handling of predictors, missing data and sample size. Admission SCr was frequently taken to represent baseline renal function. Most models were considered at high risk of bias. Area under the receiver operating characteristic curves to predict HA-AKI ranged 0.71-0.80 in derivation (reported in 8/11 studies), 0.66-0.80 for internal validation studies (n=7) and 0.65-0.71 in five external validations. For calibration, the Hosmer-Lemeshow test or a calibration plot was provided in 4/11 derivations, 3/11 internal and 3/5 external validations. A minority of the models allow easy bedside calculation and potential electronic automation. No impact analysis studies were found.

CONCLUSIONS

AKI prediction models may help address shortcomings in risk assessment; however, in general hospital populations, few have external validation. Similar predictors reflect an elderly demographic with chronic comorbidities. Reporting deficiencies mirrors prediction research more broadly, with handling of SCr (baseline function and use as a predictor) a concern. Future research should focus on validation, exploration of electronic linkage and impact analysis. The latter could combine a prediction model with AKI alerting to address prevention and early recognition of evolving AKI.

Predicting AKI in emergency admissions: an external validation study of the acute kidney injury prediction score (APS)

OBJECTIVES

Hospital-acquired acute kidney injury (HA-AKI) is associated with a high risk of mortality. Prediction models or rules may identify those most at risk of HA-AKI. This study externally validated one of the few clinical prediction rules (CPRs) derived in a general medicine cohort using clinical information and data from an acute hospitals electronic system on admission: the acute kidney injury prediction score (APS).

DESIGN, SETTING AND PARTICIPANTS

External validation in a single UK non-specialist acute hospital (2013-2015, 12 554 episodes); four cohorts: adult medical and general surgical populations, with and without a known preadmission baseline serum creatinine (SCr).

METHODS

Performance assessed by discrimination using area under the receiver operating characteristic curves (AUCROC) and calibration.

RESULTS

HA-AKI incidence within 7 days (kidney disease: improving global outcomes (KDIGO) change in SCr) was 8.1% (n=409) of medical patients with known baseline SCr, 6.6% (n=141) in those without a baseline, 4.9% (n=204) in surgical patients with baseline and 4% (n=49) in those without. Across the four cohorts AUCROC were: medical with known baseline 0.65 (95% CIs 0.62 to 0.67) and no baseline 0.71 (0.67 to 0.75), surgical with baseline 0.66 (0.62 to 0.70) and no baseline 0.68 (0.58 to 0.75). For calibration, in medicine and surgical cohorts with baseline SCr, Hosmer-Lemeshow p values were non-significant, suggesting acceptable calibration. In the medical cohort, at a cut-off of five points on the APS to predict HA-AKI, positive predictive value was 16% (13-18%) and negative predictive value 94% (93-94%). Of medical patients with HA-AKI, those with an APS ≥5 had a significantly increased risk of death (28% vs 18%, OR 1.8 (95% CI 1.1 to 2.9), p=0.015).

CONCLUSIONS

On external validation the APS on admission shows moderate discrimination and acceptable calibration to predict HA-AKI and may be useful as a severity marker when HA-AKI occurs. Harnessing linked data from primary care may be one way to achieve more accurate risk prediction.

A Clinical Approach to the Acute Cardiorenal Syndrome

Acute kidney injury is a frequent complication of acute heart failure syndromes, portending an adverse prognosis. Acute cardiorenal syndrome represents a unique form of acute kidney injury specific to acute heart failure syndromes. The pathophysiology of acute cardiorenal syndrome involves renal venous congestion, ineffective forward flow, and impaired renal autoregulation caused by neurohormonal activation. Biomarkers reflecting different aspects of acute cardiorenal syndrome pathophysiology may allow patient phenotyping to inform prognosis and treatment. Adjunctive vasoactive, neurohormonal, and diuretic therapies may relieve congestive symptoms and/or improve renal function, but no single therapy has been proved to reduce mortality in acute cardiorenal syndrome.

Utilizing electronic health records to predict acute kidney injury risk and outcomes: workgroup statements from the 15(th) ADQI Consensus Conference

The data contained within the electronic health record (EHR) is "big" from the standpoint of volume, velocity, and variety. These circumstances and the pervasive trend towards EHR adoption have sparked interest in applying big data predictive analytic techniques to EHR data. Acute kidney injury (AKI) is a condition well suited to prediction and risk forecasting; not only does the consensus definition for AKI allow temporal anchoring of events, but no treatments exist once AKI develops, underscoring the importance of early identification and prevention. The Acute Dialysis Quality Initiative (ADQI) convened a group of key opinion leaders and stakeholders to consider how best to approach AKI research and care in the "Big Data" era. This manuscript addresses the core elements of AKI risk prediction and outlines potential pathways and processes. We describe AKI prediction targets, feature selection, model development, and data display.

Clinical prediction scores for type 1 cardiorenal syndrome derived and validated in chinese cohorts

Type 1 cardiorenal syndrome is one of the major diseases threatening human life in China. The incidence of acute kidney injury (AKI) associated with acute heart failure (AHF), acute myocardial infarction (AMI), cardiac surgery, and coronary angiography has been reported to be 32.2, 14.7, 40.2, and 4.5%, respectively. In the past 2 years, we derived and validated 4 risk scores for the prediction of AKI associated with the above acute heart diseases as well as for examination and treatment in Chinese cohorts. A univariable comparison and a subsequent multivariate logistic regression analysis of the potential predictive variables of AKI in the derivation set were conducted and used to establish the prediction scores, which were then verified in the validation set. The area under the receiver operating characteristic (ROC) curve and the Hosmer-Lemeshow goodness-of-fit statistic test were performed to assess the discrimination and calibration of the prediction scores, respectively. These 4 prediction scores all showed adequate discrimination (area under the ROC curve, ≥0.70) and good calibration (p > 0.05). Both Forman's risk score (for AKI associated with AHF) and Mehran's risk score (for AKI associated with coronary angiography) are widely applied around the world. The external validation of these 2 risk scores was performed in our patients, but their discriminative power was quite low (area under the ROC curve, 0.65 and 0.57, respectively). Therefore, these prediction scores derived from Chinese cohorts might be more accurate than those derived from different races when they are applied in Chinese patients.

Nephrology in china

China has a large population and a high prevalence of chronic kidney disease (CKD). The increasing incidence of obesity and type 2 diabetes mellitus, coupled with an ageing population, will exacerbate the burden of CKD unless effective control and prevention strategies are implemented. The unmet challenges of managing the growing number of patients with end-stage renal disease (ESRD) in China are reflected by the lower rate of patients receiving dialysis relative to many Western countries, owing to a lack of financial and clinical resources, and inequalities in access to health care across regions and populations. The feasibility of expanding peritoneal dialysis is being examined, and ongoing health-care reforms provide an invaluable opportunity to improve the status and quality of dialysis for patients with ESRD in China. The Chinese Society of Nephrology (CSN) advocates for efforts focused on preventing CKD coupled with early detection, treatment, and adequate follow-up to reduce mortality and the long-term burden of CKD. In addition, rapid advances in nephrology research, from basic science to clinical epidemiology, as well as broad communication and collaboration between the CSN and other international nephrology societies, will promote the development of nephrology in China.