2280-P: Diabetic Ketoacidosis: A Systematic Review of Precipitants, Treatment, and Outcomes with a Mortality Meta-Analysis

Introduction and Objective: Diabetic ketoacidosis (DKA) is one of the most common emergencies in diabetes, yet systematic global data on presentation, management, and outcomes remain limited. This systematic review aimed to synthesise international evidence on precipitants, biochemical profiles, treatment, and outcomes of adult DKA. Methods: Seven databases were searched from inception to December 2024 for studies involving adults (≥16 years) with DKA. Data were extracted on study location, precipitants, biochemical parameters, management protocols, guideline use, and outcomes. Risk of bias was assessed using appropriate validated tools. Descriptive statistics were calculated to analyse global trends, with a random-effects meta-analysis for in-hospital mortality risk. Results: Of 21,689 records screened, 18 studies were included, encompassing 1,344 DKA episodes (mean age: 42.1 years). Heterogeneity was reflected through the use of five different guidelines globally, with 38.9% of studies not stating the guideline used. The most common precipitants of DKA episodes were inadequate diabetes medication (40.9%) and infection (31.1%). Thirteen studies reported DKA resolution times, with a mean of 18.4 hours (95% CI: 13.9-22.9 hours), respectively. Management practices, admission biochemical profiles, and guideline use also differed between continents. Meta-analysis of eight studies showed an in-hospital mortality rate summary estimate of 2.64% (95% CI: 1.22%-5.61%) with a prediction interval of 0.61-10.7%, though there was high heterogeneity. Conclusion: This systematic review highlights international variation in DKA precipitants, management, and outcomes, with low but heterogeneous in-hospital mortality. This necessitates an international consensus on DKA guidelines to reduce variation and potentially improve outcomes. Disclosure L. Bomphrey: None. J. French: None. K. Malhotra: None. A. Manta: None. K. Scandrett: None. P. Kempegowda: None.