Effects of Light Exposure on Total Parenteral Nutrition and its Implications in the Neonatal Population

Until enteral nutrition is established, total parenteral nutrition (TPN) is an important means to meet the nutritional needs of premature neonates by increasing weight gain, protein accretion, positive nitrogen balance, and energy in-take.1–7 TPN is often necessary to feed premature neonates because they have limited nutritional reserves and immature, often erratic, gastrointestinal function.7–9 It is also a common practice to rely on TPN while delaying or slowly introducing enteral feedings in premature neonates in order to decrease the risk of necrotizing enterocolitis.10,11 

A number of nutritional components are added together to form TPN. The macromolecules commonly included in TPN mixtures are water, amino acids, dextrose, and lipid emulsion. Macromolecules are used for energy and as structural components of cells. Micromolecules in TPN include vitamins, minerals and trace elements. Micromolecules are added to TPN because they are required to support metabolic activities, such as enzymatic reactions, fluid balance and regulation of electrophysiological processes.3 Some of these TPN components are susceptible to formation of peroxides, aldehydes, and other degradation products.

This review will provide a brief background of hydroperoxide and free radical reactions, outline reasons for increased neonatal susceptibility to radical damage, and summarize laboratory, animal, and human data on light exposed TPN in order to highlight the potential health issues associated with exposure of neonates to TPN degradation products.

A free radical is a chemical species that has a single, unpaired, electron in the outer orbital of one of its atoms.12 Oxygen is a common biological molecule that readily accepts electrons to form free radicals such as the superoxide anion, O₂⁻, and the peroxide ion, 2O₂⁻.12 The peroxide ion rapidly becomes protonated at physiological pH to form hydrogen peroxide, H₂O₂.12 Superoxide and H₂O₂ can become strong oxidizing agents in the presence of a transition metal, such as iron, by forming the reactive hydroxyl radical, OH⁻.12 The hydroxyl radical can combine with lipids such as polyunsaturated fatty acids to form lipid hydroperoxides and aldehydes, such as hydroxypentenal, hydroxynonenal, nonadienal, decadienal, and malondialdehyde, which are all highly cytotoxic.12–17 This hydroxyl radical is also capable of reacting with and damaging proteins and cell components in the human body such as phospholipids, carbohydrates, metalloproteins, and DNA.12,13,18,19 Peroxidation can also cause a loss in quality and potency of TPN ingredients.20,21 Once initiated, the process of peroxide formation and oxidization will continue in biological systems until it is arrested by antioxidants.

Free radical generation is adequately sequestered by antioxidants in the bodies of normal, healthy infants, children, and adults. Anti-oxidants protect the human body from damage caused by free radicals. Vitamin E, vitamin C, and superoxide dismutase are the most important antioxidants in the human body.12 Vitamin E is the body's most important extracellular lipid-soluble antioxidant while vitamin C is the body's most important extracellular water-soluble antioxidant.12 The superoxide dismutases are the major intracellular antioxidant enzyme family in the body.12 Extracellular superoxide dismutase possesses only one three-thousandth of the activity of intracellular superoxide dismutase.18 The superoxide dismutases are metalloprotein enzymes that catalyze the formation of superoxide to oxygen and H₂O₂.12 Hydrogen peroxide is converted to water by the intracellular enzymes glutathione peroxidase and catalase.12 Methionine sulfoxide reductase is also an intracellular antioxidant. Albumin, lactoferrin, and urate are other extracellular antioxidants.18 

Sick and premature neonates often receive high concentrations of peroxides in the first days of life from endogenous and exogenous sources. Examples of endogenous sources include reperfusion injury from hypoxia, acidosis, oxidative stress from birth trauma, mechanical ventilation, supplemental oxygen delivery, infection, and inflammation.18,22 A significant exogenous source of hydroperoxides is light exposed TPN. However, many neonates do not have a proper antioxidant response to this peroxide challenge because of immature physiology. Neonates have higher concentrations of bilirubin and lower concentrations of vitamin E, ß-carotene, and glutathione compared to adults.23–25 Bilirubin has protective antioxidant properties while deficiencies in the antioxidants vitamin E, ß-carotene, and glutathione are detrimental to antioxidant activity. Superoxide dismutase activity is low in premature neonates compared to term neonates, children and adults.26 This activity worsens with oxidative stress challenge. Apotransferrin and ceruloplasmin have potent ferroxidase and ferric ion sequestration properties, respectively. Their antioxidant effect is approximately 200–500 times more potent than vitamin E.27 However, both of these antioxidants are also found in lower concentrations in term neonates than adults and in premature neonates compared to term neonates.27,28 This may lead to more limited sequestration of iron and copper, resulting in increased catalysis of hydroxyl radicals.13 Increased radical formation and reduced antioxidant scavenging may exacerbate common disease states known to present in premature neonates such as hypoxic-ischemic encephalopathy, periventricular leukomalacia, chronic lung disease, necrotizing enterocolitis, intraventricular hemorrhage, and retinopathy of prematurity.19,22,29 

Phototherapy lights commonly used in neonates with hyperbilirubinemia cause the formation of lipid hydroperoxides to a greater extent than ambient light alone, thereby increasing peroxide exposure to those neonates.30,31 Lipid hydroperoxides may be responsible for increased neonatal pulmonary vascular resistance and vasoconstriction of umbilical veins.30,32–34 The vasoconstriction is thought to be caused by lipid generated hydroperoxide inhibition of the creation of prostaglandin I₂ (PGI₂), a vasodilator, from prostaglandin H₂ and increased prostaglandin F_2α, a vasoconstrictor.34 

The references that support this review article were found by using a search conducted with the MEDLINE database for published in vitro and in vivo research evaluating the effects of light on TPN. The search was conducted in August 2008. No time limit was put on the search. Search terms included: “pediatric,” “neonate,” “premature infant,” “children,” “total parenteral nutrition,” “light,” “degradation,” “TPN,” “photoprotection” and “light protection.” Other references were located in the reference sections of the documents obtained from the MEDLINE database search.

The color of light protection determines its ability to prevent the formation of peroxides in TPN and degradation of TPN components. Research was conducted by a group of investigators that quantified the amount of peroxide generated by five experimental TPN sets by varying the color of IV tubing.35 A clear bag with clear tubing was used as a control. The other four bags were covered with opaque black plastic bags, but were connected to clear, orange, yellow, or black tubing. Lipid-free TPN was also compared to lipid-containing all-in-one total nutrient admixture (TNA). Both ambient light and phototherapy were applied to the systems. The extent of peroxide formation in the presence of light over time was compared between the systems. The investigators found that lipid-free TPN with a clear bag and tubing had a 50% higher peroxide concentration than the light protected systems with colored tubing after 23 hours of light exposure. Peroxide concentrations did not differ between the systems that used orange, yellow, or black tubing after 2 and 23 hours. The peroxide concentration in the TNA control solution with a clear bag and tubing was significantly higher than in the lipid-free TPN with a clear bag and tubing. No difference was found in peroxide formation in TNA systems with yellow and black tubing. However, the orange tubing used with the TNA solution was found to have greater than a 50% rise in peroxide concentrations at 1 and 23 hours compared to yellow or black tubing, but it still generated significantly less peroxide than the control solution. No difference in peroxide formation was found between ambient light and phototherapy among the systems studied. This study shows that color and transparency of light protection is important when choosing a product for use in the pharmacy and on the patient care units. It also showed that yellow and black tubing protected TPN solution the best, but orange tubing still prevented a significant amount of peroxide formation. Yellow and orange tubing have an advantage over black tubing because they are transparent, allowing the clinician to visualize air bubbles and particulate matter in the line.

The effect of oxygen exposure to TPN solution is causative in the generation of peroxides. One study compared peroxides generated from neonatal and adult lipid-free TPN with or without oxygen washout by nitrogen and with and without light exposure.36 The ferrous oxidation of xylenol orange in methanol base (FOX-II) and glutathione peroxidase assays were used to quantify peroxides. Peroxide concentrations remained less than 15 μmol/L in light exposed adult TPN solution, but light exposed neonatal TPN generated from 190 to 300 μmol/L of peroxides. Peroxide concentrations ranged from 60 to 130 μmol/L when neonatal TPN was protected from light. Removal of oxygen from a test solution of neonatal TPN inhibited peroxide formation. However, the effect was lost when the solution was delivered through an IV infusion set. While oxygen is an important cause of increased generation of peroxides when TPN is exposed to light, it would be difficult to consistently remove oxygen exposure from the system from the point of compounding to patient delivery. This study showed that even if all of the air were withdrawn from TPN and lipid bags to limit peroxide formation, the beneficial effect is lost when the solution travels through IV tubing.

Several studies were conducted that analytically quantified vitamin degradation and peroxide formation in TPN solution upon continuous exposure to light. One study was conducted to evaluate the effect of phototherapy and vitamin C on lipid hydroperoxide generation in lipid emulsion.30 Lipid emulsion was prepared in a syringe connected to standard clear tubing. Syringes were glass, plastic, or plastic covered with aluminum foil. Tubing was exposed to either a phototherapy spotlight or ambient room light throughout the study period. After 24 hours of light exposure, all syringe reservoirs had an increase in hydroperoxides by threefold (p<0.0005). Hydroperoxides increased an additional 65-fold from the syringe reservoir concentration when exposed to phototherapy (p<0.0005). Hydroperoxide formation was similar in the syringes that were covered with aluminum foil and the syringes that received vitamin C prior to exposure to ambient light. Vitamin E did not have an effect on hydroperoxide generation in lipid emulsion.

Two publications reported on experiments where a multivitamin preparation (MVP) with vitamin C enhanced peroxidation when added to lipid emulsion. One analyzed how MVP in TPN affected peroxide load in patients.37 The investigators compared TPN and TNA exposed to 6 hours of either daylight or darkness, with and without MVP. General peroxide concentrations were measured by the FOX-II assay. Exposure to daylight for 6 hours had no effect on peroxide content in TPN solutions not containing MVPs. However, the authors did report that MVP added to TPN resulted in an immediate, 3-fold higher peroxide content compared to either TPN or TNA alone, with and without light exposure (p<0.001). The amount of peroxides in TNA solution, with and without light exposure was greater than TPN. The amount of peroxide in TNA was additive, not a compounding effect, compared to TPN. Eighty-eight percent of the peroxides in the TNA admixture were H₂O₂. Amino acid solution was found to have a protective effect on peroxide generation. In this study, the lesser concentrated amino acid solution was found to generate a higher peroxide content than the more concentrated amino acid solution at a constant MVP concentration.

The other study found dextrose solutions with vitamin C + 5′-phosphate flavin mononucleotide (FMN) or FMN + polysorbates generated peroxide concentrations that were greater than 3-fold higher than the individual components of the solution alone after 4 and 24 hours of light exposure.38 FMN is the primary form of riboflavin (vitamin B₂) found in the human body. General peroxide concentrations were measured by the FOX-II assay. The lipid based emulsion with FMN + vitamin C generated 12-fold more peroxides than lipid alone and twice as much peroxide as observed with the dextrose formulations. Peroxide formation of vitamin C and FMN solutions in darkness remained low after 24 hours. Vitamin C and FMN had a positive concentration-dependent relationship on the generation of peroxides in dextrose solution exposed to light. Vitamin C provided a protective effect on peroxide generation in TPN in the absence of FMN. Peroxide formation reached a plateau after 24 hours in systems studied up to 72 hours.

The findings from these two studies are controversial because they used the FOX-II assay for quantification of peroxides from a lipid emulsion. The FOX-II assay uses methanol that is intended to draw lipid hydroperoxides from the aliphatic phase of the experiment, but may still draw H₂O₂ from aqueous components in the system. FOX is a non-specific measurement of all peroxides. The FOX-III, n-propanol based assay may draw less H₂O₂ from the aqueous phase than the methanol based FOX-II assay. Hydrogen peroxide generated and measured from either the FOX-II or FOX-III assays may not be the result of lipid peroxidation, but drawn unintentionally from peroxidation of other, aqueous, components in the experiment.39 

A study conducted by a different group of researchers used the FOX-III assay to evaluate peroxide generation in a lipid emulsion with and without the addition of MVP.31 Lipid emulsion without MVP generated approximately 1.5-fold more peroxides when exposed to phototherapy light versus ambient light and significantly more peroxides under ambient light versus darkness (p<0.001). However, triglyceride hydroperoxide formation was inhibited by the addition of MVP to light-exposed lipid emulsion, although approximately 70% of the riboflavin (p<0.001) and 11% of the vitamin C (p<0.01) was still degraded. Vitamin C loss increased to 80% after exposure to 6 hours of phototherapy light.

The same researchers published another study that further quantified lipid peroxide formation with three variations of the FOX assay compared to a catalase assay.39 The catalase assay is high performance liquid chromatography based and is specific for lipid hydroperoxides because catalase degrades H₂O₂ in solution. The researchers found that the FOX assays reported quite different amounts of peroxide depending on the amount of MVP added to the solution with the FOX-I assay reporting higher peroxide concentrations than the FOX-III assay and the FOX-II assay reporting much higher concentrations compared with the FOX-I assay. The researchers also compared H₂O₂ concentrations in dextrose solution + MVP and lipid emulsion + MVP either exposed to ambient light or darkness. Hydrogen peroxide concentrations increased 40% less in lipid emulsion either exposed to ambient light or darkness compared to the dextrose + MVP solutions from 0 to 150 minutes. From these experiments, the researchers concluded that none of the FOX assays were reliable for accurately determining lipid hydro-peroxide concentrations from the lipid phase in experiments of lipid emulsion containing MVP because of their variable activity when exposed to MVP. They speculated that vitamin C may have interfered with the assays by either reducing ferric ions, generating H₂O₂ via autoxidation, or because riboflavin may have reacted with oxygen in the presence of a reducing agent, producing H₂O₂. The authors also concluded that MVP diminishes H₂O₂ generation if added to lipid emulsion.

One publication reported on the use of mass spectroscopy to measure vitamin C concentrations in TPN after exposure to light for three hours.40 The researchers documented degradation byproducts in three types of solutions: MVP, vitamin C + riboflavin, and vitamin C + H₂O₂ + Fe²⁺. All three solutions experienced approximately a 50% reduction in vitamin C abundance and developed the same five degradation byproducts in the mass spectra fingerprint. The researchers conducted a separate experiment that compared the peroxide content of solutions of H₂O₂, vitamin C + H₂O₂, and vitamin C + H₂O₂ + FeCl₂. It showed that a similar peroxide concentration was found in the H₂O₂ and vitamin C + H₂O₂ solutions over time, but the peroxide concentration increased by greater than 60% within 30 minutes when Fe²⁺ was added to the vitamin C + H₂O₂ solution, showing the importance of iron as a transition metal.

In summary, the research conducted on multivitamins show that vitamin C and riboflavin, in particular, are significantly involved in the formation of peroxides in light exposed TPN solution, especially during phototherapy, and are themselves significantly degraded by light. Different testing methods and research groups have confirmed these findings. Liberated peroxides may cause tissue damage. Degradation of vitamin C and riboflavin may cause vitamin deficiencies in some patients.

The effect of trace elements on lipid emulsion and TNA in light and darkness was reported in one publication.41 Lipid hydroperoxides increased 6- to 9-fold (p<0.001) and pH decreased by 0.77 units (p<0.001) in the lipid emulsion mixed with trace elements kept in darkness compared to the control lipid emulsion. Lipid hydroperoxides also increased significantly from 0.04 to 0.19 mmol/L (p<0.01) and pH dropped by 0.01 units in TNA mixed with trace elements kept under refrigeration in darkness compared to control TNA. Lipid hydroperoxides increased significantly from 0.52 to 1.92 mmol/L (p<0.001) and pH decreased from 0.03 to 0.11 units in TNA + trace elements versus TNA control, respectively, at room temperature and exposed to ambient light.

Investigators administered MVP in parenteral nutrition and lipid emulsion over 3 days in newborn guinea pigs to determine if the mode of delivery of MVP affected nutrient disposition or played a role in oxidant stress.42 The pups received TPN + MVP exposed to light, TPN + MVP protected from light, or lipid emulsion + MVP exposed to light. Urinary concentrations of creatinine, nitrogen, and vitamin C and hepatic concentrations of vitamin A, vitamin E, vitamin C, isoprostanes, and glutathione were sampled. The investigators found that light exposed TPN + MVP was associated with a 40% higher concentration of urinary nitrogen (p<0.005) and a 44% higher concentration of urinary vitamin C (p<0.05) compared to light protected TPN + MVP. The hepatic vitamin C concentration was 72% to 119% higher in the group that received lipid emulsion + MVP compared to the other two groups (p<0.05). Hepatic concentrations of vitamin A, vitamin E, isoprostanes, and glutathione were similar between the groups. The light protected TPN + MVP group and light exposed lipid emulsion + MVP group had similar urinary excretion of nitrogen and vitamin C. Urinary creatinine was similar between the groups. Total peroxides were similar in the light exposed TPN + MVP group and lipid emulsion + MVP group and 46% higher than the light protected TPN + MVP group. The authors believed that vitamin C protected MVP solution from lipid peroxidation.

The hepatobiliary effects of protecting TPN from light were studied in a number of investigations in both the rat and guinea pig model. However, the guinea pig model is believed to be more desirable because guinea pigs are the only animal species other than humans and primates that cannot synthesize vitamin C. Newborn guinea pigs also have immature glutathione synthesis that is similar to newborn humans.43 

In a study in the guinea pig model, H₂O₂ in dextrose + NaCl, tert-butylhydroperoxide (TBH) in dextrose + NaCl, MVP + dextrose + NaCl, or control solution of dextrose + NaCl was infused over 4 days.44 Study animals and solution were exposed to 12 hours of light per day. The concentrations of H₂O₂ and TBH in the study solutions approximated the amounts found in MVP solution. After the infusions were completed, the hepatic glutathione content (p<0.01) and a free-radical sensitive marker, PGI₂/total prostaglandin ratio (p<0.01), were both reduced in the groups administered solutions containing H₂O₂ and TBH. However, the solution that contained MVP did not statistically significantly affect hepatic glutathione content or PGI₂/total prostaglandin ratio compared to the control solution. These findings show that peroxides oxidize membrane-bound prostaglandins and probably consume protective glutathione, but MVI may be hepatoprotective.

Another study of guinea pig pups was designed to assess hepatic damage from light exposed TPN or two types of MVP in control solution.45 Pups were randomized to intravenously receive 4 days of either MVP in control solution (i.e., dextrose + NaCl) exposed to light, MVP in control solution protected from light, MVP in control solution exposed to light without riboflavin, TPN exposed to light, TPN protected from light, or control solution. Pups that received either light exposed MVP in control solution or TPN were found to have more hepatic steatosis on histological examination (p<0.05), larger liver weight (p<0.05), and higher isoprostane F_2α concentrations (p<0.01), a marker of free radical attack, compared to the comparator light protected solutions. Light exposed riboflavin-free MVP solution and control solution had similar low scores of hepatic steatosis (p=NS), suggesting an active role of riboflavin in the formation of hepatic steatosis.

The role of oxidant-antioxidant imbalance on decreased bile flow was determined by feeding guinea pig pups 3 days of either light exposed TNA, TNA + enteral feeding, 1% MVP solution, 3% MVP solution, H₂O₂ solution, or control enteral feeding.43 Pups fed TNA with or without enteral feedings had significantly heavier livers and a 50% decrease in bile flow. Bile flow was significantly lower in pups fed TNA versus 1% or 3% MVP solution even though the 3% MVP solution had the highest amount of peroxides in solution in the study. Bile flow was similar between 0% MVP solution, 0% MVP solution + H₂O₂, and 3% MVP with sodium metabisulfite (to quench the peroxides) indicating that peroxide content does not explain the reduction in bile flow seen with TNA.

Investigation into the effect of potentially toxic byproducts from the interaction of riboflavin-induced H₂O₂ on vitamin C was studied in newborn guinea pigs.21 Researchers administered a control solution of amino acids with or without riboflavin and with or without light exposure to the pups for 3 days. Degradation of the amino acid formulation was significantly increased with time (p<0.01) and the presence of light exposure (p<0.01) and riboflavin (p<0.01). H₂O₂ and photoexcited riboflavin induced the transformation of dehydroascorbate into new, biologically active peroxide and aldehyde molecules. High concentrations of the degradation molecules were infused into the pups, causing an increase in hepatic acetyl-coA carboxylase (ACC) activity (p<0.01). ACC regulates hepatic fatty acid synthesis and mitochondrial oxidation. Increased hepatic ACC activity was not seen in pups that received light protected amino acid solution.

It is not clear from the evidence found in these investigations of guinea pigs if light exposed MVP is hepatoprotective for oxidant damage in control solution, but histological and biochemical evidence shows that hepatic steatosis, increased liver weight, and higher hepatic ACC and isoprostane F_2α concentrations result from infusing TPN that has been exposed to light. These markers of damage are apparent in as little as 3–4 days of receiving light exposed TPN and TNA. Among the MVPs, riboflavin may provide a unique contributory role in the development of hepatic steatosis. The effect of light exposure on TNA in bile flow was not conclusive since light protected TNA was not used in the control arm.

Rat studies of light exposed TPN report similar findings of hepatotoxicity as in guinea pigs. Light exposure to TPN was found to decrease bile flow and increase hepatic necrosis, portal tract inflammation, biliary concentrations of inorganic phosphate, a marker of leaky cellular tight junctions, and biliary concentrations of oxidized glutathione, an antioxidant that is produced primarily by the liver.46,47 

Several studies have been conducted to evaluate the effect of light exposed TPN on indicators of pulmonary damage in newborn guinea pigs. One study measured the oxidant effect of light exposed solutions of MVP + control solution, TPN, TPN + MVP, and dextrose + NaCl + heparin control in newborn guinea pig pups.48 After 4 days of administration of study solution, the peroxide concentrations were similar in the MVP groups compared with a group that received H₂O₂ + control. Pulmonary glutathione decreased by 25% in the MVP + control group (p<0.05) and increased by 41% in the groups treated with TPN solution (p<0.001). Pulmonary glutathione was measured because it is a key antioxidant in the lungs, possessing antiradical and antiperoxide properties. The ratio of pulmonary (PGE₂ + PGF_2α)/total prostaglandins was increased compared to control in the MVP + control (p<0.001) and TPN + MVP groups (p<0.001), but statistically similar to control in the TPN group. The rise in the (PGE₂ + PGF_2α)/total prostaglandins ratio is important because the enzyme that is responsible for 80% of the degradation of these prostaglandins is inhibited by oxidation. A rise in the ratio may indicate oxidative challenge in the lungs.49 

Pulmonary fibrosis and oxidation from light exposed TPN elements were estimated by the measurement of pulmonary procollagen α1 (I) mRNA concentrations, a precursor of pulmonary fibrosis, and pulmonary glutathione concentrations in the lungs of newborn guinea pigs.50 Pups were administered 4 days of light protected or exposed MVP + control, light protected or exposed TPN, H₂O₂ + control, H2O₂ + glutathione + control, and 5% dextrose + 0.45% NaCl control. Light exposed MVP + control, light protected TPN, and light exposed TPN pups had increases of 67%, 93%, and 175%, respectively, in pulmonary procollagen α1 (I) mRNA and 58%, 189%, and 244%, respectively, in pulmonary glutathione compared to the light protected MVP group. These increases in procollagen α1 (I) mRNA and glutathione were significantly higher in both TPN groups compared to the light protected MVP group (p<0.05) and higher in the light exposed TPN group compared to the light protected TPN group (p<0.01). Light exposed MVP and higher potency H₂O₂ each had a similar increase in pro-collagen α1 (I) mRNA (p<0.01) and glutathione (p<0.05) compared to control.

Another study compared H₂O₂, MVP, and light exposure on lung remodeling in newborn guinea pigs.51 TPN or MVP with or without light protection, H₂O₂ with or without glutathione, MVP with or without metabisulfite, vitamin C with or without riboflavin, and control were compared. Light exposed TPN solution had 18% fewer alveolar intercepts (p<0.01), a count of alveoli on a standardized microscopic preparation of histological lung specimens, and a 3-fold higher percentage of apoptotic lung tissue (p<0.01) compared to light protected TPN. Similar findings were recorded for light exposed MVP compared with light protected MVP. The alveolarization index was also significantly lower in animals that received vitamin C plus riboflavin compared to the solutions not containing riboflavin.

The degree of pulmonary fibrosis and alveolar damage was studied in newborn guinea pigs that received 4 days of ambient light exposed TPN + MVP, lipid emulsion + MVP, or control solution in an effort to determine if the lipid emulsion adequately protected peroxide degradation of MVP.29 The researchers found that the TPN + MVP group had higher pulmonary procollagen α1 (I) mRNA concentrations than the lipid emulsion + MVP group (p<0.05), but the lipid emulsion + MVP group also had significantly higher pulmonary procollagen α1 (I) mRNA concentrations compared to the control group. Both groups had a similar lower number of alveolar intercepts compared to controls (p<0.05), indicating loss of lung tissue.

Similar to the negative hepatic evidence found in the guinea pig and rat models, light exposed TPN has been shown to increase apoptotic lung tissue, decrease alveolar intercepts, increase pulmonary procollagen α1 (I) mRNA, a precursor of pulmonary fibrosis, and increase pulmonary glutathione and (PGE₂ + PGF_2α)/total prostaglandin ratio, indicating a pulmonary oxidant challenge. Lipid emulsion + MVP also causes a decrease in alveolar intercepts and an increase in pulmonary procollagen α1 (I) mRNA concentrations. Among the vitamins, light exposed riboflavin may have a unique and negative impact on the integrity of lung tissue. In summary, these investigations of guinea pigs shows that pulmonary damage and a significant response to antioxidant challenge results in animals that receive light exposed TPN, MVP, and riboflavin solutions for even short periods of time.

A number of studies have been conducted that quantify peroxide and aldehyde byproducts, metabolic effects and clinical outcome of preterm infants that received light exposed TPN. In one study, investigators compared urinary peroxide excretion in 30 preterm infants, less than or equal to 32 weeks gestation at birth, who received 10% dextrose, dextrose + amino acids, TPN, or TNA.52 Urinary peroxide concentrations were determined on day 0 with 10% dextrose and again on days 1 and 2 of study solution. The groups of infants that received light exposed TPN and TNA had similar higher urine peroxide concentrations compared to the dextrose + amino acids group and the 10% dextrose baseline control (p<0.01). Photoprotection prevented the rise in urinary peroxide excretion in the TPN group. Patients that received light exposed TPN + MVP had higher urinary peroxide concentrations than light exposed TPN without MVP (14.7 vs. 74.5 micromol/L; p<0.05). The study findings suggested an inadequate peroxide quenching mechanism in premature neonates, an active renal peroxide excretion mechanism, or both. No correlation was found between urine peroxide concentrations and post-natal age, sex, use of oxygen supplementation, or phototherapy.

Another group of researchers compared plasma malondialdehyde (MDA) concentrations of 54 enterally fed newborn infants with concentrations found in light exposed and light protected TNA.14 MDA is a cytotoxic aldehyde oxidation byproduct. The MDA concentration of TNA solution increased from 152 to 335 nmol/L (p<0.001) after being stored for 24 hours in darkness, but equivalent solutions increased from 179 to 13,880 nmol/L (p<0.001) after light exposure over the same time period. The average serum MDA concentration measured in the enterally fed newborns was 173 nmol/L. This implied that TNA causes an infusion of extra MDA into newborns above concentrations normally found in the body. The exposure is much greater with light exposed TNA.

One study measured antioxidant concentrations in 60 term and preterm infants that received enteral feeding, enteral feeding + TPN, and TPN.53 Additional vitamin C (preterm, 25 mg/ kg/day; term, 80 mg/day) was added to the TPN solutions as an antioxidant. Plasma antioxidant concentrations were measured at baseline and on the first and fifth days of study feeding. After the first and fifth study days, plasma MDA, super-oxide dismutase (SOD), vitamin C, and vitamin E concentrations were not significantly different in any of the groups compared to baseline. However, term infants who were given enteral feeding + TPN or TPN alone had higher MDA concentrations before starting TPN compared to the fifth study day, possibly indicating that the additional vitamin C was acting as an antioxidant in the patients. No differences were found between term and preterm infants when evaluating SOD, vitamin E, and vitamin C concentrations, though preterm infants in the study had an older mean age (32 weeks gestational age at birth) than in other studies.

One group of researchers studied the effect of light exposure on TPN and its effect on establishing minimal enteral nutrition in 128 preterm infants.54 Infants were randomized either to light protected TPN or light exposed TPN. Light protection was provided through the use of amber tubing and opaque black plastic coverings for the bag and syringe. Minimal enteral nutrition was initiated by the use of a standardized protocol and increased at the discretion of the clinicians. The researchers found that infants in the light protected TPN group had significantly faster daily advancement of minimal enteral nutrition (p<0.05) and cumulative feeding volumes (p<0.001) compared to the light exposed group during the first week of the study. The study authors hypothesized that TPN light exposure may have produced peroxides that, in turn, caused mesenteric vasoconstriction and reduced feeding tolerance.

The effect of light exposed TPN on the risk of chronic lung disease was studied in 77 premature neonates.55 Neonates were randomized to either light protected or light exposed TPN from birth. Light protection was applied to both the TPN and lipid emulsion. A standardized protocol was used to advance TPN. Post hoc analysis found that light protected TPN was associated with a 30% relative reduction in chronic lung disease or death compared with the light exposed TPN group, but the result was not statistically significant (p<0.2). A sample size of 300 would be required to answer the question. Chronic lung disease and death was significantly more common in males compared to females as seen in other reports.

Variations in metabolic response were studied in 45 premature infants randomized to either light protected or light exposed TPN from birth.56 Light protection was applied to both the TPN and lipid emulsion. TPN advancement was standardized according to a protocol. The effects of light exposure on TPN were assessed once the TPN reached the full rate (approximately 7 days). The average peroxide concentration was 30% higher in light exposed TPN compared to light protected TPN. However, no difference was found between the light exposed and light protected TPN subjects with regard to average urinary nitrogen concentration.

Researchers investigated the impact of light protection of TPN on plasma glucose and triglycerides over the first 9 days of life in 59 preterm infants.57 Infants were randomized to either light protected or light exposed TPN from birth. TPN and lipid dosing was increased up to a pre-defined maximum rate. Study subjects were excluded from data analysis if enteral feeding reached 5 mL/kg/day or more. Subject demographics, macronutrient TPN intakes, and amount of enteral feedings were similar between the two groups. Plasma triglyceride concentrations increased at a 2.5 times higher rate over time in the light exposed group compared to the light protected group, achieving statistical significance on days 8 and 9 of life (p<0.05). Blood glucose concentrations over the first 9 days of life were significantly higher in the light exposed group (p<0.001). However, the number of patients presenting with hyperglycemia and the number of patients requiring insulin for hyperglycemia was statistically similar between the groups.

Studies of light exposed TPN and TNA provide evidence of slower advancement to minimal enteral nutrition and cumulative feeding volumes, increased urinary peroxides, and increased plasma triglycerides and glucose concentrations in preterm neonates compared to preterm neonates that received light protected TPN. Adding additional Vitamin C to TPN may reduce the plasma concentrations of MDA in patients. Research determining the effect of light exposed TPN on clinical outcomes such as chronic lung disease has not yet been completed with an adequate sample size to determine if a clinical benefit of light protecting TPN exists.

Free radicals, peroxides, and aldehydes are commonly found in the bodies of newborn animals and premature infants. A significant exogenous source of these byproducts is administered to the patient via TPN. Laboratory research shows that light protecting TPN with dark bags and with colored plastics limit peroxide and MDA formation and degradation of TPN components. Furthermore, studies in laboratory animals shows that light exposed TPN can cause hepatic damage, steatosis, cholestasis, and pulmonary oxidant challenge, remodeling, apoptosis, and an increased marker for pulmonary fibrosis. Human research has found increased urinary peroxide and plasma aldehyde concentrations, longer time to minimal and cumulative enteral feeding, and increased plasma glucose and triglycerides in preterm infants that received light exposed TPN.

A drawback of achieving light protection of TPN and lipid in clinical practice is the lack of availability of marketed light protection devices such as colored TPN and lipid bags and syringes. Instead, dark opaque and amber plastic bags are available for a low cost and may be used to facilitate protecting TPN and lipid bags and syringes from light after preparation, during delivery to the patient care unit, and during infusion into the patient. Use of these coverings is adequate to provide some light protection, but coverage is not complete and the coverings take more time to apply during the TPN preparation process than if the bags and syringes were available in light protection colors from vendors. Furthermore, dark plastic and foil that covers TPN and lipids may cause a safety concern because they require nurses and other practitioners to lift the covering to view the pharmacy label that is affixed to the container. Pharmacy labels could be directly applied to the external surface of the TPN and lipid bags and syringes if the bags and syringes were manufactured with light protection in the plastic. However, as opposed to TPN and lipid bags and syringes, amber-orange colored IV tubing is available for purchase for light protection while still allowing the nurse to observe the IV line for air bubbles, precipitates, or leaks. The colored tubing offers a significant advantage over older methods of tubing light protection such as tubular amber plastic, aluminum foil, or fabric coverings because these former methods required extensive manual manipulation and they obscured the patency and integrity of the IV fluid path.

The absence of larger human clinical outcomes studies precludes light protection of TPN components from becoming a global recommendation at this time. However, it would seem prudent to use this non-invasive approach, given the available data, to light protect TPN components from the point of arrival in the pharmacy throughout the duration of TPN administration on the patient care unit until more conclusive human studies are forthcoming.

Portions of data that appear in this manuscript were presented by Dr. Michaelson at the Annual Meeting of the American College of Clinical Pharmacy in Denver, Colorado on October 16, 2007.

ACC

acetyl-coA carboxylaset

FMN

5′-phosphate flavin mononucleotide

MVP

multivitamin preparation

TBH

tertbutylhydroperoxide

TNA

all-in-one total nutrient admixture

SOD

superoxide dismutase

TPN

total parenteral nutrition