Answers are grounded in this paper and cite their sources.
Nutritional support for patients with obesity and sepsis remains challenging. Guidelines generally recommend avoiding early overfeeding and providing hypocaloric feeding with adequate protein for critically ill patients with obesity, but evidence specific to ICU patients with obesity and sepsis is limited. Using MIMIC-IV, we emulated a target trial in adults with obesity meeting Sepsis-3 criteria who were alive and still in the ICU 72 h after admission. Time zero was hour 72, and the nutritional strategy grace period spanned hours 72–168. Strategies were hypocaloric nutrition (40%–70% of the ideal body weight-based energy target) and standard-calorie nutrition (70%–100%). We used a clone–censor–weight approach with inverse probability of censoring weights. We included 3,878 eligible ICU stays, generating 7,756 cloned records. Among grace-period completers, 320 met the standard-calorie target and 383 met the hypocaloric target. Mortality point estimates favored hypocaloric nutrition, but overall comparisons were not statistically significant: the odds ratios (ORs) were 0.91 (95% CI 0.75–1.10) for hospital mortality, 0.92 (95% CI 0.76–1.12) for ICU mortality, and 0.87 (95% CI 0.71–1.07) for 28-day mortality. Exploratory subgroup associations were more pronounced in patients with SOFA scores ≥ 10, early vasoactive drug use, or renal replacement therapy. In this cohort, hypocaloric nutrition coincided with lower protein intake (0.85 vs. 1.19 g/kg ideal body weight/day). In patients with obesity and sepsis who were alive and still in the ICU 72 h after admission, hypocaloric nutrition during ICU days 4–7 did not significantly reduce overall mortality compared with standard-calorie nutrition, although point estimates favored hypocaloric feeding. Exploratory subgroup findings identified populations for prospective investigation, suggesting that potential benefits may be concentrated in patients with greater illness severity or hemodynamic instability. Neither strategy has been shown to be safer or clinically preferable. Trial registration: Not applicable.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: