Length of hospital stay remains a key determinant of outcomes and resource utilization in diabetic ketoacidosis. A retrospective analysis published in Clinical Diabetology evaluated whether the systemic immune-inflammation index can predict hospitalization duration in adults admitted with DKA in the absence of infection.

The study examined patients aged 18 years and older who were admitted with DKA between December 2023 and May 2025. Of 98 consecutively screened emergency department admissions, 37 patients were excluded, resulting in a final analytic cohort of 61 patients. The mean age was 36 years, and 57.4% were male. Associations between LOS and inflammatory markers were assessed using correlation analysis, multiple linear regression, and receiver operating characteristic analysis.

SII showed a strong positive correlation with LOS (r = 0.600; p = 0.00). NLR and PLR were also significantly correlated with LOS, although with weaker associations (r = 0.539 and r = 0.403, respectively). In multivariable regression, SII emerged as the only independent predictor of LOS (p = 0.003). In predictive analysis, SII demonstrated the highest accuracy for identifying hospitalizations longer than 3 days, with an area under the curve of 0.884 (95% confidence interval 0.794 to 0.973).

These findings indicate that SII may serve as a practical biomarker for anticipating prolonged hospitalization in DKA. Early identification of patients with elevated inflammatory burden may support timely risk stratification and resource planning, pending confirmation in larger cohorts.