scholarly journals Deep Learning Algorithm for Management of Diabetes Mellitus via Electrocardiogram-Based Glycated Hemoglobin (ECG-HbA1c): A Retrospective Cohort Study

2021 ◽  
Vol 11 (8) ◽  
pp. 725
Author(s):  
Chin-Sheng Lin ◽  
Yung-Tsai Lee ◽  
Wen-Hui Fang ◽  
Yu-Sheng Lou ◽  
Feng-Chih Kuo ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Deep Learning ◽  
Glycated Hemoglobin ◽  
Learning Algorithm ◽  
Cardiac Activity ◽  
Health Examination ◽  
Primary Analysis ◽  
Noninvasive Test ◽  
All Cause Mortality ◽  
New Onset

Background: glycated hemoglobin (HbA1c) provides information on diabetes mellitus (DM) management. Electrocardiography (ECG) is a noninvasive test of cardiac activity that has been determined to be related to DM and its complications. This study developed a deep learning model (DLM) to estimate HbA1c via ECG. Methods: there were 104,823 ECGs with corresponding HbA1c or fasting glucose which were utilized to train a DLM for calculating ECG-HbA1c. Next, 1539 cases from outpatient departments and health examination centers provided 2190 ECGs for initial validation, and another 3293 cases with their first ECGs were employed to analyze its contributions to DM management. The primary analysis was used to distinguish patients with and without mild to severe DM, and the secondary analysis was to explore the predictive value of ECG-HbA1c for future complications, which included all-cause mortality, new-onset chronic kidney disease (CKD), and new-onset heart failure (HF). Results: we used a gender/age-matching strategy to train a DLM to achieve the best AUCs of 0.8255 with a sensitivity of 71.9% and specificity of 77.7% in a follow-up cohort with correlation of 0.496 and mean absolute errors of 1.230. The stratified analysis shows that DM presented in patients with fewer comorbidities was significantly more likely to be detected by ECG-HbA1c. Patients with higher ECG-HbA1c under the same Lab-HbA1c exhibited worse physical conditions. Of interest, ECG-HbA1c may contribute to the mortality (gender/age adjusted hazard ratio (HR): 1.53, 95% conference interval (CI): 1.08–2.17), new-onset CKD (HR: 1.56, 95% CI: 1.30–1.87), and new-onset HF (HR: 1.51, 95% CI: 1.13–2.01) independently of Lab-HbA1c. An additional impact of ECG-HbA1c on the risk of all-cause mortality (C-index: 0.831 to 0.835, p < 0.05), new-onset CKD (C-index: 0.735 to 0.745, p < 0.01), and new-onset HF (C-index: 0.793 to 0.796, p < 0.05) were observed in full adjustment models. Conclusion: the ECG-HbA1c could be considered as a novel biomarker for screening DM and predicting the progression of DM and its complications.

scholarly journals Comparative effects of sodium glucose cotransporter 2 (SGLT2) inhibitors and dipeptidyl peptidase-4 (DPP4) inhibitors on new-onset atrial fibrillation and stroke outcomes

2021 ◽  
Author(s):  
Sharen Lee ◽  
Jiandong Zhou ◽  
Carlin Chang ◽  
Tong Liu ◽  
Dong Chang ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Atrial Fibrillation ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Propensity Score ◽  
Lower Risk ◽  
Sglt2 Inhibitors ◽  
All Cause Mortality ◽  
New Onset

AbstractBackgroundSGLT2I and DPP4I are medications prescribed for type 2 diabetes mellitus patients. However, there are few population-based studies comparing their effects on incident atrial fibrillation or ischemic stroke.MethodsThis was a territory-wide cohort study of type 2 diabetes mellitus patients prescribed SGLT2I or DPP4I between January 1st, 2015 to December 31st, 2019 in Hong Kong. Patients with both DPP4I and SGLT2I use and patients with drug discontinuation were excluded. Patients with prior AF or stroke were excluded for the respective analysis. 1:2 propensity-score matching was conducted for demographics, past comorbidities and medications using nearest-neighbor matching method. Cox models were used to identify significant predictors for new onset heart failure (HF) or myocardial infarction (MI), cardiovascular and all-cause mortality.ResultsThe AF-free cohort included 49108 patients (mean age: 66.48 years old [SD: 12.89], 55.32% males) and the stroke-free cohort included 49563 patients (27244 males [54.96%], mean baseline age: 66.7 years old [SD: 12.97, max: 104.6 years old]). After propensity score matching, SGLT2i use was associated with a lower risk of new onset AF (HR: 0.43[0.28, 0.66]), cardiovascular mortality (HR: 0.79[0.58, 1.09]) and all-cause mortality (HR: 0.69[0.60, 0.79]) in the AF-free cohort. It was also associated with a lower risk of new onset stroke (0.46[0.33, 0.64]), cardiovascular mortality (HR: 0.74[0.55, 1.00]) and all-cause mortality (HR: 0.64[0.56, 0.74]) in the stroke-free cohort.ConclusionsThe novelty of our work si that SGLT2 inhibitors are protective against atrial fibrillation and stroke development for the first time. These findings should be validated in other cohorts.

scholarly journals Cox Proportional Hazard Regression Versus a Deep Learning Algorithm in the Prediction of Dementia: An Analysis Based on Periodic Health Examination (Preprint)

2018 ◽  
Author(s):  
Woo Jung Kim ◽  
Ji Min Sung ◽  
David Sung ◽  
Myeong-Hun Chae ◽  
Suk Kyoon An ◽  
...  
Keyword(s):  
Time Series ◽  
Deep Learning ◽  
Proportional Hazards ◽  
Proportional Hazards Model ◽  
Learning Algorithm ◽  
Repeated Measurements ◽  
Cox Proportional Hazards Model ◽  
Health Examination ◽  
Alzheimer Dementia ◽  
Deep Learning Algorithm

BACKGROUND With the increase in the world’s aging population, there is a growing need to prevent and predict dementia among the general population. The availability of national time-series health examination data in South Korea provides an opportunity to use deep learning algorithm, an artificial intelligence technology, to expedite the analysis of mass and sequential data. OBJECTIVE This study aimed to compare the discriminative accuracy between a time-series deep learning algorithm and conventional statistical methods to predict all-cause dementia and Alzheimer dementia using periodic health examination data. METHODS Diagnostic codes in medical claims data from a South Korean national health examination cohort were used to identify individuals who developed dementia or Alzheimer dementia over a 10-year period. As a result, 479,845 and 465,081 individuals, who were aged 40 to 79 years and without all-cause dementia and Alzheimer dementia, respectively, were identified at baseline. The performance of the following 3 models was compared with predictions of which individuals would develop either type of dementia: Cox proportional hazards model using only baseline data (HR-B), Cox proportional hazards model using repeated measurements (HR-R), and deep learning model using repeated measurements (DL-R). RESULTS The discrimination indices (95% CI) for the HR-B, HR-R, and DL-R models to predict all-cause dementia were 0.84 (0.83-0.85), 0.87 (0.86-0.88), and 0.90 (0.90-0.90), respectively, and those to predict Alzheimer dementia were 0.87 (0.86-0.88), 0.90 (0.88-0.91), and 0.91 (0.91-0.91), respectively. The DL-R model showed the best performance, followed by the HR-R model, in predicting both types of dementia. The DL-R model was superior to the HR-R model in all validation groups tested. CONCLUSIONS A deep learning algorithm using time-series data can be an accurate and cost-effective method to predict dementia. A combination of deep learning and proportional hazards models might help to enhance prevention strategies for dementia.

scholarly journals Association of Pulmonary Function Decline over Time with Longitudinal Change of Glycated Hemoglobin in Participants without Diabetes Mellitus

2021 ◽  
Vol 11 (10) ◽  
pp. 994
Author(s):  
Wen-Hsien Lee ◽  
Da-Wei Wu ◽  
Ying-Chih Chen ◽  
Yi-Hsueh Liu ◽  
Wei-Sheng Liao ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Lung Function ◽  
Pulmonary Function ◽  
Glycated Hemoglobin ◽  
Longitudinal Change ◽  
Type 2 Dm ◽  
Over Time ◽  
New Onset

Pulmonary damage and function impairment were frequently noted in patients with diabetes mellitus (DM). However, the relationship between lung function and glycemic status in non-DM subjects was not well-known. Here, we evaluated the association of longitudinal changes of lung function parameters with longitudinal changes of glycated hemoglobin (HbA1c) in non-DM participants. The study enrolled participants without prior type 2 DM, hypertension, and chronic obstructive pulmonary disease (COPD) from the Taiwan Biobank database. Laboratory profiles and pulmonary function parameters, including forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1), were examined at baseline and follow-up. Finally, 7055 participants were selected in this study. During a mean 3.9-year follow-up, FVC and FEV1 were significantly decreased over time (both p < 0.001). In the multivariable analysis, the baseline (unstandardized coefficient β = −0.032, p < 0.001) and longitudinal change (unstandardized coefficient β = −0.025, p = 0.026) of FVC were negatively associated with the baseline and longitudinal change of HbA1c, respectively. Additionally, the longitudinal change of FVC was negatively associated with the risk of newly diagnosed type 2 DM (p = 0.018). During a mean 3.9-year follow-up, our present study, including participants without type 2 DM, hypertension, and COPD, demonstrated that the baseline and longitudinal change of FVC were negatively and respectively correlated with the baseline and longitudinal change of HbA1c. Furthermore, compared to those without new-onset DM, participants with new-onset DM had a more pronounced decline of FVC over time.

scholarly journals Development of a predictive risk model for all-cause mortality in patients with diabetes in Hong Kong

2021 ◽  
Vol 9 (1) ◽  
pp. e001950
Author(s):  
Sharen Lee ◽  
Jiandong Zhou ◽  
Keith Sai Kit Leung ◽  
William Ka Kei Wu ◽  
Wing Tak Wong ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Deep Learning ◽  
Risk Model ◽  
All Cause Mortality ◽  
Patients With Diabetes ◽  
Predictive Risk

IntroductionPatients with diabetes mellitus are risk of premature death. In this study, we developed a machine learning-driven predictive risk model for all-cause mortality among patients with type 2 diabetes mellitus using multiparametric approach with data from different domains.Research design and methodsThis study used territory-wide data of patients with type 2 diabetes attending public hospitals or their associated ambulatory/outpatient facilities in Hong Kong between January 1, 2009 and December 31, 2009. The primary outcome is all-cause mortality. The association of risk variables and all-cause mortality was assessed using Cox proportional hazards models. Machine and deep learning approaches were used to improve overall survival prediction and were evaluated with fivefold cross validation method.ResultsA total of 273 678 patients (mean age: 65.4±12.7 years, male: 48.2%, median follow-up: 142 (IQR=106–142) months) were included, with 91 155 deaths occurring on follow-up (33.3%; annualized mortality rate: 3.4%/year; 2.7 million patient-years). Multivariate Cox regression found the following significant predictors of all-cause mortality: age, male gender, baseline comorbidities, anemia, mean values of neutrophil-to-lymphocyte ratio, high-density lipoprotein-cholesterol, total cholesterol, triglyceride, HbA1c and fasting blood glucose (FBG), measures of variability of both HbA1c and FBG. The above parameters were incorporated into a score-based predictive risk model that had a c-statistic of 0.73 (95% CI 0.66 to 0.77), which was improved to 0.86 (0.81 to 0.90) and 0.87 (0.84 to 0.91) using random survival forests and deep survival learning models, respectively.ConclusionsA multiparametric model incorporating variables from different domains predicted all-cause mortality accurately in type 2 diabetes mellitus. The predictive and modeling capabilities of machine/deep learning survival analysis achieved more accurate predictions.

Abstract 15098: New-Onset Diabetes Mellitus Risk in Statin Users Depends on Baseline HemoglobinA1c Value - Results of VA Database Analysis

Circulation ◽  
2020 ◽  
Vol 142 (Suppl_3) ◽  
Author(s):  
Anna P Ziganshina ◽  
Darren Gemoets ◽  
Laurence Kaminsky ◽  
Aidar R Gosmanov
Keyword(s):  
Diabetes Mellitus ◽  
Mortality Rate ◽  
Statin Therapy ◽  
Baseline Hba1c ◽  
Onset Diabetes ◽  
All Cause Mortality ◽  
Us Veterans ◽  
New Onset Diabetes Mellitus ◽  
Statin Use ◽  
New Onset

Introduction: Statin use in non-diabetic patients is associated with an increased risk of development of new-onset diabetes mellitus (NODM). However, little is known if baseline hemoglobin A1c (HbA1c) can affect diabetogenic risks and mortality benefits of statin therapy. Methods: In a retrospective cohort study between 01/2011 and 12/2018 we collected the data of 152358 non-diabetic US veterans on statin and not on statin therapy who had available baseline HbA1c value and full demographic and clinical information prior to DM diagnosis. The risk of statin-induced DM and all-cause mortality rate were assessed in the whole cohort and based on baseline HbA1c categories: ≤5.6%, 5.7-5.9% and 6.0-6.4%. DM rate and mortality rate were calculated by Cox proportional hazards model adjusted for case-mix. Results: After mean follow up of 6.89 (SD 2.26) years in non-statin users and 3.85 (SD 2.29) years in statin users and adjustments for multiple covariates such as age, gender, ethnicity, obesity, hypertension (HTN), cardiovascular disease, and metformin use, we found that the rate of statin-induced DM depends on baseline HbA1c (Table 1). Analysis by HbA1c categories showed that NODM rate is inversely related to HbA1c level, while statin use in patients with HbA1c 6.0-6.4% was not associated with increased rate of NODM. We did not observe increase in all-cause mortality in statin users and statin users with HTN across HbA1c categories. Atorvastatin use was associated with decrease in all-cause mortality in HTN patients (Table 1). Conclusions: The results of this largest to date analysis of non-diabetic US veterans closely matched for baseline characteristics suggest that the rate of statin-induced DM depends on baseline HbA1c. The significant increase in NODM rate is only observed in patients with baseline HbA1c <6.0% regardless of statin type prescribed. The increased NODM risk is not associated with all-cause mortality in statin users in general including patients with HTN.

Deep Learning-Based Diabetic Retinopathy Detection

2021 ◽  
Vol 11 (3) ◽  
pp. 38-48
Author(s):  
Mohamed Jebran P. ◽  
Sufia Banu
Keyword(s):  
Diabetes Mellitus ◽  
Diabetic Retinopathy ◽  
Deep Learning ◽  
Learning Algorithm ◽  
Classification Performance ◽  
Learning Problems ◽  
Deep Learning Algorithm ◽  
Lesion Level ◽  
Fundus Photographs

Artificial intelligence (AI) is rapidly evolving from machine learning (ML) to deep learning (DL), which has ignited particular interest in ophthalmology as well. Deep learning has been applied in ophthalmology to fundus photographs, which achieve robust classification performance in the detection of diabetic retinopathy (DR). Diabetic retinopathy is a progressive condition observed in people who have had multiple years of diabetes mellitus. This paper focuses on examining how a deep learning algorithm can be applied for the detection and classification of diabetic retinopathy, both at the image level and at the lesion level. The performance of various neural networks is summarized by taking into account the sensitivity, precision, accuracy with respect to the size of the test datasets. Deep learning problems are discussed at the end.

scholarly journals Incident heart failure and myocardial infarction in sodium-glucose cotransporter 2 versus dipeptidyl peptidase-4 inhibitor users

2021 ◽  
Author(s):  
Jiandong Zhou ◽  
Sharen Lee ◽  
Keith Sai Kit Leung ◽  
Abraham Ka Chung Wai ◽  
Tong Liu ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Myocardial Infarction ◽  
Heart Failure ◽  
Population Based ◽  
Dipeptidyl Peptidase 4 ◽  
Sodium Glucose Cotransporter ◽  
All Cause Mortality ◽  
Incident Heart Failure ◽  
New Onset

Objectives: To compare the rates of major cardiovascular adverse events in sodium-glucose cotransporter-2 inhibitors (SGLT2I) and dipeptidyl-peptidase-4 inhibitors (DPP4I) users in a Chinese population. Background: SGLT2I and DPP4I are increasingly prescribed for type 2 diabetes mellitus patients. However, few population-based studies are comparing their effects on incident heart failure or myocardial infarction. Methods: This was a population-based retrospective cohort study using the electronic health record database in Hong Kong, including type 2 diabetes mellitus patients receiving either SGLT2I or DPP4I between January 1st, 2015, to December 31st, 2020. Propensity-score matching was performed in a 1:1 ratio based on demographics, past comorbidities, non-SGLT2I/DPP4I medications with nearest-neighbor matching (caliper=0.1). Univariate and multivariate Cox models were used to identify significant predictors for new onset heart failure, new onset myocardial infarction, cardiovascular mortality, and all-cause mortality. Sensitivity analyses with competing risk models and multiple propensity score matching approaches were conducted. Subgroup age and gender analyses were presented. Results: A total of 41994 patients (58.89% males, median admission age at 58 years old, interquartile rage [IQR]: 51.2-65.3) were included in the study cohorts with a median follow-up duration of 5.6 years (IQR: 5.32-5.82). After adjusting for significant demographics, past comorbidities, medication prescriptions and biochemical results, SGLT2I users have a significantly lower risk for myocardial infarction (hazard ratio [HR]: 0.34, 95% confidence interval [CI]: [0.28, 0.41], P < 0.0001), cardiovascular mortality (HR: 0.53, 95% CI: [0.38, 0.74], P = 0.0002) and all-cause mortality (HR: 0.21, 95% CI: [0.18, 0.25], P= 0.0001) under multivariate Cox regression. However, the risk for heart failure is comparable (HR: 0.87, 95% CI: [0.73, 1.04], P= 0.1343). Conclusions: SGLT2 inhibitors are protective against adverse cardiovascular events compared to DPP4I. The prescription of SGLT2I is preferred especially for males and patients aged 65 or older to prevent cardiovascular risks.

scholarly journals The Impact of New-Onset Diabetes Mellitus and Hypertension on All-Cause Mortality in an Apparently Healthy Population: A Ten-Year Follow-Up Study

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Suranut Charoensri ◽  
Kittrawee Kritmetapak ◽  
Tassanapong Tangpattanasiri ◽  
Chatlert Pongchaiyakul
Keyword(s):  
Diabetes Mellitus ◽  
Tertiary Care ◽  
Healthy Population ◽  
Risk Of Death ◽  
Onset Diabetes ◽  
All Cause Mortality ◽  
The Impact ◽  
New Onset Diabetes Mellitus ◽  
New Onset ◽  
Apparently Healthy

Introduction. The comparative effect of new-onset diabetes mellitus (DM) and hypertension (HT) on long-term mortality is a matter of debate. Materials and Methods. From 2007 to 2017, a 10-year longitudinal retrospective cohort study was conducted in Thailand’s tertiary care setting. As baseline data, health check-up data from apparently healthy participants without underlying disease from 2007 were extracted. The vital status of all participants was determined in 2017, ten years after an initial examination. The impact of new-onset DM and HT at baseline on 10-year all-cause mortality was investigated using multivariable logistic regression analysis. Results. The prevalence of new-onset DM and HT was 6.4% and 28.8%, respectively, at baseline. Newly diagnosed diabetes increased the risk of all-cause mortality over 10 years (adjusted OR 4.77 and 95% CI 2.23-9.99). HT, on the other hand, did not increase the risk of death (adjusted OR 1.24 and 95% CI 0.65-2.35). Different HT and DM status combinations were compared to a nondiabetic, nonhypertensive reference. Individuals who were diabetic and hypertensive had a greater risk of death (adjusted OR 6.22 and 95% CI 2.22-17.00). Having DM without HT also increased the risk of death (adjusted OR 4.36 and 95% CI 1.35-12.87). However, having HT without DM did not result in a significant increase in 10-year mortality risk (adjusted OR 1.21 and 95% CI 0.57-2.56). Conclusion. In an apparently healthy population, new-onset DM is more strongly associated with 10-year all-cause mortality than new-onset HT. Having both DM and HT was associated with a greater risk of death when compared to having DM or HT alone.

The association between SGLT2 inhibitors and new-onset arrhythmias: a nationwide population-based longitudinal cohort study

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
W.-Z Siao ◽  
G.P Jong
Keyword(s):  
Diabetes Mellitus ◽  
Atrial Fibrillation ◽  
Cohort Study ◽  
Lower Risk ◽  
Sglt2 Inhibitor ◽  
Population Based ◽  
Sglt2 Inhibitors ◽  
Funding Source ◽  
All Cause Mortality ◽  
New Onset

Abstract Background/Introduction Clinical trials have shown the cardiovascular protective effect of sodium-glucose cotransporter-2 (SGLT2) inhibitors and reduced hospitalization for heart failure. However, no study has investigated the association between SGLT2 inhibitors and the risk of arrhythmias. Purpose To evaluate the risk of new-onset arrhythmias (NOA) and all-cause mortality with the use of SGLT2 inhibitors. Methods This was a population-based cohort study utilizing Taiwan's National Health Insurance Research Database. Each patient aged 20 years and older who took SGLT2 inhibitors was assigned to the SGLT2 inhibitor group, whereas sex-, age-, diabetes mellitus duration-, drug index date-, and propensity score-matched randomly selected patients without SGLT2 inhibitors were assigned to the non-SGLT2 inhibitor group. The study outcome was all-cause mortality and NOA. Results A total of 399,810 patients newly diagnosed with type 2 diabetes mellitus (T2DM) were enrolled. A 1:1 matching propensity method was used to match 79,150 patients to 79,150 controls in the non-SGLT2 inhibitors group for analysis. The SGLT2 inhibitor group was associated with a lower risk of all-cause mortality (adjusted hazard ratio [aHR] 0.547; 95% confidence interval [CI] 0.482–0.621; P=0.0001) and NOA (aHR 0.830; 95% CI 0.751–0.916; P=0.0002). Subgroup analysis revealed that the SGLT2 inhibitor group was associated with a lower risk of all-cause mortality in all age and severity of the adapted Diabetes Complication Severity Index (aDCSI) subgroups. Furthermore, NOA and atrial fibrillation were associated with a lower risk in subgroups with an aDCSI score of between 1 and 2. Conclusions Patients with T2DM prescribed with SGLT2 inhibitors were associated with a lower risk of all-cause mortality and NOA compared with those not taking SGLT2 inhibitors in real-world practice. The potential protective effect of NOA and atrial fibrillation is observed, especially in individuals with T2DM and an aDCSI score of between 1 and 2. Funding Acknowledgement Type of funding source: Private hospital(s). Main funding source(s): This study was supported in parts by grants from Chung Shan Medical University Hospital.

scholarly journals Alkaline phosphatase variability predicts new onset heart failure, cardiovascular mortality, and all-cause mortality in patients with type-2 diabetes mellitus: a population-based study

2021 ◽  
Author(s):  
Jiandong Zhou ◽  
Sharen Lee ◽  
Govinda Adhikari ◽  
Wing Tak Wong ◽  
Khalid Bin Waleed ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Heart Failure ◽  
Type 2 Diabetes ◽  
Alkaline Phosphatase ◽  
Type 2 Diabetes Mellitus ◽  
Cardiovascular Mortality ◽  
Study Cohort ◽  
All Cause Mortality ◽  
New Onset

Objective: To investigate the associations of alkaline phosphatase (ALP) variability measures with new onset heart failure, cardiovascular mortality, and all-cause mortality in type 2 diabetes mellitus patients with a populational-cohort study. Method: This study included patients with type 2 diabetes mellitus who presented to ambulatory, outpatient and inpatient facilities managed by the public sector in Hong Kong between January 1st, 2000 to December 31st, 2019. Comprehensive clinical and medical data including demographics, past comorbidities, medications, and laboratory examinations of complete blood, lipid/glycemic profile and their variability were collected. ALP and its variability measures were extracted. Univariable and multiple multivariable Cox regression were used to identify the associations of alkaline phosphatase variability with new onset heart failure and mortality risks. Patients were stratified into three subgroups based on the tertiles of baseline ALP level. Results: The study cohort consisted of 14289 patients (52.52% males, mean age at initial drug exposure: 74.55 years old [standard deviation (SD): 12.7]). Over a mean follow up of 2513 days [interquartile range (IQR): 1151-4173]), 10182 patients suffered from all-cause mortality (incidence rate [IR]: 71.25%), 1966 patients (IR: 13.75%) died from cardiovascular causes, and 1171 patients (IR: 8.19%) developed with new onset heart failure. Higher cumulative incidences of all three outcomes were observed for the highest tertile of ALP compared to medium/low tertiles. ALP baseline and variability level predicted new onset heart failure, cardiovascular and all-cause mortality before adjusting for subclinical biomarkers (p < 0.01). Amongst the measures of ALP variability, the hazard ratio (HR) of coefficient of variation (CV) was markedly raised in particular (new onset heart failure: HR=2.73, 95% confidence interval [CI]= [1.71-4.37], p <0.0001; all-cause mortality: HR= 5.83, 95% CI= [5.01-6.79], p <0.0001; cardiovascular mortality: HR= 4.81, 95% CI= [3.36-6.88], p <0.0001). Conclusions: Raised ALP level and variability are associated with increased risks of all-cause mortality, cardiovascular mortality and new onset heart failure amongst patients with type 2 diabetes mellitus.

Sign in / Sign up

Export Citation Format

Share Document