High-protein supplementation in critically ill patients: a systematic review, meta-analysis and umbrella review of existing evidence

Background Protein serves as a core component of nutritional support for critically ill patients; however, no unified consensus or clinical standard has been established for protein supplementation to date. This gap highlights the urgent need for an updated systematic review and meta-analysis. Methods Randomized controlled trials (RCTs) and meta-analyses that met our criteria were included in PubMed, Web of Science, Cochrane Library and Embase English databases up to April 1, 2026. We used meta-analysis, subgroup analysis and umbrella review methods to explore the effects of high-protein and conventional protein supplementation on critically ill patients. We also conducted Trial Sequential Analysis (TSA) to verify the effectiveness of the evidence. The assessment of bias risk was conducted according to the Revised Cochrane’s Risk of Bias tool (ROB 2.0) for RCTs and AMSTAR-2 tool for meta-analysis. Result A total of 18 RCTs and 8 meta-analyses are included. In the meta-analysis, no significant association between high-protein supplementation and overall mortality is found (OR 0.99, 95% CI 0.95–1.28, p = 0.69; I 2 = 0%). In the umbrella review, there are no significant differences in each outcome between high-protein supplementation and conventional protein supplementation. In addition, long term intervention for more than 14 days would slightly increase overall mortality of critically ill patients (OR 1.181, 95% CI 1.002 to 1.391, p = 0.047; I 2 = 0%). Conclusion Compared with high and conventional protein supplementation, high-protein supplementation has no significant impact on clinical outcomes of critically ill patients, and long-term intake might increase overall mortality of patients. Systematic review registration Identifier: CRD420250652684.