An Overview of Parenteral Nutrition from Birth to Adolescence, Based on a Composite Fish Oil Containing Lipid Emulsion and a Pediatric Amino Acid Solution

1. Introduction

Adequate nutrient supply is of greater relevance during childhood than at any other period in life. Due to their high basal and anabolic requirements and limited metabolic reserves, pediatric patients are particularly sensitive to energy and protein restriction [1]. Sufficient nutrients must be provided not only for the maintenance of body tissue, but also for growth, while starvation, even for few days, may be detrimental [1,2,3].

Parenteral nutrition (PN) has become a valid therapeutic option for patients with intestinal failure (IF) who require long-term PN and home-PN (HPN) [1,4,5]. The provision of PN to pediatric patients represents a distinct clinical challenge, as metabolic demands and nutritional needs change according to age, growth velocity, and clinical situations [3]. Moreover, the complex and rapidly changing fluid and electrolyte requirement is a major challenge to PN delivery [6].

In children, IF is defined as a critical reduction of functional gut mass below the minimum needed for adequate growth [7,8]. The leading cause of IF in childhood is short bowel syndrome (SBS) as a consequence of extensive small bowel resection. Other conditions of IF are neuromuscular disorders (chronic intestinal pseudo-obstruction and long segment Hirschsprung disease) and mucosal intestinal diseases [8]. Intestinal failure may be reversible or irreversible, depending on the underlying cause.

Long-term PN and home PN (HPN) are the mainstay for the management of chronic IF, that allow children to grow in their familiar environment [9].

This article is based on the sourcing of PubMed (without time restriction), Embase (since 2016), and Google Scholar (since 2018), with the last search conducted in June 2022. This narrative review paper aims to give an overview of the specific quantitative and qualitative nutritional intake in pediatric patients including term newborn infants, infants/toddlers, children, and adolescents receiving PN. Preterm infants with their highly specific demands are not included, as this would be beyond the scope of the present article. Of note, this specific population provided important data not only for developing crystalline amino acids solutions (CAASs) specifically designed for preterms as well as for the demonstration of safety and efficiency of the composite intravenous lipid emulsion (ILE) containing fish oil (FO). Moreover, this article explores the clinical experience regarding the use of a composite ILE containing FO (SMOFlipid®, Fresenius Kabi Bad Homburg, Germany) and a pediatric CAAS (Vaminolact®, Fresenius Kabi Bad Homburg, Germany) in the management of pediatric patients with acute or chronic IF depending on PN.

2. Provision of Energy and Macronutrients in Pediatric Patients on PN

Energy

Generally, energy supply should meet the requirements to cover the resting energy expenditure (REE), plus those for physical activity, growth, diet induced thermogenesis and the correction of pre-existing malnutrition when present [3,10]. ESPGHAN/ESPEN/ESPR guidelines recommend calculating total parenteral energy requirements of stable patients from the REE that may be either measured by indirect calorimetry or estimated from Schofield’s formula, which are based on gender, age, body weight, and length [11]. Energy intake must cover energy expenditure for physical activity and growth (including catch-up growth) and must be adjusted according to age and disease states [3].

In clinical practice, PN intake must be adapted to the degree of intestinal insufficiency in pediatric patients. It may be estimated by the level of PN required for normal or catch-up growth. The PN dependency index (PNDI) has been defined as the ratio of non-protein-energy intake (NPEI) provided by PN for achieving optimal growth to REE calculated by using the Schofield formula, expressed as a percentage and validated from several papers [12,13,14,15]. In clinical practice, the PNDI corresponds to the caloric requirements from PN to achieve normal growth. It reflects the degree of intestinal insufficiency more accurately than other criteria, such as the ratio of parenteral versus enteral intakes or the addition of PN and enteral intake when compared to recommended dietary allowances (RDA). The PNDI for estimating the PN requirements for optimal growth makes sense by considering the final nutritional efficiency. Based on the PNDI expressed in percentage, it can be considered as high when >120% and even higher for achieving catch-up growth and low when <50% [12,13,14,15].

Excess energy intake may increase the risk of hyperglycemia, infections, and fat deposition (fat body mass and liver steatosis). Underfeeding may result in failure to thrive, poor development, impaired immune responses, increasing the risks of morbidity, and mortality in newborn infants, infants/toddlers, and children [3]. In acute critical illness, energy provision should be reduced compared to stable patients while overfeeding (glucose) increases ventilatory workload and potentially prolongs the need for mechanical ventilation and the length of hospital stay [3,16]. During the stable phase of critical illness, requirements are approximately 1.3 times the REE and further increase in the recovery phase for catch-up growth [3]. Therefore, energy provision must be carefully managed to ensure that energy needs are met but not exceeded.

Glucose

D-glucose (dextrose) is the main source of calories in PN, but also accounts for a major part of the osmolarity in PN solutions. All body cells and organs utilize glucose and for the brain, renal medulla and erythrocytes, glucose is an obligatory energy substrate [17].

Factors guiding glucose intake with PN include age, disease status, nutritional status, the concomitant provision of other macronutrients (lipids and amino acids). The rate of glucose delivery should not exceed the maximum rate of glucose oxidation. Studies have shown a maximal oxidation rate of about 12 mg/kg body weight (BW) per min (17.2 g/kg BW per day) in term infants after surgery or infants on long-term PN [17,18,19].

The higher the glucose delivery rate, the higher is hyperinsulinism favoring a costly lipid synthesis from glucose (lipogenesis) with subsequent lipid deposition in adipose tissue and liver. Moreover glucose overfeeding induces hyperglycemia and increased CO₂ production and minute ventilation [17]. In critically ill children with insulin resistance and beta-cell dysfunction, risk of hyperglycemia is a particular concern [17,20] and represents an independent risk factor for morbidity and mortality [21,22].

Lipids

Quantitative requirements

Intravenous lipid emulsions are an integral component of pediatric PN [23]. ILEs are a low osmolarity and energy-dense source of calories. In addition, they serve as a source of essential fatty acids (EFAs) and facilitate the delivery of fat-soluble vitamins [24,25]. According to the ESPPGHAN/ESPEN/ESPR guidelines [23], lipid intake should provide 25–50% of the non-protein calories in pediatric patients on total PN. As for glucose, the maximal lipid oxidation rate has been established [26]. It decreases with age, from approximately 3 g/kg/day in young children to 1.7-2.5 g/kg/day in adults. According to ESPGHAN/ESPEN/ESPR guidelines [23], parenteral lipid intake in children should be limited to a maximum of 3 g/kg/day. (LoE 3e4, RG 0, conditional recommendation for, strong consensus).

Qualitative fatty acids requirements

Not only the quantity of lipids but also the fatty acid (FA) composition of the ILE used as part of PN is critical, as it may influence clinical outcomes in pediatric patients [27]. To prevent biochemical evidence of EFA deficiency, parenteral lipid intake should include 0.1 g/kg BW/day of the omega-6 polyunsaturated fatty acid (PUFA) linoleic acid (LA) in term infants and children [23]. The “historical” ILEs, subsequently referred to as “first generation”, were derived from pure soybean oil (SO) rich in LA. A potential disadvantage with pure SO ILEs is that high contents of omega-6 PUFAs, in particular of LA, have been attributed to pro-inflammatory and immunosuppressive properties. This led to the development of more advanced ILEs containing mixtures of SO with alternative oil sources such as medium-chain triglycerides (MCT) from coconut/palm kernel oil, olive oil (OO) and/or FO [25,28]. In particular, FO representing a source of omega-3 PUFAs with anti-inflammatory properties [29] has become an important component of the last generation of composite ILEs containing FO [28,30]. The long-chain omega-3 PUFAs docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) and the omega-6 PUFA arachidonic acid (ARA) modulate key metabolic pathways, including inflammatory and immune responses, coagulation, and cell signaling [28,31]. DHA and ARA are also required for an optimal development of brain as well as for cognition and visual acuity [32,33,34]. In order to prevent EFA deficiency in term infants to children, a lipid emulsion dosage providing a minimum LA intake of 0.1 g/kg/day can be given, which also provides an adequate intake of linolenic acid (LNA) with all 20% ILEs currently registered for pediatric use (LoE 3-4, RG 0, conditional recommendation for, strong consensus). In critically ill pediatric patients, ILE should also be an integral part of PN. Composite ILEs containing FO may be used as the first-choice treatment. Available evidence raises the important question on the best timing to provide PN support in critically ill children, but do not allow to differentiate potential effects on outcomes of the timing of introducing ILE (LoE 4, GPP, conditional recommendation for, strong consensus) [21].

The ESPGHAN/ESPEN/ESPR guidelines [23] support that in newborn infants and older children on short-term PN, pure SO ILEs may provide a less balanced FA profile than composite ILEs. For PN lasting longer than a few days, the use of pure SO-ILE is discouraged and composite ILEs with or without FO are recommended as first-line treatment [23.35]. Of note, to date there are no quantitative recommendations for the supply of EPA, DHA, or ARA in pediatric patients.

Intestinal failure associated liver disease (IFALD) is the most prevalent complication affecting children with IF receiving long-term PN [24,36,37]. The diagnosis of IFALD is usually based on the onset of cholestasis, generally defined as an elevation in conjugated serum bilirubin (CBil) concentration (≥2 mg/dL) [24,37]. Effects of pure FO based ILEs (Omegaven®, Fresenius Kabi Bad Homburg, Germany) for reversing cholestasis in children receiving a SO based ILE (Intralipid®, Fresenius Kabi Bad Homburg, Germany) have first been reported by Gura et al [38]. Ten years later, the same group reported data of patients treated with pure FO from 2004 to 2014. Most of the time, cholestasis appeared very early during the first year of life. Resolution of cholestasis was defined as sustained direct bilirubin (DBil) <2 mg/dL, and treatment failure as liver transplantation or death while DBil was >2 mg/dL. Among 182 patients treated with FO, 86% achieved resolution of cholestasis and 14% failed therapy. Factors significantly associated with therapy failure included lower birth weight, patients older than 20.4 weeks (9.9, 38.6 weeks) compared with 11.7 weeks (7.3, 21.4 weeks), and more advanced liver disease at therapy initiation compared to patients whose cholestasis resolved [39].

Provision of high amounts of the omega-6 PUFA LA with pure SO ILEs results in the endogenous formation of ARA, a precursor of lipid mediators with more pro-inflammatory properties [28,40]. Excessive intake of LA may thus promote an inflammatory state contributing to cholestasis and liver fibrosis [31,40]. Moreover, pure SO ILEs are relatively low in vitamin E (alpha-tocopherol) as antioxidant substrate and, thus, are prone to the formation of lipid peroxidation products, which may lead to macrophage activation and hepatocyte damage [24,41,42]. Finally, pure SO based ILEs also contain much higher concentrations of phytosterols compared with any of the other ILEs, especially composite ILEs containing FO. There is evidence that phytosterols may accumulate in the liver, thereby reducing bile secretion while promoting liver damage in children with IF [24,43]. Recent discovery of farnesoid X receptor (FXR) as possible target mediating altered bile secretion in SBS patients has opened new field of research for IFALD prevention [44,45].

In this regard, composite ILEs with FO offer several advantages, including high concentrations of the anti-inflammatory omega-3 PUFAs DHA and EPA and the antioxidant alpha-tocopherol (200 mg/L), reduced omega-6 PUFA content, and a reduced phytosterol load [23,24,40]. Moreover, some composite ILEs, containing FO, also provide rapidly oxidizable MCTs [46,47] while the diversity of oils ensures a balanced intake of FA and prevents EFA deficiency [36,37].

To reverse IFALD in pediatric patients, ESPGHAN recommends a discontinuation of SO ILE, a reduction of other ILE dosage and/or the use of composite ILE with FO, along with the treatment and management of other risk factors (LoE 2b, RG B, strong recommendation). The use of pure FO ILE is not recommended for general use in pediatric patients but may be used for short-term rescue treatment in patients with progression to severe IFALD, based on case reports (LoE 3e4, GPP, conditional recommendation for, strong consensus).

Amino acids and protein synthesis

Quantitative requirements

In any patient receiving PN, proteins have to be administered in the form of free amino acids (AAs). Pediatric CAASs are currently available. Their primary goal is to promote anabolism (protein synthesis and positive nitrogen balance) for cells and tissue building (e.g., muscle, skeleton) consistent with the normal growth and development typical of healthy peers [48]. Due to additional needs for growth, the protein requirements based on weight are higher in newborn infants, infants/toddlers, and children than in adults and the timely provision of adequate amounts of AAs with PN is particularly critical [1].

Generally, the total AA requirement is lower in parenterally fed newborn infants, infants/toddlers, and children than in enterally fed patients because the intestinal and hepatic “first-pass” metabolism is bypassed [49]. The ESPGHAN/ESPEN/ESPR guidelines [49] recommend a minimum AA intake of 1.0 g/kg/d for stable infants and children from 1 month to 3 years to avoid negative balance (LOE 1¡, moderate quality, RG B, strong recommendation, strong consensus). In stable children aged 3-12 years an AA intake of 1.0-2.0 g/kg per day may be considered. (LOE 4, RG GPP, conditional recommendation, strong consensus) [47]. The maximum AA intake should not exceed 3.0 g/kg BW/day in term newborn infants and 2.0 g/kg BW/day in stable children and adolescents [49]. Excessive AA intake has been shown to increase oxidation, in turn promoting high blood urea concentrations and metabolic acidosis [50]. In very low birth weight (VLBW) newborn infants, high AA and energy intakes from birth onwards result in a more anabolic state but may cause higher urea concentrations, reflecting a higher AA oxidation [51]. Of note, AA must always be co-administered with sufficient amounts of energy in the form of glucose and lipid to prevent wasting of AAs for endogenous glucose production. The optimal ratio of NPEI in kcal to 1 gr of nitrogen (N) decreases with age, from approximately 250:1 in infants below 1 year of age to 150:1 in adolescents [13].

Qualitative requirements

In the sixties at the early phase of PN development, newborn infants received protein hydrolysates as a source of nitrogen [52,53]. These protein hydrolysates were not only nutritionally unadapted but raised the risk of severe allergic reactions and were therefore abandoned. With the advent of technology for the production of pure L-AAs, a second generation of CAAS was developed, but much more adapted for adult than pediatric patients and especially for VLBW newborn infants. Extrapolation from data on oral nutrition was often unsatisfactory for the adequate formulation of such mixtures. Intestinal metabolism of specific AAs is highly variable and may also change with age [49]. For example, the parenteral requirements for the branched-chain amino acids (BCAAs) leucine, valine, and isoleucine are considerably lower compared to enteral requirements, as losses due to first-pass metabolism with enteral administration were shown to be substantial both in the newborn state [54] and in older children [55]. On the other hand, parenteral arginine requirements are higher than with enteral nutrition, as intestinal metabolism results in a net synthesis of arginine [56]. Moreover, parenteral requirements for tyrosine and cysteine are increased due to the lacking conversion of phenylalanine and methionine in the intestine that would occur after enteral intake [56]. Accordingly, the composition of CAASs for pediatric use should ideally be adapted to the specific requirements resulting from the lack of intestinal “first-pass” metabolism. In other words, composition of AAs solution for PN cannot copy the enteral intake.

In the 80’s, great attention has been given to AA supply in VLBW infants from the first day of life in order to avoid catabolism, establish anabolism, achieve in utero protein accretion rates, and promote linear growth. Numerous studies have been performed for developing CAASs with the aim of resembling the plasma AA patterns of normally growing, breast fed infants, or resembling the AA profile of cord blood and achieving positive nitrogen balance. The direct study of the parenterally nourished patient led to the third generation of CAASs providing all nine essential AAs and a varying composition of non-essential AAs. Studies have been published on the clinical efficacy of altered doses of arginine, BCAAs, cysteine, or taurine supplementation of PN [57,58,59,60,61,62,63,64,65,66].

Today, most currently used AA mixtures contain AA amounts that result from this model. These pediatric AA mixtures provide more essential and less non-essential AAs. Specificities of those mixtures such as Vaminolact or Primene® (Baxter Maurepas, France) are nowadays well established.

An AA that should be included in pediatric AA solutions is taurine. Plasma, platelet, and urinary taurine levels have been shown to be significantly decreased in children on long-term HPN compared to those in normal children of similar age [67]. It has even been suggested that taurine may be conditionally essential in pediatric patients on long-term PN [68]. ESPGHAN/ESPEN/ESPR guidelines [49] recommend that AA solutions for infants and children should contain taurine, a recommendation that is based on data indicating that adequate taurine may prevent cholestasis in newborn infants.

Factors affecting solubilities of calcium and phosphate in neonatal PN solutions containing the new AA formulation were examined [69,70]. Nowadays, those “pediatric AAs mixtures” are widely used for children and adolescents on short as well as for long-term PN.

3. PN with SMOFlipid in Newborn Infants, Infants/Toddlers, Children, and Adolescents – Clinical Experience

Special features of SMOFlipid

SMOFlipid is a composite ILE containing SO (30%), MCTs (30%), OO (25%), and FO (15%) and is supplemented with alpha-tocopherol (200 mg/L) to counteract lipid peroxidation and oxidative stress [22]. A key feature of SMOFlipid is its specific FA pattern resulting from the combination of these four different oils. In comparison to SO, SO/MCT or OO/SO-based ILEs, the pattern is more similar to human umbilical cord blood and breast milk, respectively, with regard to proportions of saturated fatty acids (SFAs), monounsaturated fatty acids (MUFAs), EFAs, and the omega-3 PUFAs EPA and DHA (Figure 1) [71,72,73]. Such a balanced FA supply is of relevance, especially for newborn infants, to maintain plasma and tissue FA patterns equivalent to those seen in breast-fed infants [32].

Several randomized clinical trials have been performed in VLBW infants comparing SO based ILEs and SMOFlipid [51,77,78,79,80,81].

In these studies, growth was not different between treatment groups, but those who received composite ILEs with FO had higher circulating EPA and DHA levels in both plasma and red blood cells (RBCs). Moreover, liver function tests (ALAT, ASAT, GGT, alkaline phosphatase) are less modified in those receiving FO based ILEs. This suggest that the reduced content of n-6 PUFAs and phytosterols in SMOFlipid can contribute to counteract inflammation and liver damages while the supply of adequate amounts of DHA supports visual, neural, and mental development [27,28,31,32,33].

In numerous clinical trials conducted in term newborn infants, infants and toddlers, older children and adolescents, SMOFlipid provided as a component of PN has been evaluated in comparison to other types of ILEs, especially SO based ILEs, in terms of effects on plasma and tissue FA patterns, inflammation and infections, cholestasis and liver function, growth and nutritional status as well as EFA status (see Table 1).

Table 1: Studies investigating PN with SMOFlipid in newborn infants, infants/toddlers, children, and adolescents [9,15,47,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107]

Effects on fatty acid patterns

Several studies evaluated effects of HPN with SMOFlipid on FA patterns in pediatric patients (Table 1). Goulet et al. conducted the first double blind RCT including infants and children on HPN receiving either SMOFlipid or a SO-based ILE [47]. Both ILEs were administered at a mean dose of 1.4 g lipids/kg BW/day. After 4 weeks, EPA and DHA contents in RBC and plasma phospholipids (PL) were significantly increased in the SMOFlipid group compared to controls. Moreover, the ratio of n-3 to n-6 FAs in plasma and RBC-PL was significantly elevated with SMOFlipid compared to SO-ILE [47]. Similar results have been published by Lezo et al. [86] from a study involving 38 pediatric patients with a median age of 5.56 (0.9-21.9) years on long-term PN receiving composite FO containing ILE (SMOFlipid) (n=23) or OO based ILE (Clinoleic) (n=15) at a dose of 1.3 g lipids/kg BW/day. Patients on SMOFlipid showed significantly higher levels of EPA and DHA as well as lower levels of ARA and Mead acid (MA)/ARA ratio in plasma and RBC compared with patients on Clinoleic and with healthy controls. Patients did not differ between groups for growth and liver status.

Potential risk of essential fatty acid deficiency

It has been claimed that pure FO based ILEs (e.g., Omegaven) were safe for achieving a normal EFA status [108]. Yet, ESPGHAN does not recommend the long-term use of Omegaven as the sole source of IV lipid [23]. As a corollary, effects of long-term use of SMOFlipid for children on long-term HPN receiving a composite FO based ILE, on the RBC FA profile and EFA status also remain to be clarified. With that aim, 46 children highly dependent on PN (median PNDI 120% and daily ILE dose of 1.5-2 g/kg BW) for ≥1 year were included in a prospective study when they had received the composite FO ILE for >6 month. Out of this baseline group, only 25 children remained highly PN dependent (SMOF1, n = 25) and could be assessed a second time, 2.4 years later (SMOF2, n = 25). An independent control group (“weaned off PN” group; n = 24) included children who had been weaned off PN for >2 years (median: 4 years). No differences for growth parameters, citrulline, and bilirubin were observed between the SMOF groups (0.2 < P < 0.8). The weaned-off group did not differ from the SMOF groups for growth parameters (0.2 < P < 0.4) but citrulline was higher (P < 0.0001), reflecting intestinal sufficiency. SMOFlipid induced higher percentages of EPA (20:5n–3) (8.4% ±2.9%) and DHA (22:6n–3) (11.7% ±2.2%) in RBC profiles in the SMOF2 group compared to weaned-off children (0.8% ± 0.4% and 6.6% ± 2.3%, respectively) but lower proportions of ARA (20:4n–6). However, the Holman index, a marker of EFA deficiency, did not vary between groups (P = 0.9). Authors concluded that long-term (>2.4 years) use of SMOFlipid was well tolerated in HPN-dependent children. The RBC-FA profile alterations were consistent with the omega-3 PUFA–enriched composition of this emulsion without evidence of EFA deficiency (see Table 1) [15].

Likewise, Lezo et al. found no significant differences in Holman index between pediatric patients of all age groups receiving HPN with SMOFlipid or OO/SO-ILE (Clinoleic) compared to healthy controls; there were no signs of EFA deficiency (Holman index ≥ 0.21) with either ILE [86].

Effects on inflammation and infections

Evidence on the effects of PN with SMOFlipid on parameters of inflammation and infections in hospitalized patients receiving PN is limited and lacking for patients receiving HPN. In a small RCT measuring levels of pro-inflammatory cytokines in 14 newborn infants receiving SMOFlipid (dose range 1-4 g lipids/kg BW/day) for 72 hours post gastrointestinal (GI) surgery, serum levels of IL-6 decreased from baseline to day 3 whereas there was an increase in IL-6 levels in controls receiving SO/MCT-ILE (p=0.007 for change between groups) [90]. The authors concluded that SMOFlipid was beneficial to ameliorate the inflammatory response post GI-surgery [90]. In a larger RCT including 160 newborn infants receiving either SMOFlipid or SO/MCT-ILE for 4-5 days after GI surgery (lipid dose 1-3 g/kg BW/day), no significant differences in pro-inflammatory cytokines or incidence of sepsis between groups were seen at 2 and 4 weeks post-surgery [95].

Effects on cholestasis/liver function

Hospitalized patients

In hospitalized newborn infants, infants/toddlers, children, and/or adolescents receiving PN with SMOFlipid administered at doses of up to 3 g lipids/kg BW/day for periods of 1 to 9 weeks, studies report significantly lower incidences of cholestasis and IFALD as reflected by significantly lower plasma conjugated bilirubin (CBil), total bilirubin (TBil), and direct bilirubin (DBil) levels compared to controls receiving SO-ILE [82,85,93,94,96,99]. Moreover, administration of SMOFlipid resulted in improved liver function as indicated by significantly lower levels of the liver enzymes alanine aminotransferase (ALT), aspartate aminotransferase (AST), and DBil levels compared to SO/MCT-ILE [95,102,105].

HPN patients

In 2010, a double blind RCT investigated the effects of SMOFlipid (lipid dose 1.4 g/kg BW/day) compared to the same dose of SO-ILE, on RBC FA profiles, and liver status in children receiving HPN over a period of 4 weeks. Parameters of liver function (ALAT, ASAT, GGT, alkaline phosphatase) were below critical values and tended to be lower after the 4 weeks study period compared to controls receiving SO-ILE. Total bilirubin decreased significantly with SMOFlipid whereas it increased with SO-ILE over the observation period, the change being significant between the treatment groups [47]. Muhammed et al. [107] published a cohort comparing serum bilirubin during 6 months in children with PN-associated jaundice who changed from SO based ILE Intralipid (n=9) to SMOFlipid (n=8). After 6 months, 5 of 8 children in the SMOFlipid and 2 of 9 children in the Intralipid group had total resolution of jaundice. The median bilirubin fell by 99 μmol/L in the SMOFlipid group but increased by 79 μmol/L in the Intralipid group (P=0.02). Belza et al. conducted a retrospective cohort study in infants with IF and found that SMOFlipid administered at doses between 2-3 g lipid/kg BW/day resulted in a lower incidence and decreased severity of IFALD over the first 12 months of administration compared with a historical cohort who received SO-ILE. Infants receiving SMOFlipid were less likely to reach critical CBil levels, and no infant receiving SMOFlipid required “rescue treatment” with pure FO-ILE for salvage of liver function [94]. In a large cross-sectional study including 578 pediatric patients enrolled in HPN centers in France between 2014 and 2019, it was reported that the use of SMOFlipid for these patients significantly increased within this period from 67.4% to 88.3%. Overall, the prevalence of cholestasis (Conjugated Bilirubin > 20mmole/L) was low and remained stable between 4.1 and 5.9% during the study period [9]. In a recent prospective cross-sectional study including 46 children on HPN who were highly dependent upon PN as indicated by the PNDI (median 120%), these investigators evaluated the clinical safety of SMOFlipid long-term administration (average duration 2.4 years) at doses between 1.5–2.0 g/kg BW/day [15]. Liver function tests (ALAT, ASAT, GGT, alkaline phosphatase) and indicators of cholestasis, especially CBil, were close to the normal range in the group receiving SMOFlipid. These results provide further evidence that PN with SMOFlipid reduces the risk of cholestasis and IFALD in the long-term [15].

Overall, studies investigating effects of HPN with SMOFlipid (doses up to 3 g lipids/kg BW/day) for varying durations ranging from 4 weeks to 6.6 years consistently report benefits in terms of correction or prevention of liver disease following SMOFlipid administration [47,84,88,107].

Effects on growth and nutritional status

Hospitalized patients

In studies evaluating effects of PN with SMOFlipid, administered in doses up to 3 g lipids /kg BW/day for 2 to 9 weeks to hospitalized pediatric patients of all age groups, no significant differences in weight gain and parameters of growth as compared to SO-ILE [82,96,105] or SO/MCT-ILE [95,102] were reported (see Table 1), indicating that SMOFlipid is at least equally effective as SO-based ILEs with regard to supporting adequate growth in these patients.

HPN patients

Several studies evaluated effects of long-term HPN with SMOFlipid on growth parameters (see Table 1). Goulet et al. found that after receiving SMOFlipid for more than 6 months, highly PN-dependent children (median age 5.8 years) on HPN showed normal body weight and height relative to growth charts for healthy children. After an extended PN duration of 2.4 years, growth parameters between patients still receiving PN with SMOFlipid and those weaned from PN did not show any significant differences and were consistent with normal growth rates [15]. Ho et al. retrospectively evaluated growth parameters in children on HPN from initiation of SMOFlipid up to 1.5 years post SMOFlipid initiation and found slight increases in body mass index (BMI) z-scores over this period that were, however, not significant [88]. In newborn infants with IF receiving PN with SMOFlipid (2-3 g lipids/kg BW/day) for an average of 421 days, Belza et al. reported significantly improved weight z-scores after 3 and 6 months compared to historic controls receiving SO-ILE. Anthropometric parameters including weight, height, and head circumference remained within the normal range with both ILEs [94]. Lezo et al. conducted a prospective multicenter study including pediatric patients of all age groups receiving HPN with SMOFlipid or OO/SO-ILE (median lipid dose 1.3 g/kg BW/day) for a mean duration of 22 and 21 months, respectively [86]. Median Z-scores of weight, height or BMI for age were not statistically significant between groups [86]. The authors concluded that both ILEs ensured adequate growth – despite the lower plasma levels of ARA seen in this study.

4. Crystalline Amino Acid Solutions for Newborn Infants, Infants/Toddlers, Children, and Adolescents – Clinical Experience

Special features

Vaminolact is a pediatric CAAS providing essential (49%) and non-essential (51%) AAs in a pattern similar to human milk (Figure 2). The AA composition of Vaminolact has a special focus on AAs that are critical for pediatric patients (Table 2). It supplies adequate amounts of cysteine and tyrosine to compensate for the insufficient synthetic capacity in newborn and young infants. Moreover, it provides higher amounts of arginine to account for reduced intestinal synthesis and lower amounts of methionine, phenylalanine, valine, and isoleucine to account for the lack of intestinal “first-pass” metabolism [54,55,56]. This helps to avoid plasma AAs imbalances in pediatric patients receiving PN.

There is no data available for supporting the use of adult CAASs such as Vamin^® (Fresenius Kabi Bad Homburg, Germany) in infants, toddlers, children and adolescents. As a matter of fact, even designed for neonatology according to the value of CAASs, there is no worry to use them also in infants/toddlers, children, and adolescents. The literature reports valuable data supporting the safety and efficiency of those pediatric AAs mixtures on the short as well as the long-term (Table 1). Moreover, Vaminolact supplies taurine, an AA that may be conditionally essential in pediatric patients requiring long-term PN [68,111]. Taurine deficiency during the neonatal period may result in retinal dysfunction [112] as well as long-term neurodevelopmental impairment [113]. Additionally, available evidence suggests that adequate taurine provision may prevent cholestasis in newborn infants [112,114]. According to ESPGHAN/ESPEN/ESPR guidelines [49], taurine should be part of AA solutions for administration to the pediatric age group.

Clinical experience with Vaminolact

Nutritional efficacy and tolerance of Vaminolact has been proven by extensive clinical experience since its introduction to the market in 1989 and demonstrated by clinical studies showing adequate weight gain, positive or improved nitrogen balance, and normal development in pediatric patients receiving PN with Vaminolact in doses in line with ESPGHAN/ESPEN/ESPR guidelines (Table 3) [49].

Table 3: Studies investigating PN with Vaminolact in newborn infants, infants/toddlers, children [13,14,47,115,116,117,118,119,120,121,122,123]

In 1989, Puntis et al. reported results from the first published study comparing Vaminolact in parenterally fed surgical newborn infants and infants/toddlers to a conventional adult CAAS based on the AA pattern of egg protein (Vamin) [123]. Vaminolact was initiated at 0.5 g AA/kg BW/day and advanced to a maximum of 2.5 g AA/kg/day during the 6 days study period. Plasma concentrations of all AAs except threonine, lysine, histidine, and cysteine/cystine were closer to the reference plasma AA pattern of breast-fed infants than with the adult AA solution. In particular, infants receiving Vaminolact had a reduced risk of developing hyperphenylalaninemia, and hypotyrosinemia that have been associated with neurotoxicity in parenterally fed newborn infants [124]. From this short term study, no significant differences in growth or nitrogen retention were seen between the two groups and no hematological or biochemical measurements changed significantly in either group. Authors concluded that Vaminolact would be likely to lessen the risk of neurotoxicity resulting from AA imbalance [123].

In a subsequent study comparing these two CAASs (Vaminolact vs. Vamin) in critically ill newborn infants, Thornton et al found that taurine levels recovered more rapidly with Vaminolact (n.s.) while phenylalanine levels were significantly lower at day 1 and day 3. Moreover, administration of Vaminolact at a mean dose of 2.3 g AAs/kg BW/day from days 3 for a mean total duration of 12 days also resulted in substantially improved nitrogen balance when compared to the adult CAAS [122].

A clinical study involving hospitalized infants recovering from severe malnutrition receiving Vaminolact as part of total PN has shown a body weight gain after the 28 days study period was 110 ± 5% of optimal weight gain for age. In this study, the mean NPEI from PN was 104.3 ± 8.0 kcal/kg BW/d with a mean ratio of NPEI over REE (PNDI) of 210 ± 20%. Authors concluded that a NPEI twice the REE was optimal to achieve the targeted body weight gain in this population of malnourished infants on TPN [14].

A retrospective study by Abi Nader [13] investigated parameters of growth and nutritional status among infants/toddlers and children on HPN receiving Vaminolact or a similar AA solution adapted to the specific requirements of infants (Primene) at a mean dose of 2.9 g AAs/kg BW/day for a mean duration of 1.9 years [14]. Growth velocity for both body weight gain and size was within the normal range.

Nowadays, one should consider these pediatric CAASs as more adapted than any adult CAAS for short as well as for long-term use from birth to adolescence. They achieve normal body weight and size growth without any reported metabolic disorder.

The role for commercial multi-chamber bags inpediatric PN

PN formulations for pediatric patients can either be individualized, tailored to meet the specific requirements of the patient, or standardized which may be suitable for a broad range of patients [125,126]. Individualized PN formulations are prescribed and prepared, usually on a daily basis, in the hospital pharmacy or in specialized compounding centers [126]. Individualized PN has long been regarded as the standard for PN in term infants, infants/toddlers, children, and adolescents. Nowadays there are concerns about hospital compounded bags regarding the adequacy of nutrient delivery as well as prescribing and compounding errors [127]. Individualizing PN in pediatric patients entails a large number of calculations and numerous patient-specific characteristics must be accounted for. It is, therefore, particularly prone to errors, in particular concentration errors with a reported error rate of 6% in a neonatal/pediatric ICU [128]. It is moreover unlikely that an expert in pediatric nutrition is readily available in each center [129].

With standardized PN formulations, in particular with ready-to-use, commercially manufactured multi-chamber bags (MCBs), this problem may be overcome since an expert nutrition care team can order the most common and suitable bags in advance [129]. Commercial MCBs can reduce infection risk, provide an adequate supply of nutrients, support adequate growth, provide ease of use, decrease prescription errors, and potentially reduce costs compared to individually compounded bags [130,131]. Thus, they can provide important benefits in pediatric patients, particularly in smaller centers where automated compounding facilities are not available. A major advantage of standardized PN bags vs. individualized PN is related to their immediate availability on demand due to easy storage and longer shelf life of approximately 2 years, while individualized formulations cannot be prepared at any time and require highly aseptic conditions, specifically trained staff, and a longer lead time for provision [132,133]. Indeed, roles for commercial MCBs have been proposed during emergency situations that make it impossible to acquire supplies, during power failures interrupting the cooling chain, travel, or shortages of single PN components have been proposed [134,135].

According to current ESPGHAN/ESPEN/ESPR guidelines [125], standard PN solutions should generally be chosen over individualized PN solutions in the majority of pediatric and newborn patients [125]. Yet, although a wide variety of commercial MCBs are available for adults, these are not optimal for infants and children as their metabolic demands and nutritional needs differ significantly, depending on age, weight, underlying disease, nutritional and hydration status, and environmental factors [3,126,136]. Accordingly, standard PN formulations designed specifically to meet the varying nutritional requirements of newborn infants, infants/toddlers, children, and adolescents have been developed and made commercially available [126]. These modern commercial MCBs contain fixed amounts of nutrients conform to current guidelines, thus, enabling an improved compliance with guidelines [125,133]. These MCBs should contain last generation ILEs and AA solutions adapted for pediatric use.

However, there seem to be pediatric patients whose requirements cannot be met by standard regimes, especially among patients requiring PN for longer periods. As described in the current literature, only few children receive standard formulas in the HPN setting. According to a large national survey form France, only 0.3 to 7.2% of HPN patients received standardized formulations between 2014 and 2019, mostly during holidays or during the period of the process of weaning from PN [9]. A European survey of 61 IF teams from 20 countries found that HPN was provided in form of individually age- and weight-specific customized bags in 78% of the IF teams, as commercial MCBs without adaptations in 25% of teams and as commercial MCBs customized by the pharmacy in 31% of teams (multiple answers per team possible) [137]. Indeed, customizing commercial MCBs by individual additions of electrolytes when required may represent a promising strategy to increase the number of HPN patients that could benefit from standardized PN formulations.

5. Summary and Conclusion

With the evolution of practices and extended use of PN in hospitalized as well as HPN pediatric patients, MCBs adapted to this age group are needed. The safety and efficacy of lipid and AA components are well documented from many reported studies, most of which are reviewed in this article. The onset of composite FO-based ILEs (SMOFlipid) calls for the development of MCBs containing these ILEs in combination with CAAS adapted for use in pediatric patients. Overall, the available evidence clearly supports the benefits of SMOFlipid administered at doses in line with ESPGHAN/ESPEN/ESPR guidelines [23] in terms of improving the RBC-FA profile without an increased risk of essential fatty acid deficiency (EFAD), of reducing or reversing IFALD, while ensuring adequate growth, both in hospitalized or HPN pediatric patients (see Table 1). The advantages of MCBs have been underlined especially in hospitalized pediatric patients while open discussion should be promoted for its use for HPN pediatric patients.