ICU nutrition research: did the evidence get better? Remaining sources of bias.

Abstract

To assess the quality of recently published (18 months from date of article request) randomized trials/systematic reviews of such trials that addressed the use of nutritional support in critically ill patients. Eligible papers were published between October 1, 2022 and April 7, 2024, only enrolled adults, described a comparison of a nutritional intervention to something else, and reported a clinically relevant outcome. Thirteen randomized trials and four systematic reviews of randomized trials were identified. Quality was assessed by determining the risks of bias of each trial. Two of these trials were at low risk of bias, six were rated as having some concern(s) about bias, and five were at high risk of bias. The four systematic reviews included 55 randomized trials; four were at low risk, 31 had some concerns, and 20 were at high risk. No randomized trial comparing nutritional support to a true control (no nutritional support) was identified in this search; seven older trials, all small and containing risks of bias, failed to demonstrate any consistent differences in clinical outcomes. The quality of the trials underlying the use of nutritional support in the intensive care unit is not very high.