A Student Teaching Assistant Program for Diversity, Equity, Inclusion, and Antiracism Curricular Enhancement.

PROBLEM: In the United States, physician bias is exhibited early in medical training and contributes to systemic inequities within the field of medicine. A lack of diversity, equity, inclusion, and antiracism (DEI-AR) content within medical curricula drives critical gaps in knowledge and deficiencies when preparing medical students to serve patients of diverse backgrounds. At the Mayo Clinic Alix School of Medicine (MCASOM), student-led curricular reviews between 2017 to 2018 and 2020 to 2021 revealed opportunities to improve DEI-AR content within preclinical courses. Course directors expressed concern of limited expertise and time to enact effective changes. APPROACH: The MCASOM DEI-AR teaching assistant (TA) program aims to curate a collaborative partnership between course directors and compensated student TAs to facilitate course enhancements responsive to the prior preclinical course review while centering standardized DEI-AR best practices. OUTCOMES: As of January 2024, the program has engaged 14 TAs and partnered with 24 preclinical courses. Postcourse student evaluation responses were collected from 8 courses for 2021 to 2022 (before enhancements) and 2022 to 2023 (after enhancements). Student satisfaction with DEI-AR content is tracked through postcourse evaluations, with preliminary data demonstrating improvement after DEI-AR curricular integration (improvement of mean preenhancement and postenhancement scores of 3.81 to 4.05; t12 = 1.79, P = .21). Qualitative student comments were sorted into general categories of positive, negative, or neutral, showing a 6.25% median increase in positive perception of DEI. NEXT STEPS: Plans for the MCASOM DEI-AR TA program include application of quality improvement strategies to improve program processes and outcomes. Development of a centralized dashboard that integrates course enhancement progress and ongoing feedback from evaluations is anticipated to facilitate this effort. The program additionally aims to develop partnerships with clinical clerkships, which would allow for a more comprehensive enhancement of the overall medical education experience related to DEI-AR.

Endosomal trafficking inhibitor EGA can control TLR7-mediated IFNα expression by human plasmacytoid dendritic cells.

Plasmacytoid dendritic cells (pDC) are the major producer of type 1 IFN in response to TLR7 agonists. Aberrant TLR7 activation and type 1 IFN expression by pDCs are linked to the pathogenesis of certain types of autoimmune diseases, including systemic lupus erythematosus (SLE). This study investigated the underlying mechanisms for TLR7-mediated cytokine expression by pDCs using a late endosome trafficking inhibitor, EGA (4-bromobenzaldehyde N-(2,6-dimethylphenyl) semicarbazone). We found that EGA treatment decreased IFNα expression by pDCs stimulated with imiquimod (R837), single-stranded RNA40, and influenza virus. EGA also decreased TNFα expression and secretion by R837-stimulated pDCs. Mechanistically, EGA treatment decreased phosphorylation of IKKα/ß, STAT1, and p38, and prolonged degradation of IκBα. Furthermore, EGA treatment decreased the colocalization of 3F, a substituted adenine TLR7 agonist, with LAMP1+ compartments in pDCs. EGA was also capable of diminishing IFNα expression by SLE pDCs treated with R837 or live PR8/A/34 influenza viruses. Therefore, we concluded that trafficking of TLR7 agonists to LAMP1+ compartments is important for IFNα expression by pDCs. Data from this study support additional examinations of the potential benefits of EGA in treating type 1 IFN-associated inflammatory diseases in the future.

The Value of Pharmacogenomics for White and Indigenous Americans after Kidney Transplantation.

BACKGROUND: There is a paucity of evidence to inform the value of pharmacogenomic (PGx) results in patients after kidney transplant and how these results differ between Indigenous Americans and Whites. This study aims to identify the frequency of recommended medication changes based on PGx results and compare the pharmacogenomic (PGx) results and patients' perceptions of the findings between a cohort of Indigenous American and White kidney transplant recipients. METHODS: Thirty-one Indigenous Americans and fifty White kidney transplant recipients were studied prospectively. Genetic variants were identified using the OneOme RightMed PGx test of 27 genes. PGx pharmacist generated a report of the genetic variation and recommended changes. Pre- and post-qualitative patient surveys were obtained. RESULTS: White and Indigenous American subjects had a similar mean number of medications at the time of PGx testing (mean 13 (SD 4.5)). In the entire cohort, 53% received beta blockers, 30% received antidepressants, 16% anticoagulation, 47% pain medication, and 25% statin therapy. Drug-gene interactions that warranted a clinical action were present in 21.5% of patients. In 12.7%, monitoring was recommended. Compared to the Whites, the Indigenous American patients had more normal CYP2C19 (p = 0.012) and CYP2D6 (p = 0.012) activities. The Indigenous American patients had more normal CYP4F2 (p = 0.004) and lower VKORC (p = 0.041) activities, phenotypes for warfarin drug dosing, and efficacy compared to the Whites. SLC6A4, which affects antidepressant metabolism, showed statistical differences between the two cohorts (p = 0.017); specifically, SLC6A4 had reduced expression in 45% of the Indigenous American patients compared to 20% of the White patients. There was no significant difference in patient perception before and after PGx. CONCLUSIONS: Kidney transplant recipients had several drug-gene interactions that were clinically actionable; over one-third of patients were likely to benefit from changes in medications or drug doses based on the PGx results. The Indigenous American patients differed in the expression of drug-metabolizing enzymes and drug transporters from the White patients.

Implementation of genomic medicine for rare disease in a tertiary healthcare system: Mayo Clinic Program for Rare and Undiagnosed Diseases (PRaUD).

BACKGROUND: In the United States, rare disease (RD) is defined as a condition that affects fewer than 200,000 individuals. Collectively, RD affects an estimated 30 million Americans. A significant portion of RD has an underlying genetic cause; however, this may go undiagnosed. To better serve these patients, the Mayo Clinic Program for Rare and Undiagnosed Diseases (PRaUD) was created under the auspices of the Center for Individualized Medicine (CIM) aiming to integrate genomics into subspecialty practice including targeted genetic testing, research, and education. METHODS: Patients were identified by subspecialty healthcare providers from 11 clinical divisions/departments. Targeted multi-gene panels or custom exome/genome-based panels were utilized. To support the goals of PRaUD, a new clinical service model, the Genetic Testing and Counseling (GTAC) unit, was established to improve access and increase efficiency for genetic test facilitation. The GTAC unit includes genetic counselors, genetic counseling assistants, genetic nurses, and a medical geneticist. Patients receive abbreviated point-of-care genetic counseling and testing through a partnership with subspecialty providers. RESULTS: Implementation of PRaUD began in 2018 and GTAC unit launched in 2020 to support program expansion. Currently, 29 RD clinical indications are included in 11 specialty divisions/departments with over 142 referring providers. To date, 1152 patients have been evaluated with an overall solved or likely solved rate of 17.5% and as high as 66.7% depending on the phenotype. Noteworthy, 42.7% of the solved or likely solved patients underwent changes in medical management and outcome based on genetic test results. CONCLUSION: Implementation of PRaUD and GTAC have enabled subspecialty practices advance expertise in RD where genetic counselors have not historically been embedded in practice. Democratizing access to genetic testing and counseling can broaden the reach of patients with RD and increase the diagnostic yield of such indications leading to better medical management as well as expanding research opportunities.

Safety and mortality outcomes for direct oral anticoagulants in renal transplant recipients.

PURPOSE: Direct oral anticoagulants (DOACs) are increasingly used in renal transplant recipients (RTR), but relatively understudied in this population. We assess the safety of post-transplant anticoagulation with DOACs compared to warfarin. METHODS: We conducted a retrospective study of RTRs at the Mayo Clinic sites (2011-present) that were anticoagulated for greater than 3 months excluding the 1st month post-transplant. The main safety outcomes were bleeding and all-cause mortality. Concomitant antiplatelet and interacting drugs were noted. DOAC dose adjustment was assessed according to common US prescribing practices, guidelines, and/or FDA labeling. RESULTS: The median follow-up was longer for RTRs on warfarin (1098 days [IQR 521, 1517]) than DOACs (449 days [IQR 338, 942]). Largely, there were no differences in baseline characteristics and comorbidities between RTRs on DOACs (n = 208; apixaban 91.3%, rivaroxaban 8.7%) versus warfarin (n = 320). There was no difference in post-transplant use of antiplatelets, immunosuppressants, most antifungals assessed, or amiodarone. There was no significant difference in incident major bleeding (8.4 vs. 5.3%, p = 0.89), GI bleeding (4.4% vs. 1.9%, p = 0.98), or intra-cranial hemorrhage (1.9% vs. 1.4%, p = 0.85) between warfarin and DOAC. There was no significant difference in mortality in the warfarin group compared to DOACs when adjusted for follow-up time (22.2% vs. 10.1%, p = 0.21). Rates of post-transplant venous thromboembolism, atrial fibrillation or stroke were similar between the two groups. 32% (n = 67) of patients on DOACs were dose reduced, where 51% of those reductions were warranted. 7% of patients that were not dose reduced should have been. CONCLUSIONS: DOACs did not have inferior bleeding or mortality outcomes compared to warfarin in RTRs. There was greater use of warfarin compared to DOACs and a high rate of improper DOAC dose reduction.

Disruption of endosomal trafficking with EGA alters TLR9 cytokine response in human plasmacytoid dendritic cells.

Plasmacytoid dendritic cells (pDCs) exhibit bifurcated cytokine responses to TLR9 agonists, an IRF7-mediated type 1 IFN response or a pro-inflammatory cytokine response via the activation of NF-κB. This bifurcated response has been hypothesized to result from either distinct signaling endosomes or endo-lysosomal trafficking delay of TLR9 agonists allowing for autocrine signaling to affect outcomes. Utilizing the late endosome trafficking inhibitor, EGA, we assessed the bifurcated cytokine responses of pDCs to TLR9 stimulation. EGA treatment of pDCs diminished both IFNα and pro-inflammatory cytokine expression induced by CpG DNAs (D- and K-type), CpG-DNAs complexed with DOTAP, and genomic DNAs complexed with LL37. Mechanistically, EGA suppressed phosphorylation of IKKα/ß, STAT1, Akt, and p38, and decreased colocalization of CpG oligodeoxynucleotides with LAMP+ endo-lysosomes. EGA also diminished type 1 IFN expression by pDCs from systemic lupus erythematosus patients. Therefore, our findings help understand mechanisms for the bifurcated cytokine responses by pDCs and support future examination of the potential benefit of EGA in treating type 1 IFN-associated inflammatory diseases in the future.

Characteristics and Graft Outcomes of Kidney Transplant Candidates with Enteric Risk Factors and Elevated Plasma Oxalate Levels.

INTRODUCTION: This study describes patient characteristics and examines graft function of kidney transplant recipients (without primary hyperoxaluria) with elevated plasma oxalate (POx) and enteric risk factors prior to transplant at our institution. METHODS: Kidney transplant recipients between 2012 and 2020 with elevated POx at the time of kidney transplant evaluation were included. A matched control cohort was gathered using patient/donor age, living/deceased donor type, panel reactive antibody, kidney donor profile index, and human leukocyte antigen mismatch as matching variables. Graft function at 1 year and at last follow-up was reported. RESULTS: A total of 106 patients with elevated POx were identified. A third of the patients had Roux-en-Y gastric bypass, a third had other enteric risks, and a third did not have an identifiable enteric risk. Median eGFR (estimated glomerular filtration rate) at 1 year and at last follow-up was similar between cases and controls except for subgroup of patients with pre-transplant POx >30 µmol/L where 1-year eGFR was lower compared to controls. Across eGFR categories, more cases were in eGFR category <30 mL/min/1.73 m2 compared to controls. CONCLUSION: Roux-en-Y gastric bypass is the most common identifiable risk for elevated POx in kidney transplant candidates. 1-year graft function was not inferior in cases compared to matched controls except for subgroup with POx >30 µmol/L pre-transplant.

GFR estimated with creatinine rather than cystatin C is more reflective of the true risk of adverse outcomes with low GFR in kidney transplant recipients.

BACKGROUND: Serum cystatin C-based estimated glomerular filtration rate (eGFRcys) generally associates with clinical outcomes better than serum creatinine-based eGFR (eGFRcr) despite similar precision in estimating measured GFR (mGFR). We sought to determine whether the risk of adverse outcomes with eGFRcr or eGFRcys was via GFR alone or also via non-GFR determinants among kidney transplant recipients. METHODS: Consecutive adult kidney transplant recipients underwent a standardized GFR assessment during a routine follow-up clinic visit between 2011 and 2013. Patients were followed for graft failure or the composite outcome of cardiovascular (CV) events or mortality through 2020. The risk of these events by baseline mGFR, eGFRcr and eGFRcys was assessed unadjusted, adjusted for mGFR and adjusted for CV risk factors. RESULTS: There were 1135 recipients with a mean baseline mGFR of 55.6, eGFRcr of 54.8 and eGFRcys of 46.8 ml/min/1.73 m2 and a median follow-up of 6 years. Each 10 ml/min/1.73 m2 decrease in mGFR, eGFRcr or eGFRcys associated with graft failure [hazard ratio (HR) 1.79, 1.68 and 2.07, respectively; P < .001 for all) and CV events or mortality outcome (HR 1.28, 1.19 and 1.43, respectively; P < .001 for all). After adjusting for mGFR, eGFRcys associated with graft failure (HR 1.57, P < .001) and CV events or mortality (HR 1.49, P < .001), but eGFRcr did not associate with either. After further adjusting for CV risk factors, risk of these outcomes with lower eGFRcys was attenuated. CONCLUSION: eGFRcr better represents the true relationship between GFR and outcomes after kidney transplantation because it has less non-GFR residual association. Cystatin C is better interpreted as a nonspecific prognostic biomarker than is eGFR in the kidney transplant setting.

Cardiologist Evaluation and Approval Was the Primary Predictor of Kidney Transplant Candidacy and Transplantation Among Patients With Reduced Left Ventricular Ejection Fraction.

End-stage kidney disease patients with concomitant heart failure (HF) with reduced ejection fraction are often denied kidney transplantation. The aims of this study were to explore factors predictive of suitability for kidney transplant and to assess cardiovascular outcomes in patients with impaired left ventricular ejection fraction (LVEF) after transplant. Methods: We evaluated 109 consecutive adults with LVEF ≤40% at the time of initial kidney transplant evaluation between 2013 and 2018. Posttransplant cardiovascular outcomes were defined as nonfatal myocardial infarction (MI), admission for HF, cardiovascular death, and all-cause mortality. Results: A cardiologist participated in kidney transplant evaluation for 87% of patients and was present at 49% of transplant selection conferences. Twenty-four patients (22%) were denied by a cardiologist for kidney transplant' and 59 (54%) were denied by the selection committee, of whom 43 were because of cardiovascular risk. Forty-two (38%) patients were approved for kidney transplant. On univariate analysis, the variables associated with denial for kidney transplant included cardiologist denial, higher cardiac troponin T, prior coronary intervention, cardiovascular event, positive stress study, lower ejection fraction, and lower VO2 max (all P < 0.05). Cardiologist denial was the most significant predictor of denial for kidney transplant in different multivariate models. At a median follow-up of 15 mo, 5 (5%) suffered nonfatal MI, 13 (12%) were hospitalized for HF exacerbation, and 17 (16%) died. Only 22 patients, 52% of those approved, underwent kidney transplant. After kidney transplant, there was 1 death, 1 nonfatal MI, and 3 hospitalizations for HF. Median LVEF improved from 38% before listing to 55% posttransplant. Conclusions: Cardiologist denial was the primary predictor of rejection for kidney transplant. Despite careful selection, prevalence of cardiovascular events and mortality after kidney transplant was 23%. There is need for a structured multidisciplinary approach for patients with impaired LVEF.

Explainable Preoperative Automated Machine Learning Prediction Model for Cardiac Surgery-Associated Acute Kidney Injury.

BACKGROUND: We aimed to develop and validate an automated machine learning (autoML) prediction model for cardiac surgery-associated acute kidney injury (CSA-AKI). METHODS: Using 69 preoperative variables, we developed several models to predict post-operative AKI in adult patients undergoing cardiac surgery. Models included autoML and non-autoML types, including decision tree (DT), random forest (RF), extreme gradient boosting (XGBoost), and artificial neural network (ANN), as well as a logistic regression prediction model. We then compared model performance using area under the receiver operating characteristic curve (AUROC) and assessed model calibration using Brier score on the independent testing dataset. RESULTS: The incidence of CSA-AKI was 36%. Stacked ensemble autoML had the highest predictive performance among autoML models, and was chosen for comparison with other non-autoML and multivariable logistic regression models. The autoML had the highest AUROC (0.79), followed by RF (0.78), XGBoost (0.77), multivariable logistic regression (0.77), ANN (0.75), and DT (0.64). The autoML had comparable AUROC with RF and outperformed the other models. The autoML was well-calibrated. The Brier score for autoML, RF, DT, XGBoost, ANN, and multivariable logistic regression was 0.18, 0.18, 0.21, 0.19, 0.19, and 0.18, respectively. We applied SHAP and LIME algorithms to our autoML prediction model to extract an explanation of the variables that drive patient-specific predictions of CSA-AKI. CONCLUSION: We were able to present a preoperative autoML prediction model for CSA-AKI that provided high predictive performance that was comparable to RF and superior to other ML and multivariable logistic regression models. The novel approaches of the proposed explainable preoperative autoML prediction model for CSA-AKI may guide clinicians in advancing individualized medicine plans for patients under cardiac surgery.

Etiology, urine metabolic risk factors, and urine oxalate patterns in patients with significant hyperoxaluria and recurrent nephrolithiasis.

PURPOSE: American Urology Association guidelines recommend genetic testing for patients with recurrent stones and urine oxalate > 75 mg/day. The goal of this study was to examine the treatment of patients in this category in a large multidisciplinary adult stone clinic. METHODS: Patients were evaluated from a single institution between 2006 and 2019. Those with at least one level of urinary oxalate excretion (uOx) above 75 mg/day were identified. A chart review identified enteric risk factors and genetic testing results. Patients without an identifiable enteric cause were considered idiopathic. RESULTS: A total of 4229 separate 24-h urine collections in 1302 patients were reviewed. At least one measurement of uOx above 75 mg/day was found in 103 (7.9%) patients. Enteric hyperoxaluria (EH) was seen in 28 (27%) and idiopathic hyperoxaluria (IH) in 76 (74%). 20 (71%) patients in the EH group had undergone gastric bypass. The median uOx was significantly higher level in the EH group (121.0 vs. 93.0 mg/day). For the entire cohort, there was a drop in uOx (- 33.8 mg/day) with medical and dietary therapy after a follow-up of 46.6 months. The final oxalate was higher in EH (88.9 vs. 60.1 mg/day). Only one patient had referral for genetic testing and was found to have primary hyperoxaluria type 2. CONCLUSIONS: The most common cause of significant hyperoxaluria in patients with recurrent nephrolithiasis remains idiopathic. Patients with IH have more significant improvement in uOx compared to EH; however, both groups had hyperoxaluria at last follow-up. Rate of genetic testing is low despite guideline recommendations.

Subtyping hospitalized patients with hypokalemia by machine learning consensus clustering and associated mortality risks.

BACKGROUND: Hospitalized patients with hypokalemia are heterogeneous and cluster analysis, an unsupervised machine learning methodology, may discover more precise and specific homogeneous groups within this population of interest. Our study aimed to cluster patients with hypokalemia at hospital admission using an unsupervised machine learning approach and assess the mortality risk among these distinct clusters. METHODS: We performed consensus clustering analysis based on demographic information, principal diagnoses, comorbidities and laboratory data among 4763 hospitalized adult patients with admission serum potassium ≤3.5 mEq/L. We calculated the standardized mean difference of each variable and used the cutoff of ±0.3 to identify each cluster's key features. We assessed the association of the hypokalemia cluster with hospital and 1-year mortality. RESULTS: Consensus cluster analysis identified three distinct clusters that best represented patients' baseline characteristics. Cluster 1 had 1150 (32%) patients, cluster 2 had 1344 (28%) patients and cluster 3 had 1909 (40%) patients. Based on the standardized difference, patients in cluster 1 were younger, had less comorbidity burden but higher estimated glomerular filtration rate (eGFR) and higher hemoglobin; patients in cluster 2 were older, more likely to be admitted for cardiovascular disease and had higher serum sodium and chloride levels but lower eGFR, serum bicarbonate, strong ion difference (SID) and hemoglobin, while patients in cluster 3 were older, had a greater comorbidity burden, higher serum bicarbonate and SID but lower serum sodium, chloride and eGFR. Compared with cluster 1, cluster 2 had both higher hospital and 1-year mortality, whereas cluster 3 had higher 1-year mortality but comparable hospital mortality. CONCLUSION: Our study demonstrated the use of consensus clustering analysis in the heterogeneous cohort of hospitalized hypokalemic patients to characterize their patterns of baseline clinical and laboratory data into three clinically distinct clusters with different mortality risks.

Hypernatremia subgroups among hospitalized patients by machine learning consensus clustering with different patient survival.

BACKGROUND: The objective of this study was to characterize hypernatremia patients at hospital admission into clusters using an unsupervised machine learning approach and to evaluate the mortality risk among these distinct clusters. METHODS: We performed consensus cluster analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 922 hospitalized adult patients with admission serum sodium of > 145 mEq/L. We calculated the standardized difference of each variable to identify each cluster's key features. We assessed the association of each hypernatremia cluster with hospital and 1-year mortality. RESULTS: There were three distinct clusters of patients with hypernatremia on admission: 318 (34%) patients in cluster 1, 339 (37%) patients in cluster 2, and 265 (29%) patients in cluster 3. Cluster 1 consisted of more critically ill patients with more severe hypernatremia and hypokalemic hyperchloremic metabolic acidosis. Cluster 2 consisted of older patients with more comorbidity burden, body mass index, and metabolic alkalosis. Cluster 3 consisted of younger patients with less comorbidity burden, higher baseline eGFR, hemoglobin, and serum albumin. Compared to cluster 3, odds ratios for hospital mortality were 15.74 (95% CI 3.75-66.18) for cluster 1, and 6.51 (95% CI 1.48-28.59) for cluster 2, whereas hazard ratios for 1-year mortality were 6.25 (95% CI 3.69-11.46) for cluster 1 and 4.66 (95% CI 2.73-8.59) for cluster 2. CONCLUSION: Our cluster analysis identified three clinically distinct phenotypes with differing mortality risk in patients hospitalized with hypernatremia.

Machine Learning Consensus Clustering Approach for Hospitalized Patients with Dysmagnesemia.

BACKGROUND: The objectives of this study were to classify patients with serum magnesium derangement on hospital admission into clusters using unsupervised machine learning approach and to evaluate the mortality risks among these distinct clusters. METHODS: Consensus cluster analysis was performed based on demographic information, principal diagnoses, comorbidities, and laboratory data in hypomagnesemia (serum magnesium ≤ 1.6 mg/dL) and hypermagnesemia cohorts (serum magnesium ≥ 2.4 mg/dL). Each cluster's key features were determined using the standardized mean difference. The associations of the clusters with hospital mortality and one-year mortality were assessed. RESULTS: In hypomagnesemia cohort (n = 13,320), consensus cluster analysis identified three clusters. Cluster 1 patients had the highest comorbidity burden and lowest serum magnesium. Cluster 2 patients had the youngest age, lowest comorbidity burden, and highest kidney function. Cluster 3 patients had the oldest age and lowest kidney function. Cluster 1 and cluster 3 were associated with higher hospital and one-year mortality compared to cluster 2. In hypermagnesemia cohort (n = 4671), the analysis identified two clusters. Compared to cluster 1, the key features of cluster 2 included older age, higher comorbidity burden, more hospital admissions primarily due to kidney disease, more acute kidney injury, and lower kidney function. Compared to cluster 1, cluster 2 was associated with higher hospital mortality and one-year mortality. CONCLUSION: Our cluster analysis identified clinically distinct phenotypes with differing mortality risks in hospitalized patients with dysmagnesemia. Future studies are required to assess the application of this ML consensus clustering approach to care for hospitalized patients with dysmagnesemia.

Clinical Outcomes and Histological Patterns in Oxalate Nephropathy due to Enteric and Nonenteric Risk Factors.

INTRODUCTION: Current knowledge of risk factors and renal histologic patterns of oxalate nephropathy (ON) not due to primary hyperoxaluria (PH) has been limited to small case series and case reports. Thus, we analyzed and compared clinical risk factors, histologic characteristics, and renal outcomes of patients with biopsy-confirmed ON among a cohort of patients with enteric and nonenteric risk factors. METHODS: A clinical data repository of native kidney pathology reports from 2009 to 2020 at all Mayo Clinic sites was used to identify 421 ON cases. RESULTS: After excluding cases in transplanted kidneys or due to PH, 64 cases remained. Enteric risk factors were present in 30 and nonenteric in 34. Roux-en-Y gastric bypass (17) and pancreatic insufficiency (6) were most common in the enteric hyperoxaluria group. In the nonenteric group, vitamin C (7) and dietary oxalate (7) were common, while no apparent risk was noted in 16. Acute kidney injury (AKI) stage III at the time of diagnosis was present in 60%, and 40.6% required dialysis. Patients in the nonenteric group had more interstitial inflammation (p = 0.01), and a greater number of tubules contained intratubular calcium oxalate (CaOx) crystals (p = 0.001) than the nonenteric group. Patients in the enteric group were more likely to have baseline chronic kidney disease (CKD) (p = 0.02) and moderate-to-severe tubulointerstitial fibrosis and atrophy (IFTA) (OR 3.49, p = 0.02). After a median follow-up of 10 months, 39% were dialysis dependent, 11% received a kidney transplant, and 32% died. On univariate analysis, >10 tubules with CaOx crystals, baseline CKD, and AKI requiring dialysis correlated with the risk of dialysis, transplant, or death. On multivariate analysis, only AKI requiring dialysis correlated with adverse renal outcomes. CONCLUSION: This is the largest cohort study of ON not due to PH. Histologic features differ in patients with enteric versus nonenteric risks. Patients in the enteric group are more likely to have baseline CKD and significant IFTA, while patients in the nonenteric group were more likely to have a greater number of tubules with CaOx crystals and corresponding interstitial inflammation. AKI requiring dialysis at the time of diagnosis was the single most significant predictor of adverse renal outcome.

Genomics Integration Into Nephrology Practice.

RATIONALE & OBJECTIVE: The etiology of kidney disease remains unknown in many individuals with chronic kidney disease (CKD). We created the Mayo Clinic Nephrology Genomics Clinic to improve our ability to integrate genomic and clinical data to identify the etiology of unexplained CKD. STUDY DESIGN: Retrospective study. SETTING & PARTICIPANTS: An essential component of our program is the Nephrology Genomics Board which consists of nephrologists, geneticists, pathologists, translational omics scientists, and trainees who interpret the patient's clinical and genetic data. Since September 2016, the Board has reviewed 163 cases (15 cystic, 100 glomerular, 6 congenital anomalies of kidney and urinary tract (CAKUT), 20 stones, 15 tubulointerstitial, and 13 other). ANALYTICAL APPROACH: Testing was performed with targeted panels, single gene analysis, or analysis of kidney-related genes from exome sequencing. Variant classification was obtained based on the 2015 American College of Medical Genetics and Genomics and the Association for Molecular Pathology guidelines. RESULTS: A definitive genetic diagnosis was achieved for 50 families (30.7%). The highest diagnostic yield was obtained in individuals with tubulointerstitial diseases (53.3%), followed by congenital anomalies of the kidney and urological tract (33.3%), glomerular (31%), cysts (26.7%), stones (25%), and others (15.4%). A further 20 (12.3%) patients had variants of interest, and variant segregation, and research activities (exome, genome, or transcriptome sequencing) are ongoing for 44 (40%) unresolved families. LIMITATIONS: Possible overestimation of diagnostic rate due to inclusion of individuals with variants with evidence of pathogenicity but classified as of uncertain significance by the clinical laboratory. CONCLUSIONS: Integration of genomic and research testing and multidisciplinary evaluation in a nephrology cohort with CKD of unknown etiology or suspected monogenic disease provided a diagnosis in a third of families. These diagnoses had prognostic implications, and often changes in management were implemented.

Machine Learning Consensus Clustering Approach for Hospitalized Patients with Phosphate Derangements.

BACKGROUND: The goal of this study was to categorize patients with abnormal serum phosphate upon hospital admission into distinct clusters utilizing an unsupervised machine learning approach, and to assess the mortality risk associated with these clusters. METHODS: We utilized the consensus clustering approach on demographic information, comorbidities, principal diagnoses, and laboratory data of hypophosphatemia (serum phosphate ≤ 2.4 mg/dL) and hyperphosphatemia cohorts (serum phosphate ≥ 4.6 mg/dL). The standardized mean difference was applied to determine each cluster's key features. We assessed the association of the clusters with mortality. RESULTS: In the hypophosphatemia cohort (n = 3113), the consensus cluster analysis identified two clusters. The key features of patients in Cluster 2, compared with Cluster 1, included: older age; a higher comorbidity burden, particularly hypertension; diabetes mellitus; coronary artery disease; lower eGFR; and more acute kidney injury (AKI) at admission. Cluster 2 had a comparable hospital mortality (3.7% vs. 2.9%; p = 0.17), but a higher one-year mortality (26.8% vs. 14.0%; p < 0.001), and five-year mortality (20.2% vs. 44.3%; p < 0.001), compared to Cluster 1. In the hyperphosphatemia cohort (n = 7252), the analysis identified two clusters. The key features of patients in Cluster 2, compared with Cluster 1, included: older age; more primary admission for kidney disease; more history of hypertension; more end-stage kidney disease; more AKI at admission; and higher admission potassium, magnesium, and phosphate. Cluster 2 had a higher hospital (8.9% vs. 2.4%; p < 0.001) one-year mortality (32.9% vs. 14.8%; p < 0.001), and five-year mortality (24.5% vs. 51.1%; p < 0.001), compared with Cluster 1. CONCLUSION: Our cluster analysis classified clinically distinct phenotypes with different mortality risks among hospitalized patients with serum phosphate derangements. Age, comorbidities, and kidney function were the key features that differentiated the phenotypes.

Clinically Distinct Subtypes of Acute Kidney Injury on Hospital Admission Identified by Machine Learning Consensus Clustering.

BACKGROUND: We aimed to cluster patients with acute kidney injury at hospital admission into clinically distinct subtypes using an unsupervised machine learning approach and assess the mortality risk among the distinct clusters. METHODS: We performed consensus clustering analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 4289 hospitalized adult patients with acute kidney injury at admission. The standardized difference of each variable was calculated to identify each cluster's key features. We assessed the association of each acute kidney injury cluster with hospital and one-year mortality. RESULTS: Consensus clustering analysis identified four distinct clusters. There were 1201 (28%) patients in cluster 1, 1396 (33%) patients in cluster 2, 1191 (28%) patients in cluster 3, and 501 (12%) patients in cluster 4. Cluster 1 patients were the youngest and had the least comorbidities. Cluster 2 and cluster 3 patients were older and had lower baseline kidney function. Cluster 2 patients had lower serum bicarbonate, strong ion difference, and hemoglobin, but higher serum chloride, whereas cluster 3 patients had lower serum chloride but higher serum bicarbonate and strong ion difference. Cluster 4 patients were younger and more likely to be admitted for genitourinary disease and infectious disease but less likely to be admitted for cardiovascular disease. Cluster 4 patients also had more severe acute kidney injury, lower serum sodium, serum chloride, and serum bicarbonate, but higher serum potassium and anion gap. Cluster 2, 3, and 4 patients had significantly higher hospital and one-year mortality than cluster 1 patients (p < 0.001). CONCLUSION: Our study demonstrated using machine learning consensus clustering analysis to characterize a heterogeneous cohort of patients with acute kidney injury on hospital admission into four clinically distinct clusters with different associated mortality risks.

Subtyping Hyperchloremia among Hospitalized Patients by Machine Learning Consensus Clustering.

Background and Objectives: Despite the association between hyperchloremia and adverse outcomes, mortality risks among patients with hyperchloremia have not consistently been observed among all studies with different patient populations with hyperchloremia. The objective of this study was to characterize hyperchloremic patients at hospital admission into clusters using an unsupervised machine learning approach and to evaluate the mortality risk among these distinct clusters. Materials and Methods: We performed consensus cluster analysis based on demographic information, principal diagnoses, comorbidities, and laboratory data among 11,394 hospitalized adult patients with admission serum chloride of >108 mEq/L. We calculated the standardized mean difference of each variable to identify each cluster's key features. We assessed the association of each hyperchloremia cluster with hospital and one-year mortality. Results: There were three distinct clusters of patients with admission hyperchloremia: 3237 (28%), 4059 (36%), and 4098 (36%) patients in clusters 1 through 3, respectively. Cluster 1 was characterized by higher serum chloride but lower serum sodium, bicarbonate, hemoglobin, and albumin. Cluster 2 was characterized by younger age, lower comorbidity score, lower serum chloride, and higher estimated glomerular filtration (eGFR), hemoglobin, and albumin. Cluster 3 was characterized by older age, higher comorbidity score, higher serum sodium, potassium, and lower eGFR. Compared with cluster 2, odds ratios for hospital mortality were 3.60 (95% CI 2.33-5.56) for cluster 1, and 4.83 (95% CI 3.21-7.28) for cluster 3, whereas hazard ratios for one-year mortality were 4.49 (95% CI 3.53-5.70) for cluster 1 and 6.96 (95% CI 5.56-8.72) for cluster 3. Conclusions: Our cluster analysis identified three clinically distinct phenotypes with differing mortality risks in hospitalized patients with admission hyperchloremia.