Nutrition provision over time in longer stay critically ill patients: A post hoc analysis of The Augmented vs Routine Approach to Giving Energy Trial

Abstract

Background

Limited literature exists on nutrition practices for long-stay patients in the intensive care unit (ICU). We aimed to compare nutrition practices in the first and second weeks of an ICU admission.

Method

A post hoc exploratory analysis of The Augmented vs Routine Approach to Giving Energy Trial (TARGET) randomized controlled trial (RCT) was undertaken. Inclusion criteria were: enrolled in TARGET on day 1 or 2 of ICU admission and ICU length of stay (LOS) >14 days. Clinical characteristics are described, and nutrition delivery and management compared between days 1–7 and 8–14. Data are n (%), mean ± SD, median (interquartile range [IQR]), or mean difference (MD) and 95% confidence interval (95% CI), with P < 0.05 considered significant.

Results

Data from 664 patients were analyzed (56.2 ± 16.3 years; 61% male; body mass index 29.2 ± 7.5 kg/m² and APACHE II 21.9 ± 8.1). When comparing days 1–7 to 8–14: (1) energy delivery was greater (all sources: 1826 ± 603 vs 1729 ± 689 (MD: 97 [95% CI: 52–140] kcal/day, P < 0.001) and nonnutrition sources: 317 ± 230 vs 192 ± 197 (MD 125 [95% CI: 111–139] kcal/day; P < 0.001); (2) protein delivery was similar (66 ± 20 vs 68 ± 24 (MD: −1.4 [95% CI: −3.2 to 0.4] g/day; P = 0.125]); and (3) fewer patients received parenteral nutrition (PN) (5% vs 9%, P < 0.001) or small intestine feeding (3% vs 8%; P < 0.001).

Conclusion

In this post hoc analysis, patients with an ICU LOS >14 days had greater energy delivery and fewer patients received PN or small intestine feeding during days 1–7 than days 8–14. Uncertainty remains regarding whether these data reflect usual practice and the clinical implications of this.