Sie sind auf Seite 1von 16


Nutrition in Clinical Practice

Volume 00 Number 0
Determination of Nutrition Risk and Status in Critically Ill xxx 2018 1–16

C 2018 American Society for

Patients: What Are Our Considerations? Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10214

Zheng-Yii Lee, MSc1,2 ; and Daren K. Heyland, MD, MSc, FRCPC3

The stress catabolism state predisposes critically ill patients to a high risk of malnutrition. This, coupled with inadequate or delayed
nutrition provision, will lead to further deterioration of nutrition status. Preexisting malnutrition and iatrogenic underfeeding
are associated with increased risk of adverse complications. Therefore, accurate detection of patients who are malnourished
and/or with high nutrition risk is important for timely and optimal nutrition intervention. Various tools have been developed for
nutrition screening and assessment for hospitalized patients, but not all are studied or validated in critically ill populations. In this
review article, we consider the pathophysiology of malnutrition in critical illness and the currently available literature to develop
recommendations for nutrition screening and assessment. We suggest the use of the (modified) Nutrition Risk in the Critically
Ill (mNUTRIC) for nutrition risk screening and the subjective global assessment (SGA) together with other criteria relevant to
the critically ill patients, such as gastrointestinal function, risk of aspiration, determination of sarcopenia and frailty, and risk
of refeeding syndrome for nutrition assessment. Further research is needed to identify suitable nutrition monitoring indicators to
determine the response to the provision of nutrition. (Nutr Clin Pract. 2018;00:1–16)

critical illness; inflammation; malnutrition; nutrition assessment; nutrition status; risk assessment; screening

Introduction Pathophysiology of Malnutrition in Critically

In critically ill patients, nutrition status is closely linked Ill Patients
with clinical outcomes. However, determination of nutrition The pathophysiology of malnutrition during critical illness
status in critically ill patients is not a straightforward can be viewed from at least the following 2 main perspec-
process. Recognizing the role of inflammation in affecting tives: stress catabolism and inadequate nutrition intake.
the nutrition status of a patient, Jensen et al1 proposed In the early phase of critical illness, catabolic hormones
a concept to define malnutrition based on etiologies. This (such as glucagon, cortisol, and catecholamines) are
concept divides malnutrition into the following 2 main eti-
ologies: pure starvation without disease (starvation-related
malnutrition) and disease-related malnutrition associated From the 1 Department of Nutrition and Dietetics, Faculty of
Medicine and Health Sciences, Universiti Putra Malaysia, Serdang,
with variable degree of inflammation (chronic disease-
Malaysia; 2 Department of Anesthesiology, Faculty of Medicine,
related and acute disease or injury-related malnutrition).1 University of Malaya, Kuala Lumpur, Malaysia; and the
This conceptual model was later adopted by the consensus 3 Department of Critical Care Medicine, Queen’s University and

between the Academy of Nutrition & Dietetics (AND) and Clinical Evaluation Research Unit, Kingston General Hospital,
the American Society for Parenteral and Enteral Nutri- Kingston, Ontario, Canada.
tion (ASPEN) for the diagnosis of malnutrition (AND- Financial disclosures: None declared.
ASPEN Consensus).2 The discussion of this article will Conflicts of interest: None declared.
focus on the definition of acute disease or injury-related This article originally appeared online on xxxx 0, 2018.
malnutrition. Based on the pathophysiology of malnutri-
Corresponding Author:
tion in critical illness and the current available studies, Zheng-Yii Lee, MSc, Department of Nutrition and Dietetics, Faculty
considerations for nutrition screening and assessment of of Medicine and Health Sciences, Universiti Putra Malaysia, Serdang,
critically ill patients are discussed and suitable tools are Malaysia.
suggested. Email:
2 Nutrition in Clinical Practice 00(0)

secreted to mobilize body nutrition reserves (muscle and 2002 (NRS-2002)10 and the Nutrition Risk in the Critically
adipose tissue) for the generation of endogenous energy Ill (NUTRIC;11 Appendixes 1 and 2).
substrate (glucose, amino acids, and free-fatty acids) and The NRS-2002 was developed from an analysis of con-
to prioritize the delivery of these energy substrates to vital trolled trials and included recent dietary intake, weight
organs such as the brain or the heart.3,4 At the same time, loss, disease severity, and age to identify patients’ nutrition
proinflammatory cytokines such as Interleukin (IL)-1, IL-6 risks.10 A score of 3 is considered to be high nutrition risk.
and tumor necrosis factor-α are also secreted in response It has been shown to have good predictive validity in various
to the body’s acute insult and further exaggerate the hospitalized patients,12,13 but not in critically ill patients.
catabolism process.4 During such inflammatory states, the Furthermore, the automatic classification of patients with
provision of nutrition is not able to completely reverse the an Acute Physiology and Chronic Health Evaluation II
loss of body cell mass.3 Such conditions predispose critically score of 10 to high nutrition risk render most ICU
ill patients to a high risk of malnutrition (loss of body patients to be classified as high risk. Furthermore, the utility
cell mass to a critical level), and the risk of complications of variables such as recent food intake and weight change
is significantly increased if malnutrition ensues.5 At this may be limited due to data unavailability or inaccuracy.
stage, the priority is to provide nutrition support to support The NUTRIC score was developed in studies of critically
vital organ system functions and preserve appropriate host ill patient populations and has been validated in many
responses while the underlying disease is treated.6 observational studies from different countries.11,14-23 As
Although the disease process has tremendous impact on IL-6 is not routinely performed in most ICUs in the world,
the nutrition status of critically ill patients, depending on the modified-NUTRIC (mNUTRIC) score without IL-6
the patients’ history, the patients may already have features was proposed and revalidated.14 However, one of the major
of malnutrition with a reduced or restricted food intake limitations of the NUTRIC/mNUTRIC score is the absence
long before intensive care unit (ICU) admission due to the of classical nutrition variables such as recent food intake
underlying chronic conditions (such as chronic obstructive and weight change.
pulmonary disease, cancer, or chronic renal failure) or The ASPEN/Society of Critical Care Medicine
have reduced intake from a hospital stay prior to ICU guidelines suggest the use of either the NRS-2002 or
adminssion.6,7 Moreover, in the ICU, the patients may NUTRIC/mNUTRIC for the nutrition screening of a
continue to have restricted nutrition intake and thus they critically ill patient.24 However, the NUTRIC/mNUTRIC
may experience prolonged fasting or frequent feeding inter- score may be a more suitable nutrition risk screening tool
ruptions due to various ICU procedures.8 These 2 factors, than the NRS-2002 based on the following reasons:
preexisting malnutrition and iatrogenic underfeeding, may
further complicate the nutrition status and worsen clinical 1. The NUTRIC/mNUTRIC score was developed in
outcomes. the critically ill patient population with a clear
conceptual model based on the etiology-based mal-
nutrition definition proposed by the AND-ASPEN
Nutrition Risk Screening consensus and considered variables for starva-
ASPEN defined nutrition screening as “a process to identify tion (acute and chronic), inflammation (acute
an individual who is malnourished or who is at risk for mal- and chronic), age, disease severity, and organ
nutrition to determine if a detailed nutrition assessment is dysfunction.11
indicated.”9 In the critically ill patient, the stress catabolism 2. The absence of classical nutrition variables such as
process associated with the degree of inflammation (which weight change and recent food intake was due to
is closely linked with disease severity) requires nutrition the difficulty in obtaining such variables in ICU
screening that considers nutrition risk in the context of patients. During the initial development study of
disease severity and clinical outcomes.5 Nutrition screening the NUTRIC score, >70% of the data from these
in this context therefore “refers to the risk of acquiring variables were unable to be obtained. Moreover, the
complications and other forms of adverse outcome that inclusion of the available data from these variables
might have been prevented by timely and adequate nutrition were not predictive of the outcome (28-day mor-
support.”5 The nutrition risk screening tool must be able tality), and therefore they were eliminated from the
to identify the patients who are most likely to benefit from final statistical model.11 Even if these variables were
more adequate nutrition provision.10,11 Because the patho- obtained, the accuracy of this information may not
physiology of malnutrition is closely linked with the under- be verifiable.
lying inflammatory status, the scoring system must include 3. The variables in the final model of the NUTRIC/
variables related to the metabolic state and disease severity.5 mNUTRIC score correlate well with the pathophysi-
The following 2 nutrition risk screening tools in the litera- ology of malnutrition presented previously, whereby
ture have such characteristics: the Nutrition Risk Screening the degree of inflammation is a more determining
Lee and Heyland 3

factor of nutrition risk especially during the acute 2 of the 6 characteristics is recommended for the diagnosis
phase of critical illness and hence the use of the dis- of malnutrition.2 The criteria to diagnose moderate or
ease severity (Acute Physiology and Chronic Health severe malnutrition in the context of acute illness or injury
Evaluation II) and organ failure scoring (sequential is shown in Appendix 3. One recent prospective study in
organ failure assessment) is reasonable. critically ill patients has found that patients who were mal-
4. The variables “number of comorbidities” considered nourished as determined by the AND-ASPEN consensus
chronic inflammation and “days from hospital to criteria had 2.5 times higher hospital mortality risk than
ICU admission” considered iatrogenic reduced or their nonmalnourished counterparts.28
restricted food intake that occurred before ICU Besides the AND-ASPEN consensus criteria, another
admission. In our opinion, both of these variables potential nutrition assessment tool is the subjective global
are more objective and include the possibility of assessment (SGA), initially developed to predict postoper-
long-term and short-term reduced food intake and ative outcomes.29 The SGA incorporated weight change,
recent weight loss. recent food intake, gastrointestinal (GI) symptoms, func-
5. The NUTRIC/mNUTRIC score has been shown in tional capacity, subcutaneous fat loss, muscle loss, and fluid
various critically ill populations to have good predic- status for the diagnosis of moderate (SGA class B) or
tive validity for clinical outcomes, although the dis- severe malnutrition (SGA class C; Appendix 4). In fact, the
criminative ability is of fair level (Table 1).11,14-16,20-23 criteria used for assessment between the SGA and the AND-
Various observational studies in different popula- ASPEN consensus are notably similar.
tions have also shown that clinical outcomes are Nevertheless, the SGA has been widely studied in the
modified by nutrition adequacy depending on the critical care setting. A recent systematic review has iden-
risk status of the patients (Table 1).11,14,16-19 tified at least 10 studies that used the SGA in critically
ill patients.30 Among the 5 studies rated as high quality,
SGA class B or C when compared with SGA class A
The NUTRIC/mNUTRIC score remains to be prospec-
were associated with increased hospital mortality, longer
tively tested in a randomized controlled trial. There have
ICU length of stay, increased risk of new infection and
been inconsistent results from post hoc analyses of prospec-
ICU readmission, and increased percentage of patients
tive randomized trials. In one trial, similar clinical outcomes
discharged to nursing home.30 Therefore, the SGA may be
were demonstrated between patients with high and low
more favorable than the AND-ASPEN consensus criteria
nutrition risk regardless of the nutrition provision.25 Al-
for the nutrition assessment of critically ill patients.
though the subgroup analysis of another pilot randomized
However, both the SGA and the AND-ASPEN consen-
trial demonstrated a trend toward lower ICU and hospital
sus criteria have at least the following limitations:
mortality among patients with mNUTRIC 5 who received
greater energy and protein delivery by supplemental par-
1. They were not developed in critically ill populations
enteral nutrition (albeit not significant, P = .19), such a
and therefore lack variables relevant for critically ill
trend was not observed in patients with mNUTRIC <5.26
Future trials that focus on enrolling patients with high nutri-
2. They have failed to incorporate the severity of
tion risk may be able to provide a more definitive answer on
acute illness and criteria associated with stress
whether the mNUTRIC score can modify the relationship
between nutrition adequacy and clinical outcomes.
3. They are not “responsive enough” to changes within
a relatively shorter time duration and thus are not
Nutrition Assessment helpful in assessing whether nutrition provided is
Nutrition assessment has a different role from nutrition optimal, hindering effective nutrition monitoring
screening. As discussed previously, nutrition screening de- and evaluation.32
termines the nutrition risk of a patient, and in this context, 4. Critically ill patients are often unconscious and
the risk of acquiring complications as a consequence of therefore weight and food history are either not avail-
failure of optimal nutrition provision. In contrast, nutrition able or obtained from family member.11 Although
assessment is a formal evaluation of patient nutrition status 1 feasibility study reported that weight and food
by a trained healthcare professional, usually a dietitian, and history can be obtained in 54.5% and 76.4% of
results in a nutrition-related diagnosis.27 the ICU patients, respectively,33 the accuracy of the
The AND-ASPEN consensus statement proposes the information provided (especially by proxy) may not
characteristics and criteria for the diagnosis of malnutrition be verifiable.
(undernutrition), which are the following: energy intake, 5. Required subjective clinical interpretation and
weight change over time, subcutaneous fat loss, muscle loss, therefore training is needed especially among
fluid accumulation, and handgrip strength.2 A minimum of nonexperts.27
Table 1. NUTRIC/mNUTRIC: Discriminative Performance and Whether Nutrition Interventions Modify the Association Between Risk Score and Clinical Outcomes.

Do nutrition intervention
modify the association
Discriminative between risk score and
Author Country N Population Score Performance clinical outcomes?

Observational studies
Heyland et al, Canada 597 - Age 18 Total: 1–10 - aROC for 28-day Yes. High risk: ↑ nutrition
201111 - Stay 24 hours in - Low risk: 0–5 mortality: 0.783 adequacy was a/w ↓ 28-day
the ICU - High risk: 6–10 mortality (P = .01)

Rahman et al, 40 ICU in 1199 - Age 18 and MV Total: 1–9 - aROC for 28-day Yes. High risk: each 25% ↑ in %
201514 Europe & - Multiorgan failure - Low risk: 0–5 mortality: 0.648 of energy received was a/w ↓
North - Expected ICU LOS - High risk: 6–9 6-month mortality by 18%
America >5 days (P < .05)

Mukhopa-dhyay Singapore 401 - Medical ICU Total: 1–9 - aROC for 28-day Yes. High risk: ↑ nutrition
et al, 201616 - Age 18 - Low risk: 0–4 mortality: 0.71 adequacy was a/w ↓ 28-d
- MV 48 - High risk: 5–9 - Se & Sp: 72% & 63% mortality (P < .001)
hours (n = 273)
- Stay 24 hours in
the ICU

Compher et al, Worldwide 202 2853 - Age 18 and MV Total: 1–9 – Yes. High risk: ↑ E & protein was
201717 ICU - Stay in ICU 4 days - Low risk: 0–4 a/w ↓ 60-day mortality & time
(n = 2853), 12 days - High risk: 5–9 to discharge alive
(n = 1605)

Lee et al, 201718 Malaysia 154 - Age 18 Total: 1–9 – Yes. Low risk: E & protein
- MV within 48 hours - Low risk: 0–5 adequacy 2/3 of requirement
of ICU admission - High risk: 6–9 was a/w ↑ 60-day mortality by
- Stay 72 hours in about 6 times
the ICU

Hsu et al, 201819 Taiwan 190 - Age 65 Total: 1–9 – Yes. High risk: energy adequacy
- MV 48 hours - Low risk: 0–4 of 80% was
- High risk: 5–9 a/w ↓ ICU and hospital
mortality than <80%;; protein
adequacy of 80% was a/w ↓
hospital mortality than <80%
Moretti et al, Argentina 368 - Adult Total: 1–9 / 1–10 (use - aROC for mortality: –
201420 - MV >24 hours CRP for IL-6) 0.671
- Low risk: 0–4/0-5 - If substitute IL-6 with
- High risk: 5–9/6-10 CRP, aROC: 0.679

Table 1. (continued)

Do nutrition intervention
modify the association
Discriminative between risk score and
Author Country N Population Score Performance clinical outcomes?

Mendes et al, Portugal 1143 - Adult Total: 1–9 - aROC for 28-day –
201721 - ICU LOS >72 hours - Low risk: 0–4 mortality: 0.718
- High risk: 5–9

Lew et al, 201722 Singapore 439 - Adult 18 Total: 1–9 - aROC for hospital –
- ICU LOS 24 hours - Low risk: 0–4 mortality: 0.66
- High risk: 5–9

Kalaiselvan et al, India 678 - Adult Total: 1–9 - aROC for mortality: –
201723 - Required MV for - Low risk: 0–4 0.642
>48 hours - High risk: 5–9 - PPV, NPV, Se & Sp for
mortality: 47.4%,
68.9%, 41.5%, 73.8%,

de Vries et al, Netherlands 475 - Age 18 Total: 1–9 - aROC for 28-day –
201815 - MV within 24 hours - Low risk: 0–4 mortality: 0.768
of ICU - High risk: 5–9 - aROC for MV >2
days: 0.666

Randomized controlled trial

Arabi et al, 201725 Saudi Arabia 894 - Age 18 Total: 1–9 – No. Similar outcomes between
& Canada - Start EN within - Low risk: 0–4 higher and lower energy
48 hours - High risk: 5–9 delivery regardless of
- Expected to stay NUTRIC score
72 hours in the

Wischmeyer Canada, 125 - Age >18 Total: 1–9 Yes. High risk: ↑ nutrition
et al, 201826 United - Acute respiratory - Low risk: 0–4 adequacy by SPN was a/w a
States, failure - High risk: 5–9 not significant trend toward
Belgium, and - Start EN within reduced ICU and hospital
France 48 hours mortality (P = .19)
- BMI <25 or >35

↑ = increased, ↓ = decreased/reduced. a/w, associated with; aROC, area under the receiver operating characteristics curve; BMI, body mass index; CRP, C-reactive protein; E, energy; EN,
enteral nutrition; ICU, intensive care unit; IL-6, Interleukin-6; LOS, length of stay; MV, mechanically ventilated; mNUTRIC, Modified-NUTRIC; NPV, negative predictive value; NUTRIC,
Nutrition Risk in the Critically Ill; PPV, positive predictive value; Se, sensitivity; Sp, specificity; SPN; supplemental parenteral nutrition.

6 Nutrition in Clinical Practice 00(0)

Table 2. Gastrointestinal Symptoms That May Be Indicative mortality.34 In critically ill patients with enteral nutrition,
of Gastrointestinal Dysfunction. the risk factors of aspiration include a documented previ-
ous episode of aspiration, reduced level of consciousness
Symptom Definition (continuous infusion of sedatives or increased intracra-
nial pressure), vomiting, persistently high gastric residual
High GRV Maximum GRV >500 mL at least once volume, and being fed in supine position and/or having
Vomiting/ Visible vomiting or regurgitation in any increased risk of delayed gastric emptying (hyperglycemia,
regurgitation amount electrolyte abnormalities, use of drugs known to reduce
Diarrhea Loose or liquid stool 3 or more times per gastric emptying).35
day (diagnosis may also be based on the
King’s Stool Charta )
The potential use of various technologies for the assess-
Bowel distension Suspected or radiologically confirmed ment of body composition such as computed tomography
bowel dilatation in any bowel segment (CT), ultrasound, and bioelectrical impedance analysis has
GI bleeding Visible appearance of blood in vomits, been explored in various studies in recent years.36 Skeletal
nasogastric aspirate, or stool muscle quality revealed by CT scan at the third lum-
Intra-abdominal Mean intra-abdominal pressure of the day bar vertebra was associated with mortality in critically ill
hypertension 12 mmHg patients.37-39 However, CT scans are costly, require transfer
Abdominal Mean intra-abdominal pressure 20 mmHg
compartment with new organ dysfunction or failure,
to a radiology unit, and involve a dose of radiation that
syndrome with intra-abdominal pressure limits its usage to patients for whom CT is ordered for other
measured in the supine position with clinical reasons. The measurement of quadricep muscle
zero-point at mid axillary line with a layer thickness by ultrasound in the ICU also showed
maximal instillation volume of 25 mL promising results.40,41 Serial ultrasound measurement of the
rectus femoris cross-sectional area has shown that muscle
GRV, gastric residual volume. Adapted from Reintam Blaser A, Poeze
M, Malbrain ML, Bjorck M, Oudemans-van Straaten HM, Starkopf mass decreases continuously from ICU admission to almost
J. Gastrointestinal symptoms during the first week of intensive care one-fifth of baseline at day 10 of an ICU stay.42 Low fat-
are associated with poor outcome: a prospective multicentre study. free mass as indicated by low phase angle or high impedance
Intensive Care Med. 2013;39(5):899-909.
ratio obtained by bioelectrical impedance analysis measure-
a Whelan K, Judd PA, Preedy VR, Taylor MA. Covert assessment of ment has also been shown to be associated with mortality
concurrent and construct validity of a chart to characterize fecal and time-discharge alive among the critically ill patients.43,44
output and diarrhea in patients receiving enteral nutrition. JPEN J However, the routine use of such body composition anal-
Parenter Enteral Nutr. 2008;32(2):160-168.
ysis techniques may be impractical in the mechanically
ventilated critically ill population. Fluid and electrolyte
abnormalities commonly seen in critically ill patients will
Besides the use of the SGA or the AND-ASPEN con- influenced the accuracy of bioelectrical impedance analysis,
sensus criteria for nutrition assessment, there are many and further research is needed to define the optimal way to
other criteria relevant to ICU patients that a dietitian conduct ultrasound measurement for a better reliability and
needs to consider during the nutrition assessment process. validity in the ICU.36
The ASPEN/Society of Critical Care Medicine guidelines Because low skeletal muscle mass is associated with
suggest an evaluation of comorbid conditions, function of poor clinical outcomes in the ICU and traditional imaging
the GI tract, and risk of aspiration as part of nutrition methods are impractical or not yet validated, alternative
assessment.24 However, no further explanation or more spe- measures to diagnosing low skeletal muscle mass may be
cific variables were provided. It is indeed useful to consider needed.37-39 The SARC-F is a simple and validated ques-
the current diagnosis as well as the comorbid conditions tionnaire used to determine sarcopenia (Appendix 5).45,46
of the patients to evaluate the degree and duration of As the main feature of sarcopenia is the loss of muscle mass
inflammation as well as the possible duration of inadequate (and strength),47 the SARC-F may be useful to identify pa-
nutrition intake. In fact, the NUTRIC/mNUTRIC score tients with low skeletal muscle mass before ICU admission.
has the number of comorbidities as one of its variables. As sarcopenia is commonly associated with frailty,48 frailty
The variables to evaluate GI tract function may be can also be measured by using the Clinical Frailty Scale
adopted from the study of Reintam Blaser et al.34 This (Appendix 6).49 In a systematic review and meta-analysis,
prospective multicenter observational study found that GI 7 of the 10 included studies used the Clinical Frailty Scale
symptoms including high gastric residual volume, vomiting to measure frailty. The presence of frailty in the ICU was
or regurgitation, diarrhea, bowel distension, GI bleeding, found to be associated with increased mortality and reduced
and abdominal compartment syndrome (Table 2) were pre- likelihood to be discharged home.50 It is most likely that
dictive of 28-day mortality, and the occurrence of 3 or more sarcopenic and frail patients had poor nutrition status and
coincident GI symptoms was associated with higher 28-day warrant a more attentive nutrition intervention.47,50
Lee and Heyland 7

Table 3. Comparison of Variables Used by Nutrition Risk Screening and Assessment Tools.

Nutrition Risk Screening

Tools NA Tools
Components Variables NUTRIC NRS-2002 SGA AND-ASPEN
√ √ √
FH Recent food intake √
FH GI symptoms (nausea, vomiting,
diarrhea, anorexia) √
FH Days in hospital before ICU
admission (acute starvation) √
AD Current weight √ √ √
AD Weight change √b
BD IL-6 or CRPa (acute inflammation) √ √
PD Subcutaneous fat loss √ √
PD Muscle loss √ √
PD Fluid status √
PD Muscle strength (handgrip) √
PD Functional capacity √
CH Number of comorbidities (chronic
inflammation) √ √
CH Age √ √
Clinical APACHE II (disease severity) √
Clinical SOFA (organ dysfunction)

AD, anthropometric data; AND-ASPEN, Academy of Nutrition and Dietetics-American Society of Parenteral and Enteral Nutrition criteria;
APACHE II, acute physiology and chronic health evaluation II; BD, biochemical data, medical test and procedures; CH; client history; CRP,
C-reactive protein; FH, food and nutrition-related history; GI, gastrointestinal; ICU, intensive care unit; IL-6, Interleukin-6; NA, nutrition
assessment; NRS-2002, Nutrition Risk Screening-2002; NUTRIC; Nutrition Risk in the Critically Ill; PD, nutrition-focused physical findings;
SGA, Subjective Global Assessment; SOFA, sequential organ failure assessment.
a Moretti et al, 2014.20
b Not required in modified-NUTRIC.

Suggestions for Nutrition Screening and hospital mortality is better (area under the receiver oper-
Assessment for Critically Ill Patients ating characteristics curve [aROC], 0.70) when compared
with either a single tool (aROC for mNUTRIC, 0.66; aROC
The AND has developed the Nutrition Care Process, which for SGA, 0.61).22 When compared with patients who had
includes nutrition assessment, nutrition diagnosis, nutri- low nutrition risk or were not malnourished, an mNUTRIC
tion intervention, and nutrition monitoring and evalua- score of 5 and SGA classification of B or C increased hos-
tion to guide and standardize dietetic practice.51 In the pital mortality by 14.4 times, whereas hospital mortality was
Nutrition Care Process, nutrition assessment consists of 5 increased by 5.3 times and 4.3 times if only the mNUTRIC
components, which are food and nutrition-related history; or SGA were used alone, respectively.22 Another study also
anthropometric data; biochemical data, medical test, and found that patients who were classified as high nutrition
procedures; nutrition-focused physical findings; and client risk (by the mNUTRIC) and malnourished (SGA) had the
history.51 longest hospital and ICU length of stay and were more likely
The variables used by the NUTRIC, NRS-2002, SGA, to require additional rehabilitation after discharge from
and AND-ASPEN as part of the components of nutrition the ICU.52 The improvement of discriminative ability for
assessment are summarized in Table 3. As severity of disease 28-day mortality cannot be demonstrated when the mNU-
plays a major role in the pathophysiology of malnutrition, TRIC was used in combination with the Malnutrition
a clinical component is added to the 5 original components. Universal Screening Tool, whereas the aROC used in com-
From the table, it is observed that the variables for the NU- bination resulted in a lower discriminative ability and was
TRIC score and the SGA are mutually exclusive (variables lower than when the mNUTRIC was used alone (0.679 vs
used by the NUTRIC are not used by the SGA and vice 0.768).15
versa). Therefore, both tools complement each other well for Therefore, a complete nutrition evaluation for a critically
the nutrition risk screening and assessment of critically ill ill patient may include the use of the NUTRIC/mNUTRIC
patients. A recent study in 439 critically ill patients showed score for nutrition risk screening and the SGA and other
that the discriminative ability of a combination of tools on relevant criteria for ICU patients (such as GI tract function
8 Nutrition in Clinical Practice 00(0)

and risk of aspiration) for nutrition assessment. The use of Monitoring of serum phosphorus and potassium level
AND-ASPEN consensus criteria may also be feasible given after the commencement of nutrition support is needed
the almost similar criteria with the SGA. However, the SGA for early detection of potential refeeding syndrome. A
is more favorable as it is well studied in the critically ill pop- validated nutrition monitoring indicator that is responsive
ulation. In addition, the SGA does not require a handgrip to short-term nutrition therapy is still lacking in the critical
dynamometer, which may not be available in a resource- care setting, and further research is warranted.
limited setting. ICU patients may also be unconscious or
too weak to perform test for handgrip strength. In addition, Statement of Authorship
the cut off values of the variables were defined arbitrarily.
Z.-Y. Lee contributed to conception/design of the research;
More studies are needed before a stronger recommendation
Z.-Y. Lee and D. K. Heyland contributed to acquisition,
for its use in the critically ill patients can be made. The incor-
analysis, or interpretation of the data; Z.-Y. Lee drafted
poration of body composition analysis for nutrition assess-
the manuscript; Z.-Y. Lee and D. K. Heyland critically
ment is an added advantage if feasible. The use of simple
revised the manuscript; and Z.-Y. Lee and D. K. Heyland
questionnaires to identify patients with sarcopenia and/or
agree to be fully accountable for ensuring the integrity and
frailty before ICU admission may also help to determine
accuracy of the work. All authors read and approved the
potential patients with low skeletal muscle mass and poor
final manuscript.
nutrition status. The monitoring of the response to nutrition
intervention may be a more difficult task as most of the nu-
trition variables are not responsive enough to assess whether References
nutrition provided is optimal.32 However, electrolyte level 1. Jensen GL, Mirtallo J, Compher C, et al. Adult starvation and
disease-related malnutrition: a proposal for etiology-based diagnosis in
(such as phosphorus and potassium) need to be monitored
the clinical practice setting from the International Consensus Guide-
frequently during the initial stage of nutrition provision line Committee. JPEN J Parenter Enteral Nutr. 2010;34(2):156-159.
to detect potential occurrence of refeeding syndrome so
that calorie intake can be adjusted accordingly.53 Future 2. White J, Guenter P, Jensen G, Malone A, Schofield M. Consen-
studies may monitor the clinical response to treatment such sus statement: Academy of Nutrition and Dietetics and American
Society for Parenteral and Enteral Nutrition: characteristics recom-
as level of inotropes, dose of insulin needed for glycemic
mended for the identification and documentation of adult malnutrition
control, ventilator setting, wound healing, and physical (undernutrition). JPEN J Parenter Enteral Nutr. 2012;36(3):275-283.
functioning as potential indicators for optimal nutrition
therapy. Ferrie and Tsang54 suggested the use of fatigue 3. Jensen GL. Inflammation as the key interface of the medical and
scoring and functional tests (such as handgrip strength and nutrition universes: a provocative examination of the future of clinical
nutrition and medicine. JPEN J Parenter Enteral Nutr. 2006;30(5):453-
forced expiratory volume in 1 second) in conscious patients
and the use of bedside ultrasound for unconscious patients 4. Preiser J-C, Ichai C, Orban J-C, Groeneveld BJ. Metabolic re-
as potential nutrition monitoring indicators in the ICU, sponse to the stress of critical illness. Br J Anaesth. 2014;113(6):1-10.
which remains to be tested in a prospective randomized trial.
It must be noted that currently there is no gold standard for 5. Kondrup J. Nutritional-risk scoring systems in the intensive
care unit. Curr Opin Clin Nutr Metab Care. 2014;17(2):177-182.
the diagnosis of malnutrition, and therefore it is necessary to
exercise appropriate clinical judgment based on individual 6. Jensen GL, Wheeler D. A new approach to defining and diag-
patient responses for the determination of optimal nutrition nosing malnutrition in adult critical illness. Curr Opin Crit Care.
provision. 2012;18(2):206-211.
7. Ziegler TR. Parenteral nutrition in the critically ill patient. N Engl J
Med. 2009;361(11):1088-1097.
Conclusion 8. Lee Z-Y, Noor Airini I, Barakatun-Nisak M-Y. Prevalence and
duration of reasons for enteral nutrition feeding interruption in
The detection of nutrition risk and diagnosis of a tertiary intensive care unit. Nutrition. September 2018;53:26-33.
malnutrition in critically ill patients are not a simple
and straightforward processes. They involve consideration 9. Mueller C, Compher C, Ellen DM. A.S.P.E.N. clinical guidelines:
nutrition screening, assessment, and intervention in adults. JPEN
of the pathophysiology of malnutrition and the use
J Parenter Enteral Nutr. 2011;35(1):16-24.
of accurate and responsive indicators to measure 0148607110389335
nutrition status. Despite their inherent limitations, the 10. Kondrup J, Rasmussen HH, Hamberg O, Stanga Z. Nutritional
NUTRIC/mNUTRIC score and SGA may be the current risk screening (NRS 2002): a new method based on an analysis of
best available tools for the evaluation of nutrition risk and controlled clinical trials. Clin Nutr. 2003;22(3):321-336.
diagnosis of malnutrition, respectively. Other assessment
11. Heyland DK, Dhaliwal R, Jiang X, Day AG. Identifying critically ill
criteria relevant to ICU patients such as GI function, risk patients who benefit the most from nutrition therapy: the development
of aspiration, and the determination of sarcopenia and and initial validation of a novel risk assessment tool. Crit Care.
frailty may also be useful for a more complete assessment. 2011;15(6):R268.
Lee and Heyland 9

12. Johansen N, Kondrup J, Plum LM, et al. Effect of nutritional sup- 26. Wischmeyer PE, Hasselmann M, Kummerlen C, et al. A random-
port on clinical outcome in patients at nutritional risk. Clin Nutr. ized trial of supplemental parenteral nutrition in underweight and
2004;23(4):539-550. overweight critically ill patients: the TOP-UP pilot trial. Crit Care.
13. Starke J, Schneider H, Alteheld B, Stehle P, Meier R. Short-term 2017;21(1):1-14.
individual nutritional care as part of routine clinical setting improves 27. Correia MITD. Nutrition screening vs nutrition assessment: what’s
outcome and quality of life in malnourished medical patients. Clin the difference? Nutr Clin Pr. 2018;33(1):62-72.
Nutr. 2011;30(2):194-201. 0884533617719669
14. Rahman A, Hasan RM, Agarwala R, Martin C, Day AG, Hey- 28. Ceniccola GD, Holanda TP, Pequeno RSF, et al. Relevance of
land DK. Identifying critically-ill patients who will benefit most AND-ASPEN criteria of malnutrition to predict hospital mortality in
from nutritional therapy: further validation of the “modified NU- critically ill patients: a prospective study. J Crit Care. 2018;44:398-403.
TRIC” nutritional risk assessment tool. Clin Nutr. 2016;35(1):158-162. 29. Detsky AS, McLaughlin JR, Baker JP, et al. What is subjective
15. de Vries MCH, Koekkoek WK, Opdam MH, van Blokland D, van Zan- global assessment of nutritional status? JPEN J Parenter Enteral Nutr.
ten ARH. Nutritional assessment of critically ill patients: validation 1987;11(1):8-13.
of the modified NUTRIC score. Eur J Clin Nutr. 2018;72(3):428-435. 30. Lew CCH, Yandell R, Fraser RJL, Chua AP, Chong MFF, Miller M. Association between malnutrition and clinical outcomes in the
16. Mukhopadhyay A, Henry J, Ong V, et al. Association of modified intensive care unit: a systematic review. JPEN J Parenter Enteral Nutr.
NUTRIC score with 28-day mortality in critically ill patients. Clin Nutr. 2017;41(5):744-758.
2017;36(4):1143-1148. 31. Patel C, Omer E, Diamond SJ, McClave SA. Can nutritional as-
17. Compher C, Chittams J, Sammarco T, Nicolo M, Heyland DK. sessment tools predict response to nutritional therapy? Curr Gas-
Greater protein and energy intake may be associated with improved troenterol Rep. 2016;18(4):1-15.
mortality in higher risk critically ill patients: a multicenter, multi- 0488-y
national observational study. Crit Care Med. 2017;45(2):156-163. 32. Ferrie S, Allman-Farinelli M. Commonly used “nutrition” indicators do not predict outcome in the critically ill: a systematic review. Nutr Clin
18. Lee Z-Y, Noor Airini I, Barakatun-Nisak M-Y. Relationship of energy Pr. 2013;28(4):463-484.
and protein adequacy with 60-day mortality in mechanically ventilated 33. Nicolo M, Compher CW, Still C, Huseini M, Dayton S, Jensen GL.
critically ill patients: a prospective observational study. Clin Nutr. Feasibility of accessing data in hospitalized patients to support diagno-
2018;37(4):1264-1270. sis of malnutrition by the Academy-A.S.P.E.N. Malnutrition Consen-
19. Hsu PH, Lee CH, Kuo LK, Kung YC, Chen WJ, Tzeng MS. Higher sus Recommended Clinical Characteristics. JPEN J Parenter Enteral
energy and protein intake from enteral nutrition may reduce hospi- Nutr. 2014;38(8):954-959.
tal mortality in mechanically ventilated critically ill elderly patients 34. Reintam Blaser A, Poeze M, Malbrain ML, Bjorck M, Oudemans-
[published online ahead of print March 26, 2018]. Int J Gerontol. van Straaten HM, Starkopf J. Gastrointestinal symptoms during the first week of intensive care are associated with poor outcome: a
20. Moretti D, Bagilet DH, Buncuga M, Settecase CJ, Quaglino MB, prospective multicentre study. Intensive Care Med. 2013;39(5):899-909.
Quintana R. Study of two variants of nutritional risk score “NU-
TRIC” in ventilated critical patients. Nutr Hosp. 2014;29(1):166-172. 35. McClave SA, Demeo MT, Delegge MH, et al. North American Summit on aspiration in the critically ill patient: consensus statement. JPEN J
21. Mendes R, Policarpo S, Fortuna P, Alves M, Virella D, Heyland Parenter Enteral Nutr. 2002;26(6 suppl):S80-S85.
DK. Nutritional risk assessment and cultural validation of the mod- 36. Earthman CP. Body composition tools for assessment of adult malnu-
ified NUTRIC score in critically ill patients—a multicenter prospec- trition at the bedside. JPEN J Parenter Enteral Nutr. 2015;39(7):787-
tive cohort study. J Crit Care. February 2017;37:45-49. https://doi. 822.
org/10.1016/j.jcrc.2016.08.001 37. Moisey LL, Mourtzakis M, Cotton BA, et al. Skeletal muscle
22. Lew CCH, Cheung KP, Chong MFF, Chua AP, Fraser RJL, predicts ventilator-free days, ICU-free days, and mortality in el-
Miller M. Combining 2 commonly adopted nutrition instruments derly ICU patients. Crit Care. 2013;17(5):1-8.
in the critical care setting is superior to administering either one cc12901
alone. JPEN J Parenter Enteral Nutr. 2018;42(5):872-876. https://doi. 38. Weijs PJ, Looijaard WG, Dekker IM, et al. Low skeletal muscle area is a
org/10.1177/0148607117726060 risk factor for mortality in mechanically ventilated critically ill patients.
23. Kalaiselvan MS, Renuka MK, Arunkumar AS. Use of nutri- Crit Care. 2014;18(1):R12.
tion risk in critically ill (NUTRIC) score to assess nutritional 39. Looijaard WGPM, Dekker IM, Stapel SN, et al. Skeletal muscle quality
risk in mechanically ventilated patients: a prospective observational as assessed by CT-derived skeletal muscle density is associated with 6-
study. Indian J Crit Care Med. 2017;21(5):253-256. https://doi. month mortality in mechanically ventilated critically ill patients. Crit
org/10.4103/ijccm.IJCCM_24_17 Care. 2016;20(1):1-10.
24. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for 40. Paris MT, Mourtzakis M, Day A, et al. Validation of bedside ul-
the provision and assessment of nutrition support therapy in trasound of muscle layer thickness of the quadriceps in the criti-
the adult critically ill patient: Society of Critical Care Medicine cally ill patient (VALIDUM Study). JPEN J Parenter Enteral Nutr.
(SCCM) and American Society for Parenteral and Enteral Nutri- 2017;41(2):171-180.
tion (A.S.P.E.N.). JPEN J Parenter Enteral Nutr. 2016;40(2):159-211. 41. Fivez T, Hendrickx A, Van Herpe T, et al. An analysis of reliability and accuracy of muscle thickness ultrasonography in critically ill
25. Arabi YM, Aldawood AS, Al-Dorzi HM, et al. Permissive under- children and adults. JPEN J Parenter Enteral Nutr. 2016;40(7):944-949.
feeding or standard enteral feeding in high and low nutritional risk
critically ill adults: post-hoc analysis of the PermiT trial. Am J 42. Puthucheary ZA, Rawal J, McPhail M, et al. Acute skeletal muscle
Respir Crit Care Med. 2017;195(5):652-662. wasting in critical illness. JAMA. 2013;310(15):1591-1600. https://doi.
rccm.201605-1012OC org/10.1001/jama.2013.278481
10 Nutrition in Clinical Practice 00(0)

43. Thibault R, Makhlouf A-M, Mulliez A, et al. Fat-free mass at 49. Rockwood K, Song X, MacKnight C, et al. A global clinical measure
admission predicts 28-day mortality in intensive care unit pa- of fitness and frailty in elderly people. CMAJ. 2005;173(5):489-495.
tients: the international prospective observational study Phase Angle
Project. Intensive Care Med. 2016;42(9):1445-1453. 50. Muscedere J, Waters B, Varambally A, et al. The impact of frailty
10.1007/s00134-016-4468-3 on intensive care unit outcomes: a systematic review and meta-
44. Kuchnia A, Earthman C, Teigen L, et al. Evaluation of analysis. Intensive Care Med. 2017;43(8):1105-1122.
bioelectrical impedance analysis in critically ill patients. JPEN 10.1007/s00134-017-4867-0
J Parenter Enteral Nutr. 2017;41(7):1131-1138. 51. Writing Group of the Nutrition Care Process/Standardized Language
10.1177/0148607116651063 Committee. Nutrition Care Process Part II: using the International
45. Malmstrom TK, Morley JE. SARC-F: a simple questionnaire to Dietetics and Nutrition Terminology to Document the Nutrition
rapidly diagnose sarcopenia. J Am Med Dir Assoc. 2013;14(8):531-532. Care Process. J Am Diet Assoc. 2008;108(8):1287-1293. 10.1016/j.jada.2008.06.368
46. Malmstrom TK, Miller DK, Simonsick EM, Ferrucci L, Morley JE. 52. Coltman A, Peterson S, Roehl K, Roosevelt H, Sowa D. Use of
SARC-F: a symptom score to predict persons with sarcopenia at risk for 3 tools to assess nutrition risk in the intensive care unit. JPEN
poor functional outcomes. J Cachexia Sarcopenia Muscle. 2016;7(1):28- J Parenter Enteral Nutr. 2015;39(1):28-33.
36. 0148607114532135
47. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, et al. Sarcopenia: European 53. Doig GS, Simpson F, Heighes PT, et al. Restricted versus continued
consensus on definition and diagnosis. Age Ageing. 2010;39(4):412-423. standard caloric intake during the management of refeeding syn- drome in critically ill adults: a randomised, parallel-group, multicentre,
48. Morley JE, Anker SD, von Haehling S. Prevalence, incidence, and clin- single-blind controlled trial. Lancet Respir Med. 2015;3(12):943-952.
ical impact of sarcopenia: facts, numbers, and epidemiology—update
2014. J Cachexia Sarcopenia Muscle. 2014;5(4):253-259. https://doi. 54. Ferrie S, Tsang E. Monitoring nutrition in critical illness: what
org/10.1007/s13539-014-0161-y can we use? Nutr Clin Pr. 2018;33(1):133-146.
Lee and Heyland 11


1) NRS-2002

(a) Simple screening:

i. Is BMI <20.5?
ii. Has intake been reduced during the last week?
iii. Has there been a recent weight loss?
iv. Is the patient severely ill?

If the answer is yes to any of these four questions, the formal screening in the table is carried out

Impaired nutritional status Severity of Illness (stress metabolism)

Absent Score 0 Normal nutritional status Absent Score 0 Normal nutritional requirements

Mild Score 1 • Weight loss >5% in 3 months Mild Score 1 • Hip fracture
or • Chronic patients in particular
• Food intake below 50–75% of normal with acute complications:
requirement in preceding week cirrhosis, COPD
• Chronic haemodialysis
• Diabetes
• Oncology
Moderate Score 2 • Weight loss >5% in 2 months Moderate Score 2 • Major abdominal surgery.
or • Stroke
• BMI 18.5 – 20.5 + impaired general • Severe pneumonia
condition • Hematologic malignancy
• Food intake below 25 – 50% of normal
requirement in preceding week
Severe Score 3 • Weight loss >5% in 1 month Severe Score 3 • Head injury
( > 15% in 3 months) • Bone marrow transplantation
or • Intensive care patients
• BMI <18.5 + impaired general (APACHE 10)
• Food intake below 0 – 25% of normal
requirement in preceding week
Score Score
Total Score

(b) Calculate the total score:

i. Find score (0–3) for Impaired nutritional status (only one: choose the variable with highest score) and Severity of
disease (stress metabolism, i.e. increase in nutritional requirements).
ii. Add the two scores (→ total score)
iii. If age 70 years: add 1 to the total score to correct for frailty of elderly
iv. If age-corrected total 3: start nutritional support
12 Nutrition in Clinical Practice 00(0)

2) The NUTRIC and the Modified NUTRIC Score

Variable Range Points

Age (year) <50 0

50 – 74 1
75 2
Acute physiology and chronic health evaluation II (APACHE II) <15 0
15 – 19 1
20 – 28 2
28 3
Sequential organ-failure assessment (SOFA) <6 0
6–9 1
10 2
Number of Comorbidities 0–1 0
2 1
Days from hospital to ICU admission 0 – <1 0
1 1
Interleukin-6 (IL-6) (pg/ml) 0 – <400 0
400 1

NUTRIC Score Scoring System (With IL-6)

Sum of points Category Explanation

6 – 10 High Score • Associated with worse clinical outcome (mortality, ventilation)

• These patients are the most likely to benefit from aggressive nutrition therapy
0–5 Low Score • These patients have a low malnutrition risk

Modified-NUTRIC Score Scoring System (Without IL-6)

Sum of points Category Explanation

5–9 High Score • Associated with worse clinical outcome (mortality, ventilation)
• These patients are the most likely to benefit from aggressive nutrition therapy
0–4 Low Score • These patients have a low malnutrition risk

(Adapted from Accessed on 1st April, 2018)

Lee and Heyland 13

3) The Academy of Nutrition and Dietetics and the American Society of Parenteral and Enteral Nutrition (AND-ASPEN)
Consensus for the Diagnosis of Malnutrition in the Context of Acute Illness or Injury

Clinical Characteristics Moderate Severe

<75% of estimated ER for 50% of estimated ER for

1) Energy Intake >7days 5 days
2) Weight change over time (note for under- or % Time % Time
overhydration) 1–2 1 week >2 1 week
*Usual weight should be used as the baseline weight 5 1 month >5 1 month
7.5 3 months >7.5 3 months
3) Loss of subcutaneous fat (eg, orbital, triceps, fat Mild Moderate
overlying the ribs)
4) Muscle Loss (eg, wasting of the temples [temporalis Mild Moderate
muscle], clavicles [pectoralis and deltoids], shoulders
[deltoids], interosseous muscles, scapula [latissimus
dorsi, trapezious, deltoids], thigh [quadriceps], and
calf [gastrocnemius])
5) Fluid Accumulation (Generalized or localized Mild Moderate to Severe
[extremities, vulvar/scrotal edema, or ascites])
6) Reduced grip strength (Refer normative standard by NA Measurably reduced
the device manufacturer)

A minimum of 2 of the 6 characteristics above is recommended for diagnosis of either severe or nonsevere malnutrition
14 Nutrition in Clinical Practice 00(0)

4) Subjective Global Assessment (SGA)

A. History

1. Weight change
Overall loss in past 6 months: Amount = kg
% loss = kg
Change in past 2 weeks: Increase
No change
2 Dietary Intake change (relative to normal)
No Change
Weeks =
Suboptimal solid diet
Hypocaloric liquids
Full liquid
3 Gastrointestinal symptoms (that persisted for >2 weeks)
4 Functional Capacity
No Dysfunction
Weeks =
Working suboptimally
5 Disease and its relation to nutritional requirements
Primary Diagnosis (Specify):
Metabolic Demand (Stress)
No stress
Low stress
Moderate stress
High stress

B. Physical (for each trait specify: 0 = normal, 1+ = mild 2+ = moderate, 3+ = severe)

Loss of subcutaneous fat (triceps, biceps, under the eyes)
Muscle wasting (temple, clavicle, shoulder, scapula/ribs, quadriceps, calf, knee, interosseous muscle)
Ankle edema
Sacral edema

C. SGA rating (select one)

Well Nourished (A)
Moderately (or suspected of being) malnourished (B)
Severely Malnourished (C)
Lee and Heyland 15


Component Question Scoring

Strength How much difficulty do you have in lifting and carrying 10 pounds None = 0
(4.5 kilogram) ? Some = 1
A lot or unable = 2
Assistance in walking How much difficulty do you have walking across a room? None = 0
Some = 1
A lot, use aids, or unable = 2
Rise from a chair How much difficulty do you have transferring from a chair or bed? None = 0
Some = 1
A lot or unable without help = 2
Climb stairs How much difficulty do you have climbing a flight of 10 stairs? None = 0
Some = 1
A lot or unable = 2
Falls How many times have you fallen in the past year? None = 0
1–3 falls = 1
4 or more falls = 2
16 Nutrition in Clinical Practice 00(0)

6) Clinical Frailty Scale

Please consider the participant’s overall condition 2 weeks prior to this admission to hospital.
How fit or frail was s/he at that time point? Check one response only.
If you have trouble deciding between two options, choose the higher functioning level.