Clinical Nutrition 39 (2020) 3813e3820

Contents lists available at ScienceDirect

Clinical Nutrition
journal homepage: http://www.elsevier.com/locate/clnu

Original article

Nutritional therapy among burn injured patients in the critical care

setting: An international multicenter observational study on “best
achievable” practices
Michail Chourdakis a, *, Emmanouil Bouras a, Beth A. Shields b, Christian Stoppe c,
Anne-Françoise Rousseau d, Daren K. Heyland e, f, **
a
School of Medicine, Faculty of Health Sciences, Aristotle University of Thessaloniki, Thessaloniki, Greece
b
Department of Nutritional Medicine and Metabolism, Burn Unit, U.S. Army Institute of Surgical Research, Ft. Sam Houston, TX, USA
c
Department of Intensive Care Medicine, University Hospital, RWTH Aachen, Germany
d
Intensive Care Department and Burn Center, University Hospital, Li ege, Belgium
e
Department of Critical Care Medicine, Queen's University, Kingston, Ontario, Canada
f
Clinical Evaluation Research Unit, Kingston General Hospital, Kingston, Ontario, Canada

a r t i c l e i n f o s u m m a r y

Article history: Background & aims: Burn patients pose a number of clinical challenges for doctors and dietitians to
Received 29 January 2020 achieve optimal nutrition practice. The objective of this study was to describe nutrition practices in burn
Accepted 10 April 2020 center intensive care units (ICUs) compared to the most recent ESPEN and SCCM/ASPEN guidelines
(hereafter referenced as “the Guidelines”) and highlight the variation in practice and what is “best
Keywords: achievable.”
Mortality
Methods: In 2014e15, we prospectively enrolled 283 mechanically ventilated patients who were
ICU
admitted to one of 14 burn ICUs for at least 72 h. Data collected included information on the estimation
Nutrition therapy
Micronutrients
of energy and protein requirements, their actual delivery as well as route and time of feeding, and
administration of micronutrients. We describe site practices and data per patient-day.
Results: Adherence to the Guidelines for the use of enteral nutrition (EN) over parenteral nutrition (PN)
was 90.5% of patient-days (site range 79.2%e97.0%). However, adherence to the Guidelines for the
measurement of energy requirements was 6.0% of patient-days (site range 0.0%e93.3%), supplementa-
tion with glutamine took place in 22.4% of patient-days (site range 0.0%e61.8%). Provision of 80% of
energy requirements within 48e72 h was achieved in 35.3% of patients (site range 0.0%e80.0%), and
provision of 80% of protein needs within 48e72 h was achieved in 34.3% of patients (site range 0.0%
e80.0%). Average nutritional adequacy was 64.9 ± 40.0% for energy (best site: 80.2%, worst site: 42.0%)
and 65.6 ± 42.1% for protein (best site: 87.3%, worst site: 43.6%).
Conclusion: The present findings indicate that despite high adherence to providing EN over PN, there is
still a large gap between many recommendations and clinical practice, and the achievement of nutrition
goals for patients in burn centers is suboptimal.
© 2020 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

1. Introduction hypermetabolic response, increased losses and subsequent needs,

and altered and dysregulated glucose metabolism [1e6], nutri-
Patients with severe burns pose several nutritional challenges tional therapy is a critical component of severe burn treatment, as
for both physicians and dietitians. Due to the perilous post-burn well as an important outcome effector [1]. The general framework
of nutritional support is based on management of the hyper-
* Corresponding author. University Campus, Aristotle University of Thessaloniki, catabolic state, glycemic control, micronutrients supplementation,
School of Medicine, Building 16A, Entrance 4, 54124, Thessaloniki, Greece. and of course adequate provision of energy and nutrients, with
** Corresponding author. Department of Critical Care Medicine, Queen's Univer- prompt feeding initiation via a tolerable and effective route [1e6].
sity, Kingston, Ontario, Canada.
E-mail addresses: [email protected] (M. Chourdakis), [email protected]
Several professional organizations, including the European So-
(D.K. Heyland). ciety for Clinical Nutrition and Metabolism (ESPEN) and the Society

https://doi.org/10.1016/j.clnu.2020.04.023
0261-5614/© 2020 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
3814 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820

of Critical Care Medicine (SCCM) with the American Society for Body mass index (BMI) was calculated as body weight (kg)
Parenteral and Enteral Nutrition (ASPEN), have published recom- divided by the square of height (m2). Upon admission, the Acute
mendations for nutrition for critically ill burn patients [2,5], albeit Physiology and Chronic Health Evaluation (APACHE) II [10] and the
given recommendations are based on low evidence levels Sequential Organ Failure Assessment (SOFA) [11,12] scores were
(Supplementary Table 1). ESPEN and SCCM/ASPEN guidelines calculated, and the percent of total body surface area (%TBSA) burn
(hereafter referenced as “the Guidelines”) agree in their recom- was determined based on conventional surface estimation
mendations for enteral nutrition (EN) be utilized for patients with a methods [13]. The abbreviated burn severity index (ABSI) [14], a
functional gastrointestinal tract and who are unable to meet fiveeitem scale used to assess burn severity, was also calculated.
nutrition goals with oral intake alone, reserving parenteral nutri- The ABSI includes sex, age, presence of inhalation injury, presence
tion (PN) for those who are unable to tolerate or have a contrain- of a full-thickness burn, and %TBSA. Nutritional risk was deter-
dication to EN; early initiation of EN (ESPEN: within 12 h from mined with the modified NUTrition Risk in the Critically ill
injury, SCCM/ASPEN: within 6 h from injury); the use of indirect (mNUTRIC) tool [15]. Glucose monitoring; insulin and feeding
calorimetry (IC) for the most accurate determination of energy protocols were followed as usual in each center. Data management
goals; protein goals of 1.5e2.0 g/kg/d (“or higher” per SCCM/ has been previously described [7].The research ethics committee at
ASPEN); supplementation of vitamin C, zinc, and selenium; and the lead site, Queens University, approved the INS. The need for
glucose control (ESPEN: glucose levels between 81 and 144 mg/dL, informed consent was waived as this was an observational study,
SCCM/ASPEN: glucose levels between 140 or 150e180 mg/dL). without interventions.
ESPEN does not address the timing for achieving nutrition goals,
but SCCM/ASPEN suggest that 80% of the energy and protein goals 2.3. Statistical analyses
should be met within 48e72 h. ESPEN recommends glutamine
supplementation; however, SCCM/ASPEN recommends against. Categorical variables were described as counts and percentages,
Evaluating compliance with nutrition guidelines is important and continuous variables were reported as means with standard
from the point of view of obtaining optimal patient outcomes. deviations and/or medians with interquartile range (IQR),
Failure to translate this knowledge into practice may result in depending on their distribution. Variables related to nutritional/
increased length of stay (LOS), but also morbidity and mortality for clinical guidelines were reported as overall averages (or percentage
thousands of critically ill patients [7,8]. The present multicenter of all patient-days across sites), along with the range of site aver-
prospective study aimed to record nutritional strategies for patients ages (average result across all patients within an individual site).
admitted to burn center ICUs, and to examine adherence to the The site ranges are interpreted as the best and worst site perfor-
Guidelines. We compared actual practices among burn centers in mances relative to the nutritional/clinical guidelines. The calorie
an effort to define or establish the “best achievable” performance in and protein intakes via the enteral or parenteral route were divided
key nutritional practices. by the amount prescribed in the baseline assessment to determine
percentage of the nutrition prescription achieved. Calories from
2. Methods propofol infusions (if > 6 h) were included in the calculation
whereas calories from intravenous (IV) dextrose were not. Nutri-
2.1. Study design and participants tional intake from oral nutrition was not included in the calculation
of nutritional adequacy. Days after permanent progression to
Burn Intensive Care Units (ICUs) were recruited during the In- exclusive oral intake were excluded from the calculation of nutri-
ternational Nutrition Survey (INS) in 2014 and 2015 for this pro- tional adequacy. We selected >80% nutritional adequacy as an in-
spective, observational study on nutrition practices. Patients were dicator of high performance, based on previous publication from
consecutively enrolled in the study when they met the following members of the group [16].
inclusion criteria: 1) age
18 years old, or
16 years old if approved All analyses were performed using IBM SPSS Statistics version
locally at site, 2) on mechanical ventilation within 48 h of admission 23.0 (International Business Machines Corp., Armonk, New York,
to the ICU, and 3) length of ICU stay
72 h from admission. United States) and SAS® 9.4 (SAS® Institute Inc., Cary, North Car-
olina, USA), and the level of significance was set at p < 0.05.
2.2. Data collection
3. Results
For each patient included in the INS, sites collected data
describing patient characteristics, ICU admission information, A total of 14 burn centers registered to participate in this survey,
baseline nutrition assessment, daily nutrition data, and 60-day including 11 in the USA, one in Canada, one in South Africa, and one
patient outcomes. As per our usual INS practices [7,9], and given in Australia, enrolling 283 patients for 3085 total patient-days, with
the observational nature of this study, baseline nutrition assess- an average of 20.2 ± 6.7 patients per site (range: 15e41). Table 1
ment was not standardized across sites. Data on total calories and describes the ICU characteristics of the participating sites, and
protein prescribed and received were recorded. For this, “total Table 2 shows the patient demographics. Table 3 summarizes the
calories and protein prescribed” referred to the total as determined adherence of our sample to the Guidelines.
at the initial assessment, whereas “total calories and protein
received” referred to the total received by EN or PN plus propofol, 3.1. Feeding route
according to each center's standard protocol. Daily nutrition data,
which included the initial feeding strategy and type and amount of Guidelines recommend that EN be used in all critically ill pa-
nutrition received, were collected from ICU admission until ICU tients with an intact gastrointestinal tract, and PN should not be
discharge or death, or for a maximum of 12 days. Oral intake was used routinely. Figure 1 displays the percentage of patient-days
not included. Patient outcomes at day 60 were collected and where EN and PN were provided. EN was the predominated
included the date mechanical ventilation was discontinued, the feeding route provided on 90.5% of the patient-days, ranging from
dates of ICU and hospital discharge, and the date of death (if rele- 79.2% to 97.0% between worst and best sites, respectively. Oral
vant). Data were abstracted from patient records and were entered feeding occurred during 18.2% of total patient-days, with 11.8%
online using a secure web-based data collection tool. patient-days including oral supplement drinks. Neither EN nor PN
 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820 3815

Table 1 3.3. Energy requirements

Characteristics of participating centers.

Total, n ¼ 14 IC was used to determine energy requirements for 6% of pa-

Geographic region Australia 1 (7.1%)
tients; the best site used IC for 93.3% of patients and the worst did
Canada 1 (7.1%) not use it at all. The most popular method for estimating energy
USA 11 (78.6%) requirements was based on body weight (22.3%), being followed by
South Africa 1 (7.1%) the HarriseBenedict equation [17] (16.3%) and the Milner formula
Hospital type Teaching 13 (92.9%)
(12.4%), whereas for the rest, several other ways were used
Non-teaching 1 (7.1%)
Size of hospital (beds) 462.5 (280e1004) (Supplementary Table 3). Actual dry body weight was used for the
Median (range) estimation of energy requirements in the majority of patients
ICU type Closed 4 (28.6%) (55.5%), followed by the use of estimated dry body weight in 21.9%
Open 7 (50.0%)
of the cases.
Other 3 (21.4%)
Multiple ICUs in the hospital Yes 13 (92.9%)
Medical Director Yes 13 (92.9%) 3.4. Meeting energy needs
Size of ICU (beds) 12.5 (4e20)
Median (range) Overall, on average the included 283 patients were prescribed
Presence of dietitian(s) Yes 14 (100%)
2648.4 ± 738.8 kcal/d or 32.3 ± 9.2 kcal/kg/d and received
Full time equivalent dietitians 0.5 (0.2e3.0)
(per 10 beds) Median (range) 64.9 ± 40.0% of this goal over the 12 days of observation (Table 4).
Figure 2 illustrates the actual received average, best and worst
ICU: Intensive Care Unit.
performances in regard to the prescribed calories across the 12 days
of observation (site average: 65.3%; 80.2% for best and 42.0% for
were provided for 9.5% of patient-days (site range: 3.0%e20.8% of worst site).
patient-days), whereas there were no days with PN alone and Provision of at least 80% of energy needs was achieved within
PN þ EN accumulated for less than 2% of patient-days. 48e72 h per the SCCM/ASPEN guidelines in 35.3% of patients, with
the best site achieving this in 80.0% of patients and the worst
completely failing in this (0.0%).
3.2. EN initiation In 39.2% of patient-days, EN was held or interrupted; reasons
can be seen in Table 5. Main reasons that led to EN interruptions
The average time to EN initiation was 22.4 ± 20.1 h from ICU were either an operating room (33.0%) or bedside (25.8%) proced-
admission. The best performing site was able to initiate EN ure, whereas EN intolerance was reported in 14.8% of cases.
4.9 ± 4.0 h after ICU admission; the worst site delayed initiation of
EN for 47.5 ± 23.0 h, on average. EN was initiated within 6 and 12 h 3.5. Protein requirements
from admission in 18.7% and 35.0% of patients, respectively. The
best site achieved EN initiation within 6 h from admission in 70.0% Protein requirements were calculated based on each patient's
of patients, whereas the worst site had no success (0% of patients) actual dry body weight for 42.6% of the participants, followed by
in this. The best site achieved EN initiation within 12 from admis- Hamwi's [18] ideal body weight formula in 19.8% of the sample. The
sion for 90.0% of patients, whereas the worst site had 5% success in range of 1.5e2.0 g/kg/d was prescribed in 33.9% of patients. The
this parameter. More information on the timing of initiation of EN protein prescription was higher than 2.0 g/kg/d for 31.1% of pa-
can be found in Supplementary Table 2. tients, and 35.0% of patients were prescribed less than 1.5 g/kg/d.

3.6. Meeting protein needs

Table 2
Patient characteristics.
Overall, participating sites prescribed 149.1 ± 52.1 g of protein
Total, n ¼ 283
per day, which translated to 1.81 ± 0.61 g protein/kg/d and the
Number of patients, per site Mean (SD) 20.2 (6.7) average adequacy of total protein over the 12 days of observation
Age, yrs Mean (SD) 46.6 (17.8) was 65.6 ± 42.1% (Table 4). When examining the amount of protein
Gender, Male/Female 190 (67.1%)/93 (32.9%)
BMI, kg/m2 Mean (SD) 28.5 (7.8)
received, 19.2% of the patient-days had actual intake within
mNUTRIC score Mean (SD) 4.0 (1.6) 1.5e2.0 g protein/kg/d, with the best site achieving this for 39.8%
APACHE II score Mean (SD) 19.0 (8.1) and the worst for 7.9% of their patient-days (Table 3). Figure 3 il-
SOFA score Mean (SD) 15.0 (5.0) lustrates the actual received average and the performance of the
TBSA, % Mean (SD) 30.0 (21.5)
best and worst sites during the 12 days of observation (site average:
ABSI Mean (SD) 7.5 (2.4)
Presence of ARDS 30 (10.6%) 65.4%; 87.3% for best and 43.6% for worst site). Provision of at least
Length of mechanical ventilation, days 80% of needs was achieved within 48e72 h per the SCCM/ASPEN
Mean (SD) 17.2 (18.1) guidelines in 34.3% of patients (80.0% for best and 0.0% for worst
Median (IQR) 10.5 (3.9, 24.0) site).
Length of ICU stay, days
Mean (SD) 28.9 (20.7)
Median (IQR) 22.7 (11.2, 48.2) 3.7. Micronutrient supplementation
Length of hospital stay, days
Mean (SD) 35.4 (20.2) Glutamine was administered in 22.4% of patient-days (highest
Median (IQR) 30.9 (16.4, 61.0)
site 61.8%, lowest site 0.0). Vitamin C was supplemented for 61.7% of
Number of Patient died within 60 days 54 (19.1%)
patient-days (best site: 98.7%, worst site: 2.4%). Zinc was supple-
ABSI: Abbreviated Burn Severity Index; APACHE II: Acute Physiology and Chronic mented for 44.2% of patient-days (best site: 97.0%, worst site: 0.0%).
Health Evaluation II; ARDS: Acute Respiratory Distress Syndrome; BMI: Body Mass
Index; ICU: Intensive Care Unit; IQR: Interquartile Range; mNUTRIC: modified
Selenium was supplemented for 24.8% of patient-days (best site:
Nutrition Risk in Critically Ill; SD: Standard Deviation; SOFA: Sequential Organ 82.3%, worst site: 0.0%). Provisions were low for IV Vitamin C (17.5%
Failure Assessment score; TBSA: Total Burn Surface Area. of patient-days, highest site 98.7%, lowest site 0.0%), zinc (15.5% of
3816 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820

Table 3
Summary of findings regarding adherence to the ESPEN and SCCM/ASPEN guidelines for the nutritional management of severe burn patients.

Recommendation Society Guidelines Overall adherence

ESPEN SCCM/ASPEN Overall Patient-Days Site average Site Range

Best site Worst site

Feeding route Suggested feeding route: EN yes yes 90.5% 90.4% 97.0% 79.2%
EN initiation Within 12 h from admission yes e 35% 36.8% 90.0% 5.0%
Within 4-6 h from admission e yes 18.7% 20.1% 70.0% 0.0%
Energy needs Measurement of Requirements: per indirect calorimetry yes yes 6.0% 7.7% 93.3% 0.0%
Time point for meeting needs: 80% of goal energy within 48-72h e yes 35.3% 37.7% 80.0% 0.0%
Protein needs Received 1.5-2.0 g/kg/d yes yes 19.2% 19.2% 39.8% 7.9%
Time point for meeting needs: 80% of goal protein within 48-72h e yes 34.3% 36.6% 80.0% 0.0%
Received 1.5-2.0 g/kg/d “or higher” e yes 35.6% 33.5% 61.1% 15.2%
Supplementation Glutamine, either route: 0.3 g/kg/d yes e 22.4% 18.5% 61.8% 0.0%
Glutamine: EN/PO e e 22.4% 18.5% 61.8% 0.0%
Glutamine: IV e e <0.1% <0.1% N/A N/A
Vitamin C: either route yes yes 61.7% 57.7% 98.7% 2.4%
Vitamin C: EN/PO e e 45.4% 42.7% 83.7% 0.0%
Vitamin C: IV e e 17.5% 16.0% 98.7% 0.0%
Zn: either route yes yes 44.2% 41.6% 97.0% 0.0%
Zn: EN/PO e e 31.7% 30.0% 76.7% 0.0%
Zn: IV e e 15.5% 13.9% 97.0% 0.0%
Se: either route yes yes 24.8% 19.9% 82.3% 0.0%
Se: EN/PO e e 12.4% 8.6% 50.8% 0.0%
Se: IV e e 14.5% 12.9% 82.3% 0.0%
No recommendation on Probiotics e e 2.6% 3.1% 34.4% 0.0%
Glucose control Glucose levels: 4.5-8 mmol/L ¼ 81e144 mg/dL yes e 57.1% 56.8% 68.0% 48.5%
Glucose levels: 7.8 or 8.3-10 mmol/L ¼ 140 or 150e180 mg/dL e yes 24.2% 24.1% 31.7% 15.0%

EN: Enteral Nutrition; ESPEN: European Society for Clinical Nutrition and Metabolism; IV: Intravenous; PO: Per Os; Se: Selenium; SCCM/ASPEN: Society of Critical Care
Medicine/American Society for Parenteral and Enteral Nutrition; Zn: Zinc.

Fig. 1. Lowest edge of error bars indicate the worst performance while the highest indicates best performance. ‘None’ refers to number of patients with neither EN nor PN,
regardless of oral intake. The best achievable practices are circled.

patient-days, highest site 97.0%, lowest site 0.0%), selenium (14.5% supplementary provision of pharmaconutrients can be seen in
of patient-days, highest site 82.3, lowest site 0.0%), and glutamine Supplementary Table 4.
(<0.1% of patient-days). High intakes of IV vitamin C (>10 mg/kg/
day) were achieved in a total of 5% of patient-days, while levels of 3.8. Glucose control
>60 mg/kg/day were only reached on one patient-day.
Finally, despite the absence of specific information available for The daily morning glucose level was 144.5 ± 40.9 mg/dL (site
the use of probiotics, in 2.6% of patient-days such was used, with a average range, 134.4e151.5 mg/dL), with 42.1% of days higher than
range of 0e34.4% for all sites. Overall, detailed information on the 144 mg/dL and 12.5% of days higher than 180 mg/dL. ESPEN's
 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820 3817

Table 4
Prescribed (goal) versus received energy and protein.

Targeted, mean (SD) Received, mean (SD) Site average Best site Worst site

n ¼ 283 patients n¼2849 patient-days n¼14 sites

Energy requirements, kcal/d 2648.4 (738.8) 1717.9 (1149.0) 1657.9 2418.4 1145.1
Protein requirements, g/d 149.1 (52.1) 98.0 (72.7) 92.7 159.5 68.5
Energy per kg of BW, kcal/kg/d 32.3 (9.2) 21.1 (14.5) 20.7 29.0 14.1
Protein per kg of BW, kcal/kg/d 1.81 (0.61) 1.2 (0.89) 1.15 1.79 0.83
Nutritional adequacy for energy, % N/A 64.9 (40.0) 65.3 80.2 42.0
Nutritional adequacy for protein, % N/A 65.6 (42.1) 65.4 87.3 43.6

BW: Body Weight; SD: Standard Deviation.

Fig. 2. The amount of calories received by EN and PN as a percentage of the calories prescribed at baseline assessment in all patients.

recommendation to keep glucose levels between 81 and 144 mg/dL

Table 5
was met for 57.1% of patient-days, as measured by the morning Reasons Enteral Nutrition Feeds interrupted.
glucose level (best site: 68.0%, worst site: 48.5%). SCCM/ASPEN's
Total (%) n ¼ 999
recommendation to keep glucose levels between 140 or
(39.2%)c
150e180 mg/dL was adhered to for 24.2% of patient-days (best site:
Number of patient-days
31.7%, worst site: 15.0%). Insulin was administered in a total of 1347
Fasting for operating room procedure 330 (33.0%)
patient-days, of which 223 (16.6%, site range: 0.0%e41.3%) reported Fasting for bedside procedure 258 (25.8%)
more than 96 units/day. EN Intolerancea 148 (14.8%)
No enteral access available/enteral access lost, 80 (8.0%)
displaced or malfunctioning
4. Discussion Trial of oral intake 73 (7.3%)
Inotropes, vasopressor requirement 27 (2.7%)
We conducted a prospective, observational study in 14 different Dressing Change 21 (2.1%)
ICUs to compare nutrition practices among burn units and deter- Subject deemed too sick to continue enteral feeding 18 (1.8%)
Fasting for radiology suite procedure 18 (1.8%)
mine the “best achievable” nutrition practices across participating Fasting for administration of medications 11 (1.1%)
sites. We observed large gaps between most of the Guidelines Otherb 15 (1.5%)
examined and current practices achieved.
CPAP: Continuous Positive Airway Pressure; EN: Enteral Nutrition; GI:
According to the Guidelines [2,3,5], EN is the preferred feeding Gastrointestinal.
route for burn patients, conferring several advantages over PN. In a
Increased gastric residuals (n ¼ 60.8%), vomiting/emesis (n ¼ 19.6%), increased
particular, early EN commencement is associated with reduced abdominal girth or abdominal distension (n ¼ 10.1%), subjective discomfort
infection [19], cost [20], and length of stay [21], shortened hyper- (n ¼ 7.4%), diarrhea or other (n ¼ 2.1%).
b
Including GI bleeding (2), Patient agitation (2), Unknown (2), Acute arterial
metabolic phase [22e24], protection of the gastrointestinal track bleeding (1), CPAP/vent weaning (1), Hydrotherapy (shower table) (1), Patient on
[25], and improved nutrient adequacy [26]. This superiority of EN CPAP (1).
c
appears to be acknowledged by the majority involved in the In a total of 2550 patient-days.
3818 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820

Fig. 3. The amount of protein received by EN and PN as a percentage of the protein prescribed at baseline assessment in all patients.

nutrition support provided by burn centers and was used for 90.5% achieve this quality indicator in at least 80% of patients suggests
of patient-days. that systematic improvements in nutritional delivery is possible.
SCCM/ASPEN [2,5] recommends initiating EN within 4e6 h of The concept of Volume Based Feedings, as recently presented by
injury. Іn our sample EN was initiated within 4e6 h of admission the PEPuP protocol, seems to increase the probability of meeting
for less than 1 out of every 5 patients. The best site was able to the nutritional goals (in both energy and protein) in a safe way, also
achieve EN initiation within 4e6 h of admission in 70% of patients, related to less episodes of hyperglycemia and lower use of PN [32].
indicating that this may be one of the more difficult guidelines to The Guidelines suggest the use of supplementation for vitamin
follow, as the recommendation is initiating EN within 4e6 h of C, zinc, and selenium among burn patients. ESPEN recommends
injury, rather than within 4e6 h of admission. Only approximately glutamine supplementation; however, SCCM/ASPEN recommends
one-third of patients received EN within 12 h of admission to the against and 22.4% of the patient-days received supplementation.
ICU, despite ESPEN guidelines. However, the best site was able to The on-going, international, multicenter, double-blind, pragmatic,
achieve EN initiation within 12 h of admission in 90% of patients randomized controlled trial of patients with severe burns (RE-
so improvements in timeliness of administration of EN are ENERGIZE) is the largest so far, and its results will shed more light
possible. on the effects on glutamine in burn patients [33].
Precise estimation of the caloric requirements of burn patients is With the exception of vitamin C, micronutrient supplements
important for nutritional management in order to optimize wound were provided in less than half of the total patient-days. However,
healing [27,28]. Measured energy needs of adult burn patients have even in the case of Vitamin C doses submitted were very low (at a
been shown to be increased by 160% on average, and no predictive supplemental level) and cannot be considered to demonstrate any
equation is considered accurate [28e30]. As a result, the Guidelines pharmacological action. Only 1 patient received >60 mg/kg/h,
[3,5] advocate for the measurement of energy requirements via IC whereas studies using IV vitamin C at 200 mg/kg/d (CITRUS ALI
at several time points during hospitalization to match the dynamic Trial) [34] failed to show an improvement in dysfunction scores or
changes in nutrient needs. In our study, the best site was able to alter markers of inflammation and vascular injury among patients
achieve measurements via IC in 93.3% of patients, with the with sepsis and acute respiratory distress syndrome.
remaining being estimated through a variety of equations. ESPEN [3] advocates for tight glycemic control among major
The Guidelines recommend goal protein prescription of burn patients, targeting a goal blood glucose between 81 and
1.5e2.0 g/kg/d (“or higher” per SCCM/ASPEN). A 31.0% of patients 144 mg/dL. Overall, 57.1% of patient-days were within the ESPEN
were prescribed >2.0 g/kg/d and only 19.2% of the recorded goal at the morning glucose check (best site: 68.0% of patient-
patient-days were within 1.5e2.0 g/kg/d (best site: 39.8%, worst days). Although SCCM/ASPEN does not suggest any burn-
site: 7.9%), leaving the majority of patient-days with an inappro- specific optimal glucose range, they provide recommendations
priately low protein administration. for 140 or 150e180 mg/dL for critically ill patients in general
According to the SCCM/ASPEN guidelines [2,5], 80% of energy [2,5]. Overall, 24.2% of the patient-days were within the SCCM/
and protein goals should be provided within 3 days of EN initiation; ASPEN recommended target (best site: 61.1% of patient-days).
any delays in initiation have been shown to lead to poorer out- According to Porter [27], post-burn glucose dysregulation is the
comes [31]. In our study, this recommendation was followed in result of diminished central and peripheral insulin sensitivity. As
approximately 1/3 of cases. The fact that one site was able to such, any strategies for improving glucose control may lead to
 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820 3819

faster recovery and reduced mortality [27]. However, a recent Formatting of funding sources
meta-analysis demonstrated only neutral results for short-term
or 3- or 6-month mortality and risk of sepsis among ICU pa- NA.
tients on intensive glycose control, and these results remained
constant among different study settings including surgical ICUs, Conflict of interest
medical ICUs or mixed ICUs [35]. Moreover, for a noteworthy
number of patient-days (223/1347, 16.6%, site range: 0.0%e41.3%) No conflict of interest.
insulin administration was reported to be more than 96 units/
day, which clearly exceeds traditional standards of metabolic
Acknowledgements
tolerance and is greater than normal non-burn insulin
requirements.
We would like to thank Dr. M. Grammatikopoulou for her input
We did not ask the sites whether they were intending to comply
in the very first draft of the paper.
to those or other guidelines, but we thought this was not necessary
as from a clinical point of view compliance with best practice
statements is a prerequisite for good clinical practice. Appendix A. Supplementary data
We acknowledge that it can be seen as a limitation that the ICU
participating in our sample are not a representative sample of ICU Supplementary data to this article can be found online at
across the globe. This is mainly attributed to the fact that partici- https://doi.org/10.1016/j.clnu.2020.04.023.
pation to the study was voluntary and this could be related to a
potential response bias, and therefor the evidence-practice gap References
found here, might be larger in reality. Despite, that all sample
[1] Clark A, Imran J, Madni T, Wolf SE. Nutrition and metabolism in burn patients.
centers are located in Anglo Saxon countries, and even if taking into Burns Trauma 2017;5:11.
account the cultural and medical specificities that come with this, [2] McClave SA, Taylor BE, Martindale RG, Warren MM, Johnson DR,
we still think that our study has a strength, which is that it does Braunschweig C, et al. Guidelines for the provision and assessment of nutri-
tion support therapy in the adult critically ill patient: society of critical care
observe actual nutrition practices among a large sample of ICUs and medicine (SCCM) and American society for parenteral and enteral nutrition
compares them to the evidence-based guidelines. We also believe (A.S.P.E.N.). J Parenter Enter Nutr 2016;40:159e211.
that the benchmarking aspect of this study highlights “best [3] Rousseau AF, Losser MR, Ichai C, Berger MM. ESPEN endorsed recommenda-
tions: nutritional therapy in major burns. Clin Nutr 2013;32:497e502.
achievable” practice.
[4] Snell JA, Loh NH, Mahambrey T, Shokrollahi K. Clinical review: the critical care
We recommend investigating strategies for removing the bar- management of the burn patient. Crit Care 2013;17:241.
riers to achieving the Guidelines within individual institutions. [5] Taylor BE, McClave SA, Martindale RG, Warren MM, Johnson DR,
Clearly, generating high levels of evidence will also help with Braunschweig C, et al. Guidelines for the provision and assessment of nutri-
tion support therapy in the adult critically ill patient: society of critical care
adherence to these clinical practice guidelines. medicine (SCCM) and American society for parenteral and enteral nutrition
(A.S.P.E.N.). Crit Care Med 2016;44:390e438.
[6] Williams FN, Branski LK, Jeschke MG, Herndon DN. What, how, and how much
5. Conclusions should patients with burns be fed? Surg Clin 2011;91:609e29.
[7] Cahill NE, Dhaliwal R, Day AG, Jiang X, Heyland DK. Nutrition therapy in the
critical care setting: what is “best achievable” practice? An international
Collectively, the present findings identified an overall significant multicenter observational study. Crit Care Med 2010;38:395e401.
lack of compliance to the Guidelines for nutrition support. The re- [8] Hegazi RA, Wischmeyer PE. Clinical review: optimizing enteral nutrition for
sults reveal a great variability in the estimation of energy re- critically ill patients-a simple data-driven formula. Crit Care 2011;15:234.
[9] Heyland DK, Lemieux M, Shu L, Quisenberry K, Day AG. What is “best
quirements, optimal timing of EN initiation, and lack of consensus achievable” practice in implementing the enhanced protein-energy provision
concerning the energy and protein needs of patients. Additionally, via the enteral route feeding protocol in intensive care units in the United
non-adherence was observed with glycemic control and adminis- States? Results of a multicenter, quality improvement collaborative. J Parenter
Enter Nutr 2016;42(2):308e17. https://doi.org/10.1177/0148607116673301.
tration of micronutrient supplements.
[10] Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Apache II: a severity of
We acknowledge that the inconsistency in standards and pro- disease classification system. Crit Care Med 1985;13:818e29.
cedures of care among different centers might have resulted in a [11] Moreno R, Vincent JL, Matos R, Mendonca A, Cantraine F, Thijs L, et al. The use
of maximum SOFA score to quantify organ dysfunction/failure in intensive
significant variability of the data. However, a strength of the study
care. Results of a prospective, multicentre study. Work Gr Sepsi Rel Prob
is that “real life” data was reported, observing actual nutrition ESICM Inten Care Med 1999;25:686e96.
practices among a large sample of ICUs. Undoubtedly, setting a [12] Ladhani HA, Sajankila N, Zosa BM, He JC, Yowler CJ, Brandt C, et al. Utility of
goal of 100% adherence to the Guidelines may not be practical. Sequential Organ Failure Assessment score in predicting bacteremia in criti-
cally ill burn patients. Am J Surg 2018;215:478e81.
However, the value of this paper is in defining a realistic, [13] Giretzlehner M, Dirnberger J, Owen R, Haller HL, Lumenta DB, Kamolz LP. The
achievable performance goal, which is important from a quality determination of total burn surface area: how much difference? Burns. J Inter
improvement perspective. It must be noted that goals of this study Soc Burn Inj 2013;39:1107e13.
[14] Tobiasen J, Hiebert JM, Edlich RF. The abbreviated burn severity index. Ann
were not to explain why this variation in adherence exists and not Emerg Med 1982;11:260e2.
to explain the mechanisms by which the best practice is achieved. [15] Heyland DK, Dhaliwal R, Jiang X, Day AG. Identifying critically ill patients who
Such an analysis may offer further insights in how to improve benefit the most from nutrition therapy: the development and initial vali-
dation of a novel risk assessment tool. Crit Care 2011;15:R268.
performance [7] by underlining factors that could act as barriers [16] Heyland DK, Cahill N, Day AG. Optimal amount of calories for critically ill
and/or enablers. patients: depends on how you slice the cake! Crit Care Med 2011;39:
2619e26.
[17] Harris JABF. A biometric study of basal metabolism in man. 1919. p. 32e40.
Statement of authorship [18] Hamwi G. Therapy: changing dietary concepts. 1964.
[19] Lu G, Huang J, Yu J, Zhu Y, Cai L, Gu Z, et al. Influence of early post-burn enteral
nutrition on clinical outcomes of patients with extensive burns. J Clin Biochem
MC lead the writing of the manuscript and participated in the Nutr 2011;48:222e5.
statistical analyses, EB lead the statistical analyses, BES, CS, AFR [20] Chen Z, Wang S, Yu B, Li A. A comparison study between early enteral
participated in various stages of the writing of the manuscript, DKH nutrition and parenteral nutrition in severe burn patients. Burns : J Inter Soc
Burn Inj 2007;33:708e12.
conducted the idea of the study and reviewed each version of the [21] Mosier MJ, Pham TN, Klein MB, Gibran NS, Arnoldo BD, Gamelli RL, et al. Early
manuscript. enteral nutrition in burns: compliance with guidelines and associated
3820 M. Chourdakis et al. / Clinical Nutrition 39 (2020) 3813e3820

outcomes in a multicenter study. J Burn Care Res : Off Pub Am Burn Asso [29] Allard JP, Pichard C, Hoshino E, Stechison S, Fareholm L, Peters WJ, et al.
2011;32:104e9. Validation of a new formula for calculating the energy requirements of burn
[22] Hart DW, Wolf SE, Chinkes DL, Beauford RB, Mlcak RP, Heggers JP, et al. patients. J Parenter Enter Nutr 1990;14:115e8.
Effects of early excision and aggressive enteral feeding on hypermetabolism, [30] Dickerson RN, Gervasio JM, Riley ML, Murrell JE, Hickerson WL, Kudsk KA,
catabolism, and sepsis after severe burn. J Trauma 2003;54:755e61. dis- et al. Accuracy of predictive methods to estimate resting energy expenditure
cussion 61-4. of thermally-injured patients. J Parenter Enter Nutr 2002;26:17e29.
[23] Williams FN, Herndon DN, Jeschke MG. The hypermetabolic response to burn [31] Wolf SE, Rose JK, Desai MH, Mileski JP, Barrow RE, Herndon DN. Mortality
injury and interventions to modify this response. Clin Plast Surg 2009;36: determinants in massive pediatric burns. An analysis of 103 children with >
583e96. or ¼ 80% TBSA burns (> or ¼ 70% full-thickness). Ann Surg 1997;225:554e65.
[24] Williams FN, Jeschke MG, Chinkes DL, Suman OE, Branski LK, Herndon DN. discussion 65-9.
Modulation of the hypermetabolic response to trauma: temperature, nutri- [32] Prest PJ, Justice J, Bell N, McCarroll R, Watson CM. A volume-based feeding
tion, and drugs. J Am Coll Surg 2009;208:489e502. protocol improves nutrient delivery and glycemic control in a surgical trauma
[25] Xiao SC, Zhu SH, Xia ZF, Lu W, Wang GQ, Ben DF, et al. Prevention and intensive care unit. J Parenter Enter Nutr 2019. https://doi.org/10.1002/
treatment of gastrointestinal dysfunction following severe burns: a sum- jpen.1712.
mary of recent 30-year clinical experience. World J Gastroenterol 2008;14: [33] Heyland DK, Wischmeyer P, Jeschke MG, Wibbenmeyer L, Turgeon AF,
3231e5. Stelfox HT, et al. A RandomizEd trial of ENtERal Glutamine to minimIZE
[26] Gottschlich MM, Jenkins ME, Mayes T, Khoury J, Kagan RJ, Warden GD. The thermal injury (The RE-ENERGIZE Trial): a clinical trial protocol. Scars, Burns
2002 Clinical Research Award. An evaluation of the safety of early vs delayed Healing 2017;3. 2059513117745241.
enteral support and effects on clinical, nutritional, and endocrine outcomes [34] Fowler 3rd AA, Truwit JD, Hite RD, Morris PE, DeWilde C, Priday A, et al. Effect
after severe burns. J Burn Care Rehabil 2002;23:401e15. of vitamin C infusion on organ failure and biomarkers of inflammation and
[27] Porter C, Tompkins RG, Finnerty CC, Sidossis LS, Suman OE, Herndon DN. The vascular injury in patients with sepsis and severe acute respiratory failure: the
metabolic stress response to burn trauma: current understanding and ther- CITRIS-ALI randomized clinical trial. J Am Med AssocJ Am Med Assoc
apies. Lancet (London, England) 2016;388:1417e26. 2019;322:1261e70.
[28] Rodriguez NA, Jeschke MG, Williams FN, Kamolz LP, Herndon DN. Nutrition in [35] Fu Y, Sun Y, Zhang J, Cheng Y. Intensive glucose control for critically ill pa-
burns: galveston contributions. J Parenter Enter Nutr 2011;35:704e14. tients: an updated meta-analysis. Endocr Connect 2018;7:1288e98.