Risk prediction models for refeeding syndrome in critically ill adults: A systematic review

The incidence of refeeding syndrome (RFS) in critically ill patients has attracted increasing attention, accompanied by the proliferation of risk prediction models designed to identify susceptible individuals. Nevertheless, the validity and practicality of these models in clinical settings and for future research are not yet fully established.

The aim of this study was to systematically review published studies on risk prediction models for RFS in critically ill patients.

The following databases were searched from their inception to March 15, 2025: China National Knowledge Infrastructure, Wanfang database, China Science and Technology Journal database (VIP), SinoMed, PubMed, Web of Science, Cochrane Library, CINAHL, and Embase. The Prediction Model Risk of Bias Assessment Tool checklist was used to evaluate the risk of bias and applicability of the included studies.

A total of 1472 studies were retrieved. This review included 24 prediction models from 16 studies after the selection process. The reported area under the receiver operating characteristic curve ranged from 0.626 to 0.980. All studies have a high risk of bias, primarily due to the limited availability of data sources and insufficient reporting of analytical methods.

All studies included in the review on risk prediction models for RFS in critically ill patients were identified as having a high risk of bias. Future studies should prioritise the development of new models with larger sample sizes, rigorous study designs, and multicentre external validation to enhance the reliability and applicability.