A large consensus supports the provision of high protein intakes during the late phase of critical illness, e.g., during recovery when the ability to increase the synthesis of muscle proteins from the pool of circulating amino acids increases [1, 2]. However, controversial views are expressed regarding the amount of proteins to be given during the early phase of critical illness, when muscle protein breakdown outweighs muscle synthesis as a result of the resistance to anabolic stimuli [3, 4]. The proportion of nitrogen losses to be compensated by protein intake in the critically ill is a matter of debate, as reflected by recommendations cited in the most recently published guidelines: 1.2–2.5 g/kg of protein per day [5, 6] and the provision of an amount of protein lower than nitrogen losses [1, 4], in agreement with the “Baby stomach” concept [7]. These discrepant views based on experts’ opinions reflect the paucity of data from adequately powered clinical studies assessing the effects of different amounts of proteins on relevant endpoints [8].

Meanwhile, industrial companies recently started to market nutritional formulas containing high amounts of proteins or amino acids and promoted their use early during the course of critical illness, following experts’ opinions mainly based on associations between high nitrogen intakes and better outcomes and on biochemical arguments. The marketing of these solutions is possible as the legal standards do not require the same sequence of testing as for a new drug, i.e., phase I clinical trials to check the safety, phase II clinical trials to assess the efficacy, and phase III clinical trials to compare the new treatment with the current standard of care. In the field of nutrition, this sequence is usually not followed; as a result the issue of safety may have been overlooked [9]. Nonetheless, the issues raised by the three phases of clinical testing are relevant for nutritional solutions as well as for any new therapeutic modality.

The community of clinicians is then left with conflicting arguments either supporting the use of high protein solutions or cautioning against this practice (Fig. 1). This manuscript intends to summarize the arguments supporting both sides and the current clinical research.

Fig. 1
figure 1

The arguments in favor and against high protein intake during the early phase of critical illness. AA amino acid

High nitrogen intake during the early phase of critical illness: the pros

The renewed enthusiasm for high protein intake results mainly from attention paid to ICU-acquired muscle weakness (ICU-AW). Indeed, the importance of ICU-AW in the outcome of critically ill patients has been underlined by the description of long-term physical impairments and disabilities impairing the quality of life of survivors and increasing healthcare-related costs. The time course of muscle wasting is characterized by an initial abrupt drop in muscle mass and function followed by a slow, progressive recovery [10,11,12,13]. Recently, decreased mitochondrial biogenesis and dysregulated lipid oxidation have been reported as contributors to compromised skeletal muscle bioenergetic status [14]. Clinically, in addition to a decrease in functional autonomy and quality of life, this prolonged muscle weakness represents a huge burden for society as a high proportion of patients who required an ICU stay lasting several days are unable to return to work or even to home [15].

The prevention of ICU-AW requires a multi-modal “bundle” approach, including the avoidance of sedation, early mobilization, and ambulation. The inclusion of high protein intakes in this bundle of measures appears logical as an adjunctive measure to limit the loss of muscle mass and function by boosting the synthesis of muscle proteins. High protein intake is expected to stimulate new protein synthesis, thereby preserving muscle mass [6]. The combination of physical activity, including active and passive mobilization, with high protein nutritional formulas or supplemental intravenous amino acids was suggested as a “must” for physical rehabilitation. Research in this area was even ranked as the number one priority by a group of experts [16]. Compelling retrospective data on large cohorts of patients support these expectations, as improved survival was observed in patients who received the highest amounts of protein, regardless of their physical activity [17,18,19,20].

The results of a recent clinical prospective study confirm that it is possible to increase the circulating pool of amino acids with an enteral solution containing high amounts of proteins [21], in spite of high splanchnic extraction of some amino acids [22]. In other interventional studies, intravenous infusion of amino acids was found to be safe in patients at risk of acute renal failure [23] and transiently improved muscle function [24]. Improved 90-day survival was even found in a post-hoc analysis in the subset of patients with normal renal function [25].

Parallel to this quantitative approach, the qualitative aspects of proteins can also represent a promising area of clinical research. For instance, whey proteins could increase muscle synthesis more efficiently than soy or casein-based solutions as a result of their higher digestibility, their higher content in leucine, and their insulinotropic properties [26, 27]. Likewise, the effects of semi-elemental or elemental solutions should be re-considered as a means to improve digestibility and protein availability during enteral nutrition [28].

High nitrogen intake during the early phase of critical illness: the cons

On the “con” side of high protein intake, no clinical benefit has been reported from interventional studies comparing solutions containing high amounts of nutrients, including proteins, with standard amounts [29,30,31,32]. However, in contrast to a potential benefit on muscle protein synthesis, the issue of the safety of high nitrogen intake during the acute phase of critical illness is an emerging concern. Indeed, a preplanned post-hoc analysis of the PEPaNIC study [33] that evaluated the effects of withholding parenteral nutrition in critically ill children suggests a linear positive correlation between the amount of amino acids provided and poorer outcome in the children randomized to the early parenteral nutrition group, until day 4 after admission [34] . The underlying mechanisms are not fully understood and are currently being investigated. Besides increased urea generation reported in the EAT-ICU (Early Goal-Directed Nutrition in ICU Patients) trial [35], increased production of glucagon leading to further oxidation of amino acids has also been reported [36].

Teleologically, muscle wasting could be considered a desirable consequence of adaptive anabolic resistance and lasts a few hours or days after injury [37]. The inability to respond to anabolic stimuli during the acute phase can be considered as a component of an adaptive response designed to provide substrates for gluconeogenesis in order to meet the requirements of vital organs and systems, an event known as auto-cannibalization or auto-cannibalism [3]. In this scenario, the loss of muscle would serve to supply gluconeogenetic organs. Likewise, the ability of muscles to build myofibrils will be limited and the provision of high amounts of amino acids will not attenuate the muscle wasting and could even amplify the degradation of amino acids.


The risk-to-benefit ratio of the provision of high amounts of proteins or amino acids during the early phase of critical illness is largely unknown. Some aspects have been investigated, while others are still unexplored.

Importantly, the optimal combination of proteins and physical activity is unknown [38]. This is a key issue when early physical activity is feasible and probably beneficial. Of note, the needs and protein metabolism of elderly and/or obese patients can differ from those of the overall ICU population [6, 39, 40].

Hopefully, some of the pending issues could be answered by some of the ongoing trials registered on (Table 1). Most of the currently recruiting studies are prospective randomized controlled trials. The inclusion criteria studies are highly variable, even though an anticipated long stay and the requirement for mechanical ventilation are mandatory in most trials. The primary outcomes tend to focus on physical function in several studies, while all-cause mortality is less commonly used as a primary outcome. A wide range of interventions are being tested and compared to the standards of care, from supplemental proteins (1.5–3.0 g/kg/day) alone to combination with standardized physical activity.

Table 1 Ongoing studies currently recruiting adult patients (source: – Jul 23, 2018)

Meanwhile, owing to the potential risks of high amounts of proteins, the principle of precaution should prevail, i.e., the provision of 0.3–0.8 g proteins/kg/day during the early phase of critical illness. We definitely need to appraise more precisely the risk-to-benefit ratio by characterizing the relevant risks and measuring muscle function at the bedside as a proxy for the benefit of high protein intake.