Comparative 3D genome analysis between neural retina and retinal pigment epithelium reveals differential cis-regulatory interactions at retinal disease loci.

BACKGROUND: Vision depends on the interplay between photoreceptor cells of the neural retina and the underlying retinal pigment epithelium (RPE). Most genes involved in inherited retinal diseases display specific spatiotemporal expression within these interconnected retinal components through the local recruitment of cis-regulatory elements (CREs) in 3D nuclear space. RESULTS: To understand the role of differential chromatin architecture in establishing tissue-specific expression at inherited retinal disease loci, we mapped genome-wide chromatin interactions using in situ Hi-C and H3K4me3 HiChIP on neural retina and RPE/choroid from human adult donor eyes. We observed chromatin looping between active promoters and 32,425 and 8060 candidate CREs in the neural retina and RPE/choroid, respectively. A comparative 3D genome analysis between these two retinal tissues revealed that 56% of 290 known inherited retinal disease genes were marked by differential chromatin interactions. One of these was ABCA4, which is implicated in the most common autosomal recessive inherited retinal disease. We zoomed in on retina- and RPE-specific cis-regulatory interactions at the ABCA4 locus using high-resolution UMI-4C. Integration with bulk and single-cell epigenomic datasets and in vivo enhancer assays in zebrafish revealed tissue-specific CREs interacting with ABCA4. CONCLUSIONS: Through comparative 3D genome mapping, based on genome-wide, promoter-centric, and locus-specific assays of human neural retina and RPE, we have shown that gene regulation at key inherited retinal disease loci is likely mediated by tissue-specific chromatin interactions. These findings do not only provide insight into tissue-specific regulatory landscapes at retinal disease loci, but also delineate the search space for non-coding genomic variation underlying unsolved inherited retinal diseases.

Surrogate indices of insulin resistance using the Matsuda index as reference in adult men-a computational approach.

Background: Overweight and obesity, high blood pressure, hyperglycemia, hyperlipidemia, and insulin resistance (IR) are strongly associated with non-communicable diseases (NCDs), including type 2 diabetes, cardiovascular disease, stroke, and cancer. Different surrogate indices of IR are derived and validated with the euglycemic-hyperinsulinemic clamp (EHC) test. Thus, using a computational approach to predict IR with Matsuda index as reference, this study aimed to determine the optimal cutoff value and diagnosis accuracy for surrogate indices in non-diabetic young adult men. Methods: A cross-sectional descriptive study was carried out with 93 young men (ages 18-31). Serum levels of glucose and insulin were analyzed in the fasting state and during an oral glucose tolerance test (OGTT). Additionally, clinical, biochemical, hormonal, and anthropometric characteristics and body composition (DEXA) were determined. The computational approach to evaluate the IR diagnostic accuracy and cutoff value using difference parameters was examined, as well as other statistical tools to make the output robust. Results: The highest sensitivity and specificity at the optimal cutoff value, respectively, were established for the Homeostasis model assessment of insulin resistance index (HOMA-IR) (0.91; 0.98; 3.40), the Quantitative insulin sensitivity check index (QUICKI) (0.98; 0.96; 0.33), the triglyceride-glucose (TyG)-waist circumference index (TyG-WC) (1.00; 1.00; 427.77), the TyG-body mass index (TyG-BMI) (1.00; 1.00; 132.44), TyG-waist-to-height ratio (TyG-WHtR) (0.98; 1.00; 2.48), waist-to-height ratio (WHtR) (1.00; 1.00; 0.53), waist circumference (WC) (1.00; 1.00; 92.63), body mass index (BMI) (1.00; 1.00; 28.69), total body fat percentage (TFM) (%) (1.00; 1.00; 31.07), android fat (AF) (%) (1.00; 0.98; 40.33), lipid accumulation product (LAP) (0.84; 1.00; 45.49), leptin (0.91; 1.00; 16.08), leptin/adiponectin ratio (LAR) (0.84; 1.00; 1.17), and fasting insulin (0.91; 0.98; 16.01). Conclusions: The computational approach was used to determine the diagnosis accuracy and the optimal cutoff value for IR to be used in preventive healthcare.

Multi-omics analysis in human retina uncovers ultraconserved cis-regulatory elements at rare eye disease loci.

Cross-species genome comparisons have revealed a substantial number of ultraconserved non-coding elements (UCNEs). Several of these elements have proved to be essential tissue- and cell type-specific cis-regulators of developmental gene expression. Here, we characterize a set of UCNEs as candidate CREs (cCREs) during retinal development and evaluate the contribution of their genomic variation to rare eye diseases, for which pathogenic non-coding variants are emerging. Integration of bulk and single-cell retinal multi-omics data reveals 594 genes under potential cis-regulatory control of UCNEs, of which 45 are implicated in rare eye disease. Mining of candidate cis-regulatory UCNEs in WGS data derived from the rare eye disease cohort of Genomics England reveals 178 ultrarare variants within 84 UCNEs associated with 29 disease genes. Overall, we provide a comprehensive annotation of ultraconserved non-coding regions acting as cCREs during retinal development which can be targets of non-coding variation underlying rare eye diseases.

Evaluation of Clinical and Machine Data of Critically Ill Adult COVID Patients with Acute Kidney Injury Exposed to Enhanced Hemoadsorption during CRRT.

INTRODUCTION: The FDA authorized the emergency use of enhanced hemoadsorption with oXiris in critically ill adult COVID patients with respiratory failure or severe disease to reduce inflammation. In this study, we evaluated critically ill adult COVID patients with acute kidney injury (AKI) who were exposed versus not exposed to enhanced hemoadsorption with oXiris during continuous renal replacement therapy (CRRT). METHODS: Retrospective cohort study of critically ill adult COVID patients with AKI requiring CRRT. Exposure to oXiris was defined as receiving oXiris for >12 cumulative hours and more than one-third of the time within the first 72 h of CRRT. Study outcomes included filter-specific performance metrics and clinical outcomes such as ventilator requirement, mortality, and dialysis dependence. Inverse probability treatment weighting was used to balance potential confounders in weighted regression models. RESULTS: 14,043 h of CRRT corresponding to 85 critically ill adult patients were analyzed. Among these, 2,736 h corresponded to oXiris exposure (n = 25 patients) and 11,307 h to a standard CRRT filter (n = 60 patients). Transmembrane pressures (TMPs) increased rapidly and were overall higher with oXiris versus standard filter, but filter life (median of 36.3 vs. 33.1 h, p = 0.913, respectively) and filter/clotting alarms remained similar in both groups. In adjusted models, oXiris exposure was not independently associated with the composite of hospital mortality and dialysis dependence at discharge (OR 2.13, 95% CI: 0.98-4.82, p = 0.06), but it was associated with fewer ventilator (ß = -15.02, 95% CI: -29.23 to -0.82, p = 0.04) and intensive care unit days (ß = -14.74, 95% CI: -28.54 to -0.95, p = 0.04) in survivors. DISCUSSION/CONCLUSION: In critically ill adult COVID patients with AKI requiring CRRT, oXiris filters exhibited higher levels of TMP when compared to a standard CRRT filter, but no differences in filter life and filter/clotting alarm profiles were observed. The use of oXiris was not associated with improvement in clinical outcomes such as hospital mortality or dialysis dependence at discharge.

Epidemiology and Outcomes of AKI Treated With Continuous Kidney Replacement Therapy: The Multicenter CRRTnet Study.

Rationale & Objective: Continuous kidney replacement therapy (CKRT) is the predominant form of acute kidney replacement therapy used for critically ill adult patients with acute kidney injury (AKI). Given the variability in CKRT practice, a contemporary understanding of its epidemiology is necessary to improve care delivery. Study Design: Multicenter, prospective living registry. Setting & Population: 1,106 critically ill adults with AKI requiring CKRT from December 2013 to January 2021 across 5 academic centers and 6 intensive care units. Patients with pre-existing kidney failure and those with coronavirus 2 infection were excluded. Exposure: CKRT for more than 24 hours. Outcomes: Hospital mortality, kidney recovery, and health care resource utilization. Analytical Approach: Data were collected according to preselected timepoints at intensive care unit admission and CKRT initiation and analyzed descriptively. Results: Patients' characteristics, contributors to AKI, and CKRT indications differed among centers. Mean (standard deviation) age was 59.3 (13.9) years, 39.7% of patients were women, and median [IQR] APACHE-II (acute physiologic assessment and chronic health evaluation) score was 30 [25-34]. Overall, 41.1% of patients survived to hospital discharge. Patients that died were older (mean age 61 vs. 56.8, P < 0.001), had greater comorbidity (median Charlson score 3 [1-4] vs. 2 [1-3], P < 0.001), and higher acuity of illness (median APACHE-II score 30 [25-35] vs. 29 [24-33], P = 0.003). The most common condition predisposing to AKI was sepsis (42.6%), and the most common CKRT indications were oliguria/anuria (56.2%) and fluid overload (53.9%). Standardized mortality ratios were similar among centers. Limitations: The generalizability of these results to CKRT practices in nonacademic centers or low-and middle-income countries is limited. Conclusions: In this registry, sepsis was the major contributor to AKI and fluid management was collectively the most common CKRT indication. Significant heterogeneity in patient- and CKRT-specific characteristics was found in current practice. These data highlight the need for establishing benchmarks of CKRT delivery, performance, and patient outcomes. Data from this registry could assist with the design of such studies.

The Optimization of Outpatient Hemodialysis Management for Acute Kidney Injury Requiring Dialysis Patients: A Quality Improvement Study.

INTRODUCTION: In 2017, the Centers for Medicare and Medicaid Services allowed survivors of hospitalized acute kidney injury requiring dialysis (AKI-D) who were ambulatory and still dependent on hemodialysis (HD) to receive treatment in outpatient dialysis facilities. This policy change generated the ongoing need to improve AKI-D care in the outpatient setting. METHODS: Quality improvement study in adult patients admitted to an outpatient HD unit with the diagnosis of AKI-D. We developed a protocol to manage these patients that included: (a) multidisciplinary evaluations; (b) personalized 3-tier HD prescription for dose/ultrafiltration rate and frequency; (c) weekly assessment of kidney recovery; and (d) patient empowerment. Patient- and protocol-specific characteristics were described. We analyzed hourly HD data and protocol adherence, and relevant hemodynamic data were compared according to HD-free survival at 90 days. RESULTS: A total of 457.3 h of HD from 9 patients under the AKI-D protocol were interrogated. Three out of 9 patients were alive and liberated from HD within the first 90 days of outpatient HD. Overall protocol adherence was 53.8% and did not differ by HD-free survival (54.5% vs. 53.7% in those that recovered vs. not). Protocol adherence was associated with fewer intradialytic hypotension events (peak to nadir blood pressure, p < 0.01), while intradialytic hypotension (pre- to post-blood pressure) occurred more frequently in patients who did not recover kidney function (p = 0.009). CONCLUSION: We demonstrated the feasibility of implementing a management protocol for AKI-D patients in an outpatient dialysis facility. We found that fewer episodes of intradialytic hypotension occurred when the outpatient HD management was adherent to the protocol. The feasibility of this protocol should be confirmed in other facilities, and importantly, efficacy testing to evaluate its impact on AKI-D outpatient care is necessary.

Prediction of Mortality and Major Adverse Kidney Events in Critically Ill Patients With Acute Kidney Injury.

RATIONALE & OBJECTIVE: Risk prediction tools for assisting acute kidney injury (AKI) management have focused on AKI onset but have infrequently addressed kidney recovery. We developed clinical models for risk stratification of mortality and major adverse kidney events (MAKE) in critically ill patients with incident AKI. STUDY DESIGN: Multicenter cohort study. SETTING & PARTICIPANTS: 9,587 adult patients admitted to heterogeneous intensive care units (ICUs; March 2009 to February 2017) who experienced AKI within the first 3 days of their ICU stays. PREDICTORS: Multimodal clinical data consisting of 71 features collected in the first 3 days of ICU stay. OUTCOMES: (1) Hospital mortality and (2) MAKE, defined as the composite of death during hospitalization or within 120 days of discharge, receipt of kidney replacement therapy in the last 48 hours of hospital stay, initiation of maintenance kidney replacement therapy within 120 days, or a ≥50% decrease in estimated glomerular filtration rate from baseline to 120 days from hospital discharge. ANALYTICAL APPROACH: Four machine-learning algorithms (logistic regression, random forest, support vector machine, and extreme gradient boosting) and the SHAP (Shapley Additive Explanations) framework were used for feature selection and interpretation. Model performance was evaluated by 10-fold cross-validation and external validation. RESULTS: One developed model including 15 features outperformed the SOFA (Sequential Organ Failure Assessment) score for the prediction of hospital mortality, with areas under the curve of 0.79 (95% CI, 0.79-0.80) and 0.71 (95% CI, 0.71-0.71) in the development cohort and 0.74 (95% CI, 0.73-0.74) and 0.71 (95% CI, 0.71-0.71) in the validation cohort (P < 0.001 for both). A second developed model including 14 features outperformed KDIGO (Kidney Disease: Improving Global Outcomes) AKI severity staging for the prediction of MAKE: 0.78 (95% CI, 0.78-0.78) versus 0.66 (95% CI, 0.66-0.66) in the development cohort and 0.73 (95% CI, 0.72-0.74) versus 0.67 (95% CI, 0.67-0.67) in the validation cohort (P < 0.001 for both). LIMITATIONS: The models are applicable only to critically ill adult patients with incident AKI within the first 3 days of an ICU stay. CONCLUSIONS: The reported clinical models exhibited better performance for mortality and kidney recovery prediction than standard scoring tools commonly used in critically ill patients with AKI in the ICU. Additional validation is needed to support the utility and implementation of these models. PLAIN-LANGUAGE SUMMARY: Acute kidney injury (AKI) occurs commonly in critically ill patients admitted to the intensive care unit (ICU) and is associated with high morbidity and mortality rates. Prediction of mortality and recovery after an episode of AKI may assist bedside decision making. In this report, we describe the development and validation of a clinical model using data from the first 3 days of an ICU stay to predict hospital mortality and major adverse kidney events occurring as long as 120 days after hospital discharge among critically ill adult patients who experienced AKI within the first 3 days of an ICU stay. The proposed clinical models exhibited good performance for outcome prediction and, if further validated, could enable risk stratification for timely interventions that promote kidney recovery.

Multi-omics approach dissects cis-regulatory mechanisms underlying North Carolina macular dystrophy, a retinal enhanceropathy.

North Carolina macular dystrophy (NCMD) is a rare autosomal-dominant disease affecting macular development. The disease is caused by non-coding single-nucleotide variants (SNVs) in two hotspot regions near PRDM13 and by duplications in two distinct chromosomal loci, overlapping DNase I hypersensitive sites near either PRDM13 or IRX1. To unravel the mechanisms by which these variants cause disease, we first established a genome-wide multi-omics retinal database, RegRet. Integration of UMI-4C profiles we generated on adult human retina then allowed fine-mapping of the interactions of the PRDM13 and IRX1 promoters and the identification of eighteen candidate cis-regulatory elements (cCREs), the activity of which was investigated by luciferase and Xenopus enhancer assays. Next, luciferase assays showed that the non-coding SNVs located in the two hotspot regions of PRDM13 affect cCRE activity, including two NCMD-associated non-coding SNVs that we identified herein. Interestingly, the cCRE containing one of these SNVs was shown to interact with the PRDM13 promoter, demonstrated in vivo activity in Xenopus, and is active at the developmental stage when progenitor cells of the central retina exit mitosis, suggesting that this region is a PRDM13 enhancer. Finally, mining of single-cell transcriptional data of embryonic and adult retina revealed the highest expression of PRDM13 and IRX1 when amacrine cells start to synapse with retinal ganglion cells, supporting the hypothesis that altered PRDM13 or IRX1 expression impairs interactions between these cells during retinogenesis. Overall, this study provides insight into the cis-regulatory mechanisms of NCMD and supports that this condition is a retinal enhanceropathy.

Assessment of prescribed vs. achieved fluid balance during continuous renal replacement therapy and mortality outcome.

BACKGROUND: Fluid management during continuous renal replacement therapy (CRRT) requires accuracy in the prescription of desired patient fluid balance (FBGoal) and precision in the attainable patient fluid balance (FBAchieved). Herein, we examined the association of the gap between prescribed vs. achieved patient fluid balance during CRRT (%FBGap) with hospital mortality in critically ill patients. METHODS: Cohort study of critically ill adults with acute kidney injury (AKI) requiring CRRT and a prescription of negative fluid balance (mean patient fluid balance goal of negative ≥0.5 liters per day). Fluid management parameters included: 1) NUF (net ultrafiltration rate); 2) FBGoal; 3) FBAchieved; and 4) FBGap (% gap of fluid balance achieved vs. goal), all adjusted by patient's weight (kg) and duration of CRRT (hours). RESULTS: Data from 653 patients (median of 102.2 patient-hours of CRRT) were analyzed. Mean (SD) age was 56.7 (14.6) years and 61.9% were male. Hospital mortality rate was 64%. Despite FBGoal was similar in patients who died vs. survived, survivors achieved greater negative fluid balance during CRRT than non-survivors: median FBAchieved -0.25 [-0.52 to -0.05] vs. 0.06 [-0.26 to 0.62] ml/kg/h, p<0.001. Median NUF was lower in patients who died vs. survived: 1.06 [0.63-1.47] vs. 1.22 [0.82-1.69] ml/kg/h, p<0.001, and median %FBGap was higher in patients who died (112.8%, 61.5 to 165.7) vs. survived (64.2%, 30.5 to 91.8), p<0.001. In multivariable models, higher %FBGap was independently associated with increased risk of hospital mortality: aOR (95% CI) 1.01 (1.01-1.02), p<0.001. NUF was not associated with hospital mortality when adjusted by %FBGap and other clinical parameters: aOR 0.96 (0.72-1.28), p = 0.771. CONCLUSIONS: Higher %FBGap was independently associated with an increased risk of hospital mortality in critically ill adults with AKI on CRRT in whom clinicians prescribed negative fluid balance via CRRT. %FBGap represents a novel quality indicator of CRRT delivery that could assist with operationalizing fluid management interventions during CRRT.

Serum IL-17 levels are higher in critically ill patients with AKI and associated with worse outcomes.

BACKGROUND: Interleukin-17 (IL-17) antagonism in rats reduces the severity and progression of AKI. IL-17-producing circulating T helper-17 (TH17) cells is increased in critically ill patients with AKI indicating that this pathway is also activated in humans. We aim to compare serum IL-17A levels in critically ill patients with versus without AKI and to examine their relationship with mortality and major adverse kidney events (MAKE). METHODS: Multicenter, prospective study of ICU patients with AKI stage 2 or 3 and without AKI. Samples were collected at 24-48 h after AKI diagnosis or ICU admission (in those without AKI) [timepoint 1, T1] and 5-7 days later [timepoint 2, T2]. MAKE was defined as the composite of death, dependence on kidney replacement therapy or a reduction in eGFR of ≥ 30% from baseline up to 90 days following hospital discharge. RESULTS: A total of 299 patients were evaluated. Patients in the highest IL-17A tertile (versus lower tertiles) at T1 had higher acuity of illness and comorbidity scores. Patients with AKI had higher levels of IL-17A than those without AKI: T1 1918.6 fg/ml (692.0-5860.9) versus 623.1 fg/ml (331.7-1503.4), p < 0.001; T2 2167.7 fg/ml (839.9-4618.9) versus 1193.5 fg/ml (523.8-2198.7), p = 0.006. Every onefold higher serum IL-17A at T1 was independently associated with increased risk of hospital mortality (aOR 1.35, 95% CI: 1.06-1.73) and MAKE (aOR 1.26, 95% CI: 1.02-1.55). The highest tertile of IL-17A (vs. the lowest tertile) was also independently associated with higher risk of MAKE (aOR 3.03, 95% CI: 1.34-6.87). There was no effect modification of these associations by AKI status. IL-17A levels remained significantly elevated at T2 in patients that died or developed MAKE. CONCLUSIONS: Serum IL-17A levels measured by the time of AKI diagnosis or ICU admission were differentially elevated in critically ill patients with AKI when compared to those without AKI and were independently associated with hospital mortality and MAKE.

Acute kidney injury contributes to worse physical and quality of life outcomes in survivors of critical illness.

OBJECTIVES: Survivors of critical illness and acute kidney injury (AKI) are at risk of increased morbidity. The purpose of this study was to compare physical, emotional, and cognitive health in survivors of critical illness with and without AKI. METHODS: Retrospective cohort study of adult (≥ 18 years old) survivors of critical illness due to sepsis and/or acute respiratory failure who attended follow-up in a specialized ICU Recovery Clinic. Outcomes were evaluated during 3-month visit and comprised validated tests for evaluation of physical function, muscle strength, cognitive and emotional health, and self-reported health-related quality of life (HRQOL). Descriptive statistics and group comparisons were performed. RESULTS: A total of 104 patients with median age of 55 [49-64] years, 54% male, and median SOFA score of 10 [8-12] were analyzed. Incidence of AKI during ICU admission was 61 and 19.2% of patients required renal replacement therapy (RRT). Patients with AKI stage 2 or 3 (vs. those with AKI stage 1 or no AKI) walked less on the 6-min walk test (223 ± 132 vs. 295 ± 153 m, p = 0.059) and achieved lower of the predicted walk distance (38% vs. 58%, p = 0.041). Similar patterns of worse physical function and more significant muscle weakness were observed in multiple tests, with overall worse metrics in patients that required RRT. Patients with AKI stage 2 or 3 also reported lower HRQOL scores when compared to their counterparts, including less ability to return to work or hobby, or reengage in driving. There were no significant differences in cognitive function or emotional health between groups. CONCLUSIONS: Survivors of critical illness and AKI stage 2 or 3 have increased physical debility and overall lower quality of life, with more impairment in return to work, hobby, and driving when compared to their counterparts without AKI or AKI stage 1 at 3 months post-discharge.

Severe acute kidney injury caused by decompression sickness syndrome.

Decompression sickness (DCS) occurs when divers are exposed to reduced barometric pressure during their ascent from depth. We report a case of DCS causing severe acute kidney injury (AKI) after an uneventful dive in which all decompression stops were made as instructed by a dive computer. A 26-year-old man presented with abdominal and bilateral flank pain ~ 24 hours after scuba diving to a depth of 23 m. Vitals and physical exam were unremarkable. Lab results revealed elevated serum creatinine at 2.3 mg/dL from a normal baseline and elevated blood urea nitrogen at 23 mg/dL. The patient was non-oliguric. Other biochemical parameters were unremarkable. Dipstick urinalysis showed presence of blood and 100 mg/dL proteinuria. Urine microscopy revealed hyaline casts and red blood cells ~ 16 - 30/HPF but no acanthocytes. Urine protein-to-creatinine ratio was 340 mg/g. Renal ultrasound showed normal sized kidneys with increased cortical echogenicity, and computed tomography of the abdomen/pelvis showed bilateral striated nephrogram with delayed excretion, both radiographic signs of acute tubular necrosis. The patient received isotonic IV fluids and 5 sessions of hyperbaric oxygen therapy. Symptomatic improvement was observed by day 3 of hospitalization with full recovery of kidney function after discharge. Due to a wide range of associated symptomology, a thorough and prompt evaluation is warranted in suspected cases of DCS, particularly if presentation is more than 24 hours following ascent.

KIT-LSTM: Knowledge-guided Time-aware LSTM for Continuous Clinical Risk Prediction.

Rapid accumulation of temporal Electronic Health Record (EHR) data and recent advances in deep learning have shown high potential in precisely and timely predicting patients' risks using AI. However, most existing risk prediction approaches ignore the complex asynchronous and irregular problems in real-world EHR data. This paper proposes a novel approach called Knowledge-guIded Time-aware LSTM (KIT-LSTM) for continuous mortality predictions using EHR. KIT-LSTM extends LSTM with two time-aware gates and a knowledge-aware gate to better model EHR and interprets results. Experiments on real-world data for patients with acute kidney injury with dialysis (AKI-D) demonstrate that KIT-LSTM performs better than the state-of-the-art methods for predicting patients' risk trajectories and model interpretation. KIT-LSTM can better support timely decision-making for clinicians.

Assessment of a modified renal angina index for AKI prediction in critically ill adults.

BACKGROUND: The renal angina index (RAI) is a useful tool for risk stratification of acute kidney injury (AKI) in critically ill children. We evaluated the performance of a modified adult RAI (mRAI) for the risk stratification of AKI in critically ill adults. METHODS: We used two independent intensive care unit (ICU) cohorts: 13 965 adult patients from the University of Kentucky (UKY) and 4789 from University of Texas Southwestern (UTSW). The mRAI included: diabetes, presence of sepsis, mechanical ventilation, pressor/inotrope use, percentage change in serum creatinine (SCr) in reference to admission SCr (ΔSCr) and fluid overload percentage within the first day of ICU admission. The primary outcome was AKI Stage ≥2 at Days 2-7. Performance and reclassification metrics were determined for the mRAI score compared with ΔSCr alone. RESULTS: The mRAI score outperformed ΔSCr and readjusted probabilities to predict AKI Stage ≥2 at Days 2-7: C-statistic: UKY 0.781 versus 0.708 [integrated discrimination improvement (IDI) 2.2%] and UTSW 0.766 versus 0.696 (IDI 1.8%) (P < 0.001 for both). In the UKY cohort, only 3.3% of patients with mRAI score <10 had the AKI event, while 16.4% of patients with mRAI score of ≥10 had the AKI event (negative predictive value 96.8%). Similar findings were observed in the UTSW cohort as part of external validation. CONCLUSIONS: In critically ill adults, the adult mRAI score determined within the first day of ICU admission outperformed changes in SCr for the prediction of AKI Stage ≥2 at Days 2-7 of ICU stay. The mRAI is a feasible tool for AKI risk stratification in adult patients in the ICU.

Kidney Recovery in Patients With Acute Kidney Injury Treated in Outpatient Hemodialysis or Rehabilitation Facilities.

RATIONALE & OBJECTIVE: Since January 2017, patients with acute kidney injury requiring dialysis (AKI-D) can be discharged to outpatient dialysis centers for continued hemodialysis (HD) support. We aimed to examine the rate of kidney recovery, time to recovery, and hospitalization-related clinical parameters associated with kidney recovery in patients with AKI-D. STUDY DESIGN: Single-center prospective cohort study. SETTING & PARTICIPANTS: 111 adult patients who were admitted to the University of Kentucky Hospital, experienced AKI-D, and were discharged with need of outpatient HD. EXPOSURE: Hospitalization-related clinical parameters were evaluated. OUTCOME: Kidney recovery as a composite of being alive and no longer requiring HD or other form of kidney replacement therapy. ANALYTICAL APPROACH: Discrete-time survival analysis and logistic regression were used to determine adjusted probabilities of kidney recovery at prespecified time points and to evaluate clinical parameters associated with recovery. RESULTS: 45 (41%) patients recovered kidney function, 25 (55.5%) within the first 30 days following discharge, 16 (35.5%) within 30 to 60 days, and 4 (9%) within 60 to 90 days. Adjusted probabilities of recovery were 36.7%, 27.4%, and 6.3%, respectively. Of the remaining patients, 49 (44%) developed kidney failure requiring chronic kidney replacement therapy and 17 (15%) died or went to hospice. Patients who did not recover kidney function were older, had more comorbid conditions, had lower estimated glomerular filtration rates at baseline, and received more blood transfusions during hospitalization when compared with those who recovered kidney function. LIMITATIONS: Selection bias given that patients included in the study were all eligible for AKI management with outpatient HD as part of Medicare/Medicaid services. CONCLUSIONS: At least one-third of AKI-D survivors discharged from an acute care hospital dependent on HD recovered kidney function within the first 90 days of discharge, more commonly in the first 30 days postdischarge. Future studies should elucidate clinical parameters that can inform risk classification and interventions to promote kidney recovery in this vulnerable and growing population.

KGDAL: Knowledge Graph Guided Double Attention LSTM for Rolling Mortality Prediction for AKI-D Patients.

With the rapid accumulation of electronic health record (EHR) data, deep learning (DL) models have exhibited promising performance on patient risk prediction. Recent advances have also demonstrated the effectiveness of knowledge graphs (KG) in providing valuable prior knowledge for further improving DL model performance. However, it is still unclear how KG can be utilized to encode high-order relations among clinical concepts and how DL models can make full use of the encoded concept relations to solve real-world healthcare problems and to interpret the outcomes. We propose a novel knowledge graph guided double attention LSTM model named KGDAL for rolling mortality prediction for critically ill patients with acute kidney injury requiring dialysis (AKI-D). KGDAL constructs a KG-based two-dimension attention in both time and feature spaces. In the experiment with two large healthcare datasets, we compared KGDAL with a variety of rolling mortality prediction models and conducted an ablation study to test the effectiveness, efficacy, and contribution of different attention mechanisms. The results showed that KGDAL clearly outperformed all the compared models. Also, KGDAL-derived patient risk trajectories may assist healthcare providers to make timely decisions and actions. The source code, sample data, and manual of KGDAL are available at https://github.com/lucasliu0928/KGDAL.

Serum renin and major adverse kidney events in critically ill patients: a multicenter prospective study.

BACKGROUND: Preliminary studies have suggested that the renin-angiotensin system is activated in critical illness and associated with mortality and kidney outcomes. We sought to assess in a larger, multicenter study the relationship between serum renin and Major Adverse Kidney Events (MAKE) in intensive care unit (ICU) patients. METHODS: Prospective, multicenter study at two institutions of patients with and without acute kidney injury (AKI). Blood samples were collected for renin measurement a median of 2 days into the index ICU admission and 5-7 days later. The primary outcome was MAKE at hospital discharge, a composite of mortality, kidney replacement therapy, or reduced estimated glomerular filtration rate to ≤ 75% of baseline. RESULTS: Patients in the highest renin tertile were more severely ill overall, including more AKI, vasopressor-dependence, and severity of illness. MAKE were significantly greater in the highest renin tertile compared to the first and second tertiles. In multivariable logistic regression, this initial measurement of renin remained significantly associated with both MAKE as well as the individual component of mortality. The association of renin with MAKE in survivors was not statistically significant. Renin measurements at the second time point were also higher in patients with MAKE. The trajectory of the renin measurements between time 1 and 2 was distinct when comparing death versus survival, but not when comparing MAKE versus those without. CONCLUSIONS: In a broad cohort of critically ill patients, serum renin measured early in the ICU admission is associated with MAKE at discharge, particularly mortality.