Dataset Information

Machine-learning-based early prediction of end-stage renal disease in patients with diabetic kidney disease using clinical trials data.

ABSTRACT:

Aim

To predict end-stage renal disease (ESRD) in patients with type 2 diabetes by using machine-learning models with multiple baseline demographic and clinical characteristics.

Materials and methods

In total, 11?789 patients with type 2 diabetes and nephropathy from three clinical trials, RENAAL (n = 1513), IDNT (n = 1715) and ALTITUDE (n = 8561), were used in this study. Eighteen baseline demographic and clinical characteristics were used as predictors to train machine-learning models to predict ESRD (doubling of serum creatinine and/or ESRD). We used the area under the receiver operator curve (AUC) to assess the prediction performance of models and compared this with traditional Cox proportional hazard regression and kidney failure risk equation models.

Results

The feed forward neural network model predicted ESRD with an AUC of 0.82 (0.76-0.87), 0.81 (0.75-0.86) and 0.84 (0.79-0.90) in the RENAAL, IDNT and ALTITUDE trials, respectively. The feed forward neural network model selected urinary albumin to creatinine ratio, serum albumin, uric acid and serum creatinine as important predictors and obtained a state-of-the-art performance for predicting long-term ESRD.

Conclusions

Despite large inter-patient variability, non-linear machine-learning models can be used to predict long-term ESRD in patients with type 2 diabetes and nephropathy using baseline demographic and clinical characteristics. The proposed method has the potential to create accurate and multiple outcome prediction automated models to identify high-risk patients who could benefit from therapy in clinical practice.

SUBMITTER: Belur Nagaraj S

PROVIDER: S-EPMC7756814 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine-learning-based early prediction of end-stage renal disease in patients with diabetic kidney disease using clinical trials data.

Belur Nagaraj Sunil S Pena Michelle J MJ Ju Wenjun W Heerspink Hiddo L HL

Diabetes, obesity & metabolism 20200922 12

<h4>Aim</h4>To predict end-stage renal disease (ESRD) in patients with type 2 diabetes by using machine-learning models with multiple baseline demographic and clinical characteristics.<h4>Materials and methods</h4>In total, 11 789 patients with type 2 diabetes and nephropathy from three clinical trials, RENAAL (n = 1513), IDNT (n = 1715) and ALTITUDE (n = 8561), were used in this study. Eighteen baseline demographic and clinical characteristics were used as predictors to train machine-learning m ...[more]

PMID: 32844582

Similar Datasets

Project description:BackgroundEnd stage renal disease (ESRD) describes the most severe stage of chronic kidney disease (CKD), when patients need dialysis or renal transplant. There is often a delay in recognizing, diagnosing, and treating the various etiologies of CKD. The objective of the present study was to employ machine learning algorithms to develop a prediction model for progression to ESRD based on a large-scale multidimensional database.MethodsThis study analyzed 10,000,000 medical insurance claims from 550,000 patient records using a commercial health insurance database. Inclusion criteria were patients over the age of 18 diagnosed with CKD Stages 1-4. We compiled 240 predictor candidates, divided into six feature groups: demographics, chronic conditions, diagnosis and procedure features, medication features, medical costs, and episode counts. We used a feature embedding method based on implementation of the Word2Vec algorithm to further capture temporal information for the three main components of the data: diagnosis, procedures, and medications. For the analysis, we used the gradient boosting tree algorithm (XGBoost implementation).ResultsThe C-statistic for the model was 0.93 [(0.916-0.943) 95% confidence interval], with a sensitivity of 0.715 and specificity of 0.958. Positive Predictive Value (PPV) was 0.517, and Negative Predictive Value (NPV) was 0.981. For the top 1 percentile of patients identified by our model, the PPV was 1.0. In addition, for the top 5 percentile of patients identified by our model, the PPV was 0.71. All the results above were tested on the test data only, and the threshold used to obtain these results was 0.1. Notable features contributing to the model were chronic heart and ischemic heart disease as a comorbidity, patient age, and number of hypertensive crisis events.ConclusionsWhen a patient is approaching the threshold of ESRD risk, a warning message can be sent electronically to the physician, who will initiate a referral for a nephrology consultation to ensure an investigation to hasten the establishment of a diagnosis and initiate management and therapy when appropriate.

Project description:BackgroundAnemia is common and is associated with impaired clinical outcomes in diabetic chronic kidney disease (CKD). It may be explained by reduced erythropoietin (EPO) synthesis, but recent data suggest that EPO-resistance and diminished iron availability due to inflammation contribute significantly. In this cohort study, we evaluated the impact of hepcidin-25--the key hormone of iron-metabolism--on clinical outcomes in diabetic patients with CKD along with endogenous EPO levels.Methods249 diabetic patients with CKD of any stage, excluding end-stage renal disease (ESRD), were enrolled (2003-2005), if they were not on EPO-stimulating agent and iron therapy. Hepcidin-25 levels were measured by radioimmunoassay. The association of hepcidin-25 at baseline with clinical variables was investigated using linear regression models. All-cause mortality and a composite endpoint of CKD progression (ESRD or doubling of serum creatinine) were analyzed by Cox proportional hazards models.ResultsPatients (age 67 yrs, 53% male, GFR 51 ml/min, hemoglobin 131 g/L, EPO 13.5 U/L, hepcidin-25 62.0 ng/ml) were followed for a median time of 4.2 yrs. Forty-nine patients died (19.7%) and forty (16.1%) patients reached the composite endpoint. Elevated hepcidin levels were independently associated with higher ferritin-levels, lower EPO-levels and impaired kidney function (all p<0.05). Hepcidin was related to mortality, along with its interaction with EPO, older age, greater proteinuria and elevated CRP (all p<0.05). Hepcidin was also predictive for progression of CKD, aside from baseline GFR, proteinuria, low albumin- and hemoglobin-levels and a history of CVD (all p<0.05).ConclusionsWe found hepcidin-25 to be associated with EPO and impaired kidney function in diabetic CKD. Elevated hepcidin-25 and EPO-levels were independent predictors of mortality, while hepcidin-25 was also predictive for progression of CKD. Both hepcidin-25 and EPO may represent important prognostic factors of clinical outcome and have the potential to further define "high risk" populations in CKD.

Project description:Risk of end-stage renal disease (ESRD) in kidney donors has been compared with risk faced by the general population, but the general population represents an unscreened, high-risk comparator. A comparison to similarly screened healthy nondonors would more properly estimate the sequelae of kidney donation.To compare the risk of ESRD in kidney donors with that of a healthy cohort of nondonors who are at equally low risk of renal disease and free of contraindications to live donation and to stratify these comparisons by patient demographics.A cohort of 96,217 kidney donors in the United States between April 1994 and November 2011 and a cohort of 20,024 participants of the Third National Health and Nutrition Examination Survey (NHANES III) were linked to Centers for Medicare & Medicaid Services data to ascertain development of ESRD, which was defined as the initiation of maintenance dialysis, placement on the waiting list, or receipt of a living or deceased donor kidney transplant, whichever was identified first. Maximum follow-up was 15.0 years; median follow-up was 7.6 years (interquartile range [IQR], 3.9-11.5 years) for kidney donors and 15.0 years (IQR, 13.7-15.0 years) for matched healthy nondonors.Cumulative incidence and lifetime risk of ESRD.Among live donors, with median follow-up of 7.6 years (maximum, 15.0), ESRD developed in 99 individuals in a mean (SD) of 8.6 (3.6) years after donation. Among matched healthy nondonors, with median follow-up of 15.0 years (maximum, 15.0), ESRD developed in 36 nondonors in 10.7 (3.2) years, drawn from 17 ESRD events in the unmatched healthy nondonor pool of 9364. Estimated risk of ESRD at 15 years after donation was 30.8 per 10,000 (95% CI, 24.3-38.5) in kidney donors and 3.9 per 10,000 (95% CI, 0.8-8.9) in their matched healthy nondonor counterparts (P?<?.001). This difference was observed in both black and white individuals, with an estimated risk of 74.7 per 10,000 black donors (95% CI, 47.8-105.8) vs 23.9 per 10,000 black nondonors (95% CI, 1.6-62.4; P?<?.001) and an estimated risk of 22.7 per 10,000 white donors (95% CI, 15.6-30.1) vs 0.0 white nondonors (P?<?.001). Estimated lifetime risk of ESRD was 90 per 10,000 donors, 326 per 10,000 unscreened nondonors (general population), and 14 per 10,000 healthy nondonors.Compared with matched healthy nondonors, kidney donors had an increased risk of ESRD over a median of 7.6 years; however, the magnitude of the absolute risk increase was small. These findings may help inform discussions with persons considering live kidney donation.

Dataset Information

Machine-learning-based early prediction of end-stage renal disease in patients with diabetic kidney disease using clinical trials data.

Aim

Materials and methods

Results

Conclusions

Publications

Machine-learning-based early prediction of end-stage renal disease in patients with diabetic kidney disease using clinical trials data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets