Clinical predictive factors and prediction models for end-stage renal disease in Chinese patients with type 2 diabetes mellitus.
Ontology highlight
ABSTRACT:
SUBMITTER: Gao Y
PROVIDER: S-EPMC10310973 | biostudies-literature | 2023 Jul
REPOSITORIES: biostudies-literature
Publications
Clinical predictive factors and prediction models for end-stage renal disease in Chinese patients with type 2 diabetes mellitus.
Gao Yueming Y Shang Zhi Z Nie Sheng S Feng Songtao S Wang Bin B Li Zuolin Z Wu Min M Wen Yi Y Xu Hong H Weng Jianping J Chen Chunbo C Liu Huafeng H Yang Qiongqiong Q Li Hua H Kong Yaozhong Y Li Guisen G Wan Qijun Q Zha Yan Y Hu Ying Y Xu Gang G Shi Yongjun Y Zhou Yilun Y Su Guobin G Tang Ying Y Gong Mengchun M Fan Hou Fan HF Liu Bicheng B
Clinical and translational medicine 20230701 7
Similar Datasets
Project description:BackgroundSince diabetes mellitus (DM) is the leading cause of end stage renal disease (ESRD), this study aimed to develop a 5-year ESRD risk prediction model among Chinese patients with Type 2 DM (T2DM) in primary care.MethodsA retrospective cohort study was conducted on 149,333 Chinese adult T2DM primary care patients without ESRD in 2010. Using the derivation cohort over a median of 5 years follow-up, the gender-specific models including the interaction effect between predictors and age were derived using Cox regression with a forward stepwise approach. Harrell's C-statistic and calibration plot were applied to the validation cohort to assess discrimination and calibration of the models.ResultsPrediction models showed better discrimination with Harrell's C-statistics of 0.866 (males) and 0.862 (females) and calibration power from the plots than other established models. The predictors included age, usages of anti-hypertensive drugs, anti-glucose drugs, and Hemogloblin A1c, blood pressure, urine albumin/creatinine ratio (ACR) and estimated glomerular filtration rate (eGFR). Specific predictors for male were smoking and presence of sight threatening diabetic retinopathy while additional predictors for female included longer duration of diabetes and quadratic effect of body mass index. Interaction factors with age showed a greater weighting of insulin and urine ACR in younger males, and eGFR in younger females.ConclusionsOur newly developed gender-specific models provide a more accurate 5-year ESRD risk predictions for Chinese diabetic primary care patients than other existing models. The models included several modifiable risk factors that clinicians can use to counsel patients, and to target at in the delivery of care to patients.
| S-EPMC5539616 | biostudies-literature
Project description:Background: To investigate the relationship between serum iron status and renal outcome in patients with type 2 diabetes mellitus (T2DM). Methods: Chinese patients (n=111) with T2DM and biopsy-proven diabetic nephropathy (DN) were surveyed in a longitudinal, retrospective study. Serum iron, total iron-binding capacity, ferritin, and transferrin were measured at the time of renal biopsy. Iron deposition and transferrin staining were performed with renal biopsy specimens of DN patients and potential kidney donors. End-stage renal disease (ESRD) was the end-point. ESRD was defined as an estimated glomerular filtration rate <15 mL/min/1.73 m2 or the need for chronic renal replacement therapy. Cox proportional hazard models were used to estimate the hazard ratios (HRs) for the influence of serum iron metabolism on ESRD. Results: During a median follow up of 30.9 months, 66 (59.5%) patients progressed to ESRD. After adjusting for age, sex, baseline systolic blood pressure, renal functions, hemoglobin, HbA1c, and pathological findings, lower serum transferrin concentrations were significantly associated with higher ESRD in multivariate models. Compared with patients in the highest transferrin quartile (≥1.65 g/L), patients in the lowest quartile (≤1.15 g/L) had multivariable-adjusted HR (95% confidence interval) of 7.36 (1.40-38.65) for ESRD. Moreover, tubular epithelial cells in DN exhibited a higher deposition of iron and transferrin expression compared with healthy controls. Conclusions: Low serum transferrin concentration was associated with diabetic ESRD in patients with T2DM. Free iron nephrotoxicity and poor nutritional status with accumulated iron or transferrin deposition might contribute to ESRD.
| S-EPMC7484672 | biostudies-literature
Project description:Background and objectivesThe study aimed to evaluate the performance of a predictive model using the kidney failure risk equation (KFRE) for end-stage renal disease (ESRD) in diabetes and to investigate the impact of glomerular filtration rate (GFR) as estimated by different equations on the performance of the KFRE model in diabetes.Design setting participants and measurementsA total of 18,928 individuals with diabetes without ESRD history from the UK Biobank, a prospective cohort study initiated in 2006-2010, were included in this study. Modification of diet in renal disease (MDRD), chronic kidney disease epidemiology collaboration (CKD-EPI) or revised Lund-Malmö (r-LM) were used to estimate GFR in the KFRE model. Cox proportional risk regression was used to determine the correlation coefficients between each variable and ESRD risk in each model. Harrell's C-index and net reclassification improvement (NRI) index were used to evaluate the differentiation of the models. Analysis was repeated in subgroups based on albuminuria and hemoglobin A1C (HbA1c) levels.ResultsOverall, 132 of the 18,928 patients developed ESRD after a median follow-up of 12 years. The Harrell's C-index based on GFR estimated by CKD-EPI, MDRD, and r-LM was 0.914 (95% CI = 0.8812-0.9459), 0.908 (95% CI = 0.8727-0.9423), and 0.917 (95% CI = 0.8837-0.9496), respectively. Subgroup analysis revealed that in diabetic patients with macroalbuminuria, the KFRE model based on GFR estimated by r-LM (KFRE-eGFRr-LM) had better differentiation compared to the KFRE model based on GFR estimated by CKD-EPI (KFRE-eGFRCKD-EPI) with a KFRE-eGFRr-LM C-index of 0.846 (95% CI = 0.797-0.894, p = 0.025), while the KFRE model based on GFR estimated by MDRD (KFRE-eGFRMDRD) showed no significant difference compared to the KFRE-eGFRCKD-EPI (KFRE-eGFRMDRD C-index of 0.837, 95% CI = 0.785-0.889, p = 0.765). Subgroup analysis of poor glycemic control (HbA1c >8.5%) demonstrated the same trend. Compared to KFRE-eGFRCKD-EPI (C-index = 0.925, 95% CI = 0.874-0.976), KFRE-eGFRr-LM had a C-index of 0.935 (95% CI = 0.888-0.982, p = 0.071), and KFRE-eGFRMDRD had a C-index of 0.925 (95% CI = 0.874-0.976, p = 0.498).ConclusionsIn adults with diabetes, the r-LM equation performs better than the CKD-EPI and MDRD equations in the KFRE model for predicting ESRD, especially for those with macroalbuminuria and poor glycemic control (HbA1c >8.5%).
| S-EPMC9205244 | biostudies-literature
Project description:ObjectivesType 2 diabetes mellitus (T2DM) imposes a great burden on healthcare systems, and these patients experience higher long-term risks for developing end-stage renal disease (ESRD). Managing diabetic nephropathy becomes more challenging when kidney function starts declining. Therefore, developing predictive models for the risk of developing ESRD in newly diagnosed T2DM patients may be helpful in clinical settings.MethodsWe established machine learning models constructed from a subset of clinical features collected from 53,477 newly diagnosed T2DM patients from January 2008 to December 2018 and then selected the best model. The cohort was divided, with 70% and 30% of patients randomly assigned to the training and testing sets, respectively.ResultsThe discriminative ability of our machine learning models, including logistic regression, extra tree classifier, random forest, gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and light gradient boosting machine were evaluated across the cohort. XGBoost yielded the highest area under the receiver operating characteristic curve (AUC) of 0.953, followed by extra tree and GBDT, with AUC values of 0.952 and 0.938 on the testing dataset. The SHapley Additive explanation summary plot in the XGBoost model illustrated that the top five important features included baseline serum creatinine, mean serum creatine within 1 year before the diagnosis of T2DM, high-sensitivity C-reactive protein, spot urine protein-to-creatinine ratio and female gender.ConclusionsBecause our machine learning prediction models were based on routinely collected clinical features, they can be used as risk assessment tools for developing ESRD. By identifying high-risk patients, intervention strategies may be provided at an early stage.
| S-EPMC10007785 | biostudies-literature
Project description:PurposeDiabetes mellitus (DM) is the most common cause of end-stage renal disease (ESRD) and an important risk factor for cardiovascular (CV) disease. We investigated the impact of DM on subclinical CV damage by comprehensive screening protocol in ESRD patients.Materials and methodsEchocardiography, coronary computed tomography angiogram, 24-h ambulatory blood pressure monitoring, and central blood pressure with pulse wave velocity (PWV) were performed in 91 ESRD patients from the Cardiovascular and Metabolic disease Etiology Research Center-HIgh risk cohort.ResultsThe DM group (n=38) had higher systolic blood pressure than the non-DM group (n=53), however, other clinical CV risk factors were not different between two groups. Central aortic systolic pressure (148.7±29.8 mm Hg vs. 133.7±27.0 mm Hg, p= 0.014), PWV (12.1±2.7 m/s vs. 9.4±2.1 m/s, p<0.001), and early mitral inflow to early mitral annulus velocity (16.7±6.4 vs. 13.7±5.9, p=0.026) were higher in the DM group. Although the prevalence of coronary artery disease (CAD) was not different between the DM and the non-DM group (95% vs. 84.4%, p=0.471), the severity of CAD was higher in the DM group (p=0.01). In multivariate regression analysis, DM was an independent determinant for central systolic pressure (p=0.011), PWV (p<0.001) and the prevalence of CAD (p=0.046).ConclusionDiabetic ESRD patients have higher central systolic pressure and more advanced arteriosclerosis than the non-DM control group. These findings suggest that screening for subclinical CV damage may be helpful for diabetic ESRD patients.
| S-EPMC5122655 | biostudies-literature
Project description:Diabetes mellitus (DM) is an important risk factor in patients with end-stage renal disease (ESRD). DM is associated with the development of cardiovascular diseases, such as atrial fibrillation (AF), due to poor glycemic control. However, few studies have focused on the risk of developing ESRD among DM patients with and without AF. This study evaluated ESRD risk among DM patients with and without AF in Taiwan. Data were retrieved from one million patients randomly sampled from Taiwan's National Health Insurance Research Database, including 6,105 DM patients with AF propensity score-matched with 6,105 DM patients without AF. Both groups were followed until death, any dialysis treatment, or December 31, 2013, whichever occurred first. AF was diagnosed by a qualified physician according to the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), using the diagnostic code 427.31. Patients aged <20 years or diagnosed with ESRD before the index date were excluded. A Cox proportional hazard regression model was used to calculate the relative ESRD risk. Among DM patients, those with AF have more comorbidities than those without AF. We also found a 1.18-fold (95% confidence interval [CI]: 1.01-1.46) increase in ESRD risk among patients with AF compared with those without AF. In addition, DM patients with hypertension, chronic kidney disease (CKD), or higher Charlson Comorbidity Index scores also have significantly increased ESRD risks than those without these complications. A 1.39-fold (95% CI: 1.04-1.86) increase in risk was observed for patients with AF among the non-CKD group. Our findings suggest that patients with DM should be closely monitored for irregular or rapid heart rates.
| S-EPMC9417190 | biostudies-literature
Project description:This study aimed to develop a model to predict the 5-year risk of developing end-stage renal disease (ESRD) in patients with type 2 diabetes mellitus (T2DM) using machine learning (ML). It also aimed to implement the developed algorithms into electronic medical records (EMR) system using Health Level Seven (HL7) Fast Healthcare Interoperability Resources (FHIR). The final dataset used for modeling included 19,159 patients. The medical data were engineered to generate various types of features that were input into the various ML classifiers. The classifier with the best performance was XGBoost, with an area under the receiver operator characteristics curve (AUROC) of 0.95 and area under the precision recall curve (AUPRC) of 0.79 using three-fold cross-validation, compared to other models such as logistic regression, random forest, and support vector machine (AUROC range, 0.929-0.943; AUPRC 0.765-0.792). Serum creatinine, serum albumin, the urine albumin-to-creatinine ratio, Charlson comorbidity index, estimated GFR, and medication days of insulin were features that were ranked high for the ESRD risk prediction. The algorithm was implemented in the EMR system using HL7 FHIR through an ML-dedicated server that preprocessed unstructured data and trained updated data.
| S-EPMC9253099 | biostudies-literature
Project description:BACKGROUND:Diabetes Mellitus is an increasingly prevalent chronic disease characterized by the body's inability to metabolize glucose. The objective of this study was to build an effective predictive model with high sensitivity and selectivity to better identify Canadian patients at risk of having Diabetes Mellitus based on patient demographic data and the laboratory results during their visits to medical facilities. METHODS:Using the most recent records of 13,309 Canadian patients aged between 18 and 90 years, along with their laboratory information (age, sex, fasting blood glucose, body mass index, high-density lipoprotein, triglycerides, blood pressure, and low-density lipoprotein), we built predictive models using Logistic Regression and Gradient Boosting Machine (GBM) techniques. The area under the receiver operating characteristic curve (AROC) was used to evaluate the discriminatory capability of these models. We used the adjusted threshold method and the class weight method to improve sensitivity - the proportion of Diabetes Mellitus patients correctly predicted by the model. We also compared these models to other learning machine techniques such as Decision Tree and Random Forest. RESULTS:The AROC for the proposed GBM model is 84.7% with a sensitivity of 71.6% and the AROC for the proposed Logistic Regression model is 84.0% with a sensitivity of 73.4%. The GBM and Logistic Regression models perform better than the Random Forest and Decision Tree models. CONCLUSIONS:The ability of our model to predict patients with Diabetes using some commonly used lab results is high with satisfactory sensitivity. These models can be built into an online computer program to help physicians in predicting patients with future occurrence of diabetes and providing necessary preventive interventions. The model is developed and validated on the Canadian population which is more specific and powerful to apply on Canadian patients than existing models developed from US or other populations. Fasting blood glucose, body mass index, high-density lipoprotein, and triglycerides were the most important predictors in these models.
| S-EPMC6794897 | biostudies-literature
Project description:End-stage renal disease (ESRD) is associated with alterations in T-cell immunity, including increased CD28null and reduced regulatory T cells (Tregs). However, whether immune disturbances are due to ESRD or primary disease is not yet clear. As diabetes mellitus is the leading cause of ESRD, we evaluated its impact on the immune profile of ESRD patients. CD28null, Tregs, and natural killer cells were initially analyzed by flow cytometry in 30 predialysis ESRD patients due to diabetes (DM), 30 non-DM (NDM), and 25 healthy controls. Measurements were repeated after 6 months on hemodialysis (HD) or peritoneal dialysis (CAPD). The percentage of CD4 + CD28null cells, CD8 + CD28null cells, and Tregs showed significant differences in DM, NDM, and controls; mean rank 33.71 vs. 25.68 vs. 18.88, p = 0.006, 37.79 vs. 28.82 vs. 17.08, p = 0.008, and 20.79 vs. 26.12 vs. 41.33, p = 0.001, respectively. DM vs. NDM had increased CD4 + CD28null and CD8 + CD28null cells, 11.5% (1.5%-24%) vs. 4.1% (0-42.3%), p = 0.02 and 61.3% (24%-76%) vs. 43% (5.7%-85%), p = 0.04, respectively. After 6 months on HD but not CAPD, DM showed a significant further increase in CD4 + CD28null cells, from 30 (14-100) to 52.7 (15-203), p = 0.02; and CD8 + CD28null cells, from 137 (56-275) to 266 (103-456), p = 0.01. Diabetes mellitus affects T-cell subtypes even at predialysis stage, though changes become more prominent after commencement on HD.
| S-EPMC8001943 | biostudies-literature
Project description:BackgroundDiabetic kidney disease (DKD) is a common and serious complication in patients with diabetic mellitus (DM), the risk of cardiovascular events and all-cause mortality also increases in DKD patients. This study aimed to detect the influencing factors of DKD in type 2 DM (T2DM) patients, and construct DKD prediction models and nomogram for clinical decision-making.MethodsA total of 14,628 patients with T2DM were included. These patients were divided into pre-DKD and non-DKD groups, depending on the occurrence of DKD during a 3-year follow-up from first clinic attendance. The influencing indicators of DKD were analyzed, the prediction models were established by multivariable logistic regression, and a nomogram was drawn for DKD risk assessment.ResultsTwo prediction models for DKD were built by multivariate logistic regression analysis. Model 1 was created based on 17 variables using the forward selection method, Model 2 was established by 19 variables using the backward elimination method. The Somers' D values of both models were 0.789. Four independent predictors were selected to build the nomogram, including age, UACR, eGFR, and neutrophil percentages. The C-index of the nomogram reached 0.864, suggesting a good predictive accuracy for DKD development.ConclusionsOur prediction models had strong predictive powers, and our nomogram provided visual aids to DKD risk calculation, which was simple and fast. These algorithms can provide early DKD risk prediction, which might help to improve the medical care for early detection and intervention in T2DM patients, and then consequently improve the prognosis of DM patients.
| S-EPMC9427038 | biostudies-literature