Nasal drug delivery has been recognized as a very promising route for delivery of therapeutic compounds. This interest arises from the different possible advantages presented by the nasal cavity, such as: the epithelium very vascularized and with a relatively large surface area available for drug absorption, the porous endothelial basement membrane, the direct transport of absorbed drugs into the systemic circulation thereby avoiding the first-pass effect hepatic present in peroral administration, the lower enzymatic activity compared with the gastrointestinal tract and the liver (Ugwoke et al., 2001). For all these reasons the nasal route can be considered a useful alternative both to parenteral and oral routes (Edman and Bjork, 1992; Turker et al., 2004). A wide range of nasal products is in development, mostly in correlation with the rapid onset of action of nasal route, for example, for the treatment of pain (nasal morphine and ketamine) and for the treatment of erectile dysfunction (nasal apomorphine) (Illum, 2003). [2]
However, there are some problems such as mucociliary clearance and low permeability of the nasal mucosa to some drugs that have a large influence on the efficiency of the nasal absorption of drugs [2]. Nasal mucociliary clearance is one of the most important limiting factor for nasal drug delivery. It severely limits the time allowed for drug absorption to occur and effectively rules out sustained nasal drug administration. However, mucoadhesive preparations have been developed to increase the contact time between the dosage form and mucosal layers of nasal cavities thus enhancing drug absorption [3,4]. Illum et al. [5] introduced mucoadhesive microsphere systems for nasal delivery and characterized them well. The microspheres form a gel-like layer, which is cleared slowly from the nasal cavity, resulting in a prolonged residence time of the drug formulation. [3,4]
Another important limiting factor in nasal application is the low permeability of the nasal mucosa for the drugs with polar and high molecular size. It seems to be necessary to consider an absorption enhancement mechanism for co-administration of drugs with either mucoadhesive polymers or penetration enhancers or combination of the two [9_/11]. [3,4]
Ketorolac tromethamine (KT) is a potent non-narcotic analgesic with moderate anti- inflammatory activity (1). [5] Its use has been implicated in the number of acute painful conditions ranging from moderate to severe pain relief such as postpartum (Bloom. eld et al. 1984) and postoperative pain (Yee et al. 1984, 1986). [6] When administered as the conventional formulation, it causes gastrointestinal complaints such as gastrointestinal bleeding, perforation and peptic ulceration [5,6]. KT has a shorter mean plasma elimination half-life of 4-6 hr. Therefore, it is imperative to design prolong releasing dosage form in order to reduce the frequency of dosing and adverse effects, especially since duration of treatment is typically longer for NSAIDS. It is been reportedly known that sustained release formulations of KT such as, sustained release tablet,[6] transdermal [7], nasal powder[8], gel[9] and microspheres[10], liposomes[11], osmotic tablets[12], ocular gel[13], parenteral microspheres[14] etc. have been attempted by many researchers. Preparation of mucoadhesive microspheres could be advantageous strategy so as to provide an intimate contact between the drug delivery system and the mucosal membrane. This could be achieved by incorporating a mucoadhesive agent in the polymer backbone of microspheres.
Various attempts have been made in order to prepare mucoadhesive microspheres by spray drying technique using hydroxyl propyl methyl cellulose, carbopol, chitosan, hydroxyl propyl cellulose, polycarbophil with excellent mucoadhesive properties (Harikarnpakdee et al., 2006; Sakagami et al., 2002; Desai & Park, 2005).[15,16,17] The objective of the present study was to prepare the KT-loaded microspheres by the spray-drying method using Carbopol (CP), Chitosan (CS) and Polycarbophil (PL). Prepared microspheres were characterized for their surface morphology, swelling behavior, mucoahesion, drug release proï¬le and ex vivo nasal cilio toxicity by using appropriate evaluative studies.
Materials and Methods
Materials
Ketorolac tromethamine (KT) was obtained as a gift sample from Symed Labs Limited, Hydrabad, (India). Carbopol 974® PNF (CP) and Noveon AA-1 (Polycarbophil, PL) were obtained as a gift sample from Lubrizol Advanced Materials Inc, Mumbai (India). Chitosan, >85% deacetylation, was kindly contributed from the Central Institute of Fisheries Technology, Kochi, India. All other reagents and solvent were of analytical grade and used without purification.
Preparation of Ketorolac mucoadhesive microspheres
Mucoadhesive microspheres were prepared by spray drying of dispersion using a LU-222 spray drier (Lab Ultima, India) with a standard 0.7 mm nozzle. [15] For the dispersion system, CP and PL were solubilized in de-ionized water at different concentrations and CS was solubilized in 1% v/v aqueous acetic acid solution. KT was dissolved in each polymeric solution per se, in order to achieve desired drug-to-polymer ratio of 1:4 and 1:5. The preliminary experiments were carried out to study the effects of process and formulation parameters on the yield and particle size of the resulting microspheres, were studied by setting the pump rate (5, 10 and 15 mL/min), inlet temperature (140°, 160° and 180°C) and drug-to-polymer concentration (1:4 and 1:5). Each formulation was prepared in triplicate for further studies.
Encapsulation efficiency
Twenty five mg of accurately weighed drug loaded mucoadhesive microspheres were added to 100 mL of 0.1 N HCl.[14] The resulting mixture was kept shaking on a mechanical shaker for 24 hr. Then the solution was filtered and 1 mL of this solution was appropriately diluted with 0.1 N HCl and analyzed spectrophotometrically at 322 nm. The drug encapsulation efficiency was calculated using equation (1):
.[1]
Particle Size
Particle size of prepared microspheres was determined using Microscopic imaging analysis technique. [18] Particle size distribution of microspheres was performed using an AXIOPALN microscope (Zeiss MPM400 Germany) that is equipped with a computer-controlled image analysis system (Zeiss KS300 Germany).
Scanning electron microscope (SEM)
A scanning electron microscope (ESEM TMP with EDAX, Philips, Holland) was used to characterize the surface topography of the microspheres. The microscope was equipped with electron optical system (EOS) consisting of 0.5-30 kV capacity electron gun and an electron detector. The microspheres were placed on a metallic support with a thin adhesive tape and were coated with gold under vacuum. The surface was scanned and photographs were taken at 30kV accelerating voltage for the drug loaded microspheres.
Swelling index
The swelling ability of the microspheres in physiological media was determined by allowing the microspheres to swell to their equilibrium. [19, 20] Accurately weighed amounts of microspheres were immersed in a little excess of Phosphate buffer (pH 6.6) and kept for 24 hr. The following formula was used for calculation of percentage of swelling:
…………..[2]
Where, Ssw = Percentage swelling of microspheres; Wo = initial weight of microspheres; and Ws = weight of microspheres after swelling.
Mucoadhesion
Mucoadhesion of different microspheres system was assessed using the method reported by Jain SK et al [19] with little modification. A strip of sheep nasal mucosa was mounted on a glass slide and accurately weighed bioadhesive microspheres in dispersion form was placed on the mucosa of the intestine. This glass slide was incubated for 15 min in a desiccator at 90% relative humidity to allow the polymer to interact with the membrane and finally placed in the cell that was attached to the outer assembly at an angle 45°. Phosphate buffer saline (pH 6.6), previously warmed to 37 ± 0.5 °C, was circulated to the cell over the microspheres and membrane at the rate of 1 mL/min. Washings were collected at different time intervals and microspheres were separated by centrifugation followed by drying at 50 °C. The weight of microspheres washed out was taken and percentage mucoadhesion was calculated by the following formula:
…..[3]
Where, Wo = Weight of microspheres applied; Wt = Weight of microspheres washed out.
Fourier Transform Infrared Spectroscopy (FTIR)
The spectra were recorded for pure drug, drug loaded microspheres and blank microspheres using FTIR. Samples were prepared in KBr disks (2 mg sample in 200 mg KBr). The scanning range was 400 - 4000 cm -1 and the resolution was 2 cm-1.
Differential Scanning Calorimetry (DSC)
Differential scanning calorimetry scans of drug, blank microspheres and drug loaded microspheres were performed using DSC-PYRIS-1. The analysis was performed with a heating range of 50 - 300oC and a rate of 10 oC min-1.
In vitro diffusion study
The in vitro drug release test of the pure drug (KT) and prepared microspheres was carried out using an apparatus called Franz diffusion cell.[20,21] A dialysis membrane (cut-off Mw 12,000) was placed between the microspheres sample and receptor compartment containing phosphate buffer solution (pH 6.6). The KT loaded microspheres equivalent to 10 mg of KT were applied to the dialysis membrane. The volume of the receptor compartment was 20 mL, which is similar to that of a nasal cavity. The temperature of the receptor medium was adjusted to 37±1 0C. The content of the receptor compartment was continuously stirred with a magnetic stirrer. Aliquot of a 1.0 mL were withdrawn from the receptor compartment at hourly intervals for 8 hr and replaced with the same amount of fresh buffer solution. The aliquot was analyzed for the drug content at 322 nm after appropriate dilutions against reference using phosphate buffer saline pH 6.6 as blank. All experiments were performed in triplicate.
Release kinetics
In order to understand the mechanism and kinetics of drug release, the results of the in vitro drug release study were fitted with various kinetic equations.[22] The kinetic models used are zero-order, first-order, Higuchi matrix, and Baker and Lonsdale models19. The Higuchi square root of time model has been derived from Fick's first law of diffusion and is suited for the modeling of drug release from a homogeneous planar matrix, assuming that the matrix does not dissolve. The Baker and Lonsdale models drug release from diffusion rate-limiting matrixes of spherical shape. In order to define a model which will represent a better fit for the formulation, drug release data were analyzed by Peppas equation. r2 values were calculated for the linear curves obtained by regression analysis of the above plots.
Ex-vivo Nasal cilio toxicity of mucoadhesive microspheres
Freshly excised sheep nasal mucosa, except for the septum, was collected from the slaughter house in saline phosphate buffer pH 6.6.[23,24] Four sheep nasal mucosa pieces (N1, N2, N3, N4) with uniform thickness were selected and mounted on Franz diffusion cells. N1 was treated with 0.5 mL of saline phosphate buffer pH 6.6 (negative control); N2 with 0.5 mL of isopropyl alcohol (positive control), N3 with 0.5 mL of KT in phosphate buffer pH 6.6 and N4 with 0.5 mL of KT loaded polymeric microspheres for 1 hr. After 1 hr, the mucosa was treated with saline phosphate buffer pH 6.6 and subjected to histological studies to evaluate the toxicities of KT loaded polymeric microspheres.
Histology
After removal of the sheep nasal mucosa from diffusion cell, the tissues were placed in 10% buffered formaldehyde solution, fixed for 72 hr. For the purpose of histological study, tissues were dehydrated in ascending degrees of ethylalcohol (70, 80, 90, 96, and 99% v/v) and sequentially embedded in paraffin wax blocks according to the standard procedure, sectioned at 5 µ thickness. They were further deparaffined with xylol, and histologic observations were performed after staining for functional nasal tissues by hematoxylin-eosin. The slides were examined using light microscope.
Statistical analyses
All the reported determinations were performed in triplicate. One-way analysis of variance (ANOVA) followed by Tuckey's multiple range test was performed to determine the least significant difference for all the reported evaluations. The differences were considered as significant at p < 0.05.
Results
Effects of process variables on the characteristics of the microspheres
The results of the influence of preparation parameters on the characteristics of the mucoadhesive microspheres are shown in Table 1. It was evident from the study that process variables such as drug to polymer ratio, inlet temperature and pump flow rate greatly affect the size and yield of microspheres. The different polymer concentrations were employed for the preparation of microspheres but low polymer concentrations (1:1 to 1:3) was not able to give the desired particle size as the droplet size was small and most of the part of drop consisted of solvent which would evaporate leaving the small particles (data not reported). Therefore, the polymer concentration was increased and optimized at 1:4 and 1:5 w/v for different formulations. The inlet temperature did not provide any influence on particle size by changing the temperature from 140 to 180 °C but it greatly affected yield of the resultant microspheres as revealed from the table 1. We found the optimum yield at 160 °C for mucoadhesive microspheres with respective polymers. The size of the prepared microspheres under the condition of a faster pump rate was large. It might be attributed to the formation of larger droplets during the process. There was an apparent decrease in particle size from 18.03 to 11.27 µm for CP microspheres, 17.20 to 9.47 µm for PL microspheres, 18.90 to 12.35 µm for CS microspheres when the air flow rate was increased from 5 to 15 ml/min. therefore, the processing conditions were selected from preliminary experiments as follow: inlet air temperature of 160 °C, pump setting of 10 ml/min, pressure bar at ~ 2 atm.
Encapsulation efficiency
KT loaded microspheres were produced with a high drug encapsulation efficiency (Table 2). The amount of KT encapsulated in the microspheres was in the range of 79.03 ± 3.08 and 92.19 ± 3.21%. More efficient drug loading was achieved for microspheres that were prepared from Chitosan than for those prepared from Carbopol or Polycarbophil. Moreover, a significant difference (p<0.05) in the entrapment efficiency with (SDKTS2) was observed when compared it with mucoadhesive microspheres prepared with CP and PL microspheres. Spray drying technique is generally characterized by high drug encapsulation efficiency [23].
Particle size and SEM
The particle size of prepared mucoadhesive microspheres is shown in Table 1. Mean sizes of the microspheres formulations ranged from 10.29 to 16.75 µm. Microspheres prepared with PL and CS as polymer did not show any remarkable differences in terms of size. SEM photomicrographs of KT loaded CP, PL and CS microspheres are reported in Figure 3. SEM analysis of the samples revealed that all prepared microspheres had spherical shape and similar surface morphology, regardless of the type and/or polymeric composition of spray-dried systems. The SEM images indicated towards smooth, nonporous and spherical microspheres. Some of the particles appeared to be in aggregates, but without confirmation of any collapsing particles.
Swelling study
The degree of swelling (Ssw) in mucoadhesive microspheres of drug in microspheres controlled the loading and release characteristics of prepared microspheres, hence the swelling of microspheres was evaluated as shown in Figure 3. There was no significant difference between swelling properties of KT-loaded CP and PL microspheres, due to hydrophilic nature of KT. The microspheres prepared with CP and PL reveal maximum degree of swelling of 398 % (SDKTC2) and 408 % (SDKTP1) respectively. While, maximum degree of swelling was decreased for CS microspheres to 372 % (SDKTS2). According to results of one way ANOVA test the swelling profiles of prepared mucoadhesive microspheres was found to be different (p < 0.05) at each time point.
Mucoadhesion
Mucoadhesion studies were carried out to ensure the adhesion of the microspheres to the mucosa for a prolonged period of time at the site of absorption. The results of in-vitro mucoadhesion tests, expressed as percent of attached microspheres are reported in Fig. 3. The results showed that the microspheres had good mucoadhesive properties as the percentages found in the range of 73-89% and could adequately adhere on nasal mucosa. The mucoadhesion of prepared microspheres were ranked, CS>CP>PL microspheres. Microspheres based on chitosan (SDKTS1) possessed significantly (P < 0.001) higher adhesion than CP and PL microspheres.
FTIR and DSC
The IR spectra of prepared microspheres were recorded in comparison with IR spectra of both pure KT and blank microspheres. The IR spectra of KT showed peaks at 3360, 1588 and 1278 nm representing the -COOH stretching, -C=O stretching and -C-N stretching respectively. The peaks at 1561 nm and 730 nm showed as major peaks for drug. All the above peaks were present in drug loaded microspheres that confirms the presence of drug in the polymer without any interaction.
The thermal behaviors of prepared microspheres were recorded in comparison with thermograms of both pure KT and blank microspheres. The DSC-thermogram of pure KT showed endothermic peak at 159 °C, corresponding to its melting point. KT loaded polymeric microspheres exhibited a single melting peak at 153 °C due to presence of KT in polymeric matrix. However there was slight decrease in the melting point of drug when prepared in the form of microspheres. The evaluation of the thermograms obtained from DSC revealed some interaction between the polymer and the drug in the microspheres.
In vitro diffusion study
The in vitro release profiles obtained from the drug-loaded microspheres, compared to the release profile of the drug alone is shown in fig. 4. The rate of dissolution of KT powder was significantly faster (approximately more than 98% of the drug dissolved in 2 h). The loading of KT into polymeric matrices led to prolonged dissolution/release rate. The decrease in the rate of release was dependent on the kind of polymer used and on the drug to polymer ratio. In fact, about 85-95 % of released drug was achieved up to 8 h from the spray-dried microspheres.
According to the results of one-way ANOVA, the drug release was found to be significantly different at each time level (P < 0.001) as well as among the drug products (P < 0.05) implying that the dissolution profiles were not parallel (Figure 5). According to results of, Tukey's Multiple Range test, it was found that the percentage released of prepared microspheres demonstrated statistically different (P < 0.05) at the time points after 30 min (> 20% drug release), 4 h (> 50% drug release) and 8 h (> 85% drug release). It is revealed from the results of the one-way ANOVA method that the release profiles had differing shapes, in terms of course of release and percent released, SDKTS1 demonstrated the satisfactory drug release property.
Release Kinetics
The in vitro release data obtained were fitted in to various kinetic equations. Correlations of individual batch with applied equation are given in Table 4. The release rates were determined from the slope of the appropriate plots. All the prepared microspheres showed higher correlation with Higuchi plot than zero order and first order. To find out release mechanism the in vitro release data were applied in Korsmeyer-Peppas equation. The release exponent n was determined and given in Table 4. Microspheres prepared with CP and CS demonstrated (n<0.5) fickian diffusion. While microspheres prepared with PL showed (n>0.5) anomalous (non-fickian) diffusion.
Nasal Cilia Toxicity
The batches which demonstrated a satisfactory encapsulation, mucoadhesion and drug release property from among all the prepared batches were chosen for ex vivo trials. Nasal cilio-toxicity studies were carried out in an attempt to evaluate the potential toxic effects of excipients and KT used in the formulations on the nasal mucosa. The nasal mucosa treated with phosphate buffer pH 6.6 (negative control) showed no nasocilliary damage and the nasal membrane remained intact, whereas an extensive damage to nasal mucosa coupled with loss of nasal cilia was observed with positive control. However, the application of KT on nasal mucosa showed mild nasal mucosal damage associated with loss of few nasal cilia. No apparent nasal mucosal damage was observed in nasal mucosa treated with KT loaded microspheres, thus substantiating the safety of the excipients and drug used in the formulations.
Discussion
Spray drying is an important method for the preparation of nasal microspheres. It will give rise to microspheres in which active drug will be in the matrix of the polymer. In this study mucoadhesive microspheres were prepared by using three different polymers namely, CP, PL and CS. The ideal microsphere particle size requirement for nasal delivery should range from 10 to 50 µm as smaller particles than this will enter the lungs.[25] In order to optimize the particle size and yield, we carried out spray drying process by varying the process parameters viz., inlet air temperature, pump flow rate and drug to polymer ratio.
The influence of the concentration of polymers on the particle size and yield studied at two different concentrations with the same amount of drug. Increasing the concentration of the polymers resulted in an increase in particle size. This is due to the greater amount of polymers enclosed in the same volume of a liquid droplet as the concentration of polymers is increased. These results are agreement with those of Pavenetto et al. (1994), Wagenaar and Mu¨ ller (1994). [26] A change in the inlet temperature of the equipment can affect the drying rate. Therefore, study was carried out using three levels of the inlet temperature i.e. 140, 160, 180°c and the effects were observed. The increase in the inlet temperature of the equipment shows minimal decline of the average particle size. The influence of pump rate on particle size was studied by using three levels of the pump speed i.e. 5, 10, 15 ml/min. Average particle size of microspheres is increased as flow rate is increasing. This may be due to the formation of the bigger droplets as more amount of the liquid is available for the spraying. But the yield of the product decreases with increase in the flow rate, which may be due to the loss of the product in the un-dried form due to deposition on the drying chamber walls since bigger droplets require more drying time to form microspheres. [25]
High encapsulation efficiency of the microspheres may be due to formation of more intact matrix network by the spray drying process. This slows down the diffusion of highly water soluble KT in the aqueous solution in which the microspheres are prepared. Percentage encapsulation efficiency for microspheres prepared with CP and PL as polymer was lower as compared to CS as polymer and was decreased as polymer amount increased this may be due to saturation concentration [mucoadhsive; cp glipzide and metoprlol Cp Pl]. [26,27,28]
The microspheres were uniform in size for each batch. Particle size was mainly governed by the polymer concentration. Particle size increased with increasing polymer concentration which may be due to increased viscosity of the dispersion, which affects the performance of spraying of the mixture and results in the formation of larger droplets. There were no drug particles on the surface of the microspheres, and no signs of recrystallization or aggregation were observed. The surface of the drug-free microspheres was smooth. [muco pl glipizide] [29]
Different swelling behavior of KT loaded mucoadhesive microspheres can be explained by considering the state of the polymers as well as the drug to polymer ratio. The high swelling property of CP and PL microspheres could be attributed to their ionized ability to uncoil the polymer into an extended structure. High molecular weight of CP and PL could be the possible reason for higher swelling of CP and PL microspheres than CS microspheres. However, the swelling capacity of the PL microspheres increased considerably on increasing the amount of PL, which was comparable to previous studies performed by using PL as polymer. The water uptake in hydrogels depends upon the extent of hydrodynamic free volume and availability of hydrophilic functional groups for the water to establish hydrogen bonds.[30,31,32]
Highest mucoadhesion of CS microspheres may be due to electrostatic attraction between CS and mucin. This can be evidenced from strong interaction between CS microspheres and mucous glycoprotein and/or mucosal surfaces. Mucoadhesion of CS microspheres increases because more amount of polymer results in higher amount of free -NH2 groups, which are responsible for binding with sialic acid groups in mucus membrane and thus results in increase in mucoadhesive properties of microspheres. Reduction in mucoadhesion of CP microspheres after KT incorporation could be explained by lower polymer concentration in the matrix. CP microspheres had negative charge in phosphate buffer (pH 6.6), causing negative charge repulsion with mucus, numerous hydrophilic functional groups such as carboxyl groups in CP molecules could form hydrogen bonds with mucous molecules, thus producing some adhesive force of this polymer. Mucoadhesion of PL microspheres was poor this may be due to its nonionic property and the presence of the drug molecule could prevent formation of hydrogen bonds, which are responsible for mucoadhesion. In addition type, amount, and molecular weight of polymer might have played a significant role on mucoadhesion.[33,34,35]
The effect of different polymers on the properties of microspheres can be visualized through in vitro drug release pattern of microspheres in a much better way. Swelling of microspheres is an important factor affecting the diffusion of incorporated drug. It has been found that drug diffusion in highly hydrated CP and PL microspheres is faster than that in less hydrated CS microspheres. Slow cumulative drug release from microspheres may be attributed to the increase in the density of the polymer matrix and also an increase in the diffusional path length that the drug molecules have to traverse. The polymeric gel might have acted as a barrier to penetration of the medium, thereby suppressing the diffusion of KT from the swollen polymeric matrix. It may be demonstrated that high swelling ability of KT loaded microspheres, large contact surface between swollen microspheres and small cores due to spray-drying lead to similar release profiles for all the microspheres. [36,37,38]
For all formulations no severe damage was found on the integrity of nasal mucosa (Fig. 4). The observed changes on nasal mucosa can be summarized as epithelium disruption and complete loss of some parts of the epithelium (Fig. 4). Morphological changes in the nasal epithelia exposed to microspheres were milder than those exposed to KT alone and Isopropyl alcohol (Fig. 4). The difference seems to lie in the formulation, i.e., there would be negative effects when the KT is applied in the form of a powder as opposed to when the application is in microspheric form. Also this result reveals the fact that the mucosa remains intact after the microsphere exposure and retains a good morphology. The untoward effect of KT powder may be due to its acidic structure [23,24].
