Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Pharmacy 237

_____________________________ _____________________________

In Situ Gelling Drug Delivery Systems for Periodontal Anaesthesia

BY

MARIE SCHERLUND

ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2000

Abstract

In this thesis local anaesthetic formulations based on PEO-PPO-PEO block copolymers(PEO and PPO being poly(ethylene oxide) and poly(propylene oxide), respectively) ornonionic cellulose ethers undergoing temperature- or dilution-induced thickening wereinvestigated. The aim of the work was to develop formulations, which can be easilyadministered to the periodontal pocket, stay at the application site, give a fast onset ofanaesthesia and have a duration sufficient to perform periodontal scaling procedures.

Emulsions, (mixed) micellar solutions and microemulsions fulfilling the requirementsstated above were achieved by combining the active ingredients lidocaine and prilocainewith the nonionic block copolymers Lutrol F127 and Lutrol F68. The criticalmicellisation and gelation temperatures of the systems were found to be interconnectedand influenced by the total polymer concentration and the polymer mixturecomposition, as well as the presence of cosolutes and pH.

A low-viscous isotropic phase that turns into a high-viscous hexagonal phase as thewater content increases was found by combining Lutrol F68, water, a eutectic mixtureof lidocaine and prilocaine and Akoline MCM. The system has a slower release ratecompared to the microemulsion formulation, which might make it suitable forindications where a longer duration is needed.

Finally, a temperature-induced gelling system was achieved by adding lidocaine andprilocaine to mixtures of ethyl(hydroxyethyl)cellulose (EHEC) and sodium dodecylsulfate (SDS), hexadecyltrimethylammonium bromide (CTAB) or myristoylcholinebromide systems at or just below the surfactant concentration found to give a maximumviscosity increase at room temperature. In particular, the myristoylcholine bromidesystem may be interesting considering its antibacterial properties and biodegradability.

Till Anders, Mamma och Mormor

5

Contents

ABBREVIATIONS AND DEFINITIONS ............................................................................................... 6

PAPERS DISCUSSED .............................................................................................................................. 7

1 AIM OF THE THESIS ..................................................................................................................... 8

2 INTRODUCTION............................................................................................................................. 9

3 LOCAL ANAESTHESIA............................................................................................................... 113.1 LOCAL ANAESTHESIA OF THE PERIODONTAL POCKET .................................................................... 11

3.1.1 Requirements........................................................................................................................ 113.2 FORMULATIONS FOR ANAESTHETISING ORAL MUCOSA .................................................................. 11

3.2.1 Important formulation factors.............................................................................................. 11

4 POLYMERS SHOWING REVERSED TEMPERATURE-DEPENDENCE............................ 134.1 BLOCK COPOLYMERS, PEO-PPO-PEO.......................................................................................... 13

4.1.1 Self-association .................................................................................................................... 134.2 CELLULOSE ETHERS, EHEC AND HM-EHEC................................................................................ 16

4.2.1 Polymer-surfactant interactions........................................................................................... 16

5 FORMULATION STRATEGIES ................................................................................................. 175.1 FORMULATIONS OF OIL AND WATER .............................................................................................. 17

5.1.1 Solubilisation ....................................................................................................................... 175.1.2 Fixation................................................................................................................................ 17

5.2 FORMULATION TYPES OF SPECIAL INTEREST.................................................................................. 18

6 METHODS ...................................................................................................................................... 206.1 PHOTON CORRELATION SPECTROSCOPY (PCS) .............................................................................. 206.2 NUCLEAR MAGNETIC RESONANCE (NMR) SELF-DIFFUSION .......................................................... 206.3 DIFFERENTIAL SCANNING CALORIMETRY (DSC) ........................................................................... 216.4 SMALL ANGLE X-RAY DIFFRACTION.............................................................................................. 216.5 RHEOLOGY .................................................................................................................................... 226.6 IN VITRO DRUG RELEASE ............................................................................................................... 22

7 RESULTS AND DISCUSSION...................................................................................................... 237.1 TEMPERATURE-INDUCED GELATION OF BLOCK COPOLYMER SYSTEMS .......................................... 23

7.1.1 Micelle and gel formation .................................................................................................... 237.1.2 In vitro drug release............................................................................................................. 297.1.3 In vivo results....................................................................................................................... 31

7.2 DILUTION-INDUCED GELATION OF BLOCK COPOLYMER SYSTEMS.................................................. 317.2.1 Phase behaviour of a polymer/water/oil system................................................................... 327.2.2 Gel formation ....................................................................................................................... 347.2.3 In vitro drug release............................................................................................................. 35

7.3 GELATION OF POLYMER-SURFACTANT INTERACTING SYSTEMS................................................. 367.3.1 Clouding behaviour ............................................................................................................. 367.3.2 Rheological behaviour ......................................................................................................... 377.3.3 In vitro drug release............................................................................................................. 40

8 CONCLUSIONS ............................................................................................................................. 42

9 OUTLINE FOR FUTURE WORK ............................................................................................... 44

10 ACKNOWLEDGEMENTS............................................................................................................ 45

11 REFERENCES................................................................................................................................ 46

6

Abbreviations and Definitions

cac Critical aggregation concentrationcmc Critical micellisation concentrationcmt Critical micellisation temperatureCP Cloud pointCTAB Hexadecyltrimethylammonium bromideD Diffusion coefficientd Distance between planes in the crystalDSC Differential scanning calorimetryDSethyl Average substitution degree of ethyl groupsEHEC Ethyl(hydroxy ethyl) celluloseEO Ethylene oxideFTPGSE Fourier transform pulsed gradient spin-echoG´ Elastic or storage modulusG´´ Viscous or loss modulusH EnthalpyH1 Hexagonal phaseH2 Reversed hexagonal phaseHM-EHEC Hydrophobically modified ethyl(hydroxy ethyl) celluloseI Cubic phasek Boltzman constantL2 Reversed micellar phaseL3 Bicontinuous phaseLα Lamellar phaseMSEO Molar substitution degree of hydroxyethyl groupsn Order of the diffraction patternNaCl Sodium chlorideNaSCN Sodium thiocyanateNMR Nuclear magnetic resonanceo/w Oil-in-waterPCS Photon correlation spectroscopyPEO-PPO-PEO Poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide)PO Propylene oxideRh Hydrodynamic radiusS EntropySDS Sodium dodecyl sulphateT Temperaturew/o Water-in-oilVAS Visual analogue scaleVRS Verbal rating scaleφ Phaseλ Radiation wavelengthθ Scattering angleη Viscosity

7

Papers Discussed

This thesis is based on the following papers, which will be referred to by their Romannumerals in the text.

I Scherlund, M., Malmsten, M. and Brodin, A., Stabilisation of a thermosettingemulsion system using ionic and nonionic surfactants. Int. J. Pharm., 173 (1998)103-116.

II Scherlund, M., Malmsten, M., Holmqvist, P. and Brodin, A., Thermosettingmicroemulsions and mixed micellar solutions as drug delivery systems forperiodontal anaesthesia. Int. J. Pharm. 194 (2000) 103-116.

III Scherlund, M., Brodin, A. and Malmsten, M., Micellisation and gelation in blockcopolymer systems containing local anaesthetics. Int. J. Pharm., Accepted.

IV Scherlund, M., Welin-Berger, K., Brodin, A. and Malmsten, M., A localanaesthetic block copolymer system undergoing phase transition on dilution withwater. Eur. J. Pharm. Sci., Submitted.

V Scherlund, M., Brodin, A. and Malmsten, M., Nonionic cellulose ethers as potentialdrug delivery systems for periodontal anaesthesia. J. Colloid Interface Sci., 229(2000) 365-374.

Reprints were made with permission from the journals.

8

1 Aim of the Thesis

The aim of this work has been to develop local anaesthetic in situ gelling systemssuitable for administration into body cavities such as the periodontal pocket. For thispurpose, poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) (PEO-PPO-PEO) block copolymers and nonionic cellulose ethers were used together with a mixtureof two amphiphilic drug substances (lidocaine and prilocaine) to make formulationsundergoing either temperature-induced or dilution-induced thickening. The systemswere investigated regarding the influence of the concentration of active ingredients andpolymers, pH and cosolutes on physico-chemical properties such as micellisation, phasebehaviour, in vitro drug release, and stability. The studies were performed usingrheology, NMR self-diffusion, photon correlation spectroscopy (PCS), differentialscanning calorimetry (DSC) and small angle X-ray diffraction techniques.

9

2 Introduction

This thesis deals with drug delivery systems undergoing transitions from a low to a highviscous state as a consequence of an increase in temperature or dilution with water(saliva). Such systems have gained significant interest among formulators within thepharmaceutical field as drug delivery vehicles for dermal, nasal, ocular, oral, buccal,vaginal, rectal and parenteral administration (Schmolka, 1977; Chen-Chow and Frank,1981; Miller and Donovan, 1982; Miyazaki et al., 1986; Abd Elbary et al., 1991; Wangand Johnston, 1995; Edsman et al., 1998). Despite the interest in these systems, topicalformulations for local anaesthesia of the periodontal pocket are scarce.

Both PEO-PPO-PEO block copolymers and ethyl(hydroxy ethyl) cellulose (EHEC)show a reversed temperature dependence and can be formulated into systems turninginto a �gel� state at body temperature, the latter in the presence of ionic surfactant. Thegelation is fully reversible and upon cooling the systems turn into a low viscous stateagain. Furthermore, due to their amphiphilic nature PEO-PPO-PEO block copolymerscan self-associate into various structures, forming, e.g., low viscous phases that turn intohigh viscous hexagonal or cubic phases upon dilution with water. Also, the ability toform micelles makes the block copolymers interesting as drug delivery vehicles whereenhanced drug solubility, reduced drug hydrolysis, controlled release of the activecomponent, or improved bioavailability is required (Collett and Tobin, 1979; Lin andKawashima, 1985; Gilbert et al., 1986; Nagarajan et al., 1986; Tarr and Yalkowsky,1987; Hurter and Hatton, 1992; Kabanov et al., 1992; Saito et al., 1994; Alexandridisand Hatton, 1995).

The local anaesthetic agents used throughout the thesis are lidocaine and prilocainewhich form a eutectic mixture (Brodin et al., 1984). The aim of the work with thesesubstances is to achieve systems which can be easily administered to the periodontalpocket, stay at the application site due to a viscosity increase, give a fast onset ofanaesthesia lasting throughout the dental procedure and thereafter be easily rinsed outwith water causing a fast decline in anaesthetic effect.

There are a number of challenges when preparing these kind of systems since the phasebehaviour of both PEO-PPO-PEO block copolymers and the nonionic cellulose ethersare affected by various factors such as temperature, copolymer composition, molecularweight, concentration, and presence of cosolutes such as surfactants, electrolytes, andhydrophobic substances (Vadnere et al., 1984; Carlsson et al., 1986; Karlström et al.,1990; Malmsten and Lindman, 1992a; Linse, 1993a,b; Mortensen and Pedersen, 1993;Wanka et al., 1994; Alexandridis et al., 1995; Alexandridis and Hatton, 1995;Alexandridis et al., 1996a; Alexandridis and Holzwarth, 1997; Jönsson et al., 1998).Some of these factors are studied for PEO-PPO-PEO block copolymer formulationsundergoing temperature-induced thickening in paper I-III, PEO-PPO-PEO blockcopolymer formulations undergoing dilution-induced thickening in paper IV andnonionic cellulose ethers undergoing temperature-induced thickening in paper V. Thepolymers investigated in the present work are shown in Table 1.

10

Table 1. Polymers investigated in the present work.

1MSEO = Molar substitution degree of hydroxyethyl groups, DSethyl = Average substitution degree of ethylgroups.

2CP = Cloud point.

3Structural formula of Lutrol® F127 a=99, b=65 and Lutrol® F68 a=79, b=28.

4Structural formula of EHEC R=H and HM-EHEC R=(CH2 - CH2O)x - C6H4 - C9H19.

Polymers Type of Polymer Meanmolecular

weight(g/mol)

MSEO/DSethyl

1Nonyl-phenol(mol%)

CP2

(°C)

Poloxamer 407;Lutrol F127

Nonionicblock copolymer3

12500 - - >100

Poloxamer 188;Lutrol F68

Nonionicblock copolymer3

8600 - - >100

EHEC Nonioniccellulose ether4

127000 0.9/1.5 - 34

HM-EHEC Nonioniccellulose ether4

120000 2.1/0.8 0.8-1.2 36

H (O - CH2 - CH2) a (O - CH - CH2) b (O - CH2 - CH2) a OHCH3

{

O

OHO

CH 2

O

OO

CH 2

CH 2

OH

HOOH

CH 2O

CH 2

3

CH 2

CH 2OCH 2CH 2

OCH 2CH 3

O O

HO

CH 2

O

OO

CH 2

CH 2

O

O

HOO

CH 2OH

CH 2

O

CH 2

CH2

OCH2

CH 3

R

{CH2

CH2

CH2

OH

CH

11

3 Local Anaesthesia

3.1 Local anaesthesia of the periodontal pocket

3.1.1 Requirements

Periodontal scaling is a dental procedure performed to remove plaque from the surfaceof a tooth in order to avoid build-up of bacteria which unattended can lead to loss of thetooth. The scaling procedure is regarded as unpleasant and painful by two-thirds ofpatients and therefore requires local anaesthesia (Svensson et al., 1994). The mainanaesthetic techniques used for this purpose are nerve block/infiltration anaesthesiaeither alone or in combination with topical anaesthesia. However, since many peoplesuffer from fear of needle insertions (Milgrom et al., 1997; Skaret et al., 1998) andwould like to avoid numbness of the lip and tongue a fast-acting and effective topicalanaesthetic which is easy to administer is desirable.

3.2 Formulations for anaesthetising oral mucosa

3.2.1 Important formulation factors

A number of topical anaesthetics are used in dentistry, although a draw-back with mostformulations available on the market is a low degree of efficacy. EMLA® cream(lidocaine/prilocaine 5%) has been found to be effective in preventing procedure-relatedpain in the mouth (Holst and Evers, 1985; Svensson et al., 1993; Svensson et al., 1994;Donaldson and Meechan, 1995) but since it was not intended for the oral cavity, iteasily spreads out (Svensson and Petersen, 1992) causing anaesthesia also in areas notwanted and a spread of its bitter taste.

Lidocaine and prilocaine, as seen in Figure 1, were selected as active ingredients for thedevelopment of a suitable formulation for anaesthetising the periodontal pocket prior toscaling procedures. Such a formulation needs to have a fast onset time, sufficient depthof penetration, sufficient duration, and should be easy to apply, stay at the applicationsite and in addition be non-irritant and stable at normal storage conditions.

Figure 1. Chemical structure of the active ingredients.

Lidocaine

CH3

NH - CO - CH2 - N

CH3 C2H5

C2H5

CH3

NH - CO - CH - NH - CH2 - CH2 - CH3

Prilocaine

CH3

12

A formulation which is easy to apply and stays at the application site once applied canbe achieved by using excipients that will undergo transition from a low viscous statebefore application to a high viscous state after application. Such transitions can beinduced, e.g., by an increase in temperature or dilution of the formulation with water(saliva). Examples of polymers showing temperature-induced gelation are the PEO-PPO-PEO block copolymers and nonionic cellulose ethers, the latter together with ionicsurfactants. Dilution-induced gelation can be achieved, e.g., by utilising the phase-behaviour of the PEO-PPO-PEO block copolymers.

13

4 Polymers Showing Reversed Temperature-dependence

Aqueous solutions of nonionic polymers and surfactants containing ethylene oxide (EO)show in many properties anomalous temperature dependencies such as clouding(reduced solubility), induced micellisation and increasing micellar size, gel formationetc. upon increasing temperature (Lindman et al., 1990). There have been severaltheories put forward in the literature as explanations for this unusual behaviour based onsolvent-solvent (Kjellander and Florin, 1981; Kjellander, 1982), solute-solvent (Langand Morgan, 1980; Goldstein, 1984) and solute-solute interactions (Karlström, 1985).The first of these deals with the temperature-dependent structuring of water around theEO units. At low temperatures it is assumed that the EO chains are surrounded by azone of increased structuring of water which breaks down as the temperature isincreased, eventually resulting in phase separation. Naturally, at sufficiently hightemperatures T∆S will be greater than ∆H resulting in complete miscibility of EO andwater (Kjellander and Florin, 1981; Kjellander, 1982). The solute-solvent model on theother hand explains the phase separation by a breakage of the hydrogen bondingbetween water molecules and ether oxygen of the EO groups at higher temperatures(Lang and Morgan, 1980; Goldstein, 1984). The third model was first introduced todescribe the phase behaviour of PEO in water and focuses on the temperature-dependentconformational changes of the polymer. The PEO chain may exist in a large number ofconformations having different energies. The conformation, which is gauche around theC-C bond and anti around the C-O, has the lowest energy but also a low statisticalweight. At low temperature this conformation, which has a large dipole moment,dominates. The importance of other conformations of higher statistical weight increaseswith increasing temperature. The resulting conformational changes will make the PEOchains progressively less polar with increasing temperature. As this occurs, the PEOchains will interact less favourably with water and more favourably among themselves,leading to an increased tendency for phase separation (Karlström, 1985).

Apart from PEO, the latter model has been used successfully to predict the solutionproperties of other clouding polymer systems such as PEO-PPO-PEO blockcopolymers and cellulose derivatives (Karlström et al., 1990; Johansson et al., 1993;Linse, 1993b; Zhang et al., 1994; Thuresson et al., 1995; Svensson, 1998). Furthermore,a similar model, assuming temperature-dependent EO-EO, PO-PO and EO-POinteractions, has been used to predict the micellisation behaviour of PEO-PPO-PEOblock copolymer solutions (Linse and Malmsten, 1992; Linse, 1993a,b).

4.1 Block copolymers, PEO-PPO-PEO

4.1.1 Self-association

Due to the amphiphilic nature of PEO-PPO-PEO block copolymers they are able to self-aggregate to form a variety of associated structures such as micelles and liquidcrystalline phases, as well as the reversed counterparts and different microemulsionstructures (Linse, 1993 a,b; Mortensen and Pedersen, 1993; Wanka et al., 1994;

14

Alexandridis et al., 1995; Alexandridis and Hatton, 1995; Zhang and Khan, 1995;Alexandridis et al., 1996b; Alexandridis et al., 1998). The micelles formed in diluteaqueous solution consist of a hydrophobic core of propylene oxide and a hydrophiliccorona of ethylene oxide (Zhou and Chu, 1988; Wanka et al., 1990; Brown et al., 1991;Linse and Malmsten, 1992; Malmsten and Lindman; 1992a; Mortensen, 1992; Yu et al.,1992; Mortensen and Pedersen, 1993; Alexandridis et al., 1994; Wanka et al., 1994;Hecht et al., 1995). Due to this structure the micellar solutions are capable ofsolubilising hydrophobic solutes (Alexandridis and Hatton, 1995). The solubilisatingcapacity of hydrophobic substances in the micelles has been found to be higher foraromatic than for aliphatic substances and larger the smaller the solubilised molecule(Nagarajan et al., 1986). Furthermore, the solubilising capacity increases withincreasing polymer concentration (Collett and Tobin, 1979; Saito et al., 1994; Kwonand Kataoka; 1995), temperature and relative propylene oxide content of the polymer(Hurter and Hatton, 1992; Saito et al., 1994). Another interesting feature of the blockcopolymer micelles is that they remain stable for long times on reducing theconcentration below the critical micellisation concentration (cmc), i.e., on dilutionwhich is a promising factor for pharmaceutical applications (Malmsten and Lindman,1992a).

A characteristic property of these systems is that the self-assembly is highly temperaturedependent. More specifically, an increasing temperature facilitates self-association,resulting in a strong decrease in the cmc (Linse and Malmsten; 1992; Linse, 1993a) andat a given polymer concentration a critical micellisation temperature (cmt) occurs(Alexandridis and Hatton, 1995). Several techniques have been described in theliterature for measuring the cmc/cmt of block copolymers, e.g., light scattering,fluorescence spectroscopy, surface tension, nuclear magnetic resonance spectroscopy(NMR) and differential scanning calorimetry (DSC) (Alexandridis and Hatton, 1995).The micellisation, detected with DSC, has been shown to be endothermic and the self-assembly therefore inferred to be driven by the entropy gain when the micelles form,presumably mostly due to a change in hydrophobic behaviour of PPO and to a lesserextent also to the increase in the non-polar conformational state of PEO with increasingtemperature (Wanka et al, 1990; Yu et al., 1992; Linse, 1993a,b; Alexandridis et al.,1994; Hecht and Hoffman; 1994; Wanka et al., 1994; Alexandridis and Hatton, 1995;Patterson et al., 1996).

Another characteristic property of PEO-PPO-PEO block copolymer systems is thethermoreversible gelation displayed by some concentrated block copolymers attemperatures close to room temperature (Wanka et al., 1990; Brown et al., 1991; Yu etal., 1992; Wang and Johnston, 1991). For example, a 20% w/w aqueous solution ofLutrol® F127 typically shows a quite low viscosity (about 0.02 Pa s) (Malmsten andLindman, 1993) at low temperatures. On increasing temperature, gel formation occursover a narrow temperature range and the elastic modulus of the gel formed is quite high(Brown et al., 1991; Malmsten and Lindman, 1993; Wanka et al., 1994; Alexandridisand Hatton, 1995). The gels display a largely elastic mechanical behaviour over a widefrequency range, contrary for example, to many semi-dilute or concentrated polymersolutions which at least at low frequencies are largely viscous in nature (Ferry, 1980).

15

The driving forces for this thermal gelation and the structure of the �gels� have beendiscussed extensively in the literature. The gel transition has been related to intrinsicchanges in the micellar properties (Rassing and Attwood, 1983), entropic changesinvolving locally ordered water molecules close to the hydrophobic propylene oxidesegments (Vadnere et al., 1984) and the possibility of an ordered three-dimensionalstructured state or network (Wanka et al., 1990; Wang and Johnston, 1991). Neutronscattering has shown that the gel formation occurs when the micelle concentrationreaches the critical volume fraction of 0.53, i.e. where the micelles are locked into ahard sphere crystalline structure due to their high volume density (Mortensen et al.,1992; Mortensen, 1992; Mortensen and Pedersen, 1993; Alexandridis et al., 1994;Hecht and Hoffman 1994; Alexandridis and Hatton, 1995; Mortensen, 1996). If shear isapplied to such a phase it may align the system, e.g., into one macroscopic monodomainwith the micelles organised on a body-centred cubic lattice (Mortensen et al., 1992;Mortensen, 1992; Mortensen, 1996). Although the shear has a dramatic influence on thetexture the phase behaviour has been found to be largely unaffected by shearing(Mortensen, 1996). The onset of gelation can easily be detected by, e.g., oscillatingrheological measurements (Wanka et al., 1990; Brown et al., 1991).

Both the micellisation and gelation are affected by a range of factors, e.g., temperature,copolymer composition, molecular weight, and concentration, as well as presence ofcosolutes such as surfactants, electrolytes, and hydrophobic substances (Malmsten andLindman, 1992a; Linse, 1993a,b; Mortensen and Pedersen, 1993; Wanka et al., 1994;Alexandridis et al., 1995; Alexandridis and Hatton, 1995; Alexandridis et al., 1996a;Alexandridis and Holzwarth, 1997). This rich behaviour of block copolymers describedabove together with their commercial availability make them interesting as drugdelivery vehicles, since they can improve solubility, reduce hydrolytic degradation,achieve controlled release, result in improved bioavailability etc.(Collett and Tobin,1979; Lin and Kawashima, 1985; Nagarajan et al., 1986; Tarr and Yalkowsky, 1987;Yokoyama et al., 1990; Hurter and Hatton, 1992; Kabanov et al., 1992; Yokoyama,1992; Saito et al., 1994; Alexandridis and Hatton, 1995).

The aspects of toxicity of the formulation components and their degradation productsare a major concern when developing pharmaceutical formulations. The toxicity ofPEO-PPO-PEO block copolymers have been investigated for different pharmaceuticalapplications and found to be quite low, and decreasing with increasing PEO content andmolecular weight (Schmolka, 1991). The two block copolymers used in this study havemostly been reported to show a low toxicity. However, Lutrol F68 has been found toinduce phospholipidosis in rats at high concentrations and Lutrol F127, despite its highmolecular weight and PEO content, has been reported to cause damage to the retinawhen applied to the eye for two weeks (Magnusson et al., 1986; Davidorf et al., 1990).The toxicity of these polymers is a complex issue and therefore, the toxicologicalaspects have to be considered from each formulation perspective. Furthermore, thesepolymers degrade in presence of oxygen forming aldehydes which can react with otherformulation components and this must naturally also be addressed (Bergh et al., 1998).

16

4.2 Cellulose ethers, EHEC and HM-EHEC

4.2.1 Polymer-surfactant interactions

Analogous to the PEO-PPO-PEO block copolymers, the nonionic cellulose ethersEHEC (ethyl(hydroxyethyl)cellulose) and hydrophobically modified EHEC (HM-EHEC) phase separate from aqueous solution when heated above a certain temperature,usually referred to as the cloud point (CP) (Jullander, 1957; Karlström, 1985; Carlssonet al., 1989; Carlsson et al., 1990; Karlström et al., 1990). The clouding phenomenon isa reversible process, i.e., when lowering the temperature below the cloud point the two-phase system will turn into a one-phase system again.

There are many factors that affect the CP, for example properties of the polymer itselfsuch as type and degree of substitution, molecular weight, and concentration, but alsoaddition of cosolutes such as surfactants, alcohols and electrolytes (Carlsson et al.,1986; Karlström et al., 1990; Jönsson et al., 1998). Cloud-point measurements are easyto perform and is a useful method for gaining knowledge about the interactions insolutions of cosolutes and thermoseparating polymers.

The interaction between cellulose ethers and surfactants is of particular interest to thepresent work since the aggregate formation between these substances may havedramatic effects on both solubility and rheology, which may be used in a formulationstrategy. When ionic surfactants bind to the cellulose ethers their electrically chargedhead group will effectively increase the solubility of the polymer due to electrostaticrepulsion between the polymer-surfactant aggregates. At the same time, however, thehydrophobic part of the added surfactant will increase the overall hydrophobicity of thepolymer which may lead to a decrease in polymer solubility. Thus, the influence of thesurfactant on the solubility of the polymer is determined by the balance between thesetwo opposing effects, and is therefore sensitive to the presence of excess electrolyte(Carlsson et al., 1986).

EHEC has been shown to form thermoreversible gels in the presence of ionicsurfactants at certain conditions (Carlsson et al., 1990; Lindman et al., 1990). Thenetwork formation is caused by a delicate interplay between connectivity provided bythe polymer chains and swelling caused by the polymer-bound ionic surfactant(Walderhaug et al., 1995; Cabane et al., 1996) and possibly also temperature-dependentbinding of the surfactant (Carlsson et al., 1990; Zana et al., 1992; Kamenka et al.,1994). Quantitatively, the effects of polymer-surfactant aggregate formation dependstrongly on the conditions, but in certain cases this might result in orders of magnitudeincrease in viscosity and elastic modulus (Carlsson et al., 1990; Iliopoulos et al., 1991;Holmberg et al., 1992; Nyström et al., 1995; Cabane et al., 1996; Holmberg andSundelöf, 1996; Kjöniksen et al., 1998).

This rather unique phenomenon is interesting from a pharmaceutical point of view, andin situ gelling systems based on EHEC have previously been reported in the literature,e.g., for nasal and ophthalmic drug delivery (Rydén and Edman, 1992; Lindell andEngström, 1993).

17

5 Formulation Strategies

5.1 Formulations of oil and water

5.1.1 Solubilisation

In developing a formulation for application to the oral mucosa the organolepticproperties, toxicity and ease of handling have to be considered when choosing a suitablesolvent. Therefore, water is usually preferred since water continuous systems areconsidered to be well perceived by patients, non-toxic and easy to rinse out after thetreatment.

The active ingredients used in the present work, i.e., lidocaine and prilocaine, are moreor less hydrophobic depending on the pH of the system and a common way ofsolubilising such substances in water is to make emulsions by adding an amphiphile tothe system. Oil-in-water (o/w) emulsions offer advantages in topical administration,including water miscibility, thus making them washable, ease of spreading onto theapplication site, and a low degree of irritation (Becher, 1985). However, sinceemulsions are kinetically stabilised, but thermodynamically unstable, these willeventually phase separate, which may cause stability problems.

Microemulsions contain water, oil and amphiphile(s) making up a single opticallyisotropic and thermodynamically stable liquid solution (Danielsson and Lindman, 1981;Florence and Attwood, 1998; Jönsson et al., 1998). They are easy to prepare and showexcellent long-term stability. Beside these advantages, microemulsions may offeradditional benefits for periodontal uses such as an increased drug release rate, due tosmaller droplet size compared to emulsions, and improved handling due to thetransparency of the formulation, the latter making it easy to see the instruments in theperiodontal working area. However, preparing microemulsions as drug delivery systemsoften requires a high surfactant concentration that might be a disadvantage from atoxicological point of view.

There have been several reports of their use in pharmaceutical formulations, frequentlywith good results (Attwood, 1994; Israelachvili, 1994; Lawrence, 1994; Constantinides,1995; Paul and Moulik, 1997; Malmsten, 1998). Particularly interesting microemulsionsystems for pharmaceutical applications are those based on nonionic surfactants, sincethese generally require no additional surfactant or cosurfactant to form a microemulsionand are frequently less toxic than those based on ionic surfactants (Schick and Fowkes,1967; Gloxhuber, 1980; Schmolka, 1991).

5.1.2 Fixation

Another requirement in preparing a formulation for anaesthetising the periodontalpocket is that it needs to remain at the application site at the same time as being easy toapply. To achieve such a system one can take advantage of the differences between the

18

ex vivo and in vivo conditions, e.g., regarding the temperature and presence of water(saliva). In particular, systems undergoing a transition from a low-viscous solution to ahigh-viscous system after administration offer interesting opportunities in this respect.

5.2 Formulation types of special interest

There have been several reports in the literature of in situ thickening systems where theviscosity increase is induced by a temperature increase, dilution or ion-exchange(Moorhouse et al., 1981; Rozier et al., 1989; Engström and Engström, 1992; Engströmet al., 1992; Norling et al., 1992; Rydén and Edman, 1992; Lindell and Engström, 1993;Esposito et al., 1996). Particularly interesting for the application in focus here, aretemperature- and dilution-induced thickening which were investigated using blockcopolymers and nonionic cellulose ethers.

The PEO-PPO-PEO block copolymer Lutrol F127 displays thermoreversible gelationat temperatures close to room temperature (Wanka et al., 1990; Brown et al., 1991;Wang and Johnston, 1991; Yu et al., 1992). The gel formation occurs over a narrowtemperature range turning such solutions from a low viscous state to a gel state of highelastic modulus (Brown et al., 1991; Malmsten and Lindman, 1993; Wanka et al., 1994;Alexandridis and Hatton, 1995). In addition to acting as a thickener PEO-PPO-PEOblock copolymers are able to form micelles which can solubilise the active ingredientsand hence, no additional surfactants are required. Another advantage when dealing withpharmaceutical preparations is their relatively low toxicity (Schmolka, 1991) as well astheir commercial availability.

Furthermore, PEO-PPO-PEO block copolymers can self-associate into a large numberof different structures such as micelles, microemulsions and liquid crystalline phases(Linse, 1993a,b; Mortensen and Pedersen, 1993; Wanka et al., 1994; Wu et al., 1994;Alexandridis et al., 1995; Alexandridis and Hatton, 1995; Zhang and Khan, 1995;Alexandridis et al., 1996b). Therefore it should be possible to develop a low viscositysolution which will turn into a liquid crystalline phase of some stiffness upon dilution(e.g., a lamellar (Lα), hexagonal (H) or cubic (I) phase). An advantage in formulatingsuch a system is that a higher amount of polymer can be used and as a consequence alarger amount of active ingredients can be incorporated into the system. However, thephases of interest can be very small resulting in limitations in choice of formulationcomposition.

Considering this, Lutrol F127 and F68 or Lutrol F68 were used together with theeutectic mixture of lidocaine and prilocaine to develop formulations undergoing eithertemperature- or dilution-induced thickening. The influence of polymer properties,temperature and presence of cosolutes on the rheology, stability, drug release rate etc. ofthese systems were investigated in paper I-IV.

EHEC has also been found to form thermoreversible gels at certain conditions inpresence of ionic surfactants (Carlsson et al., 1990, Lindman et al., 1990). Although the

19

EHEC systems are more viscous before reaching the gel state and less viscous afterreaching the gel state compared to PEO-PPO-PEO formulations such systems may stillbe sufficient to meet the formulation requirements. The ionic surfactants needed fortemperature-induced thickening may be a disadvantage from a toxicological point ofview although the amounts needed are quite low as is the release rate of the surfactantfrom the system (25% w/w released after 1 hour application time) (Lindell, 1996). Infact, using cationic surfactants with antibacterial properties could be an advantage forthe application in focus here, although caution should be applied due to the irritatingeffects of cationic surfactants. Such systems are particularly interesting if readilybiodegradable surfactants are used. Also, the surfactant micelles formed on the polymerbackbone contain less water in their interior than micelles formed by the PEO-PPO-PEO systems and may thus provide a better protection against hydrolysis of thesolubilised drug.

Therefore, EHEC together with anionic or cationic surfactants was combined withlidocaine and prilocaine to develop temperature-induced thickening systems. HM-EHEC systems were also investigated regarding its solubilising capacity. The influenceof the properties of the polymer and addition of cosolutes on CP, rheology and drugrelease rate of these systems were studied in paper V.

20

6 Methods

6.1 Photon correlation spectroscopy (PCS)

PCS is a technique for determining the average particle size and broadness of the sizedistribution of particles dispersed in liquids, within the size range from a fewnanometers to about 1 µm or to the onset of sedimentation. The measurement isconducted by illuminating the sample by a narrow monochromatic and coherent laserbeam. The light scattered by the particles at an angle (typically 90°) is recorded by adetector and the output fed to a correlator. The dispersed particles are in constantBrownian and/or thermal motion and the observed scattered intensity fluctuates along atime axis. By monitoring the autocorrection function of the intensity scatter, the PCSmeasurement describes the movement of the particles in the form of the diffusioncoefficient (D) from which the particle size (hydrodynamic radius, Rh) can be calculatedusing the Stokes-Einstein equation:

Rh = kT/6πηD (1)

where k is the Boltzman constant, T is the temperature, and η the viscosity of thecontinuous phase.

This technique has been used in several studies to measure the size of micellaraggregates in block copolymer systems (Al-Saden et al., 1982; Zhou and Chu, 1988;Wanka et al., 1990; Brown et al., 1991) and has been taken as a standard inpharmaceutical development work (International Standard ISO 13321). Since theformulations of investigation in the present work consist of micelles in the nanometerrange this technique is useful in determining the size of these aggregates.

6.2 Nuclear magnetic resonance (NMR) self-diffusion

NMR self-diffusion is a useful method for determining the structure of isotropic andanisotropic surfactant, polymer, and polymer/surfactant systems (Lindman et al., 1987;Söderman and Stilbs, 1994; Söderman and Olsson, 1997). An aspect of NMR self-diffusion measurement making it especially interesting for these types of systems is thatthe diffusion of all components may be monitored simultaneously.

For the self-diffusion NMR measurements the Fourier transform pulsed gradient spin-echo technique (FTPGSE) (Stilbs, 1987) was used. In this technique the self-diffusioncoefficients are determined by using a spin-echo sequence in combination with pulsedmagnetic field gradients which labels nuclei at different spatial positions with differentprecession frequencies. During the time of the experiment movement of spins will resultin defocusing of the echo coherence, quantified as the diffusion coefficient D. TheFTPGSE technique has been described previously in the literature as a useful method indetermining the structure of aqueous systems based on EO-PO-EO block copolymers

21

and nonionic cellulose ethers (Stilbs, 1987; Malmsten and Lindman, 1992a,b; Carlssonet al., 1989).

6.3 Differential scanning calorimetry (DSC)

DSC is a thermal analysis method for detecting changes in physical and/or chemicalproperties of materials as a function of temperature by measuring the heat changesassociated with such processes. The sample and an inert reference are placed in atemperature controlled chamber and the heat flow required to maintain the sample andthe reference at the same temperature is measured. This results in either an absorption ofheat (endothermic reaction) or a release of heat (exothermic reaction) (McNaughton andMortimer, 1975).

DSC has been described in the literature as a possible technique for studying blockcopolymer self-assembly (Alexandridis and Hatton, 1995). The micellisation processhas been shown to be endothermic and therefore inferred to be driven by the entropygain when the micelles form. This is presumably an effect of both hydrophobicallydriven self-assembly, and an increase in the non-polar conformational state of PEO withincreasing temperature (Wanka et al, 1990; Yu et al., 1992; Linse, 1993a,b;Alexandridis et al., 1994; Hecht and Hoffman; 1994; Wanka et al., 1994; Alexandridisand Hatton, 1995; Patterson et al., 1996). In the present work, this technique was usedto study the critical micellisation temperature (cmt) for some of the PEO-PPO-PEOblock copolymer formulations, particularly at higher concentrations, where lightscattering or NMR self-diffusion are more difficult to use in order to obtain informationon the onset of micellisation.

6.4 Small angle X-ray diffraction

Small angle X-ray diffraction is a useful method in studying liquid crystalline phases.The crystal structure is built up of unit cells which are the smallest building units of acrystal, describing the three-dimensional relationship between the molecules in thecrystal lattice. Both the symmetry and the dimensions of the unit cell, as well as thelong-range organisation, can be studied by X-ray diffraction. The sample is fixed andsealed between two mica windows and placed in the beam of monochromatic X-rays.The scattered radiation satisfying Bragg´s law:

nλ = 2d sin θ (2)

where n is the order of the diffraction pattern, λ is the radiation wavelength, d is thedistance between planes in the crystal and θ is the scattering angle, gives a diffractionpattern (Stout and Jensen, 1989). This technique has been extensively used in theliterature for investigating the structure of liquid crystalline phases formed by bothsurfactants and block copolymers (Luzzati, 1968; Fontell, 1992; Larsson, 1994;

22

Svensson et al., 1998) and was used in the present work to determine the structure ofsome liquid crystalline phases of a block copolymer system.

6.5 Rheology

Rheology is an important tool in determining the macroscopic behaviour of polymersolutions and its dependence on temperature, polymer-surfactant interaction, additivesetc. By subjecting a sample to oscillating shear between a cone and a plate the elasticmodulus (G´) and loss modulus (G´´) can be measured at different temperatures andfrequencies. Such measurements were performed in the present work using a stresscontrolled rheometer to investigate the thermogelling properties of block copolymersystems and the interaction between nonionic cellulose ethers and surfactants.

6.6 In vitro drug release

In vitro drug release methods are frequently used to gain information about the releaseprofiles of active ingredients when developing formulations. In the present work, twocompartment glass diffusion cells, as seen in Figure 2, were used for the experiments.The formulation was applied on a synthetic cellulose membrane (Spectra/Por 4,MWCO 12,000 - 14,000) separating the donor and receptor compartments. Degasseddistilled water was used as sink solution. The amount of active ingredients released overtime was analysed using a spectrophotometer or high performance liquidchromatography (HPLC). A reference formulation was included as a control in allexperiments.

Figure 2. A schematic illustration of the two compartment cell used for the drug diffusion studies.

Membrane

Magneticstirrer

Sampling tube

23

7 Results and Discussion

7.1 Temperature-induced gelation of block copolymer systems

7.1.1 Micelle and gel formation

The active ingredients, lidocaine and prilocaine were incorporated into an aqueoussystem as oil-in-water emulsions. Lutrol F127 was added to the system both as aviscosity controlling and surface active agent, i.e., both to emulsify the sparinglysoluble active ingredients and to act as a thickener for the system. Despite thesolubilising capacity of this polymer it was found that an additional surfactant wasrequired to improve the stability of the formulation (paper I). Interestingly, the additionof Lutrol F68 resulted in a clear one-phase system, which was investigated further inpapers II and III.

(a)

(b)

Figure 3a) Elastic modulus (G´) of a 20% w/w aqueous solution of Lutrol F127. b) The phasebehaviour of aqueous solutions of Lutrol F127.

Concentration (% w/w)14 16 18 20 22 24 26

Tem

pera

ture

(°C

)

0

5

10

15

20

25

30

35

Gel

Solution

Temperature (°C)5 10 15 20 25 30 35 40

G' (

Pa)

10-3

10-2

10-1

100101

102

103

104

105

24

Particularly for the latter system, the occurrence and properties of the micelles formed,as well as its dependence on parameters such as total polymer concentration, polymermixture composition, concentration of active ingredients, and pH, is important. Theseeffects were therefore investigated in some detail. For this purpose both PCS, NMR andDSC were employed. Of particular importance for the present work is also the capacityof these systems to solubilise local anaesthetic agents such as lidocaine and prilocaine,which was investigated, e.g., by chromatographic techniques.

As discussed previously, one of the interesting properties of the block copolymers is theability to form thermoreversible gels. This is illustrated in Figure 3a and b for aqueoussolutions of Lutrol F127. In analogy to the micellisation process, the temperature ofgelation decreases with an increasing polymer concentration. Interestingly, at evenhigher temperatures, the gels �melt� and low viscosity solutions are formed once more(Malmsten and Lindman, 1992a; Alexandridis and Hatton, 1995). However, thistransition generally occurs at temperatures significantly higher than those relevant forpharmaceutical applications (Malmsten and Lindman,1992a; Alexandridis and Hatton,1995). The transition from a low viscous system to a very stiff one occurs very abruptly,frequently within a range of 5 °C, as seen in Figure 3a. Moreover, the onset temperaturecan be controlled, e.g., by the polymer concentration, molecular weight andcomposition as well as by the presence of cosolutes such as electrolytes, organicsubstances and surfactants.

The influence on the micellisation and gelation processes by cosolutes such aselectrolytes, surfactants and hydrophobic substances was investigated in papers I, II andIII and the main results will be discussed below.

Effect of total polymer concentration and polymer mixture compositionFormulations containing a fixed amount of polymers and 5 or 10 % w/w of activeingredients at pH 7.8 or 10 were investigated regarding the effect on cmt and gelationtemperature upon diluting the systems with water. It was found that increasing the watercontent causes an increase in both cmt and gelation temperature. However the effect onthe gelation temperature is much more dramatic than that on cmt. This might beexplained by the gelation temperature being determined by both the self-assembly andthe close-packing of micelles (Mortensen, 1992; Mortensen et al., 1992; Mortensen andPedersen, 1993; Mortensen, 1996), the latter being rather sensitive to total copolymerconcentration (for further details see paper IV).

Interestingly, the Lutrol F127/F68 ratio (at a constant total polymer concentration) hasa major effect on the properties of the system. Some of the combinations of the twopolymers, in presence of the active ingredients, were found to result in clear solutionswhich were thought to be o/w microemulsions or (mixed) micellar solutions dependingon the pH of the system. Since microemulsions are thermodynamically stable one-phasesystems the combination of Lutrol F68 and 127 and lidocaine and prilocaine werestudied more in depth in paper II and III. The effect of the ratio between Lutrol F127and F68 on self-assembly was further investigated by measuring cmt and the gelationtemperature for formulations containing a total polymer concentration of 21% w/w.Both cmt and the gelation temperature were found to decrease with increasing fraction

25

of Lutrol F127, although quantitatively the cmt decrease is much smaller than thedecrease in gel transition. Also, PCS and NMR indicate that the effect of the copolymerratio on the size of the micelles formed in dilute solution is quite minor. Consideringthis, the magnitude of the effect of the copolymer mixture composition on the gelationtemperature is rather surprising. A possible explanation might be the difference inaggregation number between the two polymers. Judging from literature values theaggregation number for Lutrol F127 seem to be higher than for Lutrol F68, althoughthe spread in the results reported is quite substantial and also depend strongly on thetemperature (Rassing and Attwood, 1983; Linse 1993a; Nagarajan, 1999). Nevertheless,the observed effects on the gelation temperature clearly show that this factor must beconsidered when choosing a suitable copolymer for developing a thermogellingformulation.

Effect of electrolytes and surfactantsThe micelle and gel formation is greatly influenced by the addition of electrolytes. Forexample, both the entire gel region and the cmt are shifted to lower temperatures uponthe addition of NaCl (a typical �salting out� cosolute) to an aqueous solution of LutrolF127 whereas the opposite is found for NaSCN (a typical �salting in� cosolute). Thisdemonstrates the strong connection between the micellisation and gelation processes.Qualitatively, the results obtained agree well with previous findings (Malmsten andLindman, 1992a; Hecht and Hoffman, 1995; Pandit et al., 2000).

The influence on the gelation behaviour of Lutrol F127 of a number of ionic andnonionic surfactants were studied in paper I. Most of the surfactants investigated onlyhad a minor effect on the gelation behaviour but for the anionic bile salt sodiumglycocholate, a substantial concentration-dependent increase of the gelation temperaturewas found.

Effect of active ingredients and pHDepending on the pH of the system, which determines the charge and solubility of theactive ingredients, lidocaine and prilocaine may have either a small or a major effect onthe gelation temperature of PEO-PPO-PEO block copolymer systems. In Figure 4a itcan be seen that at high pH, when the active components are uncharged and thussparingly soluble in water, the gelation point decreases with increasing amount oflidocaine and prilocaine. This is in agreement with previous findings on the effects ofhydrophobic compounds on the gel formation of PEO-PPO-PEO block copolymersystems (Malmsten and Lindman, 1992a). The cmt is also affected by the activeingredients causing the cmt to decrease as seen in Figure 4b. On the other hand, at pH 5,where 99.9% w/w of the active ingredients are in their ionised form, the effect on cmt isminor. The pH effect on the system is clearly elucidated in Figure 4c where it can beseen that cmt decreases drastically with increasing pH until pH 7.8 where 50% of theactive ingredients are in their hydrophobic state. Increasing the pH further seems to beof minor importance for the onset of micellisation. The results are expected since thepresence of a hydrophobic (pH>>pKa) component induces micellisation (lowers cmc)and causes a micellar growth (Lindman and Wennerström, 1980), whereas at pH<< pKalidocaine and prilocaine act largely as low concentration electrolytes.

26

(a)

(b)

(c)

Figure 4a) The effect of concentration of active ingredients on the gelation temperature at pH 5 (opencircles), 7 (open triangles), 8 (open squares) and 10 (open diamonds). The concentration of Lutrol F127and F68 was 15.5 and 4% w/w, respectively, for all formulations. b) cmt as a function of concentration ofactive ingredients in a formulation containing 15.5% w/w of Lutrol F127 and 5.5% w/w of Lutrol F68at pH 5 (open circles) and 7.8 (filled circles). c) cmt as a function of pH for a formulation containing 5%w/w of active ingredients, 15.5% w/w of Lutrol F127 and 5.5% w/w of Lutrol F68.

Concentration of active ingredients (% w/w)0 2 4 6 8 10

cmt (

°C)

0

5

10

15

pH5 6 7 8 9 10

cmt (

°C)

0

5

10

15

pKa

Clido+prilo/Ctot

0.0 0.1 0.2 0.3

T gel (

°C)

20

30

40

50

27

Interestingly, the amount solubilised drug molecules has little influence on the size ofthe micelles in the range investigated (see Figure 5). Considering the relatively low ormoderate concentration of the active ingredients in comparison to that of the polymerstaken together this might perhaps be expected.

Figure 5. Influence of different concentrations of active ingredients on the hydrodynamic radius of themicelles at 35 °C. NMR results (3% w/w solutions) (filled circles) and PCS results (3% w/w solutions)(open circles) are shown. Ctot corresponds to the total concentration in weight basis of polymers andactive components.

The distribution of active ingredients in unimers and micelles were investigated usinggel filtration. Block copolymer micelles have been found to be stable for more than anhour after dilution below cmc, making it possible to study the micellisation with thistechnique (Tuzar et al., 1974; Booth et al., 1978; Malmsten and Lindman, 1992a).Considering this, it should also be possible to study the solubilisation in unimers andmicelles, and this was therefore attempted in paper II. Although the preferentialsolubilisation in micelles over unimers could be demonstrated with this technique themethod was found to suffer from some difficulties, probably relating to material loss intubings etc., which resulted in an unexpected pH-dependence of the solubilisation.Therefore, also tail-end analysis was used to study the amount of active ingredientspresent in the different domains of the system (Suzuki and Sasaki, 1971; Ayako andFumio, 1978; Nyqvist-Mayer et al., 1985).

Cl+p / Ctot

0.0 0.1 0.2 0.3

Hyd

rody

nam

ic ra

dius

(nm

)

0

5

10

15

20

25

28

(a)

(b)

Figure 6a) The concentration of active ingredients as a function of elution volume after gel filtration forformulations containing 15.5% w/w Lutrol F127, 5.5% w/w Lutrol F68 and 2.5% w/w (opentriangles), 5% w/w (filled circles) or 10% w/w (open squares) of active ingredients at pH 7.8. b) Theconcentration of active ingredients as a function of elution volume after gel filtration for formulationscontaining 15.5% w/w Lutrol F127, 5.5% w/w Lutrol F68 and 5% w/w of active ingredients at pH 5(open circles), 7.8 (filled circles) and 9 (open squares). All formulations were diluted to 25% w/w withdistilled water prior to the gel filtration.

As seen in Figure 6a and b three plateaus were observed during the elution using thistechnique. A reasonable assumption is that these correspond to the concentration of theactive components in the added formulation, micelle phase, and unimer and/or freelysoluble phase respectively. To clarify whether or not this is the case, the particle size ofsamples in each of the three plateau regions were investigated by PCS after filtrationthrough a 0.22 µm filter in order to eliminate dust particles. Such measurements showedthat samples from the first and middle fraction contained aggregates with a radius (Z

Elution volume (ml)0 10 20 30 40 50 60

Con

cent

ratio

n of

act

ive

ingr

edie

nts (

% w

/w)

0.0

0.5

1.0

1.5

2.0

2.5

Elution volume (ml)0 10 20 30 40 50 60

Con

cent

ratio

n of

act

ive

ingr

edie

nts (

% w

/w)

0.0

0.5

1.0

1.5

29

average) in the range of 10-15 nm respectively (the polydispersity being below 0.3indicating a rather uniform distribution in particle size) which correlates well withfindings for the corresponding diluted systems using PCS and NMR. Also in analogywith our PCS and NMR findings very low scattering intensities were found for theexpected unimer and freely soluble phase. From Figure 6a and b it can be seen that theconcentration of active ingredients in the first plateau corresponds very well with theinitial concentration of the formulations diluted to 25% w/w (0.625, 1.25 and 2.5% w/wrespectively). The amount of active ingredients present in the micelle and unimer andfreely soluble phases were inferred to be 50-60 and 40-50% w/w, respectively, for allformulations at pH 7.8.

The solubilisation of lidocaine and prilocaine was found to be affected by the pH-dependent ionisation of these active ingredients. As seen in Figure 6b the relativeamount of lidocaine and prilocaine present in the micelle phase at pH 5 and 9 isapproximately 0 and 80% respectively, as compared to 50-60% at pH 7.8, while theamount present in the unimer and freely soluble phase is 100% at pH 5, 40-50% at pH7.8 and 20% at pH 9. Based on the pKa values of lidocaine and prilocaine, the amountpresent in their uncharged state is 0.2% w/w at pH 5, about 50% w/w at pH 7.8, and94% w/w at pH 9, which seems to correlate quite well with our findings of the amountof active ingredients solubilised at these pH values.

7.1.2 In vitro drug release

As seen in Figure 7a, the (mixed) micellar and microemulsion formulations (describedin paper II) have a fast release rate of the active ingredients. Specifically, it is higher(except at the highest pH) than that of the emulsion-based formulations (described inpaper I) and that of the EMLA cream bench-mark formulation. Most likely, this is dueto the smaller aggregates solubilising the active components. These results areencouraging since a high initial release rate is important considering the indication forthe formulation.

The influence of pH on the release rate of the active ingredients is shown in Figure 7b,and as seen the release rate increases with decreasing pH. This is expected since thelower the pH the more of the active ingredients are present in their ionised form andhence more is present in the aqueous part of the system. This has also been found forother substances. To give one example, the release of indomethacin (a weak acid) fromPEO-poly(β-benzyl L-aspartate) block copolymer micelles in aqueous solution has beenreported to show an increased release rate above pKa of indomethacin (La et al., 1996).Interestingly, there seem to be a difference in release rate between lidocaine andprilocaine at different pH values. As seen in Figure 7c the same amount of the twoactive ingredients is released at pH 5, where both substances are highly soluble in water.With increasing pH, however, the release of prilocaine is higher than that of lidocaine.This is not surprising considering that lidocaine is less soluble at these conditions thanprilocaine (Nyqvist-Mayer et al., 1986).

30

(a)

(b)

(c)

Figure 7a) Release curves for formulations containing active ingredients 5% w/w, Lutrol F68 5.5% w/wand Lutrol F127 15.5% w/w at pH 5 (open circles), 7.5 (open triangles), 7.8 (open squares) and 10(open diamonds) compared to EMLA cream 5% w/w (filled hexagons). b) Initial release rate, taken overthe first 4 h, as a function of pH. The arrow indicates the pKa value of lidocaine and prilocaine,respectively. c) The release rate ratio of the individual components divided by the total release rate oflidocaine (open circles) and prilocaine (filled circles) at different pH values. All formulations contained15.5% w/w of Lutrol F127 and 5.5% w/w of Lutrol F68. The lines are merely guides to the eye.

Time (hours)0 1 2 3 4

Con

cent

ratio

n (µ

mol

/cm

2 )

0

20

40

60

80

100

pH5 6 7 8 9 10

dc/d

t (µm

ol/c

m2 h)

0

10

20

30pKa

pH5 6 7 8 9 10

(Sin

gle

com

pone

nt /

Tota

l)

0.44

0.46

0.48

0.50

0.52

0.54

0.56

0.58

31

7.1.3 In vivo results

The onset and duration of the effect of Oraqix (5% w/w lidocaine and prilocaine,15.5% w/w Lutrol F127 and 5.5% w/w Lutrol F68, pH 7.8) have been measured in arandomised, parallel-group, open-labelled study using active treatment only (Friskopp etal., 2000). Thirty dental patients, recruited from a periodontal specialist clinic inSweden, were randomised to receive the formulation for 30 s, 2 min, or 5 min prior toperiodontal scaling. The formulation was applied to the periodontal pockets around aselected tooth using a blunt applicator.

The intensity of pain during the periodontal scaling procedure, as well as the overallpain and degree of discomfort were rated on a visual analogue scale (VAS) and on averbal rating scale (VRS) by the patient at the end of the treatment. The VAS scale wasa 100 mm, horizontal, blank ruler where the left and right end-points were marked �nopain� (0 mm) and �worst pain imaginable� (100 mm), respectively. After completingthe procedure, the presence of anaesthesia was checked by periodontal probing every 5min until sensation returned or until 30 min had passed since the end of the periodontalscaling. Furthermore, the patients were asked about the taste of the formulation and thedental hygienist about the overall ease of application of the formulation and overall easeof performing the periodontal scaling procedure.

The median VAS pain score was 7.5 mm in the 30 s group, 28.5 mm in the 2 min group,and 15.5 mm in the 5 min group. The periodontal scaling procedure did not have to beinterrupted due to pain for the 30 s or 2 min groups. However, for two patients in the 5min group, the procedure had to be interrupted due to pain. The mean duration ofanaesthesia measured were 18.1, 17.3, and 19.9 min in the 30 s, 2 min, and 5 mingroups respectively. The patients did not report any numbness of the tongue, lip orcheek. Furthermore, no adverse local reactions were found in the oral mucosa. Thedental hygienist regarded the formulation as easy to apply and not interfering with theperiodontal scaling procedure. In conclusion the formulation provides anaesthesia afteran application time of 30 s, with a mean duration of action of about 17 to 20 min.

7.2 Dilution-induced gelation of block copolymer systems

For the temperature-induced gelling systems the total amount of polymers are restrictedto approximately 20% w/w to achieve a thermal gelation above room temperature.Consequently, this limits the solubilising capacity for higher concentrations of sparinglysoluble drugs (Nagarajan et al., 1986; Hurter and Hatton, 1992). By having a systemundergoing phase transition as a consequence of dilution with water (or saliva) a higheramount of polymer and oil/drug could in principle be used (starting, e.g., from an L2phase). Thus, a larger amount of active ingredients can be incorporated into the systemalthough the magnitude of this effect will depend strongly on the system. Due to thehigher polymer concentration, one could also expect a slower release rate of the activeingredients, making the formulation suitable for indications where a longer duration isan advantage.

32

7.2.1 Phase behaviour of a polymer/water/oil system

In paper IV a system containing Lutrol F68, water and an oil-phase, consisting of aeutectic mixture of lidocaine and prilocaine (1:1) (Brodin et al., 1984) and AkolineMCM (84.5% caprylic glyceride, C8 and 15.5% capric glyceride, C10 containing 60%w/w mono- and 40% w/w diglyceride) in a ratio of 30:70% w/w, is described. Thepurpose of this study was to investigate the possibility of formulating a system that willturn from a low viscous state to a high viscous state upon dilution with water.

(a)

(b)

Figure 8a) A phase diagram of the Lutrol F68:H2O:LP:Akoline MCM system at 20 °C. b) A phasediagram of the Lutrol F68:H2O:LP:Akoline MCM system at 37 °C. The composition of the oil-phase,LP:Akoline MCM is 3:7. The abbreviations *, L2, H1, H2, Lα and I correspond to a low-viscous isotropic,reversed micellar, hexagonal, reversed hexagonal, lamellar and cubic phase, respectively. The samplesused for small-angle X-ray diffraction measurements have been marked with a circle. No two-phase areasare indicated between the I/H1/low-viscous isotropic phase and H2/ L2 phases.

LP:Akoline MCM (30:70% w/w )

0 10 20 30 40 50 60 70

Lutrol® F68

30

4050

6070

8090

100

H2O

010

2030

4050

6070 Not Investigated

*

IH2 L2

H1

2 φ

2 φ

LP:Akoline MCM (30:70% w/w)

0 10 20 30 40 50 60 70

Lutrol® F68

3040

5060

7080

90100

H2O

010

2030

4050

60

70 Not Investigated

L2Lα H2I

*

H1

2 φ

33

Rough ternary phase diagrams of the system, at 20 and 37 °C, are shown in Figure 8aand b for a limited concentration range, showing a relatively rich phase behaviour withthe presence of hexagonal and cubic phases that turn into reversed structures withdecreasing water content. As the concentration of oil increases in the system there is aprogression from a hexagonal to a reversed hexagonal and finally a reversed micellarphase. At a Lutrol F68:H2O:LP:Akoline MCM composition of about 59:23:5.4:12.6 anisotropic phase of low viscosity was found. After measuring the water conductivity andNMR self-diffusion of this phase as well as performing X-ray diffraction measurementsand investigating the samples in this region through crossed polarisers, the most likelystructure was inferred to be either an L3, a bicontinuous cubic or a bicontinuousmicroemulsion phase. In its dilution line there is a high viscosity hexagonal region andat further dilution a cubic phase appears. The low-viscous isotropic phase is somewhatextended at 37 °C, otherwise the phase diagram is relatively independent oftemperature, within this temperature range, for the system studied.

Photographs were taken through crossed polarisers of the low-viscous isotropic phasebefore and after addition of a drop of the phase to excess water (Figures 9a-c). Thesample starts to take up the water immediately forming an optically anisotropic phase(most likely the hexagonal phase based on the phase diagram) which make the sampleappear bright in the region where it was added. In presence of such a large amount ofwater, 30 minutes is enough to form most of this phase, whereas it takes about twohours until all of the active ingredients have been released and the whole sample hasdissolved.

(a) (b) (c)

Figure 9a) The low-viscous isotropic phase. (b) 30 minutes after the addition of the low-viscousisotropic phase to a test tube of water. (c) 2 hours after the addition of the low-viscous isotropic phase toa test tube of water. The samples were photographed under polarised light.

34

7.2.2 Gel formation

The rheology for this system as a function of water content is shown in Figure 10a for afixed polymer:oil ratio. As seen, already at a water addition of 7% w/w in excess of thatof the starting formulation, a highly elastic state is reached at both 20 and 37 °C.Although generally the viscosity is higher at 20 than 37 °C, the viscosity at 37 °C ismore than needed to give a formulation that will stay at the application site. Note thatthe phase transition of this system, when applied to the oral cavity, is induced by theuptake of water and not due to the temperature increase from 20 to 37 °C which isillustrated in Figure 10b.

(a)

(b)

Figure 10a) The elastic modulus, G´ as a function of water content for a formulation with an initialcomposition of Lutrol F68:H2O:LP:Akoline MCM; 64:23:3.9:9.1 at 20 °C (open circles) and 37 °C(filled circles). b) The elastic modulus, G´ as a function of temperature for a low viscous formulation(Lutrol F68:H2O:LP:Akoline MCM; 59:23:5.4:12.6) (open circles) and a high viscous formulation(Lutrol F68:H2O:LP:Akoline MCM; 50:35:4.5:10.5) (filled circles).

Temperature (°C)20 25 30 35 40

G´ (

kPa)

10-5

10-4

10-3

10-2

10-1

100

101

102

103

Water content (% w/w)25 30 35 40

G´ (

kPa)

0

5

10

15

20

35

Furthermore, the influence of the concentration of the active ingredients and pH on therheology was studied in the ranges of 5.4-12.6 % w/w and 6.6 to 9 respectively. Theresults showed that the system is quite robust regarding both concentration of activeingredients and pH.

7.2.3 In vitro drug release

The influence of the amount of active ingredients and pH on the release rate of theactive ingredients is shown in Figure 11a and b.

(a)

(b)

Figure 11a) The amount released of active ingredients over time for Lutrol F68:H2O:LP:AkolineMCM; 59:23:5.4:12.6 (open squares), Lutrol F68:H2O:LP:Akoline MCM;59:23:9:9 (open triangles),Lutrol F68:H2O:LP:Akoline MCM; 59:23:12.6:5.4 (open circles) and EMLA cream 5% w/w (opendiamonds). All formulations had a pH of 9. b) The amount released of active ingredients over time forLutrol F68:H2O:LP:Akoline MCM; 59:23:9:9 pH 9 (open squares), pH 7.7 (open triangles), pH 6.6(open circles), EMLA cream 5% w/w (open diamonds) and a microemulsion formulation at pH 7.8containing 5% w/w lidocaine and prilocaine, 5.5% w/w Lutrol F68 and 15.5% w/w Lutrol F127 (filledcircles).

Time (minutes)0 20 40 60 80 100 120

Con

cent

ratio

n (µ

mol

/cm

2 )

0

1

2

3

4

5

6

Time (minutes)0 20 40 60 80 100 120

Con

cent

ratio

n ( µ

mol

/cm

2 )

0

5

10

15

20

36

The phase transition was induced by adding a fixed amount of water to each diffusioncell prior to the addition of the low-viscous isotropic phase and each cell was inspectedat the experiments to ensure that a phase transition into a hardened system occurred. Ascan be expected the amount released increases with increasing concentration of theactive ingredients and decreasing pH. Considering the pKa values of lidocaine andprilocaine being 7.86 and 7.89 respectively (Nyqvist-Mayer et al., 1986), the lower thepH the more of the active ingredients are in their ionised form and hence the higher therelease rate. As with the thermoresponsive systems described above, the release rate ofthe system can be changed by simply changing the pH as seen in Figure 11b.

Comparing the release profiles of these formulations with a microemulsion formulationcontaining 5% w/w of the same active ingredients (see paper II) where instead thethickening is induced by a change in temperature, the release rate is considerably slowerand thus the presently investigated formulations may have a potential for applicationwhere a longer duration is needed.

7.3 Gelation of polymer-surfactant interacting systems

7.3.1 Clouding behaviour

Two nonionic cellulose ethers displaying a CP of about 34-36 °C at higherconcentrations (in the absence of cosolutes) were investigated in paper V. Their low CPmakes them interesting from a drug delivery perspective and the influence of theproperties of the polymer and addition of cosolutes are described below. The interactionof EHEC and HM-EHEC with anionic, cationic and nonionic surfactants wereinvestigated. In some cases nonionic polymers interact weakly with nonionic surfactantsbut generally no association occurs (Saito, 1987). Therefore, it was not surprising tofind that the addition of the nonionic surfactant Arlatone® 289 (PEG-54, 54 mol ofethylene oxide, hydrogenated castor oil), does not affect the CP of either EHEC or HM-EHEC, even at relatively high concentrations.

On the other hand, both anionic and cationic surfactants interact more or less effectivelywith nonionic cellulose ethers. For example, the anionic surfactant SDS (sodiumdodecyl sulphate) causes an onset of the increase in CP of EHEC and HM-EHEC atabout 2 mmolal, which closely matches the critical aggregation concentration (cac) forthese systems. Addition of glycocholic acid to the polymers, on the other hand, gives aless pronounced increase in CP with increasing surfactant concentration, and initialaddition of glycocholic acid is found to result in a significantly more extensivereduction of CP than the corresponding SDS addition. It is possible that the bile saltwith its low aggregation number compared to that of SDS effectively increases theoverall hydrophobicity of the polymer leading to a lowering of the polymer solubility inanalogy to, e.g., short-chain alcohols (Malmsten and Lindman, 1990).

Also cationic surfactants cause an increase on CP at high surfactant concentrations. Forexample, CTAB (hexadecyltrimethylammonium bromide) display a similar trend as

37

observed for the anionic surfactant SDS on the CP of the polymers. However, thelowering effect of CP at low surfactant concentrations is larger than observed with SDS.This might be related to a more extensive binding of CTAB than that of SDS to thepolymer at low polymer concentration as previously reported by Walderhaug et al.(Walderhaug et al., 1995). Overall the interaction between surfactants and polymerseems to be more extensive for the hydrophobically modified polymer compared to theunmodified one. This is expected, since HM-EHEC has a more blocky polymerstructure than EHEC with parts of the polymer more hydrophobic than the rest of thepolymer backbone. Therefore, the surfactant can self-associate, forming micelle-likeclusters at the hydrophobic chains, which therefore can act as nucleation sites(Thuresson et al., 1996; Thuresson and Lindman, 1997).

Effect of active ingredientsThe effect of increasing amount of lidocaine and prilocaine (1:1) on the CP wereinvestigated at pH 5, where 99.9% w/w of the active ingredients are in their ionisedform and thus expected to act as electrolytes, or possibly short-chain cationicsurfactants. There is a slight increase in CP for both polymers with increasing amount oflidocaine and prilocaine but quantitatively, the effect is minor. This can be expectedconsidering the charged nature of the active ingredients and the short hydrocarbonchains of these substances.

The effect of the active ingredients on both polymers was also investigated at pH 10where the local anaesthetic agents are in their uncharged form. However, since thisresulted in turbid solutions (no surfactant present) for both EHEC and HM-EHEC, CPcould not be measured. Nevertheless, it is interesting to note that upon the addition ofthe uncharged active ingredients to HM-EHEC opalescent homogeneous solutions werefound over a range of concentrations whereas the addition of the same amount of activeingredients to EHEC gives totally white systems, eventually leading to phase separation.These findings indicate an interaction between added hydrophobic agents and HM-EHEC, suggesting that this polymer through its ability to self-associate and as opposedto the unmodified EHEC is able to solubilise the active ingredients to some extent.

7.3.2 Rheological behaviour

As discussed previously, ionic surfactants bind co-operatively to the polymer backboneand polymer-bound micelles occur above the critical aggregation concentration (cac).These micelles may act as transient cross-links, thus effectively facilitating formation ofa gel structure. With increasing concentration of the surfactant a point is reached wherethe number of micelles are sufficient to saturate all hydrophobic domains of the polymerand the viscosity decreases again (Carlsson et al., 1990; Holmberg et al., 1992; Nyströmet al., 1995; Cabane et al., 1996; Holmberg and Sundelöf, 1996; Kjöniksen et al., 1998).As the temperature increases the solvency decreases causing an increased transientnetwork formation. The resulting increased viscosity is determined by a delicate balancebetween connectivity, swelling (Walderhaug et al., 1995; Cabane et al., 1996) andpossibly also temperature-dependent binding of the surfactant (Carlsson et al., 1990;

38

Zana et al., 1992; Kamenka et al., 1994). The magnitude of the effect depends on manyfactors such as the properties of the polymer and surfactant and presence of cosolutes(Carlsson et al., 1990; Walderhaug et al., 1995).

Effect of surfactantsContrary to the nonionic surfactant Arlatone 289 which does not affect the elasticmodulus (G´) or the viscous modulus (G´´) of EHEC or HM-EHEC, at least for asurfactant concentration of 60 mmolal, both anionic and cationic surfactants were foundto affect the rheological properties of EHEC and HM-EHEC. Analogous to the CPresults, there is an increase in G´ and G´´ for SDS (as seen in Figure 12) and CTABinitiated at about 2 mmolal which seem to be close to the cac.

For EHEC the magnitude of the maximum in the case of CTAB is smaller than for SDS,which might suggest a weaker and /or less extensive binding of CTAB to the polymer.The G´ and G´´ for EHEC and HM-EHEC have previously been reported to increasewith increasing amount of SDS up to a maximum value at approximately 5 mmolal forthe SDS/HM-EHEC system and 10 mmolal for the SDS/EHEC system, which is ingood agreement with our findings (Nyström et al., 1995). The G´max and G´´max valuescorrelate well with values reported previously in the literature for HM-EHEC but differfor EHEC where G´max exceeds G´´max as seen in Figure 12. A possible explanation forthis is the substitution degree, i.e., MSEO and DSethyl, and/or nonylphenol, of thepolymers being similar for HM-EHEC but different for EHEC as compared to the onesreported in the literature.

Figure 12. The effect of SDS on the elastic modulus (G´) (circles) and loss modulus (G´´) (triangles) onEHEC (filled symbols) and HM-EHEC (open symbols). The arrow indicates the cmc of SDS.

Concentration of SDS (mmolal)0 10 20 30 40

G´ a

nd G

´´ (P

a)

0

5

10

15

20

25

30

39

Interestingly, the thickening due to SDS is much more pronounced for EHEC than forHM-EHEC whereas the opposite is found for CTAB. This might be due to thedifference in binding between SDS and CTAB found by Walderhaug et al. (Walderhauget al., 1995) or differences in hydrophobic domains between the two polymers (Nyströmand Lindman, 1995). However, it is interesting to note that the hydrophobicallymodified polymer seems to be less sensitive towards the nature of the charge of thesurfactant as indicated by similar G´max and G´´max values for both the anionic andcationic surfactants investigated which is opposite to the behaviour of EHEC.

A number of other cationic surfactant, like CTAB also having antibacterial properties,and therefore being of interest, e.g., in a disinfecting local anaesthetic formulation forthe periodontal cavity, were also investigated yielding essentially similar results as forCTAB. This is expected from the structural similarity of these cationic surfactants aswell as the similar behaviour regarding CP effects on EHEC and HM-EHEC solutions.

Effect of active ingredients and ionic surfactantAddition of lidocaine and prilocaine in their uncharged form to a mixed SDS/EHEC orSDS/HM-EHEC system corresponding to either at or before the maximum thickeningthere is a decrease in both G´ and G´´ with increasing concentration of these substances.The same behaviour is also observed for the mixed CTAB/EHEC and CTAB/HM-EHEC systems. One would perhaps have expected solubilisation of the activecomponents to result in incorporation of further surfactants at the polymer backbone,and hence possibly an increase in G´ and G´´ instead of a decrease. A possibleexplanation for the observed behaviour could be that the added lidocaine and prilocainecauses a reduction in the cmc for the mixed system and hence a reduction in the freemonomer concentration (Lindman and Wennerström, 1980).

One of the most interesting features of mixed surfactant/EHEC systems is the increasein viscosity and G´ on increasing temperature. As can be seen in Figure 13a and b theincrease in G´ when increasing the temperature from 20 to 37 °C is quite substantial.The onset of increase in the elastic modulus at an SDS concentration of 3 mmolal islowered by the addition of the hydrophobic active ingredients (pH 9.9) possibly causedby surfactant self-association (in analogy to the cmc lowering due to hydrophobicsubstances (Lindman and Wennerström, 1980)).

A temperature-induced thickening is also observed with the cationic surfactants, also inthe presence of the active components (Figure 13b). The findings for myristoylcholinebromide are especially interesting for this kind of application since this surfactant isantibacterial and at the same time readily biodegradable (Ahlström et al., 1995).

40

(a)

(b)

Figure 13a) Elastic modulus (G´) as a function of temperature, of formulations containing EHEC 1%w/w and SDS 3 mmolal with no addition (open circles) and 0.5% w/w (filled circles) of active ingredients,pH 9.9. b) G´ as a function of temperature, of formulations containing EHEC 1% w/w andmyristoylcholine bromide 3 mmolal with no addition (open circles), and 0.5%w/w (filled circles) of activeingredients, pH 9.2. The formulations were not pH adjusted and no isotonic agent was added.

7.3.3 In vitro drug release

The release rate of the active ingredients from formulations containing anionic andcationic surfactants is displayed in Figure 14. As expected, the release rate increaseswith increasing amount of active ingredients for formulations containing the sameamount of SDS. The release rate seems to decrease somewhat with increasing amount ofSDS for formulations containing the same amount of active ingredients, butquantitatively the effects of the surfactant concentration at a fixed concentration ofactive ingredients are limited. Furthermore, it is interesting to note that the release ratefor a formulation containing myristoylcholine bromide seems to be somewhat higher

Temperature (°C)25 30 35 40

G´ (

Pa)

0

10

20

30

Temperature (°C)25 30 35 40

G´ (

Pa)

0

10

20

30

41

than formulations containing SDS when comparing formulations containing the sameamount of active ingredients. The origin of this effect is currently unknown, anddeserves further investigation.

Figure 14. Release curves for formulations containing 3 mmolal SDS and 0.5% w/w active ingredientspH 9.9 (filled squares), 10 mmolal SDS and 0.5% w/w active ingredients pH 10.2 (open triangles), 10mmolal SDS and 5% w/w active ingredients pH 11.1 (open squares), 3 mmolal myristoylcholine bromideand 0.5% w/w active ingredients pH 9.2 (open diamonds), 10 mmolal CTAB and 0.5% w/w activeingredients pH 9.8 (open inverted triangles) compared to EMLA® cream 5% w/w (open circles) and anEMLA cream formulation with 0.5% w/w of active ingredients (filled circles). All formulations except thereference contained 1% w/w EHEC. The formulations were not pH adjusted and no isotonic agent wasadded.

Most of the formulations showed a comparable release rate of lidocaine and prilocaineas to the EMLA cream reference formulations. In particular, the results shown in Figure13 for the formulation containing 3 mmolal myristoylcholine bromide is promising bothfrom the point of view of its antibacterial effect and biodegradability as well as its ratherrapid release of the active ingredients. Considering the intended use of the formulationthere is an additional benefit from having an antibacterial excipient present in theformulation since this could be expected to help preventing the reoccurence of plaqueafter this has been removed by the scaling procedure. Furthermore, the use ofmyristoylcholine bromide could be expected to be beneficial from a toxicological pointof view since the small amounts of cationic surfactant present in the system ishydrolysed by butyrylcholine esterase present in mucosal membranes into degradationproducts that are significantly less toxic than the original substance (Ahlström et al.,1995).

Time (minutes)0 10 20 30 40 50 60

Con

cent

ratio

n (µ

mol

/cm

2 )

0

1

2

3

42

8 Conclusions

Systems displaying temperature-induced gelation were achieved by combining theactive ingredients lidocaine and prilocaine with the nonionic block copolymers Lutrol

F127 and Lutrol F68. For the diluted formulations over a certain concentration andcomposition range, this resulted in clear stable micellar solutions with a micellardiameter of 20-30 nm. The cmt and gelation temperature of the systems were found tobe interconnected and influenced by the total polymer concentration and the polymermixture composition, as well as the presence of cosolutes such as electrolytes andhydrophobic substances, the latter as found particularly for the eutectic mixture oflidocaine and prilocaine. Both the cmt and the gelation temperature decrease withincreasing pH of the system, i.e., at reduced solubility of the active ingredients.

The active ingredients are preferentially solubilised in micelles in these microemulsionsand (mixed) micellar formulations and depending on the pH of the system,approximately 0% w/w at pH 5, 50-60% at pH 7.8 and 80% at pH 9 of the activeingredients are present in the micellar phase. By choosing a pH in the area of 7.8 it ispossible to have a system with a gelation point between room and body temperaturedisplaying a high initial release rate. Such a formulation has been tested on dentalpatients during a periodontal scaling procedure and found to fulfil the requirements ofhaving a fast onset time, a sufficient anaesthetic effect, little or no adverse effects at thesame time as being easy to administer and handle.

A dilution-induced thickening system was found when combining Lutrol F68, water, aeutectic mixture of lidocaine and prilocaine and Akoline MCM. The phase behaviour ofthe system is interesting with the presence of a low-viscous isotropic phase that turnsinto a hexagonal phase as the water content increases. The viscosity and elastic modulusof the former is low whereas that of the hexagonal phase formed on dilution with wateris high, the difference between the phases being 1000- to 10 000-fold. The behaviour ofthe system is promising from a pharmaceutical point of view, since it facilitatesinjection of the low-viscous isotropic phase into, e.g., the periodontal pocket where thepresence of saliva will cause a transition into a rigid hexagonal phase (followed by acubic phase), thereby contributing to the formulation staying at the application site. Theslower release rate compared to the microemulsion or (mixed) micellar formulationdescribed above might also make the formulation suitable for indications where a longerduration is required. Furthermore, since these systems contain more polymer they areable to solubilise higher amounts of hydrophobic agents.

A temperature-induced gelling system could also be achieved by adding lidocaine andprilocaine to an EHEC/SDS, EHEC/CTAB, or EHEC/myristoylcholine bromide systemat or just below the surfactant concentration found to give a maximum viscosityincrease at room temperature. The results show that at least small amounts of activeingredients can be incorporated into the systems without affecting the gelationbehaviour in a detrimental way. The stability and drug release also seem to indicate apossibility of formulating a local anaesthetic drug delivery system suitable foradministration into the periodontal pocket. A disadvantage of using ionic surfactants isthe toxicological risks involved. However, it has been shown that the release of SDS

43

from gelling EHEC is quite slow with approximately 25% of the surfactant beingreleased within 1 hour (Lindell, 1996). Also the surfactant concentration needed to formgelling systems is usually rather low. Therefore, the presently investigated systems areinteresting as drug delivery systems for at least some applications where a temperatureinduced viscosity increase is an advantage. In particular, the myristoylcholine bromidesystem may be interesting in this respect considering its antibacterial properties andbiodegradability.

Table 2. A summary of the different formulations.

Investigatedsystems

Advantages Disadvantages

PEO-PPO-PEOO/W Emulsions

• Easy to apply• Stays at the application site• Water miscible → can be washed away

• Additional surfactant/copolymerneeded for stabilisation → toxicityaspects

• White appearance → hard to see theinstruments

• Kinetically unstable → willeventually phase separate

PEO-PPO-PEOMicroemulsions/mixed micellarsystems

• Easy to apply• Stays at the application site• Water miscible → can be washed away• Clear appearance → easy to see the

instruments• Small particle size → fast release of

active ingrediens• High solubilising capacity• Thermodynamically stable → excellent

long-term stability

• Requires high amount ofsurfactant/copolymer

EHEC/surfactantsystems

• Easy to apply• Stays at the application site• Water miscible → can be washed away• Biodegradable• Surfactants with antibacterial properties

can be used• Protection against hydrolysis of drug

(less water in the micellar core)• From natural sources

• Ionic surfactant needed to inducethickening → toxicity aspects

• Limited solubilising capacity →limited amount of active ingredient

• More viscous before and less viscousafter gelation than PEO-PPO-PEOblock copolymer systems

Low-viscousisotropic phase→ hexagonalphase

• Easy to apply• Stays at the application site• Water miscible → can be washed away• High solubilising capacity• Good long-term stability

• Small area of the low-viscousisotropic phase

• Slower release of active componentscompared to microemulsion/mixedmicellar formulations (can be anadvantage depending on indication)

44

9 Outline for Future Work

The temperature-induced thickening system consisting of the active ingredientslidocaine and prilocaine and the PEO-PPO-PEO block copolymers Lutrol F127 andLutrol F68 was investigated quite extensively in this work regarding micellisation andgelation behaviour. However, studies on the influence of polymer mixture compositionon these processes show that the effect on gelation temperature is much morepronounced than that on cmt and the micellar size, which is quite surprising. Therefore,more information is needed regarding mixed micelle formation in solutions containingLutrol F68 and F127 to fully understand these processes. A way to learn more wouldbe to study the influence of polymer mixture composition on the micellar structure andaggregation number with, e.g., small angle neutron scattering (SANS) and time-resolved fluorescence spectroscopy.

For the dilution-induced thickening systems consisting of Lutrol F68, water, theeutectic mixture of lidocaine and prilocaine, and Akoline MCM it would be interestingto investigate the influence of the active ingredients and pH on the phase behaviourmore thoroughly. Furthermore, it would be interesting to perform some in vivo tests tolearn further if this system can be used for controlled drug delivery. Moreover, othercopolymer/oil systems displaying such behaviour would be valuable to identify andcharacterise.

For the temperature-induced thickening systems of nonionic cellulose ether, ionicsurfactant and active ingredients a surprising feature was found when comparing therheological behaviour of EHEC and HM-EHEC in presence of SDS and CTAB. Thus,addition of SDS gave a more pronounced viscosity increase for EHEC than for HM-EHEC, whereas the opposite was found for CTAB. In order to investigate the origin ofthis behaviour, further experiments with identical backbone polymers should beperformed.

Finally, for all systems mentioned above it would be interesting to test their suitabilityfor other administration routes, e.g., nasal, ocular, rectal and parenteral, and also thepossibility of solubilising other types of active ingredients and substances.

45

10 Acknowledgements

I would like to express my warmest thanks to:

Martin Malmsten for being encouraging, reliable, demanding and a personification ofthe word �efficiency�. In other words the best supervisor I could ever have wished for.

Arne Brodin for giving me the opportunity to do my Ph. D. studies at AstraZeneca andfor supporting my work especially by giving me the chance to do research at the OhioState University for a year.

Christer Nyström for valuable discussions and academic guidance.

Sylvan Frank for letting me use his labs and equipment and making my stay at OhioState University a true pleasure.

Anna Shenderova, Chao Ye, Jessica Miller and Karen Lawler for all the good times weshared during my stay in the USA and for keeping your promises of staying in touch.

My co-authors Peter Holmqvist and Katy Welin-Berger for good collaboration.

Karin Schillén, Anna Karin Morén, Mats Blom, Magnus Nydén, Istvan Furo and LinaRagnarsson for experimental help.

Coworkers and friends at AstraZeneca for actually making it fun to go to work.

Mats Berg for all your help with experiments and for always having a smart practicalsolution to all my questions in the laboratory.

Katy Welin-Berger, my soul-mate and laughing partner, for always being there for me,both in good and bad times.

All my wonderful friends who I plan to see a lot more of in the future.

My mother and grandmother for teaching me about life, for making me want to be abetter person and for loving me unconditionally no matter what.

Anders, my addiction in life, for always making me happy and for being so patient.Without your love, encouragement and constant belief in me this thesis would neverhave been written.

46

11 References

Abd Elbary, A., Foda, N., Tayel, S., 1991. Bioavailability of Glibencamide from NasalDelivery Systems. Pharm. Ind., 53, 1151-1155.

Ahlström, B., Chelminska-Bertilsson, M., Thompson, R.A. and Edebo, L., 1995. Long-Chain Alkanoylcholines, a New Category of Soft Antimicrobial Agents That areEnzymatically Degradable. Antimicrob. Agents Chemother., 39, 50-55.

Alexandridis, P., Holzwarth, J.F., Hatton, T.A., 1994. Micellisation of Poly(ethyleneoxide)-Poly(propylene oxide)-Poly(ethylene oxide) Triblock Copolymers inAqueous Solutions: Thermodynamics of Copolymer Association. Macromolecules,27, 2414-2425.

Alexandridis, P., Hatton, T.A., 1995. Poly(ethylene oxide) - Poly(propylene oxide) -Poly(ethylene oxide) Block Copolymer Surfactants in Aqueous Solutions and atInterfaces: Thermodynamics, Structure, Dynamics, and Modeling. Colloids Surf. A, 96, 1-46.

Alexandridis, P., Olsson, U., Lindman, B., 1995. Self-Assembly of Amphiphilic BlockCopolymers: The (EO)13(PO)30(EO)13-Water-p-Xylene System. Macromolecules,28, 7700-7710.

Alexandridis, P., Zhou, D., Khan, A., 1996a. Lyotropic Liquid Crystallinity inAmphiphilic Block Copolymers: Temperature Effects on Phase Behavior andStructure for Poly(ethylene oxide)-b-poly(propylene oxide)-b-poly(ethylene oxide)Copolymers of Different Composition. Langmuir, 12, 2690-2700.

Alexandridis, P., Olsson, U., Lindman, B., 1996b. Phase Behaviour of AmphiphilicBlock Copolymers in Water-Oil Mixtures: The Pluronic 25R4-Water-p-XyleneSystem. J. Phys. Chem., 100, 280-288.

Alexandridis, P., Holzwarth, J.F., 1997. Differential Scanning Calorimetry Investigationof the Effect of Salts on Aqueous Solution Properties of an Amphiphilic BlockCopolymer (Poloxamer). Langmuir, 13, 6074-6082.

Alexandridis , P., Olsson, U., Lindman, B., 1998. A Record Nine Different Phases (FourCubic, Two Hexagonal, and One Lamellar Lyotropic Liquid Crystalline and TwoMicellar Solutions) in a Ternary Isothermal System of an Amphiphilic BlockCopolymer and Selective Solvents (Water and Oil). Langmuir, 14, 2627-2638.

Al-Saden, A.A., Whateley, T.L., Florence, A.T., 1982. Poloxamer Association inAqueous Solution. J. Colloid Interface Sci., 90, 303-309.

Attwood, D., 1994. Microemulsions. In Colloidal Drug Delivery Systems, Kreuter, J.,Ed., Marcel Dekker, New York, vol. 66, pp 31-71.

Ayako, G., Fumio, E., 1978. The Distribution of Alkyl Paraben in Aqueous SodiumLauryl Sulfate Micellar Solutions. J. Colloid Interface Sci., 66, 26-32.

Becher, P., 1985. Encyclopedia of Emulsion Technology - Applications, Marcel Dekker,Inc., New York.

Bergh, M., Magnusson, K., Nilsson, J.L.G., Karlberg, A.T., 1998. Formation ofFormaldehyde and Peroxides by Air Oxidation of High Purity PolyoxyethyleneSurfactants. Contact Dermatitis, 39, 14-20.

Booth, C., Naylor, T.D., Price, C., Rajab, N.S., Stubbersfield, R.B., 1978. Investigationof the Size Distribution of Nonionic Micelles Formed from a Polystyrene-Polyisoprene Block Copolymer in N,N-Dimethylacetamide. J. Chem. Soc. FaradayTrans., 1, 74, 2352-2362.

47

Brodin, A., Nyqvist-Mayer, A., Wadsten, T., Forslund, B., Broberg, F., 1984. PhaseDiagram and Aqueous Solubility of the Lidocaine - Prilocaine Binary System. J.Pharm. Sci., 73, 481-484.

Brown, W., Schillen, K., Almgren, M., Hvidt, S., Bahadur, P., 1991. Micelle and GelFormation in a Poly(ethylene oxide)/Poly(propylene oxide)/Poly(ethylene oxide)Triblock Copolymer in Water Solution. Dynamic and Static Light Scattering andOscillatory Shear Measurements. J. Phys. Chem., 95, 1850-1858.

Cabane, B., Lindell, K., Engström, S., Lindman, B., 1996. Microphase Separation inPolymer + Surfactant Systems. Macromolecules, 29, 3188-3197.

Carlsson, A., Karlström, G., Lindman, B., 1986. Synergistic Surfactant-ElectrolyteEffect in Polymer Solutions. Langmuir, 2, 536-537.

Carlsson, A., Karlström, G., Lindman, B., 1989. Characterisation of the Interactionbetween a Nonionic Polymer and a Cationic Surfactant by Fourier Transform NMRSelf-Diffusion Technique. J. Phys. Chem., 93, 3677-3683.

Carlsson, A., Karlström, G. and Lindman, B., 1990. Thermal Gelation of NonionicCellulose Ethers and Ionic Surfactants in Water. Colloids Surf., 47, 147-165.

Chen-Chow, P., Frank, S.G., 1981. In vitro release of lidocaine from Pluronic F-127gels. Int. J. Pharm., 8, 89-100.

Collett, J.H., Tobin, E.A., 1979. Relationships between poloxamer structure and thesolubilization of some para-substituted acetanilides. J. Pharm. Pharmacol., 31,174-177.

Constantinides, P.P., 1995. Lipid Microemulsions for Improving Drug Dissolution andOral Absorption: Physical and Biopharmaceutical Aspects. Pharm. Res., 12, 1561-1572.

Davidorf, F.H., Chambers, R.B., Kwon, O.W, Doyle, W., Gresak, P., Frank, S.G., 1990.Ocular Toxicity of Vitreal Pluronic Polyol F-127. Retina, 10, 297-300.

Danielsson, I., Lindman, B., 1981. The Definition of Microemulsion. Colloids Surf., 3,391-392.

Donaldson, D., Meechan, J.G., 1995. A comparison of the effects of EMLA cream andtopical 5% lidocaine on discomfort during gingival probing. Anesthesia Progress,42, 7-10.

Edsman, K., Carlfors, J., Petersson, R., 1998. Rheological Evaluation of Poloxamer asan In Situ Gel for Ophtalmic Use. Eur. J. Pharm. Sci., 6, 105-112.

Engström, S., Engström, L., 1992. Phase behaviour of the lidocaine-monoolein-watersystem. Int. J. Pharm., 79, 113-122.

Engström, S., Lindahl, L., Wallin, R., Engblom, J., 1992. A study of polar lipid drugcarrier systems undergoing a thermoreversible lamellar-to-cubic phase transition.Int. J. Pharm., 86, 137-145.

Esposito, E., Carotta, V., Scabbia, A., Trombelli, L., D�Antona, P., Menegatti, E.,Nastruzzi, C., 1996. Comparative Analysis of Tetracycline-Containing Dental Gels:Poloxamer- and Monoglyceride-Based Formulations. Int. J. Pharm., 142,9-23.

Ferry, J.D., 1980. Viscoelastic Properties of Polymers, John Wiley & Sons. Inc., NewYork.

Florence, A.T., Attwood, D., 1998. Physicochemical Principles of Pharmacy, ThirdEdition, Macmillian Press LTD, London.

48

Fontell, K., 1992. Some Aspects on the Cubic Phases in Surfactant and Surfactant-likeLipid Systems. Adv. Colloid Interface Sci., 41, 127-47.

Friskopp, J., Nilsson, M., Isacsson, G., 2000. The Anesthetic Onset and Duration of aNew Lidocaine/prilocaine Gel Intra-pocket Anesthetic (Oraqix®) for PeriodontalScaling/root Planing. J. Clin. Periodontol., in press.

Gilbert, J.G., Hadgraft, J., Bye, A., Brookes, L.G., 1986. Drug release from Pluronic F-127 gels. Int. J. Pharm., 32, 223-228.

Gloxhuber, C., 1980. Anionic Surfactants � Biochemistry, Toxicology, Dermatology,Surfactants Science Series, Vol. 10, Marcel Dekker, Inc., New York.

Goldstein, R.E., 1984. On the Theory of Lower Critical Solution Points in Hydrogen-bonded Mixtures. J. Phys. Chem., 80, 5340-5341.

Hecht, E., Hoffman, H., 1994. Interaction of ABA Block Copolymers with IonicSurfactants in Aqueous Solution. Langmuir, 10, 86-91.

Hecht, E., Hoffman, H., 1995. Kinetic and Calorimetric Investigations on MicelleFormation of Block Copolymers of the Poloxamer Type. Colloids Surf. A, 96, 181-197.

Hecht, E., Mortensen, K., Hoffman, H., 1995. L3 Phase in a Binary BlockCopolymer/Water System. Macromolecules, 28, 5465-5476.

Holmberg, C., Nilsson, S., Singh, S.K., Sundelöf, L-O., 1992. Hydrodynamic andThermodynamic Aspects of the SDS-EHEC-Water System. J. Phys. Chem., 96,871-876.

Holmberg, C., Sundelöf, L-O., 1996. Temperature Dependence of HydrodynamicProperties and Surfactant-polymer Interaction in Solution. The EHEC/SDS/WaterSystem. Langmuir, 12, 883-889.

Holst, A., Evers, H., 1985. Experimental studies of new topical anaesthetics on the oralmucosa. Swedish Dental Journal, 9, 185-191.

Hurter, P.N., Hatton, A.T., 1992. Solubilization of polycyclic aromatic hydrocarbons bypoly(ethylene oxide-propylene oxide) block copolymer micelles: effects of polymerstructure. Langmuir, 8, 1291-1299.

Iliopoulos, I., Wang, T.K. and Audebert, R., 1991. Viscometric Evidence of Interactionsbetween Hydrophobically Modified Poly(sodium acrylate) and Sodium DodecylSulfate. Langmuir, 7, 617-619.

International Standard, ISO 13321, 1996. Particle Size Analysis � Photon CorrelationSpectroscopy, First edition.

Israelachvili, J., 1994. The Science and Applications of Emulsions - An Overview.Colloid Surf. A, 91, 1-8.

Johansson, H.O., Karlström, G., Tjerneld, F., 1993. Experimental and Theoretical Studyof Phase Separation in Aqueous Solutions of Clouding Polymers and CarboxylicAcids. Macromolecules, 26, 4478-4483.

Jullander, I., 1957. Water-soluble Ethylhydroxyethylcellulose. Ind. Eng. Chem., 49,364-368.

Jönsson, B., Lindman, B., Holmberg, K. and Kronberg, B., 1998. Surfactants andPolymers in Aqueous Solution, John Wiley & Sons. Inc., Chichester.

Kabanov, A.V., Batrakova, E.V., Melik-Nubarov, N.S., Fedoseev, N.A., Dorodnich,T.Y., Alakhov, V.Y., Chekonin, V.P., Nazarova, I.R., Kabanov, V.A., 1992. A NewClass of Drug Carriers: Micelles of Poly (oxyethylene)-poly (oxypropylene) Block

49

Copolymers as Microcontainers for Drug Targeting from Blood in Brain. J.Controlled Release, 22, 141-158.

Kamenka, N., Burgaud, I., Zana, R., Lindman, B., 1994. Electrical Conductivity, Self-diffusion, and Fluorescence Probe Investigations of the Interaction between SodiumDodecyl Sulfate and Ethyl(hyroxyethyl)cellulose. J. Phys. Chem., 98, 6785-6789.

Karlström, G.J., 1985. A New Model for Upper and Lower Critical SolutionTemperatures in Poly(ethylene oxide) Solutions. J. Phys. Chem., 89, 4962-4964.

Karlström, G., Carlsson, A., Lindman, B., 1990. Phase Diagrams of Nonionic Polymer-water Systems. Experimental and Theoretical Studies of the Effects of Surfactantsand Other Cosolutes. J. Phys., Chem. 94, 5005-5015.

Kjellander, R., Florin, E., 1981. Water Structure and Changes in Thermal Stability ofthe System Poly(ethylene oxide)-water. J. Chem. Soc. Faraday Trans., 77, 2053-2077.

Kjellander, R., 1982. Phase Separation of Non-ionic Surfactant Solutions. J. Chem. Soc.Faraday Trans., 78, 2025-2042.

Kjöniksen, A.L., Nyström, B., Lindman, B., 1998. Dynamic Viscoelasticity of Gellingand Nongellig Aqueous Mixtures of Ethyl(hydroxyethyl)cellulose and an IonicSurfactant. Macromolecules, 31, 1852-1858.

Kwon, G.S., Kataoka, K., 1995. Block Copolymer Micelles as Long Circulating DrugVehicles. Adv. Drug Del. Rev., 16, 295-309.

La, S.B., Okano, T., Kataoka, K., 1996. Preparation and Characterization of theMicelle-Forming Polymeric Drug Indomethacin - Incorporated Poly(ethyleneoxide)-Poly(beta-benzyl L-aspartate) Block Copolymer Micelles. J. Pharm. Sci.,85, 85-90.

Lang, J.C., Morgan, R.D., 1980. Nonionic Surfactant Mixtures. I. Phase Equilibria inC10E4-H2O and Closed-loop Coexistence. J. Chem. Phys., 73, 5849-5861.

Larsson, K., 1994. Lipids-Molecular Organization, Physical Functions and TechnicalApplications, vol. III, The Oily Press Ltd, West Ferry, Dundee, Scotland.

Lawrence, M.J., 1994. Surfactant Systems: Microemulsions and Vesicles as Vehiclesfor Drug Delivery. Eur. J. Drug Metab. Pharmacokinet., 19, 257-269.

Lin, S.Y., Kawashima, Y., 1985. Kinetic studies on the stability of indomethacin inalkaline aqueous solution containing poly(oxyethylene)poly(oxypropylene) surface-active block copolymers. Pharm. Acta Helv., 60, 345-350.

Lindell, K., 1996. An Investigation of Thermogelling Aqueous Systems ofEthyl(hydroxyethyl) Cellulose and Ionic Surfactants, Ph.D. Thesis, University ofLund, Sweden.

Lindell, K., Engström, S., 1993. In Vitro Release of Timolol Maleate from an In SituGelling Polymer System. Int. J. Pharm., 95, 219-228.

Lindman, B. and Wennerström, H., 1980. In Topics in current Chemistry; Boschke, F.,Ed.; Vol. 87, p. 1-83, Springer Verlag, Berlin.

Lindman, B., Söderman, O., Wennerström, H. 1987. In NMR Sudies of SurfactantSystems; In Surfactant Solutions. New Methods of Investigation; Zana, R., Ed.;Marcel Dekker, New York, ch.6.

Lindman, B., Carlsson, A., Karlström, G., Malmsten, M., 1990. Nonionic Polymers andSurfactants � Some Anomalies in Temperature Dependence and in Interactions withIonic Surfactants. Adv. Colloid Interface Sci., 32, 183-203.

50

Linse, P., 1993a. Micellization of Poly(ethylene oxide)-Poly(propylene oxide) BlockCopolymers in Aqueous Solution. Macromolecules, 26, 4437-4449.

Linse, P., 1993b. Phase-Behavior of Poly(ethylene oxide)-Poly(propylene oxide) BlockCopolymers in Aqueous Solution. J. Phys. Chem., 97, 13896-13902.

Linse, P., Malmsten, M., 1992. Temperature - Dependent Micellization in AqueousBlock Copolymer Solutions. Macromolecules, 25, 5434-5439.

Luzzati, V., 1968. X-ray diffraction studies of lipid-water systems. BiologicalMembranes, 1, 71.

Magnusson, G., Olsson, T., Nyberg, J.A., 1986. Toxicity of Pluronic F68. Toxicol.Letts., 30, 203-207.

Malmsten, M., Lindman, B., 1990. Ellipsometry Studies of the Adsorption of CelluloseEthers. Langmuir, 6, 357-364.

Malmsten, M., Lindman, B., 1992a. Self-Assembly in Aqueous Block CopolymerSolutions. Macromolecules, 25, 5440-5445.

Malmsten, M., Lindman, B., 1992b. Water Self-Diffusion in Aqueous Block CopolymerSolutions. Macromolecules, 25, 5446-5450.

Malmsten, M., Lindman, B.,1993. Effects of Homopolymers on the Gel Formation inAqueous Block Copolymer Solutions. Macromolecules, 26, 1282-1286.

Malmsten, M., 1998. In Microemulsions: Fundamentals and Applied Aspects, Kumar,P., Mittal, K.L., Eds., Marcel Dekker, Inc., New York.

McNaughton, J.L., Mortimer, C.T., 1975. Differential Scanning Calorimetry, PhysicalChemistry Series 2.

Miller, S.C., Donovan, M.D., 1982. Effect of Poloxamer 407 Gel on the Miotic Activityof Pilocarpine Nitrate in Rabbits. Int. J. Pharm., 12, 147-152.

Milgrom, P., Coldwell, S.E., Getz, T., Weinstein, P., Ramsay, D.S., 1997. Fourdimensions of fear of dental injections. J. Am. Dent. Assoc., 128, 756-766.

Miyazaki, S., Yokouchi, C., Nakamura T., Hashiguchi, N., Hou, W.M., Takada M.,1986. Pluronic F127 Gels as a Novel Vehicle for Rectal Administration ofIndomethacin. Chem. Pharm. Bull., 34, 1801-1808.

Moorhouse, R., Colegrove, G.T., Sandford, P.A., Baird, J.K., Kang, K.S., 1981. InBrandt, D.A., Ed., Solution Properties of Polysaccharides, Washington, DC, pp.111-124.

Mortensen, K., 1992. Phase Behaviour of Poly(ethylene oxide)-Poly(propylene oxide)-Poly(ethylene oxide) Triblock-Copolymer Dissolved in Water. Europhys. Lett., 19,599-604.

Mortensen, K., Brown, W., Nordén, B., 1992. Inverse Melting Transition and Evidenceof Three-Dimensional Cubatic Structure in a Block-Copolymer Micellar System.Phys. Rev. Lett., 68, 2340-2343.

Mortensen, K., Pedersen J.S., 1993. Structural Study on the Micelle Formation ofPoly(ethylene oxide)-Poly(propylene oxide)-Poly(ethylene oxide) TriblockCopolymer in Aqueous Solution. Macromolecules, 26, 805-812.

Mortensen, K., 1996. Structural Study of Aqueous Solutions of PEO-PPO-PEOTriblock Copolymers, their Micellar Aggregates and Mesophases; a Small-AngleNeutron Scattering Study. J. Phys. Condens. Matter, 8, 103-124.

Nagarajan, R., Barry, M., Ruckenstein, E., 1986. Unusual selectivity in solubilization byblock copolymer micelles. Langmuir, 2, 210-215.

51

Nagarajan, R., 1999. Solubilization of Hydrocarbons and Resulting Aggregate ShapeTransitions in Aqueous Solutions of Pluronic (PEO-PPO-PEO) BlockCopolymers. Colloids Surf. B, 16, 55-72.

Norling, T., Lading, P., Engström, S., Larsson, K., Krog, N., Nissen, S.S., 1992.Formulation of a drug delivery system based on a mixture of monoglycerides andtriglycerides for use in the treatment of periodontal disease. J. Clin. Periodontol.,19, 687-692.

Nyqvist-Mayer, A.A., Brodin, A.F., Frank, S.G., 1985. Phase Distribution Studies on anOil-Water Emulsion Based on a Eutectic Mixture of Lidocaine and Prilocaine as theDispersed Phase. J. Pharm. Sci., 74, 1192-1195.

Nyqvist-Mayer, A.A., Brodin, A.F., Frank, S.G., 1986. Drug Release Studies on an Oil-Water Emulsion Based on a Eutectic Mixture of Lidocaine and Prilocaine as theDispersed Phase. J. Pharm. Sci., 75, 365-373.

Nyström, B., Thuresson, K., Lindman, B., 1995. Rheological and Dynamic Light-scattering Studies on Aqueous Solutions of a Hydrophobically Modified NonionicCellulose Ether and its Unmodified Analogue. Langmuir, 11, 1994-2002.

Nyström, B., Lindman, B., 1995. Dynamic and Viscoelastic Properties During theThermal Gelation Process of a Nonionic Cellulose Ether Dissolved in Water- In thePresence of Ionic Surfactant. Macromolecules, 28, 967-974.

Pandit, N., Trygstad, T., Croy, S., Bohorquez, M., Kock, C., 2000. Effect of Salts on theMicellization, Clouding, and Solubilization Behavior of Pluronic F127 Solutions. J.Coll. Interface Sci., 222, 213-220.

Patterson, I., Chowdhry, B.Z., Leharne, S., 1996. Evaluation of Surfactant MonomerConcentrations in Equilibrium with Micelles and their Temperature Dependence forOxyethylene/Oxypropylene/Oxyethylene Triblock Copolymers using DSC.Colloids Surf. A, 111, 213-222.

Paul, B.K., Moulik, S.P., 1997. Microemulsions an Overview. J. Dispersion Sci.Technol., 18, 301-367.

Rassing, J., Attwood, D., 1983. Ultrasonic Velocity and Light-Scattering Studies on thePolyoxyethylene-Polyoxypropylene Copolymer Pluronic F127 in AqueousSolution. Int. J. Pharm., 13, 47-55.

Rozier, A., Mazuel, C., Grove, J., Plazonnet, B., 1989. Gelrite : A Novel Ion-activated,In-situ Gelling Polymer for Ophtalmic Vehicles. Effect on Bioavailability ofTimolol. Int. J. Pharm., 57, 163-168.

Rydén, L., Edman, P., 1992. Effect of Polymers and Microsperes on the NasalAbsorption of Insulin in Rats. Int. J. Pharm., 83, 1-10.

Saito, S., 1987. In Nonionic Surfactants: Physical Chemistry, Schick, M.J., Ed.,Surfactant Science Ser., Marcel Dekker, New York, vol. 23, p. 881.

Saito, Y., Kondo, Y., Abe, M., Sato, T., 1994. Solubilization behavior of estriol in anaqueous solution of pluronic L-64 as a function of concentration and temperature.Chem. Pharm. Bull., 42, 1348-1350.

Schick, M.J., Fowkes, F.M., 1967. Nonionic Surfactants, Surfactants Science SeriesVol. 1, Marcel Dekker, Inc., New York.

Schmolka, I.R., 1977. A Review of Block Polymer Surfactants. J. Amer. Oil Chem.Soc., 54, 110-116.

Schmolka, I.R., 1991. Polymers for Controlled Drug Delivery, Tarcha, P.J., Ed., CRCPress, Boca Raton.

52

Skaret, E., Raadal, M., Berg, E., Kvale, G., 1998. Dental anxiety among 18-yr-olds inNorway. Prevalence and related factors. Eur. J. Oral Sci., 106, 835-843.

Stilbs, P., 1987. Fourier Transform Pulsed-gradient Spin-echo Studies of MolecularDiffusion. Progress in NMR Spectroscopy, 19, 1-45.

Stout, G.H., Jensen, L.H., 1989. X-ray Structure Determination. A Practical Guide.John Wiley and Sons, Inc., New York.

Suzuki, H., Sasaki, T., 1971, Gel Filtration of Aqueous Sodium Dodecyl Sulfate. Bull.Chem. Soc. Jpn., 44, 2630-2634.

Svensson, B., Alexandridis, P., Olsson, U., 1998. Self-assembly of a Poly(ethyleneoxide)/poly(propylene oxide) Block Copolymer (Pluronic P104,(EO)27(PO)61(EO)27) in Water and Xylene. J. Phys. Chem. B, 102, 7541-7548.

Svensson, M., 1998. Aqueous Block Copolymer Systems � Self-association, PhaseBehaviour and Partitioning of Biomolecules, Ph. D. Thesis, University of Lund,Sweden.

Svensson, P., Petersen, J.K., 1992. Anesthetic effect of EMLA occluded with Orahesiveoral bandages on oral mucosa. A placebo-controlled study. Anesth. Prog., 39, 79-82.

Svensson, P., Arendt-Nielsen, L., Bjerring, P., Kaaber, S., 1993. Oral mucosal analgesiaquantitatively assessed by argon laser-induced thresholds and single-evoked vertexpotentials. Anesth. Pain Control Dent., 2, 154-161.

Svensson, P., Petersen, J.K., Svensson, H., 1994. Efficacy of a topical anesthetic onpain and unpleasantness during scaling of gingival pockets. Anesth. Prog., 41, 35-39.

Söderman, O., Stilbs, P., 1994. NMR Studies of Complex Surfactant Systems. Prog.Nucl. Magn. Reson. Spectrosc., 26, 445-482.

Söderman, O., Olsson, U., 1997. Dynamics of Amphiphilic Systems Srudied UsingNMR Relaxation and Pulsed Field Gradient Experiments. Current Op. ColloidInterface Sci., 2, 131-136.

Tarr, B.D., Yalkowsky, S.H., 1987. A new parenteral vehicle for the administration ofsome poorly water soluble anti-cancer drugs. J. Parenter. Sci. Technol., 41, 31-33.

Thuresson, K., Karlström, G., Lindman, B., 1995. Phase Diagrams of Mixtures of aNonionic Polymer, Hexanol and Water. An Experimental and Theoretical Study ofthe Effect of Hydrophobic Moodification. J. Phys. Chem., 99, 3823-3831.

Thuresson, K., Söderman, O., Hansson, P. and Wang, G., 1996. Binding of SDS toEthyl(hydroxyethyl)cellulose. Effect of Hydrophobic Modification of the Polymer.J. Phys. Chem., 100, 4909-4918.

Thuresson, K., Lindman, B., 1997. Effect of Hydrophobic Modification of a NonionicCellulose Derivative on the Interaction with Surfactants. Phase behavior andAssociation. J. Phys. Chem. B, 101, 6460-6468.

Tuzar, Z., Petrus, V., Kratochvil, P., 1974. Sedimentation and Light Scattering Study ofBlock Copolymer Association. Makromol. Chem., 175, 3181-3192.

Vadnere, M., Amidon, G.L., Lindenbaum, S., Haslam, J.L., 1984. ThermodynamicStudies on the Gel-sol Transition of Some Pluronic Polyols. Int. J. Pharm., 22, 207-218.

Walderhaug, H., Nyström, B., Hansen, F. K., Lindman, B., 1995. Interactions of IonicSurfactants with a Nonionic Cellulose Ether in Solution and in the Gel StateStudied by Pulsed Field Gradient NMR. J. Phys. Chem., 99, 4672-4678.

53

Wang, P.L., Johnston, T.P., 1991. Kinetics of Sol-to-Gel Transition for PoloxamerPolyols. J. Appl. Polym. Sci., 43, 283-292.

Wang, P.L., Johnston, T.P., 1995. Sustained Release Interleukin-2 FollowingIntramuscular Injection in Rats. Int. J. Pharm., 113, 73-81.

Wanka, G., Hoffman, H., Ulbricht, W., 1990. The Aggregation of Poly-(oxyethylene)-Poly(oxypropylene)-Block-Copolymers in Aqueous Solution. Colloid Polym. Sci.,268, 101-117.

Wanka, G., Hoffman, H., Ulbricht, W., 1994. Phase-Diagrams and AggregationBehavior of Poly(oxyethylene)-Poly(oxypropylene)-Poly(oxyethylene) TriblockCopolymers in Aqueous Solutions. Macromolecules, 27, 4145-4159.

Wu, G.W., Ying, Q.C., Chu, B., 1994. Lamellar Structure of Block CopolymerPoly(oxyethylene-oxypropylene-oxyethylene) in Xylene/Water Mixtures.Macromolecules, 27, 5758-5765.

Yokoyama, M., Miyauchi, M., Yamada, N., Okano, T., Sakurai, Y., Kataoka; K., Inoue,S., 1990. Polymer Micelles as Novel Drufg Carriers: Adriamycin-conjugatedPoly(ethylebe glycol)-poly(aspartic acid) Block Copolymer. J. Controlled Release,11, 269-278.

Yokoyama, M., 1992. Block Copolymers as Drug Carriers. Crit. Rev. Ther. DrugCarrier Syst., 9, 213-248.

Yu, G., Deng, Y., Dalton, S., Wang, Q-G., Attwood, D., Price, C., Booth, C., 1992.Micellisation and Gelation of Triblock Copoly(oxyethylene/oxypropylene/oxyethylene), F127. J. Chem. Soc., Faraday Trans., 88, 2537-2544.

Zana, R., Binana-Limbelé, W., Kamenka, N., Lindman, B., 1992.Ethyl(hydroxyethyl)cellulose � Cationic Surfactant Interactions: ElectricalConductivity, Self-diffusion, and Time-resolved Fluorescence QuenchingInvestigations. J. Phys. Chem., 96, 5461-5465.

Zhang, K., Karlström, G., Lindman, B., 1994. Ternary Aqueous Mixtures of NonionicPolymer with a Surfactant or a Second Polymer. A Theoretical and ExperimentalInvestigation of the Phase Behavior. J. Phys. Chem., 98, 4411-4421.

Zhang, K.W., Khan, A., 1995. Phase Behavior of Poly(ethylene oxide)-Poly(propyleneoxide)-Poly(ethylene oxide) Triblock Copolymers in Water. Macromolecules, 28,3807-3812.

Zhou, Z., Chu, B., 1988. Light-Scattering Study on the Association Behavior of triblockPolymers of Ethylene Oxide and Propylene Oxide in Aqueous Solution. J. ColloidInterface Sci., 126, 171-180.