EDITORIAL

C URRENT
OPINION Personalized nutrition therapy in critical illness
and convalescence: moving beyond one-size-fits-all
to phenotyping and endotyping
Arthur R.H. van Zanten

INTRODUCTION is not suppressed by nutrition therapy and adminis-

Recent research has highlighted the limitations of tration of insulin [7]. Therefore, full feeding in the early
traditional feeding guidelines for patients, which phase may induce overfeeding. Considering whether
fail to consider individual requirements, body com- the provided energy promotes anabolic processes or
position, and metabolic stress. However, personal- induces insulin resistance and hyperglycaemia is cru-
ized nutrition therapy has emerged as a promising cial [8]. Optimal glycaemic targets are addressed
approach to improve outcomes and prevent adverse by Gunst et al. in this journal [9]. Currently, we lack
effects. Phenotyping involves considering patient- biomarkers to guide energy intake in critical illness and
specific data and body composition to determine recovery phases. However, a high respiratory quotient
the ideal macronutrients and micronutrients (RQ), elevated insulin needs, and exceeding 110% of
during various stages of illness and recovery. By REE based on indirect calorimetry measurements may
employing endotyping, which identifies distinct indicate metabolic intolerance (endotyping) [10–13].
disease-related mechanisms and metabolic bio- In a recent meta-analysis on individualized energy
markers, tailored nutritional plans can be devel- strategies [14], a 23% decrease in mortality was
oped. This edition of Current Opinion in Critical observed with personalized energy provision based
Care focuses on the latest insights into personalized on indirect calorimetry (IC). However, concerns
nutrition therapy. were raised regarding study size, inconsistent mortal-
ity data, and the lack of mortality reduction in
individual studies or impact on other patient out-
INDIVIDUALISED ENERGY STRATEGY comes [15]. Nevertheless, experts and guidelines
Energy provision in nutritional support is crucial, endorse the use of IC-based strategies in guiding
but the outdated approach of using total body nutrition therapy [1,16].
weight and predictive equations for dosing should Composition and routes of administration of
be discarded [1]. Phenotyping, which considers energy sources are also relevant. Typically, paren-
patient characteristics like BMI and nonintentional teral nutrition contains all three macronutrients.
calorie intake, is essential [2–3]. Patients with higher Haines et al., in this journal edition, address that
BMIs may need increased energy intake to maintain with similar energy doses better outcomes are
metabolic balance, while those receiving noninten- observed when soy-based lipids are switched to
tional calories (e.g., propofol, citrate, or dextrose-
containing fluids) may require lower energy intake Department of Intensive Care Medicine & Research, Gelderse Vallei
to prevent overfeeding [4]. Additionally, energy Hospital, Ede, The Netherlands
requirements can significantly differ throughout Correspondence to Arthur R.H. van Zanten, MD, PhD, Chair Department
the different phases of critical illness [5]. of Intensive Care Medicine & Research, Gelderse Vallei Hospital, Willy
Endotyping involves measuring variations in Brandtlaan 10, 6716 RP Ede, The Netherlands/Division of Human
energy expenditure using indirect calorimetry Nutrition and Health, chair group Nutritional Biology, Wageningen Uni-
versity & Research, HELIX (Building 124), Stippeneng 4, 6708 WE
(IC), recently recommended by international nutri-
Wageningen, The Netherlands.
tion guidelines, and assessing endogenous energy E-mail: zantena@zgv.nl or arthur.vanzanten@wur.nl
production to determine individual energy require- Curr Opin Crit Care 2023, 29:281–285
ments. Although relevant, simple bedside determi-
DOI:10.1097/MCC.0000000000001060
nation of endogenous energy production is not
This is an open access article distributed under the Creative Commons
possible [6]. Attribution License 4.0 (CCBY), which permits unrestricted use, dis-
Early full-dose feeding is associated with adverse tribution, and reproduction in any medium, provided the original work is
outcomes as endogenous energy production then properly cited.

1070-5295 Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc. www.co-criticalcare.com
Metabolic support

fish-oil-based lipid emulsions [17]. Mitochondrial in the subgroup of patients with acute kidney
dysfunction may induce an intracellular energy cri- injury and high SOFA scores, increased mortality
sis [18]. Alternative energy sources then should be was observed [28]. More studies on high protein
considered. The role of ketones with anticatabolic interventions in ICU are ongoing such as the PRE-
properties as promising alternatives is addressed in CISE trial (NCT04633421) and the Target protein
this edition by Watson et al. [19]. trial (ACTRN12621001484831).
Personalized energy strategies, incorporating Metabolic tolerance of amino acids varies
phenotyping and endotyping, may surpass the tra- among patients and changes during critical illness
ditional one-size-fits-all approach for critically ill and convalescence. Endotyping, using intracellular
patients. However, more research is necessary to biomarkers, can identify specific protein needs
establish standardized methods for determining when phenotyping is insufficient. In ICU patients,
individual energy needs and evaluating the long- despite similar protein digestion, absorption and
term effects of personalized energy provision on plasma amino acid levels compared to healthy con-
clinical outcomes. trols, severe anabolic resistance leads to significantly
lower muscle protein synthesis, indicating the
impact of critical illness on protein metabolism
INDIVIDUALISED PROTEIN STRATEGY [29]. The mechanisms involved in these processes
Adequate protein intake is crucial for muscle mass, are not fully understood, but potential factors
immune function, and wound healing in critically include mitochondrial dysfunction, intracellular
ill patients. However, a one-size-fits-all approach to energy depletion, proteostatic phenotype induc-
protein dosing based on total body weight may not tion, and disruption of cellular metabolic processes
be effective. Individualized protein dosing based through autophagy suppression [30–31]. The role of
on phenotyping and endotyping is recommended. autophagy in critical illness is discussed in detail in a
Phenotyping involves identifying patient charac- comprehensive overview by Vanhorebeek et al. in
teristics that may affect protein requirements, such this journal [32].
as body mass index (BMI), sex, lean body mass Proteins and amino acids not used for muscle
(LBM), and age. Studies have shown that men protein synthesis can be metabolized in various
generally require higher protein intake than ways. Amino acids can be converted to glucose
women due to their larger body size and higher through gluconeogenesis in the liver. Excess amino
LBM [20]. However, in patients of the same sex and acids are metabolized in the liver, producing
total body weight, variations in LBM can result in ammonia, which is then converted to urea for safe
potential protein overfeeding or underfeeding. Eld- excretion. The urea –creatinine-ratio (UCR) may be
erly patients and those with lower BMIs may also a useful biomarker for monitoring whether
have higher protein requirements to prevent ingested proteins are utilized for muscle protein
muscle wasting. Increasing protein dose has been synthesis, oxidized for energy, or metabolized to
associated with reduced but also enhanced muscle urea. High UCR levels in patients from the REDOXS
mass loss [21]. trial were linked to adverse clinical outcomes, par-
Early high protein intake may worsen clinical ticularly increased mortality with high-dose gluta-
outcomes, as has been shown in an observational mine supplementation [33]. Incorporating UCR
study and posthoc analyses of randomized con- into the analysis revealed that the association
trolled trials (RCTs) [22–24]. Recently, our group between glutamine and increased mortality disap-
also found associations with increased mortality peared, indicating that it was the higher amino
of high protein intake during the emergence of acid dosage, as indicated by elevated UCR levels,
refeeding hypophosphatemia [25]. Therefore, guide- rather than glutamine itself, that was metabolically
lines have recommended gradual progression to the intolerable. These findings may also help explain
protein intake target [1]. For this reason, in our ICU, the adverse effects of high-protein intake in
we apply daily steps of 25% increase both for pro- patients with high SOFA scores, early-phase critical
teins and energy [26]. illness, and AKI patients, as inflammation is
A recent meta-analysis found no evidence the primary driver of catabolic responses and ana-
of benefits on clinical outcomes, including mortal- bolic resistance.
ity and length of stay, when increasing protein Personalized protein dosing based on phenotyp-
intake by approximately 0.5 g/kg per day during ing and endotyping could help optimize nitrogen
critical illness [27]. This was confirmed in the balance, prevent muscle wasting, and improve clin-
EFFORT trial by Heyland et al. comparing an ical outcomes in critically ill patients while avoiding
achieved protein target of 1.6 vs. 0.9 g/kg per day the harmful effects of excessive protein intake. Addi-
that showed no differences in outcomes. Moreover, tionally, monitoring the loss of muscle mass via

282 www.co-criticalcare.com Volume 29 ! Number 4 ! August 2023

 Personalized nutrition therapy in critical illness and convalescence van Zanten

FIGURE 1. Moving from one-size-fits-all strategies in nutrition therapy to phenotyping and endotyping critically ill patients to
meet individual metabolic demands in all phases of disease and convalescence. Instead of relying on general population-
based averages for energy, protein, and micronutrient intake, individualized nutrition strategies should be tailored to patients
based on specific phenotypic factors such as sex, age, body composition, and treatments. Additionally, endotyping can be
used to assess the metabolic condition of the patient and predict how nutrients will be utilized, either for beneficial metabolic
processes or potentially harmful effects. Personalized nutrition plans based on individualized daily demands and endotyping
potentially can improve patient outcomes. ATP, adenosine triphosphate; BIA, bioelectrical impedance assessment; BW, body
weight; CO2, carbon dioxide; CRRT, continuous renal replacement therapy; g, grams; kcal, kilocalories; kg, kilograms; LBM,
lean body mass; RDA, recommended daily allowances; REE, resting energy expenditure; RQ, respiratory quotient; VCO2,
production of CO2 per minute.

muscle ultrasound or BIA during ICU stay may be a Phenotyping involves identifying patient char-
helpful tool for following up with individual acteristics, such as age, sex, and medical history, that
patients at risk (Fig. 1). may affect micronutrient requirements. For exam-
ple, elderly patients may have reduced absorption
and utilization of certain vitamins like vitamin D,
INDIVIDUALISED MICRONUTRIENT while those with renal impairment may experience
STRATEGY increased losses of micronutrients during continu-
Micronutrients, such as vitamins and trace ele- ous renal replacement therapy. To prevent micro-
ments, play a critical role in maintaining an indi- nutrient deficiencies, early multisupplementation
vidual’s overall health, particularly during times of with low doses of various micronutrients can be
stress and illness. However, feeding ICU patients has considered for specific phenotypes. This is especially
traditionally relied on a one-size-fits-all approach, relevant when enteral feeding administration is pro-
using recommended daily allowances as a guideline gressed gradually over 4–5 days and the target of
[34]. Instead, a personalized approach based on 1500 kcal or 1500 ml of tube feeding, with sufficient
phenotyping and endotyping may be more appro- micronutrients to meet the recommended daily
priate. allowance (RDA), is not achieved.

1070-5295 Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc. www.co-criticalcare.com 283
Metabolic support

Endotyping involves measuring intracellular requires adequate vitamins and trace elements. How-
biomarkers or downstream metabolic activity to ever, the traditional method of providing micronu-
identify specific micronutrient deficiencies not trients based on recommended daily allowances may
reflected in serum levels. not suffice, calling for a personalized approach based
Micronutrients in critical illness are excellently on phenotyping and endotyping. Identifying specific
addressed by Koekkoek and Berger in this journal deficiencies through intracellular and downstream
[35]. In critical illness, ionized calcium levels are biomarkers is crucial. Nonetheless, implementing
typically lower. Melchers and Van Zanten discuss in this complex personalized approach poses chal-
this issue whether this indicates adaptation or lenges. High-dose micronutrient supplementation
necessitates calcium supplementation [36]. is not recommended due to potential adverse effects
Radke et al. [37] provide the latest insights on and lack of proven benefits.
high-dose vitamin C supplementation in this jour- This special edition of Current Opinion in Crit-
nal. Although some studies indicate potential ben- ical Care offers valuable insights for effective and
efits of high-dose micronutrient supplementation individualized nutrition therapy.
for critically ill patients, it is crucial to recognize the
adverse effects of excessive doses. Currently, high-
Acknowledgements
dose supplementation is not recommended for ICU
patients. In summary, a personalized approach to I like to express my sincere gratitude to all the contrib-
micronutrient supplementation based on pheno- utors and the editor-in-chief, who have dedicated their
time and effort to share their insights and expertise and
typing and endotyping may be more effective in
have made this special edition possible.
optimizing the nutritional status of ICU patients.
However, further research is needed to develop
this approach. Financial support and sponsorship
None.
OVERCOMING BARRIERS TO ACHIEVING Conflicts of interest
NUTRITIONAL TARGETS IN CRITICAL
Professor Van Zanten reported receiving honoraria for
ILLNESS
advisory board meetings, lectures, research, and travel
Nutrition therapy in critical illness faces barriers hin- expenses from Abbott, AOP Pharma, Baxter, Cardinal
dering its effectiveness. Viner Smith et al. note low Health, Danone-Nutricia, Dim-3, Fresenius Kabi, GE
intake in patients receiving nasal high-flow oxygen Healthcare, Medcaptain, Mermaid, Nestle!, PAION,
therapy or noninvasive ventilation [38]. Gastrointes- Lyric, and Rousselot.
tinal (GI) dysfunction further hampers nutrition ther-
apy by affecting GI motility, nutrient digestion and
absorption, and gut metabolism. To address these REFERENCES AND RECOMMENDED
challenges, researchers have explored strategies such READING
Papers of particular interest, published within the annual period of review, have
as energy-dense and elemental formulas, prokinetics, been highlighted as:
intermittent feeding, and postpyloric feeding, with & of special interest
&& of outstanding interest
varying success. Van Gassel et al. [39] provide a com-
prehensive exploration of this topic. In the post-ICU 1. Singer P, Blaser AR, Berger MM, et al. ESPEN guideline on clinical nutrition in
phase, De Waele et al. [40] present updated insights on the intensive care unit. Clin Nutr 2019; 38:48–79.
2. Graf S, Pichard C, Genton L, et al. Energy expenditure in mechanically
feeding practices and strategies to improve macronu- ventilated patients: the weight of body weight! Clin Nutr 2017; 36:224–
trient intake during convalescence. 228.
3. Bousie E, van Blokland D, Lammers HJ, van Zanten AR. Relevance of
nonnutritional calories in mechanically ventilated critically ill patients. Eur J
CONCLUSION Clin Nutr 2016; 70:1443–1450.
4. Bousie E, van Blokland D, van Zanten ARH. Effects of implementation of a
Personalized nutrition therapy is essential for better computerized nutritional protocol in mechanically ventilated critically ill pa-
tients: a single-centre before and after study. Clin Nutr ESPEN 2016; 11:
outcomes in critically ill patients, necessitating an e47–e54.
individualized approach based on phenotyping and 5. van Zanten ARH, De Waele E, Wischmeyer PE. Nutrition therapy and critical
illness: practical guidance for the ICU, post-ICU, and long-term convales-
endotyping. Energy strategies tailored to the patient’s cence phases. Crit Care 2019; 23:368.
phenotype and guided by indirect calorimetry can 6. Udin I, Habisreutinger M, Tappy L, et al. Magnitude of gluconeogenesis and
endogenous glucose production: are they predictable in clinical settings?
be more effective than traditional weight-based Clin Nutr 2021; 40:3807–3814.
approaches. Likewise, protein dosing should be per- 7. Fraipont V, Preiser JC. Energy estimation and measurement in critically ill
patients. JPEN J Parenter Enteral Nutr 2013; 37:705–713.
sonalized to prevent underfeeding or overfeeding, 8. Fong KM, Au SY, Ng GWY. Glycemic control in critically ill patients with or
considering metabolic tolerance. Biomarkers like without diabetes. BMC Anesthesiol 2022; 22:227.
9. Gunst J, Verbruggen SC. Insulin resistance in critical illness: consequences
the UCR show promise in assessing protein metabo- for nutrition therapy and glucose management. Curr Opin Crit Care 2023;
lism. Cellular homeostasis during critical illness 29:286–292.

284 www.co-criticalcare.com Volume 29 ! Number 4 ! August 2023

 Personalized nutrition therapy in critical illness and convalescence van Zanten

10. Frankenfield DC. Impact of feeding on resting metabolic rate and gas exchange Nutrition in Acute Lung Injury Trial (INTACT): a post hoc analysis. Am J Clin
in critically ill patients. JPEN J Parenter Enteral Nutr 2019; 43:226–233. Nutr 2017; 105:411–416.
11. Chapple LS, Weinel L, Ridley EJ, et al. Clinical sequelae from overfeeding in 25. Slingerland-Boot H, Rooijakkers E, Wac K, et al., Macronutrient intake and
enterally fed critically ill adults: where is the evidence? JPEN J Parenter Enteral outcomes of ICU patients with refeeding hypophosphatemia, Clin Nutr
Nutr 2020; 44:980–991. ESPEN, https://doi.org/10.1016/j.clnesp.2023.03.003.
12. Zusman O, Theilla M, Cohen J, et al. Resting energy expenditure, calorie and 26. Koekkoek KWAC, van Zanten ARH. Nutrition in the ICU: new trends versus
protein consumption in critically ill patients: a retrospective cohort study. Crit old-fashioned standard enteral feeding? Curr Opin Anaesthesiol 2018;
Care 2016; 20:367. 31:136–143.
13. Weijs PJ, Looijaard WG, Beishuizen A, et al. Early high protein intake is 27. Lee ZY, Yap CSL, Hasan MS, et al. The effect of higher versus lower protein
associated with low mortality and energy overfeeding with high mortality in delivery in critically ill patients: a systematic review and meta-analysis of
nonseptic mechanically ventilated critically ill patients. Crit Care 2014; 18:701. randomized controlled trials. Crit Care 2021; 25:260.
14. Duan JY, Zheng WH, Zhou H, et al. Energy delivery guided by indirect 28. Heyland DK, Patel J, Compher C, et al. EFFORT Protein Trial team. The effect
calorimetry in critically ill patients: a systematic review and meta-analysis. of higher protein dosing in critically ill patients with high nutritional risk
Crit Care 2021; 25:88. (EFFORT Protein): an international, multicentre, pragmatic, registry-based
15. Sundström Rehal M, Tatucu-Babet OA, Oosterveld T. Indirect calorimetry: randomised trial. Lancet 2023; 401:568–576.
should it be part of routine care or only used in specific situations? Curr Opin 29. Chapple LS, Kouw IWK, Summers MJ, et al. Muscle protein synthesis after
Clin Nutr Metab Care 2023; 26:154–159. protein administration in critical illness. Am J Respir Crit Care Med 2022;
16. Moonen HPFX, Beckers KJH, van Zanten ARH. Energy expenditure and 206:740–749.
indirect calorimetry in critical illness and convalescence: current evidence 30. Wang X, Middleton FA, Tawil R, Chen XJ. Cytosolic adaptation to mitochon-
and practical considerations. J Intensive Care 2021; 9:8. dria-induced proteostatic stress causes progressive muscle wasting.
17. Haines K, Grisel B, Gorenshtein L, Wischmeyer PE. Lipid emulsions in iScience 2021; 25:103715.
parenteral nutrition: does it matter? Curr Opin Crit Care 2023; 29:293–299. 31. Gunst J, Casaer MP, Preiser JC, et al. Toward nutrition improving outcome of
18. Moonen HPFX, Van Zanten ARH. Mitochondrial dysfunction in critical illness critically ill patients: How to interpret recent feeding RCTs? Crit Care 2023;
during acute metabolic stress and convalescence: consequences for nutrition 27:43.
therapy. Curr Opin Crit Care 2020; 26:346–354. 32. Vanhorebeek I, Casaer M, Gunst J. Nutrition and autophagy deficiency in
19. Watson N, McClelland TJ, Puthucheary Z. Is there a role for ketones as critical illness. Curr Opin Crit Care 2023; 29:306–314.
alternative fuel in critical illness? Curr Opin Crit Care 2023; 29:300–305. 33. Haines RW, Fowler AJ, Wan YI, et al. Catabolism in critical illness: a reanalysis
20. Moonen HP, Hermans AJ, Jans I, van Zanten AR. Protein requirements and of the REducing Deaths due to OXidative Stress (REDOXS) Trial. Crit Care
provision in hospitalised COVID-19 ward and ICU patients: Agreement Med 2022; 50:1072–1082.
between calculations based on body weight and height, and measured 34. Berger MM. Nutrition and micronutrient therapy in critical illness should be
bioimpedance lean body mass. Clin Nutr ESPEN 2022; 49:474–482. individualized. JPEN J Parenter Enteral Nutr 2020; 44:1380–1387.
21. Ishibashi N, Plank LD, Sando K, Hill GL. Optimal protein requirements during 35. Koekkoek KWA, Berger MM. An update on essential micronutrients in critical
the first 2 weeks after the onset of critical illness. Crit Care Med 1998; illness. Curr Opin Crit Care 2023; 29:315–329.
26:1529–1535. 36. Melchers M, van Zanten ARH. Management of hypocalcaemia in the critically
22. Koekkoek WACK, van Setten CHC, Olthof LE, et al. Timing of PROTein ill. Curr Opin Crit Care 2023; 29:330–338.
INtake and clinical outcomes of adult critically ill patients on prolonged 37. Radke DI, Homayr AL, Stoppe C, Elke G. Vitamin C in critical illness: end of
mechanical VENTilation: The PROTINVENT retrospective study. Clin Nutr the story or still a place? Curr Opin Crit Care 2023; 29:339–345.
2019; 38:883–890. 38. Smith EV, D’Cruz RF, Chapple L. Nutrition therapy during noninvasive
23. Casaer MP, Wilmer A, Hermans G, et al. Role of disease and macronutrient ventilation: oral or enteral? Curr Opin Crit Care 2023; 29:346–353.
dose in the randomized controlled EPaNIC trial: a post hoc analysis. Am J 39. van Gassel RJJ, Bels JLM, van de Poll MCG. Nutritional strategies during
Respir Crit Care Med 2013; 187:247–255. gastrointestinal dysfunction. Curr Opin Crit Care 2023; 29:354–359.
24. Braunschweig CL, Freels S, Sheean PM, et al. Role of timing and dose of 40. De Waele E, Rosseel Z, Pen JJ. Post-ICU nutrition: the neglected side of
energy received in patients with acute lung injury on mortality in the Intensive metabolic support. Curr Opin Crit Care 2023; 29:360–362.

1070-5295 Copyright © 2023 The Author(s). Published by Wolters Kluwer Health, Inc. www.co-criticalcare.com 285