Development and Internal Validation of a Clinical Prediction Model for Refeeding Syndrome in Adult Intensive Care...

Refeeding syndrome (RFS) is a potentially life-threatening complication during nutritional rehabilitation in malnourished patients, especially those in intensive care units (ICUs). This study aimed to develop and internally validate a clinical prediction model for assessing the risk of RFS in adult ICU patients.

This retrospective observational study was conducted at Beijing Jishuitan Hospital from January 2022 to November 2023. Adult ICU patients at high risk for RFS, identified by nutritional assessment, were included. RFS was defined as a >10% decrease in serum phosphorus, potassium, or magnesium within 5 days after refeeding. Demographic, clinical, and biochemical data were collected from electronic medical records. For the biochemical data, the baseline (the day before refeeding), peak (highest measurements during the first five days after refeeding), and latest value (the fifth day after refeeding) value were analyzed. Univariable and multivariable logistic regression, with stepwise selection, identified independent predictors. Model performance was evaluated by receiver operating characteristic (ROC) curve analysis (area under the curve, AUC), calibration plots, and decision curve analysis (DCA), with internal validation performed using bootstrap resampling.

A total of 132 ICU patients were included (RFS group, n=86; non-RFS group, n=46). Baseline characteristics, illness severity scores, and comorbidities, were generally comparable between groups. Multivariable analysis showed that higher peak urine epithelial cell count (OR=1.145, 95% CI: 1.023-1.282), lower baseline total bilirubin (OR=0.969, 95% CI: 0.940-1.000), lower peak potassium (OR=0.383, 95% CI: 0.147-0.995), and lower latest relative lymphocyte count (OR=0.946, 95% CI: 0.897-0.997) were independently associated with RFS risk. The model demonstrated good discrimination (AUC=0.78, 95% CI: 0.69-0.87) and calibration. DCA indicated clinical utility across a range of risk thresholds.

This internally validated model accurately predicts RFS risk in high-risk adult ICU patients, potentially improving early identification and individualized nutritional management. Further external validation is needed before wider clinical application.