OBJECTIVES: Across guidelines, protein dosing for critically ill patients with obesity varies considerably. The objective of this analysis was to evaluate whether this population would benefit from higher doses of protein.
DESIGN: A post hoc subgroup analysis of the effect of higher protein dosing in critically ill patients with high nutritional risk (EFFORT Protein): an international, multicenter, pragmatic, registry-based randomized trial.
SETTING: Eighty-five adult ICUs across 16 countries.
PATIENTS: Patients with obesity defined as a body mass index (BMI) greater than or equal to 30 kg/m 2 ( n = 425).
INTERVENTIONS: In the primary study, patients were randomized into a high-dose (≥ 2.2 g/kg/d) or usual-dose protein group (≤ 1.2 g/kg/d).
MEASUREMENTS AND MAIN RESULTS: Protein intake was monitored for up to 28 days, and outcomes (time to discharge alive [TTDA], 60-d mortality, days of mechanical ventilation [MV], hospital, and ICU length of stay [LOS]) were recorded until 60 days post-randomization. Of the 1301 patients in the primary study, 425 had a BMI greater than or equal to 30 kg/m 2 . After adjusting for sites and covariates, we observed a nonsignificant slower rate of TTDA with higher protein that ruled out a clinically important benefit (hazard ratio, 0.78; 95% CI, 0.58-1.05; p = 0.10). We found no evidence of difference in TTDA between protein groups when subgroups with different classes of obesity or patients with and without various nutritional and frailty risk variables were examined, even after the removal of patients with baseline acute kidney injury. Overall, 60-day mortality rates were 31.5% and 28.2% in the high protein and usual protein groups, respectively (risk difference, 3.3%; 95% CI, -5.4 to 12.1; p = 0.46). Duration of MV and LOS in hospital and ICU were not significantly different between groups.
CONCLUSIONS: In critically ill patients with obesity, higher protein doses did not improve clinical outcomes, including those with higher nutritional and frailty risk.
* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.