O033 - FACTORS ASSOCIATED WITH REFEEDING SYNDROME IN CRITICALLY ILL PATIENTS FOLLOWING EMERGENCY ICU ADMISSION: A RETROSPECTIVE COHORT STUDY USING A LARGE-SCALE CLAIMS DATABASE

O033

FACTORS ASSOCIATED WITH REFEEDING SYNDROME IN CRITICALLY ILL PATIENTS FOLLOWING EMERGENCY ICU ADMISSION: A RETROSPECTIVE COHORT STUDY USING A LARGE-SCALE CLAIMS DATABASE

M. Yoshida^1,*, S. Shimizu², N. Kanda³, Y. Hashimoto¹, H. Tanoue⁴, H. Wakatake⁵, K. Nakamura⁶, G. Tadahiro²

¹Department of Clinical Research Management, National Hospital Organization Headquarters, Tokyo, ²Department of Health Data Science, Yokohama City University, Yokohama, ³Division of General Internal Medicine, Jichi Medical University, Tochigi, ⁴Department of Nutrition, St. Marianna University School of Medicine, Yokohama City Seibu Hospital, Yokohama, ⁵Department of Emergency and Critical Care Medicine, St. Marianna University School of Medicine, Kawasaki, ⁶Department of Disaster and Emergency Medicine, Graduate School of Medicine, Kobe University, Kobe, Japan

Rationale: Refeeding syndrome (RFS) is a frequent ICU metabolic complication, yet specific risk factors in critically ill patients remain unclear. This study aimed to identify clinical factors associated with RFS.

Methods: We conducted a retrospective cohort study using a Japanese administrative claims database integrated with laboratory data. We included adult critically ill patients emergently admitted between April 2018 and June 2025 with available serum phosphate measurements. The primary outcome was RFS, defined according to the ESPEN definition as a decrease in serum phosphate of>0.16 mmol/L from baseline or a serum phosphate level of <0.65 mmol/L within 3 days after nutrition initiation. A stricter secondary outcome required both criteria. After multiple imputation for missing data, multivariable logistic regression identified  factors associated with RFS.

Results: RFS incidence was 7,865 (66.5%) by ESPEN definition and 2,245 (19.0%) by the strict definition. In multivariable analysis in ESPEN definition, RFS was associated with female sex (odds ratio (OR) 1.17, 95% confidence interval (CI) 1.08–1.27), lower body mass index (OR 1.03 per 1 kg/m² decrease, 95% CI 1.02–1.04), emergency surgery (OR 1.42, 95% CI 1.15–1.76), higher severity score (OR 1.11 per 1-point increase, 95% CI 1.09–1.12), elevated C-reactive protein levels (OR 1.01 per 1 mg/dL increase, 95% CI 1.00–1.01), and lower serum albumin levels (OR 1.40 per 1 g/dL decrease, 95% CI 1.30–1.50). In contrast, nursing facility admission and higher Charlson Comorbidity Index were inversely associated with RFS (Figure). Findings were consistent using the strict definition.

Image:

Conclusion: In emergently admitted critically ill patients, catabolic indicators including severity, invasiveness, and inflammation, may serve as key RFS risk factors alongside conventional nutritional markers.

Disclosure of Interest: M. Yoshida Grant / Research Support from: No COI, but this study is supported by a designated research grant (FY2025) from the Clinical Trial Group Committee of the Japanese Society of Intensive Care Medicine and by the Japan Society for the Promotion of Science KAKENHI., S. Shimizu: None declared, N. Kanda: None declared, Y. Hashimoto: None declared, H. Tanoue: None declared, H. Wakatake: None declared, K. Nakamura: None declared, G. Tadahiro: None declared