RESEARCH QUESTION: In critically ill patients, what is the association between preexisting malnutrition and time to discharge alive (TTDA), and does high protein treatment modify this association?

STUDY DESIGN AND METHODS: This multicenter randomized controlled trial involving 16 countries was designed to investigate the effects of high vs usual protein treatment in 1,301 critically ill patients. The primary outcome was TTDA. Multivariable regression was used to identify if preexisting malnutrition was associated with TTDA and if protein delivery modified their association.

RESULTS: The prevalence of preexisting malnutrition was 43.8%, and the cumulative incidence of live hospital discharge by day 60 was 41.2% vs 52.9% in the groups with and without preexisting malnutrition, respectively. The average protein delivery in the high vs usual treatment groups was 1.6 g/kg per day vs 0.9 g/kg per day. Preexisting malnutrition was independently associated with slower TTDA (adjusted hazard ratio, 0.81; 95% CI, 0.67-0.98). However, high protein treatment in patients with and without preexisting malnutrition was not associated with TTDA (adjusted hazard ratios of 0.84 [95% CI, 0.63-1.11] and 0.97 [95% CI, 0.77-1.21]). Furthermore, no effect modification was observed (ratio of adjusted hazard ratio, 0.84; 95% CI, 0.58-1.20).

INTERPRETATION: Malnutrition was associated with slower TTDA, but high protein treatment did not modify the association. These findings challenge current international critical care nutrition guidelines.

CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov; No.: NCT03160547; URL: www.

METHODS: Four electronic databases were searched from inception to December 2022 (MEDLINE via Ovid, EMBASE via Ovid, CINAHL via Healthcare Databases Advanced Search, CENTRAL via Cochrane). Studies were included if they examined at least one clinimetric property of a CONCISE measurement instrument or recognised variation in adults ≥ 18 years with critical illness or recovering from critical illness in any language. The COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) checklist for systematic reviews of Patient-Reported Outcome Measures was used. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses were used in line with COSMIN guidance. The COSMIN checklist was used to evaluate the risk of bias and the quality of clinimetric properties. Overall certainty of the evidence was rated using a modified Grading of Recommendations, Assessment, Development and Evaluation approach. Narrative synthesis was performed and where possible, meta-analysis was conducted.

RESULTS: A total of 4316 studies were screened. Forty-seven were included in the review, reporting data for 12308 participants. The Short Form-36 Questionnaire (Physical Component Score and Physical Functioning), sit-to-stand test, 6-m walk test and Barthel Index had the strongest clinimetric properties and certainty of evidence. The Short Physical Performance Battery, Katz Index and handgrip strength had less favourable results. There was limited data for Lawson Instrumental Activities of Daily Living and the Global Leadership Initiative on Malnutrition criteria. The risk of bias ranged from inadequate to very good. The certainty of the evidence ranged from very low to high.

CONCLUSIONS: Variable evidence exists to support the clinimetric properties of the CONCISE measurement instruments. We suggest using this review alongside CONCISE to guide outcome selection for future trials of nutrition and metabolic interventions in critical illness.