Aqueous parenteral nutrition (PN) provides intravenous carbohydrates, electrolytes, amino acids, and minerals to meet nutritional needs when modes of enteral nutrition are compromised. Although caloric requirements can be achieved with carbohydrates and proteins alone, patients may suffer complications ranging from essential fatty acid deficiency to developmental issues, growth restriction, and dermatitis in the absence of fat provision.
Fats are essential macronutrients that play a critical role throughout the body. Fatty acids provide sources of energy as well as structural components of cell membranes, precursors to cellular pathways and immune responses. Over the years, intravenous lipid emulsions (ILEs) have been an essential component of PN, especially in preterm and very low birth infants. Intravenous lipid emulsions have revolutionized the management of infants with gastrointestinal (GI) anomalies and diagnoses such as intestinal failure, allowing nutritional support for intestinal growth and adaptation to occur. In addition, PN and lipids are vital in supporting preterm infants when enteral therapy is contraindicated. However, prolonged PN usage still increases the risk of infection and intestinal failure associated liver disease (IFALD) in both populations which contribute significantly to infant mortality.1 Furthermore, it has been suggested certain intravenous lipid emulsions contribute to the pathogenesis of IFALD, however the pathogenesis of IFALD still remains highly complex.
In the United States, the most common standard-of-care ILE, such as Intralipid, contains 100% soybean oil, rich in omega-6 polyunsaturated fatty acids (PUFA). An alternative lipid formulation that is gaining prevalence in the literature and clinical practice for prevention of cholestasis is SMOF lipid, composed of 30% soybean oil, 30% medium chain triglycerides, 25% olive oil, and 15% fish oil. Another combination ILE that has been investigated, however is not available in the United States, is Lipofundin, composed of equal parts soybean oil and MCT oil. Intralipid and SMOF lipid share a similar dosing strategy, with patients typically receiving 3 grams/kg/day, therefore the same total calories from fat. These alternative lipid emulsions have been trialed to help prevent IFALD as they have been proposed to reduce hepatic inflammation and have also been observed to have positive benefits on other morbidities, such as retinopathy of prematurity and osteopenia. Finally, although not the focus of this review, Omegaven, a 100% fish oil-based ILE, has been studied in the pediatric population and is FDA-approved specifically for reversal of IFALD, however is not indicated for empiric therapy in parenterally fed infants. With countless new studies in the literature, the current standard of care of providing exclusive soybean-based lipid emulsions initially to all patients comes into question.
Whereas the pathogenesis of IFALD seems to be multifactorial and is not fully understood, there has been extensive research reviewing the role of ILEs contributing to inflammation. Traditional soybean oil-based lipids are high in omega-6 PUFA precursors of pro-inflammatory eicosanoids. The high amounts of omega-6 PUFA have a large linoleic acid component which leads to arachidonic acid, a substrate for interleukin-6, platelet activating factor, and tumor necrosis factor-alpha, which are all implicated as pro-inflammatory factors.2 This pro-inflammatory state leads to potentially increased lipid peroxidation, oxidative stress and inflammation, and in turn cholestasis and hepatitis. Alternative lipid emulsions that contain fish oil have rich sources of omega-3 fatty acids which have been shown to have anti-inflammatory properties. Omega-3 fatty acids are precursors for prostaglandins, leukotrienes and factors promoting resolution of inflammation. Low ratios of omega-6 to omega-3 fatty acids, as seen in SMOF lipid compared to soybean lipids, have been shown to be correlate with lower proinflammatory markers, suggesting the potential for less inflammation and cellular damage. Because of this, the opportunity of choosing alternative lipid emulsions for specific clinical scenarios is useful.
First line lipid delection – patients with GI issues
Consequently to the proposed mechanism of action and benefit, use of SMOF lipid in infants has been investigated in patients for the prevention of IFALD with evidence of early liver disease, or who were at high risk of developing liver disease secondary to a known gastrointestinal (GI) or surgical diagnosis. An early randomized evaluation of this was published in 2016 from a population of 26 infants with short bowel disease or known intestinal failure, who had progressed to hepatic dysfunction (conjugated bilirubin >1 mg/dL) requiring ongoing parenteral nutrition.3 Among these patients, the primary outcome of conjugated bilirubin during the final week of trial lipids was found to be significantly lower in the SMOF lipid group. At the endpoint of the trial, fewer patients in the SMOF lipid group had a persistent conjugated bilirubin greater than 2.9 mg/dL compared to Intralipid (27% vs. 69%, P=0.04). Together, this suggested that SMOF lipid could reduce the risk of progressive IFALD in neonates with intestinal failure, without any significant impact on growth.
In the years following, several other prospective and retrospective analyses of SMOF lipid versus soy- and soy-medium chain triglyceride (MCT)-based comparators were published. Periera-da-Silva and colleagues conducted a trial of infants born at greater than 34 weeks gestation with a history of surgery for a major GI anomaly or another anomaly impacting the function of the GI tract, who were randomized to either SMOF lipid or Lipofundin.4 The study was forced to cease recruitment due to the loss of availability of Lipofundin during the study period, however, no significant difference was detected in the incidence or severity of cholestasis or other labs associated with hepatic dysfunction. Hypertriglyceridemia was notably significantly more frequent in the Lipofundin group compared to the SMOF lipid group, however the authors could not conclude that lipid choice of SMOF versus Lipofundin was significantly beneficial on parenteral nutrition associated cholestasis. Jiang and colleagues also conducted a trial of SMOF lipid versus a Lipofundin control group in 160 infants status-post GI surgery who had received at least 2 weeks of ILEs at the time of enrollment.5 The primary outcome of postoperative weight gain was not found to be different at any of the weekly follow-up timepoints. After two weeks of study lipid, no significant differences existed in the aspartate transaminase (AST), alanine aminotransferase (ALT) or direct bilirubin, however patients who received SMOF lipid exhibited decreased ALT compared to the Lipofundin group at the end of study weeks three and four (P <0.05 for each). Significantly lower AST and direct bilirubin were also observed in the SMOF lipid group at the end of study week four (P <0.05). This study indicates that SMOFlipid may be a preferable lipid emulsion in infants who are post-GI surgery and require PN for four or more weeks. Use of a Lipofundin comparator is a complicating factor as the MCT oil component and ratios may have different impacts on IFALD than Intralipid.
Finally, Casson and colleagues conducted a retrospective review of 44 infants at two intestinal rehabilitation centers, one of which used SMOF lipid and one of which used Intralipid.6 Patients at each center remained on their primary lipid therapy unless their conjugated bilirubin levels rose to beyond 6 mg/dL, at which point they either received a reduced dose, or were initiated on 100% fish oil lipid emulsion as rescue therapy. The authors did not report a difference in the primary outcome of cholestasis (defined as a conjugated bilirubin >2 mg/dL for at least two consecutive weeks), as this outcome was observed in 91% of the Intralipid patients and 76% of the SMOF lipid patients (p=0.18). In the subgroup of infants on full PN without enteral intake (including 9 patients in the SMOF lipid group and 13 patients in the Intralipid group), there was a trend towards a higher incidence of cholestasis in infants receiving Intralipid (92% vs 78%, p=0.057). There was not a difference in the peak conjugated bilirubin level or rate of patients requiring additional interventions for cholestasis, however the conjugated bilirubin values of infants receiving Intralipid normalized slower compared to those on SMOF lipid (120 vs 83 days, p=0.04). Overall, in this mixed group of patients on at least partial PN, early SMOF lipid use was not associated with a decreased incidence or severity of cholestasis, but among patients on full PN there may be a benefit of early SMOF lipid initiation.
Together, use of SMOF lipid in patients with GI comorbidities, short bowel disease or intestinal failure may be associated with beneficial effects on prevention and resolution of IFALD. This effect may be most clinically significant in patients on prolonged durations of parenteral nutrition, or in patients who are not receiving any enteral nutrition. Limitations in these studies and others regarding sample size, uniformity of patients and adherence to protocols make interpreting and applying the data in clinical practice challenging.
First line lipid selection – preterm neonates, or before development of GI issues
As the evidence suggests, using SMOF lipid is associated with benefit in infants with GI reasons for prolonged PN use, there is understandably interest in empiric use in another patient population that is highly likely to have prolonged PN requirements: preterm neonates.
Tomsits and colleagues conducted one of the first randomized controlled trial of SMOF lipid versus Intralipid initiated in the first week of life in 60 preterm neonates (born ≤ 34 weeks gestational age).7 The authors investigated the impact of each lipid on triglyceride levels and relative change in weight in the first 8 days of lipid therapy, which were both similar between groups. This study did not address direct bilirubin values, however found no difference in total bilirubin between study entry and conclusion. The authors reported a significant difference between the gamma glutamyl-transferase (GGT) at study entry and end between groups- in the Intralipid group, GGT increased between day zero and study end, whereas in the SMOF lipid group, GGT actually decreased from day zero to the final measurement. Although this study did not directly address measures of IFALD, it demonstrated the tolerability of SMOF lipid in preterm neonates was not significantly worse than Intralipid, which is relevant as lipid intolerance manifested as hypertriglyceridemia is a noteworthy adverse effect requiring monitoring. Following this, Rayaan et al addressed a similar patient population as the prior study in preterm neonates born at <34 weeks who were expected to require a minimum of 7 days of PN.8 The primary outcome focused on triglyceride values, which were not significantly different between groups, though both groups had a significant difference between baseline and post-intervention triglyceride values, endorsing that this is an important monitoring parameter in clinical care. In addition, this study reported the direct bilirubin values following exposure of 8 days of study lipid, and found that the mean direct bilirubin the Intralipid group was significantly higher than that of the SMOF lipid group (mean (SD): 12.83 (7.2) umol/L [0.75 (0.42) mg/dL] vs 10.26 (5.8) umol/L [0.6 (0.34) mg/dL]). Among these patients, the authors similarly concluded that SMOF lipid was well tolerated and had a favorable impact on conjugated bilirubin.
The remaining available literature for use of SMOF lipid empirically in preterm neonates is retrospective, most commonly evaluating outcomes before and after an institutional policy change to utilize SMOF lipid in all preterm neonates. Choudhary, et al retrospectively reviewed 222 patients who were born at <32 weeks gestation who required ILEs for >14 days.9 The authors investigated the impact of the practice change from empiric Intralipid to empiric SMOF lipid on mortality and major neonatal morbidities. There was no significant difference in mortality or the majority of morbidities, including IFALD. The SMOF lipid group did have greater weight at 36 weeks corrected gestational age as well as a greater change in weight from birth to 36 weeks (a mean difference of each of about 125 g) though no differences were observed in length or head circumference measurements at 36 weeks. Different from other studies, the authors found comparable hepatic outcomes with SMOF lipid use. Together though, the favorable growth outcomes may someday associate with long-term benefits in developmental outcomes, but further research is needed.
A large retrospective study of 1297 very low birth weight preterm neonates who had received at least one week of ILE was conducted by Torgalkar and colleagues to evaluate the impact of SMOF lipid versus Intralipid on mortality prior to discharge and major morbidities.10 No significant difference was observed in mortality or morbidity outcomes including chronic lung disease, necrotizing enterocolitis, or infections. Time to regain birth weight was shorter in the SMOF group (median [IQR]: 11 [8, 13] days vs 8 [6, 11] days, p<0.01), and reduced rates of lipid interruptions, retinopathy of prematurity, and osteopenia of prematurity were also observed in the SMOF group. After adjusting for confounding factors, SMOF lipid was significantly associated with lower odds of lipid interruptions, retinopathy of prematurity, osteopenia of prematurity as well as neonatal cholestasis (OR (95% CI) 0.69 (0.51, 0.95)). Further prospective studies are still necessary, particularly considering the impact on cholestasis in the setting of a shorter PN duration and time to full enteral feeds in the SMOF era, but this large retrospective evaluation indicates that there is benefit of empiric SMOF lipid use in very low birth weight neonates.
An additional large retrospective cohort study including 1332 term and preterm neonates who were on an average of six days of parenteral nutrition found no significant difference in rates of IFALD, however did find significantly fewer patients with direct bilirubin values greater than 1 mg/dL and 1.5 mg/dL after SMOF lipid treatment.11 Even after adjusting for low birth weight and prematurity, the rate of IFALD remained similar between groups. The rate of liver disease was found to be different in subgroups of patients with chronic lung disease, sepsis, or concomitant hepatotoxic medication use, suggesting that there may be an opportunity for future research into the impact of SMOF lipid on cholestasis in patients specifically with those criteria or other hyper-inflammatory conditions. In addition, inclusion of term infants provides value to a more general NICU population but is different than the population included in other empiric SMOF lipid studies, which may have impacted the ultimate outcomes observed.
In summary, there is some literature support for the use of empiric SMOF lipid in preterm neonates. This is likely of additional benefit in those who are expected to have prolonged PN use (> 4 weeks), and potentially in those who are receiving hepatotoxic medications or experiencing the impacts of other concomitant inflammatory conditions, which both may contribute to further hepatic inflammation and cholestasis. Finally, it is promising that infants in the SMOF lipid groups across investigations did not experience poorer growth outcomes than those in the standard-of-care groups, supporting that beyond cholestasis, this intervention continues to support the goal of healthy neonatal growth and development.
Lipid emulsions available on the market have evolved over time, and due to new research suggesting added benefits from alternative lipid sources such as SMOF lipid, consideration of empiric use of alternative lipid sources should be addressed. Although SMOF lipid has been approved by the FDA for use in adults, approval has not yet been granted for in infants by the FDA as large randomized trials are awaited. Previous systematic reviews evaluating lipid emulsions for term, late preterm and preterm infants have been inconclusive in determining if there are added benefits, most likely due to the lack of highly powered data and specific patient inclusion criteria.12,13 The expanding body of literature continues to suggest potential benefits on cholestasis outcomes from alternative lipid emulsions containing fish, MCT, olive and soybean oil, as well as no overall clinical harm, with favorable impacts on triglycerides and growth. Consideration of patient-specific characteristics that were most associated with benefit in the studies is a reasonable starting point when assessing use of SMOF lipid in an off-label manner, particularly patients with GI diagnoses, anomalies, or surgeries that are anticipated to require PN for a prolonged period (> 4 weeks), and those preterm neonates that are anticipated to require prolonged PN, regardless of a GI diagnosis.
1. Goulet OJ, Cai W, Seo JM. Lipid Emulsion Use in Pediatric Patients Requiring Long-Term Parenteral Nutrition. JPEN J Parenter Enteral Nutr. 2020;44 Suppl 1:S55-S67. doi:10.1002/jpen.1762.
2. Hojsak I, Colomb V, Braegger C, et al. ESPGHAN Committee on Nutrition Position Paper. Intravenous Lipid Emulsions and Risk of Hepatotoxicity in Infants and Children: a Systematic Review and Meta-analysis. J Pediatr Gastroenterol Nutr. 2016;62(5):776-792. doi:10.1097/MPG.000000000000112.
3. Diamond I, Grant R, Pencharz P et al. Preventing the Progression of Intestinal Failure–Associated Liver Disease in Infants Using a Composite Lipid Emulsion: A Pilot Randomized Controlled Trial of SMOFlipid. Journal of Parenteral and Enteral Nutrition. 2016;41(5):866-877. doi:10.1177/0148607115626921.
4. Pereira-da-Silva L, Nóbrega S, Rosa M et al. Parenteral nutrition-associated cholestasis and triglyceridemia in surgical term and near-term neonates: A pilot randomized controlled trial of two mixed intravenous lipid emulsions. Clin Nutr ESPEN. 2017;22:7-12. doi:10.1016/j.clnesp.2017.08.007.
5. Jiang, W., Chen, G., Zhang, J. et al. The effects of two mixed intravenous lipid emulsions on clinical outcomes in infants after gastrointestinal surgery: a prospective, randomized study. Pediatr Surg Int 35, 347–355 (2019).
6. Casson C, Nguyen V, Nayak P et al. A Comparison of Smoflipid® and Intralipid® in the Early Management of Infants with Intestinal Failure. J Pediatr Surg. 2020;55(1):153-157. doi:10.1016/j.jpedsurg.2019.09.073.
7. Tomsits E, Pataki M, Tölgyesi A, Fekete G, Rischak K, Szollár L. Safety and Efficacy of a Lipid Emulsion Containing a Mixture of Soybean Oil, Medium-chain Triglycerides, Olive Oil, and Fish Oil: A Randomised, Double-blind Clinical Trial in Premature Infants Requiring Parenteral Nutrition. J Pediatr Gastroenterol Nutr. 2010;51(4):514-521. doi:10.1097/mpg.0b013e3181de210c. A
8. Rayyan M, Devlieger H, Jochum F, Allegaert K. Short-Term Use of Parenteral Nutrition With a Lipid Emulsion Containing a Mixture of Soybean Oil, Olive Oil, Medium-Chain Triglycerides, and Fish Oil. Journal of Parenteral and Enteral Nutrition. 2012;36(1_suppl):81S-94S. doi:10.1177/0148607111424411.
9. Choudhary, N., Tan, K. & Malhotra, A. Inpatient outcomes of preterm infants receiving ω-3 enriched lipid emulsion (SMOFlipid): an observational study. Eur J Pediatr 177, 723–731 (2018).
10. Torgalkar, R., Dave, S., Shah, J. et al. Multi-component lipid emulsion vs soy-based lipid emulsion for very low birth weight preterm neonates: A pre-post comparative study. J Perinatol 39, 1118–1124 (2019). A
11. Stramara L, Hernandez L, Bloom B, Durham C. Development of Parenteral Nutrition–Associated Liver Disease and Other Adverse Effects in Neonates Receiving SMOFlipid or Intralipid. Journal of Parenteral and Enteral Nutrition. 2020. doi:10.1002/jpen.1774.
12. Kapoor V, Malviya MN, Soll R. Lipid emulsions for parenterally fed preterm infants. Cochrane Database of Systematic Reviews 2019, Issue 6. Art. No.: CD013163. DOI: 10.1002/14651858.CD013163.pub2.
13. Kapoor V, Malviya MN, Soll R. Lipid emulsions for parenterally fed term and late preterm infants. Cochrane Database Syst Rev. 2019;6(6):CD013171. Published 2019 Jun 4. doi:10.1002/14651858.CD013171.pub2.