Dataset Information

Kidney Histopathology and Prediction of Kidney Failure: A Retrospective Cohort Study.

ABSTRACT:

Rationale & objective

The use of kidney histopathology for predicting kidney failure is not established. We hypothesized that the use of histopathologic features of kidney biopsy specimens would improve prediction of clinical outcomes made using demographic and clinical variables alone.

Study design

Retrospective cohort study and development of a clinical prediction model.

Setting & participants

All 2,720 individuals from the Biopsy Biobank Cohort of Indiana who underwent kidney biopsy between 2002 and 2015 and had at least 2 years of follow-up.

New predictors & established predictors

Demographic variables, comorbid conditions, baseline clinical characteristics, and histopathologic features.

Outcomes

Time to kidney failure, defined as sustained estimated glomerular filtration rate ≤ 10mL/min/1.73m².

Analytical approach

Multivariable Cox regression model with internal validation by bootstrapping. Models including clinical and demographic variables were fit with the addition of histopathologic features. To assess the impact of adding a histopathology variable, the amount of variance explained (r²) and the C index were calculated. The impact on prediction was assessed by calculating the net reclassification index for each histopathologic variable and for all combined.

Results

Median follow-up was 3.1 years. Within 5 years of biopsy, 411 (15.1%) patients developed kidney failure. Multivariable analyses including demographic and clinical variables revealed that severe glomerular obsolescence (adjusted HR, 2.03; 95% CI, 1.51-2.03), severe interstitial fibrosis and tubular atrophy (adjusted HR, 1.99; 95% CI, 1.52-2.59), and severe arteriolar hyalinosis (adjusted HR, 1.53; 95% CI, 1.14-2.05) were independently associated with the primary outcome. The addition of all histopathologic variables to the clinical model yielded a net reclassification index for kidney failure of 5.1% (P < 0.001) with a full model C statistic of 0.915. Analyses addressing the competing risk for death, optimism, or shrinkage did not significantly change the results.

Limitations

Selection bias from the use of clinically indicated biopsies and exclusion of patients with less than 2 years of follow-up, as well as reliance on surrogate indicators of kidney failure onset.

Conclusions

A model incorporating histopathologic features from kidney biopsy specimens improved prediction of kidney failure and may be valuable clinically. Future studies will be needed to understand whether even more detailed characterization of kidney tissue may further improve prognostication about the future trajectory of estimated glomerular filtration rate.

SUBMITTER: Eadon MT

PROVIDER: S-EPMC7483298 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background: A mechanism to predict graft failure before the actual kidney transplantation occurs is crucial to clinical management of chronic kidney disease patients. Several kidney graft outcome prediction models, developed using machine learning methods, are available in the literature. However, most of those models used small datasets and none of the machine learning-based prediction models available in the medical literature modelled time-to-event (survival) information, but instead used the binary outcome of failure or not. The objective of this study is to develop two separate machine learning-based predictive models to predict graft failure following live and deceased donor kidney transplant, using time-to-event data in a large national dataset from Australia. Methods: The dataset provided by the Australia and New Zealand Dialysis and Transplant Registry will be used for the analysis. This retrospective dataset contains the cohort of patients who underwent a kidney transplant in Australia from January 1 st, 2007, to December 31 st, 2017. This included 3,758 live donor transplants and 7,365 deceased donor transplants. Three machine learning methods (survival tree, random survival forest and survival support vector machine) and one traditional regression method, Cox proportional regression, will be used to develop the two predictive models (for live donor and deceased donor transplants). The best predictive model will be selected based on the model's performance. Discussion: This protocol describes the development of two separate machine learning-based predictive models to predict graft failure following live and deceased donor kidney transplant, using a large national dataset from Australia. Furthermore, these two models will be the most comprehensive kidney graft failure predictive models that have used survival data to model using machine learning techniques. Thus, these models are expected to provide valuable insight into the complex interactions between graft failure and donor and recipient characteristics.

Project description:BackgroundInitiation of maintenance dialysis therapy for patients with chronic kidney failure is a period of high risk for adverse patient outcomes. Whether indications for dialysis therapy initiation are associated with mortality in this population is unknown.Study designRetrospective cohort study.Setting & participants461 patients who initiated dialysis therapy (hemodialysis, 437; peritoneal dialysis, 24) from January 1, 2004, through December 31, 2012, and were treated in facilities operated by a single dialysis organization. Follow-up for the primary outcome was through December 31, 2013.PredictorClinically documented primary indication for dialysis therapy initiation, as categorized into 4 groups: laboratory evidence of kidney function decline (reference category), uremic symptoms, volume overload or hypertension, and other/unknown.OutcomesAll-cause mortality.ResultsDuring a median follow-up of 2.4 years, 183 (40%) patients died. Crude mortality rates were 10.0 (95% CI, 6.8-14.7), 12.7 (95% CI, 10.2-15.7), 21.7 (95% CI, 16.4-28.6), and 12.2 (95% CI, 6.8-14.7) deaths/100 patient-years among patients initiating dialysis therapy primarily for laboratory evidence of kidney function decline, uremic symptoms, volume overload or hypertension, and other/unknown reason, respectively. Following adjustment for demographic variables, coexisting illnesses, and estimated glomerular filtration rate, initiation of dialysis therapy for uremic symptoms, volume overload or hypertension, or other/unknown reasons was associated with 1.12 (95% CI, 0.72-1.77), 1.69 (95% CI, 1.02-2.80), and 1.28 (95% CI, 0.73-2.26) times higher risk, respectively, for subsequent mortality compared to initiation for laboratory evidence of kidney function decline.LimitationsPossibility of residual confounding by unmeasured variables; reliance on clinical documentation to ascertain exposure.ConclusionsPatients initiating dialysis therapy due to volume overload may have increased risk for mortality compared with patients initiating dialysis due to laboratory evidence of kidney function decline. Further studies are needed to identify and test interventions that might reduce this risk.

Project description:Rationale & objectiveThis study describes the epidemiology, characteristics, and outcomes of patients with immunoglobulin A nephropathy (IgAN)-attributed kidney failure in the US Renal Data System (USRDS) from 2008 to 2018, including health care resource utilization and costs among patients with Medicare-linked data.Study designRetrospective cohort study.Setting & populationPatients with IgAN-attributed kidney failure in the USRDS.OutcomesPrevalence/incidence, clinical/demographic characteristics, time to kidney transplant, and health care resource utilization and costs.Analytical approachPatients with IgAN as primary cause of kidney failure (IgAN cohort) were followed from USRDS registration (index date) until data end/death. Prevalence/incidence were calculated per 1,000,000 US persons. Demographic and clinical characteristics at index and treatment modality during follow-up were summarized. Time from index to kidney transplant was assessed using Kaplan-Meier and competing risk analyses. Health care resource utilization and health care costs were reported among patients with 1 year Medicare Part A+B coverage postindex, including or excluding those who died (Medicare Coverage and 1-year Medicare Coverage subgroups, respectively).ResultsThe IgAN cohort, Medicare Coverage, and 1-year Medicare Coverage subgroups included 10,101, 1,696, and 1,510 patients, respectively. Mean annual period prevalence and incidence of IgAN-attributed kidney failure were 39.3 and 2.9 per 1,000,000 US persons, respectively. Initial treatment was in-center hemodialysis (63.1%) or kidney transplant (15.1%). Year 1 and 5 kidney transplant rates were 5% and 17%, respectively, accounting for competing risk of death. In the Medicare Coverage and 1-year Medicare Coverage subgroups, 74.4% and 72.3%, respectively, required inpatient admission, 67.3% and 64.4%, respectively, visited the emergency room, and mean total health care costs were $6,293 (SD: $6,934) and $5,284 ($3,455), respectively, per-patient-per-month in the year postindex.LimitationsDrug costs may be underestimated as Medicare Part D coverage was not required; kidney acquisition costs were unavailable.ConclusionsIgAN-attributed kidney failure is associated with substantial clinical and economic burdens. Novel therapies for IgAN that delay kidney failure are needed.

Project description:BackgroundThe current acute kidney injury (AKI) risk prediction model for patients undergoing percutaneous coronary intervention (PCI) from the American College of Cardiology (ACC) National Cardiovascular Data Registry (NCDR) employed regression techniques. This study aimed to evaluate whether models using machine learning techniques could significantly improve AKI risk prediction after PCI.Methods and findingsWe used the same cohort and candidate variables used to develop the current NCDR CathPCI Registry AKI model, including 947,091 patients who underwent PCI procedures between June 1, 2009, and June 30, 2011. The mean age of these patients was 64.8 years, and 32.8% were women, with a total of 69,826 (7.4%) AKI events. We replicated the current AKI model as the baseline model and compared it with a series of new models. Temporal validation was performed using data from 970,869 patients undergoing PCIs between July 1, 2016, and March 31, 2017, with a mean age of 65.7 years; 31.9% were women, and 72,954 (7.5%) had AKI events. Each model was derived by implementing one of two strategies for preprocessing candidate variables (preselecting and transforming candidate variables or using all candidate variables in their original forms), one of three variable-selection methods (stepwise backward selection, lasso regularization, or permutation-based selection), and one of two methods to model the relationship between variables and outcome (logistic regression or gradient descent boosting). The cohort was divided into different training (70%) and test (30%) sets using 100 different random splits, and the performance of the models was evaluated internally in the test sets. The best model, according to the internal evaluation, was derived by using all available candidate variables in their original form, permutation-based variable selection, and gradient descent boosting. Compared with the baseline model that uses 11 variables, the best model used 13 variables and achieved a significantly better area under the receiver operating characteristic curve (AUC) of 0.752 (95% confidence interval [CI] 0.749-0.754) versus 0.711 (95% CI 0.708-0.714), a significantly better Brier score of 0.0617 (95% CI 0.0615-0.0618) versus 0.0636 (95% CI 0.0634-0.0638), and a better calibration slope of observed versus predicted rate of 1.008 (95% CI 0.988-1.028) versus 1.036 (95% CI 1.015-1.056). The best model also had a significantly wider predictive range (25.3% versus 21.6%, p < 0.001) and was more accurate in stratifying AKI risk for patients. Evaluated on a more contemporary CathPCI cohort (July 1, 2015-March 31, 2017), the best model consistently achieved significantly better performance than the baseline model in AUC (0.785 versus 0.753), Brier score (0.0610 versus 0.0627), calibration slope (1.003 versus 1.062), and predictive range (29.4% versus 26.2%). The current study does not address implementation for risk calculation at the point of care, and potential challenges include the availability and accessibility of the predictors.ConclusionsMachine learning techniques and data-driven approaches resulted in improved prediction of AKI risk after PCI. The results support the potential of these techniques for improving risk prediction models and identification of patients who may benefit from risk-mitigation strategies.

Project description:BACKGROUND:Artificial intelligence-enabled electronic health record (EHR) analysis can revolutionize medical practice from the diagnosis and prediction of complex diseases to making recommendations in patient care, especially for chronic conditions such as chronic kidney disease (CKD), which is one of the most frequent complications in patients with diabetes and is associated with substantial morbidity and mortality. OBJECTIVE:The longitudinal prediction of health outcomes requires effective representation of temporal data in the EHR. In this study, we proposed a novel temporal-enhanced gradient boosting machine (GBM) model that dynamically updates and ensembles learners based on new events in patient timelines to improve the prediction accuracy of CKD among patients with diabetes. METHODS:Using a broad spectrum of deidentified EHR data on a retrospective cohort of 14,039 adult patients with type 2 diabetes and GBM as the base learner, we validated our proposed Landmark-Boosting model against three state-of-the-art temporal models for rolling predictions of 1-year CKD risk. RESULTS:The proposed model uniformly outperformed other models, achieving an area under receiver operating curve of 0.83 (95% CI 0.76-0.85), 0.78 (95% CI 0.75-0.82), and 0.82 (95% CI 0.78-0.86) in predicting CKD risk with automatic accumulation of new data in later years (years 2, 3, and 4 since diabetes mellitus onset, respectively). The Landmark-Boosting model also maintained the best calibration across moderate- and high-risk groups and over time. The experimental results demonstrated that the proposed temporal model can not only accurately predict 1-year CKD risk but also improve performance over time with additionally accumulated data, which is essential for clinical use to improve renal management of patients with diabetes. CONCLUSIONS:Incorporation of temporal information in EHR data can significantly improve predictive model performance and will particularly benefit patients who follow-up with their physicians as recommended.

Project description:BACKGROUND:Initial reports indicate a high incidence of acute kidney injury (AKI) in Coronavirus Disease 2019 (COVID-19), but more data are required to clarify if COVID-19 is an independent risk factor for AKI and how COVID-19-associated AKI may differ from AKI due to other causes. We therefore sought to study the relationship between COVID-19, AKI, and outcomes in a retrospective cohort of patients admitted to 2 acute hospitals in Derby, United Kingdom. METHODS AND FINDINGS:We extracted electronic data from 4,759 hospitalised patients who were tested for COVID-19 between 5 March 2020 and 12 May 2020. The data were linked to electronic patient records and laboratory information management systems. The primary outcome was AKI, and secondary outcomes included in-hospital mortality, need for ventilatory support, intensive care unit (ICU) admission, and length of stay. As compared to the COVID-19-negative group (n = 3,374), COVID-19 patients (n = 1,161) were older (72.1 ± 16.1 versus 65.3 ± 20.4 years, p < 0.001), had a greater proportion of men (56.6% versus 44.9%, p < 0.001), greater proportion of Asian ethnicity (8.3% versus 4.0%, p < 0.001), and lower proportion of white ethnicity (75.5% versus 82.5%, p < 0.001). AKI developed in 304 (26.2%) COVID-19-positive patients (COVID-19 AKI) and 420 (12.4%) COVID-19-negative patients (AKI controls). COVID-19 patients aged 65 to 84 years (odds ratio [OR] 1.67, 95% confidence interval [CI] 1.11 to 2.50), needing mechanical ventilation (OR 8.74, 95% CI 5.27 to 14.77), having congestive cardiac failure (OR 1.72, 95% CI 1.18 to 2.50), chronic liver disease (OR 3.43, 95% CI 1.17 to 10.00), and chronic kidney disease (CKD) (OR 2.81, 95% CI 1.97 to 4.01) had higher odds for developing AKI. Mortality was higher in COVID-19 AKI versus COVID-19 patients without AKI (60.5% versus 27.4%, p < 0.001), and AKI was an independent predictor of mortality (OR 3.27, 95% CI 2.39 to 4.48). Compared with AKI controls, COVID-19 AKI was observed in a higher proportion of men (58.9% versus 51%, p = 0.04) and lower proportion with white ethnicity (74.7% versus 86.9%, p = 0.003); was more frequently associated with cerebrovascular disease (11.8% versus 6.0%, p = 0.006), chronic lung disease (28.0% versus 19.3%, p = 0.007), diabetes (24.7% versus 17.9%, p = 0.03), and CKD (34.2% versus 20.0%, p < 0.001); and was more likely to be hospital acquired (61.2% versus 46.4%, p < 0.001). Mortality was higher in the COVID-19 AKI as compared to the control AKI group (60.5% versus 27.6%, p < 0.001). In multivariable analysis, AKI patients aged 65 to 84 years, (OR 3.08, 95% CI 1.77 to 5.35) and ?85 years of age (OR 3.54, 95% CI 1.87 to 6.70), peak AKI stage 2 (OR 1.74, 95% CI 1.05 to 2.90), AKI stage 3 (OR 2.01, 95% CI 1.13 to 3.57), and COVID-19 (OR 3.80, 95% CI 2.62 to 5.51) had higher odds of death. Limitations of the study include retrospective design, lack of urinalysis data, and low ethnic diversity of the region. CONCLUSIONS:We observed a high incidence of AKI in patients with COVID-19 that was associated with a 3-fold higher odds of death than COVID-19 without AKI and a 4-fold higher odds of death than AKI due to other causes. These data indicate that patients with COVID-19 should be monitored for the development of AKI and measures taken to prevent this. TRIAL REGISTRATION:ClinicalTrials.gov NCT04407156.