Evaluation of the impact of critical quality attributes and critical process parameters on quality and stability of parenteral nutrition nanoemulsions

Accepted Manuscript Evaluation of the impact of critical quality attributes and Critical Process Parameters on quality and stability of parenteral nutrition nanoemulsions Dušica Mirković, Svetlana Ibrić, Bojana Balanč, Željko Knez, Branko Bugarski PII:

S1773-2247(16)30592-5

DOI:

10.1016/j.jddst.2017.04.004

Reference:

JDDST 340

To appear in:

Journal of Drug Delivery Science and Technology

Received Date: 4 December 2016 Revised Date:

28 February 2017

Accepted Date: 2 April 2017

Please cite this article as: Duš. Mirković, S. Ibrić, B. Balanč, Ž. Knez, B. Bugarski, Evaluation of the impact of critical quality attributes and Critical Process Parameters on quality and stability of parenteral nutrition nanoemulsions, Journal of Drug Delivery Science and Technology (2017), doi: 10.1016/ j.jddst.2017.04.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Evaluation of the Impact of Critical Quality Attributes and ACCEPTED MANUSCRIPT Critical Process Parameters on Quality and Stability of Parenteral Nutrition Nanoemulsions Dušica Mirkovića,*, Svetlana Ibrićb, Bojana Balančc, Željko Knezd, Branko Bugarskic a Department of Pharmaceutical Technology, Military Medical Academy, Belgrade, Serbia Department of Pharmaceutical Technology and Cosmetology, Faculty of Pharmacy, University of Belgrade, Serbia c Faculty of Technology and Metallurgy, University of Belgrade, Serbia d Faculty of Chemistry and Chemical Engineering, University of Maribor, Slovenia

RI PT

b

Abstract

M AN U

SC

The aim of this study was to develop, characterize and evaluate concentrated nanoemulsions (20%) for parenteral nutrition. Those systems were developed by the high-pressure homogenization method. Optimal conditions for the nanoemulsion production were identified using 24-1 fractional factorial design. The characterization of physicochemical parameters was carried out immediately after the nanoemulsion production, and after 10 and 30 days. The biological control was conducted 30 days after their preparation as well. The oil phase contained the combination of the soybean (SO) and fish oil (FO) as well as the fish oil and medium-chain triglycerides (MCT), while the aqueous phase was composed of water for injections. The egg yolk phospholipids (EP) were used as surfactants alone, or in combination with Poloxamer 188 (Pl). The obtained results were in accordance with (1) the literature date e.i. quality requirements for parenteral emulsions (the droplet diameter ≤ 500 nm, PDI ≤ 0.25, absolute value of ζ–potential ≥ 25 mV, pH-value in the range of 6 to 9). It was shown that the combination of two surfactants (the egg yolk phospholipids that provides the electrostatic stabilization and Poloxamer 188 – steric stabilizer) used as emulsifiers ensures the optimal quality of the obtained nanoemulsions for parenteral nutrition.

1. Introduction

TE D

Keywords: Nanoemulsions; Parenteral nutrition; Stability; Polydispersity index; Fractional factorial design.

AC C

EP

Nanoemulsions (also called mini emulsions, ultrafine emulsions or submicron emulsions) that are used for parenteral nutrition are polydispersed, isotropic, kinetically stable, but thermodynamically unstable systems of oil-in-water (O/W) type, which should be sterile and apyrogenic [1]. Due to a very small size of droplets, parenteral nanoemulsions are „resistant” to aggregation processes (flocculation, coagulation, coalescence), what explains their long shelf life. Since they are characterized by a high degree of uniformity of droplet size, nanoemulsions are an almost ideal pharmaceutical form which has found its widespread use in many pharmaceutical areas. The quality and stability of those colloidal systems may be affected by numerous factors. The factors mostly affecting nanoemulsion properties include a proper choice, the type, quality and concentration of the used substances, the preparation method as well as the process parameters (the number of homogenization cycles and the homogenization pressure) [2]. (2) Nanoemulsions that have so far been used for parenteral nutrition in clinical practice contain either SO or FO only, or the combination of the SO and MCT as well as the combination of the SO, MCT, olive and FO.

ACCEPTED MANUSCRIPT

2

AC C

EP

TE D

M AN U

SC

RI PT

In this study, two oil combinations were used (the first containing the SO and FO, and the later composed of MCT and the FO. The utilization of the soybean oil in the emulsion for the parenteral nutrition production has the longest tradition. It is composed of long-chain triglycerides (LCT) which contain polyunsaturated essential fatty acids (linoleic – 18:2n-6 and linolenic–18:3n-3). These fatty acids are precursors of arachidonic (ω-6) and eicosapentaenoic acids (ω-3) [3]. The introduction of Medium Chain Triglycerides into the parenteral emulsion technology has brought considerable changes. These triglycerides have a greater solubilization effect, a lower accumulation in adipose tissues and the liver, a faster clearance and the resistance to peroxidation [4,5]. They are mostly used in combination with other oil types for they do not contain essential fatty acids [3]. Over the recent years, however, there has been a great deal of interest among numerous researchers in the use of fish oil. Due to the fact that the fish oil provides a rich source of ω-3 polyunsaturated fatty acids (PUFAs), especially eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), it is more and more used in critically ill patients [6]. These essential fatty acids inhibit the production of pro-inflammatory cytokines (TNF-α, IL-6, and IL-1β) and modulate the production of anti-inflammatory cytokines (IL-10) [7,8]. The application of this oil is also important in posttraumatic and postsurgical patients, early stages of sepsis, patients with inflammatory bowel diseases (Crohn’s ulceratic colitis), acute pancreatitis [9], as well as for the retinal and brain development [10,11]. If the combination of soybean and fish oil is used, the proper ratio of ω-6/ω-3 polyunsaturated fatty acids (PUFAs) could be achieved [12]. The proper selection of an emulsifying agent is necessary for the nanoemulsion stabilization. There is a small number of surfactants that are considered safe for the parenteral application. In this study, the egg yolk phospholipids (EP) and Poloxamer 188 (Pl) were used as emulsifying agents [2,3]. EP are ionic emulsifiers of electrostatic effect [13,14], while Pl is a non-ionic emulsifier of amphiphilic character, which, due to its steric effect, serves as an emulsifier. The combination of EP and Pl resulted in the formation of a better close-packed mixed film at the oil-water interface of emulsified droplets. The synergetic effect of these two emulsifiers enhances the stability of prepared nanoemulsions [15,16]. Based on the experience of a great number of researchers, it has been shown that the nanoemulsion stability can be increased considerably by adding smaller amounts of co-emulsifiers to the formula. In this experiment, Sodium Oleate (SOl) was used and remained in constant concentration (0,03%, w/w). Used as a coemulsifying and a pH-adjusting agent, it enables the achievement of better stability and enhances the ζpotential value [2]. Its localization in the interfacial layer ensures the strengthening of the integral film formed with emulsifiers by strengthening molecular bonds (actions) between EP and Pl, what, thus improves mechanical properties of the film [4,14]. It is well-known that nanoemulsions for parenteral nutrition should be isotonic. In preparing nanoemulsions glycerol (G), the most suitable isotonizing agent was used. To prevent unwanted peroxidation reactions, antioxidants – α-tocopherol (Toc) and thioglycolic acid (TA) were used [3,14,17]. The known fact is that nanoemulsions for parenteral nutrition are the most stable at the pH value range of 6 – 9 [18]. During the sterilization process and the shelf life as well, the pH value decreases due to the hydrolysis of triglycerides to their constituents, i.e. free fatty acids. To ensure the upper pH limit, 0,1mol/l aqueous solution of Sodium Hydroxide is added to nanoemulsions prior to their sterilization. In addition, the proper of the preparation methodology, as well as the optimally adjusted manufacturing parameters are considered also important. There are various high- and low-energy methods used for the nanoemulsion preparation [2,19,20]. The high-pressure homogenization method that is mostly used for the nanoemulsion preparation is also used for these study. Synergetic effects of several forces such as cavitation, hydraulic shear and intense turbulence are used for carrying out this process.

ACCEPTED MANUSCRIPT

3

The aim of the study was to prepare the nanoemulsions for parenteral nutrition and investigate the impact of independent variables (Critical Material Attributes – the types of oil and surfactant phases as well as the Critical Process Parameters – the number of homogenization cycles and the homogenization pressure) on the Critical Quality Attributes (the mean size of droplets, the droplet size distribution i.e. polydispersity index – PDI and ζ–potential). 2. Materials and methods 2.1. Materials

SC

RI PT

Soybean oil (Lipoid Purified Soybean Oil 700) was purchased from Lipoid GmbH, Germany, the Fish oil – Oleum Jecoris (Ph. Eur. 7.5), Miglyol 812® contains MCT – Medium-Chain Triglycerides, α-tocopherol were obtained from Caelo, GmbH, Hilden, Germany, egg phospholipids with 80% phosphatidylcholine (Lipoid® E80) and Lipoid Sodium Oleate B were kindly supplied by the Lipoid GmbH, Germany, Kolliphor® P 188 – Poloxamer 188 (Ph. Eur.) was also a kind gift from BASF (Ludwigshafen, Germany). Glycerol (Ph. Eur.) and Sodium hydroxide (Ph. Eur.) were purchased from Merck, Germany, Thioglycolic acid was purchased from Sigma–Aldrich Chemie GmbH (Steinheim, Germany). Water used in the experiment was double distilled and obtained from the Milli Q-water purification system (Millipore, MA). 2.2. Methods

M AN U

2.2.1. Experimental design

In order to evaluate the influence of formulation and process variables on the performance of O/W nanoemulsions for parenteral nutrition, 24-1 fractional factorial design was applied. The type of oil phase, the surfactant (egg yolk phospholipids) with or without the use of the second surfactant (Poloxamer 188), the number of homogenization cycles and the pressure were recognized as the critical quality attributes for formulations and processes (input parameters). These variables were varied according to the 24-1 fractional factorial design, as presented in Table 1.

1 2 3 4 5 6 7 8

X1

X2

X3

X4

A1

1

4

300

A2

1

4

700

A1

2

4

700

A2

2

4

300

A1

1

10

700

A2

1

10

300

A1

2

10

300

A2

2

10

700

EP

Formulation

TE D

Table 1. Experimental matrix according to the 24-1 fractional factorial design

AC C

X1 - oil phase (A1 – mixture of FO and SO, A2 – mixture of FO and MCT); X2 - surfactant (1 –EP, 2 – mixture of EP and Pl); X3 – number of cycles; X4 – pressure (bar) In order to identify and define the influence of independent variables on a dependent variable, obtained values were fitted into the following model:

ACCEPTED MANUSCRIPT

4

y = b0 + b1X1 + b2X2+ b3X3 + b4X4 + b12X1X2 + b13X1X3 + b14X1X4 Where y – the dependent variable; X1-X4 – independent variables; b0-b4 – regression coefficients that demonstrate the influence of independent variables on the dependent variable y; b12, b13, b14 – regression coefficients that demonstrate the interaction between corresponding variables.

RI PT

Since this type of design is fractional, it was possible to calculate only selected interaction terms. 2.2.2. Preparation of nanoemulsions

Table 2. Composition of nanoemulsion formulations

SC

Nanoemulsions for parenteral nutrition studied in this paper are prepared by the high-pressure homogenization method, according to the technology standards laid down in numerous literature sources [20].

COMPOSITION OF NANOEMULSION FORMULATIONS (%, w/w) Formulation FO

MCT

EP

1.

10

10

―

1.20

2.

―

10

10

1.20

3.

10

10

―

1.20

4.

―

10

10

1.20

5.

10

10

―

1.20

6.

―

10

10

1.20

7.

10

10

―

1.20

8.

―

10

10

1.20

Pl

SOl

G

Toc

M AN U

SO

TA

Water to [g] 100

―

0.03

2.50

0.01

0.01

―

0.03

2.50

0.01

0.01

100

0.60

0.03

2.50

0.01

0.01

100

0.60

0.03

2.50

0.01

0.01

100

―

0.03

2.50

0.01

0.01

100

―

0.03

2.50

0.01

0.01

100

0.60

0.03

2.50

0.01

0.01

100

0.60

0.03

2.50

0.01

0.01

100

TE D

SO – Soybean oil, FO – Fish oil, MCT – Medium-Chain Triglycerides, EP – Egg yolk phospholipids, Pl – Poloxamer 188, SOl – Sodium Oleate, G – Glycerol, Toc – α-tocopherol, TA – Thioglicolic acid For a more complete insight, it could be said that the preparation process for parenteral nanoemulsions was carried out in three phases: the pre-emulsification, the homogenization and the sterilization phase.

EP

I. Pre-emulsification phase

AC C

During the pre-emulsification phase, a crude emulsion (pre-emulsion) was prepared. The EP emulsifier was dissolved in the mixture of the FO and SO, or the mixture of the FO and MCT (Table 2). In spite of the fact that high oil concentrations result in an increase of the droplet size [21,22], it was, however, agreed that, in this experiment, the level of concentration would be high (20%, w/w in total). The reason for that was our intention to investigate to what extent such a high oil concentration would affect the droplet size. Then, antioxidants (Toc and TA) at the unchanged concentration were added. The aqueous phase, i.e. external phase, (see the Table 2) was, as planned, composed of hydrosoluble emulsifier Pl, then G, SOl and double-distilled

ACCEPTED MANUSCRIPT

5

water at 100%, w/w (the amount of water changed, depending on whether or not Pl was added). The (1) aqueous and oil phase were then separately heated at the temperature of around 70°C under constant stirring using a magnetic stir bar (IKA RCT Basic, Germany) at 800 rpm until they were completely dissolved. After that, the aqueous and oil phase were mixed, and the obtained mixture was pre-emulsified using the high-shear mixer Ultra-Turax T25 (Janke & Kunkel Ika-Labortechnik, Germany) at 13500 rpm for five minutes. In that way, a crude emulsion with the droplet size of around 2 µm was obtained. II. The high-pressure homogenization

RI PT

The nanoemulsion droplets were obtained by processing a crude emulsion through a high-pressure homogenizer (Gea Niro Panda plus 2000, Italy). The pressure range that this device can achieve is from 150 to 2000 bars. During this experimental phase, the pressure was 300, i.e. 700 bars, while the number of homogenization cycles was four, i.e. ten. These values varied as shown in the Table 1. The temperature of the entire homogenization process was maintained at 40 °C. The procedure was repeated as many times as defined by the design of the experiment (Table 1).

SC

III. Sterilization

M AN U

After the prepared formulations were cooled to the temperature of 25 °C, their pH value was adjusted to around 8,5 (Table 7) by adding 0,1 mol/l solution of Sodium hydroxide [3]. The nanoemulsions were sterilized in the autoclave („Zirbus“ technology, Germany) at the temperature of 121 °C and the pressure of 2 bars for 15 minutes [14]. By that, the third phase of the preparation of these nanoemulsions was completed. In order to monitor the stability of prepared samples, they were kept at the room temperature before starting the measurement of defined parameters as was recommended by manufacturers of those nanoemulsions for parenteral nutrition (B. Braun, Fresenius Kabi, Baxter), 2.3. Characterization of obtained nanoemulsions

2.3.1. Particle size analysis

TE D

The physico-chemical stability characteristics of the obtained nanoemulsions were monitored immediately after their preparation (0h), as well as after the periods of 10 and 30 days. The monitoring included the visual examination and measurements of the mean size of lipid droplets, the PDI, ζ–potential and the pH value. (2) The biological control including the sterility and endotoxin tests of the obtained nanoemulsions was conducted as well after their preparation and after 30 days according to the Ph. Eur. standards [23]. All samples were protected from light.

AC C

EP

The mean size of lipid droplets (the hydrodynamic droplet diameter) was measured by the Photon Correlation Spectroscopy method (also known as the Dynamic Light Scattering). The Zetasizer Nano–ZS instrument (Malvern Instruments, Ltd. UK) with the measurement range of 0.6 nm up to 6 µm was used. In this experiment, we also utilized the capability of the instrument to calculate the droplet size distribution (PDI) parallely with the measurement of the mean value of the hydrodynamic diameter of lipid droplets. All the measurements were done at the temperature of 25 °C, and each sample was diluted 500 times with highly purified water before it was measured. The measurements were repeated three times, and the obtained results were presented as the mean value.

ACCEPTED MANUSCRIPT

6

2.3.2. ζ–Potential Zetasizer Nano – ZS was also used to determine the ζ–potential. Samples were diluted 1:500 (v/v) with purified water. The determination was performed at the temperature of 25 °C. The electrophoretic mobility (m/s) was converted into the ζ–potential (mV). 2.3.3. pH value measurements

RI PT

The pH measurement of the prepared nanoemulsions was conducted prior to and after the sterilization process, and during the period of monitoring the prepared nanoemulsion stability. For the result verification purposes, the measurements were repeated three times, and presented results represented their mean value. The measurements were done by direct submerging the electrode of the calibrated pH-meter (Mettler Toledo, Seven Go, Swiss) into the samples at the room temperature. 3. Results and discussion

M AN U

SC

On visual examination, it was determined that all prepared nanoemulsions were homogenous with milkywhite appearance. When stored at the room temperature, these properties of nanoemulsions remained the same. Results of the mean droplet size, the PDI and the ζ–potential immediately after the emulsion is prepared, as well as 10 and 30 days after its preparation are presented in Table 3. All the results are presented as mean values of the three measurements and include standard deviations (StD). (1) Table 3. Results of mean droplet size, PDI and ζ–potential ± Std (0h, after 10 and 30 days)

Formulation

DEPENDANT VARIABLES

Mean droplet size (nm ± StD)

0h

after 10

Polydispersity index ± StD

after 30

days

days

0h

after 10

after 30

days

days

ζ-potential (mV ± StD)

0h

after 10 days

after 30 days

251±5

323±4

311±9

0.237±0.019

0.216±0.015

0.246±0.023

-38.8±0.551

-41.1±0.404

-42.4±0.850

2.

284±3

247±1

274±6

0.170±0.009

0.225±0.012

0.250±0.021

-34.9±0.945

-40.4±1.150

-37.7±1.069

3.

194±1

191±0.5

208±1

0.106±0.006

0.116±0.023

0.197±0.011

-32.0±0.577

-29.8±0.902

-27.7±0.404

4.

190±1

188±0.7

189±0.5

0.119±0.019

0.108±0.008

0.099±0.004

-29.7±1.910

-28.0±0.900

-28.4±1.887

5.

217±2

311±4

342±5

0.176±0.019

0.235±0.250

0.250±0.025

-34.2±0.954

-44.0±0.100

-41.0±1.331

6.

227±3

308±8

354±1

0.142±0.007

0.234±0.019

0.250±0.031

-31.7±0.551

-37.3±0.643

-40.8±1.569

7.

192±2

192±0.6

190±1

0.069±0.005

0.121±0.010

0.104±0.012

-29.8±0.557

-29.3±1.420

-31.3±1.452

0.068±0.016

0.060±0.007

0.067±0.006

-28.9±0.557

-27.7±1.400

-33.2±1.415

EP

AC C

8.

TE D

1.

181±2

183±1

183±1

ACCEPTED MANUSCRIPT

7

3.1. The influence of independent variables (critical material attributes and critical process parameters) on the mean droplet size Estimated regression coefficients representing the influence of varied variables on the mean droplet size are presented in Table 4. Table 4. Regression coefficients which demonstrate the influence of independent variables and the interaction between corresponding variables on a mean droplet size.

10 days after the preparation -11.36 -54.34 5.31 -9.96 8.46 8.36 -6.66

30 days after the preparation -6.23 -63.80 10.78 -4.60 -0.100 7.53 -16.75

RI PT

Immediately after the preparation 3.52 -27.78 -12.70 1.95 -7.30 -3.72 10.13

SC

Coefficient Variable X1 X2 X3 X4 X 1X 2 X 1X 3 X 1X 4

X1 - oil phase; X2 - surfactant; X3 – number of cycles; X4 – pressure (bar)

AC C

EP

TE D

M AN U

Based on the results presented in the table, it is evident that the X2 factor has the greatest influence on the mean droplets size. If two surfactants (EP and Pl) are used for these nanoemulsions, the droplets of the smallest size are obtained. This effect is negative, and it increased during aging of the emulsion. The existence of large-size droplets that might be formed during the shelf life (so-called natural aging) or under the influence of some destabilizing factors might cause the pulmonary embolism [14, 17, 18]. Generally, the droplet size of the parenteral emulsion must be smaller than the diameter of the smallest capillaries. With regard to the assessed impact of the droplet size to the parenteral emulsion stability, the most complete parameters are defined by the United States Pharmacopeia USP 36, Chapter 729 „Globule size Distribution in Lipid Injectable Emulsion“, Method I – Light-Scattering Method and Method II – Measurement of Large Globule Content by Light Obscuration or Extinction Method“). According to this standard, the mean droplet size must not exceed 500 nm, but if droplets larger than 5 µm are present, their number should be less than 0.05% [24]. In this experiment, the mean droplet size of all measured samples ranged from 181 to 353 nm (Table 3). If only one emulsifier (for instance, EP) is used, the interfacial film does not have a sufficient strength to resist the shear forces occurring in the pre-homogenization phase, then during the high-pressure homogenization and the sterilization process as well. The combination of EP and Pl enables a dual system stabilization. The ionic emulsifier, EP, provides the electrostatic stabilization of the system, while the non-ionic emulsifier, Pl, contributes to the improvement of the stability of prepared nanoemulsions by its steric stabilizing effects. In fact, the duplex interfacial film enables the formation of droplets of smaller size and the stabilization of the system by the synergetic interaction of its mechanical and electrostatic properties [25-27]. The amount of the emulsifier is also a very important parameter that affects the production technology and the quality of parenteral nanoemulsions. Insufficient amounts of the emulsifier do not provide effective barrier among the droplets, and may also enhance the possibility of emulsion destabilization [28]. The amount of surfactants in the 20% standard parenteral emulsions (Intralipid® 20%, Lipofundin MCT/LCT® 20%) is 1.2%, w/w. In this experiment, when only EP were used, the amount of 1.2%, w/w emulsifier varied to the maximum of 1.8% w/w in formulations

ACCEPTED MANUSCRIPT

8

to which Pl was added. The FDA recommended maximum amount of Pl which can be used for the production of parenteral preparations is 0.6%, w/w [29]. Large amounts of surfactants may be toxic, and, due to the fact that emulsions thus formed are very viscous, their application may be very painful [3]. The results of this investigation showed that the droplets are smaller when the A1 oil phase (the combination of the soybean and fish oil) is used. However, during storage, this effect (X2) is opposite – droplets are smaller in the A2 oil phase (the combination of MCT and the FO). The number of homogenization cycles (X3) has a positive effect on the droplet size – greater number of cycles give larger droplets while with increasing the homogenization pressure (X4), the droplet size decreases.

RI PT

3.2. The influence of independent variables (critical material attributes and critical process parameters) on the polydispersity index – PDI The PDI indicates the width of the droplet size distribution and the homogeneity of these preparations. According to Muller at all, the PDI value for parenteral nanoemulsions should be less or equal than 0.25 [30]. The estimated regression coefficients representing the influence of varied variables on the droplet size distribution – PDI immediately after the preparation and 10 and 30 days after that are presented in Table 5.

SC

Table 5. Regression coefficients which demonstrate the influence of independent variables and the interaction between corresponding variables on a PDI.

Immediately after the preparation 0.14 -0.011 -0.045 -0.022 0.005 0.0023 0.0001

10 days after the preparation 0.0048 -0.076 0.011 -0.018 -0.022 0.004 -0.021

30 days after the preparation -0.019 -0.11 -0.018 -0.032 -0.015 0.047 -0.014

M AN U

Coefficient Variable X1 X2 X3 X4 X 1X 2 X 1X 3 X 1X 4

AC C

EP

TE D

X1 - oil phase; X2 - surfactant; X3 – number of cycles; X4 – pressure (bar)

Fig. 1. Graphical interpretation of the interaction between the oil phase type and the number of homogenization cycles

ACCEPTED MANUSCRIPT

9

RI PT

It is evident that the presence of two surfactants has the highest influence on the PDI. This effect is negative, i.e. when Pl is added to the emulsion, the PDI is smaller. This effect increases during aging of emulsion. The PDI is smaller when the A1 oil phase is used. However, during storing, this effect is opposite – droplets are smaller in the A2 oil phase. The number of homogenization cycles and pressure also have the negative effect on the PDI. It is interesting that during storage, 30 days after the nanoemulsion preparation, the interaction term between the oil phase and the number of cycles increased. This interaction is presented in Figure 1. When the A1 oil phase is used, with increasing the number of cycles, the PDI decreased after storage for 30 days. In the formulations with the A2 oil phase, the PDI increased with increasing the number of cycles. According to these findings, when the A1 oil phase is used, it is recommended to use a greater number of homogenization cycles. On the other hand, when the A2 oil phase is used, it is better if the number of cycles is smaller. 3.3. The influence of independent variables (critical material attributes and critical process parameters) on the ζ – potential

SC

There are numerous factors that affect the ζ–potential value of nanoemulsions, of which the most important are: the pH, the ionic strength, the type and concentration of emulsifiers. Stable nanoemulsions are considered to have absolute values greater than ±25 mV [31]. The estimated regression coefficients representing the influence of varied variables on ζ–potential immediately after the preparation as well as 10 and 30 days after that are presented in Table 6.

M AN U

Table 6. Regression coefficients which demonstrate the influence of independent variables and the interaction between corresponding variables on a ζ–potential.

Immediately after the preparation 1.20 2.40 1.35 0 -0.40 -0.35 -0.60

TE D

Coefficient Variable X1 X2 X3 X4 X 1X 2 X 1X 3 X 1X 4

10 days after the preparation 1.39 6.04 0.16 -0.74 -0.54 0.69 0.037

30 days after the preparation 0.29 5.16 -1.26 0.41 -0.94 -0.71 -0.84

X1 - oil phase; X2 - surfactant; X3 – number of cycles; X4 – pressure (bar)

AC C

EP

According to the obtained results, the type and quantity of surfactants have the greatest impact on the ζ– potential. The presence of Pl significantly increases the zeta potential. This effect is more pronounced after storage. The oil phase type has also a positive effect. That means that higher values of ζ–potential are obtained when A2 oil phase is used. However, this effect is decreasing during the emulsion storage, and it is negligible 30 days after the emulsion preparation. The number of homogenization cycles has a positive effect immediately after the preparation. However, after storage for 10 days, this effect is decreasing, and after 30 days, it becomes negative – a greater number of homogenization cycles give a smaller value of the ζ–potential. The pressure has no effect on the ζ–potential immediately after the emulsion preparation. However, this effect is positive after the emulsion storage for 30 days.

ACCEPTED MANUSCRIPT

10

3.4. pH values of nanoemulsions

RI PT

It is known that the pH value is a very crucial parameter characterizing the quality and stability of parenteral nanoemulsions, what particularly comes to the fore after the sterilization of this pharmaceutical form. The USP Pharmacopoeia defines the pH value range (6 – 9) within which parenteral nanoemulsions are stable [32]. In case of nanoemulsions used for the parenteral nutrition, the pH adjustment to 9 prior to the sterilization process is very important since there is a small possibility of hydrolyzing phospholipids and triglycerides to free fatty acids in such an environment. The pH – values of nanoemulsions (before sterilization, after sterilization and after 10 and 30 days) are presented in Table 7. There were no significant changes within this period. Table 7. The pH – values of nanoemulsions (before sterilization, after sterilization and after 10 and 30 days).

After sterilization 7.80

8.65 8.60 8.50 8.65 8.55 8.60 8.55

7.75 7.85 7.80 7.90 7.80 7.90 7.85

After 10 days 7.80 7.75 7.80 7.80 7.80 7.80 7.85 7.85

SC

Before sterilization 8.55

M AN U

Formulation 1. 2. 3. 4. 5. 6. 7. 8.

After 30 days 7.70 7.75 7.80 7.75 7.80 7.70 7.85 7.85

(2) Finally, the monitoring of the biological stability of prepared nanoemulsions showed that all the preparations remained sterile and endotoxin – free during the investigation. 4. Conclusions

AC C

EP

TE D

In this study, the physical stability of prepared O/W nanoemulsions for parenteral nutrition formulated according to the 24-1 fractional factorial design were monitored. All the formulations were prepared by the high-pressure homogenization method. It was shown that all the samples remained stable during the investigation. Immediately after the preparation, as well as after 10 and 30 days, the characterization of their physical properties (the visual examination, the mean droplets size, the droplet size distribution (PDI), ζ– potential, and the pH-value) was carried out. The obtained results were in accordance with the quality requirements for parenteral emulsions (the droplet diameter ≤ 500 nm, PDI ≤ 0.25, absolute value of ζ– potential ≥ 25 mV, pH-value in the range of 6 to 8). Based on the set objectives and obtained results, it was shown that the type and amount of surfactants have the greatest impact on the mean droplet size, the droplet size distribution and the ζ–potential. It was also shown that the synergetic effects of two surfactants (the egg yolk phospholipids and Poloxamer 188), when used as emulsifiers, ensures the optimal quality of the obtained nanoemulsions for parenteral nutrition. (2) These results correlate to the literature data indicating that nanoemulsions of optimal quality can be obtained if electrostatic and steric stabilizers are combined. During the investigation period, all the nanoemulsions remained sterile and endotoxin-free.

ACCEPTED MANUSCRIPT

11

Acknowledgments The authors are grateful to Lipoid GmbH, Germany for donating Lipoid® E80 and Lipoid Sodium Oleate B. The authors also would like to thankful BASF (Ludwigshafen, Germany) for kindly donating Kolliphor® P 188.

References

AC C

EP

TE D

M AN U

SC

RI PT

[1] McClements DJ, Nanoemulsions versus microemulsions: terminology, differences, and similarities. Soft Matter 8 (2012) 1719–29. [2] Klang V, Valenta C, Lecithin-based nanoemulsions. J. Drug Del. Sci. Tech 21 (2011) 55–76. [3] Hippalgaonkar K, Majumdar S, Kansara V, Injectable lipid emulsions advancements, opportunities and challenges, AAPS PharmSciTech 11 (2010) 1526-40. [4] Bach AC, Storck D, Meraihi Z, Medium-chain triglyceride-based fat emulsions: an alternative energy supply in stress and sepsis, JPEN 12 (1988) 82-88. [5] Driscoll DF, Nehne J, Franke R, et al., The influence of medium-chain triglycerides on the stability of all-in-one formulations, Int J Pharm 240 (2002) 1-10. [6] Singer P, Shapiro H, Theilla M, et al., Anti-inflammatory properties of omega-3 fatty acids in critical illness: Novel mechanisms and an integrative perspective, Intensive Care Med 34 (2008) 1580-92. [7] Waitzberg DL, Torrinhas RS, Fish oil lipid emulsions and immune response: what clinicians need to know? Nutr Clin Pract 24 (2009) 487–99. [8] Calder PC, Grimble RF, Polyunsaturated fatty acids, inflammation and immunity, Eur J Clin Nutr 56, suppl 3 (2002) 14–19. [9] Lei QC, Wang XY, Xia XF et al., The Role of Omega-3 Fatty Acids in Acute Pancreatitis: A MetaAnalysis of Randomized Controlled Trials, Nutrients 7 (2015) 2261-73. [10] Furst P, Kuhn KS, Fish oil emulsions: what benefits can they bring? Clin Nutr 19 (2000) 7–14. [11] Waitzberg DL, Torrinhas RS, Jacintho TM, New parenteral lipid emulsions for clinical use, J Parenter Enter Nutr 30 (2006) 351-67. [12] Grimm H, Tibell A, Norrlind B, et al., Immunoregulation by parenteral lipids: impact of the n-3 to n6 fatty acid ratio, JPEN 18 (1994) 417–21. [13] Yang R, Zhang X, Li F, et al., Role of phospholipids and copolymers in enhancing stability and controlling degradation of intravenous lipid emulsions, Colloid Surf A Physicochem Eng Asp 436 (2013) 434-42. [14] Wabel C, Influence of lecithin on structure and Stability of Parenteral Fat Emulsions [dissertation], Nürnberg: Universität Erlangen 1998. [15] Benita S, Levy MY, Submicron emulsions as colloidal drug carriers for intravenous administration: comprehensive physicochemical characterization, J Pharm Sci 82 (1993) 1069-79. [16] Levy MY, Schultze W, Fuhrer C, et al., Characterization of diazepam submicron interface: role of oleic acid, J Microencapsul 11 (1994) 79-92. [17] Floyd AG, Top ten considerations in the development of parenteral emulsions, Pharm Sci. Technol. Today 2, (1999) 134–43. [18] Driscoll DF, Lipid Injectable Emulsions: 2006, Nutrition in Clinical Practice 21 (2006) 381-6. [19] Sonal S, Sushama T, Razdan BK, Nanoemulsions: Formulation methods and stability aspects, World Journal of Pharmacy and Pharmaceutical Sciences 3 (2013) 2214-28.

ACCEPTED MANUSCRIPT

12

AC C

EP

TE D

M AN U

SC

RI PT

[20] Tadros T, Izquierdo P, Esquena J, et al., Formation and stability of nano-emulsions, Adv. Colloid Interface Sci (2004) 108–109, 303–18. [21] Ferezou J, Bach AC, Structure and metabolic fate of triacylglycerol and phospholipid-rich particles of commercial parenteral fat emulsions, Nutrition 15 (1999) 44–50. [22] Ferezou J, Nguyen TL, Leray C et al., Lipid composition and structure of commercial parenteral emulsions, Biochim Biophys Acta 1213 (1994) 149–58. [23] Biological tests. 2010. European Pharmacopoeia-7th ed., Vol. 1, Council of Europe, Strasbourg, France. p. 153-198. [24] Chapter 729. The United States Pharmacopeia 36 and The National Formularity 31 (USP36/NF31), Rockville, Marylend, USA, Globule Size Distribution in Lipid Injectable Emulsions, (2013) 321–23. [25] Tadros TF, Emulsion stability. In: P. Becher, editor. Encyclopedia of Emulsion Technology, Vol. 1, New York: Marcel Dekker (1983) 129–285. [26] Benita S, Friedman D, Weinstock M, Physostigmine emulsion: a new injectable controlled release delivery system, Int J Pharm 30 (1986) 47–55. [27] Trotta M, Pattarino F, Ignoni T, Stability of drug-carrier emulsions containing phosphatidylcholine mixtures, European Journal of Pharmaceutics and Biopharmaceutics 53 (2002) 203-8. [28] Han F, Li S, Yin R, et al., Effect of surfactants on the formation and characterization of a new type of colloidal drug delivery system: Nanostructured lipid carriers, Colloids and Surfaces A: Physicochem. Eng. Aspects 315 (2008) 210-16. [29] Inactive Ingredient Search for Approved Drug Products. Available online: http://www.accessdata. fda.gov/scripts/cder/iig/index.cfm). [30] Müller RH, Schmidt S, Buttle I, et al., SolEmuls® – novel technology for the formulation of i.v. emulsions with poorly soluble drugs, Int J Pharm 269 (2004) 293–302. [31] Marín-Quintero D, Fernández-Campos F, Calpena-Campmany AC, et al., Formulation, design and optimization for the improvement of nystatin–loaded lipid intravenous emulsion. J Pharm Sci 102, 11 (2013) 4015-23. [32] Driscoll DF, Lipid Injectable Emulsions: Pharmacopeial and Safety Issues; Pharmaceutical Research 23 (2006) 1959-69.