FORMULATION AND EVALUATION OF pH TRIGGERED IN SITU GELLING OPHTHALMIC SUSTAINED DRUG DELIVERY SYSTEM OF CIPROFLOXACIN HYDROCHLORIDE

Published online: 31 May 2020 To cite this article: KSHITIS CHANDRA BARAL, UTTAM BUDHATHOKI & PANNA THAPA (2020) Formulation and evaluation of pH triggered in situ gelling ophthalmic sustained drug delivery system of ciprofloxacin hydrochloride, Malaysian Journal of Pharmaceutical Sciences, 18(1): 31–59, https://doi.org/10.21315/mjps2020.18.1.3. To link to this article: https://doi.org/10.21315/mjps2020.18.1.3


INTRODUCTION
The common method of ocular drug delivery is topical administration of ophthalmic drops into the lower cul-de-sac. Such drops are outflow quickly due to the eye blinking reflux and the pre-corneal region returns to maintain resident volume of around 7 μl. The available concentration of drug in pre-corneal fluid is the driving force for passive transport of drug across the cornea.
The cornea is an optically transparent tissue and is a protective barrier (multiple layers) to the interior of the eye though it has a small surface area (~1 cm 2 ). Therefore, it is a major barrier for traditional topical drug delivery in the treatment of anterior segment diseases such as glaucoma, keratitis and bacterial and viral infections. Pre-corneal drainage, tears washout and limited contact time are major challenges to the anterior segment drug delivery upon topical administration. To be clinically effective, topical formulation has to possess a balance between the hydrophilicity and lipophilicity with higher contact time (Kansara, Hao and Mitra 2007). Design of modern ocular drug delivery systems is based on the drug application pathways, absorption mechanism and the overall ocular pharmacokinetic/ pharmacodynamic profile. Various ocular drug delivery such as hydrogels, micro-particles, nanoparticles, liposomes, insert, in situ gelling system and other colloidal systems, as well as solid inserts and shields or surgically applied polymeric implants have been proposed for prolonging the release of drugs and enhancing corneal bioavailability.
The progress has been made in gel technology for the development of droppable gel. They are liquid upon instillation and undergo phase transition in the ocular cul-de-sac to form visco-elastic gel that provides a response to environmental changes. Ideally, an in situ gelling system should be a low viscous, free flowing liquid to allow for reproducible administration to the eye as drops, gel formed following phase transition should be strong enough to withstand the shear forces in the cul-de-sac and demonstrates long residence time. Three methods, viz. change in pH, change in temperature and ion activation, have been used for phase transition in the eye surface.
In human body, there are remarkable changes in pH that can be used to direct therapeutic agents to a specific body area, tissue or cell compartment. Owing to wide variations in the pH value of the physiological fluids, sol-gel transitions induced by pH changes seem to be an ideal approach for enhancing the pharmacological efficacies of the topical drug delivery, especially ophthalmic and intravaginal applications. All the pH-sensitive polymers contain pendant acidic or basic groups that either accept or release protons in response to changes in environmental pH. Swelling of hydrogel increases as the external pH increases in case of weakly acidic (anionic) groups, but decreases if polymer contains weakly basic (cationic) groups. The most of anionic pH-sensitive polymers are based on polyacrylic acid (PAA) (Carbopol® or carbomer) or its derivatives. Likewise, polyvinylacetal diethylaminoacetate (AEA) solutions with a low viscosity at pH 4 form hydrogel at neutral pH condition. Other polymers like cellulose acetate phthalate latexes (CAP), polymethacrylic acid (PMA), polyethylene glycol (PEG) and pseudo latexes also have been used as pH sensitive system to achieve gelation (Almeida et al. 2013). In this study, authors aim to develop and evaluate pH triggered in situ gelling ophthalmic drug delivery approach to prolong the effect of ciprofloxacin hydrochloride (HCl).

Analytical Method Development
A robust high performance liquid chromatographic (HPLC) based analytical method has been developed and validated for the quantification of ciprofloxacin HCl in pH triggered in situ gelling formulation based on validation parameters viz. linearity, specificity, accuracy, repeatability, recovery, robustness, limit of quantification and limit of detection as per Q2B validation of analytical procedures (ICH 1996).

Chromatographic Method
The chromatographic analysis of the drug was carried out in HPLC (Shimadzu Prominence ® ultrafast liquid chromatography [UFLC] Nexera liquid chromatograph) with the conditions as in Table 1.

Buffer System Selection
In order to ensure product stability and desired drug solubility for ciprofloxacin HCl in in situ gelling system, various buffer systems viz. acetate buffer IP (pH 4.0, 4.5, 5.0, 5.5, 6.0), phosphate buffer USP (pH 4.0, 5.0, 6.0, 6.5), citrophosphate buffer BP (pH 4.0, 5.0, 6.0, 6.5) were used. The solubility of drug in these buffer systems was carried out at a concentration of 0.3% w/v.

Drug-Excipients Compatibility Studies
Compatibility studies were carried out by isothermal stress testing (IST) technique. In this technique, drug and different excipients were weighed in unimolar ratio and then was transferred accurately into glass vials (5 ml; n = 2). The composite systems were mixed on a vortex mixer for 2 minutes. Then water (10%) was added to each of the vials and the drugexcipient blend was further mixed with a both end heat-sealed glass capillary. Each vial was sealed using a teflon-lined screw cap and stored at 50 ± 0.5°C in hot air oven. Similarly, control sample were prepared and stored under refrigerated condition for comparative study.
The samples were periodically examined for any unusual colour changes. After three weeks of storage under above condition, samples were quantitatively analysed using validated HPLC method (Table 1) with reference to refrigerated control sample and qualitatively by Fourier transform infrared spectroscopy (FTIR) (Dong and Choi 2008;Nihar, Lila and Sujata 2011;Kannan 2013). The qualitative data from FTIR and quantitative data from HPLC were compared for drug-excipient compatibility study.

FTIR Study of Ciprofloxacin HCl
Ciprofloxacin HCl (10 mg) was diluted with potassium bromide to get around 1000 mg of mixture and triturated to ensure sample homogeneity. For each sample, FTIR spectra and purity index were recorded in the wave number ranged between 2000 cm −1 and 400 cm −1 averaging 45 scans per sample using a resolution of 3.86 cm −1 (FTIR-21 prestige spectrophotometer, Shimadzu). The IR solution software was used for data collection and qualitative study.

Preparation of Resinate (Drug-resin Complex)
In this technique, the activated resins (Indion® 214 and 254) were dried and then grounded to fine powder; typically in the range of 40 µm-150 µm. Resins were mixed with a drug solution by batch process by allowing sufficient time for drug loading and equilibrium of ion exchange process. The resin/fluid slurry was filtered, filtrate washed and then dried in an oven at 60 ± 0.5°C (Jain et al. 2008).

Drug Loading Efficiency Study in Resinates
Ion exchange activated resins (Indion® 214 and 254) were separately dispersed uniformly in water using an overhead stirrer for 15 minutes in resin: drug equals to 1:1 and 2:1. The ciprofloxacin HCl was then added to the dispersion while stirring and this continued until equilibrium of resin and complex were achieved. Periodically sample were withdrawn and centrifuged at 2000 rpm (REMI-R-8CBL centrifuge). The spectrum of supernatant was scanned in UV-1800 and λmax was observed at 274 nm which was used for quantification of uncomplexed ciprofloxacin HCl using Equation (1).

Preparation Method of In situ Gelling System
Buffer salt, sodium acetate trihydrate, disodium edetate and BKC were dissolved in water (75 ml). Then, HPMCK15M was added to the above solution and allowed to hydrate for 30 minutes. Carbopol 934p was dispersed in the solution and allowed to hydrate overnight with stirring. Resinate equivalent to 0.3% w/v of ciprofloxacin was added to above solution under stirring in order to make uniform suspension of resinate. The pH of the system was adjusted to 4.5 ± 0.2 using glacial acetic acid/0.5 N NaOH. After pH adjustment, prepared formulation were dispensed in amber glass vials (5 ml), closed with grey butyl rubber closures and sealed with aluminum caps. Finally formulations were terminally sterilised in autoclave (Thermolab) at 121.5°C (15 psi) for 20 minutes (Jain et al. 2008).

pH of the formulation
The pH of the formulation was measured by calibrated pH meter (Hanna).

In situ gelling capacity
The gelling capacity of formulation was assessed by placing a drop of the system in a vial containing 1 ml of artificial tear fluid (ATF), freshly prepared with maintaining temperature at 37 ± 0.5°C under Neo lab heating block system. The composition of the ATF contains (NaCl = 0.670 g, NaHCO 3 = 0.20 g, CaCl 2 .2H 2 0 = 0.008 g) with purified water quantum satis (qs) 100 ml (Gupta et al. 2007;Jain et al. 2008).

Assay
The assay of ciprofloxacin HCL in in situ gelling system was carried by validated HPLC method (Table 1).

Standard preparation
Ciprofloxacin HCl RS (equivalent to 60 mg) was taken into volumetric flask (100 ml) and distilled water (60 ml) was added. The solution was sonicated for 10 minutes followed by shaking for 15 minutes in mega stirrer at 500 rpm. The volume was made up to 100 ml with distilled water. The reference standard solution was diluted further with 0.1 M HCL in order to make final concentration of 60 µg/ml.

Sample preparation
In order to estimate ciprofloxacin content in formulation, 2 ml of in situ gel formulation was pipetted accurately and then transferred into volumetric flask (100 ml). The resinate in the formulation was dissolved in 0.1 M HCl in ultrasonic bath for 30 minutes followed by shaking for 30 minutes in mega stirrer at 500 rpm. The volume was made up to 100 ml with 0.1 M HCl. The solution was filtered through Whatman No.1 discarding first few filtrates.

Procedure
Each of standard reference and sample solutions (25 µl) were injected in UFLC-Nexera with chromatographic condition (

In vitro release study
In vitro evaluations of formulation were carried out with modification of USP type-I dissolution apparatus. The lower portion of basket was wrapped with Spectra/por® 7 (cellophane membrane). The in vitro release study was carried out using ATF as a medium (100 ml) at 50 rpm for 8 hours. Spectra/Por ® 7 (cellophane membranes) was activated by soaking in distilled water for 30 minutes and then overnight in ATF before wrapping at the base of basket (Bottari et al.1974;Desai and Blanchard 1998).

Standard preparation
Ciprofloxacin HCl RS equivalent to 30 mg of ciprofloxacin was taken in a volumetric flask (100 ml) and ATF (60 ml) was added. The solution was sonicated for 10 minutes followed by shaking for 15 minutes in mega stirrer at 500 rpm. The volume was made up to 100 ml with same medium. The reference standard solution was diluted further with medium in order to make final drug concentration of 30 µg/ml.

Sample preparation
The formulated sample (1 ml) was placed in basket wrapped with cellophane membrane on basement surface. In vitro study was assessed by withdrawing sample (1 ml) at an interval of 1 hour for 8 hours with replenishment by ATF under thermostatically controlled hydrodynamics condition.

Rheological Study
The rheological study of the formulation were determined using the spindle (S64) at 25 ± 1°C for both sol (pH = 4.5 ± 0.2) and gel phase with increase in pH of the system. The developed formulations (pH = 4.5 ± 0.2) were poured into the beaker, the angular velocity was increased gradually from 1 rpm to 100 rpm at different angular velocities (1, 2, 2.5, 4, 5, 10, 20, 50,100 rpm). The angular velocity was reversed and average viscosity was calculated. The same formulation were again evaluated for viscosity determination in a similar fashion with increase in pH of the system (pH = 7.4 or above) by addition of 0.5 M NaOH using a Brookfield programmable DV-II+ Viscometer. The viscosity measured at both the conditions versus angular velocity was plotted for selected formulations to study rheological behaviour (Srividya, Cardova and Amin 2001).

Ocular Irritation Test
Ocular irritation study was carried out by modified hen's egg test or Huhner-embroynentest-chorioallantoic membrane (HET-CAM) test. Fertilised hen's eggs weighing between 50 g and 60 g were obtained from local poultry farm. They were candled to discard the defective ones. The selected eggs were incubated in a humidified incubator at a temperature of 37 ± 0.5°C for three days. The trays containing eggs were rotated manually in a gentle manner after every 12 hours. On day 3, egg albumin (3 ml) was removed by using sterile techniques from the pointed end of the egg. The hole was sealed by 70% alcohol sterilised parafilm with the help of sterilised spatula. The eggs were kept in the equatorial position for the development of chorioallantoic membrane (CAM) away from the shell. The eggs were candled on the 5th day of incubation, and every day thereafter non-viable embryos were removed. On 10th day, a window (2 cm × 2 cm) was made on the equator of the eggs through which formulations (0.5 ml) were instilled. A 0.9% NaCl solution and dioctyl sodium sulphosuccinate were used as negative and positive controls, respectively (Velpandian et al. 2006).

Antimicrobial Efficacy Testing
The antimicrobial efficacy test of formulated in situ gelling system was carried out using materials mentioned in Table 2.

Preparation of Muller Hinton Agar Media (MHA) and MHA Petri Plates
MHA (38 g) was suspended in 1000 ml distilled water and then heated gently to dissolve the medium completely. The dissolve media was then autoclaved at 15 lbs (121°C) for 15 minutes. After autoclaving, 25 ml of MHA media was poured in sterilised eight petri plates under horizontal laminar flow cabinet with avoiding air entrapment. The MHA in petri plates were solidified for the purpose of inoculums culture on its surface (Mueller and Hinton 1941).

Inoculums Preparation
In order to prepare inoculums, strain of Staphylococcus aureus (ATCC 25903) and Pseudomonas aeruginosa (ATCC 10145) were sub-cultured in soybean-casein digest agar media (SCDA). SCDA (40 g) in distilled water (1000 ml) and then heated gently to dissolve the medium completely. The dissolve media was then sterilised by moist heat sterilisation (autoclave) at 15 lbs (121°C) for 15 minutes. The strain microorganisms were sub-cultured in SCDA slant media with the help of sterilised wire loop. Finally, slants were incubated at 36 ± 1°C for 2-3 days. After 2-3 days, the media was found turbid due to growth of microorganisms (Bauer et al. 1966).

Inoculate Transfer to MHA Petri Plates
After adequate microbial growth of sub-cultured strain of Staphylococcus aureus (ATCC 25903) and Pseudomonas aeruginosa (ATCC 10145) in SCDA, each was homogenously mixed in vortex mixer for 2-3 minutes. The turbidity of liquid suspension was compared with 0. dispensed over MHA plates separately for both strains of bacteria aseptically. The MHA plates were incubated at 36 ± 1°C for two days for the antibiotics inhibitory effect study. The antimicrobial efficacy testing of optimised batch was compared with marketed ciprofloxacin eye drop (Microflox) on the basis of diameter of zone of inhibition (Bauer et al. 1966;Murray and Baron 2007).

Formulation Optimisation
Contour plot and response surface methodology (RSM) were used to optimise the formulations based on polynomial equation (Equation 4) that show interaction and quadratic terms for release of drug from gelling system generated for all the response variables using multiple linear regression analysis (MLRA) approach.
where, Y = dependent variable B 0 = intercept representing the arithmetic average of thirteen batches B 1 = estimated coefficient for factor X 1 X 1 and X 2 = coded levels of the independent variables

Mathematical Modeling for Drug Release Kinetics
The following models have been tested for release data of the formulations:

Zero order kinetics
Zero order kinetics is the one whose rate is independent of the concentration of drug undergoing reaction (Equation 5) (Murray and Baron 2007).
where, C = concentration of drug at time = t k 0 = zero order release constant

First order kinetics
First order kinetics is the one whose rate is directly proportional to the concentration of drug undergoing reaction. It is also called linear order kinetics because of proportionality between rate of reaction and the concentration of drug (Equation 6),

Higuchi equation
Higuchi square root model is based on Fick's laws of diffusion and is applicable to porous hydrophobic drug delivery system in homogenous and granular matrices. The model is expressed as Equation (7), where Q = amount of drug released in time t by surface unity C = initial concentration of the drug ε = matrix porosity τ = tortuosity factor of the capillary system C s = drug solubility in the matrix D = diffusion constant of the drug molecules in that liquid

Korsmeyer-Peppas model (power law)
The semi-empirical equation developed by Korsmeyer and Peppas in 1983 is based on the Fick's laws of diffusion. It can be used to analyse data of controlled release water soluble drugs from polymers. According to this model drug release kinetics, the fractional release of drug is exponentially related to release time, and is expressed as Equations (8)  Under some experimental situations release mechanism deviated from the Fick equation and follows an anomalous behaviour (non-Fickian). In these cases more generic equation can be used as:

Similarity factor
Similarity between the two products is assessed by using similarity factor (F s ). F s is a logarithmic transformation of the sum-squared error of differences between the test T j and reference products R j over all points (Equation 10).

× 100 (10) n
Where n is the sampling number, R j and T j are the % dissolved of reference and the test products at each time points j, respectively. f s value higher than 50 and close to 100 show the similarity of the dissolution profiles.

Dissimilarity factor
The difference factor (F d ) measures the percent error between two curves over all time points (Equation 11).
The percentage error is zero when the test and drug reference profiles are identical and increase proportionally with the dissimilarity between the two dissolution profiles. The F d values should be close to 0 to be similar.

Specificity
The developed analytical procedure had completely discriminates from its placebo with prominent peak at 278 nm having signal to noise (S/N) ratio of 100:1. The respective noise and drift upon 25 µl placebo injection were 141.85 and −1207.96, respectively. Similarly, the respective noise and drift upon 25 µl analyte injection were 382.36 and −165638.99, respectively.

Accuracy
The analytical method was found to be accurate across the different concentration level within linearity range. Similarly, the recovery of analyte at below and above 20% from assay level was within limit. The relative standard and standard error were below 2% (table not shown).

Precision
The procedure was found to be precise at assay level upon repetitive performance at different time interval within a day (intermediate precision) and day after (interday precision). The relative standard deviations at different conditions were below 2% indicated that process had precision (table not shown).

Detection limit
Based on standard deviation of responses and slope of calibration curve (figure not shown) the detection limit was found to be 0.1427 µg/ml.

Quantification limit
Based on standard deviation of responses and slope of calibration curve (figure not shown), the quantification limit was found to be 0.4759 µg/ml.

Robustness
As the % RSD of robustness below 2%, the robustness result upon deliberate variation of pH in mobile phase (± 0.05 unit), oven temperature (± 5°C), and organic solvent (± 5%), different columns (different lots and/or suppliers) were found within the limit (tables not shown) and hence the method had robustness.

Buffer system selection
Out of three buffer systems of different pH, acetate buffer IP (pH 4.5) (solubility = 50.59 mg/ ml) was selected for formulation.

FTIR compatibility study
The drug-excipient composite stored under respective condition as mention in Table 3 were quantified by reverse phase (RP)-HPLC method. The drug content under both control and stressed conditions was found within a range of 97.09% to 101.42%. The quantity of drug remaining under stressed condition of 10% moisture at 50°C with reference to control sample suggested that there were no significant drug degradation phenomenon.
Similarly, FTIR spectra of Carbopol 934p which showed the prominent peak between 1750 cm −1 and 1700 cm −1 was assigned to carbonyl C=O stretching band i.e. C=O while the peak at 1450 cm −1 to 1400 cm −1 was assigned to U C-O /δ O-H. The band at 1250 cm −1 to 1200 cm −1 was assigned to U C-O-C of acrylates. The ethereal cross linking is indicated by the prominent peak at 1160 cm −1 , represented a stretching vibration of U C-O-C group. The band between 850 cm −1 and 800 cm −1 indicated out of plane bending of C=CH, i.e. δ=C-H. The absorbance band of ciprofloxacin-Carbopol 934p at 1650 cm −1 to 1600 cm −1 was assigned to O-C=O i.e. carbonyl asymmetric stretching vibration. A prominent peak at 1450 cm −1 was for O-C-O group of symmetric stretching vibration. The band from 1300 cm −1 to 1250 cm −1 was assigned to C-O-C stretching vibration of acrylates and esters. The peak between 1100 cm −1 and 1000 cm −1 represented C-F groups while the band at 800 cm −1 indicated the meta distribution of δ-Ar-H group. Ciprofloxacin/EDTA composite was shown with their absorbance peak band within a wave number range between 2000 cm −1 and 400 cm −1 [ Figure 1(g)].
In the FTIR spectra of the composites, it is obvious that the band position of C=O group was affected by esterification. Furthermore, conjugation involving C=O group lowering the C=O frequency between 1650 cm −1 and 1600 cm −1 might be due to the formation of β-ketoesters. The FTIR peaks assigned to U C-O and U C-O-C representing acrylates and esters confirm the esterification between polymeric OH group and -COOH group of the drug (ciprofloxacin). The stretching vibration of C-F group remained nearly unaltered. The bending vibration of O-H group gives medium to strong bands in the region around 1450 cm −1 . The FTIR peak at 800 cm −1 gives the probability of out of plane bending of -ene bond and m-substitution of δ Ar-H hydrogen atom. Both spectra showed prominent peaks for the stretching vibration of O-C-O and C=O groups, which prove the formation of esters between the drug and polymer. Both intermolecular and polymeric hydrogen bonding are also evident from the spectra of the polymeric composites. Thus, it may be predicted from FTIR analysis that the increases in atomic densities in the particular plane as well as the small change in orientation of the crystal lattice of the composites were due to formation of the esters and intermolecular hydrogen bonding between the carboxylic group of ciprofloxacin and carbopol polymers. FTIR analyses indicated that although there are intermolecular hydrogen bondings and esterification between ciprofloxacin and Carbopol 934p, ciprofloxacin retained its crystallinity in carbopol composites. The purity index of each composite system was observed above 0.95 with reference control standard (Table 5) and overlapping the FTIR spectra with each other suggested towards correlation between them (Nihar, Lila and Sujata 2011).

Pre-treatment of Drug
Ciprofloxacin HCl exhibited instantaneous incompatibility with Carbopol 934p due to its acidic nature which turns the solution towards lower pH range, resulting in a lumpy precipitate. Drug entrapment within a resin system has been evaluated spectrophotometrically at different time interval based on Equation 1. The amount of drug loading was found higher in Indion ® 254 than Indion ® 214 at resin: drug equals 1:1 (91.84%) and 2:1 (95.30%). Based on drug loading efficiency, Indion ® 254 was selected. The optimised resin (Indion ® 254):drug was found to be 2.1:1.

FTIR Study of Resinates
The FTIR spectra of resinate (Indion ® 254: ciprofloxacin HCl = 2:1) and ciprofloxacin HCl exhibited similar absorbance bands, peak height, correction height, base height, base length, area and the corrected areas (Table 6). Similarly peak purity index of resinate (drug: resin; 1:2) was found to be 0.99 with overlapping of its peculiar functional groups with reference to standard ciprofloxacin HCl which suggest complete complexation phenomenon within binary mixture [ Figure 1(h)].

Physicochemical Characterisation
The physiochemical characterisation of in situ gelling system was carried out based on pH and in situ gelling capacity. The pH of the 13 formulations of central composite design was observed within range of 4.44 to 4.55.
Similarly, the in situ gelling capacity of the formulations was observed in a thermostatically maintained ATF in a vial. Out of the 13 formulations, F1, F6 and F11 had shown strong gel forming capacity. The gel formation was observed within a minutes and it remained for more than eight hours. F3, F4, F5, F8, F10, F12, and F13 had shown gel formation after few minutes and remained for 6-8 hours and F2 and F7 had formed gel after few minutes and the developed gel was dissolved within one hour. F2 had no gel forming capacity.
The drug content in developed and optimised formulation was found from 96.82% to 104.45%. The assay values of all the formulations are given in Table 4.

100.05
Notes: pH of the final system was adjusted to 4.50 ± 0.2 with glacial acid and 0.5 M NaOH.

Rheological Study
The viscosity of developed formulations at pH 4.5 ± 0.2 (before gel formation) and at higher pH range (pH ≥ 7.40) were evaluated with using spindle S64. The sol and gel phase of all the formulations showed shear thinning and pseudoplastic flow behaviour, which is desirable for ophthalmic formulations in order to prolong contact time with corneal membrane. The administration of ophthalmic preparations should have minimum influence over pseudoplastic character of pre-corneal tear film. As ocular shear rate is very large ranging from 0.03 s−1 during interblinking to 4250 s−1 -28500 s−1 during blinking, viscoelastic fluids with a higher viscosity at low shear rate and low under high shear rate are desirable. All the formulations were found to be in liquid phase at pH 4.5 ± 0.2. Increase in the viscosity was observed when the pH of the formulations was raised to 7.4 which transform into gel phase. Similarly, the viscosity (cps) reading was decreased when angular velocity (shear rate) increased in both sol and gel phase and at higher angular velocities, viscosities are nearly constant at both pH state of system ( Figure 2).

Figure 2:
Rheological study of optimised formulation.

In vitro Release Study
Cumulative percentage drug release versus time of 13 formulations is depicted in Figure 3. Carbopol 934p is tightly coiled acidic molecules. Once dispersed in water, the molecules begin to hydrate and loose its coil structure. The polymers achieve maximum thickening behaviour upon conversion to a salt. Such behaviour can easily achieved by neutralising the polymers towards higher pH ranges with the use of common bases like NaOH or by other neutraliser. As the polymer has acidic behaviour having pH in the range of 2.5-3.5, it turns the pH of the solution towards lower range. As the pH of ciprofloxacin HCl is in the range of 3.5-4.5, the acidic carboxylic moieties of the polymer further lower the pH. As drug beyond its range may generate lumpy mass, therefore during addition of polymer the buffer system should be properly adjusted to avoid pH decrease. The instantaneous lumpy mass has been avoided by pre-treatment of drug prior addition. The modification has been carried out by entrapment of Indion ® 254 which is strongly cationic. The release of drug from resinate is based on ion-drug complex dissociation in presence of counter ion. The amount of resin in resinate also plays a pivotal role for drug release, as it plays rate limiting action. Similarly, the neutralising capacity of carbopol based formulation basically depends upon its concentration. The ratio of neutralising agent with carbopol should be considered during formulation. The non-neutralised dispersions of carbopol have very lower viscosity at a range of 4.5 and higher at neutral pH. As tear fluid has pH 7.40 which is neutral even itself, the triggering of physiochemical parameters phase transformation takes place from sol-gel by pH sensitive polymers like Carbopol 934p. Besides viscoelastic behaviour of polymer and phase transition feature, the release rate from drug-gelling matrix system should be considered. As Carbopol 934p has dominantly gelling property, the developed gel Malay J Pharm Sci, Vol. 18, No. 1 (2020): 31-59 gradually gets dissolved in tear fluid. The addition of hydrophilic polymer like HPMCK15M acts as release retardant. Similarly, considering upon weak buffering capacity of tear fluid and slightly acidic pH (4.50 ± 0.2), other additives should be added in formulation in order to mask ocular discomfort. The added HPMCK15M also minimises the pH based irritation with its lubricating action within cul-de-sac region.

Ocular Irritancy Test
F optimized had shown mild degree of irritation having average score 0.33, 0.66 and 0.66 at 6th, 7th and 8th hour, respectively in HET-CAM technique. Since there was no high degree of irritation for a period of 8 hours, it has ocular comfort during delivery. The scoring pattern and obtained score of irritancy is depicted in Tables 5 and 6 where, X 1 = Carbopol 934p X 2 = HPMCK15M

Drug Release Kinetics
Out of 13 formulations, F1, F3, F4, F5, F6, F9 and F11 followed first order kinetics, F8 and F10 followed Higuchi model and F2, F7, F9, F12 and F optimized followed Krosmeyer-Pepas as model. Carbopol 934p showed dominant effect against drug release over its gelling capacity. Swelling of Carbopol 934p in ATF could be due to neutral pH environment. Carbopol 934p causes ionic repulsion of polymer due to presence of electrolyte in tear fluid which is manifested on the macro level as swelling (Durrani et al. 1994). The hydration of polymer resulted in rapid decrease in its glass transition temperature to the temperature of the medium.
Microscopically, there is a relaxation response of the polymer chains due to stresses introduced by the presence of the dissolution medium. This results in an increase in the radius of gyration and end-to-end distances of the polymer chains, causing a significant increase in the molecular volume of the hydrated polymer (Ranga-Rao and Devi 1988). This reduces the free volume due to presence of the micropores, which may manifest itself as a shift in the drug release mechanism. Increasing the amount of Carbopol 934p in the formulations resulted in reduction in the drug release rate with swelling controlled mechanism.
Similarly, HPMCK15M takes up water and polymer starts hydrating to form a gel layer. An initial burst of soluble drug may occur due to surface leaching when a composite containing a swellable glassy polymer comes in contact with an aqueous medium. There is an abrupt change from a glassy to a rubbery state which is associated with swelling process with time and water infiltrated deep into the case increasing the thickness by the gel layer. After its hydration, it follows states of dissolving or eroding. When water reaches the center of the system and the concentration of drug falls below the solubility value, the release rate of drug begins to reduce. At the same time, an increase in thickness of the barrier layer with time increases the diffusion path length, reducing the rate of drug release. Drug release kinetic associated with these gels layer dynamic, range initially from Fickian to anomalous (non-Fickian). In general, two major factors control the drug release from swelling controlled system by a relaxation process (hydration, gelatin or swelling) followed by erosion. The optimised formulation followed Korsmeyer-Peppas equation (R 2 = 0.985) and diffusion exponent 0.874 that indicate non-Fickian diffusion in gelling system. As a result of these simultaneous processes, two front are evident, a swelling front, where the polymer get hydrated and an eroding front. The distance between these two fronts are called diffusion layer thickness. Diffusion layer thickness depends on the selective rate at which the swelling and eroding fronts move in relation to each other.

Similarity and Dissimilarity Factor
The similarity and dissimilarity study of drug release was carried out between predicted values obtained by using regression equations 12, 13 and 14 and observed values. Similarity factor F s (87.01) and the dissimilarity factor F d (1.81) showed there is similarity between observed and predicted values.

Drug Release Comparison of Conventional Marketed Eye-Drop (Microflox) with F optimized in situ Gelling System
The drug release of optimised in situ gelling formulation was compared with Microflox purchased from retail pharmacy. The in vitro drug release study was carried out in modified dissolution apparatus through cellophane membrane. The cumulative percentage drug release of F optimized and eye drop versus time (minute) plot ( Figure 6) showed that Microflox took two hours to release the total drug but F optimized took 8 hours as Microflox did not contain rate retarding polymers.

Time (Minutes)
Microflox Eye Drop F optimize Cumulative % drug release Figure 6: Comparison of in vitro release of F optimized in situ gelling system with conventional eye-drop (Microflox).

CONCLUSION
A significant challenge to the formulator in ophthalmic drug delivery is to circumvent the protective barriers of the eye without causing permanent tissue damage. Due to unique physiological anatomy of eye and its underlying barriers, ophthalmic drug delivery system still seems to be challenging. The lower residence contact time of drug, higher tear turnover, limited area of contact and impermeability of corneal epithelium layer make lower bioavailability of conventional dosage form. Phase changing polymer can help to overcome these limitations to enhance bioavailability. This study indicates use of Carbopol 934P and HPMCK15M for in situ gelling formulation of ciprofloxacin HCl.