- Email: [email protected]
Oesophageal bioadhesion of sodium alginate suspensions: particle swelling and mucosal retention
European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Oesophageal bioadhesion of sodium alginate suspensions: particle swelling and mucosal retention J. Craig Richardson a , Peter W. Dettmar b,1 , Frank C. Hampson b,1 , Colin D. Melia a,∗ a
Formulation Insights, School of Pharmacy, University of Nottingham, University Park, Nottingham NG7 2RD, UK b Reckitt Benckiser Healthcare (UK) Ltd., Dansom Lane, Hull HU8 7DS, UK Received 3 March 2004; accepted 4 May 2004 Available online 10 July 2004
Abstract This paper describes a prospective bioadhesive liquid dosage form designed to specifically adhere to the oesophageal mucosa. It contains a swelling polymer, sodium alginate, suspended in a water-miscible vehicle and is activated by dilution with saliva to form an adherent layer of polymer on the mucosal surface. The swelling of alginate particles and the bioadhesion of 40% (w/w) sodium alginate suspensions were investigated in a range of vehicles: glycerol, propylene glycol, PEG 200 and PEG 400. Swelling of particles as a function of vehicle dilution with artificial saliva was quantified microscopically using 1,9-dimethyl methylene blue (DMMB) as a visualising agent. The minimum vehicle dilution to initiate swelling varied between vehicles: glycerol required 30% (w/w) dilution whereas PEG 400 required nearly 60% (w/w). Swelling commenced when the Hildebrand solubility parameter of the diluted vehicle was raised to 37 MPa1/2 . The bioadhesive properties of suspensions were examined by quantifying the amount of sodium alginate retained on oesophageal mucosa after washing in artificial saliva. Suspensions exhibited considerable mucoretention and strong correlations were obtained between mucosal retention, the minimum dilution to initiate swelling, and the vehicle Hildebrand solubility parameter. These relationships may allow predictive design of suspensions with specific mucoretentive properties, through judicious choice of vehicle characteristics. © 2004 Elsevier B.V. All rights reserved. Keywords: Bioadhesion; Mucoadhesion; Sodium alginate; Suspension; Oesophagus; Hildebrand solubility parameter
1. Introduction Alginates are linear polysaccharides of 1,4 linked ␣-l-guluronic acid and -d-mannuronic acid and are utilised extensively as gelling and viscosity-increasing additives in the pharmaceutical and food industries (Draget et al., 1994; Smidsrød and Draget, 1996). An important pharmaceutical application of sodium alginate is in the management of gastro–oesophageal reflux, the involuntary movement of the contents of the stomach into the oesophagus (Mehta, 2000). Recurrent reflux damages the lower oesophageal mucosa and may have serious long-term clinical consequences (Katz, 1998; Lagergren et al., 1999). Sodium alginate is used to prevent reflux by forming a raft in the stomach that ∗ Corresponding author. Tel.: +44 115 9515032; fax: +44 115 9515102. E-mail address: [email protected] (C.D. Melia). 1 Present address: Technostics Ltd., The Deep Business Centre, Kingston Upon Hull, HU1 4BG, UK.
0928-0987/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ejps.2004.05.001
acts as a physical barrier against the movement of the stomach contents (Mandel et al., 2000). Recently, it has been suggested that the mucosal damage caused by reflux may also be treated by administering an oesophageal protectant, a liquid that when ingested coats the oesophagus with a polymeric layer (Potts et al., 2000). Putative oesophageal protectants include solutions of sucralfate, sodium alginate and Smart HydrogelTM (Dobrozsi et al., 1999; Potts et al., 2000; Batchelor et al., 2002 ). An ideal protectant would be specifically retained within the oesophagus and either form a protective barrier against reflux or act as a carrier system to deliver a therapeutic agent to the damaged mucosa. This work introduces the concept of using suspensions of sodium alginate in a water-miscible vehicle, as a delivery system for targeting the oesophageal mucosa. It is envisaged that sodium alginate suspensions would be specifically retained on the oesophageal mucosa as during oesophageal transit, in situ dilution of the suspension by saliva may act as a trigger for sodium alginate swelling and mucosal retention. This concept is based on the observation that for
50
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
water-swelling polymers, swelling is an important factor in the establishment of a bioadhesive interaction (Chen and Cyr, 1970; Mortazavi and Smart, 1993). The aim of this work was to investigate (i) how dilution of water-miscible vehicles with saliva, influenced sodium alginate swelling and (ii) if differences in vehicle composition may be utilised to modulate the bioadhesive properties of sodium alginate suspensions. The model vehicles chosen represent a series of water-miscible liquids with a suitable range of physico-chemical characteristics.
2.3. Hydration fluids for particle swelling The rate of alginate particle swelling was determined in a range of water miscible vehicles, diluted with increasing concentrations of artificial saliva. The vehicles used were glycerol, propylene glycol, a 70/30% (w/w) mixture of glycerol and propylene glycol (70/30), a 40/60% (w/w) mixture of glycerol and propylene glycol (40/60), polyethylene glycol 200 (PEG 200) and polyethylene glycol 400 (PEG 400). 2.4. Visualisation of swelling particles
2. Materials and methods 2.1. Materials Sodium alginate (Protanal LF120L) was a gift from FMC BioPolymer AS, Drammen, Norway. Latex beads (5 m diameter plain fluorescent pink PL-MicroSpheres) were obtained from Polymer Laboratories Ltd., UK. All other reagents were obtained from Sigma–Aldrich, Dorset, UK. The composition of artificial saliva (Lentner, 1981) was 5 mM sodium bicarbonate, 7.36 mM sodium chloride, 20 mM potassium chloride, 6.6 mM sodium dihydrogen phosphate monohydrate, 1.5 mM calcium chloride dihydrate in water (Maxima HPLC grade, maximum conductance 18.2 M/cm, USF Elga, Buckinghamshire, UK). 2.2. Measurement of particle swelling Sodium alginate particles were selected randomly from a 90–125 m sieve fraction, placed in the centre of a haemocytometer counting chamber (Thoma, Hawksley, UK) and covered with a weighted cover slip. Fifteen microlitres of hydration fluid was injected into the chamber to immerse the particle and a video microscope system, Nikon Labophot, 2× objective (Nikon UK Ltd., Surrey, UK) attached to a COHU High Performance CCD Camera (Brian Reece Scientific Ltd., Berkshire, UK), used to capture images of the radial 2D swelling of the particle with time. The particle was trapped in a chamber of a fixed depth and therefore changes in the 2D area represented a swelling volume. Image analysis software (Image-Pro Plus Version 4.0, Media Cybernetics, USA) was used to measure the normalised swollen area as defined in Eq. (1). Normalised area of particle = area of particle at time t − area of particle at t = 0 area of particle at t = 0
(1)
As the normalised swollen area was found to increase linearly with swelling time, the rate of particle swelling could be calculated from the gradient of the area:time plot.
When a sodium alginate particle was hydrated it became translucent and the swelling perimeter rapidly became invisible. To delineate the swollen particle, a visualisation aid was added to the hydration fluid. Two visualisation aids were chosen, 1,9-dimethyl methylene blue (DMMB) a cationic dye that complexes soluble sodium alginate (Halle et al., 1993) and 5 m latex beads. Particles were hydrated with hydration fluid containing 0.1% (w/v) 5 m latex beads or 250 M DMMB and the swelling of the particle imaged using video microscopy. To assess the effect of DMMB on swelling rate, particles were hydrated in 250 M to 2000 M DMMB in artificial saliva and the rate of swelling compared with a control experiment in which the alginate particle was hydrated in artificial saliva in the absence of DMMB. To delineate the swollen particle in the control experiment, a 0.7% (w/v) suspension of 5 m latex beads in artificial saliva was injected into the haemocytometer chamber after a predetermined period of hydration. 2.5. Preparation of suspensions in water-miscible vehicles Suspensions of sodium alginate (40%, w/w) (particle size: 90–125 m) were prepared by mixing alginate with the vehicle with a spatula in a glass vial, for 1 min, and were used immediately. Suspensions were prepared in the following vehicles: glycerol, a 70/30% (w/w) mixture of glycerol and propylene glycol, a 40/60% (w/w) mixture of glycerol and propylene glycol, propylene glycol, PEG 200 and PEG 400. 2.6. Measurement of the oesophageal adhesion of sodium alginate suspensions The retention of alginate suspensions on freshly prepared pig oesophagus was investigated using an adaptation of the method of Dobrozsi et al. (1999). Fresh oesophagus was collected in phosphate buffered saline immediately after slaughter and transported on ice. Within 1 h of slaughter, the musculature was removed and an 8 cm segment of the lower oesophagus everted onto a plastic rod. The tissue was gently rinsed and allowed to equilibrate for 1 min in 0.9% sodium chloride before being attached to a USP disintegration tester
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
(United States Pharmocopeia, 2002). The tissue was then lowered into 16 g of suspension and allowed to stand for 5 s to coat the mucosa with suspension. The disintegration tester was started and the tissue dipped in and out of 18 ml artificial saliva at 37 ◦ C. After 60 s of washing, the tester was stopped and the amount of sodium alginate retained on the mucosa (obtained by scraping) and the amount eluted into the washing container, were quantified using a dye-complexation assay (Halle et al., 1993). We have shown previously that the assay was not affected by mucosal tissue (Richardson et al., 2004). The amount of sodium alginate applied initially to the mucosa, was back-calculated by adding the amount of sodium alginate washed from the mucosa to the amount retained, and the mucosal retention expressed as follows: percentage sodium alginate retained = amount of alginate retained after washing (mg) × 100 amount of alginate initially applied to the mucosa (mg)
(2) 2.7. Calculation of Hildebrand solubility parameter values The Hildebrand solubility parameter is a measure of the cohesive forces within a solvent and is used to describe solvency behaviour (Barton, 1991; Greenhalgh et al., 1999). It can be calculated using Eq. (3): E 1/2 δ= (3) V δ is the Hildebrand solubility parameter, E the enthalpy of vaporisation at zero pressure, V the molar volume and E/V is the cohesive energy density. The enthalpy of vaporisation and molar volume of a solvent can be calculated using the group contribution method proposed by Fedors (1974) who considered a solvent molecule as a series of molecular fragments, and proposed that the enthalpy of vaporisation and molar volume of an entire molecule could be calculated by summing the contributions from each fragment (Eq. (4)): E = ei and V = vi (4) E is the enthalpy of vaporisation, ei is the sum of the enthalpy of vaporisation of each molecular fragment, V the molar volume and vi the sum of the molar volumes of each molecular fragment. Fedor calculated the enthalpy of vaporisation and molar volume for a range of atoms and
51
Table 2 Hildebrand solubility parameter of the water-miscible vehicles and water Vehicle
Hildebrand solubility parameter (MPa1/2 )
Glycerol 70/30a 40/60b Propylene glycol PEG 200 PEG 400 Water
33.55 31.77 29.98 27.61 26.10 22.70 47.90
a b
70/30% (w/w) glycerol/propylene glycol mixture. 40/60% (w/w) glycerol/propylene glycol mixture.
molecular groups and this allowed the Hildebrand solubility parameter to be calculated as follows: ei 1/2 (5) δ= vi This equation was used in this work to calculate the Hildebrand solubility parameter of each vehicle. Table 1 illustrates the calculation of the Hildebrand solubility parameter for glycerol, and Table 2 lists the values calculated for the other vehicles. The Hildebrand solubility parameter of binary mixtures of vehicle and artificial saliva were also calculated using Eq. (6) (Stengele et al., 2002): δmixture = V1 δ1 + V2 δ2
(6)
Where δmixture is the Hildebrand solubility parameter of the mixture of artificial saliva and vehicle, V1 the volume fraction of artificial saliva in mixture, δ1 the Hildebrand solubility parameter of artificial saliva, V2 the volume fraction of vehicle in mixture and δ2 the Hildebrand solubility parameter of vehicle. In the absence of published values, the Hildebrand solubility parameter for artificial saliva was taken as that of water (Barton, 1991) although it is acknowledged that the presence of electrolytes may influence this value. 2.8. Statistical analysis Statistical calculations were performed using GraphPad InStat Version 3.05 (GraphPad Software Inc., San Diego, CA, USA). One-way analysis of variance (ANOVA) and Tukey’s multiple comparison test were undertaken at a significance level (α) of 0.05. Non-linear regression analysis was performed using SigmaPlot 2002, Version 8.0 (SPSS Inc., Chicago, IL, USA).
Table 1 Calculation of the Hildebrand solubility parameter of glycerol using the Fedors group contribution method Number of molecular groups present
Enthalpy of vaporisation for each group (ei, J mol−1 )
Sum of the enthalpy of vaporisation ( ei, J mol−1 )
Molar volume for each group (vi, cm3 mol−1 )
Sum of the molar volume ( vi, cm3 mol−1 )
2 × CH2 3 × OH 1 × CH
CH2 = 4940 OH = 21900 CH = 3430
2 × 4940 = 9880 3 × 21900 = 65700 1 × 3430 = 3430 Total ei = 79010
CH2 = 16.1 OH = 13 CH = −1.0
2 × 16.1 = 32.2 3 × 13 = 39 1 × −1.0 = −1.0 Total vi = 70.2
Using Eq. (5), the Hildebrand solubility parameter of glycerol = (79010/70.2)1/2 = 33.55 MPa1/2 .
52
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
3. Results and discussion 3.1. Visualisation of swelling alginate particles When the alginate particle was placed in the haemocytometer chamber and hydrated with artificial saliva, it quickly became invisible (Fig. 1A and B). Latex beads were added to the artificial saliva and used to hydrate the particle but unfortunately the beads became enmeshed in the swollen particle and could not be used to measure the swollen area. DMMB was also added to the artificial saliva and Fig. 1C–E show the swelling of alginate particles hydrated in artificial saliva with DMMB as a visualisation agent. When hydrated in a solution of DMMB, the perimeter of the swollen particle was demarcated by a purple ring. This indicates the formation of a complex between DMMB and alginate (Halle et al., 1993) and allowed continuous monitoring of the size of the swollen area.
To investigate if the visualisation aid DMMB influenced particle swelling, swelling in a range of DMMB concentrations was examined. As a control, alginate particles were hydrated in artificial saliva without DMMB. In artificial saliva alone the particle disappeared, therefore to delineate the swollen area a second injection of a 0.7% (w/v) suspension of 5 m latex beads in artificial saliva was injected into the haemocytometer chamber after a predetermined period of hydration. Fig. 1F shows that the latex beads filled the haemocytometer chamber, but were unable to penetrate the area occupied by the swollen particle. This area therefore represents the extent of particle swelling at the time the beads were injected. By varying the time period between the first injection of artificial saliva and the second injection of beads, the swollen area of an alginate particle could be calculated at a range of different time points. As the alginate particle was hydrated in only artificial saliva, changes in the swollen area over time would not be influenced by DMMB.
Fig. 1. The swelling of sodium alginate particles in artificial saliva. (A) An unhydrated particle in the haemocytometre chamber. (B) After hydration in artificial saliva for 30 s. (C) Hydration of a particle in artificial saliva containing 250 M DMMB after 10 s (D) 30 s and (E) 60 s. (F) Alginate hydrated with artificial saliva alone followed by a second injection of latex beads after 60 s (control).
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Fig. 2. Effect of DMMB concentration on the swelling rate of sodium alginate particles. Mean (n = 10) ± 1S.D.
Fig. 2 shows that regardless of the DMMB concentration used there was no significant difference between these rates of swelling (ANOVA, P > 0.05). It was concluded that within this range, DMMB in artificial saliva did not influence the alginate swelling rate and therefore provided an excellent method for visualising alginate swelling. Future experiments were undertaken using the lowest concentration of DMMB required for visualisation of the perimeter, the area within the perimeter being used as the measure of particle swelling. 3.2. Influence of dilution on the swelling behaviour of sodium alginate particles in different vehicles Fig. 3 shows the swelling profiles of alginate particles in glycerol and illustrates how particle swelling behaviour changed on dilution. In 100% glycerol, alginate particles remained unswollen, but dilution with artificial saliva initiated swelling with the swelling rate thereafter being related to the extent of dilution. The swelling/time profiles were virtually linear over the timescales examined, and the rate of swelling was taken as a simple gradient of these curves. Fig. 4 shows the rates of swelling observed in different vehicles as a function of their dilution. It was clear that the degree of vehicle dilution required to initiate swelling was
53
Fig. 4. Effect of the vehicle dilution on the rate of sodium alginate particle swelling. Mean (n = 10) ±1S.D.
vehicle dependent: for example, glycerol had to be diluted with approximately 30% (w/w) artificial saliva before significant swelling was detected, whereas PEG 400 required dilution by nearly 60% (w/w). The shape of these curves makes it difficult to determine precisely the vehicle dilution required to initiate swelling, and therefore they were fitted to a power function of the form y = a + bxc . In all cases the fit was excellent (R2 ∼ 0.99) and the extent of dilution to initiate swelling could then be predicted with some confidence (Table 3). The initiation of swelling was defined as the dilution necessary to produce a swelling rate of 5% of the alginate in artificial saliva. 3.3. Relationship between alginate swelling and the Hildebrand solubility parameter To investigate the relationship between alginate swelling and vehicle characteristics, the Hildebrand solubility parameter was calculated for the dilution required to initiate swelling in the case of each vehicle. Table 4 shows that in each case, swelling of alginate was initiated only when sufficient artificial saliva had been added to the vehicle to raise its Hildebrand solubility parameter to approximately 37 MPa1/2 . It is usually accepted that the smaller the difference between the solubility parameter of solute and solvent, the greater the likelihood of solubility (Errede, 1986; Greenhalgh et al., 1999) but it is not possible to calculate the solubility parameter of sodium alginate directly as Fedor’s group contribution method is only suitable for low molecular weight molecules. However, this work suggests that at 37 MPa1/2 , the difference in solubility parameter between alginate and solvent is suitably minimised for alginate dissolution and swelling to commence. 3.4. Influence of the vehicle on the mucosal retention of sodium alginate suspensions
Fig. 3. Effect of glycerol concentration on the swelling of sodium alginate particles. Mean (n = 10) ± 1S.D. Glycerol diluted with artificial saliva.
Sodium alginate was applied to the oesophageal mucosa by ‘dipping’ an everted section of tissue into a 40%w/w suspension. The mean amount of alginate applied in this way
54
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Table 3 The relationship between vehicle dilution and rate of particle swelling and the predicted composition required to initiate particle swelling Vehicle
Glycerol 70/30a 40/60b Propylene glycol PEG 200 PEG 400 a b
Power function describing the vehicle dilution profile
y y y y y y
= −8.734e−3 + 2.99e−5 x2.066 = −5.543e−3 + 1.321e−6 x2.744 = −6.271e−3 + 4.952e−7 x2.957 = −5.30e−3 + 9.961e−9 x3.806 = −2.424e−3 + 5.057e−9 x3.95 = −5.769e−3 + 4.635e−11 x4.967
Coefficient of determination (R2 )
0.995 0.996 0.995 0.996 0.995 0.996
Predicted composition required to initiate alginate swelling Artificial saliva (% w/w)
Vehicle (% w/w)
27.6 36.3 39.5 48.0 47.9 57.5
72.4 63.7 60.5 52.0 52.1 42.5
70/30% (w/w) glycerol/propylene glycol mixture. 40/60% (w/w) glycerol/propylene glycol mixture.
Table 4 Hildebrand solubility parameter of the artificial saliva: vehicle mixtures required to initiate swelling of alginate particles Vehicle
Glycerol 70/30a 40/60b Propylene glycol PEG 200 PEG 400 a b
Predicted composition required to initiate alginate swelling Vehicle (% w/w)
Artificial saliva (% w/w)
Hildebrand solubility parameter (MPa1/2 )
27.6 36.3 39.5 48.0 47.9 57.5
72.4 63.7 60.5 52.0 52.1 42.5
37.51 37.62 37.06 37.35 36.54 37.19
70/30% (w/w) glycerol/propylene glycol mixture. 40/60% (w/w) glycerol/propylene glycol mixture.
varied between 140 mg and 190 mg, but there were no significant differences (ANOVA, P > 0.05) between suspensions prepared using different vehicles. Fig. 5 shows the percentage alginate retained after dipping in artificial saliva for 60 s. There were large differences in retention between suspensions in different vehicles. Glycerol suspensions retained almost 85% of their alginate content on the mucosa whilst PEG 400 suspensions retained only 32%. The rank order appeared similar to that observed for the swelling rate determinations, and in the next section this relationship is examined further.
3.5. Relationship between mucosal retention, swelling and Hildebrand solubility parameters
Fig. 5. Influence of the suspension vehicle on the mucosal retention of sodium alginate suspensions. Mean (n = 3) ±1S.D. 70/30 = 70/30% (w/w) glycerol/propylene glycol mixture; 40/60 = 40/60% (w/w) glycerol/propylene glycol mixture.
Fig. 6. Relationship between the mucosal retention of sodium alginate suspensions and the extent of vehicle dilution required to initiate alginate swelling. Mean (n = 3) ±1S.D. 70/30 = 70/30% (w/w) glycerol/propylene glycol mixture; 40/60 = 40/60% (w/w) glycerol/propylene glycol mixture.
Fig. 6 shows alginate retention plotted as a function of the vehicle dilution required to initiate single particle swelling. It can be seen that vehicles requiring the least dilution, exhibited greater levels of mucosal retention as suspensions. This may appear intuitive, but it is interesting to consider the underlying processes. Hydration of the suspensions will occur from contact with the wet mucosa, and also from wetting by the external medium. The results suggest that in ve-
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
55
late their bioadhesive characteristics in a predictable manner. The choice of vehicle was shown to influence the extent of vehicle dilution required to initiate particle swelling and this had an impact on bioadhesive performance. Suspensions containing a vehicle that required a low level of dilution to initiate swelling were more mucoretentive. Suspensions of sodium alginate, and similar swelling polymers, in water miscible vehicles may putatively provide a design for bioadhesive liquid formulations in which in situ vehicle dilution by saliva acts as a controllable trigger for polymer mucoretention. These suspensions may therefore act as a delivery system where targeting of the oesophageal mucosa is required.
Fig. 7. The mucosal retention of sodium alginate suspensions as a function of the Hildebrand solubility parameter of the suspension vehicle. Mean (n = 3) ±1S.D. 70/30 = 70/30% (w/w) glycerol/propylene glycol mixture; 40/60 = 40/60% (w/w) glycerol/propylene glycol mixture.
hicles such as glycerol, particle hydration is more extensive at an earlier stage of dilution and these suspensions rapidly swell and form an adhesive layer on the mucosa. Needleman et al. (1998) have shown how the viscosity of a retained polymer layer can influence its resistance to displacement from a tissue surface during washing. Therefore, it appears that glycerol suspensions rapidly hydrate and swell to form a viscous adherent layer on the mucosa that can resist elution. In contrast, vehicles such as PEG 400 show significant loss of material during the 60 s washing period (and this was seen visually as hydrated particles in the washing fluid). This would be consistent with a delay in the development of the necessary degree of hydration and viscosity to resist elution before the effects of washing are felt. The caveats we must apply to these correlations are that it is not known how rapidly these suspensions dilute or how water is partitioned between vehicle and alginate. Overall, the strong relationship in Fig. 6 suggests that particle swelling initiation might be an important modulator of mucosal retention of sodium alginate in these conditions. Fig. 7 plots the bioadhesive retention of the suspensions against the vehicle Hildebrand solubility parameter. An excellent correlation is obtained indicating this relationship may be a useful tool to apply to the design of suspensions with predicted bioadhesive performance. From this relationship, the composition of suspensions which deliver defined amounts of bioadhesive alginate can be predicted based primarily on the selection of a vehicle having the correct Hildebrand solubility parameter.
4. Conclusions It has been demonstrated that sodium alginate suspensions can adhere to oesophageal mucosa and by varying the composition of the vehicle it was possible to modu-
Acknowledgements We would like to thank Reckitt Benckiser Healthcare (UK) Ltd., FMC BioPolymer and the Biotechnology and Biological Sciences Research Council for the provision of a case award to Craig Richardson. The advice of Dr. Stephen Hibberd, School of Mathematics, University of Nottingham, is also gratefully acknowledged.
References Barton, A.F.M., 1991. CRC Handbook of Solubility Parameters and Other Cohesion Parameters. CRC Press Inc., Boca Raton, FL, USA. Batchelor, H.K., Banning, D., Dettmar, P.W., Hampson, F.C., Jolliffe, I.G., Craig, D.Q.M., 2002. An in vitro mucosal model for prediction of the bioadhesion of alginate solutions to the oesophagus. Int. J. Pharm. 238, 123–132. Chen, J.L., Cyr, G.N., 1970. Compositions producing adhesion through hydration. In: R.S. Manly (Ed.), Adhesion in Biological Systems. Academic Press, NY, USA, pp. 163–181. Dobrozsi, D.J., Smith, R.L., Sakr, A.A., 1999. Comparative mucoretention of sucralfate suspensions in an everted rat esophagus model. Int. J. Pharm. 189, 81–89. Draget, K.I., Skjåk Braek, G., Smidsrød, O., 1994. Alginic acid gels: the effect of alginate chemical composition and molecular weight. Carbohydr. Polym. 25, 31–38. Errede, L.A., 1986. Polymer Swelling. 5. Correlation of relative swelling of poly(styrene-co-divinylbenzene) with the Hildebrand solubility parameter of the swelling liquid. Macromolecules 19, 1522–1525. Fedors, R.F., 1974. A method for estimating both the solubility parameters and molar volumes of liquids. Polym. Eng. Sci. 14, 147–154. Greenhalgh, D.J., Williams, A.C., Timmins, P., York, P., 1999. Solubility parameters as predictors of miscibility in solid dispersions. J. Pharm. Sci. 88, 1182–1190. Halle, J.P., Landry, D., Fournier, A., Beaudry, M., Leblond, F.A., 1993. Method for the quantification of alginate in microcapsules. Cell Transp. 2, 429–436. Katz, P.O., 1998. Gastroesophageal reflux disease. J. Am. Geriatr. Soc. 46, 1558–1565. Lagergren, J., Bergstrom, R., Lindgren, A., Nyren, O., 1999. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. New Engl. J. Med. 340, 825–831. Lentner, C., 1981. Geigy Scientific Tables. Units of Measurement, Body Fluids, Composition of the Body and Nutrition, eighth ed. CIBA Geigy, Basle, Switzerland.
56
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Mandel, K.G., Daggy, B.P., Brodie, D.A., Jacoby, H.I., 2000. Review article: alginate-raft formulations in the treatment of hearburn and acid reflux. Aliment. Pharmacol. Ther. 14, 669–690. Mehta, D., 2000. British National Formulary, vol. 39. BMA and RPSGB, London, UK. Mortazavi, S.A., Smart, J.D., 1993. An investigation into the role of water movement and mucus gel dehydration in mucoadhesion. J. Control. Release 25, 197–203. Needleman, I.G., Martin, G.P., Smales, F.C., 1998. Characterisation of bioadhesives for periodontal and oral mucosal drug delivery. J. Clin. Periodontol. 25, 74–82. Potts, A.M., Wilson, C.G., Stevens, H.N.E., Dobrozsi, D.J., Washington, N., Frier, M., Perkins, A.C., 2000. Oesophageal bandaging: a new
opportunity for thermosetting polymers. STP Pharm. Sci. 10, 293– 301. Richardson, J.C., Dettmar, P.W., Hampson, F.C., Melia, C.D., 2004. A simple high throughput method for the quantification of sodium alginates on oesophageal mucosa. Eur. J. Pharm. Biopharm. 57, 299– 305. Smidsrød, O., Draget, K.I., 1996. Chemistry and physical properties of alginates. Carbohydr. Eur. 14, 6–13. Stengele, A., Rey, S., Leuenberger, H., 2002. A novel approach to the characterization of polar liquids. Part 2: binary mixtures. Int. J. Pharm. 241, 231–240. United States Pharmocopeia and the National Formulary, 2002. The Stationary Office, Rockville, MD, USA.
Oesophageal bioadhesion of sodium alginate suspensions: particle swelling and mucosal retention J. Craig Richardson a , Peter W. Dettmar b,1 , Frank C. Hampson b,1 , Colin D. Melia a,∗ a
Formulation Insights, School of Pharmacy, University of Nottingham, University Park, Nottingham NG7 2RD, UK b Reckitt Benckiser Healthcare (UK) Ltd., Dansom Lane, Hull HU8 7DS, UK Received 3 March 2004; accepted 4 May 2004 Available online 10 July 2004
Abstract This paper describes a prospective bioadhesive liquid dosage form designed to specifically adhere to the oesophageal mucosa. It contains a swelling polymer, sodium alginate, suspended in a water-miscible vehicle and is activated by dilution with saliva to form an adherent layer of polymer on the mucosal surface. The swelling of alginate particles and the bioadhesion of 40% (w/w) sodium alginate suspensions were investigated in a range of vehicles: glycerol, propylene glycol, PEG 200 and PEG 400. Swelling of particles as a function of vehicle dilution with artificial saliva was quantified microscopically using 1,9-dimethyl methylene blue (DMMB) as a visualising agent. The minimum vehicle dilution to initiate swelling varied between vehicles: glycerol required 30% (w/w) dilution whereas PEG 400 required nearly 60% (w/w). Swelling commenced when the Hildebrand solubility parameter of the diluted vehicle was raised to 37 MPa1/2 . The bioadhesive properties of suspensions were examined by quantifying the amount of sodium alginate retained on oesophageal mucosa after washing in artificial saliva. Suspensions exhibited considerable mucoretention and strong correlations were obtained between mucosal retention, the minimum dilution to initiate swelling, and the vehicle Hildebrand solubility parameter. These relationships may allow predictive design of suspensions with specific mucoretentive properties, through judicious choice of vehicle characteristics. © 2004 Elsevier B.V. All rights reserved. Keywords: Bioadhesion; Mucoadhesion; Sodium alginate; Suspension; Oesophagus; Hildebrand solubility parameter
1. Introduction Alginates are linear polysaccharides of 1,4 linked ␣-l-guluronic acid and -d-mannuronic acid and are utilised extensively as gelling and viscosity-increasing additives in the pharmaceutical and food industries (Draget et al., 1994; Smidsrød and Draget, 1996). An important pharmaceutical application of sodium alginate is in the management of gastro–oesophageal reflux, the involuntary movement of the contents of the stomach into the oesophagus (Mehta, 2000). Recurrent reflux damages the lower oesophageal mucosa and may have serious long-term clinical consequences (Katz, 1998; Lagergren et al., 1999). Sodium alginate is used to prevent reflux by forming a raft in the stomach that ∗ Corresponding author. Tel.: +44 115 9515032; fax: +44 115 9515102. E-mail address: [email protected] (C.D. Melia). 1 Present address: Technostics Ltd., The Deep Business Centre, Kingston Upon Hull, HU1 4BG, UK.
0928-0987/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ejps.2004.05.001
acts as a physical barrier against the movement of the stomach contents (Mandel et al., 2000). Recently, it has been suggested that the mucosal damage caused by reflux may also be treated by administering an oesophageal protectant, a liquid that when ingested coats the oesophagus with a polymeric layer (Potts et al., 2000). Putative oesophageal protectants include solutions of sucralfate, sodium alginate and Smart HydrogelTM (Dobrozsi et al., 1999; Potts et al., 2000; Batchelor et al., 2002 ). An ideal protectant would be specifically retained within the oesophagus and either form a protective barrier against reflux or act as a carrier system to deliver a therapeutic agent to the damaged mucosa. This work introduces the concept of using suspensions of sodium alginate in a water-miscible vehicle, as a delivery system for targeting the oesophageal mucosa. It is envisaged that sodium alginate suspensions would be specifically retained on the oesophageal mucosa as during oesophageal transit, in situ dilution of the suspension by saliva may act as a trigger for sodium alginate swelling and mucosal retention. This concept is based on the observation that for
50
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
water-swelling polymers, swelling is an important factor in the establishment of a bioadhesive interaction (Chen and Cyr, 1970; Mortazavi and Smart, 1993). The aim of this work was to investigate (i) how dilution of water-miscible vehicles with saliva, influenced sodium alginate swelling and (ii) if differences in vehicle composition may be utilised to modulate the bioadhesive properties of sodium alginate suspensions. The model vehicles chosen represent a series of water-miscible liquids with a suitable range of physico-chemical characteristics.
2.3. Hydration fluids for particle swelling The rate of alginate particle swelling was determined in a range of water miscible vehicles, diluted with increasing concentrations of artificial saliva. The vehicles used were glycerol, propylene glycol, a 70/30% (w/w) mixture of glycerol and propylene glycol (70/30), a 40/60% (w/w) mixture of glycerol and propylene glycol (40/60), polyethylene glycol 200 (PEG 200) and polyethylene glycol 400 (PEG 400). 2.4. Visualisation of swelling particles
2. Materials and methods 2.1. Materials Sodium alginate (Protanal LF120L) was a gift from FMC BioPolymer AS, Drammen, Norway. Latex beads (5 m diameter plain fluorescent pink PL-MicroSpheres) were obtained from Polymer Laboratories Ltd., UK. All other reagents were obtained from Sigma–Aldrich, Dorset, UK. The composition of artificial saliva (Lentner, 1981) was 5 mM sodium bicarbonate, 7.36 mM sodium chloride, 20 mM potassium chloride, 6.6 mM sodium dihydrogen phosphate monohydrate, 1.5 mM calcium chloride dihydrate in water (Maxima HPLC grade, maximum conductance 18.2 M/cm, USF Elga, Buckinghamshire, UK). 2.2. Measurement of particle swelling Sodium alginate particles were selected randomly from a 90–125 m sieve fraction, placed in the centre of a haemocytometer counting chamber (Thoma, Hawksley, UK) and covered with a weighted cover slip. Fifteen microlitres of hydration fluid was injected into the chamber to immerse the particle and a video microscope system, Nikon Labophot, 2× objective (Nikon UK Ltd., Surrey, UK) attached to a COHU High Performance CCD Camera (Brian Reece Scientific Ltd., Berkshire, UK), used to capture images of the radial 2D swelling of the particle with time. The particle was trapped in a chamber of a fixed depth and therefore changes in the 2D area represented a swelling volume. Image analysis software (Image-Pro Plus Version 4.0, Media Cybernetics, USA) was used to measure the normalised swollen area as defined in Eq. (1). Normalised area of particle = area of particle at time t − area of particle at t = 0 area of particle at t = 0
(1)
As the normalised swollen area was found to increase linearly with swelling time, the rate of particle swelling could be calculated from the gradient of the area:time plot.
When a sodium alginate particle was hydrated it became translucent and the swelling perimeter rapidly became invisible. To delineate the swollen particle, a visualisation aid was added to the hydration fluid. Two visualisation aids were chosen, 1,9-dimethyl methylene blue (DMMB) a cationic dye that complexes soluble sodium alginate (Halle et al., 1993) and 5 m latex beads. Particles were hydrated with hydration fluid containing 0.1% (w/v) 5 m latex beads or 250 M DMMB and the swelling of the particle imaged using video microscopy. To assess the effect of DMMB on swelling rate, particles were hydrated in 250 M to 2000 M DMMB in artificial saliva and the rate of swelling compared with a control experiment in which the alginate particle was hydrated in artificial saliva in the absence of DMMB. To delineate the swollen particle in the control experiment, a 0.7% (w/v) suspension of 5 m latex beads in artificial saliva was injected into the haemocytometer chamber after a predetermined period of hydration. 2.5. Preparation of suspensions in water-miscible vehicles Suspensions of sodium alginate (40%, w/w) (particle size: 90–125 m) were prepared by mixing alginate with the vehicle with a spatula in a glass vial, for 1 min, and were used immediately. Suspensions were prepared in the following vehicles: glycerol, a 70/30% (w/w) mixture of glycerol and propylene glycol, a 40/60% (w/w) mixture of glycerol and propylene glycol, propylene glycol, PEG 200 and PEG 400. 2.6. Measurement of the oesophageal adhesion of sodium alginate suspensions The retention of alginate suspensions on freshly prepared pig oesophagus was investigated using an adaptation of the method of Dobrozsi et al. (1999). Fresh oesophagus was collected in phosphate buffered saline immediately after slaughter and transported on ice. Within 1 h of slaughter, the musculature was removed and an 8 cm segment of the lower oesophagus everted onto a plastic rod. The tissue was gently rinsed and allowed to equilibrate for 1 min in 0.9% sodium chloride before being attached to a USP disintegration tester
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
(United States Pharmocopeia, 2002). The tissue was then lowered into 16 g of suspension and allowed to stand for 5 s to coat the mucosa with suspension. The disintegration tester was started and the tissue dipped in and out of 18 ml artificial saliva at 37 ◦ C. After 60 s of washing, the tester was stopped and the amount of sodium alginate retained on the mucosa (obtained by scraping) and the amount eluted into the washing container, were quantified using a dye-complexation assay (Halle et al., 1993). We have shown previously that the assay was not affected by mucosal tissue (Richardson et al., 2004). The amount of sodium alginate applied initially to the mucosa, was back-calculated by adding the amount of sodium alginate washed from the mucosa to the amount retained, and the mucosal retention expressed as follows: percentage sodium alginate retained = amount of alginate retained after washing (mg) × 100 amount of alginate initially applied to the mucosa (mg)
(2) 2.7. Calculation of Hildebrand solubility parameter values The Hildebrand solubility parameter is a measure of the cohesive forces within a solvent and is used to describe solvency behaviour (Barton, 1991; Greenhalgh et al., 1999). It can be calculated using Eq. (3): E 1/2 δ= (3) V δ is the Hildebrand solubility parameter, E the enthalpy of vaporisation at zero pressure, V the molar volume and E/V is the cohesive energy density. The enthalpy of vaporisation and molar volume of a solvent can be calculated using the group contribution method proposed by Fedors (1974) who considered a solvent molecule as a series of molecular fragments, and proposed that the enthalpy of vaporisation and molar volume of an entire molecule could be calculated by summing the contributions from each fragment (Eq. (4)): E = ei and V = vi (4) E is the enthalpy of vaporisation, ei is the sum of the enthalpy of vaporisation of each molecular fragment, V the molar volume and vi the sum of the molar volumes of each molecular fragment. Fedor calculated the enthalpy of vaporisation and molar volume for a range of atoms and
51
Table 2 Hildebrand solubility parameter of the water-miscible vehicles and water Vehicle
Hildebrand solubility parameter (MPa1/2 )
Glycerol 70/30a 40/60b Propylene glycol PEG 200 PEG 400 Water
33.55 31.77 29.98 27.61 26.10 22.70 47.90
a b
70/30% (w/w) glycerol/propylene glycol mixture. 40/60% (w/w) glycerol/propylene glycol mixture.
molecular groups and this allowed the Hildebrand solubility parameter to be calculated as follows: ei 1/2 (5) δ= vi This equation was used in this work to calculate the Hildebrand solubility parameter of each vehicle. Table 1 illustrates the calculation of the Hildebrand solubility parameter for glycerol, and Table 2 lists the values calculated for the other vehicles. The Hildebrand solubility parameter of binary mixtures of vehicle and artificial saliva were also calculated using Eq. (6) (Stengele et al., 2002): δmixture = V1 δ1 + V2 δ2
(6)
Where δmixture is the Hildebrand solubility parameter of the mixture of artificial saliva and vehicle, V1 the volume fraction of artificial saliva in mixture, δ1 the Hildebrand solubility parameter of artificial saliva, V2 the volume fraction of vehicle in mixture and δ2 the Hildebrand solubility parameter of vehicle. In the absence of published values, the Hildebrand solubility parameter for artificial saliva was taken as that of water (Barton, 1991) although it is acknowledged that the presence of electrolytes may influence this value. 2.8. Statistical analysis Statistical calculations were performed using GraphPad InStat Version 3.05 (GraphPad Software Inc., San Diego, CA, USA). One-way analysis of variance (ANOVA) and Tukey’s multiple comparison test were undertaken at a significance level (α) of 0.05. Non-linear regression analysis was performed using SigmaPlot 2002, Version 8.0 (SPSS Inc., Chicago, IL, USA).
Table 1 Calculation of the Hildebrand solubility parameter of glycerol using the Fedors group contribution method Number of molecular groups present
Enthalpy of vaporisation for each group (ei, J mol−1 )
Sum of the enthalpy of vaporisation ( ei, J mol−1 )
Molar volume for each group (vi, cm3 mol−1 )
Sum of the molar volume ( vi, cm3 mol−1 )
2 × CH2 3 × OH 1 × CH
CH2 = 4940 OH = 21900 CH = 3430
2 × 4940 = 9880 3 × 21900 = 65700 1 × 3430 = 3430 Total ei = 79010
CH2 = 16.1 OH = 13 CH = −1.0
2 × 16.1 = 32.2 3 × 13 = 39 1 × −1.0 = −1.0 Total vi = 70.2
Using Eq. (5), the Hildebrand solubility parameter of glycerol = (79010/70.2)1/2 = 33.55 MPa1/2 .
52
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
3. Results and discussion 3.1. Visualisation of swelling alginate particles When the alginate particle was placed in the haemocytometer chamber and hydrated with artificial saliva, it quickly became invisible (Fig. 1A and B). Latex beads were added to the artificial saliva and used to hydrate the particle but unfortunately the beads became enmeshed in the swollen particle and could not be used to measure the swollen area. DMMB was also added to the artificial saliva and Fig. 1C–E show the swelling of alginate particles hydrated in artificial saliva with DMMB as a visualisation agent. When hydrated in a solution of DMMB, the perimeter of the swollen particle was demarcated by a purple ring. This indicates the formation of a complex between DMMB and alginate (Halle et al., 1993) and allowed continuous monitoring of the size of the swollen area.
To investigate if the visualisation aid DMMB influenced particle swelling, swelling in a range of DMMB concentrations was examined. As a control, alginate particles were hydrated in artificial saliva without DMMB. In artificial saliva alone the particle disappeared, therefore to delineate the swollen area a second injection of a 0.7% (w/v) suspension of 5 m latex beads in artificial saliva was injected into the haemocytometer chamber after a predetermined period of hydration. Fig. 1F shows that the latex beads filled the haemocytometer chamber, but were unable to penetrate the area occupied by the swollen particle. This area therefore represents the extent of particle swelling at the time the beads were injected. By varying the time period between the first injection of artificial saliva and the second injection of beads, the swollen area of an alginate particle could be calculated at a range of different time points. As the alginate particle was hydrated in only artificial saliva, changes in the swollen area over time would not be influenced by DMMB.
Fig. 1. The swelling of sodium alginate particles in artificial saliva. (A) An unhydrated particle in the haemocytometre chamber. (B) After hydration in artificial saliva for 30 s. (C) Hydration of a particle in artificial saliva containing 250 M DMMB after 10 s (D) 30 s and (E) 60 s. (F) Alginate hydrated with artificial saliva alone followed by a second injection of latex beads after 60 s (control).
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Fig. 2. Effect of DMMB concentration on the swelling rate of sodium alginate particles. Mean (n = 10) ± 1S.D.
Fig. 2 shows that regardless of the DMMB concentration used there was no significant difference between these rates of swelling (ANOVA, P > 0.05). It was concluded that within this range, DMMB in artificial saliva did not influence the alginate swelling rate and therefore provided an excellent method for visualising alginate swelling. Future experiments were undertaken using the lowest concentration of DMMB required for visualisation of the perimeter, the area within the perimeter being used as the measure of particle swelling. 3.2. Influence of dilution on the swelling behaviour of sodium alginate particles in different vehicles Fig. 3 shows the swelling profiles of alginate particles in glycerol and illustrates how particle swelling behaviour changed on dilution. In 100% glycerol, alginate particles remained unswollen, but dilution with artificial saliva initiated swelling with the swelling rate thereafter being related to the extent of dilution. The swelling/time profiles were virtually linear over the timescales examined, and the rate of swelling was taken as a simple gradient of these curves. Fig. 4 shows the rates of swelling observed in different vehicles as a function of their dilution. It was clear that the degree of vehicle dilution required to initiate swelling was
53
Fig. 4. Effect of the vehicle dilution on the rate of sodium alginate particle swelling. Mean (n = 10) ±1S.D.
vehicle dependent: for example, glycerol had to be diluted with approximately 30% (w/w) artificial saliva before significant swelling was detected, whereas PEG 400 required dilution by nearly 60% (w/w). The shape of these curves makes it difficult to determine precisely the vehicle dilution required to initiate swelling, and therefore they were fitted to a power function of the form y = a + bxc . In all cases the fit was excellent (R2 ∼ 0.99) and the extent of dilution to initiate swelling could then be predicted with some confidence (Table 3). The initiation of swelling was defined as the dilution necessary to produce a swelling rate of 5% of the alginate in artificial saliva. 3.3. Relationship between alginate swelling and the Hildebrand solubility parameter To investigate the relationship between alginate swelling and vehicle characteristics, the Hildebrand solubility parameter was calculated for the dilution required to initiate swelling in the case of each vehicle. Table 4 shows that in each case, swelling of alginate was initiated only when sufficient artificial saliva had been added to the vehicle to raise its Hildebrand solubility parameter to approximately 37 MPa1/2 . It is usually accepted that the smaller the difference between the solubility parameter of solute and solvent, the greater the likelihood of solubility (Errede, 1986; Greenhalgh et al., 1999) but it is not possible to calculate the solubility parameter of sodium alginate directly as Fedor’s group contribution method is only suitable for low molecular weight molecules. However, this work suggests that at 37 MPa1/2 , the difference in solubility parameter between alginate and solvent is suitably minimised for alginate dissolution and swelling to commence. 3.4. Influence of the vehicle on the mucosal retention of sodium alginate suspensions
Fig. 3. Effect of glycerol concentration on the swelling of sodium alginate particles. Mean (n = 10) ± 1S.D. Glycerol diluted with artificial saliva.
Sodium alginate was applied to the oesophageal mucosa by ‘dipping’ an everted section of tissue into a 40%w/w suspension. The mean amount of alginate applied in this way
54
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Table 3 The relationship between vehicle dilution and rate of particle swelling and the predicted composition required to initiate particle swelling Vehicle
Glycerol 70/30a 40/60b Propylene glycol PEG 200 PEG 400 a b
Power function describing the vehicle dilution profile
y y y y y y
= −8.734e−3 + 2.99e−5 x2.066 = −5.543e−3 + 1.321e−6 x2.744 = −6.271e−3 + 4.952e−7 x2.957 = −5.30e−3 + 9.961e−9 x3.806 = −2.424e−3 + 5.057e−9 x3.95 = −5.769e−3 + 4.635e−11 x4.967
Coefficient of determination (R2 )
0.995 0.996 0.995 0.996 0.995 0.996
Predicted composition required to initiate alginate swelling Artificial saliva (% w/w)
Vehicle (% w/w)
27.6 36.3 39.5 48.0 47.9 57.5
72.4 63.7 60.5 52.0 52.1 42.5
70/30% (w/w) glycerol/propylene glycol mixture. 40/60% (w/w) glycerol/propylene glycol mixture.
Table 4 Hildebrand solubility parameter of the artificial saliva: vehicle mixtures required to initiate swelling of alginate particles Vehicle
Glycerol 70/30a 40/60b Propylene glycol PEG 200 PEG 400 a b
Predicted composition required to initiate alginate swelling Vehicle (% w/w)
Artificial saliva (% w/w)
Hildebrand solubility parameter (MPa1/2 )
27.6 36.3 39.5 48.0 47.9 57.5
72.4 63.7 60.5 52.0 52.1 42.5
37.51 37.62 37.06 37.35 36.54 37.19
70/30% (w/w) glycerol/propylene glycol mixture. 40/60% (w/w) glycerol/propylene glycol mixture.
varied between 140 mg and 190 mg, but there were no significant differences (ANOVA, P > 0.05) between suspensions prepared using different vehicles. Fig. 5 shows the percentage alginate retained after dipping in artificial saliva for 60 s. There were large differences in retention between suspensions in different vehicles. Glycerol suspensions retained almost 85% of their alginate content on the mucosa whilst PEG 400 suspensions retained only 32%. The rank order appeared similar to that observed for the swelling rate determinations, and in the next section this relationship is examined further.
3.5. Relationship between mucosal retention, swelling and Hildebrand solubility parameters
Fig. 5. Influence of the suspension vehicle on the mucosal retention of sodium alginate suspensions. Mean (n = 3) ±1S.D. 70/30 = 70/30% (w/w) glycerol/propylene glycol mixture; 40/60 = 40/60% (w/w) glycerol/propylene glycol mixture.
Fig. 6. Relationship between the mucosal retention of sodium alginate suspensions and the extent of vehicle dilution required to initiate alginate swelling. Mean (n = 3) ±1S.D. 70/30 = 70/30% (w/w) glycerol/propylene glycol mixture; 40/60 = 40/60% (w/w) glycerol/propylene glycol mixture.
Fig. 6 shows alginate retention plotted as a function of the vehicle dilution required to initiate single particle swelling. It can be seen that vehicles requiring the least dilution, exhibited greater levels of mucosal retention as suspensions. This may appear intuitive, but it is interesting to consider the underlying processes. Hydration of the suspensions will occur from contact with the wet mucosa, and also from wetting by the external medium. The results suggest that in ve-
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
55
late their bioadhesive characteristics in a predictable manner. The choice of vehicle was shown to influence the extent of vehicle dilution required to initiate particle swelling and this had an impact on bioadhesive performance. Suspensions containing a vehicle that required a low level of dilution to initiate swelling were more mucoretentive. Suspensions of sodium alginate, and similar swelling polymers, in water miscible vehicles may putatively provide a design for bioadhesive liquid formulations in which in situ vehicle dilution by saliva acts as a controllable trigger for polymer mucoretention. These suspensions may therefore act as a delivery system where targeting of the oesophageal mucosa is required.
Fig. 7. The mucosal retention of sodium alginate suspensions as a function of the Hildebrand solubility parameter of the suspension vehicle. Mean (n = 3) ±1S.D. 70/30 = 70/30% (w/w) glycerol/propylene glycol mixture; 40/60 = 40/60% (w/w) glycerol/propylene glycol mixture.
hicles such as glycerol, particle hydration is more extensive at an earlier stage of dilution and these suspensions rapidly swell and form an adhesive layer on the mucosa. Needleman et al. (1998) have shown how the viscosity of a retained polymer layer can influence its resistance to displacement from a tissue surface during washing. Therefore, it appears that glycerol suspensions rapidly hydrate and swell to form a viscous adherent layer on the mucosa that can resist elution. In contrast, vehicles such as PEG 400 show significant loss of material during the 60 s washing period (and this was seen visually as hydrated particles in the washing fluid). This would be consistent with a delay in the development of the necessary degree of hydration and viscosity to resist elution before the effects of washing are felt. The caveats we must apply to these correlations are that it is not known how rapidly these suspensions dilute or how water is partitioned between vehicle and alginate. Overall, the strong relationship in Fig. 6 suggests that particle swelling initiation might be an important modulator of mucosal retention of sodium alginate in these conditions. Fig. 7 plots the bioadhesive retention of the suspensions against the vehicle Hildebrand solubility parameter. An excellent correlation is obtained indicating this relationship may be a useful tool to apply to the design of suspensions with predicted bioadhesive performance. From this relationship, the composition of suspensions which deliver defined amounts of bioadhesive alginate can be predicted based primarily on the selection of a vehicle having the correct Hildebrand solubility parameter.
4. Conclusions It has been demonstrated that sodium alginate suspensions can adhere to oesophageal mucosa and by varying the composition of the vehicle it was possible to modu-
Acknowledgements We would like to thank Reckitt Benckiser Healthcare (UK) Ltd., FMC BioPolymer and the Biotechnology and Biological Sciences Research Council for the provision of a case award to Craig Richardson. The advice of Dr. Stephen Hibberd, School of Mathematics, University of Nottingham, is also gratefully acknowledged.
References Barton, A.F.M., 1991. CRC Handbook of Solubility Parameters and Other Cohesion Parameters. CRC Press Inc., Boca Raton, FL, USA. Batchelor, H.K., Banning, D., Dettmar, P.W., Hampson, F.C., Jolliffe, I.G., Craig, D.Q.M., 2002. An in vitro mucosal model for prediction of the bioadhesion of alginate solutions to the oesophagus. Int. J. Pharm. 238, 123–132. Chen, J.L., Cyr, G.N., 1970. Compositions producing adhesion through hydration. In: R.S. Manly (Ed.), Adhesion in Biological Systems. Academic Press, NY, USA, pp. 163–181. Dobrozsi, D.J., Smith, R.L., Sakr, A.A., 1999. Comparative mucoretention of sucralfate suspensions in an everted rat esophagus model. Int. J. Pharm. 189, 81–89. Draget, K.I., Skjåk Braek, G., Smidsrød, O., 1994. Alginic acid gels: the effect of alginate chemical composition and molecular weight. Carbohydr. Polym. 25, 31–38. Errede, L.A., 1986. Polymer Swelling. 5. Correlation of relative swelling of poly(styrene-co-divinylbenzene) with the Hildebrand solubility parameter of the swelling liquid. Macromolecules 19, 1522–1525. Fedors, R.F., 1974. A method for estimating both the solubility parameters and molar volumes of liquids. Polym. Eng. Sci. 14, 147–154. Greenhalgh, D.J., Williams, A.C., Timmins, P., York, P., 1999. Solubility parameters as predictors of miscibility in solid dispersions. J. Pharm. Sci. 88, 1182–1190. Halle, J.P., Landry, D., Fournier, A., Beaudry, M., Leblond, F.A., 1993. Method for the quantification of alginate in microcapsules. Cell Transp. 2, 429–436. Katz, P.O., 1998. Gastroesophageal reflux disease. J. Am. Geriatr. Soc. 46, 1558–1565. Lagergren, J., Bergstrom, R., Lindgren, A., Nyren, O., 1999. Symptomatic gastroesophageal reflux as a risk factor for esophageal adenocarcinoma. New Engl. J. Med. 340, 825–831. Lentner, C., 1981. Geigy Scientific Tables. Units of Measurement, Body Fluids, Composition of the Body and Nutrition, eighth ed. CIBA Geigy, Basle, Switzerland.
56
J.C. Richardson et al. / European Journal of Pharmaceutical Sciences 23 (2004) 49–56
Mandel, K.G., Daggy, B.P., Brodie, D.A., Jacoby, H.I., 2000. Review article: alginate-raft formulations in the treatment of hearburn and acid reflux. Aliment. Pharmacol. Ther. 14, 669–690. Mehta, D., 2000. British National Formulary, vol. 39. BMA and RPSGB, London, UK. Mortazavi, S.A., Smart, J.D., 1993. An investigation into the role of water movement and mucus gel dehydration in mucoadhesion. J. Control. Release 25, 197–203. Needleman, I.G., Martin, G.P., Smales, F.C., 1998. Characterisation of bioadhesives for periodontal and oral mucosal drug delivery. J. Clin. Periodontol. 25, 74–82. Potts, A.M., Wilson, C.G., Stevens, H.N.E., Dobrozsi, D.J., Washington, N., Frier, M., Perkins, A.C., 2000. Oesophageal bandaging: a new
opportunity for thermosetting polymers. STP Pharm. Sci. 10, 293– 301. Richardson, J.C., Dettmar, P.W., Hampson, F.C., Melia, C.D., 2004. A simple high throughput method for the quantification of sodium alginates on oesophageal mucosa. Eur. J. Pharm. Biopharm. 57, 299– 305. Smidsrød, O., Draget, K.I., 1996. Chemistry and physical properties of alginates. Carbohydr. Eur. 14, 6–13. Stengele, A., Rey, S., Leuenberger, H., 2002. A novel approach to the characterization of polar liquids. Part 2: binary mixtures. Int. J. Pharm. 241, 231–240. United States Pharmocopeia and the National Formulary, 2002. The Stationary Office, Rockville, MD, USA.