Clinical Nutrition xxx (2016) 1e6

Contents lists available at ScienceDirect

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

Original article

Does the type of parenteral lipids matter? A clinical hint in critical

illness
J.-C. Devaud a, b, *, M.M. Berger c, A. Pannatier a, b, F. Sadeghipour a, b, P. Voirol a, b
a
Service of Pharmacy, Lausanne University Hospital, Switzerland
b
School of Pharmaceutical Sciences, University of Geneva and University of Lausanne, Geneva, Switzerland
c
Service of Adult Intensive Care Medicine & Burns, Lausanne University Hospital, Switzerland

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

Article history: Background & aims: An altered lipid prole is common among intensive care unit (ICU) patients, but
Received 15 October 2015 evidence regarding the impact of different fatty acid (FA) emulsions administered to patients requiring
Accepted 12 January 2016 parenteral nutrition (PN) is scarce. This study aimed to compare the plasma triglycerides (TG) response to
two types of commercial lipid emulsions: a structured mixture of long- and medium-chain triglycerides
Keywords: (LCT/MCT) or LCTs with n-9 FA (LCT) in ICU patients.
Hypertriglyceridemia
Methods: In this retrospective observational study conducted in a multidisciplinary ICU: two groups
ICU
were dened by the type of emulsion used. Inclusion criteria were: consecutive patients on PN staying
Nutrition
Propofol
4 days with one TG determination before commencing PN and at least one during PN. Recorded var-
Sedation iables included energy intake, amount and type of nutritional lipids, propofol dose, glucose and protein
Lipid metabolism intake, laboratory parameters, and all drugs received. Hypertriglyceridemia (hyperTG) was dened as TG
>2 mmol/L.
Results: The dynamic impact of the emulsion was analyzed in 187/757 patients completing the inclusion
criteria (112 LCT/MCT and 75 LCT). The demographic variables, severity indices, diagnostic categories,
and outcomes did not differ between the two groups. Seventy-seven patients (41%) presented hyperTG.
Both groups received similar daily energy (1604 versus 1511 kcal/day), lipids (60 versus 61 g/day), and
glucose intake (233 versus 197 g/day). There was no increase of TG concentration in those receiving the
LCT/MCT emulsion compared to those receiving the LCT emulsion (0 and 0.2 mmol/L, respectively,
p < 0.05).
Conclusion: LCT/MCT emulsions are associated with a less pronounced increase of plasma TG levels than
LCT emulsions.
2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.

1. Introduction associated with overfeeding and its deleterious consequences such

Critical illness is characterized by nutritional and metabolic
disorders, resulting in increased muscle catabolism, fat-free mass
loss, hyperglycemia, and hypertriglyceridemia (hyperTG) [1,2]. In
patients with contraindications to enteral feeding or insufcient
enteral feeding, parenteral nutrition (PN) is the standard of care [3].
Intravenous lipids are a vital component of PN as an important
source of energy because they maintain integral components of cell
membranes and prevent the development of essential fatty acid
deciency [4,5]. It has been demonstrated that PN is frequently
* Corresponding author. Service of Pharmacy, CHUV BH04 524, 1011 Lausanne,
Switzerland. Tel.: 41 79 556 14 99; fax: 41 21 314 42 99.
E-mail address: Jean-Christophe.Devaud@chuv.ch (J.-C. Devaud).
as hyperglycemia, hyperTG, liver steatosis, endocrine dysfunction,
impairment of immunity, infections, and increased mortality
[3,6,7]. The association between PN and morbidity is multi-factorial
and has often been suggested to be linked to the fat emulsions used
[3,8,9]. Although the evidence for this suggestion has never been
conclusive, many centers limit the use of fat emulsions, especially
when hyperTG is present [3,6,9].
The rst available lipid emulsions contained only long-chain
triglycerides (LCTs) [10] and metabolic disorders related to their
use have been published, and efforts at further developing and
optimizing lipid emulsions have focused on replacing part of the
LCTs with medium-chain triglycerides (MCTs) [11]. Indeed, the fatty
acids (FAs) composing emulsions do inuence the clinical re-
sponses, depending on their chemical structure. Compared with

http://dx.doi.org/10.1016/j.clnu.2016.01.009
0261-5614/ 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism.

Please cite this article in press as: Devaud J-C, et al., Does the type of parenteral lipids matter? A clinical hint in critical illness, Clinical Nutrition
(2016), http://dx.doi.org/10.1016/j.clnu.2016.01.009
2 J.-C. Devaud et al. / Clinical Nutrition xxx (2016) 1e6
LCTs, MCT-based emulsions are cleared more rapidly from the
plasma [8,12]. Other benets of MCTs include their less pronounced
tendency for deposition in tissues and their favorable effect on
protein metabolism [11,13]. Further advances include the devel-
opment of lipid emulsions containing structured triglycerides,
which are metabolized even more efciently than LCTs and MCTs
[13]. As an alternative to the physical mixture of MCT and LCT
emulsions a mixture of medium-chain and long-chain fatty acids
can be obtained by interesterifying the different fatty acids to create
a mixed triglyceride molecule called a structured triglyceride (STG).
Olive oil-based emulsions in which 60% of the LCTs consist of
mono-unsaturated fatty acid (n-9 FA) were then developed [14]
with the latter having a theoretical limited impact on lipid meta-
bolism. In the most recent emulsions developed, the LCTs and MCTs
are progressively replaced by other fatty acids, particularly omega-
3.
Having observed an increasing incidence of hyperTG after
changing the commercial parenteral lipid emulsion in our intensive
care unit (ICU), this study aimed to compare the metabolic re-
sponses to commercial PN solutions containing the LCT n-9 FA
(LCT) versus structured LCT/MCT lipids in critically ill patients.
2. Methods
This retrospective observational study was approved by the
ethics committee of the Canton of Vaud and was conducted over a
50-month period (November 2008 to December 2013) in a 32-bed
adult mixed ICU. Inclusion criteria were as follows: consecutive
patients on PN staying 4 days and <18 days with one TG deter-
mination before commencing PN and at least one during PN.
Exclusion criteria were as follows: less than 3 days of continuous PN
or if TG levels were not determined after/before and during PN
administration. HyperTG was dened as a plasma TG level
>2 mmol/L, according to the current guidelines of the American
Heart Association [15]. Patients were grouped according to the type
of lipid in the PN (see below).
2.1. Patient data
Patient data included age, admission weight, body mass index
(BMI), type of admission (surgical or medical), severity of disease
(SAPSII), and mortality during the ICU and hospital stay. All data
were collected during the stay in order to have a dynamic view and
to establish a temporal relationship with the type of lipid emulsion
used. Nutritional data included intravenous and enteral energy,
with details of the quantity of lipids, carbohydrates, and proteins.
The cumulative energy intake included nutritional (i.e., enteral
nutrition (EN) and PN) and non-nutritional energy intake (i.e.
propofol and dextrose 5% perfusions). Laboratory data included the
levels of alanine transaminase, aspartate transaminase, albumin,
pancreatic amylase, direct and indirect bilirubin, gamma-glutamyl
transferase, alkaline phosphatase, procalcitonin, prealbumin, and
C-reactive protein.
2.2. Study periods
This study was divided into 2 periods. During the rst period
(2008e2011), the predominant commercial solution was a struc-
tured mix of 64% LCT/36% MCT, made from soja and coco oil
respectively, emulsion (Structokabiven, Fresenius Kabi, Oberdorf,
Switzerland); while an 100% LCT emulsion, made from 80:20 mix of
olive and soybean oil, was used from 2011 (Olimel 5.7%, Baxter AG,
Volketswil, Switzerland). Some patients also received the CHUV
local-compounded PN in our Service of pharmacy called ALISIA
(ALImentation aux Soins Intensifs Adultes), which is an ICU patient
Please cite this article in press as: Devaud J-C, et al., Does the type of parenteral lipids matter? A clinical hint in critical illness, Clinical Nutrition
(2016), http://dx.doi.org/10.1016/j.clnu.2016.01.009
adapted, concentrated (1230 mL), low-fat (20% of energy as LCT/
MCT mix), high-protein (25% energy), and high-glucose solution
(55% energy) [16,17]. This solution is recommended when PN lasts
>3 days. The compositions of these solutions are shown in Table 1.
2.3. Nutritional management
Nutritional requirements were based on the ESPEN guidelines
[3] evolved over time as follows: energy targets were 25e30 kcal/
kg/day (medical and surgical conditions) during the rst period,
and they decreased to 20e25 kcal/kg/day during the second period
(with downregulation in elderly and obese patients). During the
rst and the second period there was a switch of the commercial
PN. Indirect calorimetry was recommended after 1 week. The
protein target was 1.2e1.3 g/kg/day, and continuous EN was
encouraged. Combined EN and PN was considered when EN was
insufcient, otherwise, PN was restricted to gastrointestinal failure.
Lipid prole monitoring was an integral part of the ICU nutrition
protocol (blood sampling three times weekly at 6 a.m. for deter-
mination of TG and C-reactive protein levels). Non-nutritional
intake was taken into account when devising nal feeding
regimen, and the sedation protocol was based on ESICM recom-
mendations [18], discouraging the use of high-dose propofol
(>4 mg/kg/h) while integrating daily sedation pauses. Overfeeding
was dened as 28 kcal/kg in the absence of indirect calorimetric
determination based on the large multicenter Spanish ICU study
including 725 patients receiving either EN or PN [19].
2.4. Laboratory analysis
All analyses were performed on a Cobas 8000 modular analyzer
(Roche Diagnostics, Switzerland), except for prealbumin, which
was determined on an Integra instrument (Roche Diagnostics,
Switzerland). Enzymatic methods were used to determine TG levels
(GPO-PAP), and an immunoturbidimetric assay was used for the
determination of albumin and C-reactive protein. Pancreatic
amylase was determined by an immunoinhibition assay, and an
electrochemical immunoassay was used for procalcitonin antigen
detection. The hospital's Laboratory of Clinical Chemistry is ISO
15189:2012 certied.
2.5. Data collection and analysis
Data were extracted from the clinical information system (CIS)
MetaVision (iMDSoft, version 5.45.5403, Tel Aviv, Israel). The CIS
was customized to provide detailed composition information and
quantities of the enteral and parenteral feeding solutions, including
the respective amounts of LCTs and MCTs [20]. It was customized to
show the total energy with detailed information including non-
nutritional substrate intake.
The data and results are presented as medians and interquartile
ranges or as number of subjects and percentage. Two-way analysis
of variance and linear regression were used for analysis with the
software programs R language (R Foundation, version 2.10.0) and
JMP V5.1 (SAS Institute, Cary, NC, USA). Statistical signicance was
considered at the level of p < 0.05.
3. Results
During the study period, 10,656 patients were admitted to the
ICU, of whom 757 (7%) were on PN and eligible for the study based
on the length of stay. Altogether, 228 patients (30.1%) presented at
least one episode of hyperTG during their time on PN, but there was
no signicant difference between the LCT/MCT and LCT groups
(p 0.53). Only 187 patients could be analyzed for dynamic TG
 J.-C. Devaud et al. / Clinical Nutrition xxx (2016) 1e6 3

Table 1
PN compositions.

Parameter Structokabivena (64% LCT/36% MCT) Olimel 5.7%b (100% LCT) ALISIAa (50% LCT/50% MCT)

Energy (kcal/L) 1054 1028 1399

Lipids (g/L) 38 40 27
Amino acids (g/L) 51 56.9 86
Glucose (g/L) 127 110 203
a
Soya and coconut oil.
b
Soya and olive oil.

changes during PN due to missing TG determinations during their
ICU stay. The clinical characteristics of the 112 (60%) patients on
LCT/MCT PN and the 75 (40%) patients on LCT are summarized in
Table 2. In the LCT/MCT group, there was more combined nutrition
and a shorter length of stay (p 0.0292) (Table 2).
3.1. Evolution of plasma TG
Among patients with at least two TG determinations, including
at least one TG during and one before or after PN, 77 patients (41%)
presented at least one hyperTG during their time on PN. The overall
incidence of hyperTG did not differ between the two groups
(p 0.66). The increase in TG over time was signicantly greater in
the LCT group than in the MCT/LCT group (Table 2) and is illus-
trated in Fig. 1. However, in patients who were shifted from a
commercial PN to ALISIA, the TG level decreased overall and was
signicantly different compared with both commercial PN prepa-
rations (Table 3). Seventeen patients under structured triglycerides
and 5 patients under LCT received >1.5 g/kg of lipids and all
received similar amount of propofol (p 0.506) with less than
4 mg/kg/h.
Propofol didn't exacerbate the TG level of patients who were
receiving >1.5 g/kg (p 0.685) compared to patients who received
a normal amount of lipids.
3.2. Feeding
The duration of PN was similar in both groups. The median
intravenous energy deliveries were 24.7 kcal/kg/day [18.3; 28.3]
and 21.9 kcal/kg/day [18.6; 25.3] for the LCT/MCT and LCT groups,
respectively (p 0.0897). Measured values did not differ between
the two groups; the lipids, intravenous and/or enteral energy, and
carbohydrate as well as protein delivery remained within the rec-
ommended ranges (Table 4).
Thirty-two patients received more than 28 kcal/kg/day, the
target being based on indirect calorimetry in three patients
(Table 2). Overfeeding was observed less often in the LCT group
(n 5) than in the LCT/MCT group (n 20) (p 0.0027). Among the
patients with overfeeding, one patient in the LCT/MCT group and
two patients in the LCT group were on strict PN, while the other
patients were on combined nutrition. In addition, a positive cor-
relation was observed between lipid delivery and propofol dose
(r2 0.28; p < 0.0001), and a negative correlation was found be-
tween enteral energy (kcal/day) and intravenous energy (kcal/day)
(r2 0.37; p < 0.0001), reecting progression towards EN.
There were no signicant differences between the levels of
aspartate transaminase, alanine transaminase, alkaline phospha-
tase, gamma-glutamyl transferase, C-reactive protein, total bili-
rubin, direct bilirubin, or prealbumin between the two groups
(Table 4) or during and before or after PN.
4. Discussion
The main nding in our study was that commercial PN con-
taining structured triglycerides did not impact the TG prole as
much as commercial PN containing LCTs alone, when delivered at
similar fat doses. This nding might reect a faster clearance by the
lipoprotein lipase, which hydrolyzes MCTs more rapidly and
completely than LCTs alone [12]. It has been proposed that MCTs,
because of their rapid and rather complete oxidation, are associated
with a lower risk of developing hyperTG than LCTs alone [9].
However, this hypothesis has remained controversial [21] probably
because previous studies cited in the literature [9,12] had no clear
TG baseline and investigated the incidence of hyperTG rather than
changes in the TG prole. The small size of the studies and the low

Table 2
Patient characteristics and outcome variables.

Variable LCT/MCT (N 112) LCT (N 75) p

Age (years) 67.5 (58.8e76) 64 (55.5e74) NS

Preadmission weight (kg) 75 (65e87) 78 (68.5e86) NS
Body mass index (kg/m2) 24.7 (23.2e28.8) 25.8 (23e29.6) NS
Women (%) 28.6% (32) 26.6% (20) NS
SAPS II 48 (39e60) 50 (39.5e60.5) NS
Medical/surgical ratio 28.8%/73.2% (30/82) 30.1%/69.8% (23/52) NS
Propofol dose during PN (mg/kg/d) 13.0 (3.3e30.4) 9.4 (1.4e32.4) NS
Patients on combined EN PN (N) 77 5 <0.0001
TG prole
TG concentration (mmol/L) before PN 1.5 (1.1e2.2) 1.4 (1.1e1.8) NS
TG concentration (mmol/L) during PN 1.6 (1.1e2.2) 1.7 (1.2e2.3) NS
Change of TG concentration (mmol/L) 0 (0.3e0.4) 0.2 (0.1e0.7) 0.01297
Outcome
Length of ICU stay (days) 10.6 (8.7e13.7) 13 (9.1e16.1) 0.0292
Combined nutrition (days) 2 (1e4) 2 (0e4) NS
Number of patients receiving 28 kcal/kg/day 22% (25) 9% (7) 0.0223
ICU mortality (%) 14.3% (16) 16% (12) NS
Hospital mortality after discharge from ICU (%) 25.9% (29) 26.6% (20) NS

Results are expressed as median (interquartile range) or as percentage (number of subjects). TG triglyceride, PN parenteral nutrition, EN enteral nutrition, NS not signicant.

Please cite this article in press as: Devaud J-C, et al., Does the type of parenteral lipids matter? A clinical hint in critical illness, Clinical Nutrition
(2016), http://dx.doi.org/10.1016/j.clnu.2016.01.009
4 J.-C. Devaud et al. / Clinical Nutrition xxx (2016) 1e6

Fig. 1. Analyses of plasma TG and C-reactive protein levels over 11 days in patients receiving PN containing MCT/LCT (A) or LCT (B), with the total daily dose of fatty acids (from PN
and propofol) indicated. The patients presented increases of 0.2 mmol/L and 0.4 mmol/L TG, respectively.

patient numbers may also explain the limited ndings in the
literature [8,22]. Moreover, our results give a tool to normalize
patients' triglycerides which may well be abnormal and worsen
under n-9 FA emulsions. The switch to a commercial PN containing
a mix of LCT/MCT might thus contribute to avoid any metabolic
disorder such a possible acute pancreatitis due to
hypertriglyceridemia.
Our ICU feeding protocol was based on the ESPEN guidelines
[3,23], which recommend 1.2e1.5 g/kg/day of protein/amino acids.
Thus, the protein delivery was similar to that observed in other

Please cite this article in press as: Devaud J-C, et al., Does the type of parenteral lipids matter? A clinical hint in critical illness, Clinical Nutrition
(2016), http://dx.doi.org/10.1016/j.clnu.2016.01.009
 J.-C. Devaud et al. / Clinical Nutrition xxx (2016) 1e6 5

Table 3 we investigated individual fat intake in those patients presenting

Changes in TG concentration under the different PN types.
Type of emulsion Change of TG concentration (mmol/L)
LCT 0.4 (0e1)
LCT and ALISIA 0.2 (0.02e0.72)
LCT/MCT 0.25 (0.1 to 0.7)
LCT/MCT and ALISIA 0.05 (0.4 to 0.4)
Results are expressed as median (interquartile range).
TG triglyceride, PN parenteral nutrition, N number of patients.
European studies [24,25]. It is recommended that the glucose de-
livery should not exceed 6 g/kg/day and that the lipid supply should
not exceed 1.5 g/kg or 35% of the total energy input [7]. These
recommendations were followed in our cohort.
As our energy target recommendations decreased during the
second period (from 25e30 kcal/kg/day to 20e25 kcal/kg/day), the
LCT emulsion was chosen because of its higher amino acid content
in order to maintain an adequate nitrogen intake. This modest
reduction in energy resulted in a reduction of delivery of approxi-
mately 100 kcal/day. Despite this reduction, there were larger (but
not more frequent) increases in the TG level; however, the differ-
ence in the lipid prole cannot be attributed to a higher substrate
and energy delivery. Even the propofol intake was similar, thus
excluding a factor that might favor hyperTG [17].
We observed that the LCT/MCT group had a signicantly shorter
hospital stay of 2.4 days. However, a meta-analysis by Wirtitsch
et al. did not demonstrate any statistically signicant reduction in
the length of hospital stay in any of the emulsion groups studied
[22]. Meanwhile, Bauer et al. showed that the length of hospital
stay was signicantly (p 0.0022) reduced by 2.5 days in the group
that had beneted from early combined EN and PN to fulll the
target [26]. There was no difference between our two groups for
patients on combined nutrition, nor any difference in death rate,
but mortality is multi-factorial in the ICU. The ALISIA substrate
composition is close to the optimal PN recommendations, with only
20% of energy as fat [5,8,10,24,27]. Its use was associated with a
decrease in the plasma TG level, conrming the impact of a 40%
lower proportion of fat than in commercial PN. Since parenteral fat
intake is known to exacerbate hyperTG during acute illness [9,28],
hyperTG. We observed that the guidelines for parenteral fat intake
N p [3] (i.e., 0.7e1.5 g/kg/d) were respected.
71 0.0251
Overfeeding, with glucose and/or fat by enteral or intravenous
10 routes, is a common cause of hyperTG [29]; however, in our cohort,
66 overfeeding was an exception. Indeed, to prevent overfeeding, all
40 patients were monitored daily against a standard energy target,
which was veried by indirect calorimetry in several cases. Our CIS
demonstrated the cumulative carbohydrate dose from both enteral
and intravenous routes, thereby preventing the administration of
doses and proportions known to induce de novo lipogenesis [3,30]
and reducing the risk of de novo lipid synthesis [31]. Our results
showed that reducing the energy target in the second period
contributed to a lower rate of overfeeding.
Although cases of acute pancreatitis, fatty liver, delayed awak-
ening, retinal lipemia [9,28,32], and elevated mortality, particularly
in association with hypocholesterolemia [33], have been described,
the clinical consequences of acute and transient hyperTG have not
been investigated to date. An increased risk of infection by
disruption of the reticuloendothelial system [9,34], coagulopathy
[35], neurological disturbance, or respiratory failure [9] also have
been reported, none of which occurred in our patients. Therefore,
further randomized controlled trials are needed to determine
which alternative PN emulsion should be used in critically ill pa-
tients to improve lipid metabolism and clinical outcomes [36]. Thus
it appears to be important to know what kind of lipid is adminis-
tered in order to lead to a better control of the ICU patients' tri-
glyceridemia and then to prevent the risks associated with
HyperTG. A signicant median elevation of serum triglyceride
levels (i.e. 0.2 mmol/L in our study) can exacerbate metabolic dis-
orders of acute illness [9].
The principal limitation of this retrospective exploratory
observational study is the modest size of the cohort, which limits
the analysis (n 187 with 41% hyperTG). The low number was due
to the twice weekly monitoring protocol not being systematically
applied, leading to many patients having only one TG determina-
tion available. However, the relatively small number of observa-
tions has to be weighed by the fact that our cohort included very
sick patients with a median stay of 11 days and 15% ICU mortality

Table 4
Details of feeding, outcomes, and other laboratory results.

Parameter LCT/MCT (N 112) LCT (N 75) p

Feeding
Days to rst PN (days) 2 (2e4) 2 (1.5e4) NS
PN days (n) 6 (4e7.25) 6 (5e8) NS
Prescribed energy (kcal/day) 1800 (1700e2000) 1800 (1500e2000) NS
Recommended target (kcal/day) 1875 (1625e2175) 1716 (1507e1892) 0.0006
Harris & Benedict predicted REE (kcal/day) 1497 (1331e1703) 1557 (1346e1744) NS
Measured substrate and energy delivery during PN
Energy delivery (kcal/day) 1604 (1231e1827) 1511 (1225e1771) NS
Protein (g/day) 84 (72.1e94.8) 85 (73e94.8) NS
Protein (g/kg/day) 1.3 (1.1e1.5) 1.3 (1.1e1.4) NS
Lipids (g/day) 60 (50e71) 61 (50e68) NS
Lipids (g/kg/day) 0.9 (0.8e1.1) 0.89 (0.7e1) NS
Glucose (g/day) 233 (213e261) 197 (176e222) NS
Glucose (g/kg/day) 3.6 (3.2e3.9) 3.1 (2.7e3.4) NS
Aspartate transaminase (UI/day) 49.5 (32.4e80) 57.5 (33e104) NS
Alanine transaminase (UI/day) 39.75 (25.8e67.3) 42.5 (24.6e96.3) NS
Alkaline phosphatase (UI/day) 117 (78e164) 124.5 (92.1e205.6) NS
Gamma-glutamyl transferase (UI/day) 129 (65e259.8) 148 (81.4e251.6) NS
C-reactive protein (UI/day) 120.5 (71.5e165.3) 121 (67.3e162) NS
Total bilirubin (UI/day) 18 (14.5e49.3) 28.25 (15.6e53) NS
Direct bilirubin (UI/day) 26 (14.5e53) 27 (17e51) NS
Prealbumin (UI/day) 0.1 (0.1e0.1) 0.09 (0.1e0.1) NS

Results are expressed as median (interquartile range).

PN parenteral nutrition, REE resting energy expenditure, NS not signicant.

Please cite this article in press as: Devaud J-C, et al., Does the type of parenteral lipids matter? A clinical hint in critical illness, Clinical Nutrition
(2016), http://dx.doi.org/10.1016/j.clnu.2016.01.009
6 J.-C. Devaud et al. / Clinical Nutrition xxx (2016) 1e6
(26% hospital mortality) and that the ICU's computerized database
contains very complete metabolic information.
Another limitation of this study is that a mixture of both types of
fatty acids was used in several cases due to the clinical use of
propofol (an LCT emulsion), as shown in both depicted cases.
Therefore, the observed changes in TG levels were not due to a
unied process. Another limitation is the wide spectrum of pa-
thologies present and presumably the genetic characteristics in the
cohort, both of which modulate the metabolic responses to feeding.
In the absence of more information, these aspects cannot be
explored.
In conclusion, hyperTG is frequent during critical illness. In
addition to the previously identied factors associated with
hyperTG, the present study demonstrated that the type of fat
emulsion directly inuences the TG prole. The MCT-containing PN
was associated with a lower impact on the plasma TG level than the
LCT emulsion in this cohort of critically ill patients. Therefore, the
TG levels of critically ill patients should be regularly monitored,
probably before the initiation of PN and then twice weekly during
PN or during propofol sedation. Furthermore, more extensive
knowledge regarding the effects of various types of emulsions on
outcomes such as the TG prole, length of hospital stay, and clinical
consequences of hyperTG is necessary.
Statement of authorship
All authors read and approved the nal manuscript as
submitted.
Conict of interest
None.
References
[1] Berger M, Pichard C. Development and current use of parenteral nutrition in
critical care e an opinion paper. Crit Care 2014;18:478.
[2] Preiser J, van Zanten A, Berger M, Biolo G, Casaer M, Doig G, et al. Metabolic
and nutritional support of critically ill patients: consensus and controversies.
Crit Care 2015;19:35.
[3] Singer P, Berger MM, Van den Berghe G, Biolo G, Calder P, Forbes A, et al.
ESPEN Guidelines on parenteral nutrition: intensive care. Clin Nutr 2009;28:
387e400.
[4] Bach AC, Storck D, Meraihi Z. Medium-chain triglyceride-based fat emulsions:
an alternative energy supply in stress and sepsis. J Parenter Enteral Nutr
1988;12:82Se8S.
[5] Calder PC. Lipids for intravenous nutrition in hospitalised adult patients: a
multiple choice of options. Proc Nutr Soc 2013;72:263e76.
[6] Shams MR, Tavassoli N, Plicaud H, Genestal M. Incidence and risk factors of
hypertriglyceridemia in the ICU. Crit Care Med 2009;13:130.
[7] Ziegler TR. Parenteral nutrition in the critically ill patient. N Engl J Med
2009;361:1088e97.
[8] Manzanares W, Dhaliwal R, Jurewitsch B, Stapleton RD, Jeejeebhoy KN,
Heyland DK. Alternative lipid emulsions in the critically ill: a systematic re-
view of the evidence. Intensive Care Med 2013;39:1683e94.
[9] Marik PE. Dyslipidemia in the critically ill. Crit Care Clin 2006;22:151e9. viii.
[10] Berger MM. The 2013 Arvid Wretlind lecture: evolving concepts in parenteral
nutrition. Clin Nutr 2014 Aug;33(4):563e70.
[11] Adolph M. Lipid emulsions in parenteral nutrition. Ann Nutr Metab 1999;43:
1e13.
Please cite this article in press as: Devaud J-C, et al., Does the type of parenteral lipids matter? A clinical hint in critical illness, Clinical Nutrition
(2016), http://dx.doi.org/10.1016/j.clnu.2016.01.009
[12] Ton MN, Chang C, Carpentier YA, Deckelbaum RJ. In vivo and in vitro prop-
erties of an intravenous lipid emulsion containing only medium chain and sh
oil triglycerides. Clin Nutr 2005;24:492e501.
[13] Wanten G. An update on parenteral lipids and immune function: only smoke,
or is there any re? Curr Opin Clin Nutr Metab Care 2006;9:79e83.
[14] Reimund JM, Rahmi G, Escalin G, Pinna G, Finck G, Muller C, et al. Efcacy and
safety of an olive oil-based intravenous fat emulsion in adult patients on
home parenteral nutrition. Aliment Pharmacol Ther 2005;21:445e54.
[15] Gibbons RJ, Abrams J, Chatterjee K, Daley J, Deedwania PC, Douglas JS, et al.
ACC/AHA 2002 guideline update for the management of patients with chronic
stable angina e summary article: a report of the American College of Cardi-
ology/American Heart Association Task Force on practice guidelines (Com-
mittee on the Management of Patients with Chronic Stable Angina). J Am Coll
Cardiol 2003;41:159.
[16] Berger M, Chiole ro R, Pannatier A, Cayeux C, Tappy L. A 10-year survey of
nutritional support in a surgical ICU: 1986e1995. Nutrition 1997;13:870e7.
[17] Devaud JC, Berger MM, Pannatier A, Marques-Vidal P, Tappy L, Rodondi N,
et al. Hypertriglyceridemia: a potential side effect of propofol sedation in
critical illness. Intensive Care Med 2012;38:1990e8.
[18] Oddo M. Sedat and analgesia skills and techniques. ESICM; 2010. p. 56.
[19] Grau T, Bonet A, Rubio M, Mateo D, Farre  M, Acosta JA, et al. Liver dysfunction
associated with articial nutrition in critically ill patients. Crit Care 2007;11:
R10.
[20] Berger MM, Revelly JP, Wasserfallen JB, Schmid A, Bouvry S, Cayeux MC, et al.
Impact of a computerized information system on quality of nutritional sup-
port in the ICU. Nutrition 2006;22:221e9.
[21] Druml W, Fischer M, Pidlich J, Lenz K. Fat elimination in chronic hepatic
failure: long-chain vs medium-chain triglycerides. Am J Clin Nutr 1995;61:
812e7.
[22] Wirtitsch M, Wessner B, Spittler A, Roth E, Volk T, Bachmann L, et al. Effect of
different lipid emulsions on the immunological function in humans: a sys-
tematic review with meta-analysis. Clin Nutr 2007;26:302e13.
[23] Kreymann KG, Berger MM, Deutz NE, Hiesmayr M, Jolliet P, Kazandjiev G, et al.
ESPEN guidelines on enteral nutrition: intensive care. Clin Nutr 2006;25:
210e23.
[24] Allingstrup MJ, Esmailzadeh N, Wilkens Knudsen A, Espersen K, Hartvig
Jensen T, Wiis J, et al. Provision of protein and energy in relation to measured
requirements in intensive care patients. Clin Nutr 2012;31:462e8.
[25] Weijs P, Stapel S, de Groot S, Driessen R, de Jong E, Girbes A, et al. Optimal
protein and energy nutrition decreases mortality in mechanically ventilated,
critically ill patients: a prospective observational cohort study. J Parenter
Enteral Nutr 2012;36:60e8.
[26] Bauer P, Charpentier C, Bouchet C, Nace L, Raffy F, Gaconnet N. Parenteral with
enteral nutrition in the critically ill. Intensive Care Med 2000;26:893e900.
[27] Jeejeebhoy KN. Parenteral nutrition in the intensive care unit. Nutr Rev
2012;70:623e30.
[28] Crook MA. Lipid clearance and total parenteral nutrition: the importance of
monitoring plasma lipids. Nutrition 2000;16:774e5.
[29] Nordenstrom J, Carpentier YA, Askanazi J, Robin AP, Elwyn DH, Hensle TW,
et al. Metabolic utilization of intravenous fat emulsion during total parenteral
nutrition. Ann Surg 1982;196:221e31.
[30] Tappy L, Schwarz J, Schneiter P, Cayeux C, Revelly J, Fagerquist C, et al. Effects
of isoenergetic glucose-based or lipid-based parenteral nutrition on glucose
metabolism, de novo lipogenesis, and respiratory gas exchanges in critically ill
patients. Crit Care Med 1998;26:860e7.
[31] Tappy L, Berger MM, Schwarz JM, Schneiter P, Kim S, Revelly JP, et al. Meta-
bolic effects of parenteral nutrition enriched with n-3 polyunsaturated fatty
acids in critically ill patients. Clin Nutr 2006;25:588e95.
[32] Llop J, Sabin P, Garau M, Burgos R, Perez M, Masso J, et al. The importance of
clinical factors in parenteral nutrition-associated hypertriglyceridemia. Clin
Nutr 2003 Dec;22(6):577e83.
[33] Chiarla C, Giovannini I, Giuliante F, Zadak Z, Vellone M, Ardito F, et al. Severe
hypocholesterolemia in surgical patients, sepsis, and critical illness. J Crit Care
2010;25:361. e7ee12.
[34] Chambrier C, Lauverjat M, Bouletreau P. Emulsions lipidiques: indication des
differentes emulsions lipidiques. Nut Clin Metabol 2006;20:73e8.
[35] McLeod G, Dick J, Wallis C, Patterson A, Cox C, Colvin J. Propofol 2% in critically
ill patients: effect on lipids. Crit Care Med 1997;25:1976e81.
[36] Edmunds CE, Brody RA, Parrott JS, Stankorb SM, Heyland DK. The effects of
different IV fat emulsions on clinical outcomes in critically ill patients. Crit
Care Med 2014;42:1168e77.