International Journal of Pharmaceutics (v.261, #1-2)
TITLE PAGE(EDI BOARD) (iii).
Effect of camphor/cyclodextrin complexation on the stability of O/W/O multiple emulsions by Shan-Chen Yu; Amélie Bochot; Geneviève Le Bas; Monique Chéron; Jacqueline Mahuteau; Jean-Louis Grossiord; Monique Seiller; Dominique Duchêne (1-8).
Camphor (CA) encapsulation in oil/water/oil multiple emulsions prepared with cyclodextrin disturbs the emulsifier potential of α- and β-natural cyclodextrins (CD). It was suggested that the size and geometrical fit between the CD cavity and CA could induce CD/CA complex formation in place of emulsifier formation leading to perturbation of emulsion stability. The complexation between CA and α-, β- or γ-CD in solution in the presence of oil phase are confirmed by phase-solubility diagrams, circular dichroism and 1 H NMR. Furthermore, in order to mimic the emulsion system, CD/CA/soybean oil ternary dispersions were prepared to observe the complexation behavior of α-, β- or γ-CD/CA by circular dichroism. X-ray diffraction on emulsion samples prepared with α- and β-CD confirms that the precipitates observed in emulsions are probably composed of crystals of CD/CA complexes. A preliminary study of the interaction between drug and CD before the formulation seems indispensable to prevent the risk of incompatibility.
Keywords: Camphor; Complexation; Cyclodextrin; Multiple emulsion;
Design of controlled-release formulation for ivermectin using silicone by H Maeda; M Brandon; A Sano (9-19).
The purpose of this study was to design a formulation using silicone as carrier, so that release of ivermectin (IVM) can be controlled for a long period of time. The lateral side of a cylindrical matrix-type formulation composed of IVM and silicone was covered with silicone to obtain a covered-rod (CR) formulation. With this formulation, linear release of IVM was obtained. With addition of polyethylene glycol 4000 (PEG), release of IVM was accelerated. In a trial with subcutaneous administration to mice, blood concentration of IVM was maintained within one-order over a period of 3 months. The velocity of release of IVM from CR preparation depended on the change in solubility of IVM by additives, and in the case of a formulation with addition of desoxycholate sodium, linear in vitro release of IVM was observed over a period of 1 year.
Keywords: Silicone; Controlled release; Ivermectin; Covered-rod; Solubility;
The absorption behavior of cyclosporin A lecithin vesicles in rat intestinal tissue by Ying Chen; Qineng Ping; Jianxin Guo; Wenli Lv; Jing Gao (21-26).
The purpose of the study was to investigate the absorption behavior of lecithin vesicles of cyclosporin A (CsA-VES), prepared by the rotary evaporation method and treated further with sonication. The everted gut sac technique and in situ circulation method were used to examine: (1) relationship between the CsA-VES absorption velocity and the CsA-VES content; (2) the influence of the intestinal mucus, blank vesicles, concentration of Na+, energy inhibitor and P-gp inhibitor on the absorption of CsA-VES; and (3) the respective accumulated content of CsA in the incubating medium and the sacs after incubation with Sandimmum Neoral®(CsA-NEO) and CsA-VES. Our results showed there was a saturated absorption of CsA. Most CsA-VES accumulated in the mucus before it reached the intestinal tissue. There was no significant difference in the accumulated absorption content in the incubating medium and the sacs of CsA-NEO and CsA-VES. The addition of blank vesicles and concentration of Na+ had no significant influence on the accumulated absorption of CsA (P>0.05).The energy inhibitor and P-gp inhibitor influenced the accumulated absorption of CsA significantly (P<0.05). CsA-VES may be transported by phagocytosis. Mucus was a barrier blocking the diffusion of CsA-VES. CsA-NEO and CsA-VES showed equal absorption levels in the intestine.
Keywords: Lecithin vesicles; Cyclosporin A; Everted gut sac technique; In situ circulation method; Absorption behavior;
Design of controlled release delivery systems using a modified pharmacokinetic approach: a case study for drugs having a short elimination half-life and a narrow therapeutic index by Anurag Sood; Ramesh Panchagnula (27-41).
The objectives of peroral controlled release drug delivery systems (CRDDS) are to maintain therapeutically effective plasma drug concentration levels for a longer duration thereby reducing the dosing frequency and to minimise the plasma drug concentration fluctuations at steady state by delivering drug in a controlled and a reproducible manner. Drug delivery rate, duration of delivery and the dosing interval are the target features for any temporal CRDDS. The classical pharmacokinetic model for designing CRDDS [Drug Dev. Ind. Pharm. 15 (1989) 1073] assumes the time of drug delivery (t del) to be less than the dosing interval. However, termination of drug release from such a CRDDS at t del and/or a declining drug input function towards the terminal phase of t del from a first order kinetic CRDDS can have severe implications on plasma drug concentration and steady state fluctuations for a drug with very short half-life. A case study is presented in this paper, wherein by means of theoretical calculations using a classical pharmacokinetic approach, it is shown that a first order kinetic CRDDS for hypothetical drugs with short elimination half-life and different pharmacokinetic conditions would result in sub-therapeutic plasma concentrations at least for some time during the dosing interval at steady state. In order to avoid sub-therapeutic plasma drug concentrations a modification in classical pharmacokinetic model is proposed and discussed through theoretical calculations for different hypothetical pharmacokinetic situations and a practical single dose pharmacokinetic study with a first order kinetic CRDDS for nifedipine (a short half-life drug; about 2 h). It is shown that improved therapeutic efficacy could be expected from a CRDDS developed based on proposed modification in the classical pharmacokinetic model.
Keywords: Controlled release; Pharmacokinetic model; Nifedipine; Diltiazem;
RGD-anchored magnetic liposomes for monocytes/neutrophils-mediated brain targeting by Sanyog Jain; Vivek Mishra; Paramjit Singh; P.K Dubey; D.K Saraf; S.P Vyas (43-55).
Negatively charged magnetic liposomes were prepared using soya lecithin (Soya PC), cholesterol and phosphatidyl serine (PS) for their preferential presentation to circulating blood phagocytes (monocytes and neutrophils). PS ratio was optimized in terms of drug and magnetite loading, in vitro magnetic responsiveness and ex vivo monocytes/neutrophils uptake. RGD peptide was covalently coupled to the negatively charged liposomes composed of PC, cholesterol, PS and phopsphatidyl ethanolamine (PE) via carbodiimide-mediated coupling. In vivo cellular sorting study under magnetic guard indicated an increase in relative count of neutrophils and monocytes. Results suggest that selective uptake of RGD-anchored magnetic liposomes by these cells imparts them magnetic property. High levels of a model drug diclofenac sodium was quantified in target organ brain. In case of negatively charged uncoated magnetic liposomes brain levels of the drug was 5.95-fold compared to free drug and 7.58-fold in comparison to non-magnetic formulation, while for RGD-coated magnetic liposomes this ratio was 9.1-fold compared to free drug solution, 6.62-fold compared to non-magnetic RGD-coated liposomes and 1.5-fold when compared to uncoated magnetic liposomes. Liver uptake was significantly bypassed (37.2% and 48.3% for uncoated and RGD-coated magnetic liposomes, respectively). This study suggested the potential of negatively charged and RGD-coated magnetic liposomes for monocytes/neutrophils-mediated active delivery of drugs to relatively inaccessible inflammatory sites, i.e. brain. The study opens a new perspective of active delivery of drugs for a possible treatment of cerebrovascular diseases.
Keywords: Magnetic liposomes; Brain targeting; RGD; Blood–brain barrier; Monocytes; Neutrophils;
Preparation and characterization of ibuprofen–cetyl alcohol beads by melt solidification technique: effect of variables by Manish Maheshwari; Anant R Ketkar; Bhaskar Chauhan; Vinay B Patil; Anant R Paradkar (57-67).
Ibuprofen (IBU) exhibits short half-life, poor compressibility, flowability and caking tendency. IBU melt has sufficiently low viscosity and exhibits interfacial tension sufficient to form droplet even at low temperature. A single step novel melt solidification technique (MST) was developed to produce IBU beads with lower amounts of excipient. Effect of variables was studied using a 32 factorial approach with speed of agitation and amount of cetyl alcohol (CA) as variables. The beads were evaluated using DSC, FT-IR and scanning electron microscope (SEM). Yield, micromeritic properties, crushing strength and release kinetics were also studied. Spherical beads with a method yield of above 90% were obtained. The data was analyzed by response surface methodology. The variables showed curvilinear relationship with yield in desired particle size range, crushing strength and, bulk and tap density. The drug release followed non-Fickian case II transport and the release rate decreased linearly with respect to amount of CA in the initial stages followed by curvilinearity at later stages of elution. The effect of changing porosity and tortuosity was well correlated.
Keywords: Ibuprofen; Cetyl alcohol; Melt solidification technique; Factorial design; Response surface methodology;
PLGA–PEG microspheres of teverelix: influence of polymer type on microsphere characteristics and on teverelix in vitro release by Delphine Mallardé; François Boutignon; Fabien Moine; Edith Barré; Sandrine David; Hélène Touchet; Paolo Ferruti; Romano Deghenghi (69-80).
Teverelix microspheres were produced by coacervation using a new type of poly(ester-carbonates) made of block copolymers of poly(lactic-glycolic acid) (PLGA) and poly(ethylene glycol) (PEG). Five different PLGA–PEG copolymers and one PLGA were used. The ‘stability window’ has been determined for all polymers. It varied depending on the molecular weight and the weight percentage of PEG. With increasing core loading (from 9.4 to 34.2%), the microparticle size increased from 10–50 to 5–1000 μm. The core loading did not have any influence on encapsulation yield, which remained above 80%. The influence of polymer type on microsphere characteristics was studied at two different core loadings: 9.4 and 28%. At a low core loading, the nature of the polymer had no influence on microsphere characteristics whereas at 28%, only PLGA–PEG copolymers gave acceptable microparticles in term of particle size. At 28%, the glass transition temperature (T g) of loaded particles was 1–8 °C higher than the T g of the corresponding polymer. Increasing the core loading increased teverelix release whereas polymer degradation was decreased. All microparticles made of PLGA–PEG copolymers showed a faster release of teverelix than PLGA-based microspheres, whatever the core loading. One PLGA–PEG was selected on the basis of in vitro release rate for further in vivo investigations.
Keywords: Microspheres; Coacervation; PLGA–PEG; PEG; In vitro release; GnRH antagonists;
Enzyme-mediated precipitation of parent drugs from their phosphate prodrugs by Tycho Heimbach; Doo-Man Oh; Lilian Y Li; Naı́r Rodrı́guez-Hornedo; George Garcia; David Fleisher (81-92).
Many oral phosphate prodrugs have failed to improve the rate or extent of absorption compared to their insoluble parent drugs. Rapid parent drug generation via intestinal alkaline phosphatase can result in supersaturated solutions, leading to parent drug precipitation. The purpose was to (1) investigate whether parent drugs can precipitate from prodrug solutions in presence of alkaline phosphatase; (2) determine whether induction times are influenced by (a) dephosphorylation rate, (b) parent drug supersaturation level, and (c) parent drug solubility. Induction times were determined from increases in optical densities after enzyme addition to prodrug solutions of TAT-59, fosphenytoin and estramustine phosphate. Apparent supersaturation ratios (σ) were calculated from parent drug solubility at intestinal pH. Precipitation could be generated for all three prodrugs. Induction times decreased with increased enzyme activity and supersaturation level and were within gastrointestinal residence times for TAT-59 concentration≥21 μM (σ≥210). Induction times for fosphenytoin were less than the GI residence time (199 min) for concentrations of approximately 352 μM (σ=4.0). At approximately 475 μM (σ=5.3) the induction times were less than 90 min. For estramustine-phosphate, no precipitation was observed within GI residence times. Enzyme-mediated precipitation will depend on apparent supersaturation ratios, parent drug dose, solubility and solubilization by the prodrug.
Keywords: Induction time; Phosphate prodrug; Precipitation; TAT-59; Supersaturation; Enzyme;
The sustained granulopoietic effect of progenipoietin encapsulated in multivesicular liposomes by Mysore P Ramprasad; Arjang Amini; Tugrul Kararli; Nandini V Katre (93-103).
Progenipoietin (ProGP), a dual receptor agonist of fetal liver tyrosine kinase-3 (flt3) and granulocyte colony-stimulating factor (G-CSF) receptors, has been shown to significantly enhance production of both polymorphonuclear leukocytes and dendritic cells (DCs) in the peripheral blood and spleen of mice, when administered as daily s.c. injections for about 10 days. Here, we have successfully designed a sustained-delivery formulation for this novel chimeric protein using multivesicular liposomes (DepoFoam), and studied the effects of changing both the triglyceride and phospholipid composition of the lipid matrix to modulate its delivery profile. Encapsulation of ProGP in these particles led to retention of its structural integrity, and maintenance of its biological activity in vivo. Administration of a single s.c. dose of 1 mg/kg of an optimized DepoProGP formulation in rats, led to significant elevation of absolute neutrophil counts (ANC) that were maintained at levels >10,000 μl−1 for 9–11 days, in a reproducible manner. In contrast, administration of the unencapsulated ProGP at the same dose, resulted in elevation of neutrophils by day 1, followed by a quick decline to base line levels by day 3. These data suggest the possibility of administering a single dose of DepoFoam-encapsulated ProGP to improve hematopoietic recovery time after chemotherapy, and for other indications that require multiple daily doses of ProGP.
Keywords: Progenipoietin; Hematopoietic growth factors; Multivesicular liposome; DepoFoam; Sustained granulopoiesis; G-CSF;
Localized delivery to CT-26 tumors in mice using thermosensitive liposomes by Jeffrey Wells; Arindam Sen; Sek Wen Hui (105-114).
A heat-sensitive liposomal drug delivery system was tested using Colon-26 (CT-26) cultured cells and tumors in mice. Lucifer yellow iodoacetamide (LY) was used as a fluorescence marker. The heat-sensitive liposomes exploit the temperature-dependence of critical micellar concentrations of the poloxamer, F127. LY release from unilamellar liposomes at different temperatures was measured. Onset of LY release occurred near 33 °C, and reached plateau above 42 °C when 90% of the LY was released. Temperature-treated liposomes were mixed with CT-26 cells to measure the binding of the released LY to cell surface. Temperature-dependency of cell-bound LY corresponds to the release curve. CT-26 tumors were grown subcutaneously in both hind legs of Balb/c mice. Mice received heat-sensitive or plain liposomes via tail vein injections, or no liposomes. For each mouse, one tumor was kept at 31.5 °C, while the counterlateral tumor was heated to 42 °C during injection and for 30 min after. LY released in tumors was determined from fluorescence intensity. Tumors receiving heat-sensitive liposomes plus heat treatment showed 2.5-fold greater fluorescence than all other tumors, which were at the background level. This study demonstrates the possible use of poloxamer-containing liposomes as a heat-sensitive drug delivery system in vivo.
Keywords: Liposome; Poloxamer; Hyperthermia; Controlled release; Fluorescence; Colon tumor;
The influence of formulation variables on in vitro transfection efficiency and physicochemical properties of chitosan-based polyplexes by Kristine Romøren; Solveig Pedersen; Gro Smistad; Øystein Evensen; Beate J Thu (115-127).
The aim of this study was to investigate how a selection of formulation variables affects the in vitro transfection efficiency and physicochemical properties (particle size, zetapotential and chitosan–plasmid association) of chitosan-based polyplexes. Experimental designs in combination with multivariate data analysis were applied to reveal the effects of the formulation variables on the responses. The following formulation variables were studied: molecular weight and degree of acetylation of chitosan, pH and ionic strength of the buffer in which chitosan was dissolved, charge ratio of polyplexes, plasmid concentration and inclusion of a coacervation agent in the plasmid solution. The in vitro transfection efficiency in Epithelioma papulosum cyprini (EPC) cells was affected by the polyplex charge ratio, the DNA concentration in the complexes as well as the molecular weight and degree of acetylation of the chitosans. Two favourable formulations were identified in a more thorough investigation. These formulations were made of SC113 (theoretical charge ratio 10) and SC214 (theoretical charge ratio 3). The size of the complexes was affected by the degree of acetylation, concentration of DNA, pH, inclusion of a coacervation agent and the charge ratio. The charge ratio, pH and ionic strength determined the zetapotential of the particles, while the charge ratio was important for the association between the plasmid and chitosan.
Keywords: DNA vaccine; Chitosan; In vitro transfection efficiency; Physicochemical properties; Multivariate data analysis;
Effects of various absorption enhancers on transport of clodronate through Caco-2 cells by Johanna Raiman; Soili Törmälehto; Kirsi Yritys; Hans E Junginger; Jukka Mönkkönen (129-136).
The major disadvantage concerning clinical use of bishosphonate drugs, like clodronate, is their poor and variable absorption after oral administration. The objective of this study was to assess the effects of four different absorption enhancers—palmitoyl carnitine chloride (PCC), N-trimethyl chitosan chloride (TMC), sodium caprate (C10), and ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA)—on the transport of clodronate using Caco-2 cell culture model. The transport experiments were performed in a normal (1.3 mM) and in a minimum-calcium concentration (apically calcium-free medium and basolaterally 100 μM calcium concentration). In the normal calcium concentration, a strong enhancement in clodronate permeation was observed with the enhancers: EGTA (2.5 mM), TMC (1.5% w/v), and PCC (0.2 mM) increased the transport of 1 mM clodronate 190-, 20-, and 10-fold, respectively, and the transport of 10 mM clodronate 130-, 70-, and 35-fold. In the minimum-calcium concentration, the effects of the absorption enhancers on the transport of clodronate were not so potent: TMC, PCC, and EGTA caused 2- to 20-fold enhancement in clodronate permeation whereas C10 (10 mM) was without any effect. According to the results, the permeation of clodronate through Caco-2 cells could be significantly promoted by the absorption enhancers, which cause widening of the tight junctions and, thus, increase the permeability of the paracellular route.
Keywords: Clodronate; Bisphosphonates; Absorption; Absorption enhancers; Caco-2 cells;
An automatic dose dispenser for microtablets—a new concept for individual dosage of drugs in tablet form by Susanne Bredenberg; Dag Nyholm; Sten-Magnus Aquilonius; Christer Nyström (137-146).
A new concept for individualising the dosage of drugs in solid form is presented. The principle is based on the use of standardised units (microtablets), each containing a subtherapeutic amount of the active ingredient. The required dose is fine-tuned by counting out a specific number of these units. The microtablets are counted electronically from the attached cassette by the automatic dispensing device. The individual dose is set and the dispenser counts and delivers the correct number of microtablets. The usefulness of the automatic dispenser concept and acceptability of the apparatus were evaluated in patients with Parkinson’s disease (PD). After initial instruction on use of the dispenser, 20 patients operated it themselves. All patients were generally satisfied with their management of the automatic dispenser and most would be happy to use the device again. Further technical development is required before use in clinical practice, but the current prototype may be acceptable for some patients. It is concluded that the final version of the automatic dose dispenser concept will offer potential for improvement of drug administration for patients with PD or other diseases requiring individual dosage.
Keywords: Individual dosage; Drug dispensing device; Parkinson’s disease; Levodopa; Microtablets;
Comparative in vitro permeability of human vaginal, small intestinal and colonic mucosa by Pieter van der Bijl; Armorel D van Eyk (147-152).
Our previous experience with a continuous flow-through perfusion system has demonstrated its usefulness for studying diffusion kinetics of drugs across small intestinal mucosa for bioavailability/bioequivalence (BA/BE) studies. During the last decade, delivery of drugs to the colon for systemic absorption as well as for local delivery in certain colonic diseases, has been extensively investigated. For this reason, we sought to assess the in vitro comparative permeability of human vaginal, small intestinal and colonic mucosa using a flow-through perfusion method. It was clear from our studies that human colonic epithelium was statistically significantly (P<0.05) more permeable to water, 17β-estradiol, arecoline and arecaidine than intestinal mucosa. However, both these mucosae were statistically significantly less permeable to the above four permeants than human vaginal mucosa. As previously shown for small intestinal mucosa, the low in vitro permeability of colonic mucosa to drugs with molecular weight >300 Da may necessitate using other epithelial membranes, e.g. vaginal mucosa, as alternative barriers for in vitro BA/BE studies. We also concluded that the flow-through mucosal perfusion system used in our laboratory is therefore also potentially useful for determining the permeability of a therapeutic agent from the colon for registration purposes.
Keywords: Human intestinal mucosa; Permeability studies; Various compounds;
Formation and stabilisation of triclosan colloidal suspensions using supersaturated systems by S.L Raghavan; K Schuessel; A Davis; J Hadgraft (153-158).
The aim of this paper is to prepare and stabilise, in situ, colloidal microsuspensions of triclosan using the polymer, hydroxypropyl methylcellulose (HPMC). The suspensions were prepared from supersaturated solutions of triclosan. The cosolvent technique was used to create supersaturation. Propylene glycol and water were used as the cosolvents. The triclosan particles had a large needle-shaped morphology, when grown in the absence of the polymer. Moreover, the particles grew rapidly to sizes greater than 5 μm over a period of 7 h. When HPMC was added, the particle sizes were in the range 90–250 nm depending on the amount of polymer present in the solutions. The stability of the solutions was evaluated over a period of 40 days during which the particle sizes did not vary. The results were consistent with the mechanism proposed by Raghavan et al. [Int. J. Pharm. 212 (2001b) 213].
Keywords: Triclosan; Hydroxypropyl methylcellulose; Cosolvent technique; Colloidal dispersions; Crystal growth; Supersaturation;
Electrostatic charge on spacer devices and salbutamol response in young children by J.C. Dubus; C. Guillot; M. Badier (159-164).
Electrostatic charge on plastic spacer devices may affect the efficacy of inhaled drugs, but its consequences have never been evaluated in asthmatic children with airflow limitation. At the end of a positive metacholine challenge, 64 children (51.3±12.9 months, 32 boys, specific airway resistance (SRaw) 257.1±56.7% and forced expiratory volume in 1 s (FEV1) 64.2±17.9% of the predicted value) inhaled one puff of hydrofluoroalkane-134a (HFA-134a) salbutamol (Ventoline®), and 15 min later two other puffs (total dose of 300 μg), delivered through either a new static Babyhaler® (n=21), a detergent-coated, reduced static, Babyhaler® (n=20), or a metal NES-Spacer® (n=23) equipped with facemask. SRaw and FEV1 were measured after each treatment and compared between groups by a Kruskal–Wallis test. The first 100 μg salbutamol induced a 151.7±43.9% decrease in SRaw and a 19.9±10.6% increase in FEV1. Additional 200 μg salbutamol allowed a supplementary decrease of 35.1±25.7% in SRaw and increase of 12.1±11.8% in FEV1, without significant difference between the spacer devices. Electrostatic charge on spacer devices does not affect bronchodilation with HFA-134a salbutamol in metacholine-challenged pre-school children. This could be in part explained by the use of supramaximal doses of salbutamol.
Keywords: Asthma; Children; Electrostatic charge; Salbutamol; Spacer devices;
The in vitro delivery of NSAIDs across skin was in proportion to the delivery of essential fatty acids in the vehicle—evidence that solutes permeate skin associated with their solvation cages? by Charles M Heard; Simon J Gallagher; John Harwood; Paula B Maguire (165-169).
As part of our investigations into novel dual action topical anti-arthritis systems, the permeation of ibuprofen or ketoprofen plus eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) were determined from a fish oil vehicle across pig ear skin in vitro. The steady state fluxes of ibuprofen and ketoprofen were 9.17±1.98 μg cm−2 h−1 and 6.12±2.39 μg cm−2 h−1, respectively. At 24 h, 5.7 μg cm−2 EPA and 3.1 μg cm−2 DHA permeated when the solute was ibuprofen; 1.4 μg cm−2 EPA and 1.0 μg cm−2 DHA when ketoprofen was the solute. At 12 h, the ketoprofen/ibuprofen ratio of the moles permeated was 0.27, the ratio of EPA permeated simultaneously with ketoprofen and ibuprofen was 0.22 and the ratio of DHA permeated simultaneously with ketoprofen and ibuprofen was 0.24. We believe this is the first time that simultaneous permeation across skin of a solute and its vehicle has been determined purposefully. The data successfully demonstrated that simultaneous permeation of NSAIDs and essential fatty acids, EPA and DHA from a formulation containing fish oil is feasible. In addition, for both NSAIDs, the relative rates of permeation of EPA and DHA, were in proportion to their levels in the fish oil and the permeation rate of either fatty acid was higher when the permeation rate of the solute was greater. This suggested that the greater the rate of permeation of the NSAID, the greater the rate of permeation of the vehicle, and that a solute permeates skin complete with its vehicular solvation cage. This apparent relationship between solute and vehicle fluxes may be of more widespread significance to skin permeation experimentation.
Keywords: Arthritis; Ibuprofen; Ketoprofen; Fish oil; Transcutaneous delivery; n−3 fatty acids; Vehicle;
Transdermal Drug Delivery by A.C. Williams (171).