WO1998046090A1

WO1998046090A1 - Anti-oxydant multivitamin preparation

Info

Publication number: WO1998046090A1
Application number: PCT/CA1998/000342
Authority: WO
Inventors: Philippe Chessex; Jean-Claude Lavoie
Original assignee: Centre Hospitalier Universitaire Saint Justine
Current assignee: Centre Hospitalier Universitaire Saint Justine
Priority date: 1997-04-15
Filing date: 1998-04-08
Publication date: 1998-10-22
Anticipated expiration: 1999-10-15
Also published as: AU7019098A

Abstract

The present invention relates to a liquid multivitamin preparation with reduced amounts of peroxide. The present invention also relates to a liquid multivitamin preparation with reduced amounts of peroxide wherein the multivitamin preparation comprises iron bound to a polyose molecule or a combination of riboflavin and vitamin C, wherein both compounds are not in contact with each other. The present invention also concerns a method for administering liquid multivitamin with reduced amounts of peroxide. The present invention also concerns a dispensing means for administering liquid multivitamin with reduced amounts of peroxide and in order to protect the preparation from light.

Description

ANTI-OXYDANT MULTIVITAMIN PREPARATION

BACKGROUND OF THE INVENTION

(a) Field of the Invention The present invention relates to multivitamin preparation for enteral or parenteral administration. The present invention also relates to a method of administering multivitamin preparations.

(b) Description of the prior art There has been a lot of research in the area of the oxidation of lipids found in foodstuffs mainly because of the negative economic impact caused by the unpleasant tastes and odors generated by the peroxidation of lipids. At the cellular level, the biological effects of peroxides are well documented. Their effect in the organism as a whole, where there is interactions between all anti-oxidant systems is not well documented. For newborn infants with an immature antioxidant capacity, the clinical effects of infused per- oxides is a cause of alarm. Parenteral nutrition is often required in the early stages of feeding immature very low birth weight infants. When compared to intravenous glucose only, high levels of hydroperoxides have been detected in nutrient solutions and in complete preparations of intravenous nutrition. Hydroperoxides are reactive and can form free radicals in turn these oxygen species may disrupt cell membrane integrity and mediate tissue injury. They can also interfere with the production of vasoactive eicosanoids and stimulate neovascularization which is responsible for the retino- pathy of prematurity. Because antioxidant systems of premature infants are immature, the potential risk of the peroxides to cause oxidant stress is high. There is a correlation between markers of free oxygen radi- cal-induced lipid peroxidation and outcome in immature newborn infants. The role of those peroxides is questioned in the pathogenesis of bronchopulmonary dyspla- sia, necrotizing enterocolitis, retinopathy of premature infants. Under the conditions met during intra-venous nutrition of newborn infants, the multivitamin solution is the major source of hydroperoxides. It is well known that the generation of hydroperoxides is stimulated by phototherapy or even by day-light. There is therefore a need to decrease the oxidant load received by immature infants. Riboflavin and polysorbate are light sensitive molecules which could be implicated in the generation of peroxides. Most parenteral multivitamin solutions contain 5 '-phosphate flavin mononucleo- tide (FMN) as a source of riboflavin, a vitamin implicated in a large number of metabolic pathways, including respiratory chain reaction. Polysorbates (PS) are detergents added in certain multivitamin solutions to dissolve liposoluble and hydrosoluble vitamins in the same medium.

Oxidation of perfused lipids

The abundance and the chemical structure of fatty acids present in cell membranes renders them a preferred target for radical attack. A free radical (RL) is an atom of which the last electronic layer is occupied by a free electron. This particularity ren^¬ ders the free radical very reactive. In order to lower its energy level, the free radical will try to gain a further electron or loose the free electron. Double bonds of polyinsaturated fatty acids are a source of electrons for free radical. Free radicals will react in a non-specific way with different molecules. The free radicals will react by three different mechanisms. They can give their electron and become ionic, they can extract an electron from an other molecule thereby gen- erating an other free radical or they can bind with the target molecule, this new molecule will then become a radical. The damages caused by the generation of free radicals are varied depending where free radicals are released. Damages include inactivation of proteins, modification of DNA and activation of other molecules. The list of pathologies related to oxidation is exhaustive and includes certain types of cancer, aging, rheumatoid arthritis, fibrosis, ischemia repression syn- drome and artherosclerosis .

Peroxidation of lipids is a three-step chain reaction. The three steps are initiation, propagation, and termination. Initiation occurs when a free radical binds to a fatty acid(L) or extracts an electron from that fatty acid. The resulting molecule is then a radical (L») which can react with molecular oxygen to form a hydroperoxide (LOO») . That radical will then attract an electron from an other fatty acid to yield a hydroxyperoxide (LOOH) and the reaction will continue with the radical generated in the process. The chain will stop when there is no more substrate or when an anti-oxidant molecule such as alpha-tocopherol provides the missing electron. Depending on the type of fatty acid and the site of the radical attack, there is for- mation of different molecules. Since the dissociation energy of a C-H bond of a bi-allyl structure is weaker then that of a mono-allyl structure (75 Vs 88 Kcal/mol) , polyunsaturated fatty acids will more likely loose their electron to become a radical with which oxygen will bind to form a hydroperoxide. During the addition, a conjugated diene is produced which can be detected by spectroscopy at 230-235 nm. The LO-OH bond of the hydroxyperoxide has a dissociation energy of about 44 Kcal/mol. An electron donor such as Fe²⁺ or Cu⁺ will be sufficient to generate an alkoxyl radical. That alkoxyl radical can attract an electron from a fatty acid and maintain the lipid peroxidation. However, a break in a C-C bond will generate an aldehyde and an alcane. If a molecule of 02 is added to the double bond near the methyl moiety of the fatty acid, pentane is formed from n-6 fatty acids, and ethane is derived from n-3 fatty acids. These gases are measured in breath as markers of peroxidation of endogenous and/or exogenous lipids. The most commonly measured aldehydes are hexanal, 4-hydroxy-2-nonenal (HNE) and malondialdehyde (MDA) .

Quantification of hydroperoxides, alcanes, aldehydes, conjugated dienes, as well as molecules produced by the action of free radicals on proteins or DNA are used as markers of oxidative stress. However, the presence or absence of such markers are not absolute proof that oxidation did or did not take place. These chemical compounds take part in the general metabolism, therefore their generation is not solely due to lipid peroxidation. For instance, MDA, hexanal and HNE can all be catabolized by either aldehyde dehydrogenase or glutathione-S-transferase. Sources of MDA are diverse, for instance thromboxane synthetase activity will produce one molecule of MDA for each thromboxane A2. Alcanes produced by fermentation in the gastrointestinal tract can contaminate alcanes measured in breath.

The presence of markers of lipid peroxidation suggests that the antioxidant capacity of the cell was at one point overwhelmed. Furthermore, cell damage can appear before or at the same time as the markers of peroxidation. Certain markers of lipid peroxidation are toxic. Aldehydes can form carbonyl moieties on macromolecules, or react with the lysine moieties of proteins. Such reactions can inactivate glucose-6- phosphate dehydrogenase or activate collagen formation. Furthermore, peroxides are potentially highly oxidant since providing them with one electron will produce a reactive oxygen radical. Hence peroxides can influence, alone or as free radicals, the formation of vasoactive molecules such as prostaglandins or thromboxanes .

Because of their high polyunsaturated fatty acid concentration, lipid emulsions are a target for peroxidation. Hence, emulsions containing a mix of medium and long chain fatty acids can be advantageous as they decrease the amount of unsaturation. Other nutrients also generate peroxides such as certain amino acids, vitamins, trace metals. Because of the immaturity of the antioxidant system of premature infants, infused peroxides are potentially cytotoxic and enhance mortality and morbidity. Lavoie et al . ( Pediatrics, 1997, 99 (3): 1-5) discuss the role of a multivitamin preparation on the actual peroxide load received by patients. This publication describes that even when protected from light, the multivitamin preparation infused produced a 10 fold increase in peroxide

In adult patients, parenteral concentrations of multivitamins are lower than those required for newborn infants, and the rate of infusion is faster. There- fore, one can speculate that the infused peroxide load is lower in adult patients. However, in compromised subjects the role of infused peroxides on the patho- genesis of morbid complications needs to be further investigated. Indeed, in patients with sepsis and adult respiratory distress syndrome, urinary peroxides concentrations were consistently higher in patients who did not survived.

Multivitamin solutions contain several antioxi- dants such as ascorbate (vitamin C) , tocopherol (vitamin E) and vitamin A. They have free radical scavenging properties but no antiperoxide activity. To date, no known source of enteral or parenteral multivitamin solutions protects against the generation of peroxides .

SUMMARY OF THE INVENTION

It is an aim of the present invention to provide a liquid multivitamin preparation with reduced amounts of peroxide. It is an aim of the present invention to provide a liquid multivitamin preparation for enteral or parenteral administration with reduced amounts of peroxide .

It is an aim of the present invention to provide a liquid multivitamin preparation comprising riboflavin and vitamin C for enteral or parenteral administration with reduced amounts of peroxide wherein the riboflavin is not in contact with the vitamin C.

It is an aim of the present invention to provide a liquid multivitamin preparation with reduced amounts of peroxide wherein the multivitamin preparation is solubilized in the presence of iron bound to a polyose molecule .

It is an aim of the present invention to provide a liquid multivitamin preparation for enteral or parenteral administration with reduced amounts of peroxide wherein the multivitamin preparation is solubilized in the presence of iron bound to a polyose molecule.

It is also an aim of the present invention to provide a liquid multivitamin preparation comprising riboflavin and vitamin C for enteral or parenteral administration comprising reduced amounts of peroxide wherein the multivitamin preparation comprises two phases until the time of enteral or parenteral admini- stration, a riboflavin phase separated from a vitamin C phase up to a molar ratio about 1/10,000 between concentrations of riboflavin and vitamin C.

It is also an aim of the present invention to provide a liquid multivitamin preparations for enteral or parenteral administration comprising reduced amounts of peroxide wherein the multivitamin preparation does not comprise riboflavin, which is administrated separately.

It is also an aim of the present invention to provide a method of administering multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the multivitamin preparation from light. It is also an aim of the present invention to provide a substantially opaque dispensing means for administering multivitamin preparation in order to protect the multivitamin preparation from light.

In accordance with the present invention there is provided a liquid multivitamin preparation comprising riboflavin and vitamin C for enteral or parenteral administration and a reduced amount of peroxide, wherein the riboflavin is not in contact with the vitamin C. In accordance with the present invention there is also provided a liquid multivitamin preparation for parenteral or enteral administration comprising reduced amounts of peroxide, wherein the multivitamin preparation is solubilized in the presence of iron bound to a polyose molecule.

The iron bound polyose molecule may be selected from the group consisting of Iron Saccharide, Iron Dex- tran and Iron Sorbitol. The preferred iron bound polyose molecule is Iron Dextran. In accordance with the present invention there is also provided a liquid multivitamin preparation comprising riboflavin and vitamin C for enteral or parenteral administration and a reduced amount of perox- ide, wherein the multivitamin preparation is in two phases until the time of enteral or parenteral administration, a riboflavin phase separated from a vitamin C phase up to a molar ratio of about 1/10,000 between concentrations of riboflavin and vitamin C. In accordance with the present invention there is also provided a method of administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide, wherein the method comprises the step of protecting the preparation from light.

The light may have a wavelength of about 200nm to about 800nm, of about 300nm to about 600nm, or more particularly of about 325nm to about 550nm.

The step of protecting the preparation from light may be effected by means of a substantially opaque dispensing means for administering the multivitamin preparation, or by means of a colored dispensing means for administering the multivitamin preparation. The dispensing means may be colored by at least one color or a combination thereof selected from the group consisting of black, gray, brown, orange, yellow, green, blue, and red, more preferably by a color or a combination thereof selected from the group consisting of black, brown, orange and yellow. In accordance with the present invention there is also provided a colored container or a substantially opaque container for administering a liquid multivitamin preparation in order to protect the multivitamin preparation from light. BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 demonstrates the influence of lipid -emulsion and day light on peroxide concentration of parenteral nutrient solution freshly prepared and compris- ing multivitamins;

Fig. 2 demonstrates the effect of two different sources of iron bound to a polyose molecule namely Iron Dextran (FeD) and Iron sorbitol (FeS) on the generation of peroxides in parenteral nutrient solution; Fig. 3 demonstrates the effect of polysorbate detergents 80(1%) and 20(0.02%) on the generation of peroxides in a control solution (glucose 5%g + NaCl 0.45g% + heparin lU/ml) and in a multivitamin solution (NIT ) ; Fig. 4 demonstrates the role of riboflavin and ascorbic acid (vitamin C) with the generation of peroxide in the multivitamin solution and in the control solution (glucose 5%g + ΝaCl 0.45g% + heparin lU/ml) ;

Fig. 5 demonstrates the level of peroxides in the TPΝ solution (all in one) exposed 3 hours to daylight (control) in presence of increasing concentrations of 4 sources of iron;

Fig. 6 demonstrates the effect of increasing concentrations of 4 sources of iron on 500μM H2O2 (A) , TBH: tert-butyl hydroperoxide (B) and cumene hydroperoxide (cumene) (C) ;

Fig. 7 demonstrates the free oxygen radicals produced by the Fenton reaction are detected by oxidation of 3 μM scopoletin; Fig. 8 demonstrates the effect of iron on the generation of peroxides in the TPΝ solution (all in one) exposed 0 to 3 hours to day-light;

Fig. 9 demonstrates the effect of increasing concentrations of 3 sources of iron on the availability of the thiol function of cysteine; Fig. 10 represents the experiment design of Example 3;

Fig. 11 represents the peroxide concentrations in fat-free TPN delivered with the 5 different intrave- nous settings, exposed to ambient light;

Fig. 12 represents the thiol functions in fat- free TPN delivered with the 5 different intravenous settings, exposed to ambient light;

Fig. 13 represents the peroxide concentrations in fat-free TPN delivered with 5 intravenous settings exposed to phototherapy;

Fig. 14 represents peroxide concentrations in a all-in-one TPN solution (with lipid emulsion) delivered with the 5 different intravenous settings, exposed to ambient light; and

Fig. 15 represents the spectra of absorbance of tested amber tubing (yellow, orange) and of 5' -phosphate flavin mononucleotide .

DETAILED DESCRIPTION OF THE INVENTION

Specifically, the present invention concerns multivitamin solutions with a reduced peroxide content.

More specifically, the present invention concerns parenteral and enteral multivitamin solutions with a reduced peroxide content. The present invention also concerns parenteral and enteral multivitamin solutions with a reduced peroxide content.

The generation of peroxides in multivitamin solutions is caused by the interaction between ribofla- vin and ascorbate in presence of light. Because of the immaturity of the antioxidant capacity of premature infants, . infused peroxides are potentially cytotoxic and might increase morbidity and mortality

By the term "polyose molecule" is meant a poly- saccharide or an analogue or derivative thereof. The term "polysaccharide" include naturally occurring poly- saccharides and synthetic polysaccharides . Polysaccha- rides can be easily derivatized at functional groups by methods well known in the art. For example, modifica- tions include but are not limited to adding an amino acid, a peptide or a polypeptide to the polysaccharide moiety.

A suitable polyose molecule is capable of binding to an iron atom to form a complex. The polyose molecule has to be selected such that when it is bound with the iron atom the complex formed does not promote formation of free oxygen radical when it is added to a multivitamin preparation.

The term "multivitamin preparation", is meant to be interpreted broadly and include suitable preparation and solution used in intravenous nutrition and oral vitamin supplement.

By the term "reduced amounts of peroxide" is meant a reduction of the amounts of peroxide in a test preparation when compared to a control preparation.

By the term "peroxides" is meant harmful peroxides generated in multivitamin preparation and nutrient solution.

Preferably the iron bound polyose molecule is selected from the group consisting of Iron Saccharide, Iron Dextran and Iron Sorbitol.

More preferably the iron bound polyose molecule is Iron Dextran or Iron sorbitol.

Most preferably the iron bound polyose molecule is Iron Dextran.

In a preferred embodiment, there is provided a method of administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light having a wavelength of about 200nm to about 800nm.

In a preferred embodiment, a method is provided for administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light having a wavelength of about 300nm to about 740nm.

In a preferred embodiment, a method is provided for administering multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light having a wavelength of about 300nm to about 600nm. In a most preferred embodiment, a method is provided for administering multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light having a wave- length of about 300nm to about 550nm.

In a most preferred embodiment, a method is provided for administering multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light having a wavelength of about 325nm to about 550nm.

In a preferred embodiment, a method is provided for administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light having a wavelength of about 500nm to about βOOnm.

In a preferred embodiment, a method is provided for administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light by means of a substantially opaque dispensing means for administering the multivitamin preparation. In a preferred embodiment, a method is provided for administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide wherein the method comprises the step of protecting the preparation from light by means of a colored dispensing means for administering the multivitamin preparation.

In a further preferred embodiment the dispensing means is colored by one or more color or a combination thereof selected from the group consisting of black, gray, brown, orange, yellow, green, blue, and red.

More preferably, the color is selected from the group consisting of black, brown , orange and yellow. Most preferably, the color is orange or yellow. In a most preferred embodiment, the color is yellow.

The multivitamin preparation can be administered either parenterally or enterally. Depending of the case, the physician can select the appropriate route of administration. The terms "parenteral or parenterally" refer to an administration that occurs elsewhere in the body than in the alimentary canal.

The terms "enteral and enterally" refer to an administration that occurs in the alimentary canal and intestine.

By the term "dispensing means" is meant a suitable device or apparatus for administering the multivitamin preparation to the patient. Dispensing means include but are not limited to medical bag for intrave- nous feeding, bottle for intravenous feeding, and con- tainer for storing multivitamin solution (including container used for commercial sales of multivitamin solutions) . The term is also meant to include tubing and other parts which may be necessary for administer- ing the multivitamin preparation.

Substantially opaque or colored medical bags, tubing, and bottle can be prepared by methods well known in the art. The medical bags, tubing and bottle will have to be made of materials which do not react with the solution to be placed therein. Colored plastic can be used to manufacture the medical bags, tubing or bottle in order to achieve the aim of the present invention. It is also possible to cover the dispensing means with suitable substantially opaque or colored material in order to achieve the aim of the present invention. It is preferable however, that the multivitamin solution or nutrient solution be protected from light as soon as it is prepared since the generation of peroxides is a photoreaction which occurs very rapidly. Suitable medical bags include Clintec™ Niaflex™ container manufactured by Baxter.

Suitable tubing include light protected tubing set by Baxter s.p.a. under the name Miraflo™ and Mixieva™ LPA, Miramed product line, Mirandola, Italy. Substantially opaque or colored medical bags can be prepared in the same fashion as the Miraflo™ tubing.

Factors involved in the generation of peroxides in parenteral nutritive solutions Using an HPLC technique, Helbock et al . showed that different batches of the lipid emulsion Lyposin 20%™ (Abbott, USA), made with soybean and safflower oils, were contaminated with 290 mM peroxides, while the Lyposin 10% preparation contained half that amount. Using a colorimetric method (Jiang Z.Y. et al., Lipids 1991, 26:853-856) it was confirmed that Intralipid™ 10% (Kabi-Vitrum, Sweden) made of soybean oil contained a similar concentration of peroxides (134 mM) as the 10% emulsion with the mixture of soybean and safflower oils (Lavoie J.C. et al . , Pediatrics, 1997, 99 (3):l-5). The colorimetric method was used to measure hydroperoxides is based on the oxidation of xylenol orange (Jiang Z.Y. et al., Lipids 1991; 26:853-856). In brief, at low pH, Fe^⁺ is oxidized in presence of hydroperoxides. The formed Fe³⁺ reacts with a chromo- phore (xylenol orange) which absorbs at 560 nm, linearly with a wide variety of peroxides such as H2O2, tert-butylhydroperoxide, cumene hydroperoxide (Lavoie J.C. et al., Biochem. Pharmacol . , 1994, 47:871-876).

Because of their high polyunsaturated fatty acid content, lipid emulsions are thought of as an important source of peroxides. Hence, the advantage of mixed emulsions containing medium and long chain fatty acids, which decrease the amount of unsaturation. Other nutrients are known to generate peroxides, such as cer- tain amino acids, vitamins, trace metals. Bhatia and his colleges documented that in the presence of mul- tivitamins certain amino acids (tryptophan, tyrosine, methionine, cysteine and phenylalanine) were photooxi- dized by blue light. The production of H2O2 was pro- portional to the concentration of riboflavin. Furthermore, in presence of free iron, H2O2 can generate, via the Fenton reaction (Fe2+ + H2O2 → Fe3⁺ + OH* + OH^~) , highly reactive radical oxygen species. This underlines the importance of adding only bound iron to par- enteral solutions. Depending on their final concentrations, bisulfites added as stabilizing agents to a variety of intravenous preparations have antiperoxide as well as prooxidant properties (Lavoie J.C. et al., Biochem. Pharmacol . , 1994, 47:871-876). Metabisulfite has antiperoxide properties at the concentration (1.5 mM) found in the amino acid solution Travasol™ from Baxter, whereas at higher concentrations it can become prooxidant due to the presumed formation of a sulfite radical. The multivitamin preparation MVI Pediatric™ from Rhόne-Poulenc-Rohrer contains several antioxidants such as ascorbate, tocopherol, vitamin A, mannitol, butylhydroxytoluene. But it also contains detergents to improve the solubility of the liposoluble elements. The multivitamin preparation MVI Pediatric™ which is widely used in neonatology throughout North America contains polysorbates 20 & 80 at concentrations of 0.02 and 1% (v/v) , respectively. These detergents are susceptible to peroxidation following photooxidation.

Researchers have mostly limited the investiga- tion of the level of peroxidation to individual components of parenteral solutions. Recently, the concentration of peroxides found in complete solutions of TPN was measured the in vi tro and in vivo in order to verify if the addition of a multivitamin preparation might protect the nutrients from oxidation (Lavoie J.C. et al., Pediatrics, 1997, 99 (3):l-5). It was found that, in the presence of light, the 1% (v/v) multivitamin preparation MVI was the major source of peroxides in the complete TPN solution. The admixture of a lipid emulsion had only a 10% additive effect over the initial peroxide concentration (Fig. 1) . The concentrations of nutrients used was clinically relevant. The data from Fig. 1 represent the mean ± SEM, n=3; variability is not shown when its value is smaller then that of the symbols. The level of peroxides increased (p<0.001) over time, and light exposure had a positive significant effect on the generation of peroxides (p<0.001). The lipid emulsion had only an additive effect (Lavoie J.C. et al . , Pediatrics, 1997, 99(3) :l-5) . This observation leads to the conclusion that in vi tro this multivitamin preparation which is known for its antiradical properties, does not protect the TPN solution against peroxidation. It is important to note that in presence of iron bound to carbohydrate with an elevated molecular weight (polyose) , the generation of peroxides does not take place. Fig. 2 documents that clinically relevant concentrations of iron bound to a polyose protects a TPN solution against the generation of peroxides, even in the presence of light.

As the multivitamin preparation MVI contains polysorbate, known to be photooxidizable, these experiments were repeated with a further polysorbate-free multivitamin preparation (Cernevit™ "VIT" Clintec Nutrition, Sevre, France) which allowed the comparison of the levels of peroxides in the presence and the absence of added polysorbates 80 (1%) and 20 (0.02%) . The multivitamin preparation VIT is commercially avail- able in Europe for intravenous use in adult patients. It contains higher concentrations of most vitamins. The experiments were performed in vi tro at room temperature and day light with 1% VIT in 5% glucose + 0.45% NaCl + lU/ml heparin to avoid contamination by peroxides generated by amino acids and lipids. Fig. 3 shows that VIT generates levels of peroxides comparable to MVI (Lavoie J.C. et al . , Pediatrics, 1997, 99 (3):1- 5) . Furthermore, polysorbate alone has no effect, and its admixture to VIT modifies only the rate of peroxide generation but not the level. The data from Fig. 3 represent the mean ± SEM, n=3; variability is not shown when its. value is smaller than that of the symbols.

The level of peroxides in VIT increases over time

(p<0.001). Adding the detergents to VIT at the same concentration as found in MVI does not modify significantly the level of peroxides.

When left in darkness the level of peroxides does not rise. Hence polysorbates are not the major source of peroxides.

Since this photoinduced generation of peroxides in the multivitamin preparations is independent from the detergents we evaluated the role of further components of the multivitamin preparation reported to be oxidizable, namely riboflavin and ascorbate. Indeed, ascorbate is an antioxidant known to become prooxidant at high concentration. Fig. 4 shows that separately riboflavin (22mM) and ascorbate (1.42 mM) did not generate peroxides. The data from Fig. 4 represent the mean ± SEM, n=3; variability is not shown when its value is smaller than that of the symbols. However, the association of riboflavin and ascorbate induced an increase in peroxides over time up to concentrations similar to those found with the commercialized multivi- tamin preparations.

The association of these two vitamins at concentrations found in 1% VIT produced an increase in peroxides over time which was similar to that found with VIT (Fig. 3) or MVI (Lavoie J.C. et al., Pediatrics, 1997, 99 (3): 1-5). The generation of peroxides is completely inhibited by catalase (50U/ml) suggesting that we are dealing with H2O2. This observation is corroborated by studies documenting by resonance spectroscopy that there is indeed an interaction between riboflavin and ascorbate during photooxidation. These results confirm that multivitamins represent the major source of peroxides infused to premature infants. These data suggests that infusing separately the hydrosoluble vitamins from the liposoluble vitamins as done by several teams will not modify the generation of peroxides caused by the association of riboflavin and ascorbate.

At the bedside, the concentration of peroxides received by parenterally fed premature infants was measured. Light exposure had a dramatic effect on the generation of peroxides (Lavoie J.C. et al . , Pediatrics, 1997, 99 (3): 1-5). During the 4 hrs transit time from the TPN bag to the site of infusion, the level of peroxides increased by 50% (from 141 to 215 mM) . But, when protecting the intravenous tubing from light, the level of infused peroxides remained at the same level as in the TPN bag. This was observed in the absence of lipids. Therefore, peroxidation is not limited to lipid emulsions. It is questionable whether these per- oxides will produce detectable biological effects in the whole organism where all the antioxidants interplay.

Biological and clinical effects of infused peroxides The level of peroxides infused with a solution of complete TPN (200-400 mM) corresponds with the amount of H2O2 released by 10*000 granulocytes/ml/24 h during phagocytosis. This comparison underlines that the load of infused peroxides is of biological signifi- cance and that the body should therefore be equipped to contain such an oxidative load. But a weak peroxidase activity as found in the lung and liver of newborn and especially of premature infants suggests that infused peroxides are potentially cytotoxic. There is a wide body of evidence that at the cellular level lipids and peroxides may have deleterious effects. Adding 10% Intralipid™ to erythrocytes will cause hemolysis. This reaction is catalyzed by free iron and inhibited by antioxidants such as ascor- bate and vitamin E. This observation suggest that the oxidant load associated with the lipid emulsion is responsible for the cytotoxicity. This is supported by the fact that tert-butylhydroperoxides stimulates pro- teolysis of erythrocytes at concentrations lower than what we found in vi tro in TPN solutions. The addition of this same peroxide to red cells produces an increase in a marker of peroxidation (malondialdehyde) at concentrations proportional to membrane arachidonic acid. In turn, this points out the effect of oxidants on mem- brane fatty acids. Similarly, the addition of long chain triglycerides to liver, lung or intestinal homogenates results in an increase in malondialdehyde. On the other hand, it is worth noting that the cellular effects of peroxides are not all deleterious. As men- tioned before, they can interact with DNA and thereby stimulate the synthesis of antioxidant enzymes. Along the same line, Lee Frank has hypothesized that, at the end of gestation, an increase in free radical formation in the mitochondrion was the stimulus responsible for triggering in the lungs the expression of the genes coding for the antioxidant enzymes superoxide dismu- tase, catalase and glutathione peroxidase.

Hydroperoxides contaminating solutions of parenteral nutrition modify the production of certain vasoactive eicosanoids leading to a vasoconstriction which results in a decreased pulmonary perfusion and a deterioration in gas exchanges. There is a wide body of evidence suggesting an effect of oxidants on eicosanoids. Indeed, in an umbilical vein model we confirmed that a free radical generating system contributes to decrease the production of prostacyclin (PGI2) , while peroxides, stimulate globally the production of prosta- glandins. Lipids have a combined effect resulting in vasoconstriction due to a decrease in the production of the vasodilating PGI2 and a stimulation of the vasocon- stricting eicosanoid thromboxane A2 (TxA2). A similar stimulation of TxA2 was observed in the lungs of rabbits perfused with reactive oxygen species.

In the course of studies on the effect of energy substrates on respiratory gas exchanges in premature infants, we observed that a lipid-rich regimen induced a deterioration compared to an isocaloric glucose-rich regimen. Following normal physiological ventilatory adaptations to an elevated glucose load, there was an increased CO2 production and pC02 remained unchanged. But, there was a statistical difference in pθ2 between both isocaloric regimens, with the lipid regimen exhibiting the lower value. Three different publications report a similar observation under comparable experi- mental conditions. Although pθ2 values were within a normal range, the fact that different groups report the same observation in different populations suggests a biological effect. These results may be explained by an increase in minute ventilation with the glucose-rich regimen. Alternatively, based on our in si tu findings of a vasopressor effect of the lipid emulsion in the umbilical vein model we believe that the drop in pθ2 during the lipid-rich regimen is related to a vasoconstriction mediated by prostaglandins . However, if the production of prostaglandins was induced mainly by the peroxides contaminating the TPN solutions, and if the multivitamin preparation was the major source of peroxides, one would not expect to find any difference between the isocaloric regimens differing by 4 g/kg/d of glucose and 1.2 g/kg/d of lipids. Hence, it is possible that the effects on the respiratory gas exchange is not mediated solely by the infused peroxides. As suggested by others, the prostaglandin precursors infused with the lipid emulsion might also stimulate directly eicosanoid synthesis. Following published reports of an increased morbidity represented by bronchopulmonary dysplasia in preterm infants receiving a lipid emulsion, it is questionable whether this association is related to infused peroxides. In an animal model, pentane measured in the breath of newborn pups was proportional to the dose of infused lipids. Therefore to investigate the clinical response induced by infused peroxides, the effect of this oxidant has to be separated from that of the lipid emulsion. This is what Pitkanen et al . did when studying premature infants less than five days of age who were receiving a fat-free nutrient intake and who were all ventilated. They found that the infants with elevated indices of peroxidation represented by exhaled ethane and pentane had a greater (p<0.05) probability of death or of developing bronchopulmonary dysplasia. This suggests that peroxides are of clinical significance in premature infants.

To reduce the deleterious effects associated with oxidants, it would be relevant to stimulate the antioxidant capacity of newborn infants at risk. Furthermore, we need to find ways to reduce the generation of peroxides induced by our therapeutic approaches and find preparations with antiradical as well as antiper- oxide properties.

Other characteristics and advantages of the pre^¬ sent invention will appear from the following examples. The following examples are intended to document the invention, without limiting its scope.

Example I

Preparation of a multivitamin solution

A multivitamin preparation suitable to be administrated parenterally with reduced amount of infused peroxides can be prepared as generally described below. Riboflavin is solubilized with liposoluble vitamins while vitamin C (ascorbate acid) is solubilized with hydrosoluble vitamins. The liposoluble vitamins being separated from the hydrosoluble vitamins and administrated separately. Since the riboflavin is not in contact with the vitamin C there is no generation of peroxides .

Example II Prevention of generation of peroxides in multivitamin solution with Iron Dextran

Iron dextran (Infufer™) and iron sorbitol (Jectofer™⁾ at 50 mg iron/ml were obtained from Sabex International (Montreal, Que, Canada) . FeCl2, FeSθ4, FeCl3, H2O2, tert-butyl hydroperoxide (TBH) , cumene hydroperoxide, scopoletin, HCl-cysteine, 5,5'-dithio- bis (2-nitrobenzoic acid) (DTNB) , xylenol orange

[o-cresolsulfonphthalein-3 ' , 3"-bis (methyliminodiacetic acid sodium salt)], 2, 6-di-tert-butyl-4-methyl-phenol (BHT) and sorbitol were purchased from Aldrich Chemical

Corp (Milwaukee, Wis, USA) . Dextran 5 kDa was provided by Sabex, as the same source of dextran used for FeD, while dextran 10 kDa was purchased from Pharmacia Ine

(Mississauga, Ont, Canada) . TRIS and EDTA-Na2 were obtained from Boehringer-Mannheim (Laval, Que, Canada) . The 10% Intralipid™ emulsion was purchased form Pharmacia. The 10% amino acid solution Travasol™ in glass containing 300 mg/L of m-bisulfite (BlendC) was obtained from the Clintec Nutrition Company (Mississauga, Ont, Canada) . The multivitamin preparation MVI-Paediatric™ (5 ml provided 2300 IU vitamin A , 400 IU vitamin D, 7 IU vitamin E, 200 μg vitamin K, 80 μg vitamin C, 1.2 mg thiamine, 1.4 mg riboflavin, 1.0 mg pyridoxine, 17 mg niacin amide, 5 mg pantothen- ate, 140 μg folic acid, 1 μg vitamin B12 and 20 μg biotin) was purchased from Rhone Poulenc Rorer (Montreal, Que, Canada) . The trace element preparation "Micro + 6 Pediatric™" (1 ml provided 3 mg zinc, 0.4 mg copper, 0.1 mg manganese, 4 μg chromium, 20 μg sele- nium, 60 μg iodine) was obtained from Sabex International (Montreal, Que, Canada) .

The levels of hydroperoxides were measured in 4 protocols:

• in the TPN solution, after a 3-hour incubation at day light and room temperature (control) in presence of different sources of iron. FeCl2, FeS04, FeD (iron bound to dextran) and FeS (iron bound to sorbitol) were used at concentrations ranging from 0 to 0.8 mM;

• in non-TPN solutions containing 500 μM H2O2, TBH and cumene hydroperoxide, after a 0.5-hour incubation at day light and room temperature in presence of the 4 sources of iron, at concentrations ranging from 0 to 0.8 mM;

• in the TPN solution after a 0, 1, 2 and 3-hour incu- bation at day light and room temperature in presence of 100 μM FeD, FeS, and equivalent amounts of dextran 5 kDa & 10 kDa and sorbitol; and

• in the lipid-free TPN solution after a 0 to 3-hour incubation at day light and room temperature in pres- ence of 100 μM FeD.

The TPN solution was designed to provide clinically relevant concentrations of individual nutrients delivered to newborn infants: 2.5% (w/v) amino acids + 10% (w/v) dextrose + standard electrolytes and trace elements + 1% (v/v) multivitamin preparation MVI Ped + 5.5% (v/v) lipid emulsion. The solution was prepared in water.

To quantify hydroperoxides, the ferrous oxida- tion/xylenol orange technique was used as described previously (Jiang ZY et al., Lipids, 1991, 26:853-856); at low pH, Fe²⁺ is oxidized in the presence of hydroperoxides; the formed Fe³⁺ reacts with xylenol orange to produce a chro ophore which absorbs at 560 nm linearly with the concentration of a wide range of hydro- peroxides (Lavoie JC et al., Biochem. Pharmacol . , 1994, 47:871-876). In the present study, 22.5 mM H2SO4, 90 μM xylenol orange, 225 μM FeCl2 and 3.6 mM BHT were added to 100 μL of sample, for a total volume of 1 ml. The absorbance was read (Beckman spectrophotometer DU-6) after 30 min. incubation at room temperature. Intra- and inter-assay coefficients of variation were 4 and 5%, respectively. Hydroperoxides were expressed in μM equivalents of TBH (μM equ TBH) .

Free oxygen radicals were detected using the fluorescent property of Scopoletin, which is modified in relation to its level of oxidation (Lavoie JC et al., Biochem. Pharmacol . , 1994, 47:871-876). Briefly, in TRIS buffer (50mM TRIS, 0.1 mM EDTA, pH 7.7), 100 μM iron (FeCl2 or FeSθ4) were added to 3 μM reduced sco- poletin, in presence of H2O2 (0, 50 and 125 μM) . After activation at 350 nm, a drop in fluorescence at 460 nm, was associated with an oxidation of scopoletin by reactive oxygen species. Separately, iron or H2O2 had no effect on oxidation of scopoletin. The proportion of iron freed by FeD and FeS was quantified by taking advantage of the iron binding property of the thiol function. The masking effect of FeD and FeS on the thiol function of cysteine, using FeCl3 as a positive control were compared. Iron (0 to 0.4 mM) was added to a solution of TRIS buffer containing lOOuM HCl-cysteine. After incubation for 30 min. at room temperature, 0.6 mM of DTNB was added. The absorbance was read at 412 nm (Ellman GL, Arch . Biochem . Biophys . , 1959, 82:70-77) ; corresponding cys- teine-free blanks were subtracted. Fig. 5 shows the dose-response relationships between increasing concentrations of 4 different sources of iron on the level of peroxides in a TPN solution. At 100 μM iron, peroxide concentrations were 75% lower than the control (TPN solution) . At iron concentrations >200 μM, FeD appeared to have a separate effect by reaching a steady state. Since it is well known that the level of peroxides in TPN increases with time, the low iron-induced level of peroxide could be caused either by a consumption in a Fenton-like reaction or a protection against the generation of peroxides .

In order to verify if the interaction of iron with peroxides resulted from the consumption or the inhibition of peroxide generation, the direct effect of iron on different sources of peroxides were measured

(Fig. 6) . As opposed to bound-iron, FeCl2 and FeSθ4 produced a marked drop in peroxides . The consumption of peroxides by both these sources of free iron is con- firmed by the concomitant increase in free oxygen radical formation shown in Fig. 7. The absence of a consumption of peroxides by bound iron sources (Fig. 2) led us to verify if FeD and FeS were involved in the generation of peroxides. The results from Fig. 8 show that the spontaneous light-induced generation of peroxides over time was contained by both sources of bound iron, and not associated with an initial consumption. In the lipid-free TPN solution, FeD had a similar protective effect on the light induced generation of per- oxides over time. Indeed, the lipid-free TPN solution was associated with a significant increase in peroxides over 3 hours of exposure to light (from 52.2±8.3 mM eq. TBH at 0 h to 104.0±11.1 mM eq. TBH at 3h) , which was abolished with FeD (42.6±2.6 mM eq. TBH at 3 h) . The role of dextran 5 kDa, lOkDa and sorbitol was excluded as these carbohydrates had no inhibitory effect on the generation of peroxides over time. Since Fe³+ reacts with xylenol orange a further technique was used to evaluate the proportion of Fe³⁺ freed by FeD and FeS. The results from Fig. 9 show that, as expected, the positive control FeCl3 binds strongly with cysteine, as opposed to FeD and FeS. From the slope of the linear regressions for FeS (r =0.96) and FeD (r²=0.99), it appears that 9% of FeS can bind with the thiol function of cysteine, and only 2% in the case of FeD.

Example III

Protection by infusions sets of peroxidation of parenteral nutrition solutions The effect on TPN solutions of five sets of bags and tubing offering different levels of protection against light was evaluated (Fig. 10) . The solution infused through transparent bag and tubing unprotected from light was used as control (C) . In all other set- tings TPN bags (Viaflex, Baxter, Mississauga, Ont, Canada) were protected from light with an opaque black plastic bag. The TPN solutions were run at 6 ml/h through 4 different types of tubing cut to a length of 2 m (inner volume of 10-12ml) : (1) transparent tubing (IVAC Corporation, San Diego, CA, USA); (2) orange tubing (MRT, Montreal, Qc, Canada) ; (3) yellow tubing (Mixieva™ LPA, Miramed product line, Mirandola, Italy) ;

TM

(4) black tubing completely opaque (Miraflo , Miramed product line, Mirandola, Italy) . The results obtained in ambient light (16-32 fc:foot candel) were compared with those observed under phototherapy (65 fc) . Two types of TPN solutions differing by their lipid content were used: 1. Fat-free TPN solution with the following composi- tion (corresponding nutrient intake) : total volume 150 ml (150 ml/kg/d) ; glucose 10% wt/vol (15 g/kg/d); amino acids 1.67% wt/vol (2.5 g/kg/d)

TM

(Pπmene , Baxter-Cl tec, Mississauga, Ont, Canada) ; Na 20 mM (3 mmol/kg/d) ; K 13 mM (2 mmol/kg/d); Ca 6.6 mM (1 mmol/kg/d); monobasic phosphate 233 mg/L (35 mg/kg/d) ; SO : 0.87 mM (0.13 mmol/kg/d); Mg 0.87 mM (0.13 mmol/kg/d); acetate 30 mM (4.496 mmol/kg/d); Cl 16.5 mM (2.5 mmol/kg/d); Se 13 μg/L ; I 20μg/L; Cr 2.7μg/L; Mn 66.7μg/L; Cu 266μg/L; Zn 2 mg/L; multivitamins 1.6% vol/vol (2.5 ml/d) (MVI pediatric™, Rhone- Poulenc-Rohrer, Montreal, Qc, Canada) . 2. The same TPN solution containing a 8.5% vol/vol lipid emulsion (2.5g/kg/d) (20% Travemulsion, Baxter-Clintec, Mississauga, Ont, Canada) in a all-in-one preparation.

All TPN solutions used on a given day were initially prepared in one bag which was kept in darkness (<0.06 fc) after the addition of multivitamins. The initial solution was divided in darkness between the 5 experimental set ups shown in Fig. 10.

After 2 and 23 hours in ambient light, an aliquot of the solution was taken at the end of the tubing for the immediate determination of peroxide concentra- tions. Because a pilot study showed a faster generation of peroxides under those extreme conditions, a further determination was added at 1 hour with phototherapy and with the all-in-one TPN solution. After 3 and 24 h an aliquot was sampled for the immediate determination of thiol functions. All measurements were performed in darkness (light intensity <0.06 fc) and run in triplicate. The absorbance spectra of the different tubing and of 5' -phosphate flavin mononucleo- tide were determined on a spectrophotometer (Model DU- 6, Beckman, Palo Alto, Ca, USA) between 235 and 900 nm. —_. y—

Hydroperoxide concentrations were measured by the ferrous oxidation/xylenol orange technique as described previously (Lavoie J.C. et al . , Pediatrics,

1997, 99 (3):l-5; Jiang ZY et al . , Lipids, 1991, 26:853-56) . A 20 μL sample was incubated during 30 minutes with 1ml of a solution containing H2SO4

(22.5 mmol/L) , xylenol orange (90 μmol/L) , FeCl2

(225 μmol/L), and BHT (3.6 mmol/L) in methanol. When required, lipids were separated by centrifugation (5500g, 3 min.) . The absorbency of the solution or the supernatant was read 560 nm. Results were expressed as μmol/L equivalents of TBH (tert-butylhy- droperoxide) (μmol/L equ TBH) . Intra- and inter-assay coefficients of variation were 4 and 5% respectively. The thiol functions related to the concentration of cysteine in Primene™ amino acid solution were measured as described previously (Lavoie JC et al., J. Pediatr. Gastroenterol . Nutr. , 1997, 25:307-311). A 50μL sample was added to 950 μL TRIS buffer (pH 7.6) containing 0.6 mmol/L 5, 5, -dithiobis (2-nitrobenzoic acid) . The absorbency was read at 412 nm and compared to a standard curve made with known concentrations of cysteine (Ellman GL., Arch . Biochem. Biophys. , 1959, 82:70-77). Results were expressed as mmol/L equivalent of cysteine (mmol/L equ Cys) .

All reagents were purchased from Sigma Chemical

Co. (St Louis, MI, USA) or Aldrich Chemical Co.

(Milwaukee, Wis, USA) . Results were expressed as mean

± SEM, n=3, and compared by Shcheffe's s-test. The level of significance was set at p< 0.05.

Fig. 11 presents the levels of peroxides in the fat-free TPN solution delivered with the different infusion sets. After 23 h, the TPN solution with the unprotected bag and transparent tubing exhibited the highest peroxides concentration. Even if the TPN bag was protected from light, the peroxides concentration was > 50% higher with a transparent tubing than with the other tubing. Peroxides concentrations were in the same range in the settings with the yellow, orange or black tubing after 2 and 23 h. The concentration of peroxides increased significantly over time only with the settings with the transparent tubing. The level of thiol functions (Fig. 12) , declined significantly by 33% between 3 and 24 h in the setting with the black, yellow or orange tubing, and by 45% with the transparent tubing, and by 60% in the unprotected control.

When exposed to phototherapy (Fig. 13) , the protective effect of the colored tubing was even more evident, since the peroxides concentrations were 2 to 4 fold higher with the transparent tubing. The orange tubing had significantly less protective effects on peroxides concentrations after 1, 2 or 23 hours than the black one, with a 50% difference in peroxides concentrations after 24 h. The yellow tubing exhibited intermediate results after 24 h, with a 25% higher peroxide concentration than with the black setting. Phototherapy did not modify significantly the thiol functions in solutions infused through colored tubing. However, this type of light exposure accelerated significantly the disappearance of thiol functions in both solutions infused with a transparent tubing (at 3h: 58±2% with the protected bag; 75±1% with the unprotected bag) . In the control setting only 8±1% of the initial thiol function (2.6 mmol/L equ Cys) remained after 24 h.

Fig. 14 presents the levels of peroxides in the TPN solution containing lipids delivered with the different infusion sets. Peroxides concentrations were not significantly different with the yellow or the black tubing. However, a rise (> 50%) in peroxides concentration was noted with the orange tubing after 1 and 24 hours, and a 2 to 3.5 fold increase with the transparent tubing.

The spectra of absorption of the orange and the yellow tubing covers these wavelengths of absorption of MVI and riboflavin, however the absorbance is 2 fold higher with the yellow than with the orange tubing within the range of wavelengths of riboflavin absorption (Fig. 15) . The results show that the yellow tubing is as efficient as a completely opaque black tubing in preventing further peroxidation of TPN solutions in ambient light. As opposed to the black tubing, the yellow one is, from a clinical stand point, more appropriate to visually detect air bubbles, flocculation or precipitation during the infusion. This type of tubing associated with a light protected bag is a practical solution to diminish the peroxide loads received by parenterally fed patients. In view of the rapid rate of formation of these peroxides (Figs. 11 & 14) they can be generated during the preparation and storage of the TPN solution if the bag is not protected from light. This is all the more important if an all-in-one preparation is considered. Therefore, the type of pro- tection bestowed by the yellow tubing could even be extended to TPN bags.

If no form of light protection is available, it appears from the results presented in Fig. 14 that one should attempt to decrease the duration of exposure of a complete TPN admixture to light. Therefore, to minimize peroxide generation the lipid emulsion should be "piggy-backed" on to the transparent tubing, as close as possible to the infusion site on the patient.

The multivitamin preparation is the major con- tributor to peroxide generation in TPN solutions. Darkness prevents the generation of peroxides in the multivitamin preparation. And 5 '-phosphate flavin mononucleotide is a photoexcitable component of the multivitamin preparation which catalyzes this generation of peroxides in the presence of electron donors in TPN solutions.

Claims

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows :

1. A liquid multivitamin preparation comprising riboflavin and vitamin C for enteral or parenteral administration and a reduced amount of peroxide, wherein said riboflavin is not in contact with said vitamin C.

2. A liquid multivitamin preparation for parenteral or enteral administration comprising reduced amounts of peroxide, wherein said multivitamin preparation is solubilized in the presence of iron bound to a polyose molecule .

3. A multivitamin preparation according to claim 2, wherein said iron bound polyose molecule is selected from the group consisting of Iron Saccharide, Iron Dextran and Iron Sorbitol.

4. A multivitamin preparation according to claim 2, wherein the iron bound polyose molecule is Iron Dextran.

5. A liquid multivitamin preparation comprising riboflavin and vitamin C for enteral or parenteral administration and a reduced amount of peroxide, wherein said multivitamin preparation is in two phases until the time of enteral or parenteral administration, a riboflavin phase separated from a vitamin C phase up to a molar ratio of about 1/10,000 between concentrations of riboflavin and vitamin C.

6. A method of administering a liquid multivitamin preparation comprising riboflavin and vitamin C with reduced amounts of peroxide, wherein said method comprises the step of protecting the preparation from light.

7. The method of claim 6, wherein said method comprises the step of protecting the preparation from light having a wavelength of about 200nm to about 800nm.

8. The method according to claim 7, wherein said light has a wavelength of about 300nm to about 600nm

9. The method according to claim 7, wherein said light has a wavelength of about 325nm to about 550nm.

10. The method of claim 6, wherein said step of protecting the preparation from light is effected by means of a substantially opaque dispensing means for administering the multivitamin preparation.

11. The method of claim 6, wherein said step of protecting the preparation from light is effected by means of a colored dispensing means for administering the multivitamin preparation.

12. The method according to claim 11, wherein the dispensing means is colored by at least one color or a combination thereof selected from the group consisting of black, gray, brown, orange, yellow, green, blue, and red.

13. A method according to claim 12, wherein the color is selected from the group consisting of black, brown, orange and yellow.

14. A colored container or a substantially opaque container for administering a liquid multivitamin preparation in order to protect the multivitamin preparation from light.