Derivation and external validation of a risk prediction algorithm to estimate future risk of cardiovascular death among patients with type 2 diabetes and incident diabetic nephropathy: prospective cohort study

ObjectiveTo derive, and externally validate, a risk score for cardiovascular death among patients with type 2 diabetes and newly diagnosed diabetic nephropathy (DN).Research design and methodsTwo independent prospective cohorts with type 2 diabetes were used to develop and externally validate the risk score. The derivation cohort comprised 2282 patients with an incident, clinical diagnosis of DN. The validation cohort includes 950 patients with incident, biopsy-proven diagnosis of DN. The outcome was cardiovascular death within 2 years of the diagnosis of DN. Logistic regression was applied to derive the risk score for cardiovascular death from the derivation cohort, which was externally validated in the validation cohort. The score was also estimated by applying the United Kingdom Prospective Diabetes Study (UKPDS) risk score in the external validation cohort.ResultsThe 2-year cardiovascular mortality was 12.05% and 11.79% in the derivation cohort and validation cohort, respectively. Traditional predictors including age, gender, body mass index, blood pressures, glucose, lipid profiles alongside novel laboratory test items covering five test panels (liver function, serum electrolytes, thyroid function, blood coagulation and blood count) were included in the final model.C-statistics was 0.736 (95% CI 0.731 to 0.740) and 0.747 (95% CI 0.737 to 0.756) in the derivation cohort and validation cohort, respectively. The calibration slope was 0.993 (95% CI 0.974 to 1.013) and 1.000 (95% CI 0.981 to 1.020) in the derivation cohort and validation cohort, respectively.The UKPDS risk score substantially underestimated cardiovascular mortality.ConclusionsA new risk score based on routine clinical measurements that quantified individual risk of cardiovascular death was developed and externally validated. Compared with the UKPDS risk score, which underestimated the cardiovascular disease risk, the new score is a more specific tool for patients with type 2 diabetes and DN. The score could work as a tool to identify individuals at the highest risk of cardiovascular death among those with DN.

Download Full-text

Development and External Validation of Risk Scores for Cardiovascular Hospitalization and Rehospitalization in Patients With Diabetes

The Journal of Clinical Endocrinology & Metabolism ◽

10.1210/jc.2017-02293 ◽

2018 ◽

Vol 103 (3) ◽

pp. 1122-1129 ◽

Cited By ~ 10

Author(s):

Dahai Yu ◽

Yamei Cai ◽

Jonathan Graffy ◽

Daniel Holman ◽

Zhanzheng Zhao ◽

...

Keyword(s):

Type 2 Diabetes ◽

Prediction Models ◽

Validation Cohort ◽

External Validation ◽

Developed Countries ◽

Risk Scores ◽

Derivation Cohort ◽

Patients With Diabetes ◽

Post Trial

Abstract Context Cardiovascular disease (CVD) is a common and costly reason for hospitalization and rehospitalization among patients with type 2 diabetes. Objective This study aimed to develop and externally validate two risk-prediction models for cardiovascular hospitalization and cardiovascular rehospitalization. Design Two independent prospective cohorts. Setting The derivation cohort includes 4704 patients with type 2 diabetes from 18 general practices in Cambridgeshire. The validation cohort comprises 1121 patients with type 2 diabetes from post-trial follow-up data. Main Outcome Measure Cardiovascular hospitalization over 2 years and cardiovascular rehospitalization after 90 days of the prior CVD hospitalization. Results The absolute rate of cardiovascular hospitalization and rehospitalization was 12.5% and 6.7% in the derivation cohort and 16.3% and 7.0% in the validation cohort. Discrimination of the models was similar in both cohorts, with C statistics above 0.70 and excellent calibration of observed and predicted risks. Conclusion Two prediction models that quantify risks of cardiovascular hospitalization and rehospitalization have been developed and externally validated. They are based on a small number of clinical measurements that are available for patients with type 2 diabetes in many developed countries in primary care settings and could serve as the tools to screen the population at high risk of cardiovascular hospitalization and rehospitalization.

Download Full-text

Interpretable Machine Learning Algorithm Reveals Novel Gut Microbiome Features in Predicting Type 2 Diabetes

Current Developments in Nutrition ◽

10.1093/cdn/nzaa062_016 ◽

2020 ◽

Vol 4 (Supplement_2) ◽

pp. 1559-1559

Author(s):

Wanglong Gou ◽

Chu-Wen Ling ◽

Yan He ◽

Zengliang Jiang ◽

Yuanqing Fu ◽

...

Keyword(s):

Machine Learning ◽

Type 2 Diabetes ◽

Gut Microbiome ◽

Large Scale ◽

Validation Cohort ◽

Learning Algorithm ◽

External Validation ◽

Positive Association ◽

Interpretable Machine Learning

Abstract Objectives The gut microbiome-type 2 diabetes (T2D) relationship among human cohorts have been controversial. We hypothesized that this limitation could be addressed by integrating the cutting-edge interpretable machine learning framework and large-scale human cohort studies. Methods 3 independent cohorts with >9000 participants were included in this study. We proposed a new machine learning-based analytic framework — using LightGBM to infer the relationship between incorporated features and T2D, and SHapley Additive explanation(SHAP) to identified microbiome features associated with the risk of T2D. We then generated a microbiome risk score (MRS) integrating the threshold and direction of the identified microbiome features to predict T2D risk. Results We finally identified 15 microbiome features (two of them are indicators of microbial diversity, others are taxa-related features) associated with the risk of T2D. The identified T2D-related gut microbiome features showed superior T2D prediction accuracy compared to host genetics or traditional risk factors. Furthermore, we found that the MRS (per unit change in MRS) consistently showed positive association with T2D risk in the discovery cohort (RR 1.28, 95%CI 1.23-1.33), external validation cohort 1 (RR 1.23, 95%CI 1.13-1.34) and external validation cohort 2 (GGMP, RR 1.12, 95%CI 1.06-1.18). The MRS could also predict future glucose increment. We subsequently identified dietary and lifestyle factors which could prospectively modulate the microbiome features, and found that body fat distribution may be the key factor modulating the gut microbiome-T2D relationship. Conclusions Taken together, we proposed a new analytical framework for the investigation of microbiome-disease relationship. The identified microbiome features may serve as potential drug targets for T2D in future. Funding Sources This study was funded by National Natural Science Foundation of China (81903316, 81773416), Westlake University (101396021801) and the 5010 Program for Clinical Researches (2007032) of the Sun Yat-sen University (Guangzhou, China).

Download Full-text

Novel Model Predicts Diabetic Nephropathy in Type 2 Diabetes

American Journal of Nephrology ◽

10.1159/000505145 ◽

2019 ◽

Vol 51 (2) ◽

pp. 130-138 ◽

Cited By ~ 2

Author(s):

Shimin Jiang ◽

Jinying Fang ◽

Tianyu Yu ◽

Lin Liu ◽

Guming Zou ◽

...

Keyword(s):

Type 2 Diabetes ◽

Diabetic Nephropathy ◽

Kidney Disease ◽

External Validation ◽

Curve Analysis ◽

Calibration Plot ◽

The Novel ◽

Decision Curve Analysis ◽

Novel Model

Background: Clinical indicators for accurately distinguishing diabetic nephropathy (DN) from non-diabetic renal disease in type 2 diabetes (T2D) are lacking. This study aimed to develop and validate a nomogram for predicting DN in T2D patients with kidney disease. Methods: A total of 302 consecutive patients with T2D who underwent renal biopsy at China-Japan Friendship Hospital between January 2014 and June 2019 were included in the study. The data were randomly split into a training set containing 70% of the patients (n = 214) and a validation set containing the remaining 30% of patients (n = 88). Multivariable logistic regression analyses were applied to develop a prediction nomogram incorporating the candidates selected in the least absolute shrinkage and selection operator regression model. Discrimination, calibration, and clinical usefulness of the prediction model were assessed using a concordance index (C-index), calibration plot, and decision curve analysis. Both internal and external validations were assessed. Results: A multivariable model that included gender, diabetes duration, diabetic retinopathy, hematuria, glycated hemoglobin A1c, anemia, blood pressure, urinary protein excretion, and estimated glomerular filtration rate was represented as the nomogram. The model demonstrated very good discrimination with a C-index of 0.934 (95% CI 0.904–0.964). The calibration plot diagram of predicted probabilities against observed DN rates indicated excellent concordance. The C-index value was 0.91 for internal validation and 0.875 for external validation. Decision curve analysis demonstrated that the novel nomogram was clinically useful. Conclusion: The novel model was very useful for predicting DN in patients with T2D and kidney disease, and thereby could be used by clinicians either in triage or as a replacement for biopsy.

Download Full-text

Remnant cholesterol and cardiovascular mortality in patients with type 2 diabetes and incident diabetic nephropathy

The Journal of Clinical Endocrinology & Metabolism ◽

10.1210/clinem/dgab533 ◽

2021 ◽

Author(s):

Dahai Yu ◽

Zheng Wang ◽

Xiaoxue Zhang ◽

Bingjie Qu ◽

Yamei Cai ◽

...

Keyword(s):

Type 2 Diabetes ◽

Diabetic Nephropathy ◽

Cardiovascular Mortality ◽

Density Lipoprotein ◽

Atherogenic Dyslipidemia ◽

Categorical Variables ◽

Newly Diagnosed ◽

Risk Of Death ◽

Remnant Cholesterol

Abstract Purpose This study examined the association between remnant cholesterol (remnant-C) and cardiovascular mortality in patients with type 2 diabetes (T2D), chronic kidney disease (CKD) stages 3-5 and newly diagnosed diabetic nephropathy (DN). Methods This study determined the baseline lipid profile and searched for deaths with cardiovascular disease (CVD) within 2 years of baseline among 2282 adults enrolled between 01/01/2015 and 31/12/2016 who had T2D, CKD stages 3-5 and newly diagnosed DN. Adjusted Logistic regression models were used to assess the associations between lipid, especially remnant-C concentration (either as continuous or categorical variables), and risk of cardiovascular mortality. Results In multivariable-adjusted analyses, low-density lipoprotein cholesterol (LDL-C) (odds ratio [OR]:1.022; 95% confidence interval [CI]: 1.017-1.026, per 10mg/dl), HDL-C (0.929 [0.922-0.936], per 5 mg/dl), Non-HDL-C (1.024 [1.021-1.028], per 10mg/dl), and remnant-C (1.115 [1.103-1.127], per 10mg/dl), but not triglyceride were associated with cardiovascular mortality. Atherogenic dyslipidemia (triglycerides >150 mg/dl [1.69 mmol/l] and HDL-C <40 mg/dl in men or <50 mg/dl in women) was also associated with cardiovascular mortality (1.073[1.031-1.116]). Remnant-C ≥30 mg/dl differentiated patients at a higher risk of cardiovascular mortality compared with those with lower concentrations, especially with interaction with LDL-C level >100 mg/dl: the highest risk was found in patients with higher levels of both remnant-C and LDL-C (1.696 [1.613-1.783]). Conclusions In patients with T2D, CKD stages 3-5 and incident DN, remnant-C was associated with higher risk of death with CVD. Different from the general population, the interaction of remnant-C and LDL-C was associated with the highest risk of cardiovascular mortality.

Download Full-text

Risk stratification on admission for predicting death or need for surgery in patients with acute type A intramural hematoma receiving medical therapy – validation cohort

European Heart Journal ◽

10.1093/ehjci/ehaa946.2338 ◽

2020 ◽

Vol 41 (Supplement_2) ◽

Author(s):

S Kageyama

Keyword(s):

Risk Score ◽

Primary Endpoint ◽

Validation Cohort ◽

Intramural Hematoma ◽

Type A ◽

Cardiovascular Death ◽

Ascending Aorta ◽

Acute Type ◽

Derivation Cohort ◽

Aorta Diameter

Abstract Background There has been continuing discussion regarding the treatment strategy for acute type A intramural hematoma (IMH). We previously examined the risk factors of death or need for surgery for acute type A IMH in patients receiving medical treatment using clinical findings on hospital arrival and developed a simple risk score using the factors. Purpose We examined the accuracy of the risk score in the validation cohort. Methods From 2009 to 2014, 57 consecutive patients diagnosed with acute type A IMH who were receiving initial medical treatment were retrospectively included for derivation cohort. Primary endpoint was a composite of cardiovascular death and operation within 1 year after onset. On admission, the primary endpoint could be predicted with 89.7% sensitivity and 75% specificity if the patient had ulcer-like projection (ULP) and/or ≥2 of the following factors: systolic blood pressure (SBP) <120 mmHg, ascending aorta diameter>45 mm, and pericardial effusion (PE). In the current study, validation cohort study was performed from 2015 to 2020 in 73 consecutive patients who met the same inclusion criteria for derivation cohort to evaluate the risk factors and the accuracy of the risk score. Result Mean age of onset was 74 years old. Mean SBP on arrival was 134 mmHg. Computed tomography (CT) on arrival showed a mean ascending aorta diameter of 46 mm. ULP and PE were seen in 27% and 41% of cases, respectively. Thirty-three patients (45.2%) reached the primary endpoint (cardiovascular death, 8 cases [11%]; operation, 25 cases [34.2%]). In univariate analysis of admission values, the primary endpoint group had significantly lower SBP (116±29 vs 147±35 mmHg), higher ascending aorta diameter (49±8 vs 45±9 mm), and higher frequency of ULP (50% vs 10%) and PE (56% vs 29%) than did the event-free group. Multivariate analysis showed that ULP and SBP were significant predictors of the primary endpoint. The total risk score ≥2 could predict the primary endpoint with 87.5% sensitivity and 71.7% specificity (area under the receiver operating characteristic curve, 0.791). Conclusion The risk score was useful to predict cardiovascular death and the need for surgery in patients with acute type A IMH receiving medical therapy in the validation cohort study. ROC curve for the risk score Funding Acknowledgement Type of funding source: None

Download Full-text

Personalized Risk Prediction for Type 2 Diabetes: the Potential of Genetic Risk Scores

10.1101/041731 ◽

2016 ◽

Cited By ~ 1

Author(s):

Kristi Lall ◽

Reedik Magi ◽

Andrew Morris ◽

Andres Metspalu ◽

Krista Fischer

Keyword(s):

Type 2 Diabetes ◽

Risk Prediction ◽

Risk Score ◽

Cardiovascular Mortality ◽

Genetic Risk ◽

Association Studies ◽

Genetic Risk Score ◽

Genome Wide Association Studies ◽

Net Reclassification Improvement

Purpose: The study aims to develop a Genetic Risk Score (GRS) for the prediction of Type 2 Diabetes (T2D) that could be used for risk assessment in general population. Methods: Using the results of genome-wide association studies, we develop a doubly-weighted GRS for the prediction of T2D risk, aiming to capture the effect of 1000 single nucleotide polymorphisms. The GRS is evaluated in the Estonian Biobank cohort (n=10273), analysing its effect on prevalent and incident T2D, while adjusting for other predictors. We assessed the effect of GRS on all-cause and cardiovascular mortality and its association with other T2D risk factors, and conducted the reclassification analysis. Results: The adjusted hazard for incident T2D is 1.90 (95% CI 1.48, 2.44) times higher and for cardiovascular mortality 1.27 (95% CI 1.10, 1.46) times higher in the highest GRS quintile compared to the rest of the cohort. No significant association between BMI and GRS is found in T2D-free individuals. Adding GRS to the prediction model for 5-year T2D risks results in continuous Net Reclassification Improvement of 0.26 (95% CI 0.15, 0.38). Conclusion: The proposed GRS would considerably improve the accuracy of T2D risk prediction when added to the set of predictors used so far. Keywords: genetic risk score, Type 2 Diabetes, risk prediction, genetic risk, precision medicine

Download Full-text