Administration of an Intravenous Fat Emulsion Enriched with Medium-Chain Triglyceride/ω-3 Fatty Acids is Beneficial Towards Anti-Inflammatory Related Fatty Acid Profile in Preterm Neonates: A Randomized, Double-Blind Clinical Trial

Intravenous administration of pure soybean oil emulsions high in linoleic acid may lead to inflammation and lipid peroxidation in preterm neonates. We aimed to investigate the effects of a medium-chain triglyceride (MCT)/ω-3 polyunsaturated fatty acid (PUFA)-enriched intravenous fat emulsion (IVFE) on plasma fatty acid (FA) profile and serum interleukin-6 (IL-6) in preterm neonates. In this double-blind randomized study, 92 preterm neonates (gestational age < 32 weeks, birth weight < 1500 g) were assigned to receive either MCT/ω-3 PUFA-enriched IVFE (Intervention Group) or soybean oil-based IVFE (Control Group). Levels of FAs were measured at baseline (day 0) and day 15 of parenteral nutrition with gas-chromatography mass-spectrometry. Serum IL-6 was measured with sandwich ELISA in 59 neonates. Plasma FAs changed significantly over time; the MCT/ω-3 PUFA-IVFE group showed higher ω-3 PUFAs (p = 0.031), eicosapentaenoic acid (p = 0.000), and oleic acid (p = 0.003), and lower ω-6/ω-3 PUFAs ratio (p = 0.001) and ω-6 PUFAs (p = 0.023) compared to control group. Linoleic acid was higher in the soybean oil (SO)-based IVFE arm compared to the MCT/ω-3 PUFAs-IVFE arm (p = 0.006). Both fat emulsion types decreased IL-6 compared to baseline, but changes were insignificant between groups. Administration of MCT/ω-3 PUFA-enriched IVFE in preterm neonates is beneficial in changing the FA profile consistent with attenuated inflammatory response.


Introduction
Preterm neonates are in an energy-deficient state due to oxygen desaturation and hypoxia events, painful and stressful stimuli, illness, and rapid neurodevelopment [1,2]. Inadequate energy supply may be associated with impaired growth, increased severity of postnatal diseases, and adverse neurodevelopment [3][4][5]. Administration of emulsions enriched with fatty acids can provide high

Participants
Preterm neonates with gestational age <32 weeks and birth weight <1500 g that were admitted to a tertiary neonatal intensive care unit (between February 2017 and February 2018), within 12 h after birth, were assessed for eligibility. Exclusion criteria were anticipated needs for parenteral nutrition (PN) at >70% of total daily energy for <10 days, evidence of intrauterine infection, perinatal asphyxia, major congenital anomalies, and refusal of parental consent.

Bioethics
The study protocol was approved by the Scientific and Ethical Committee of (removed for blind review) (#355/15-2-18). The study was conducted in accordance with the Declaration of Helsinki. Neonates' parents gave their informed consent for inclusion before they participated in the study. This work is registered in ClinicalTrials.gov (Protocol Record 201802).

Intervention Procedures
The preterm neonates were randomly allocated to one of the following two groups: the intervention group, which received an intravenous fat emulsion enriched with MCT/ω-3 PUFAs, and the control group, receiving a SO-based formula. Fat emulsion was added in the PN solution on the first or second day of life at a dose of 1 g/kg/day, which was increased by 1 g/kg/day up to a maximum amount of 3 g/kg/day.
In the intervention group (MCT/ω-3 PUFA-IVFE group), the source of parenteral fat was Smoflipid (Fresenius Kabi HELLAS, Athens, Greece), a formulation containing MCTs, soybean oil, olive oil, fish oil, and alpha-tocopherol. The conventional soybean oil-based lipid formulation Intralipid 20% (Fresenius Kabi HELLAS, Athens, Greece) was administered to the control group (SO-IVFE group). Compositions of the two fat emulsions are presented in Table S1. Amino acids derived from Vamin Infant (Fresenius Kabi HELLAS, Athens, Greece) were administered in both groups.
Neonates were fed exclusively with PN for at least the first 3 days of life. Then, enteral feedings were provided in volumes that were determined by neonates' weight and energy requirements on a daily basis. Parenteral solutions were administered to neonates (in both groups) until oral feedings reached a minimum of 80% of total energy intake. Eligibility, based on the inclusion and exclusion criteria, was assessed by the neonatologists of the neonatal intensive care unit.

Clinical Data
Before the start of the trial, the following data were collected: Gestational age, birth weight, perinatal history, and neonatal medical conditions and treatment. In addition, parenteral and enteral nutrition intakes, as well as changes in body weight, were recorded daily. Other parameters such as white blood cell and platelet counts, hematocrit, and C-reactive protein levels were assessed only if it was indicated for clinical reasons. Measurements were performed at baseline (day 0) and day 15 of PN. Blood samples were obtained after PN was temporally ceased for 4 h and before oral feeding.

Fatty Acid Assessment with Gas Chromatography
Plasma was separated from ethylenediaminetetraacetic acid treated blood samples after centrifugation at 1500 g for 15 min at 4 • C and were stored at −80 • C for further analyses. Initially, plasma samples were thawed and homogenized with stirring using Vortex. They were transferred to screw-on glass test tubes and then cooled to −80 • C and dehydrated by lyophilization. Fatty acid methyl esters were prepared by the method of Lepage G. and Roy C.C. (1986), modified by Rodríguez-Palmero et al. (1997) using acetyl chloride and methanol [27,28]. Agilent HP6890 gas chromatograph equipped with a Flame Ionization Detector (FID), MS 6890 Mass Selective Detector (MSD), and HP 7673 auto sampler was used. Methyl esters of the fatty acids were separated on a SGE BPX 70 capillary column, 60 m in length, 0.25 mm in diameter, and 0.25 µm in inner thickness. Helium was used as a carrier gas at a flow rate of 0.8 mL/min while the temperature of the sample imager and detector was 230 • C and 290 • C, respectively.
For the analysis, 1 µL of sample was injected into the split mode chromatograph at a ratio of 20:1. Thus, finally 1/20 of 1 µL was introduced into the column. The fatty acid retention times were initially recorded by analyzing 37 multi-standard fatty acid methyl esters of Sigma (Sigma L9405, Sigma-Aldrich, Seelze, Germany) using the MSD and with the help of the NIST and WILEY electronic libraries. The flame ionization detector was then used and the standards were intermixed at regular intervals to record any slight changes in the retention times of the constituents and to calculate the response factors (RFs) of the various fatty acids, so that the areas of the corresponding chromatographic peaks were corrected. The relative proportions of fatty acids (% of total fatty acids) were determined by the completion of chromatographic peaks and corrected with the use of RFs.

Serum Interleukin-6 Measurement
Serum isolation was performed in a subset of the enrolled neonates (n = 59) due to evidence of hemolysis. Sera were separated from blood samples after centrifugation at 1800 g for 10 min at 4 • C and were stored at −80 • C for further analyses.
Sandwich Enzyme Linked-Immunosorbent Assay (ELISA) kit was used to measure serum IL-6 levels (pg/mL) according to the manufacturer's instructions (R&D Systems Inc., Abingdon, UK).

Randomization, Sequence Generation, and Implementation
The sample randomization was conducted by a computer-generated randomization list. A pharmacist received the list, prepared the different parenteral formulations in identical bags, and finally assigned each neonate to control or intervention group. The pharmacist was not involved in neonates' care. All medical personnel and participants were blinded to treatment assignment during the whole study period.

Primary Outcomes and Sample Size
Our primary outcome was the detection of clinically significant differences in plasma ω-3 PUFAs, ω-6 PUFAs and EPA in the MCT/ω-3 PUFA-enriched IVFE group compared to the SO-IVFE group post-intervention. Based on previously published work of our research team [29], a minimum sample size of 14 neonates (7 per arm) was sufficient to result in a clinically important difference of 2 in plasma ω-3 PUFAs change between the control-[standard deviation of mean (SD) = 1] and the MCT/ω-3 PUFAs arm (SD = 1.5) using a two-tailed t test with 80% power and a 5% level of significance. Similarly, to detect a significant difference of 2 in total ω-6 PUFAs, 50 neonates (25 per arm) were needed (control: SD = 2.5, intervention: SD = 2.5). Additionally, a significant difference of 0.5 in EPA required 20 neonates per arm (control: SD = 0.5, intervention: SD = 0.6). Secondary outcomes of our study were significant changes in serum IL-6, plasma ω-6/ω-3 PUFAs ratio, and levels of linoleic acid, DHA, and oleic acid. A minimum sample size of 318 neonates (159 per arm) was needed to observe a significant difference of 3.5 in serum IL-6 concentration between the two groups post-intervention (control: SD = 13, intervention: SD = 11.5).

Statistical Analysis
All analyses were conducted by applying the Statistical Package for the Social Sciences (SPSS 21.0 for Windows, Chicago, IL, USA). Descriptive statistics were calculated for all parameters and the Kolmogorov-Smirnov test was applied to investigate if all measures were characterized by normal distribution. Parametric data are expressed as mean values (±SD), while non parametric data are expressed as medians and interquartile ranges. For variables with normal distribution, the independent samples t-test was applied to compare the differences between the two arms pre-and post-intervention, while for variables without normal distribution, the Mann-Whitney test was applied. Before the intervention, this test served to ensure that the study population was characterized by homogeneity. For investigating possible intra-group differences, a paired samples t-test was applied for parametric variables and the Wilcoxon test for non-parametric ones. For all statistical analyses, significance was set at p < 0.05.

Results
A total of 92 preterm neonates were included and completed the study: 46 in the control and 46 in the intervention group. Clinical characteristics of each study group at baseline are shown in Table 1. No statistically significant differences between the two groups were observed for gestational age (p = 0.126), birth weight (p = 0.995), serum IL-6 concentration (p = 0.383), and plasma fatty acid (FA) profile.
Body weight increased significantly in both the control (1331.30 ± 237.56 g vs. 1223.04 ± 215.42 g, p = 0.000) and the intervention arm (1339.05 ± 211.02 g vs. 1222.74 ± 211.90 g, p = 0.000) at day 15 compared to baseline. The increase in body weight did not differ between groups (p = 0.722). Safety and tolerability: both IVFEs were well tolerated. Serum triglyceride levels were within normal values for age, whereas no local reaction, thrombocytopenia that could be attributed to IVFE, or dropout related to any IVFE-associated adverse effect were observed.

Dietary Intake
The energy intake via parenteral nutrition, as well as the amount of enteral feeding of preterm neonates in the control (SO-IVFE) and the intervention (MCT/ω-3 PUFAs-IVFE) group, are presented in Table 2. No significant differences were observed at baseline and at the endpoint of the study between the two groups. At baseline, no neonate received enteral feeding (maternal milk or formula). Nutrient and mineral intake via enteral and parenteral routes did not differ at day 15 between the two arms (Table S2).

Plasma Fatty Acids
Plasma fatty acid profiles of the two study arms pre-and post-intervention are presented in Table 3. Parenteral nutrition significantly increased levels of total PUFAs at day 15 compared to baseline, both in the control (p = 0.000) and the intervention arm (p = 0.000). However, the change in total PUFAs did not differ between groups post-intervention (p = 0.099). Table 3. Differences in serum fatty acid profile at baseline (day 0) and at the endpoint of intervention (day 15).  Table 3. Cont.

Polyunsaturated Fatty Acids
In regards to plasma total ω-6 PUFAs, both groups showed significant elevations at day 15 compared to day 0, but the raise was significantly lower in the MCT/ω-3 PUFAs group than the SO group (p = 0.023). Augmentation of total ω-3 PUFAs in the intervention arm did not reach statistical significance (p = 0.070), but post-intervention, the mean difference between arms was significant (p = 0.031). In the control arm, total ω-3 PUFAs did not change throughout the trial (p = 0.131).
The ω-6/ω-3 PUFAs ratio was not altered in the intervention group (p = 0.202), but it was significantly increased in the SO group (p = 0.000). This change remained significant when compared to the MCT/ω-3 PUFAs group post-intervention (p = 0.001).
Among ω-6 PUFAs, levels of LA (C18:2ω-6) were significantly increased in both the SO group (p = 0.000) and the intervention group (p = 0.000) compared to baseline, but the increment was more profound in the control than the intervention group (p = 0.006). Concentration of γ-linolenic acid (GLA) (C18:3ω-6) was raised only in the SO arm (p = 0.040), but this increase was not significant when compared to the MCT/ω-3 PUFAs arm post-intervention (p = 0.231). At day 15, AA (C20:4ω-6) concentration was significantly decreased in both the control (p = 0.000) and intervention groups (p = 0.000) compared to baseline, but changes between groups were found insignificant (p = 0.204).

Monounsaturated Fatty Acids
Total MUFAs did not change throughout the study (control arm: p = 0.390; intervention arm: p = 0.575). Nevertheless, oleic acid (OA) (C18:1ω-9) was significantly elevated in the intervention arm (p = 0.000) but not in the control arm (p = 0.933) at day 15 compared to day 0, and this increase remained significant when compared to control group post-intervention (p = 0.003).

Discussion
In this double-blind controlled clinical trial, premature neonates were randomly allocated to receive either a mixed intravenous fat emulsion containing MCTs, soybean oil, olive oil, and fish oil (MCT/ω-3 PUFAs-IVFE), or the conventional soybean oil-based IVFE (SO-IVFE), starting on the first or second day after birth.
The results of our study showed that parenteral treatment with MCT/ω-3 PUFAs-IVFE for 15 days led to significantly elevated plasma total ω-3 PUFAs, EPA, and oleic acid concentrations compared to the SO-IVFE group. Furthermore, after 15 days of parenteral nutrition, significantly lower ω-6/ω-3 PUFAs ratio and ω-6 PUFAs were observed in the intervention versus the control group. The raise of LA and ALA levels were found to be higher in the soybean oil-based IVFE compared to the MCT/ω-3 PUFAs-IVFE. The fat emulsion type, however, did not affect serum IL-6 change.
The high LA and ALA content of pure soybean oil emulsions can induce low blood concentrations of their bioactive LC-PUFA metabolites, especially EPA and DHA, and they may be associated with increased rates of infection and lipid peroxidation, exacerbating oxidative stress [30][31][32]. Compared to treatment with SO-based emulsions, administration of lipid emulsions rich in MCT/LC-PUFAs and low in LA and ALA resulted in similar essential fatty acid content in plasma phospholipids and triacylglycerols of preterm neonates [31]. Moreover, excessive intake of LA may promote a persistent inflammatory state that contributes to progressive hepatocyte damage and/or portal inflammation, leading to cholestasis and fibrosis [31]. Literature suggests that a fish oil-containing lipid emulsion including pure fish oil instead of soybean oil alone may be beneficial in prevention of cholestasis in preterm neonates with parenteral nutrition-associated liver disease/cholestasis [33][34][35][36][37][38]. Moreover, pure SO-containing lipid emulsion contains low amounts of the antioxidant a-tocopherol [39]. The low a-tocopherol content may further enhance deleterious lipid peroxidation of the high parenteral PUFA supply. According to recent randomized controlled trials, the MCT/LC-PUFA lipid emulsions enriched with fish and olive oil attenuates the production of pro-inflammatory cytokines and increases total antioxidant capacity in preterm neonates [29,40,41]. In our study, neonates receiving either MCT/ω-3 PUFAs or SO-based lipid emulsions demonstrated significant reductions in serum IL-6, but no significant difference was found between groups.
There are very few clinical trials that investigated the effects of SO-based IVFE and MCT/ω-3 PUFAs-IVFE on the fatty acid profile of preterm neonates. A randomized control trial by Vlaardingerbroek and co-workers (2014) showed that the concentrations of EPA and DHA in triglycerides and phospholipids were significantly higher in the MCT/ω-3 PUFAs compared to the SO group, on days 6 and 14 of parenteral feeding [42]. Other studies showed that the MCT/ω-3 PUFAs-IVFE leads to a lower ω-6/ω-3 fatty acids ratio [43] and lower AA [44] than the soybean oil emulsion.
To our knowledge, this is the first study showing a favorable effect towards plasma oleic acid concentration after the administration of MCT/ω-3 PUFAs-IVFE, compared to SO-IVFE. In a previously published study conducted by our research team, we observed a rise of plasma oleic acid in the MCT/ω-3 PUFAs group, but no difference was evident between the two groups post-intervention [29]. Another asset of the present study was the detection and quantification of numerous plasma fatty acids after the administration of both MCT/ω-3 PUFAs and SO-IVFE.
Nevertheless, our study has some limitations. Primarily, the sample size of the study is relatively small due to low availability of preterm neonates needing parenteral nutrition for 15 days. The amount of blood that could be obtained from preterm neonates was limited; consequently, it was impossible to measure serum IL-6 levels in the overall study population or to assess more inflammatory mediators that would probably strengthen our conclusions. The effects of MCT/ω-3 PUFA-enriched IVFE on a combination of serum inflammatory mediators and fatty acid profiles, as well as the associations with nutritional status, morbidity, and long-term outcomes, should be further investigated, in order to determine the safety and the efficacy of this type of IVFE.

Conclusions
The present study suggests that administration of MCT/ω-3 PUFA-enriched IVFE in preterm neonates is associated with fatty acid profiles consistent with attenuated inflammatory response.
The results indicate that the administration of MCT/ω-3 PUFA-IVFE may be beneficial in preterm neonates (for whom parenteral nutrition is necessary) and who are at risk of developing ω-3 PUFAs deficiency. Studies including greater sample sizes and more data such as long-term outcome parameters are of paramount importance in order to determine the efficacy and safety of MCT/ω-3 PUFAs-IVFE administration on preterm neonates.
Author Contributions: P.P. and T.S. conceptualized the study. P.P., M.S., D.N., and T.S. contributed to methodology. P.P. and O.-N.K. performed investigation. A.G. was responsible for data curation. P.P. and D.N. carried out writing-Original draft preparation. A.G. and D.N. contributed to writing-Review and editing. Supervision was carried out by M.S and T.S. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.