Trends in fluid overload-related hospitalisations among patients with diabetes mellitus: The impact of chronic kidney disease

Tan, Joshua Kuan; Kadir, Hanis Abdul; Lim, Gek Hsiang; Thumboo, Julian; Bee, Yong Mong; Lim, Cynthia Ciwei

doi:10.47102/annals-acadmedsg.2024136

Cited by 3 publications

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Recurrent Hospitalizations for Fluid Overload in Diabetes with Kidney Failure Treated with Dialysis

Phang,

Ng,

Kadir

et al. 2024

Cardiorenal Med

View full text Add to dashboard Cite

Background & Aims Diabetes Mellitus is the most common cause of end-stage kidney disease (ESKD) in Singapore. ESKD patients have high disease burden and are at increased risk of recurrent hospitalizations, including fluid overload. This study aimed to characterize the risk factors associated with readmissions for fluid overload that will identify high-risk hospitalizations for interventions to reduce readmissions. Methods Retrospective cohort study of all hospitalizations for fluid overload in adults with diabetes and ESKD on dialysis in SingHealth hospitals between 2018 and 2021. Fluid overload was defined by discharge codes for fluid overload, heart failure, pulmonary edema, and generalized edema until 30th December 2022. Multivariable Cox regression analysis using the Prentice, Williams and Peterson Total Time (PWP-TT) model was performed for the outcomes of readmissions for fluid overload within 30 days and 90 days of discharge. Results Among 3234 hospitalizations for fluid overload, readmission for fluid overload within 30-day and 90-day occurred in 585 (18.1%) and 967 (29.9%) hospitalizations, respectively. Ischemic heart disease, peripheral vascular disease and lower hemoglobin level were independently associated with readmissions for fluid overload within 30 and 90 days. Additionally, heart failure, hemodialysis (compared to peritoneal dialysis) and lack of statin at discharge were associated with increased 90-day readmission risk. Conclusion Modifiable (hemoglobin level, statin use) and non-modifiable factors (ischemic heart disease, peripheral vascular disease and heart failure) influenced the risk of readmission for fluid overload. These results may guide risk stratification and inform targeted interventions to reduce avoidable, unplanned readmissions for recurrent fluid overload among individuals with diabetes and ESKD.

show abstract

Recurrent Hospitalizations for Fluid Overload in Diabetes with Kidney Failure Treated with Dialysis

Phang,

Ng,

Kadir

et al. 2024

Cardiorenal Med

View full text Add to dashboard Cite

show abstract

Comprehensive bioinformatics analysis identifies biomarkers for cardiovascular risk in end-stage renal diseases

2024

Preprint

View full text Add to dashboard Cite

Objective End-stage renal disease (ESRD) can increase the risk of cardiovascular disease (CV). We aimed to investigate the pathways and mechanisms associated with potential protective genes linked to CV (CVP). Methods We conducted a systematic bioinformatics analysis using publicly available datasets from the Gene Expression Omnibus (GEO). Differentially expressed genes (DEGs) were identified in patients with ESRD with and without arrhythmia using stringent statistical criteria. Functional enrichment analyses were performed using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways to elucidate the biological roles of these DEGs. Receiver Operating Characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of the identified biomarkers for CV risk prediction. Results Our analysis revealed a distinct set of DEGs in ESRD patients with arrhythmia compared to those without arrhythmia. GO and KEGG pathway analyses indicated that these DEGs were involved in key biological processes and pathways relevant to cardiovascular disorders and renal function, including wound healing, platelet activation, and fluid-level regulation. Moreover, this study identified four downregulated genes (ABLIM3, TREML1, VCL, and AVPR1A) and two upregulated genes (BHLHA15 and FZD8), which exhibited significant alterations in expression levels, with some showing robust discriminatory power, as evidenced by high Area Under the Curve (AUC) values in ROC analysis for predicting patients without CV risks. Conclusion This study identified a panel of genes (including a miRNA and an unknown gene) in the plasma that may serve as promising biomarkers for predicting arrhythmia risk in ESRD patients. These findings provide a foundation for future validation studies aimed at integrating plasma biomarkers into clinical practice to improve risk stratification and management of CV in patients with ESRD.

show abstract

Machine Learning–Based Prediction for High Health Care Utilizers by Using a Multi-Institutional Diabetes Registry: Model Training and Evaluation

Tan,

Quan,

Salim

et al. 2024

JMIR AI

View full text Add to dashboard Cite

Background The cost of health care in many countries is increasing rapidly. There is a growing interest in using machine learning for predicting high health care utilizers for population health initiatives. Previous studies have focused on individuals who contribute to the highest financial burden. However, this group is small and represents a limited opportunity for long-term cost reduction. Objective We developed a collection of models that predict future health care utilization at various thresholds. Methods We utilized data from a multi-institutional diabetes database from the year 2019 to develop binary classification models. These models predict health care utilization in the subsequent year across 6 different outcomes: patients having a length of stay of ≥7, ≥14, and ≥30 days and emergency department attendance of ≥3, ≥5, and ≥10 visits. To address class imbalance, random and synthetic minority oversampling techniques were employed. The models were then applied to unseen data from 2020 and 2021 to predict health care utilization in the following year. A portfolio of performance metrics, with priority on area under the receiver operating characteristic curve, sensitivity, and positive predictive value, was used for comparison. Explainability analyses were conducted on the best performing models. Results When trained with random oversampling, 4 models, that is, logistic regression, multivariate adaptive regression splines, boosted trees, and multilayer perceptron consistently achieved high area under the receiver operating characteristic curve (>0.80) and sensitivity (>0.60) across training-validation and test data sets. Correcting for class imbalance proved critical for model performance. Important predictors for all outcomes included age, number of emergency department visits in the present year, chronic kidney disease stage, inpatient bed days in the present year, and mean hemoglobin A1c levels. Explainability analyses using partial dependence plots demonstrated that for the best performing models, the learned patterns were consistent with real-world knowledge, thereby supporting the validity of the models. Conclusions We successfully developed machine learning models capable of predicting high service level utilization with strong performance and valid explainability. These models can be integrated into wider diabetes-related population health initiatives.

show abstract

Trends in fluid overload-related hospitalisations among patients with diabetes mellitus: The impact of chronic kidney disease

Cited by 3 publications

References 24 publications

Recurrent Hospitalizations for Fluid Overload in Diabetes with Kidney Failure Treated with Dialysis

Recurrent Hospitalizations for Fluid Overload in Diabetes with Kidney Failure Treated with Dialysis

Comprehensive bioinformatics analysis identifies biomarkers for cardiovascular risk in end-stage renal diseases

Machine Learning–Based Prediction for High Health Care Utilizers by Using a Multi-Institutional Diabetes Registry: Model Training and Evaluation

Contact Info

Product

Resources

About