A new scoring model for the prediction of mortality in patients with acute kidney injury

Assessment of severity scoring systems for predicting mortality in critically ill patients receiving continuous renal replacement therapy

The incidence of acute kidney injury (AKI) is increasing every year and many patients with AKI admitted to the intensive care unit (ICU) require continuous renal replacement therapy (CRRT). This study compared and analyzed severity scoring systems to assess their suitability in predicting mortality in critically ill patients receiving CRRT. Data from 612 patients receiving CRRT in four ICUs of the Korea University Medical Center between January 2016 and November 2018 were retrospectively collected. The mean age of all patients was 67.6 ± 14.8 years, and the proportion of males was 59.6%. The endpoints were in-hospital mortality and 7-day mortality from the day of CRRT initiation to the date of death. The Program to Improve Care in Acute Renal Disease (PICARD), Demirjian's, Acute Physiology and Chronic Health Evaluation (APACHE) II, Simplified Acute Physiology Score (SAPS) 3, Sequential Organ Failure Assessment (SOFA), Multiple Organ Dysfunction Score (MODS), and Liano's scores were used to predict mortality. The in-hospital and 7-day mortality rates in the study population were 72.7% and 45.1%, respectively. The area under the receiver operator characteristic curve (AUROC) revealed the highest discrimination ability for Demirjian's score (0.770), followed by Liano's score (0.728) and APACHE II (0.710). The AUROC curves for the SAPS 3, MODS, and PICARD were 0.671, 0.665, and 0.658, respectively. The AUROC of Demirjian's score was significantly higher than that of the other scores, except for Liano's score. The Hosmer-Lemeshow test on Demirjian's score showed a poor fit in our analysis; however, it was more acceptable than general severity scores. Kidney-specific severity scoring systems showed better performance in predicting mortality in critically ill patients receiving CRRT than general severity scoring systems.

Associations of systolic blood pressure and in-hospital mortality in critically ill patients with acute kidney injury

PURPOSE

Although systolic blood pressure (SBP) is associated with acute renal injury (AKI), the relationship between baseline SBP and prognosis in critically ill patients with AKI is unclear. We aimed to assess the linearity and profile of the relationship between SBP at intensive care unit (ICU) admission and in-hospital mortality in these patients.

METHODS

Data of AKI patients in the ICU settings were extracted from the Medical Information Mart for Intensive Care III database. The association between seven SBP categories (< 100, 100-109, 110-119, 120-129, 130-139, 140-149, and ≥ 150 mmHg) and all-cause in-hospital mortality was assessed by Cox proportional hazard models. Restricted cubic spline analysis for the multivariate Cox model was performed to explore the shape of the relationship between SBP and mortality.

RESULTS

A total of 24,202 patients with AKI were included in this study. A typically U-shaped relationship was found between SBP at admission and in-hospital mortality. Among all SBP categories, the lowest risk of death was observed in patients with SBP around 110-119 mmHg, whereas the highest was noted in patients with extremely low SBP (< 100 mmHg), followed by those with extremely high SBP (≥ 150 mmHg). SBP showed a significant interaction with vasopressor use and AKI stage in relation to the risk of in-hospital mortality.

CONCLUSIONS

SBP upon admission showed a non-linear association with all-cause in-hospital mortality in critically ill patients with AKI. Patients with low or high SBP show an increased risk of mortality compared to patients with normal SBP.

Characteristics and outcomes of COVID-19 patients assisted by intensivists and nonintensivists

OBJECTIVE:

The aim of this study was to assess the outcomes of critically ill patients with COVID-19 in an intensive care unit seen by a care team formed by intensive and nonintensive physicians and treatment guided by processes and protocols linked to the “choosing wisely” concept, comparing them with similar data recently published.

METHODS:

An observational cohort including adult patients with COVID-19 admitted to the intensive care unit of Hospital Independence between August 2020 and August 2021. Inclusion criteria were 18 years of age or older and there were no exclusion criteria.

RESULTS:

The study included 449 patients, of which 64.1% were referred from the ward, 21.6% from emergency rooms, and 14.2% from another hospital (continuity of attendance). The overall mortality was 48.5%, occurring mainly in the elderly and or those undergoing mechanical ventilation. We did not find any associations between different strata of body mass index and mortality. In the multivariate analysis, the time elapsed between the onset of symptoms and hospital admission, mechanical ventilation, C-reactive protein value at the end of the first week in the intensive care unit, and renal failure were independently associated with mortality. Vaccinated people comprised 8.8% of the sample, with no differences in mortality among the different vaccines, and 13.4% of patients underwent palliative treatment.

CONCLUSIONS:

Patients admitted for acute respiratory syndrome due to SARS-CoV-2 are severe and have a high mortality rate, mainly if submitted to invasive mechanical ventilation. The emergence of acute renal failure marks an especially severe subgroup with increased mortality. Processes and protocols linked to the “choosing-wisely” concept seemed to significantly benefit our intensive care unit since it had a large contingent of nonspecialist physicians.

Nomogram Prediction Model of Serum Chloride and Sodium Ions on the Risk of Acute Kidney Injury in Critically Ill Patients

Purpose

This study aims to investigate the effect of serum chloride and sodium ions on AKI occurrence in ICU patients, and further constructs a prediction model containing these factors to explore the predictive value of these ions in AKI.

Methods

The clinical information of patients admitted to ICU of Beijing Friendship Hospital Affiliated to Capital Medical University was collected for retrospective analysis. Logistic regression analysis was used to analyzing the influencing factors. A nomogram for predicting AKI risk was constructed with R software and validated by repeated sampling. Afterwards, the effectiveness and accuracy of the model were tested and evaluated.

Results

A total of 446 cases met the requirements of this study, of which 178 developed AKI during their stay in ICU, with an incidence rate of 39.9%. Hypernatremia, heart failure, sepsis, APACHE II score, and initial creatinine value and BE value at ICU admission before the diagnosis of AKI were identified as independent risk factors for developing AKI during ICU stay. These predictors were incorporated into the nomogram of AKI risk in critically ill patients, which was constructed by using R software. Receiver operating characteristic curve analysis was further used and showed that the area under the curve of the model was 0.7934 (95% CI 0.742–0.8447), indicating that the model had an ideal value. Finally, further evaluated its clinical effectiveness. The clinical effect curve and decision curve showed that most areas of the decision curve of this model were greater than 0, indicating that this model owned a certain clinical effectiveness.

Conclusion

The nomogram based on hypernatremia, heart failure, sepsis, APACHE II score, and initial creatinine and BE value in ICU can predict the individualized risk of AKI with satisfactory distinguishability and accuracy.

Prediction of Mortality Risk After Ischemic Acute Kidney Injury With a Novel Prognostic Model: A Multivariable Prediction Model Development and Validation Study

Background and Objectives:

Acute kidney injury (AKI) that results from ischemia is a common clinical syndrome and correlates with high morbidity and mortality among hospitalized patients. However, a clinical tool to predict mortality risk of ischemic AKI is not available. In this study, we aimed to develop and validate models to predict the 30-day and 1-year mortality risk of hospitalized patients with ischemic AKI.

Methods

A total of 1,836 admissions with ischemic AKI were recruited from 277,898 inpatients admitted to three affiliated tertiary general hospitals of Central South University in China between January 2015 and December 2015. Patients in the final analysis were followed up for 1 year. Study patients were randomly divided in a 7:3 ratio to form the training cohort and validation cohort. Multivariable regression analyses were used for developing mortality prediction models.

Results

Hepatorenal syndrome, shock, central nervous system failure, Charlson comorbidity index (≥2 points), mechanical ventilation, renal function at discharge were independent risk factors for 30-day mortality after ischemic AKI, while malignancy, sepsis, heart failure, liver failure, Charlson comorbidity index (≥2 points), mechanical ventilation, and renal function at discharge were predictors for 1-year mortality. The area under the receiver operating characteristic curves (AUROCs) of 30-day prediction model were 0.878 (95% confidence interval (CI): 0.849-0.908) in the training cohort and 0.867 (95% CI: 0.820–0.913) in the validation cohort. The AUROCs of the 1-year mortality prediction in the training and validation cohort were 0.803 (95% CI: 0.772–0.834) and 0.788 (95% CI: 0.741–0.835), respectively.

Conclusion

Our easily applied prediction models can effectively identify individuals at high mortality risk within 30 days or 1 year in hospitalized patients with ischemic AKI. It can guide the optimal clinical management to minimize mortality after an episode of ischemic AKI.

Prediction Models for One-Year Survival of Adult Patients with Acute Kidney Injury: A Longitudinal Study Based on the Data from the Medical Information Mart for Intensive Care III Database

Acute kidney injury (AKI) is a common complication of acute illnesses with unfavorable outcomes. This cohort study aimed at constructing prediction models for one-year survival in adult AKI patients based on prognostic nutritional index (PNI), platelet-to-lymphocyte ratio (PLR), neutrophil percentage-to-albumin ratio (NPAR), or neutrophil-to-lymphocyte ratio (NLR), respectively. In total, 6050 patients from Medical Information Mart for Intensive Care III (MIMIC-III) were involved. The least absolute shrinkage and selection operator (LASSO) regression was utilized to screen possible covariates. The samples were randomly divided into the training set and the testing set at a ratio of 7.5 : 2.5, and the prediction models were constructed in the training set by random forest. The prediction values of the models were measured via sensitivity, specificity, negative prediction value (NPV), positive prediction value (PPV), area under the curve (AUC), and accuracy. We found that NLR (OR = 1.261, 95% CI: 1.145–1.388), PLR (OR = 1.295, 95% CI: 1.152–1.445), and NPAR (OR = 1.476, 95% CI: 1.261–1.726) were associated with an increased risk, while PNI (OR = 0.035, 95% CI: 0.020–0.059) was associated with a decreased risk of one-year mortality in AKI patients. The AUC was 0.964 (95% CI: 0.959–0.969) in the training set based on PNI, age, gender, length of stay (LOS) in hospital, platelets (PLT), ethnicity, LOS in ICU, systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate, glucose, AKI stage, atrial fibrillation (AF), vasopressor, renal replacement therapy (RRT), and mechanical ventilation. The testing set was applied as the internal validation of the model with an AUC of 0.778 (95% CI: 0.754–0.801). In conclusion, PNI accompanied by age, gender, ethnicity, SBP, DBP, heart rate, PLT, glucose, AF, RRT, mechanical ventilation, vasopressor, AKI stage, LOS in ICU, and LOS in hospital exhibited a good predictive value for one-year mortality of AKI patients.

Application of interpretable machine learning for early prediction of prognosis in acute kidney injury

Background

This study aimed to develop an algorithm using the explainable artificial intelligence (XAI) approaches for the early prediction of mortality in intensive care unit (ICU) patients with acute kidney injury (AKI).

Methods

This study gathered clinical data with AKI patients from the Medical Information Mart for Intensive Care IV (MIMIC-IV) in the US between 2008 and 2019. All the data were further randomly divided into a training cohort and a validation cohort. Seven machine learning methods were used to develop the models for assessing in-hospital mortality. The optimal model was selected based on its accuracy and area under the curve (AUC). The SHapley Additive exPlanation (SHAP) values and Local Interpretable Model-Agnostic Explanations (LIME) algorithm were utilized to interpret the optimal model.

Results

A total of 22,360 patients with AKI were finally enrolled in this study (median age, 69.5 years; female, 42.8%). They were randomly split into a training cohort (16770, 75%) and a validation cohort (5590, 25%). The eXtreme Gradient Boosting (XGBoost) model achieved the best performance with an AUC of 0.890. The SHAP values showed that Glasgow Coma Scale (GCS), blood urea nitrogen, cumulative urine output on Day 1 and age were the top 4 most important variables contributing to the XGBoost model. The LIME algorithm was used to explain the individualized predictions.

Conclusions

Machine-learning models based on clinical features were developed and validated with great performance for the early prediction of a high risk of death in patients with AKI.

Mortality Prediction in Patients with Severe Acute Kidney Injury Requiring Renal Replacement Therapy

Background and Objective: Acute kidney injury (AKI) remains a serious health condition around the world, and is related to high morbidity, mortality, longer hospitalization duration and worse long-term outcomes. The aim of our study was to estimate the significant related factors for poor outcomes of patients with severe AKI requiring renal replacement therapy (RRT). Materials and Methods: We retrospectively analyzed data from patients (n = 573) with severe AKI requiring RRT within a 5-year period and analyzed the outcomes on discharge from the hospital. We also compared the clinical data of the surviving and non-surviving patients and examined possible related factors for poor patient outcomes. Logistic regression was used to analyze the odds ratio for patient mortality and its related factors. Results: In 32.5% (n = 186) of the patients, the renal function improved and RRT was stopped, 51.7% (n = 296) of the patients died, and 15.9% (n = 91) of the patients remained dialysis-dependent on the day of discharge from the hospital. During the period of 5 years, the outcomes of the investigated patients did not change statistically significantly. Administration of vasopressors, aminoglycosides, sepsis, pulmonary edema, oliguria, artificial pulmonary ventilation (APV), patient age ≥ 65 y, renal cause of AKI, AKI after cardiac surgery, a combination of two or more RRT methods, dysfunction of three or more organs, systolic blood pressure (BP) ≤ 120 mmHg, diastolic BP ≤ 65 mmHg, and Sequential Organ Failure Assessment (SOFA) score on the day of the first RRT procedure ≥ 7.5 were related factors for lethal patient outcome. Conclusions: The mortality rate among patients with severe AKI requiring RRT is very high—52%. Patient death was significantly predicted by the causes of AKI (sepsis, cardiac surgery), clinical course (oliguria, pulmonary edema, hypotension, acidosis, lesion of other organs) and the need for a continuous renal replacement therapy.

Effect of ondansetron on reducing ICU mortality in patients with acute kidney injury

The purpose of this study is to identify medications with potentially beneficial effects on decreasing mortality in patients with acute kidney injury (AKI) while in the intensive care unit (ICU). We used logistic regression to investigate associations between medications received and ICU mortality in patients with AKI in the MIMIC III database. Drugs associated with reduced mortality were then validated using the eICU database. Propensity score matching (PSM) was used for matching the patients’ baseline severity of illness followed by a chi-square test to calculate the significance of drug use and mortality. Finally, we examined gene expression signatures to explore the drug’s molecular mechanism on AKI. While several drugs demonstrated potential beneficial effects on reducing mortality, most were used for potentially fatal illnesses (e.g. antibiotics, cardiac medications). One exception was found, ondansetron, a drug without previously identified life-saving effects, has correlation with lower mortality among AKI patients. This association was confirmed in a subsequent analysis using the eICU database. Based on the comparison of gene expression signatures, the presumed therapeutic effect of ondansetron may be elicited through the NF-KB pathway and JAK-STAT pathway. Our findings provide real-world evidence to support clinical trials of ondansetron for treatment of AKI.

Establishment of a novel scoring model for mortality risk prediction in HIV-infected patients with cryptococcal meningitis

Background

Cryptococcal meningitis (CM) remains a leading cause of death in HIV-infected patients, despite advances in CM diagnostic and therapeutic strategies. This study was performed with the aim to develop and validate a novel scoring model to predict mortality risk in HIV-infected patients with CM (HIV/CM).

Methods

Data on HIV/CM inpatients were obtained from a Multicenter Cohort study in China. Independent risk factors associated with mortality were identified based on data from 2013 to 2017, and a novel scoring model for mortality risk prediction was established. The bootstrapping statistical method was used for internal validation. External validation was performed using data from 2018 to 2020.

Results

We found that six predictors, including age, stiff neck, impaired consciousness, intracranial pressure, CD4⁺ T-cell count, and urea levels, were associated with poor prognosis in HIV/CM patients. The novel scoring model could effectively identify HIV/CM patients at high risk of death on admission (area under curve 0.876; p<0.001). When the cut-off value of 5.5 points or more was applied, the sensitivity and specificity was 74.1 and 83.8%, respectively. Our scoring model showed a good discriminatory ability, with an area under the curve of 0.879 for internal validation via bootstrapping, and an area under the curve of 0.886 for external validation.

Conclusions

Our developed scoring model of six variables is simple, convenient, and accurate for screening high-risk patients with HIV/CM, which may be a useful tool for physicians to assess prognosis in HIV/CM inpatients.

Development of mortality prediction model in the elderly hospitalized AKI patients

Acute kidney injury (AKI) correlates with increased health-care costs and poor outcomes in older adults. However, there is no good scoring system to predict mortality within 30-day, 1-year after AKI in older adults. We performed a retrospective analysis screening data of 53,944 hospitalized elderly patients (age > 65 years) from multi-centers in China. 944 patients with AKI (acute kidney disease) were included and followed up for 1 year. Multivariable regression analysis was used for developing scoring models in the test group (a randomly 70% of all the patients). The established models have been verified in the validation group (a randomly 30% of all the patients). Model 1 that consisted of the risk factors for death within 30 days after AKI had accurate discrimination (The area under the receiver operating characteristic curves, AUROC: 0.90 (95% CI 0.875–0.932)) in the test group, and performed well in the validation groups (AUROC: 0.907 (95% CI 0.865–0.949)). The scoring formula of all-cause death within 1 year (model 2) is a seven-variable model including AKI type, solid tumor, renal replacement therapy, acute myocardial infarction, mechanical ventilation, the number of organ failures, and proteinuria. The area under the receiver operating characteristic (AUROC) curves of model 2 was > 0.80 both in the test and validation groups. Our newly established risk models can well predict the risk of all-cause death in older hospitalized AKI patients within 30 days or 1 year.

Recent advances in nanotherapeutics for the treatment and prevention of acute kidney injury

Acute kidney injury (AKI) is a serious kidney disease without specific medications currently except for expensive dialysis treatment. Some potential drugs are limited due to their high hydrophobicity, poor in vivo stability, low bioavailability and possible adverse effects. Besides, kidney-targeted drugs are not common and small molecules are cleared too quickly to achieve effective drug concentrations in injured kidneys. These problems limit the development of pharmacological therapy for AKI. Nanotherapeutics based on nanotechnology have been proved to be an emerging and promising treatment strategy for AKI, which may solve the pharmacological therapy dilemma. More and more nanotherapeutics with different physicochemical properties are developed to efficiently deliver drugs, increase accumulation and control release of drugs in injury kidneys and also directly as effective antioxidants. Here, we discuss the recent nanotherapeutics applied in the treatment and prevention of AKI with improved effectiveness and few side effects.

Role of Circular RNA in Kidney-Related Diseases

The kidney is vital in maintaining fluid, electrolyte, and acid–base balance. Kidney-related diseases, which are an increasing public health issue, can happen to people of any age and at any time. Circular RNAs (circRNAs) are endogenous RNA that are produced by selective RNA splicing and are involved in progression of various diseases. Studies have shown that various kidney diseases, including renal cell carcinoma, acute kidney injury, and chronic kidney disease, are linked to circRNAs. This review outlines the characteristics and biological functions of circRNAs and discusses specific studies that provide insights into the function and potential of circRNAs for application in the diagnosis and treatment of kidney-related diseases.

Acute kidney injury in patients with malignant middle cerebral artery infarction undergoing hyperosmolar therapy with mannitol

PURPOSE

To assess the kidney safety profile of mannitol in patients with malignant middle cerebral artery (MCA) infarction.

MATERIAL AND METHODS

We studied consecutive patients with malignant MCA infarction (01/2008-01/2018). Malignant MCA infarction was defined according to DESTINY criteria. We compared clinical endpoints including acute kidney injury (AKI; according to Kidney Disease: Improving Global Outcomes [KDIGO]) and dialysis between patients with and without mannitol. Multivariable model was built to explore predictor variables of AKI and in-hospital death.

RESULTS

Overall, 219 patients with malignant MCA infarction were included. Mannitol was administered in 93/219 (42.5%) patients with an average dosage of 650 g (250-950 g). Patients treated with mannitol more frequently suffered from AKI (39.8% vs. 11.9%; p < 0.001) and required hemodialysis (7.5% vs. 0.8%; p = 0.01) than patients without mannitol. At discharge, more patients in the mannitol group had persistent AKI than control patients (23.7% vs. 6.4%, p < 0.001). In multivariable model, mannitol emerged as independent predictor of AKI (OR 5.02, 95%CI 2.36-10.69; p < 0.001).

CONCLUSIONS

Acute kidney injury appears to be a frequent complication of hyperosmolar therapy with mannitol in patients with malignant MCA infarction. Given the lack of evidence supporting effectiveness of mannitol in these patients, its routine use should be carefully considered.

Novel risk models to predict acute kidney disease and its outcomes in a Chinese hospitalized population with acute kidney injury

Acute kidney disease (AKD) is a state between acute kidney injury (AKI) and chronic kidney disease (CKD), but the prognosis of AKD is unclear and there are no risk-prediction tools to identify high-risk patients. 2,556 AKI patients were selected from 277,898 inpatients of three affiliated hospitals of Central South University from January 2015 to December 2015. The primary point was whether AKI patients developed AKD. The endpoint was death or end stage renal disease (ESRD) 90 days after AKI diagnosis. Multivariable Cox regression was used for 90-day mortality and two prediction models were established by using multivariable logistic regression. Our study found that the incidence of AKD was 53.17% (1,359/2,556), while the mortality rate and incidence of ESRD in AKD cohort was 19.13% (260/1,359) and 3.02% (41/1,359), respectively. Furthermore, adjusted hazard ratio of mortality for AKD versus no AKD was 1.980 (95% CI 1.427–2.747). In scoring model 1, age, gender, hepatorenal syndromes, organic kidney diseases, oliguria or anuria, respiratory failure, blood urea nitrogen (BUN) and acute kidney injury stage were independently associated with AKI progression into AKD. In addition, oliguria or anuria, respiratory failure, shock, central nervous system failure, malignancy, RDW-CV ≥ 13.7% were independent risk factors for death or ESRD in AKD patients in scoring model 2 (goodness-of fit, P₁ = 0.930, P₂ = 0.105; AUROC₁ = 0.879 (95% CI 0.862–0.896), AUROC₂ = 0.845 (95% CI 0.813–0.877), respectively). Thus, our study demonstrated AKD was independently associated with increased 90-day mortality in hospitalized AKI patients. A new prediction model system was able to predict AKD following AKI and 90-day prognosis of AKD patients to identify high-risk patients.

Post-contrast acute kidney injury in a hospitalized population: short-, mid-, and long-term outcome and risk factors for adverse events

Objectives

To investigate the prognosis including major adverse kidney events within 30 days (MAKE30) and 90-day and 1-year adverse outcome in hospitalized patients with post-contrast acute kidney injury (PC-AKI) to identify high-risk factors.

Methods

This retrospective observational study included 288 PC-AKI patients selected from 277,898 patients admitted to hospitals from January 2015 to December 2015. PC-AKI was defined according to the 2018 guideline of European Society of Urogenital Radiology. Multivariable Cox regression and logistic regression analyses were used to analyze main outcome and risk factors.

Results

PC-AKI patients with AKI stage ≥ 2 had much higher incidence of MAKE30 than those with AKI stage 1 (RR = 7.027, 95% CI 4.918–10.039). Persistent renal dysfunction, heart failure, central nervous system failure, baseline eGFR < 60 mL/min/1.73 m², oliguria or anuria, blood urea nitrogen ≥ 7.14 mmol/L, respiratory failure, and shock were independent risk factors of 90-day or 1-year adverse prognosis (p < 0.05). Compared with transient renal dysfunction, PC-AKI patients with persistent renal dysfunction had a higher all-cause mortality rate (RR = 3.768, 95% CI 1.612–8.810; RR = 4.106, 95% CI 1.765–9.551) as well as combined endpoints of death, chronic kidney disease, or end-stage renal disease (OR = 3.685, 95% CI 1.628–8.340; OR = 5.209, 95% CI 1.730–15.681) within 90 days or 1 year.

Conclusions

PC-AKI is not always a transient, benign creatininopathy, but can result in adverse outcome. AKI stage is independently correlated to MAKE30 and persistent renal dysfunction may exaggerate the risk of long-term adverse events.

Key Points

• PC-AKI can result in adverse outcome such as persistent renal dysfunction, dialysis, chronic kidney disease (CKD), end-stage renal disease (ESRD), or death.

• AKI stage is independently correlated to MAKE30.

• Persistent renal dysfunction may exaggerate the risk of long-term adverse events.

Electronic supplementary material

The online version of this article (10.1007/s00330-020-06690-3) contains supplementary material, which is available to authorized users.

Non-coding RNA-Associated ceRNA Networks in a New Contrast-Induced Acute Kidney Injury Rat Model

Contrast-induced acute kidney injury (CI-AKI) is a severe complication of intravascular applied radial contrast media, and recent progress in interventional therapy and angiography has revived interest in explaining detailed mechanisms and developing effective treatment. Recent studies have indicated a potential link between CI-AKI and microRNA (miRNA). However, the potential non-coding RNA-associated-competing endogenous RNA (ceRNA) pairs involved in CI-AKI still remain unclear. In this study, we systematically explored the circRNA or lncRNA-associated-ceRNA mechanism in a new rat model of CI-AKI through deep RNA sequencing. The results revealed that the expression of 38 circRNAs, 12 lncRNAs, 13 miRNAs and 127 mRNAs were significantly dysregulated. We performed Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses for mRNAs with significantly different expression and then constructed comprehensive circRNA or lncRNA-associated ceRNA networks in kidney of CI-AKI rats. Thereafter, two constructed ceRNA regulatory pathways in this CI-AKI rat model—novel_circ_0004153/rno-miR-144-3p/Gpnmb or Naglu and LNC_000343/rno-miR-1956-5p/KCP—were validated by real-time qPCR. This study is the first one to provide a systematic dissection of non-coding RNA-associated ceRNA profiling in kidney of CI-AKI rats. The selected non-coding RNA-associated ceRNA networks provide new insight for the underlying mechanism and may profoundly affect the diagnosis and therapy of CI-AKI.

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.