Integrated multiomics implicates dysregulation of ECM and cell adhesion pathways as drivers of severe COVID-associated kidney injury

COVID-19 has been a significant public health concern for the last four years; however, little is known about the mechanisms that lead to severe COVID-associated kidney injury. In this multicenter study, we combined quantitative deep urinary proteomics and machine learning to predict severe acute outcomes in hospitalized COVID-19 patients. Using a 10-fold cross-validated random forest algorithm, we identified a set of urinary proteins that demonstrated predictive power for both discovery and validation set with 87% and 79% accuracy, respectively. These predictive urinary biomarkers were recapitulated in non-COVID acute kidney injury revealing overlapping injury mechanisms. We further combined orthogonal multiomics datasets to understand the mechanisms that drive severe COVID-associated kidney injury. Functional overlap and network analysis of urinary proteomics, plasma proteomics and urine sediment single-cell RNA sequencing showed that extracellular matrix and autophagy-associated pathways were uniquely impacted in severe COVID-19. Differentially abundant proteins associated with these pathways exhibited high expression in cells in the juxtamedullary nephron, endothelial cells, and podocytes, indicating that these kidney cell types could be potential targets. Further, single-cell transcriptomic analysis of kidney organoids infected with SARS-CoV-2 revealed dysregulation of extracellular matrix organization in multiple nephron segments, recapitulating the clinically observed fibrotic response across multiomics datasets. Ligand-receptor interaction analysis of the podocyte and tubule organoid clusters showed significant reduction and loss of interaction between integrins and basement membrane receptors in the infected kidney organoids. Collectively, these data suggest that extracellular matrix degradation and adhesion-associated mechanisms could be a main driver of COVID-associated kidney injury and severe outcomes.

Comparison of predicting cardiovascular disease hospitalization using individual, ZIP code-derived, and machine learning model-predicted educational attainment in New York City

BACKGROUND

Area-level social determinants of health (SDOH) based on patients' ZIP codes or census tracts have been commonly used in research instead of individual SDOHs. To our knowledge, whether machine learning (ML) could be used to derive individual SDOH measures, specifically individual educational attainment, is unknown.

METHODS

This is a retrospective study using data from the Mount Sinai BioMe Biobank. We included participants that completed a validated questionnaire on educational attainment and had home addresses in New York City. ZIP code-level education was derived from the American Community Survey matched for the participant's gender and race/ethnicity. We tested several algorithms to predict individual educational attainment from routinely collected clinical and demographic data. To evaluate how using different measures of educational attainment will impact model performance, we developed three distinct models for predicting cardiovascular (CVD) hospitalization. Educational attainment was imputed into models as either survey-derived, ZIP code-derived, or ML-predicted educational attainment.

RESULTS

A total of 20,805 participants met inclusion criteria. Concordance between survey and ZIP code-derived education was 47%, while the concordance between survey and ML model-predicted education was 67%. A total of 13,715 patients from the cohort were included into our CVD hospitalization prediction models, of which 1,538 (11.2%) had a history of CVD hospitalization. The AUROC of the model predicting CVD hospitalization using survey-derived education was significantly higher than the model using ZIP code-level education (0.77 versus 0.72; p < 0.001) and the model using ML model-predicted education (0.77 versus 0.75; p < 0.001). The AUROC for the model using ML model-predicted education was also significantly higher than that using ZIP code-level education (p = 0.003).

CONCLUSION

The concordance of survey and ZIP code-level educational attainment in NYC was low. As expected, the model utilizing survey-derived education achieved the highest performance. The model incorporating our ML model-predicted education outperformed the model relying on ZIP code-derived education. Implementing ML techniques can improve the accuracy of SDOH data and consequently increase the predictive performance of outcome models.

Relationship Between Cognitive Impairment and Depression Among Middle Aged and Older Adults in Primary Care

Objectives: To determine rates of previously undetected cognitive impairment among patients with depression in primary care. Methods: Patients ages 55 and older with no documented history of dementia or mild cognitive impairment were recruited from primary care practices in New York City, NY and Chicago, IL (n = 855). Cognitive function was assessed with the Montreal Cognitive Assessment (MoCA) and depression with the Patient Health Questionnaire-8. Results: The mean age was 66.8 (8.0) years, 45.3% were male, 32.7% Black, and 29.2% Latinx. Cognitive impairment increased with severity of depression: 22.9% in persons with mild depression, 27.4% in moderate depression and 41.8% in severe depression (p = .0002). Severe depression was significantly associated with cognitive impairment in multivariable analysis (standardized β = -.11, SE = 0.33, p < .0001). Discussion: Depression was strongly associated with previously undetected cognitive impairment. Primary care clinicians should consider screening, or expand their screening, for both conditions.

Machine learning models to predict end-stage kidney disease in chronic kidney disease stage 4

INTRODUCTION

End-stage kidney disease (ESKD) is associated with increased morbidity and mortality. Identifying patients with stage 4 CKD (CKD4) at risk of rapid progression to ESKD remains challenging. Accurate prediction of CKD4 progression can improve patient outcomes by improving advanced care planning and optimizing healthcare resource allocation.

METHODS

We obtained electronic health record data from patients with CKD4 in a large health system between January 1, 2006, and December 31, 2016. We developed and validated four models, including Least Absolute Shrinkage and Selection Operator (LASSO) regression, random forest, eXtreme Gradient Boosting (XGBoost), and artificial neural network (ANN), to predict ESKD at 3 years. We utilized area under the receiver operating characteristic curve (AUROC) to evaluate model performances and utilized Shapley additive explanation (SHAP) values and plots to define feature dependence of the best performance model.

RESULTS

We included 3,160 patients with CKD4. ESKD was observed in 538 patients (21%). All approaches had similar AUROCs; ANN yielded the highest AUROC (0.77; 95%CI 0.75 to 0.79) and LASSO regression (0.77; 95%CI 0.75 to 0.79), followed by random forest (0.76; 95% CI 0.74 to 0.79), and XGBoost (0.76; 95% CI 0.74 to 0.78).

CONCLUSIONS

We developed and validated several models for near-term prediction of kidney failure in CKD4. ANN, random forest, and XGBoost demonstrated similar predictive performances. Using this suite of models, interventions can be customized based on risk, and population health and resources appropriately allocated.

Joint Modeling of Social Determinants and Clinical Factors to Define Subphenotypes in Out-of-Hospital Cardiac Arrest Survival: Cluster Analysis

Background

Machine learning clustering offers an unbiased approach to better understand the interactions of complex social and clinical variables via integrative subphenotypes, an approach not studied in out-of-hospital cardiac arrest (OHCA).

Objective

We conducted a cluster analysis for a cohort of OHCA survivors to examine the association of clinical and social factors for mortality at 1 year.

Methods

We used a retrospective observational OHCA cohort identified from Medicare claims data, including area-level social determinants of health (SDOH) features and hospital-level data sets. We applied k-means clustering algorithms to identify subphenotypes of beneficiaries who had survived an OHCA and examined associations of outcomes by subphenotype.

Results

We identified 27,028 unique beneficiaries who survived to discharge after OHCA. We derived 4 distinct subphenotypes. Subphenotype 1 included a distribution of more urban, female, and Black beneficiaries with the least robust area-level SDOH measures and the highest 1-year mortality (2375/4417, 53.8%). Subphenotype 2 was characterized by a greater distribution of male, White beneficiaries and had the strongest zip code-level SDOH measures, with 1-year mortality at 49.9% (4577/9165). Subphenotype 3 had the highest rates of cardiac catheterization at 34.7% (1342/3866) and the greatest distribution with a driving distance to the index OHCA hospital from their primary residence >16.1 km at 85.4% (8179/9580); more were also discharged to a skilled nursing facility after index hospitalization. Subphenotype 4 had moderate median household income at US $51,659.50 (IQR US $41,295 to $67,081) and moderate to high median unemployment at 5.5% (IQR 4.2%-7.1%), with the lowest 1-year mortality (1207/3866, 31.2%). Joint modeling of these features demonstrated an increased hazard of death for subphenotypes 1 to 3 but not for subphenotype 4 when compared to reference.

Conclusions

We identified 4 distinct subphenotypes with differences in outcomes by clinical and area-level SDOH features for OHCA. Further work is needed to determine if individual or other SDOH domains are specifically tied to long-term survival after OHCA.

Peripheral Transcriptomics in Acute and Long-Term Kidney Dysfunction in SARS-CoV2 Infection

Background

Acute kidney injury (AKI) is common in hospitalized patients with SARS-CoV2 infection despite vaccination and leads to long-term kidney dysfunction. However, peripheral blood molecular signatures in AKI from COVID-19 and their association with long-term kidney dysfunction are yet unexplored.

Methods

In patients hospitalized with SARS-CoV2, we performed bulk RNA sequencing using peripheral blood mononuclear cells(PBMCs). We applied linear models accounting for technical and biological variability on RNA-Seq data accounting for false discovery rate (FDR) and compared functional enrichment and pathway results to a historical sepsis-AKI cohort. Finally, we evaluated the association of these signatures with long-term trends in kidney function.

Results

Of 283 patients, 106 had AKI. After adjustment for sex, age, mechanical ventilation, and chronic kidney disease (CKD), we identified 2635 significant differential gene expressions at FDR<0.05. Top canonical pathways were EIF2 signaling, oxidative phosphorylation, mTOR signaling, and Th17 signaling, indicating mitochondrial dysfunction and endoplasmic reticulum (ER) stress. Comparison with sepsis associated AKI showed considerable overlap of key pathways (48.14%). Using follow-up estimated glomerular filtration rate (eGFR) measurements from 115 patients, we identified 164/2635 (6.2%) of the significantly differentiated genes associated with overall decrease in long-term kidney function. The strongest associations were 'autophagy', 'renal impairment via fibrosis', and 'cardiac structure and function'.

Conclusions

We show that AKI in SARS-CoV2 is a multifactorial process with mitochondrial dysfunction driven by ER stress whereas long-term kidney function decline is associated with cardiac structure and function and immune dysregulation. Functional overlap with sepsis-AKI also highlights common signatures, indicating generalizability in therapeutic approaches.

SIGNIFICANCE STATEMENT

Peripheral transcriptomic findings in acute and long-term kidney dysfunction after hospitalization for SARS-CoV2 infection are unclear. We evaluated peripheral blood molecular signatures in AKI from COVID-19 (COVID-AKI) and their association with long-term kidney dysfunction using the largest hospitalized cohort with transcriptomic data. Analysis of 283 hospitalized patients of whom 37% had AKI, highlighted the contribution of mitochondrial dysfunction driven by endoplasmic reticulum stress in the acute stages. Subsequently, long-term kidney function decline exhibits significant associations with markers of cardiac structure and function and immune mediated dysregulation. There were similar biomolecular signatures in other inflammatory states, such as sepsis. This enhances the potential for repurposing and generalizability in therapeutic approaches.

Development and Validation of a Formative Assessment Tool for Nephrology Fellows' Clinical Reasoning

BACKGROUND

Diagnostic errors are commonly driven by failures in clinical reasoning. Deficits in clinical reasoning are common among graduate medical learners, including nephrology fellows. We created and validated an instrument to assess clinical reasoning in a national cohort of nephrology fellows and established performance thresholds for remedial coaching.

METHODS

Experts in nephrology education and clinical reasoning remediation designed an instrument to measure clinical reasoning through a written patient encounter note from a web-based, simulated AKI consult. The instrument measured clinical reasoning in three domains: problem representation, differential diagnosis with justification, and diagnostic plan with justification. Inter-rater reliability was established in a pilot cohort ( n =7 raters) of first-year nephrology fellows using a two-way random effects agreement intraclass correlation coefficient model. The instrument was then administered to a larger cohort of first-year fellows to establish performance standards for coaching using the Hofstee method ( n =6 raters).

RESULTS

In the pilot cohort, there were 15 fellows from four training program, and in the study cohort, there were 61 fellows from 20 training programs. The intraclass correlation coefficients for problem representation, differential diagnosis, and diagnostic plan were 0.90, 0.70, and 0.50, respectively. Passing thresholds (% total points) in problem representation, differential diagnosis, and diagnostic plan were 59%, 57%, and 62%, respectively. Fifty-nine percent ( n =36) met the threshold for remedial coaching in at least one domain.

CONCLUSIONS

We provide validity evidence for a simulated AKI consult for formative assessment of clinical reasoning in nephrology fellows. Most fellows met criteria for coaching in at least one of three reasoning domains, demonstrating a need for learner assessment and instruction in clinical reasoning.

Deep Learning on Electrocardiograms for Prediction of In-hospital Intradialytic Hypotension in Patients with ESKD

Intradialytic hypotension is common in patients who are on hemodialysis. We applied deep learning techniques to ECGs to predict patients at risk of IDH. The performance of the model was good with an AUC of 0.763 and AUPRC of 0.35.

Evaluation of Plasma Biomarkers to Predict Major Adverse Kidney Events in Hospitalized Patients With COVID-19

RATIONALE & OBJECTIVE

Patients hospitalized with COVID-19 are at increased risk for major adverse kidney events (MAKE). We sought to identify plasma biomarkers predictive of MAKE in patients hospitalized with COVID-19.

STUDY DESIGN

Prospective cohort study.

SETTING & PARTICIPANTS

A total of 576 patients hospitalized with COVID-19 between March 2020 and January 2021 across 3 academic medical centers.

EXPOSURE

Twenty-six plasma biomarkers of injury, inflammation, and repair from first available blood samples collected during hospitalization.

OUTCOME

MAKE, defined as KDIGO stage 3 acute kidney injury (AKI), dialysis-requiring AKI, or mortality up to 60 days.

ANALYTICAL APPROACH

Cox proportional hazards regression to associate biomarker level with MAKE. We additionally applied the least absolute shrinkage and selection operator (LASSO) and random forest regression for prediction modeling and estimated model discrimination with time-varying C index.

RESULTS

The median length of stay for COVID-19 hospitalization was 9 (IQR, 5-16) days. In total, 95 patients (16%) experienced MAKE. Each 1 SD increase in soluble tumor necrosis factor receptor 1 (sTNFR1) and sTNFR2 was significantly associated with an increased risk of MAKE (adjusted HR [AHR], 2.30 [95% CI, 1.86-2.85], and AHR, 2.26 [95% CI, 1.73-2.95], respectively). The C index of sTNFR1 alone was 0.80 (95% CI, 0.78-0.84), and the C index of sTNFR2 was 0.81 (95% CI, 0.77-0.84). LASSO and random forest regression modeling using all biomarkers yielded C indexes of 0.86 (95% CI, 0.83-0.89) and 0.84 (95% CI, 0.78-0.91), respectively.

LIMITATIONS

No control group of hospitalized patients without COVID-19.

CONCLUSIONS

We found that sTNFR1 and sTNFR2 are independently associated with MAKE in patients hospitalized with COVID-19 and can both also serve as predictors for adverse kidney outcomes.

PLAIN-LANGUAGE SUMMARY

Patients hospitalized with COVID-19 are at increased risk for long-term adverse health outcomes, but not all patients suffer long-term kidney dysfunction. Identification of patients with COVID-19 who are at high risk for adverse kidney events may have important implications in terms of nephrology follow-up and patient counseling. In this study, we found that the plasma biomarkers soluble tumor necrosis factor receptor 1 (sTNFR1) and sTNFR2 measured in hospitalized patients with COVID-19 were associated with a greater risk of adverse kidney outcomes. Along with clinical variables previously shown to predict adverse kidney events in patients with COVID-19, both sTNFR1 and sTNFR2 are also strong predictors of adverse kidney outcomes.

Patient-Specific Pharmacokinetics and Dasatinib Nephrotoxicity

BACKGROUND

Dasatinib has been associated with nephrotoxicity. We sought to examine the incidence of proteinuria on dasatinib and determine potential risk factors that may increase dasatinib-associated glomerular injury.

METHODS

We examined glomerular injury through urine albumin-creatinine ratio (UACR) in 82 patients with chronic myelogenous leukemia who were on tyrosine-kinase inhibitor therapy for at least 90 days. t tests were used to compare mean differences in UACR, while regression analysis was used to assess the effects of drug parameters on proteinuria development while on dasatinib. We assayed plasma dasatinib pharmacokinetics using tandem mass spectroscopy and further described a case study of a patient who experienced nephrotic-range proteinuria while on dasatinib.

RESULTS

Participants treated with dasatinib ( n =32) had significantly higher UACR levels (median 28.0 mg/g; interquartile range, 11.5-119.5) than participants treated with other tyrosine-kinase inhibitors ( n =50; median 15.0 mg/g; interquartile range, 8.0-35.0; P < 0.001). In total, 10% of dasatinib users exhibited severely increased albuminuria (UACR >300 mg/g) versus zero in other tyrosine-kinase inhibitors. Average steady-state concentrations of dasatinib were positively correlated with UACR ( ρ =0.54, P = 0.03) and duration of treatment ( P = 0.003). There were no associations with elevated BP or other confounding factors. In the case study, kidney biopsy revealed global glomerular damage with diffuse foot process effacement that recovered on termination of dasatinib treatment.

CONCLUSIONS

Exposure to dasatinib was associated with a significant chance of developing proteinuria compared with other similar tyrosine-kinase inhibitors. Dasatinib plasma concentration significantly correlated with higher risk of developing proteinuria while receiving dasatinib.

PODCAST

This article contains a podcast at https://dts.podtrac.com/redirect.mp3/www.asn-online.org/media/podcast/CJASN/2023_09_08_CJN0000000000000219.mp3.

Proteomic characterization of acute kidney injury in patients hospitalized with SARS-CoV2 infection

BACKGROUND

Acute kidney injury (AKI) is a known complication of COVID-19 and is associated with an increased risk of in-hospital mortality. Unbiased proteomics using biological specimens can lead to improved risk stratification and discover pathophysiological mechanisms.

METHODS

Using measurements of ~4000 plasma proteins in two cohorts of patients hospitalized with COVID-19, we discovered and validated markers of COVID-associated AKI (stage 2 or 3) and long-term kidney dysfunction. In the discovery cohort (N = 437), we identified 413 higher plasma abundances of protein targets and 30 lower plasma abundances of protein targets associated with COVID-AKI (adjusted p < 0.05). Of these, 62 proteins were validated in an external cohort (p < 0.05, N = 261).

RESULTS

We demonstrate that COVID-AKI is associated with increased markers of tubular injury (NGAL) and myocardial injury. Using estimated glomerular filtration (eGFR) measurements taken after discharge, we also find that 25 of the 62 AKI-associated proteins are significantly associated with decreased post-discharge eGFR (adjusted p < 0.05). Proteins most strongly associated with decreased post-discharge eGFR included desmocollin-2, trefoil factor 3, transmembrane emp24 domain-containing protein 10, and cystatin-C indicating tubular dysfunction and injury.

CONCLUSIONS

Using clinical and proteomic data, our results suggest that while both acute and long-term COVID-associated kidney dysfunction are associated with markers of tubular dysfunction, AKI is driven by a largely multifactorial process involving hemodynamic instability and myocardial damage.

Dasatinib nephrotoxicity correlates with patient-specific pharmacokinetics

Introduction

Dasatinib has been associated with nephrotoxicity. We sought to examine the incidence of proteinuria on dasatinib and determine potential risk factors that may increase dasatinib-associated glomerular injury.

Methods

We examine glomerular injury via urine albumin-to-creatinine ratio (UACR) in 101 chronic myelogenous leukemia patients who were on tyrosine-kinase inhibitor (TKI) therapy for at least 90 days. We assay plasma dasatinib pharmacokinetics using tandem mass spectroscopy, and further describe a case study of a patient who experienced nephrotic-range proteinuria while on dasatinib.

Results

Patients treated with dasatinib (n= 32) had significantly higher UACR levels (median 28.0 mg/g, IQR 11.5 - 119.5) than patients treated with other TKIs (n=50; median 15.0 mg/g, IQR 8.0 - 35.0; p < 0.001). In total, 10% of dasatinib users exhibited severely increased albuminuria (UACR > 300 mg/g) versus zero in other TKIs. Average steady state concentrations of dasatinib were positively correlated with UACR (ρ = 0.54, p = 0.03) as well as duration of treatment ( p =0.003). There were no associations with elevated blood pressure or other confounding factors. In the case study, kidney biopsy revealed global glomerular damage with diffuse foot process effacement that recovered upon termination of dasatinib treatment.

Conclusions

Exposure to dasatinib is associated a significant chance of developing proteinuria compared to other similar TKIs. Dasatinib plasma concentration significantly correlates with increased risk of developing proteinuria while receiving dasatinib. Screening for renal dysfunction and proteinuria is strongly advised for all dasatinib patients.

Major cardiovascular events and subsequent risk of kidney failure with replacement therapy: a CKD Prognosis Consortium study

AIMS

Chronic kidney disease (CKD) increases risk of cardiovascular disease (CVD). Less is known about how CVD associates with future risk of kidney failure with replacement therapy (KFRT).

METHODS AND RESULTS

The study included 25 903 761 individuals from the CKD Prognosis Consortium with known baseline estimated glomerular filtration rate (eGFR) and evaluated the impact of prevalent and incident coronary heart disease (CHD), stroke, heart failure (HF), and atrial fibrillation (AF) events as time-varying exposures on KFRT outcomes. Mean age was 53 (standard deviation 17) years and mean eGFR was 89 mL/min/1.73 m2, 15% had diabetes and 8.4% had urinary albumin-to-creatinine ratio (ACR) available (median 13 mg/g); 9.5% had prevalent CHD, 3.2% prior stroke, 3.3% HF, and 4.4% prior AF. During follow-up, there were 269 142 CHD, 311 021 stroke, 712 556 HF, and 605 596 AF incident events and 101 044 (0.4%) patients experienced KFRT. Both prevalent and incident CVD were associated with subsequent KFRT with adjusted hazard ratios (HRs) of 3.1 [95% confidence interval (CI): 2.9-3.3], 2.0 (1.9-2.1), 4.5 (4.2-4.9), 2.8 (2.7-3.1) after incident CHD, stroke, HF and AF, respectively. HRs were highest in first 3 months post-CVD incidence declining to baseline after 3 years. Incident HF hospitalizations showed the strongest association with KFRT [HR 46 (95% CI: 43-50) within 3 months] after adjustment for other CVD subtype incidence.

CONCLUSION

Incident CVD events strongly and independently associate with future KFRT risk, most notably after HF, then CHD, stroke, and AF. Optimal strategies for addressing the dramatic risk of KFRT following CVD events are needed.

Proteomic Characterization of Acute Kidney Injury in Patients Hospitalized with SARS-CoV2 Infection

Background Acute kidney injury (AKI) is a known complication of COVID-19 and is associated with an increased risk of in-hospital mortality. Unbiased proteomics using biological specimens can lead to improved risk stratification and discover pathophysiological mechanisms. Methods Using measurements of ~4000 plasma proteins in two cohorts of patients hospitalized with COVID-19, we discovered and validated markers of COVID-associated AKI (stage 2 or 3) and long-term kidney dysfunction. In the discovery cohort (N= 437), we identified 413 higher plasma abundances of protein targets and 40 lower plasma abundances of protein targets associated with COVID-AKI (adjusted p <0.05). Of these, 62 proteins were validated in an external cohort (p <0.05, N =261). Results We demonstrate that COVID-AKI is associated with increased markers of tubular injury ( NGAL ) and myocardial injury. Using estimated glomerular filtration (eGFR) measurements taken after discharge, we also find that 25 of the 62 AKI-associated proteins are significantly associated with decreased post-discharge eGFR (adjusted p <0.05). Proteins most strongly associated with decreased post-discharge eGFR included desmocollin-2 , trefoil factor 3 , transmembrane emp24 domain-containing protein 10 , and cystatin-C indicating tubular dysfunction and injury. Conclusions Using clinical and proteomic data, our results suggest that while both acute and long-term COVID-associated kidney dysfunction are associated with markers of tubular dysfunction, AKI is driven by a largely multifactorial process involving hemodynamic instability and myocardial damage.

Impact of Vaccination on Major Adverse Cardiovascular Events in Patients With COVID-19 Infection

Taken from the largest U.S. cohort of patients with SARS-CoV2, our results demonstrate the association of even partial vaccination with lower risk of MACE after SARS-CoV-2 infection.

Unplanned 30-Day Readmission Rates for Autosomal Dominant Polycystic Kidney Disease: Insight from the Nationwide Readmissions Database

INTRODUCTION

Among end-stage kidney disease (ESKD) patients on dialysis with autosomal dominant polycystic kidney disease (ADPKD), relatively little is known about the epidemiology and risk factors for 30-day readmissions in the USA. Therefore, we evaluated the 30-day unplanned readmission rates and predictors and inpatient care costs among ESKD patients with and without ADPKD using a nationally representative, all-payer database.

METHODS

We utilized the Nationwide Readmissions Database from 2013 to 2018 to identify patients admitted for ESKD on dialysis with and without ADPKD using ICD-9 and ICD-10 codes. The primary outcome was a 30-day, unplanned readmission rate. Secondary outcomes were readmission reasons and timing, mortality, cost of hospitalization and rehospitalization, and adjusted predictors of readmissions. We used χ2 tests, t tests, and Wilcoxon rank-sum tests for descriptive analyses and survey logistic regression to calculate adjusted odds ratios and 95% confidence intervals for associations with readmissions adjusting for confounders.

RESULTS

From 2013 to 2018, in a cohort of 1,404,144 hospitalizations with ESKD on dialysis as the primary and secondary diagnosis on index admission, there were 8,213 (0.58%) patients with ADPKD and 1,395,932 patients without ADPKD. Those who had ADPKD during index admissions had fewer 30 days readmissions (18.8 vs. 23.8%, p < 0.0001). The cost of hospitalizations and readmissions in ESKD on-dialysis patients with ADPKD was higher than non-ADPKD patients. Compared to ESKD patients without ADPKD who were readmitted, readmitted ADPKD patients were more likely to be younger with a lower Elixhauser Comorbidity Index (ECI) score; have received kidney transplant, lower source of income, elective index admissions, private insurance; and be discharged routinely, admitted in hospitals with larger bed size, in teaching hospitals, and less likely to get admitted through the emergency department. Younger age (<75 years), higher ECI score, longer length of stay, Medicare and Medicaid insurance, self-pay, discharge to a short-term hospital, specialized care, home health care, and against medical advice were associated with significantly increased odds of readmission. ADPKD patients were 31% less likely to get readmitted and 43% less likely to die during readmissions.

DISCUSSION/CONCLUSION

Nationwide, ESKD on-dialysis patients with ADPKD were less likely to have 30-day readmission than patients without ADPKD. Inpatient mortality during readmissions in patients admitted with ESKD on dialysis was lower with ADPKD as compared to those without ADPKD at the cost of higher health care expenses.

Higher Symptom Frequency and Severity After the Long Interdialytic Interval in Patients on Maintenance Intermittent Hemodialysis

Introduction

Patients on intermittent hemodialysis (HD) have a high symptom burden. Though studies report higher hospitalizations and mortality after the long interdialytic interval, whether symptoms vary based on the interdialytic interval is unclear.

Methods

This is a prospective observational study of patients over the age of 18 who received in-center HD. Patients were surveyed on the presence and severity of 20 different symptoms at the end of 12 HD sessions. Wilcoxon signed-rank test was used for comparison of severity for each symptom by the interval. Multivariable generalized estimating equation with Poisson regression by repeated measure method was used to determine the association of interdialytic interval and symptom frequency while adjusting for potential confounders.

Results

From the 97 patients enrolled, the most common symptoms were fatigue (60.8%), cramping (58.8%), and dry skin (52.6%). There was large variability in the frequency of symptoms, ranging 0% to 8% of treatments. The most severe symptoms were bone pain (mean severity score 2.2±0.9) and diarrhea (mean severity score 2.2±0.7). Eight of the 20 symptoms were significantly more common after the long interdialytic interval including fatigue (22% vs. 15%, P < 0.001) and cramping (21% vs. 16%, P = 0.003). The long interval had a 37% higher incidence rate for symptoms compared to the short interval even after adjustment. Results were similar across genders.

Conclusion

Symptoms are more common after the long interdialytic interval. Clinical assessment and research evaluating patient symptoms need to be cognizant of when patients are surveyed or include the length of interdialytic interval as a confounding variable.

Natural Language Processing in Nephrology

Unstructured data in the electronic health records contain essential patient information. Natural language processing (NLP), teaching a computer to read, allows us to tap into these data without needing the time and effort of manual chart abstraction. The core first step for all NLP algorithms is preprocessing the text to identify the core words that differentiate the text while filtering out the noise. Traditional NLP uses a rule-based approach, applying grammatical rules to infer meaning from the text. Newer NLP approaches use machine learning/deep learning which can infer meaning without explicitly being programmed. NLP use in nephrology research has focused on identifying distinct disease processes, such as CKD, and extraction of patient-oriented outcomes such as symptoms with high sensitivity. NLP can identify patient features from clinical text associated with acute kidney injury and progression of CKD. Lastly, inclusion of features extracted using NLP improved the performance of risk-prediction models compared to models that only use structured data. Implementation of NLP algorithms has been slow, partially hindered by the lack of external validation of NLP algorithms. However, NLP allows for extraction of key patient characteristics from free text, an infrequently used resource in nephrology.

Genome-first recall of healthy individuals by polygenic risk score reveals differences in coronary artery calcium

Genetic risk for coronary artery disease (CAD) is commonly measured with polygenic risk scores (PRS); yet, the relationship of atherosclerotic burden with PRS in healthy individuals not at high clinical risk for CAD (ie, without a high pooled cohort equations [PCE] score) is unknown. Here, we implemented a novel recall-by-PRS strategy to measure coronary artery calcium (CAC) scores prospectively in 53 healthy individuals with extreme high PRS (median [IQR] PRS = 94% [83-98]) and low PRS (median [IQR] PRS = 3.6% [1.2-10]). The high PRS group was associated with a 2.8-fold greater CAC than the low PRS group, adjusted for age, sex, BMI, smoking, and statin use, and had a 6.7-fold greater proportion of individuals with CAC exceeding 300 HU. These findings reveal that extreme PRS tracks with CAD risk even in those without high clinical risk and demonstrate proof of principle for recall-by-PRS approaches that should be assessed prospectively in larger trials.

The promise of artificial intelligence for kidney pathophysiology

PURPOSE OF REVIEW

We seek to determine recent advances in kidney pathophysiology that have been enabled or enhanced by artificial intelligence. We describe some of the challenges in the field as well as future directions.

RECENT FINDINGS

We first provide an overview of artificial intelligence terminologies and methodologies. We then describe the use of artificial intelligence in kidney diseases to discover risk factors from clinical data for disease progression, annotate whole slide imaging and decipher multiomics data. We delineate key examples of risk stratification and prognostication in acute kidney injury (AKI) and chronic kidney disease (CKD). We contextualize these applications in kidney disease oncology, one of the subfields to benefit demonstrably from artificial intelligence using all if these approaches. We conclude by elucidating technical challenges and ethical considerations and briefly considering future directions.

SUMMARY

The integration of clinical data, patient derived data, histology and proteomics and genomics can enhance the work of clinicians in providing more accurate diagnoses and elevating understanding of disease progression. Implementation research needs to be performed to translate these algorithms to the clinical setting.

Automated Determination of Left Ventricular Function Using Electrocardiogram Data in Patients on Maintenance Hemodialysis

BACKGROUND AND OBJECTIVES

Left ventricular ejection fraction is disrupted in patients on maintenance hemodialysis and can be estimated using deep learning models on electrocardiograms. Smaller sample sizes within this population may be mitigated using transfer learning.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

We identified patients on hemodialysis with transthoracic echocardiograms within 7 days of electrocardiogram using diagnostic/procedure codes. We developed four models: (1) trained from scratch in patients on hemodialysis, (2) pretrained on a publicly available set of natural images (ImageNet), (3) pretrained on all patients not on hemodialysis, and (4) pretrained on patients not on hemodialysis and fine-tuned on patients on hemodialysis. We assessed the ability of the models to classify left ventricular ejection fraction into clinically relevant categories of ≤40%, 41% to ≤50%, and >50%. We compared performance by area under the receiver operating characteristic curve.

RESULTS

We extracted 705,075 electrocardiogram:echocardiogram pairs for 158,840 patients not on hemodialysis used for development of models 3 and 4 and n=18,626 electrocardiogram:echocardiogram pairs for 2168 patients on hemodialysis for models 1, 2, and 4. The transfer learning model achieved area under the receiver operating characteristic curves of 0.86, 0.63, and 0.83 in predicting left ventricular ejection fraction categories of ≤40% (n=461), 41%-50% (n=398), and >50% (n=1309), respectively. For the same tasks, model 1 achieved area under the receiver operating characteristic curves of 0.74, 0.55, and 0.71, respectively; model 2 achieved area under the receiver operating characteristic curves of 0.71, 0.55, and 0.69, respectively, and model 3 achieved area under the receiver operating characteristic curves of 0.80, 0.51, and 0.77, respectively. We found that predictions of left ventricular ejection fraction by the transfer learning model were associated with mortality in a Cox regression with an adjusted hazard ratio of 1.29 (95% confidence interval, 1.04 to 1.59).

CONCLUSION

A deep learning model can determine left ventricular ejection fraction for patients on hemodialysis following pretraining on electrocardiograms of patients not on hemodialysis. Predictions of low ejection fraction from this model were associated with mortality over a 5-year follow-up period.

PODCAST

This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2022_06_06_CJN16481221.mp3.

Proteomic Characterization of Acute Kidney Injury in Patients Hospitalized with SARS-CoV2 Infection.. [PMID: 36093350 PMCID: PMC9460972 DOI: 10.1101/2021.12.09.21267548]

Acute kidney injury (AKI) is a known complication of COVID-19 and is associated with an increased risk of in-hospital mortality. Unbiased proteomics using biological specimens can lead to improved risk stratification and discover pathophysiological mechanisms. Using measurements of ∼4000 plasma proteins in two cohorts of patients hospitalized with COVID-19, we discovered and validated markers of COVID-associated AKI (stage 2 or 3) and long-term kidney dysfunction. In the discovery cohort (N= 437), we identified 413 higher plasma abundances of protein targets and 40 lower plasma abundances of protein targets associated with COVID-AKI (adjusted p <0.05). Of these, 62 proteins were validated in an external cohort (p <0.05, N =261). We demonstrate that COVID-AKI is associated with increased markers of tubular injury (NGAL) and myocardial injury. Using estimated glomerular filtration (eGFR) measurements taken after discharge, we also find that 25 of the 62 AKI-associated proteins are significantly associated with decreased post-discharge eGFR (adjusted p <0.05). Proteins most strongly associated with decreased post-discharge eGFR included desmocollin-2, trefoil factor 3, transmembrane emp24 domain-containing protein 10, and cystatin-C indicating tubular dysfunction and injury. Using clinical and proteomic data, our results suggest that while both acute and long-term COVID-associated kidney dysfunction are associated with markers of tubular dysfunction, AKI is driven by a largely multifactorial process involving hemodynamic instability and myocardial damage.

Prevalence and Outcomes Associated with Hyperuricemia in Hospitalized Patients with COVID-19

INTRODUCTION

Coronavirus 2019 (COVID-19) can increase catabolism and result in hyperuricemia. Uric acid (UA) potentially causes kidney damage by alteration of renal autoregulation, inhibition of endothelial cell proliferation, cell apoptosis, activation of the pro-inflammatory cascade, and crystal deposition. Hyperuricemia in patients with COVID-19 may contribute to acute kidney injury (AKI) and poor outcomes.

METHODS

We included 834 patients with COVID-19 who were >18 years old and hospitalized for >24 h in the Mount Sinai Health System and had at least 1 measurement of serum UA. We examined the association between the first serum UA level and development of acute kidney injury (AKI, defined by KDIGO criteria), major adverse kidney events (MAKE, defined by a composite of all-cause in-hospital mortality or dialysis or 100% increase in serum creatinine from baseline), as well as markers of inflammation and cardiac injury.

RESULTS

Among the 834 patients, the median age was 66 years, 42% were women, and the median first serum UA was 5.9 mg/dL (interquartile range 4.5-8.8). Overall, 60% experienced AKI, 52% experienced MAKE, and 32% died during hospitalization. After adjusting for demographics, comorbidities, and laboratory values, a doubling in serum UA was associated with increased AKI (odds ratio [OR] 2.8, 95% confidence interval [CI] 1.9-4.1), MAKE (OR 2.5, 95% CI 1.7-3.5), and in-hospital mortality (OR 1.7, 95% CI 1.3-2.3). Higher serum UA levels were independently associated with a higher level of procalcitonin (β, 0.6; SE 0.2) and troponin I (β, 1.2; SE 0.2) but were not associated with serum ferritin, C-reactive protein, and interleukin-6.

CONCLUSION

In patients admitted to the hospital for COVID-19, higher serum UA levels were independently associated with AKI, MAKE, and in-hospital mortality in a dose-dependent manner. In addition, hyperuricemia was associated with higher procalcitonin and troponin I levels.

The Association between Prevalence of Peritoneal Dialysis versus Hemodialysis and Patients' Distance to Dialysis-Providing Facilities

Background

Accessibility to dialysis facilities plays a central role when deciding on a patient's long-term dialysis modality. Studies investigating the effect of distance to nearest dialysis-providing unit on modality choice have yielded conflicting results. We set out to investigate the association between patients' dialysis modality and both the driving and straight-line distances to the closest HD- and PD-providing units.

Methods

All patients with ESKD who initiated in-center HD and PD in 2017, were 18-90 years old, and were on dialysis for ≥30 days were included. Patients in residence zip codes in nonconterminous United States or lived >90 miles from the nearest HD-providing unit were excluded.

Results

A total of 102,247 patients in the United States initiated in-center HD and PD in 2017. Compared with patients on HD, patients on PD had longer driving distances to their nearest PD unit (4.4 versus 3.4 miles; P<0.001). Patients who lived >30 miles from the nearest HD unit were more likely to be on PD if the nearest PD unit was a distance equal to/less than that of the HD unit. PD utilization increased with increasing distance from patients' homes to the nearest HD unit. No change in this association was found regardless of if the PD unit was farther from/closer than the nearest HD unit. This association was not seen with straight-line distance analysis.

Conclusions

With increasing distances from the nearest dialysis-providing units (HD or PD), PD utilization increased. Using driving distance rather than straight-line distance affects data analysis and outcomes. Increasing the number of PD units may have a limited effect on increasing PD utilization.

Discriminatory accuracy of the SOFA score for determining clinical decompensation in patients presenting with COVID-19

Introduction

While the global dissemination of vaccines targeting the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in a decline in the incidence of infections, the case fatality rates have remained relative stable. A major objective of managing hospitalized patients with documented or suspected COVID-19 infection is the rapid identification of features associated with severe illness using readily available laboratory tests and clinical tools. The sequential organ failure assessment (SOFA) score is a validated tool to facilitate the identification of patients at risk of dying from sepsis.

Purpose

The aim of this study was to assess the discriminatory accuracy of the SOFA score in predicting clinical decompensation in patients hospitalized with COVID-19 infection.

Methods

We conducted a retrospective analysis at a three-hospital health system, comprised of one tertiary and two community hospitals, located in the Chicago metropolitan area. All patients had positive SARS-CoV-2 testing and were hospitalized for COVID-19 infection. The primary outcome was clinical decompensation, defined as the composite endpoint of death, ICU admission, or need for intubation. We utilized the most abnormal laboratory values observed during the admission to calculate the SOFA score. Receiver Operating Curves (ROC) were then constructed to determine the sensitivity and specificity of SOFA scores.

Results

Between March 1st and May 31st 2020, 1029 patients were included in our analysis with 367 patients meeting the study endpoint. The median SOFA score was 2.0 IQR (Q1, Q3 1,4) for the entire cohort. Patients who had in-hospital mortality had a median SOFA score of 4.0 (Q1,Q3 3,7). In patients that met the primary composite endpoint, the median SOFA score was 3.0, IQR (Q1, Q3 2,6). The ROC was 0.776 (95% CI 0.746–0.806, p<0.01).

Conclusion

The SOFA score demonstrates strong discriminatory accuracy for prediction of clinical decompensation in patients presenting with COVID-19 at our urban hospital system.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Loyola University Medical Center

Predictors of clinical decompensation in patients presenting with COVID-19 in an urban hospital health system

Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in a pandemic which has infected more than 128 million people and led to over 2.8 million deaths worldwide. Although the introduction of efficacious vaccines has led to overall declines in the incidence of SARS-CoV-2 infection, there has been a recent increase in infections once more due to the appearance of mutant strains with higher virulence. It therefore remains vital to identify predictors of poor outcomes in this patient population.

Purpose

The objective of our study was to identify predictors of prolonged hospitalization, intensive care unit (ICU) admission, intubation, and death in patients infected with SARS-CoV-2.

Methods

We conducted a retrospective analysis of all patients hospitalized with SARS-CoV-2 at our health system that includes one tertiary care center and two community hospitals located in the Chicago metropolitan area. The main outcome was a composite endpoint of hospitalization >28 days, ICU admission, intubation, and death.

Explanatory variables associated with the primary outcome in the bivariate analysis (p<0.05) were included in the multivariable logistic regression model. Statistical analysis was performed using IBM SPSS 25.0.

Results

Between March 1, 2020 and May 31, 2020, 1029 patients hospitalized with SARS-CoV-2 were included in our analysis. Of these patients, 379 met the composite endpoint. Baseline demographics are described in Table 1. Of note, our cohort consisted of a predominantly minority patient population including 47% Hispanic, 17% African American, 16% Caucasian, and 16% other.

In bivariate analysis, age, hypertension, tobacco and alcohol abuse, obesity, coronary artery disease, arrhythmias, valvular heart disease, dyslipidemia, hypertension, stroke, diabetes, documented thrombosis, troponin, CRP, ESR, ferritin, LDH, BNP, D-dimer >5x the upper limit of normal, lactate, and right ventricular outflow tract velocity time integral <9.5 were significant.

After multivariable adjustment, explanatory variables associated with the composite endpoint included troponin (OR 2.36; 95% CI 1.08–5.17, p 0.03), D-dimer (OR 1.5; 95% CI 1.23–1.98, p<0.01, lactate (OR 1.58; 95% CI 1.28–1.95, p<0.01), and documented thrombosis (OR 3.56; 95% CI 1.30–8.70, p<.05). Race was not a predictor of poor outcomes in the bivariate or multivariate analysis (Table 2).

Conclusions

In a large urban cohort with a predominantly minority population, we identified several clinical predictors of poor outcomes. Of note, race was not a predictor of the primary endpoint in this study. While recent literature has demonstrated worse outcomes among racial minorities infected with SARS-CoV-2, our data suggests these variations are related to social determinants of health rather than biologic causes.

Funding Acknowledgement

Type of funding sources: Public hospital(s). Main funding source(s): Loyola University Medical Center

Estimating premorbid intelligence in persons with traumatic brain injury: an examination of the Test of Premorbid Functioning

Knowledge of intelligence is essential for interpreting cognitive performance following traumatic brain injury (TBI). The Test of Premorbid Functioning (ToPF), a word reading test co-normed with the Wechsler Adult Intelligence Scale 4th Edition (WAIS-IV), was examined as a tool for estimating premorbid intelligence in persons with a history of TBI. Fifty-two participants with mild, moderate, or severe TBI were administered the ToPF and WAIS-IV between two weeks and 19 months post-injury. The independent ability of the ToPF/demographic score and the Verbal Comprehension Index (VCI) to predict WAIS-IV Full Scale IQ (FSIQ) was examined, as were discrepancies between ToPF and WAIS-IV scores within and between participants. The ToPF/demographic predicted FSIQ accounted for a significant proportion of variability in actual FSIQ, above and beyond that accounted for by education or time since injury. ToPF and WAIS-IV scores did not differ by injury severity. In our sample, the ToPF/demographic predicted FSIQ underestimated intelligence in a substantial portion of our participants (31%), particularly in those with high average to superior intelligence. Finally, VCI scores were more predictive of actual FSIQ than the ToPF/demographic predicted FSIQ. The ToPF frequently underestimated post-injury intelligence and is therefore not accurately measuring premorbid intelligence in our sample, particularly in those with above average to superior intelligence. Clinicians are encouraged to administer the entire WAIS-IV, or at minimum the VCI subtests, for a more accurate measure of intelligence in those with above average intelligence and history of TBI.

P-296 Examining the link between environmental toxin exposure and uterine leiomyoma: a systematic review

Study question

Is there an association between exposure to certain environmental toxins and the prevalence of uterine leiomyoma in women?

Summary answer

Some evidence was obtained to suggest an association between phthalate esters, bisphenol A, heavy metals, persistent organic pollutants and the prevalence of uterine fibroids.

What is known already

Environmental toxins are naturally occurring, or human made chemicals that can act as endocrine disrupting chemicals (EDCs) by binding and activating estrogen receptors in the body. Uterine fibroids, often called leiomyoma are non-cancerous growths occurring in the uterus. Though often asymptomatic, they can cause pain, infertility, pregnancy complications and are a leading cause for hysterectomy. The aetiology of leiomyoma is not fully understood but both estrogen and progesterone have been implicated in their growth. We aimed to investigate the epidemiological evidence for the association between EDCs and the prevalence of fibroids. 

Study design, size, duration

We undertook a systematic review and in keeping with PRISMA guidelines, a structured search of Medline, Embase, Scopus, and Web of Science was conducted (to October 2020). Case-control, cross-sectional, cohort and experimental studies were included. 

Participants/materials, setting, methods

The included studies analyzed the association between one or more toxins and the occurrence, or growth of leiomyoma in humans, including human cell lines. The types of toxins, patient characteristics, association and outcome, body concentration of toxin and confounding variables were extracted and analyzed. Quality assessment was performed using the Newcastle-Ottawa Scale.

Main results and the role of chance

In total, 34 studies were included. The majority (76%) of studies revealed a significant association between the exposure studied and the prevalence of uterine leiomyoma. In examining body burden in cases vs controls, phthalate esters showed an association with increased odds of uterine leiomyoma, except in one case where a negative association was observed. In vitro experimental studies examining the effect of alkyl-phenols such as bisphenol A (BPA), octylphenol (OP) and nonylphenol (NP) demonstrated that these environmental estrogens can act to promote the proliferation of leiomyoma cells through a number of mechanisms, typically including the estrogen receptor alpha (ERa) signalling pathway. There were conflicting results for the association between alkyl-phenols and fibroids in case-control studies. A positive association between cadmium was demonstrated in only two studies. There were conflicting results for the association between lead, mercury, arsenic and uterine fibroids. Several metabolites of organophosphate esters, alternative plasticizers, and persistent organic pollutants were associated with an increased risk of uterine fibroids. 

Limitations, reasons for caution

Separating these exposures from the multiple other factors that could affect the outcome of leiomyoma is challenging, but an important issue for future research. 

Wider implications of the findings

The link between some environmental toxins and uterine fibroids discussed is in agreement with previous literature. However, our review provides a more in depth analysis on specific dosage effects, odds ratios, and potential gene mechanisms of the exposures. This information could contribute to more accurate preventative measures.

Trial registration number

not applicable

P–296 Examining the link between environmental toxin exposure and uterine leiomyoma: a systematic review

Study question

Is there an association between exposure to certain environmental toxins and the prevalence of uterine leiomyoma in women?

Summary answer

Some evidence was obtained to suggest an association between phthalate esters, bisphenol A, heavy metals, persistent organic pollutants and the prevalence of uterine fibroids.

What is known already

Environmental toxins are naturally occurring, or human made chemicals that can act as endocrine disrupting chemicals (EDCs) by binding and activating estrogen receptors in the body. Uterine fibroids, often called leiomyoma are non-cancerous growths occurring in the uterus. Though often asymptomatic, they can cause pain, infertility, pregnancy complications and are a leading cause for hysterectomy. The aetiology of leiomyoma is not fully understood but both estrogen and progesterone have been implicated in their growth. We aimed to investigate the epidemiological evidence for the association between EDCs and the prevalence of fibroids.

Study design, size, duration

We undertook a systematic review and in keeping with PRISMA guidelines, a structured search of Medline, Embase, Scopus, and Web of Science was conducted (to October 2020). Case-control, cross-sectional, cohort and experimental studies were included.

Participants/materials, setting, methods

The included studies analyzed the association between one or more toxins and the occurrence, or growth of leiomyoma in humans, including human cell lines. The types of toxins, patient characteristics, association and outcome, body concentration of toxin and confounding variables were extracted and analyzed. Quality assessment was performed using the Newcastle-Ottawa Scale.

Main results and the role of chance

In total, 34 studies were included. The majority (76%) of studies revealed a significant association between the exposure studied and the prevalence of uterine leiomyoma. In examining body burden in cases vs controls, phthalate esters showed an association with increased odds of uterine leiomyoma, except in one case where a negative association was observed. In vitro experimental studies examining the effect of alkyl-phenols such as bisphenol A (BPA), octylphenol (OP) and nonylphenol (NP) demonstrated that these environmental estrogens can act to promote the proliferation of leiomyoma cells through a number of mechanisms, typically including the estrogen receptor alpha (ERa) signalling pathway. There were conflicting results for the association between alkyl-phenols and fibroids in case-control studies. A positive association between cadmium was demonstrated in only two studies. There were conflicting results for the association between lead, mercury, arsenic and uterine fibroids. Several metabolites of organophosphate esters, alternative plasticizers, and persistent organic pollutants were associated with an increased risk of uterine fibroids.

Limitations, reasons for caution

Separating these exposures from the multiple other factors that could affect the outcome of leiomyoma is challenging, but an important issue for future research.

Wider implications of the findings: The link between some environmental toxins and uterine fibroids discussed is in agreement with previous literature. However, our review provides a more in depth analysis on specific dosage effects, odds ratios, and potential gene mechanisms of the exposures. This information could contribute to more accurate preventative measures.

Trial registration number

Not applicable

Predictive Approaches for Acute Dialysis Requirement and Death in COVID-19

BACKGROUND AND OBJECTIVES

AKI treated with dialysis initiation is a common complication of coronavirus disease 2019 (COVID-19) among hospitalized patients. However, dialysis supplies and personnel are often limited.

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS

Using data from adult patients hospitalized with COVID-19 from five hospitals from the Mount Sinai Health System who were admitted between March 10 and December 26, 2020, we developed and validated several models (logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), random forest, and eXtreme GradientBoosting [XGBoost; with and without imputation]) for predicting treatment with dialysis or death at various time horizons (1, 3, 5, and 7 days) after hospital admission. Patients admitted to the Mount Sinai Hospital were used for internal validation, whereas the other hospitals formed part of the external validation cohort. Features included demographics, comorbidities, and laboratory and vital signs within 12 hours of hospital admission.

RESULTS

A total of 6093 patients (2442 in training and 3651 in external validation) were included in the final cohort. Of the different modeling approaches used, XGBoost without imputation had the highest area under the receiver operating characteristic (AUROC) curve on internal validation (range of 0.93-0.98) and area under the precision-recall curve (AUPRC; range of 0.78-0.82) for all time points. XGBoost without imputation also had the highest test parameters on external validation (AUROC range of 0.85-0.87, and AUPRC range of 0.27-0.54) across all time windows. XGBoost without imputation outperformed all models with higher precision and recall (mean difference in AUROC of 0.04; mean difference in AUPRC of 0.15). Features of creatinine, BUN, and red cell distribution width were major drivers of the model's prediction.

CONCLUSIONS

An XGBoost model without imputation for prediction of a composite outcome of either death or dialysis in patients positive for COVID-19 had the best performance, as compared with standard and other machine learning models.

PODCAST

This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2021_07_09_CJN17311120.mp3.

Machine Learning Prediction of Death in Critically Ill Patients With Coronavirus Disease 2019

OBJECTIVES

Critically ill patients with coronavirus disease 2019 have variable mortality. Risk scores could improve care and be used for prognostic enrichment in trials. We aimed to compare machine learning algorithms and develop a simple tool for predicting 28-day mortality in ICU patients with coronavirus disease 2019.

DESIGN

This was an observational study of adult patients with coronavirus disease 2019. The primary outcome was 28-day inhospital mortality. Machine learning models and a simple tool were derived using variables from the first 48 hours of ICU admission and validated externally in independent sites and temporally with more recent admissions. Models were compared with a modified Sequential Organ Failure Assessment score, National Early Warning Score, and CURB-65 using the area under the receiver operating characteristic curve and calibration.

SETTING

Sixty-eight U.S. ICUs.

PATIENTS

Adults with coronavirus disease 2019 admitted to 68 ICUs in the United States between March 4, 2020, and June 29, 2020.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The study included 5,075 patients, 1,846 (36.4%) of whom died by day 28. eXtreme Gradient Boosting had the highest area under the receiver operating characteristic curve in external validation (0.81) and was well-calibrated, while k-nearest neighbors were the lowest performing machine learning algorithm (area under the receiver operating characteristic curve 0.69). Findings were similar with temporal validation. The simple tool, which was created using the most important features from the eXtreme Gradient Boosting model, had a significantly higher area under the receiver operating characteristic curve in external validation (0.78) than the Sequential Organ Failure Assessment score (0.69), National Early Warning Score (0.60), and CURB-65 (0.65; p < 0.05 for all comparisons). Age, number of ICU beds, creatinine, lactate, arterial pH, and Pao2/Fio2 ratio were the most important predictors in the eXtreme Gradient Boosting model.

CONCLUSIONS

eXtreme Gradient Boosting had the highest discrimination overall, and our simple tool had higher discrimination than a modified Sequential Organ Failure Assessment score, National Early Warning Score, and CURB-65 on external validation. These models could be used to improve triage decisions and clinical trial enrichment.

Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City

Federated learning is a technique for training predictive models without sharing patient-level data, thus maintaining data security while allowing inter-institutional collaboration. We used federated learning to predict acute kidney injury within three and seven days of admission, using demographics, comorbidities, vital signs, and laboratory values, in 4029 adults hospitalized with COVID-19 at five sociodemographically diverse New York City hospitals, between March-October 2020. Prediction performance of federated models was generally higher than single-hospital models and was comparable to pooled-data models. In the first use-case in kidney disease, federated learning improved prediction of a common complication of COVID-19, while preserving data privacy.

Derivation and validation of a machine learning risk score using biomarker and electronic patient data to predict progression of diabetic kidney disease

AIM

Predicting progression in diabetic kidney disease (DKD) is critical to improving outcomes. We sought to develop/validate a machine-learned, prognostic risk score (KidneyIntelX™) combining electronic health records (EHR) and biomarkers.

METHODS

This is an observational cohort study of patients with prevalent DKD/banked plasma from two EHR-linked biobanks. A random forest model was trained, and performance (AUC, positive and negative predictive values [PPV/NPV], and net reclassification index [NRI]) was compared with that of a clinical model and Kidney Disease: Improving Global Outcomes (KDIGO) categories for predicting a composite outcome of eGFR decline of ≥5 ml/min per year, ≥40% sustained decline, or kidney failure within 5 years.

RESULTS

In 1146 patients, the median age was 63 years, 51% were female, the baseline eGFR was 54 ml min-1 [1.73 m]-2, the urine albumin to creatinine ratio (uACR) was 6.9 mg/mmol, follow-up was 4.3 years and 21% had the composite endpoint. On cross-validation in derivation (n = 686), KidneyIntelX had an AUC of 0.77 (95% CI 0.74, 0.79). In validation (n = 460), the AUC was 0.77 (95% CI 0.76, 0.79). By comparison, the AUC for the clinical model was 0.62 (95% CI 0.61, 0.63) in derivation and 0.61 (95% CI 0.60, 0.63) in validation. Using derivation cut-offs, KidneyIntelX stratified 46%, 37% and 17% of the validation cohort into low-, intermediate- and high-risk groups for the composite kidney endpoint, respectively. The PPV for progressive decline in kidney function in the high-risk group was 61% for KidneyIntelX vs 40% for the highest risk strata by KDIGO categorisation (p < 0.001). Only 10% of those scored as low risk by KidneyIntelX experienced progression (i.e., NPV of 90%). The NRIevent for the high-risk group was 41% (p < 0.05).

CONCLUSIONS

KidneyIntelX improved prediction of kidney outcomes over KDIGO and clinical models in individuals with early stages of DKD.

Genome-wide polygenic risk score for retinopathy of type 2 diabetes

Diabetic retinopathy (DR) is a common consequence in type 2 diabetes (T2D) and a leading cause of blindness in working-age adults. Yet, its genetic predisposition is largely unknown. Here, we examined the polygenic architecture underlying DR by deriving and assessing a genome-wide polygenic risk score (PRS) for DR. We evaluated the PRS in 6079 individuals with T2D of European, Hispanic, African and other ancestries from a large-scale multi-ethnic biobank. Main outcomes were PRS association with DR diagnosis, symptoms and complications, and time to diagnosis, and transferability to non-European ancestries. We observed that PRS was significantly associated with DR. A standard deviation increase in PRS was accompanied by an adjusted odds ratio (OR) of 1.12 [95% confidence interval (CI) 1.04-1.20; P = 0.001] for DR diagnosis. When stratified by ancestry, PRS was associated with the highest OR in European ancestry (OR = 1.22, 95% CI 1.02-1.41; P = 0.049), followed by African (OR = 1.15, 95% CI 1.03-1.28; P = 0.028) and Hispanic ancestries (OR = 1.10, 95% CI 1.00-1.10; P = 0.050). Individuals in the top PRS decile had a 1.8-fold elevated risk for DR versus the bottom decile (P = 0.002). Among individuals without DR diagnosis, the top PRS decile had more DR symptoms than the bottom decile (P = 0.008). The PRS was associated with retinal hemorrhage (OR = 1.44, 95% CI 1.03-2.02; P = 0.03) and earlier DR presentation (10% probability of DR by 4 years in the top PRS decile versus 8 years in the bottom decile). These results establish the significant polygenic underpinnings of DR and indicate the need for more diverse ancestries in biobanks to develop multi-ancestral PRS.

Hyperglycemia Is Associated With Increased Mortality in Critically Ill Patients With COVID-19

Objective: To explore the relationship between diabetes mellitus (DM), hyperglycemia, and adverse outcomes in critically ill patients with coronavirus disease 2019 (COVID-19).

Research Design and Methods: The study population comprised 133 patients with COVID-19 admitted to an intensive care unit (ICU) at an academic, urban, quaternary-care center between March 10th and April 8th, 2020. Patients were categorized based on the presence of DM and early-onset hyperglycemia (EHG), defined as a blood glucose &gt;180 mg/dL during the first two days of ICU admission. The primary outcome was 14-day in-hospital mortality; also examined were 60-day in-hospital mortality and the levels of C-reactive protein (CRP), interleukin 6, procalcitonin, and lactate. Results: Compared to non-DM patients without EHG, non-DM patients with EHG exhibited higher adjusted hazard ratios (HR) for in-hospital mortality at 14 days (HR 5.76, p=0.008) and 60 days (HR 7.28, p=0.004). Non-DM patients with EHG also featured higher levels of mean CRP (322.3±177.7 mg/L, p=0.036), procalcitonin (34.75±69.33 ng/mL, p=0.028), and lactate (2.7±2.1 mmol/L, p=0.023). Conclusions: In patients with critical illness from COVID-19, those without DM with EHG were at greatest risk of 14-day and 60-day in-hospital mortality. The limitations of our study include its retrospective design, and relatively small cohort. However, our results raise the possibility that the combination of elevated glucose and lactate may identify a specific cohort of individuals at high mortality risk from COVID-19, and suggest that glucose testing and control are important in individuals with COVID-19, even in those without pre-existing diabetes.

Impact of the COVID-19 Pandemic on Nephrology Fellow Training and Well-Being in the United States: A National Survey

BACKGROUND

The coronavirus disease 2019 (COVID-19) pandemic's effects on nephrology fellows' educational experiences, preparedness for practice, and emotional wellbeing are unknown.

METHODS

We recruited current adult and pediatric fellows and 2020 graduates of nephrology training programs in the United States to participate in a survey measuring COVID-19's effects on their training experiences and wellbeing.

RESULTS

Of 1005 nephrology fellows-in-training and recent graduates, 425 participated (response rate 42%). Telehealth was widely adopted (90% for some or all outpatient nephrology consults), as was remote learning (76% of conferences were exclusively online). Most respondents (64%) did not have in-person consults on COVID-19 inpatients; these patients were managed by telehealth visits (27%), by in-person visits with the attending faculty without fellows (29%), or by another approach (9%). A majority of fellows (84%) and graduates (82%) said their training programs successfully sustained their education during the pandemic, and most fellows (86%) and graduates (90%) perceived themselves as prepared for unsupervised practice. Although 42% indicated the pandemic had negatively affected their overall quality of life and 33% reported a poorer work-life balance, only 15% of 412 respondents who completed the Resident Well-Being Index met its distress threshold. Risk for distress was increased among respondents who perceived the pandemic had impaired their knowledge base (odds ratio [OR], 3.04; 95% confidence interval [CI], 2.00 to 4.77) or negatively affected their quality of life (OR, 3.47; 95% CI, 2.29 to 5.46) or work-life balance (OR, 3.16; 95% CI, 2.18 to 4.71).

CONCLUSIONS

Despite major shifts in education modalities and patient care protocols precipitated by the COVID-19 pandemic, participants perceived their education and preparation for practice to be minimally affected.

Thrombosis, Bleeding, and the Observational Effect of Early Therapeutic Anticoagulation on Survival in Critically Ill Patients With COVID-19

BACKGROUND

Hypercoagulability may be a key mechanism of death in patients with coronavirus disease 2019 (COVID-19).

OBJECTIVE

To evaluate the incidence of venous thromboembolism (VTE) and major bleeding in critically ill patients with COVID-19 and examine the observational effect of early therapeutic anticoagulation on survival.

DESIGN

In a multicenter cohort study of 3239 critically ill adults with COVID-19, the incidence of VTE and major bleeding within 14 days after intensive care unit (ICU) admission was evaluated. A target trial emulation in which patients were categorized according to receipt or no receipt of therapeutic anticoagulation in the first 2 days of ICU admission was done to examine the observational effect of early therapeutic anticoagulation on survival. A Cox model with inverse probability weighting to adjust for confounding was used.

SETTING

67 hospitals in the United States.

PARTICIPANTS

Adults with COVID-19 admitted to a participating ICU.

MEASUREMENTS

Time to death, censored at hospital discharge, or date of last follow-up.

RESULTS

Among the 3239 patients included, the median age was 61 years (interquartile range, 53 to 71 years), and 2088 (64.5%) were men. A total of 204 patients (6.3%) developed VTE, and 90 patients (2.8%) developed a major bleeding event. Independent predictors of VTE were male sex and higher D-dimer level on ICU admission. Among the 2809 patients included in the target trial emulation, 384 (11.9%) received early therapeutic anticoagulation. In the primary analysis, during a median follow-up of 27 days, patients who received early therapeutic anticoagulation had a similar risk for death as those who did not (hazard ratio, 1.12 [95% CI, 0.92 to 1.35]).

LIMITATION

Observational design.

CONCLUSION

Among critically ill adults with COVID-19, early therapeutic anticoagulation did not affect survival in the target trial emulation.

PRIMARY FUNDING SOURCE

None.

Molecular Analysis of the Kidney From a Patient With COVID-19-Associated Collapsing Glomerulopathy

Recent case reports suggest that coronavirus disease 2019 (COVID-19) is associated with collapsing glomerulopathy in African Americans with apolipoprotein L1 gene (APOL1) risk alleles; however, it is unclear whether disease pathogenesis is similar to HIV-associated nephropathy. RNA sequencing analysis of a kidney biopsy specimen from a patient with COVID-19–associated collapsing glomerulopathy and APOL1 risk alleles (G1/G1) revealed similar levels of APOL1 and angiotensin-converting enzyme 2 (ACE2) messenger RNA transcripts as compared with 12 control kidney samples downloaded from the GTEx (Genotype-Tissue Expression) Portal. Whole-genome sequencing of the COVID-19–associated collapsing glomerulopathy kidney sample identified 4 indel gene variants, 3 of which are of unknown significance with respect to chronic kidney disease and/or focal segmental glomerulosclerosis. Molecular profiling of the kidney demonstrated activation of COVID-19–associated cell injury pathways such as inflammation and coagulation. Evidence for direct severe acute respiratory syndrome coronavirus 2 infection of kidney cells was lacking, which is consistent with the findings of several recent studies. Interestingly, immunostaining of kidney biopsy sections revealed increased expression of phospho-STAT3 (signal transducer and activator of transcription 3) in both COVID-19–associated collapsing glomerulopathy and HIV-associated nephropathy as compared with control kidney tissue. Importantly, interleukin 6–induced activation of STAT3 may be a targetable mechanism driving COVID-19–associated acute kidney injury.