The current study validates the assay used to measure cholesterol, phytosterols, and squalene in a variety of commercially available parenteral lipid emulsions. The assay produced good separation of the various phytosterols measured, had sensitivity for detection of the various phytosterols of 0.2–0.9 μg/mL, and a recovery of 93%–103%. We measured the levels of the various sterols and squalene in eight of the most commonly used lipid emulsions that are based upon soybean oil, coconut/soybean oils, and olive/soybean oils. Concentrations of cholesterol, phytosterols, and squalene varied greatly between the different lipid emulsions. Cholesterol varied from 63.57 to 280.43 μg/mL (441% variation); total phytosterols varied from 384.08 to 839.31 μg/mL (219% variation); and squalene varied from 4.52 to 387.45 μg/mL (8572% variation). β-Sitosterol levels were the highest amongst the phytosterols, followed by stigmasterol and campesterol. The large variations of concentrations of these compounds, all of which possess biologic activity, may have clinical significance upon efficacy and safety of the various lipid emulsions. Future studies should be performed to determine the exact effects of the various sterols and squalene upon biological functions.
Cholesterol is an important component of cell membranes and lipoproteins. High levels of cholesterol found in low density and very low density lipoproteins are associated with an increased risk of cardiovascular complications. A variety of drugs are available that reduce levels of cholesterol (i.e., statins) and decrease the risk of cardiovascular disease. On the other hand, very low levels of cholesterol occurring in underfed individuals are also associated with increased morbidity and mortality. Thus, the optimal levels of dietary intake of cholesterol for individuals with different diseases remain unclear.
Levels of cholesterol intake in most patients consuming weight maintaining diets is more than adequate for tissue functions, since cholesterol is found in most foods of animal origin. Median cholesterol intakes in western diets range from 250 to 325 mg/day in men and 180 to 205 mg/day in women [41
]. Levels of cholesterol intake with the different lipid emulsions would range from 33 to 140 mg per day in individuals consuming 100 g of lipid (approximately 1000 Kcal). However, many parenteral nutrition-dependent patients receive substantially lower amounts of lipid each day (i.e.
, 20–60 mg/day), and these patients would receive significantly less cholesterol. Cholesterol is not required in the diet of humans, since tissues can synthesize cholesterol. Thus, there is no recommended “Adequate Intake” or “Recommended Dietary Allowance” for cholesterol [41
]. However, the ability of the tissues of patients with severe organ injuries, infection, systemic inflammation and/or starvation to synthesize cholesterol is unclear, and the exact biological consequences of low intakes of cholesterol in various patient groups require further investigation.
Hepatobiliary complications are a major cause of morbidity and mortality in patients receiving long-term parenteral nutrition [2
]. The pathogenesis of PNALD is likely multifactorial. Proposed mechanisms include alterations in cholesterol metabolism, bile composition and/or bile transport, lack of enteral gastrointestinal stimulation due to short bowel (with loss of trophic factors and other gut derived hormones; loss of the enterohepatic circulation of bile acids), endotoxins or exotoxins released into the circulation due to loss of gut barrier functions, recurrent systemic infection, and direct hepatotoxicity from substances such as phytosterols in the lipid emulsions [1
The mechanisms by which phytosterols may cause liver disease are unclear. However, phytosterols have been reported to inhibit enzymes involved in cholesterol and bile acid synthesis and metabolism [11
]. Inhibition of bile acid synthesis would limit excretion of phytosterols and bilirubin, leading to their accumulation in the liver [46
]. Cholestasis and liver injury could be the result. Infants may be particularly susceptible to liver injury by phytosterols due to reduced bile acid synthesis and immature bile secretory systems [4
]. Sterols are important constituents of cell membranes and modulate membrane fluidity, cell membrane proteins such as transporters, and endocytosis and exocytosis. Replacement of membrane cholesterol with phytosterols may alter membrane functions [4
] that interfere with cell function. For example, after intravenous injection, β-sitosterol and campesterol accumulate in the liver and may replace 20%–60% of hepatic microsomal cholesterol [4
]. Phytosterols may also interfere with cholesterol metabolism [4
]. For example, β-sitosterol and cholestanol can inhibit cholesterol-7α-hydroxylase, the rate limiting enzyme required for the conversion of cholesterol to bile acids [4
]. Farnesoid X receptor (FXR) is a nuclear receptor that acts as a bile acid sensor that maintains safe intrahepatic bile acid levels [47
]. FXR reduces bile acid import through the hepatic sinusoids, decreases bile acid synthesis, and increases bile acid efflux across canalicular and sinusoidal membranes. Mice lacking the receptor are ultrasensitive to bile acid-induced hepatic injury while FXR agonists protect from cholestasis [47
]. Stigmasterol, a phytosterol present in plant derived lipid emulsions, is a potent antagonist of FXR [47
]. Stigmasterol has also been reported to antagonize the nuclear receptor PXR (pregnane X receptor), which plays a role in detoxifying the liver from excess bile acids [47
]. These effects of phytosterols may result in hepatic dysfunction, but they may also interfere with the function of other cells such as erythrocytes and leukocytes.
The presence of elevated levels of phytosterols in the circulation during administration of parenteral lipid emulsions has been associated with the development of liver dysfunction in patients treated with parenteral nutrition [6
]. Studies by Clayton et al.
], in children on long-term parenteral lipid infusion, reported high levels of plasma phytosterols that were associated with cholestatic liver disease. Clayton et al.
] also associated increased phytosterols levels with the development of thrombocytopenia and reported improvement in liver disease and platelet counts when intake of the lipid emulsions was reduced. Ellegard et al.
] reported a significant elevation in plasma phytosterol levels in adults with short bowel receiving parenteral nutrition with lipid emulsion compared to healthy controls and short bowel patients not requiring parenteral nutrition. However, these authors did not relate the levels to liver dysfunction. More recently, Llop et al
] also reported the development of liver dysfunction in adult home parenteral nutrition patients infused with parenteral lipid emulsion. Bilirubin and aspartate aminotranferase levels in the blood of the patients significantly correlated with blood phytosterols concentrations, and blood levels in the patients receiving parenteral nutrition were significantly higher than healthy controls on normal diets (55.4 vs.
14.8 μg/mL). Hallikainen et al.
] reported on a patient with parenteral nutrition-associated liver disease. When the patient was switched from a soybean oil lipid emulsion to an olive/soybean oil lipid emulsion, the concentrations of phytosterols were lowered while, simultaneously, the liver enzymes returned to lower values. Lowering the intake of soybean based lipid emulsions that contain high levels of phytosterols has also been shown to reduce the incidence of liver disease [1
]. Others have reported resolution of liver disease when patients receiving soybean oil lipid emulsions were switched to a lipid emulsion based upon fish oil (fish contains very low levels of phytosterols) [49
]. These studies indicate an association between high intakes of parenteral phytosterols and the development of liver dysfunction. Our study indicates that β-sitosterol, stigmasterol, and campesterol are the three phytosterols found at the highest concentrations in the various plant based commercial lipid emulsions. Future studies should evaluate these three phytosterols for their effects upon liver integrity. Clearly, there is a need to perform animal studies that directly link the various phytosterols to liver toxicity. Most data in the literature represent associations of liver disease with lipid emulsion administration, but the exact etiology remains elusive. A linking of phytosterols with liver disease would provide impetus for lipid emulsion manufacturers to remove or reduce levels of the phytosterols in their lipid emulsions.
Western diets provide approximately 150–350 mg/day of phytosterols, while vegetarian diets provide up to 500 mg/day [22
]. Approximately 5% of plant sterols are absorbed from the diet by the small intestine [22
], representing approximately 5–17 mg/day of absorbed phytosterol. The absorption of phytosterols contrasts with that of cholesterol, where approximately 55% is absorbed. Most of the phytosterols are transported into the portal circulation and metabolized by the liver [48
]. Phytosterols are normally excreted in the bile. During parenteral administration, phytosterols enter the systemic circulation through the vena cava, bypassing first pass hepatic elimination. Thus, phytosterols given parenterally result in higher levels in non-hepatic tissues than when given via the gastrointestinal tract. Total phytosterol intake from the lipid emulsions ranged from 137 mg/day (ClinOleic®
; olive/soy lipid emulsion) to 311 mg/day (Lipofundin®
N; soybean lipid emulsion) in patients receiving 100 g of lipid per day (). These intakes are substantially higher than the typical daily amounts that enter the circulation from the diet, and differ greatly depending upon the plant source used to manufacture the lipid emulsion. The intakes remain high even with more moderate intake of lipid (i.e.
, 50 g/day). Since the duration and dose of lipid emulsion administration are associated with the development of liver dysfunction [1
], the choice of lipid emulsion (i.e.
, using an emulsion with lower levels of phytosterols) may have clinical significance in limiting the development of liver injury. Decreasing intake of phytosterols by lowering the dose of lipid emulsion administered has been shown to limit and reverse the degree of hepatic dysfunction [1
]. In addition, the clinical impression of greater incidences of liver dysfunction with soybean versus MCT/LCT and olive/soy lipid emulsions may be reflective of the levels of phytosterols in these emulsions. These observations support but do not prove that phytosterols in the lipid emulsions are the cause of the liver disease. Future studies should compare the incidence of liver dysfunction using lipid emulsions with high versus low levels of phytosterols.
Our results differ slightly from the results of Forchielli et al.
], who measured the levels of cholesterol, β-sitosterol, campesterol, and stigmasterol in seven commercial lipid emulsions. Our results confirm that β-sitosterol, campesterol, and stigmasterol are the most common sterols in commercial lipid emulsions. However, we also detected significant quantities of β-sitostanol and lanosterol in the lipid emulsions. Our results also indicate that the lipid emulsions lack detectable quantities of desmosterol, brassicasterol, lathosterol, and ergosterol. We detected squalene in all of the lipid emulsions; levels were significantly higher in the olive oil emulsion compared to the soy and soy/coconut emulsions. In general, we detected slightly higher levels of phytosterols in the emulsions (approximately 13%–40%) than those reported by Forchielli et al.
]. Some of this variability relates to levels of the two additional phytosterol compounds that we measured. Other variability may result from use of different batches and assays. Overall, the relative differences in sterol concentrations between emulsions are similar in our study and the study of Forchielli et al.
Squalene is an open-chain, 30-carbon isoprenoid that is produced in both plants and animals [50
]. Squalene can be cyclized in different ways and can serve as a precursor for the synthesis of 20 different compounds, including cholesterol, steroid hormones, and vitamin D in the human body, and dammarane, hopane, oleanane and friedelane in plants and microorganisms [51
]. Interestingly, studies in rabbits and humans found that high intakes of squalene are not associated with an increased risk of atherosclerosis, despite the fact that dietary intake of squalene can augment cholesterol and bile acid production [52
]. On the contrary, studies have shown that long-term increases in squalene from rat and human diets fails to elevate serum cholesterol [53
]. Furthermore, squalene inhibits activity of β-hydroxy-β-methylglutaryl-CoA reductase (HMGCoA reducatse), a regulatory enzyme in cholesterol biosysnthesis. Inhibition of this enzyme is felt to have antiatherosclerotic activity [55
]. Squalene has been reported to have other functional activities, that include acting as a scavenger for reactive oxygen species [50
], anti-inflammatory properties [50
], anticancer effects (particularly for skin cancers) [52
], and coadjuvant effects when administered with vaccines [51
]. Further studies are needed to evaluate potential benefits of squalene in patients receiving parenteral nutrition.
In this report, we evaluated levels of phytosterols in lipid emulsions based upon plant oils. However, some recent lipid emulsions also include fish oils. Fish oils do not contain significant levels of phytosterols, since they originate from animal sources where the primary sterol is cholesterol. In preliminary analyses, we measured the level of phytosterols and cholesterol in a fish oil emulsion (Omegaven®, Fresenius Kabi, Germany). Total sterol levels were comparable between Omegaven® and the olive oil-based lipid emulsion, ClinOleic® (372.9 vs. 384.1 mg/L, respectively). However, the distribution of sterols was significantly different between the two emulsions. Omegaven®’s content included almost exclusively cholesterol, while ClinOleic® contained a mixture of cholesterol and phytosterols. Omegaven® was also low in squalene (26.7 mg/L). In addition, we analyzed a lipid emulsion based upon a mixture of soybean oil, medium-chain triglycerides, olive oil, and fish oil (SMOFlipid®, Fresenius Kabi, Germany). SMOFlipid® contained a total sterol content of 657 mg/L, composed of a mixture of cholesterol (450 mg/L) and phytosterols (207 mg/L). The squalene content of SMOFlipid® was 52 mg/L.