An International Publisher for Academic and Scientific Journals
Author Login 
Scholars Journal of Applied Medical Sciences | Volume-13 | Issue-08
Investigation of Short-Term Safe Discharge Predictors in Patients Presenting to the Emergency Department with Hyperglycemia
Mustafa Furkan Uzun
Published: Aug. 30, 2025 | 23 13
Pages: 1636-1637
Downloads
Abstract
Objective: To identify reliable predictors for short-term safe discharge in hyperglycemic patients presenting to the emergency department (ED). Methods: A retrospective cohort study was conducted on 206 patients presenting with blood glucose ≥200 mg/dL to Kocaeli University Faculty of Medicine Hospital ED between September 2018 and March 2019. Demographics, laboratory parameters, comorbidities, and 72-hour ED revisit rates were analyzed. Multivariate logistic regression was used to determine predictors of one-week hospitalization. Model performance was evaluated using ROC curve analysis. Results: 21.4% of patients revisited the ED within 72 hours. Patients who returned had significantly higher initial blood glucose, BUN, and creatinine levels, and lower bicarbonate and pH. Non-diabetics were more likely to present with DKA or HHS. Multivariate analysis revealed that abnormal physical examination findings (OR: 4.4, 95% CI: 1.4-13.4) and elevated creatinine (OR: 3.5, 95% CI: 1.3-9.9) independently predicted one-week hospitalization. The AUC for the model was 0.76, indicating acceptable discriminative performance. Conclusion: Clinical examination and renal function testing are essential for predicting short-term risk in ED hyperglycemia cases. Implementation of these findings into discharge protocols may optimize resource utilization and patient outcomes. Further prospective validation is recommended.