Please sir, may I have some more? The case against underfeeding

In the critically ill patient, current recommendations endorse administering nutritional support according to indirect calorimetry measurements or if not available, according to formulae, such as 25 kcal/kg ideal body weight/day or other predictive equations (1,2). However, it is universally recognized that these equations are largely inaccurate and may result in both under- or over-feeding in 40-70% of patients. Over the past few years, numerous studies have compared outcomes of patient receiving by different routes, various amounts of calories and protein (3-5). One of the most recent, by Arabi et al. (6), namely the PermiT study, compared a standard caloric regimen consisting of 70-100% of estimated caloric requirements as determined by the Penn State equation, to a permissive underfeeding regimen consisting of 40-60% of caloric requirements.