Predictive scoring systems support clinicians in decision-making by estimating the prognosis of patients in intensive care units (ICUs). However, there is limited evidence on the accuracy of these systems in predicting mortality and organ dysfunction in special populations. The aim of this review is to assess the performance of predictive scoring systems in forecasting mortality in adult ICU patients in relation to baseline kidney function. It is anticipated that the assessment of predictive scoring systems’ performance and patient outcomes in this review may reveal information that will contribute to improve the quality of care and outcomes for special or under-represented ICU patient populations. It might also inform future research and contribute to the development of novel risk prediction models to address identified gaps or unanswered questions.

This review will include only observational studies, as these allow us to assess the real-world performance of predictive scoring systems in ICU settings by examining the original validation studies. By excluding randomised trials, paediatric studies, case reports and machine learning-derived models, this review focuses on the direct practical use of the scoring systems in adult ICU patients. A comprehensive search of MEDLINE, Embase and Scopus was conducted from database inception to 10 October 2024. The data will be extracted on study characteristics, patient outcomes and performance metrics.

This review will analyse data from previously published studies; no ethical approval is required. All data that will be included in the analysis will be publicly available and will be included in the final manuscript. Results will be disseminated through publication in a peer-reviewed journal and will also be presented at seminars and conferences.

CRD42024611547.