Development of diclofenac sodium-loaded alginate-PVP K 30 microbeads using central composite design.

BACKGROUND AND THE PURPOSE OF THE STUDY: Diclofenac sodium is a non-steroidal anti-inflammatory agent with a short biological half-life (1-2 hr) and requires multiple dosing. This research was carried out to develop and optimize diclofenac sodium loaded alginate-PVP K 30 microbeads to eliminate the need for multiple dosing and adverse effects.
METHODS: Diclofenac sodium loaded alginate-PVP K 30 microbeads were prepared by ionotropic gelation. Particle size, drug release, swelling, FTIR and SEM analyses were performed.
RESULTS: Optimized microbeads showed particle size of 0.589±0.054 to 0.620±0.067 mm, and drug entrapment efficiency of 97.88±2.86 to 98.60±3.55%. The in vitro drug release from microbeads was sustained over 10 hrs and followed controlled-release pattern. FTIR analysis indicated the possibility of intermolecular hydrogen bonding interactions, i.e., -OH…O=C in microbeads.
CONCLUSION: Microbeads for oral controlled delivery of diclofenac sodium were successfully developed by ionotropic gelation.

INTRODUCTION

Alginate, the monovalent form of alginic acid, belong to the family of linear copolymers composed of β-D- mannuronic acid monomers (M), regions of ∞-L- guluronic acid residues (G), and regions of interspersed M and G units (1). It is used as matrix material in various formulations due to its hydrogel-forming properties (2). Alginates undergo gelation due to ionic interaction between carboxylic acid groups located on polymer backbone and these cations like Ca2+, Al3+, etc (3, 4). Various drugs have been successfully incorporated in alginate hydrogels and have exhibited different drug release profiles (1, 4–6).

Diclofenac sodium (DS) is a non-steroidal anti-inflammatory agent widely used as analgesic (7). Its biological half-life is 1–2 hr (2) and requires multiple dosing to maintain therapeutic concentration in blood. Hence, the controlled release systems of DS can eliminate the need for multiple dosing and adverse effects. Several investigations on formulation of DS-loaded alginate-based microparticles/beads using different polymer-blend have been reported (7–10). However, no attempt has been taken to formulate DS-loaded bead system using alginate-polyvinyl pyrrolidone (PVP) blend. Therefore in the present investigation, an attempt was made to develope and optimize DS-loaded alginate-PVP K 30 microbeads. The effects of ratios of sodium alginate to PVP K 30 and CaCl2 (cross-linker) concentrations on drug entrapment and release were analyzed using central composite design (CCD).

MATERIAL AND METHODS

Materials

DS (Techno Remedies, India), calcium chloride, sodium alginate and PVP K 30 (Loba Chemie, India) were used in this study. Other chemicals used were of analytical grades.

Microbead preparation

DS-loaded alginate-PVP K 30 microbeads were prepared by ionotropic gelation (2). Sodium alginate and PVP K 30 aqueous solutions were mixed together. Then, DS was added to the mixture and homogenized (for 10 min, 1000 rpm). The resulting mixture (drug : polymer=1:2) was dropped into CaCl2 solution via 26-gauge needle. After 15 min, beads were collected by decantation, washed repeatedly with deionized water and dried at 45°C for 12 hrs.

DS-loaded alginate microbeads were prepared by the same method but without addition of PVP K 30.

Experimental design

The experimental factors were selected as polymer-blend ratio and CaCl2 concentration; while responses were Drug Entrapment Efficiency (DEE) and drug release after 10 hrs (R10hr). The factors and levels are reported in table 1. The quadratic model was used to evaluate responses using Design-Expert® Software according to CCD (11):

Table 1

Factors and levels of the circumscribed central composite design

	Experimental Settings
Normalized levels	SA:PVP K 30a (X1)	CaCl2 (% w/v) (X2)
−1.414	1.00	1.00
−1	1.50	2.50
0	2.80	6.30
1	4.00	10.00
1.414	4.50	11.60

a SA:PVP K 30=Sodium alginate to polyvinyl pyrrolidone K 30 ratio

where, Y=response; b0=intercept, and b1-b5=regression coefficients. X1 and X2=individual effects; X12 and X22=quadratic effects; X1X2=interaction effect.

Determination of the Drug Entrapment Efficiency (DEE)

Drug-loaded microbeads (100 mg) were dispersed in 500 ml of phosphate buffer of pH 7.4 and sonicated for 10 min. The mixture was stirred at 1000 rpm for 4 hrs and filtered through Whatman® filter paper. Drug contents in filtrates were determined using UV-VIS spectrophotometer (Shimadzu, Japan) at 276 nm wavelength. The DEE of beads was calculated using following formula:

Particle size and morphological analyses

Microbeads were measured for particle size using optical microscope (Olympus, Japan). Surface morphology was examined by scanning electron microscopy (SEM) (Hitachi, Japan).

Fourier transform infrared (FTIR) spectroscopy

Spectra were obtained by FTIR spectrophotometer (Perkin-Elmer, USA) in the wavelength region of 4000-400 cm−1 using pellets of KBr.

In vitro drug release studies

Dissolution apparatus USP (Campbell Electronics, India) was employed to study the drug release at 37°C 50 rpm. Microbeads containing DS equivalent to 100 mg were taken in 900 ml of dissolution medium (0.1 N HCl of pH 1.2 for 2 hrs and phosphate buffer of pH 7.4 for next hrs). Drug content in samples were measured using UV-VIS spectrophotometer (Shimadzu, Japan) at 276 nm wavelength.

Swelling evaluation

Microbeads (100 mg) were soaked in phosphate buffer of pH 7.4 and 0.1 N HCl of pH 1.2. Swelled microbeads were removed and weighed. Swelling indexes were determined using the formula:

RESULTS AND DISCUSSION

DS-loaded alginate-PVP K 30 microbeads were prepared by ionotropic gelation according to CCD (Table 2). Quadratic models were selected as better-fit models due to smaller PRESS, and insignificant lack of fit (Table 3). Results of ANOVA indicated all models were significant (Table 4). After eliminating non-significant (p>0.05) terms, the models became:

Table 2

Experimental plan and observed response values from randomized run in central composite design.

	Normalized levels of factors		Responsesa
Experimental formulations	SA:PVP K30b (X1)	CaCl2 (% w/v) (X2)	DEE (%)c	R10hr (%)d
SP-1	−1	−1	75.32±2.14	94.44±3.45
SP-2	−1	1	98.77±2.76	66.32±1.08
SP-3	1	−1	60.33±1.64	99.56±3.45
SP-4	1	1	87.87±2.23	78.92±2.40
SP-5	−1.414	0	94.82±3.42	77.98±2.33
SP-6	1.414	0	77.68±1.96	92.36±3.02
SP-7	0	−1.414	58.59±1.12	99.64±3.42
SP-8	0	1.414	98.88±2.25	67.31±1.88
SP-9	0	0	89.35±2.16	85.87±2.43
SP-10	0	0	88.82±2.07	85.07±2.11
SP-11	0	0	89.20±2.96	85.22±2.40
SP-12	0	0	90.63±2.88	85.22±2.23
SP-13	0	0	88.43±2.13	84.95±2.17

Observed response values: Mean±SD (n=3);

SA:PVP K30=Sodium alginate to polyvinyl pyrrolidone K 30 ratio;

DEE=Drug entrapment efficiency (%);

R10hr=% Drug released in 10 hrs

Table 3

Summary of the model analysis (A), lack of fit (B), and R2 analysis (C) for the measured responses.

Source	DEE (%)		R10hr(%)
	Sum of squares	p-value	Sum of squares	p-value
(a) Model analysis
Mean vs Total	92855.36	93597.19
Linear vs Mean	1771.28	< 0.0001	1296.88	< 0.0001
2FIc vs Linear	4.18	0.7105	13.99	0.0028
Quadratic vs 2FI	241.45	< 0.0001	4.57	0.0395
Cubic vs Quadratic	4.82	0.3780	2.01	0.0635
Residual	10.14	1.00
Total	94887.24	94915.63
(b) Lack of fit
Linear	257.83	0.0007	21.04	0.0034
2FI	253.65	0.0005	7.05	0.0195
Quadratic	12.20	0.0600	2.49	0.0533
Cubic	7.37	0.0310	0.48	0.1293
Pure error	2.77	0.52
(c) R2 analysis	Adjusted		Predicted		Adjusted		Predicted
	R2	R2	R2	PRESSd	R2	R2	R2	PRESSd
Linear	0.8717	0.8461	0.7762	454.69	0.9836	0.9804	0.9651	45.98
2FI	0.8738	0.8317	0.6734	663.68	0.9943	0.9923	0.9860	18.52
Quadratic	0.9926	0.9874	0.9552	91.05	0.9977	0.9961	0.9860	18.49
Cubic	0.9950	0.9880	0.7656	476.18	0.9992	0.9982	0.9763	31.24

DEE=Drug entrapment efficiency(%);

R10hr =% Drug released in 10 hrs;

2FI=Two factor interaction;

PRESS=predicted residual sum of squares.

Table 4

Summary of ANOVA for the response parameters

Source	Sum of squares	Degree of freedom	Mean square	Fvalue	p-value Prob>F
(A) For DEE (%)a
Model	2016.91	5	403.38	188.70	< 0.0001
X1	314.12	1	314.12	146.94	< 0.0001
X2	1457.15	1	1457.15	681.64	< 0.0001
X1X2	4.18	1	4.18	1.96	0.2046
X12	27.77	1	27.77	12.99	0.0087
X22	230.44	1	230.44	107.80	< 0.0001
(B) For R10hr (%)b
Model	1315.43	5	263.09	612.17	< 0.0001
X1	181.04	1	181.04	421.25	< 0.0001
X2	1115.84	1	1115.84	2596.44	<0.0001
X1X2	13.99	1	13.99	32.55	0.0007
X12	0.02	1	0.02	0.05	0.8376
X22	4.3	1	4.39	10.22	0.0151

DEE=Drug entrapment efficiency (%); R10hr=% Drug released in 10 hrs.

X1 and X2 represent the factors; X1 2 and X2 2 are the quadratic effect; X1X2 is the interaction effect.

Three-dimensional response surface (Fig. 1), and contour plots (Fig. 2) demonstrate changes in DEE, and R10hr due to variation of factors. For optimization, numerical analysis was performed by restricting desirable ranges to 95 ≤ DEE ≤ 100%, and 60 ≤ R10hr ≤ 65%. The optimized microbeads were formulated and evaluated (Table 5). Models to produce optimized responses were well fitted (R2=0.9912 for DEE, and 0.9967 for R10hr).

Figure 1

Effects of experimental factors presented by response surface plots (a, b).

Figure 2

Effect of main effects on responses presented by contour plots (a, and b).

Table 5

Results of experiments for confirmation of the optimization capability.

	Factors		Responses
			DEE (%)b		R10 hr(%)c
Trial Code	SA:PVP K 30a	CaCl2 (% w/v)	Predicted	Observedd	Predicted	Observedd
O-1	1.00	8.60	99.45	98.27±3.42	69.55	71.28±2.22
O-2	1.50	9.20	99.79	98.60±3.55	69.41	71.02±2.67
O-1	2.20	10.30	99.16	97.88±2.86	68.46	69.88±2.82

SA:PVP K 30=Sodium alginate to polyvinyl pyrrolidone K 30 ratio;

DEE=Drug entrapment efficiency (%);

R10hr=% Drug released within 10 hrs;

Observed response values: Mean±SD (n=3).

Optimized microbeads showed DEE of 97.88±2.86 to 98.60±3.55%. Among DS-loaded alginate microbeads, highest DEE of 85.77±2.92% was observed in the case of S-3 (Table 6).

Table 6

Processing parameters and responses of plain alginate microbeads containing diclofenac sodium.

	Processing parameters		Responsesa
Trial Code	SAb	CaCl2 (% w/v)	DEE (%)c	R10 hr (%)d
S-1	2.00	8.60	78.23±2.06	95.38±3.62
S-2	2.00	9.20	81.48±3.15	91.22±3.23
S-1	2.00	10.30	85.77±2.92	81.03±3.05

Observed response values: Mean±SD (n=3);

SA=Sodium alginate;

DEE=Drug entrapment efficiency (%);

R10hr =% Drug released in 10 hrs

The particle sizes of alginate-PVP K 30 microbeads decreased by increase in the amount of incorporated PVP K 30 (due to decrease in viscosity and droplet sizes of polymer solution), and the CaCl2 concentration (due to high degree cross-linking) (Table 7). The surface morphology of optimized beads was visualized by SEM, which showed rough surface (Fig. 3).

Table 7

Average diameter of alginate-PVP K 30 and plain alginate microbeads containing diclofenac sodium, measured by optical microscopic method.

Formulation codesa	Average diameter (mm)b
SP-1	0.93±0.08
SP-2	0.59±0.07
SP-3	0.95±0.10
SP-4	0.84±0.07
SP-5	0.79±0.05
SP-6	0.87±0.09
SP-7	0.97±0.07
SP-8	0.60±0.06
SP-9	0.86±0.08
SP-10	0.86±0.10
SP-11	0.86±0.08
SP-12	0.84±0.06
SP-13	0.85±0.09
O-1	0.59±0.05
O-2	0.62±0.08
O-3	0.62±0.07
S-1	0.87±0.07
S-2	0.83±0.08
S-3	0.82±0.08

SP-1 to O-3 were alginate-PVP K 30 microbeads; whereas S-1-S-3 were plain alginate microbeads containing diclofenac sodium.

Mean±SD (n=3)

Figure 3

Photomicrograph of the surface morphology of DS-loaded alginate-PVP K 30 microbead (O-2).

In FTIR spectra, incorporation of DS in alginate showed characteristic peaks of DS, and alginate (Fig. 4). However, in DS-loaded alginate-PVP K 30 microbeads, two characteristic shifts for C=O stretching of PVP, and –OH groups of alginate compared to pure component (Fig. 4) appeared which strongly supports the idea of an intermolecular hydrogen-bonding between C=O groups of PVP, and –OH groups of alginate in alginate-PVP K 30 microbeads.

Figure 4

FTIR spectra of DS, sodium alginate, PVP K 30, DS-loaded alginate (S-3) and DS-loaded alginate-PVP K 30 microbeads (O-2).

Microbeads were found to release negligible amounts of DS in acidic medium which probably could be the surface adhered drug. Alginate-PVP K 30 microbeads showed prolonged drug release over 10 hrs (Figs. 5 and 6). In contrast, alginate microbeads showed high percent of DS release (Fig. 6). The drug release from optimized microbeads was evaluated using various kinetic models, and it was found that the drug release followed zero-order model (Table 8), indicating controlled-release pattern. The Korsmeyer-Peppas model was employed to distinguish Fickian-release when n ≤0.43 and case-II transport when n ≥0.85 (in Korsmeyer-Peppas Model, n is diffusional exponent) (12, 13). The values of n ranged between 0.9632 to 0.9950 (Table 8), indicating the drug release followed case-II transport.

Figure 5

(a). In vitro drug release from alginate-PVP K 30 microbeads containing DS (SP-1 to SP-6) (Mean±SD, n=3). (b). In vitro drug release from alginate-PVP K30 microbeads containing DS (SP-7 to SP-13) (Mean±SD, n=3)

Figure 6

In vitro drug release from optimized alginate-PVP K 30 microbeads (O-1 to O-3) and alginate microbeads (S-1 to S-3) containing DS (Mean±SD, n=3).

Table 8

Results of curve fitting of the in-vitro diclofenac sodium release data from different optimized alginate-PVP K30 microbeads.

Formulation codes		O-1	O-2	O-3
Zero-order Model	Ko	0.0711	0.0705	0.0730
	R2	0.9962	0.9937	0.9960
First-order Model	K1st	0.1426	0.1394	0.1427
		R2	0.9795	0.9736	0.9602
Higuchi Model	KH	0.3486	0.3456	0.3579
	R2	0.9870	0.9825	0.9845
Korsmeyer-Peppas	KP	0.0779	0.0724	0.0728
Model	n	0.9632	0.9884	0.9950
	R2	0.9955	0.9871	0.9952

The swelling of optimized DS-loaded alginate-PVP K 30 microbeads was lower in acidic medium (pH 1.2) in comparison with that of alkaline medium (pH 7.4) (Fig. 7). Maximum swelling was noticed after 2–3 hrs in alkaline pH after which, erosion and dissolution took place. The swelling of optimized microbeads in alkaline pH could be explained by the exchange of ions between the calcium ion of microbeads and the sodium ions present in phosphate buffer, under influence of calcium-sequestrant phosphate ions, which could result in disaggregation of alginate-PVP K 30-matrix structure leading to matrix erosion and dissolution of swollen microbeads (14, 15).

Figure 7

Swelling behavior of the optimized alginate-PVP K 30 microbeads containing DS in pH 1.2 and pH 7.4. (Mean±SD, n=3).

CONCLUSION

The optimized formulation of alginate-PVP K 30 microbeads containing DS was developed based on central composite design. The DEE of optimized alginate-PVP K 30 microbeads containing DS were found to be 97.88±2.86 to 98.60±3.55 % with a controlled-release pattern (zero-order) and case-II transport drug release. The swelling behaviour of the developed microbeads was influenced by the pH of the test medium. The FTIR spectroscopy showed an intermolecular hydrogen-bonding which could be formed between C=O groups of PVP K 30 and –OH groups of alginate in alginate-PVP K 30 microbeads which might sustaine drug release from alginate-PVP K 30 microbeads and minimize drug leaching during preparation to facilitate increase in DEE. The methods of preparation of DS-loaded alginate-PVP K 30 for controlled release characteristics were found to be simple and reproducible. In conclusion, an oral alginate-PVP K 30 microbeads for controlled delivery system of DS was successfully developed by alginate-PVP K 30 blending using ionotropic gelation.