Dataset Information

Development and external validation of a diagnostic model for biopsy-proven acute interstitial nephritis using electronic health record data.

ABSTRACT:

Background

Patients with acute interstitial nephritis (AIN) can present without typical clinical features, leading to a delay in diagnosis and treatment. We therefore developed and validated a diagnostic model to identify patients at risk of AIN using variables from the electronic health record.

Methods

In patients who underwent a kidney biopsy at Yale University between 2013 and 2018, we tested the association of >150 variables with AIN, including demographics, comorbidities, vital signs and laboratory tests (training set 70%). We used least absolute shrinkage and selection operator methodology to select prebiopsy features associated with AIN. We performed area under the receiver operating characteristics curve (AUC) analysis with internal (held-out test set 30%) and external validation (Biopsy Biobank Cohort of Indiana). We tested the change in model performance after the addition of urine biomarkers in the Yale AIN study.

Results

We included 393 patients (AIN 22%) in the training set, 158 patients (AIN 27%) in the test set, 1118 patients (AIN 11%) in the validation set and 265 patients (AIN 11%) in the Yale AIN study. Variables in the selected model included serum creatinine {adjusted odds ratio [aOR] 2.31 [95% confidence interval (CI) 1.42-3.76]}, blood urea nitrogen:creatinine ratio [aOR 0.40 (95% CI 0.20-0.78)] and urine dipstick specific gravity [aOR 0.95 (95% CI 0.91-0.99)] and protein [aOR 0.39 (95% CI 0.23-0.68)]. This model showed an AUC of 0.73 (95% CI 0.64-0.81) in the test set, which was similar to the AUC in the external validation cohort [0.74 (95% CI 0.69-0.79)]. The AUC improved to 0.84 (95% CI 0.76-0.91) upon the addition of urine interleukin-9 and tumor necrosis factor-α.

Conclusions

We developed and validated a statistical model that showed a modest AUC for AIN diagnosis, which improved upon the addition of urine biomarkers. Future studies could evaluate this model and biomarkers to identify unrecognized cases of AIN.

SUBMITTER: Moledina DG

PROVIDER: S-EPMC9755995 | biostudies-literature | 2022 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Development and external validation of a diagnostic model for biopsy-proven acute interstitial nephritis using electronic health record data.

Moledina Dennis G DG Eadon Michael T MT Calderon Frida F Yamamoto Yu Y Shaw Melissa M Perazella Mark A MA Simonov Michael M Luciano Randy R Schwantes-An Tae-Hwi TH Moeckel Gilbert G Kashgarian Michael M Kuperman Michael M Obeid Wassim W Cantley Lloyd G LG Parikh Chirag R CR Wilson F Perry FP

Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association 20221001 11

<h4>Background</h4>Patients with acute interstitial nephritis (AIN) can present without typical clinical features, leading to a delay in diagnosis and treatment. We therefore developed and validated a diagnostic model to identify patients at risk of AIN using variables from the electronic health record.<h4>Methods</h4>In patients who underwent a kidney biopsy at Yale University between 2013 and 2018, we tested the association of >150 variables with AIN, including demographics, comorbidities, ...[more]

PMID: 34865148

Similar Datasets

Project description:BackgroundIgA vasculitis (IgAV) is the most common form of systemic vasculitis in childhood and frequently involves the kidney. A minority of patients with IgA vasculitis nephritis (IgAVN), especially those presenting with heavy proteinuria and/or kidney failure at onset, are at risk of chronic end-stage kidney disease. For deciding upon treatment intensity, knowledge of the short-term clinical course of IgAVN is needed to improve treatment algorithms.MethodsFor this retrospective multicenter study, the medical records of 66 children with biopsy-proven IgAVN were reviewed. Age, gender, medical history and therapeutic interventions were recorded. Laboratory data included serum creatinine, albumin, urinary protein excretion (UPE) and glomerular filtration rate (eGFR). Threshold values were determined for each parameter, full remission was defined as no proteinuria and eGFR > 90 ml/min/1.73m2.ResultsMedian age at onset of IgAVN was 8.9 years. 14.1% of the children presented with nephrotic syndrome, 50% had an eGFR below 90 ml/min/1.73 m2 and 51.5% showed cellular crescents in renal histology. The treatment regimens varied notably. Forty-four patients were treated with immunosuppression; 17 patients with crescents or nephrotic syndrome were treated with corticosteroid (CS) pulse therapy. After 6 months, UPE had decreased from 3.7 to 0.3 g/g creatinine and the proportion of patients with a decreased eGFR had fallen from 50.0% to 35.5%. Thirteen children (26.5%) achieved full remission within 6 months.ConclusionsIn most patients with IgAVN proteinuria decreases slowly and kidney function improves, but full remission is reached only in a minority after 6 months. Persistent heavy proteinuria in the first two months rarely developed into long-term proteinuria. Therefore, decisions for more intense treatment should take into account the course of UPE over time. For a comparison of treatment effects, patient numbers were too small. Prospective, randomized controlled trials are necessary to clarify risk factors and the effect of immunosuppressive therapies.

Project description:IntroductionLong-term data pertaining to childhood-onset lupus nephritis (cLN) remain extremely scarce.MethodsWe conducted a retrospective cohort study of biopsy-proven cLN with onset age <18 years diagnosed from 2001 to 2020 to ascertain the long-term patient and kidney survival rates, and the incidence of advanced chronic kidney disease (CKD) (estimated glomerular filtration rate [eGFR] <60 ml/min per 1.73 m2).ResultsA total of 92 subjects (78 female; age 13.7 ± 3.3 years; all Chinese) were included, with follow-up duration of 10.3 years (interquartile range, 5.8-15.9). Of these, 83 children (90%) had proliferative lupus nephritis (LN) (Class III/IV ± V). Mycophenolate was used for induction in 36%, whereas 34% received cyclophosphamide (CYC); 55% received mycophenolate as maintenance immunosuppression. The rates of complete remission (CR) and partial remission (PR) at 6 months and 12 months, respectively, were 65% and 20% and 78% and 8%. Two patients died (mortality rate 2.1/1000 patient-years), with a standardized mortality ratio of 22.3. Three patients (3.2%) developed end-stage kidney disease (ESKD), and advanced CKD occurred in 5 patients (5.4%). Survival rates without advanced CKD, ESKD, or death were 96.7%, 94.2%, 92.7%, 83.2% and 83.2% at 1 year, 5 years, 10 years, 15 years, and 20 years, respectively. Multivariate analysis revealed that severe kidney failure necessitating dialysis at presentation (adjusted hazard ratio 37.7, 95% confidence interval [CI] 4.0-355.6, P = 0.002), nonresponse (NR) after 12 months of treatment (adjusted hazard ratio 11.2, 95% CI 2.3-54.9, P = 0.003), and multiple nephritis flares (adjusted hazard ratio 2.6, 95% CI 1.1-6.2, P = 0.03) were predictive of advanced CKD, ESKD, or death. Other adverse outcomes included infections (2.9 episodes/100 patient-years), osteopenia (32%), hypertension (17%), short stature (14%), and avascular necrosis (7%).ConclusionThe long-term outcomes of cLN appeared to have improved in the present era with effective immunosuppression, cautious drug tapering, and assurance of medication adherence. There is still an unacceptably high prevalence of adverse outcomes.

Project description:BackgroundAcute tubulointerstitial nephritis (AIN) is one of the few causes of acute kidney injury with diagnosis-specific treatment options. However, due to the need to obtain a kidney biopsy for histological confirmation, AIN diagnosis can be delayed, missed, or incorrectly assumed. Here, we identify and validate urinary CXCL9, an IFN-γ-induced chemokine involved in lymphocyte chemotaxis, as a diagnostic biomarker for AIN.MethodsIn a prospectively enrolled cohort with pathologist-adjudicated histological diagnoses, termed the discovery cohort, we tested the association of 180 immune proteins measured by an aptamer-based assay with AIN and validated the top protein, CXCL9, using sandwich immunoassay. We externally validated these findings in 2 cohorts with biopsy-confirmed diagnoses, termed the validation cohorts, and examined mRNA expression differences in kidney tissue from patients with AIN and individuals in the control group.ResultsIn aptamer-based assay, urinary CXCL9 was 7.6-fold higher in patients with AIN than in individuals in the control group (P = 1.23 × 10-5). Urinary CXCL9 measured by sandwich immunoassay was associated with AIN in the discovery cohort (n = 204; 15% AIN) independently of currently available clinical tests for AIN (adjusted odds ratio for highest versus lowest quartile: 6.0 [1.8-20]). Similar findings were noted in external validation cohorts, where CXCL9 had an AUC of 0.94 (0.86-1.00) for AIN diagnosis. CXCL9 mRNA expression was 3.9-fold higher in kidney tissue from patients with AIN (n = 19) compared with individuals in the control group (n = 52; P = 5.8 × 10-6).ConclusionWe identified CXCL9 as a diagnostic biomarker for AIN using aptamer-based urine proteomics, confirmed this association using sandwich immunoassays in discovery and external validation cohorts, and observed higher expression of this protein in kidney biopsies from patients with AIN.FundingThis study was supported by National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) awards K23DK117065 (DGM), K08DK113281 (KM), R01DK128087 (DGM), R01DK126815 (DGM and LGC), R01DK126477 (KNC), UH3DK114866 (CRP, DGM, and FPW), R01DK130839 (MES), and P30DK079310 (the Yale O'Brien Center). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Project description:BackgroundEffective fall prevention interventions in hospitals require appropriate allocation of resources early in admission. To address this, fall risk prediction tools and models have been developed with the aim to provide fall prevention strategies to patients at high risk. However, fall risk assessment tools have typically been inaccurate for prediction, ineffective in prevention, and time-consuming to complete. Accurate, dynamic, individualized estimates of fall risk for admitted patients using routinely recorded data may assist in prioritizing fall prevention efforts.ObjectiveThe objective of this study was to develop and validate an accurate and dynamic prognostic model for inpatient falls among a cohort of patients using routinely recorded electronic medical record data.MethodsWe used routinely recorded data from 5 Australian hospitals to develop and internally-externally validate a prediction model for inpatient falls using a Cox proportional hazards model with time-varying covariates. The study cohort included patients admitted during 2018-2021 to any ward, with no age restriction. Predictors used in the model included admission-related administrative data, length of stay, and number of previous falls during the admission (updated every 12 hours up to 14 days after admission). Model calibration was assessed using Poisson regression and discrimination using the area under the time-dependent receiver operating characteristic curve.ResultsThere were 1,107,556 inpatient admissions, 6004 falls, and 5341 unique fallers. The area under the time-dependent receiver operating characteristic curve was 0.899 (95% CI 0.88-0.91) at 24 hours after admission and declined throughout admission (eg, 0.765, 95% CI 0.75-0.78 on the seventh day after admission). Site-dependent overestimation and underestimation of risk was observed on the calibration plots.ConclusionsUsing a large dataset from multiple hospitals and robust methods to model development and validation, we developed a prognostic model for inpatient falls. It had high discrimination, suggesting the model has the potential for operationalization in clinical decision support for prioritizing inpatients for fall prevention. Performance was site dependent, and model recalibration may lead to improved performance.

Dataset Information

Development and external validation of a diagnostic model for biopsy-proven acute interstitial nephritis using electronic health record data.

Background

Methods

Results

Conclusions

Publications

Development and external validation of a diagnostic model for biopsy-proven acute interstitial nephritis using electronic health record data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets