Although a meta-analysis does not replace a large, multicenter,
randomized, controlled trial that compares EN with PN in the
critically ill patient, it does provide useful information and can
guide us in the development of such a trial to specifically assess
treatment effects of EN versus PN. This would also require a
change in how the nutrition community performs such studies, so
The main clinical implication of our data concerns the use of
nutritional support in critically ill patients who can tolerate some
EN. Our findings suggest that EN is the preferred method to
provide nutritional support to critically ill patients. Although we
did not find any difference in mortality rate between patients
administered EN and PN, the meta-analysis lacked power to detect
a small but meaningful treatment effect. Moreover, a difference in
infectious complications alone warrants a preferential recommen-
dation of EN. Acquired infection, in particular ventilator-
associated pneumonia, is a major problem for critically ill patients,
which results in increased morbidity and mortality rates and health
care costs.
49–51
Perceived barriers to using EN for nutritional support
include concerns over the risk of aspiration pneumonia, high gastric
residuals and bowel irregularities, and an inability to reach targeted
nutritional goal rates. In those patients on pressors, there is the
added concern of the potential to increase the oxygen demand of
the gastrointestinal tract in those who are fed with EN