Abstract: Objective: The main objective of the current study was to figure out the influence of MTR instrumental variables on the pellets quality and performance. Materials and Methods: The current study also compared between hand-made granules and pellets using wet granulation. In addition, a formula composed of Avicel PH-101 and dibasic calcium phosphate (90:10 % w/w) was selected for studying the ability to scale up and possibility of transfer form MTR to larger mixer types. Results: It has been shown that the hand-made granules showed the faster drug release within 20 min which may be related to their non compacted nature and higher porosity. The maximum torque was in direct proportion with the mixing speed (rpm) value of MTR rotating blades. Pellets made of binder solution volume constituting the maximum torque position were of high quality. It has been shown that increasing binder volume led to decreasing friability of pellets and hand-made granules. Results showed that increasing binder volume during wet massing led to decreased pellets particle size and decreased friability. The decreased flowability of hand-made granules may be attributed to the scaly nature of their morphology which might increase the friction to each other. It has been shown that surface roughness of pellets was increased as binder volume increased. Scaling up from 30 to 500 and 1000 g was highly successful confirming suitability of MTR as a preformulation tool.
INTRODUCTION
Granulation is the process of agglomerating small particles together into larger, semi-permanent aggregates. In spite of its economic importance and over 50 years of research, granulation has until recently been more of an art than a science (Iveson et al., 2001). Wet granulation as special type of granulation is still widely used in the pharmaceutical industry. It has not been replaced by direct compression technology, partly because of development cost considerations and habits and partly because it remains in some cases an attractive technique (Faure et al., 2001).
Pellets as multiparticulates drug delivery system offer some therapeutic advantages such as less irritation of the gastro-intestinal tract and a lowered risk of side effects due to dose dumping. They also provide some technological advantages such as better flowability, less friable dosage form, ease of coating and uniform packing (Vervaet et al., 1995).
A specially designed Mixer Torque Rheometer (MTR) has demonstrated as successful tool monitoring the wet mass consistency in addition to ability to scale up (Rowe and Sadeghnejad, 1987; Rowe, 2000). By definition, scale-up process is the transfer of a controlled process from one scale to another (Tardos, 2005; Faure et al., 1999).
Extrusion-spheronization has been described as the most popular method of producing pellets and the method of choice in the preparation of spherical particles (Chukwumezie et al., 2002). Some publications pointed out the influence of the extrusion screen specifications on the pellet quality and performance (Pinto et al., 1992; Vervaet et al., 1994). Change in operating conditions such as the shearing stress or mixing speed (rpm) could produce granules of very different characteristics in terms of size, porosity and friability (Oulahna et al., 2003). It has been shown that homogeneity of binder distribution inside the wet mass will affect on the tensile strength of the final product (Kiekens et al., 1998). Other factors such as drying conditions are of great impact on pellets characteristics (Wlosnewski et al., 2010). Drying could affect on pellets in terms of porosity and shrinkage as the case with excipient known to absorb water (Perez and Rabiskova, 2002).
Presence of water is essential feature of formulations containing Microcrystalline cellulose (MCC) for the preparation of spherical granules by the process of extrusion-spheronization (Fielden et al., 1993). Granule size distribution and sphericity were found to be dependent on the operating conditions and moisture content (Wan et al., 1993). Different phases of liquid-solid interaction are proved to exist during wet granulation such as pendular, funicular, capillary and droplet phases that depend on the degree of liquid saturation between solid particles (Podczeck and Wood, 2003). In wet granulation, the amount of binder liquid just below the quantity needed to initiate the capillary state is regarded to be optimal (Leuenberger, 1994).
Literature reveals some publications referred to effect of mixing procedure and blades orientation inside the MTR which have a role in wet massing process (Rowe, 1996). It has been shown that torque measurement can allow a good control over the granulation process allowing us to define optimum binder requirements (Chitu et al., 2011).
Variations in the amount of water in the formulations and the speed of extrusion affected both the production and the quality of the extrudate and their ability to provide pellets (Pinto et al., 2001; Baert et al., 1993). The amount of liquid binder added at the maximum torque was found to be comparable with that found for the optimum production of pellets by spheronization (Rowe and Sadeghnejad, 1987). Other investigators showed that pharmaceutical granules were indicated to be formed at binder immediately prior to the torque maximum. The solubility of materials used (both drugs and fillers) play an important role in the quantity of water required to form satisfactory pellets and on the physical characteristics of pellets (Sousa et al., 2002). The aim of our study was to investigate the effect of instrumental variables such as rpm and torque position on pellets quality in addition to ability to scale-up. Other aim was to compare between hand-made granules as an old method to that of the new method such as pellets produced by extrusion-spheronization.
MATERIALS AND METHODS
Materials: The following materials were used as received: Avicel PH-101 was kindly donated by FMC International, Wallingstown, Little Island Co. Cork, Ireland. Polyvinylpyrrolidone (PVP K30) was purchased from Fluka chemica (Buch, Switzerland). Dibasic calcium phosphate (DBCP) lot No. 9965260 was purchased from BDH (Germany). Theophylline was purchased from (Fluka AG, Switzerland). All other chemicals used in the current study were of high purity analytical grade and were purchased from Fisher Scientific UK Ltd. (Loughborough, UK).
Methods
Rheometry: The Mixer Torque Rheometer (MTR) used in the present study was
of a 135 mL capacity stainless steel bowl inside which two oppositely rotating
blades with rotational speed ranging 0-250 rpm (MTR-3, Caleva, Dorset; United
Kingdom).
Mixing procedure in MTR: An accurately weighed 15 g of solid powder formulation was used in the wet massing studies. Five milliliters of liquid binder were added manually in multiple additions over predetermined wet massing intervals. Each wet massing interval consisted of a 60 s mixing period and a 20 s data logging (collection) period with the MTR operating at the adjusted rpm at room temperature. A 40 sec period was allowed for the MTR to auto-zero when empty and also on the dry powder before liquid addition. Data were collected as a relation between the mean line torques versus the binder ratio during the granulation process at specific time intervals.
Repeatability of results: Specific formula composed of DBCP and MCC PH101 (10:90), was selected for this test. MTR apparatus was operated at different speeds (50, 100 and 150 rpm). At each rpm the test was repeated three times and the Standard Deviations (SD) were calculated.
Toque position and optimum pelletization: For both extrusion-spheronization and hand-made granulation, three different binder ratios (volume) around the maximum value were used based on MTR results (Table 1). The three binder volumes used, were the volume causing the maximum torque, 85 and 115% of that volume. Hand-made granulation was made by manual sieving of the wet mass through ASTM sieve of 1 mm pore size.
Extrusion-spheronization: The wet mass was extruded at room temperature (18-20 °C) at a speed of 100 rpm through a perforated die with a screen of pore diameter 1 mm and internal length 1.25 mm using Mini Screw Extruder (Caleva, Dorset; UK). Spheronization was performed in a spheronizer (Model 120, Caleva, Dorset; UK) with a rotating plate of regular cross-hatch geometry at a speed of 900 rpm for 5 min.
| Table 1: | Different formulation manual and instrumental granulation using Avicel PH-101 and 3% w/v | |
| Table 2: | Effect of changing mixing speed of MTR (rpm) and repeatability of results at each point | |
| Data are showed as (Mean±SD) n = 3 | ||
Pellets were then dried as a monolayer on tray in a hot oven at 60±0.5°C for 8 h using (Heraeus instruments D-6450 type B6060, Hanau).
Scale-up: The formula tested in the MTR was of 30 g batch size which was the suitable sample size to be mixed inside the MTR mixing chamber. Based on the MTR results for the 30 g formula, pellets were made as mentioned under extrusion-spheronization procedure. Scaling up was tested for two larger batch sizes 500 and 1000 g. Mixing was done in larger mixer (Kenwood) followed by extrusion-spheronization of a sample from each batch as the same procedure of the 30 g batch.
Pellets evaluation
Pellets morphology: The morphological characteristics of pellets
were done using Scanning Electron Microscopy (SEM), (Jeol, JSM-1600 Tokyo, Japan).
For this purpose, samples were sputter-coated with a thin film of gold-palladium
under an atmosphere of argon using a gold sputter module in a high-vacuum evaporator.
The coated samples were then scanned at different focusing levels taken with
an SEM.
Bulk and Tapped density and Carr’s index: The tapped density was performed with 100 taps using a tapped density tester (ERWEKA, GmbH, Germany). The results mean are the average of three determinations and are summarized in Table 3. The Carr’s index (C) was calculated using the following equation:
where, ρB is the bulk density of the powder and ρT is the tapped density of the particulates.
Friability: The pellets friability as an indication to their mechanical strength to test the resistance of the pellets to abrasion was carried out using a USP-friabilator (Electrolab, India), equipped with an 8 inches diameter abrasion drum. A 10 g pellet sample from each formulation was placed in the friabilator drum containing 200 glass beads each of 4 mm diameter. The speed of the friabilator was 20 rpm. The friabilator was rotated for 10 min and the percent weight loss of the pellets was determined. The pellet friability data are given in Table 3.
Particle size analysis for pellets: The particle size analysis was performed using Malvern Mastersizer (Mastersizer, Scirocco 2000, Malvern Instruments, Grove-wood Road, UK) equipped with MSX dry powder feeder. The median particle size data are the mean of three determinations and are summarized in (Table 3). For a typical experiment, about 500 mg of sample were fed in the sample micro feeder. All samples were analyzed 3 times and average results were taken. The sizes below which are 10 (d 0.1), 50 (d 0.5) and 90% (d 0.9) of the sample, Specific Surface Area (SSA) and Span = (d 0.9-d 0.1)/(d 0.5) were used to characterize the pellets size distribution in addition to the mean particle size.
Uniformity of the drug content: Theophylline standard calibration curves were prepared in triplicates covering the whole linear range of UV-absorbance (0.1-1.0) using a concentration range between 2.0 and 20 μg mL-1. The regression equation was linear, with a correlation coefficient (R2) of 0.9998. The determination of theophylline content of the manufactured pellets was determined spectrophotometrically at 272 nm which is the maximum absorption of wavelength of theophylline done in triplicate (USP30-NF25). Pellets were crushed in a porcelain mortar then accurately weighed 50 mg of the crushed pellets were dispersed in 100 mL phosphate buffer (pH 6.8) under sonication for 5 min. The supernatant was filtered and measured using UV-visible spectrophotometer (Beling Rayleigh, Japan).
Dissolution studies: The in-vitro release measurements were performed using USP dissolution apparatus type II (LOGAN UDT-804) with paddle rotating at 100 rpm and containing 6 vessels (Young et al., 2002). In each vessel a 500 mL phosphate buffer, pH 6.8 was added. The temperature was maintained at 37±0.05°C. An accurately weighed 300 mg of the prepared formulation was added to each vessel. For each formula, release was run in triplicate and absorbance was determined at specific intervals at 272 nm up to 2 h. The cumulative percentage of drug released was determined as a function of time.
Statistical analysis: Statistical analysis was performed using Graph Pad, PRISM® 5.01 (Graph Pad software, Inc.) using one way ANOVA. Differences between formulations were considered to be significant at p≤0.05.
| Table 3: | Evaluation of pellets from different formulations. (A) Pellets (FI-III) and made granules (FIV-VI), (B) scale-up batches | |
| Data are showed as (Mean ±SD), n = 3 | ||
RESULTS AND DISCUSSION
The formula composed of DBCP and MCC PH-101 (10:90) was tested for its maximum torque value and the optimum binder ratio using the MTR operated at 50 rpm as shown in Fig. 1. Polyvinylpyrrolidone (PVP) K30 3% solution was used as binding agent all over the study. The maximum torque value was about 0.676 (0.39) Nm, while the optimum binder ratio was 1.333 mL g-1 (30 mL /30 g).
None of the formulations showed disintegration while dissolution tests which may be due to absence of external disintegrant. The hand-made granules showed the faster drug release within 20 min. Table 3 which may be related to their non compacted nature and higher porosity as depicted from the SEM Fig. 5.
Changing the mixing speed (rpm) of the MTR (Fig. 2) was of high impact on the results especially on the value of the maximum torque (Table 2). It has been shown that the maximum torque was in direct proportion with the rpm value according the following equation: y = 0.008x + 0.280 with R2 = 0.997 where y: maximum torque, x: rpm and R2: correlation coefficient. The optimum binder ratio showed an increase when changing rpm from 50 to 100 rpm followed by no effect when changed from 100 to 150 rpm. The Standard Deviations (SD) resulted from repeatability of results were ≤0.106 indicating the high testing accuracy.
Pellets and granules content of theophylline were uniform among all batches and ranged from 90-110% (Table 3) which may be due to drug solubility in the binder solution. Content uniformity of theophylline among hand-made granules and pellets might be referred to the nature of wet granulation which supports homogenous drug distribution.
As listed in Table 1, three binder volumes 25, 30 and 35 mL were used for making pellets (FI-FIII) and for hand-made granules (FIV-FVI). It has been shown that changing binder volume showed some physical differences between pellets and granules. These changes were obvious as concluded from particle size analysis and SEM (Fig. 5). Granules made of binder solution volume at the maximum torque position (FII) were of the highest quality in terms of roundness and surface homogeneity. Hand-made granules showed no difference in their morphological characterization upon changing binder volume. Dissolution data (Fig. 3) came to prove that changing binder volume caused differences in pellets dissolution while it was non-significant in dissolution profile of hand-made granules. This may be due to the compacted nature of pellets subjected to the harsh conditions during extrusion process.
| Fig. 1: | MTR curve showing the maximum torque position (a) at definite value binder ratio (b) using MCC PH-101 and PVP 3% solution | |
| Fig. 2: | MTR curve showing the effect of changing RPM on the
repeatability of the method and apparatus accuracy using MCC PH-101 and
PVP 3% solution |
|
The non significant difference between hand-made granules (FIV, FV and FVI) concerning dissolution profiles, may be due to the porous and loose nature of granules (Fig. 5) which facilitate drug release.
Friability of both pellets and hand-made granules was correlated with binder volume. It has been shown that increasing binder volume lead to decreasing friability (Table 3). This may be attributed to increased content of the polymeric binder and consequently more binding strength.
Scaling up was highly successful confirming suitability of MTR as a preformulation tool. The three batches 30, 500 and 1000 g used binder volumes 30, 500 and 1000 mL, respectively showed high similarity in dissolution profile (Fig. 4), particle size distribution (Table 3) and pellets morphology (Fig. 5).
| Fig. 3: | Dissolution curve showing differences between pellets (FI-III) and hand-made granules (FIV-VI) using different binder ratios | |
| Fig. 4: | Dissolution curves of pellets made from scale up batches 30, 500 and 1000 g | |
Surface roughness (Fig. 5) was increased as binder volume increased as appeared from the SEM of (FI, FII and FIII). It has been also shown that particle sizes of pellets were decreased as binder ratio was increased (Fig. 5).
Pellets had cars index less than 15% indicating good flowability (Table 3A), while hand-made (Table 3B) showed cars index more than 15% indicating less flowability. The decreased flowability of hand-made granules may be attributed to the scaly nature of their morphology which might increase the friction to each other.
| Fig. 5: | Scanning electron microscopic photomicrographs showing
the morphological characterization of pellets (Formulations I-VI) and scaling
batches with different zooming levels (A) X80, (B) X1000 |
|
CONCLUSION
Based on the MTR data, scaling-up was successfully done when transferred to mixer of higher capacity. The binder ratio required for the maximum torque was shown to be suitable for pellets production of high quality. The compact nature of pellets was the cause of the different dissolution profile, while the hand-made granules were of non significant dissolution difference. Scaly morphology of hand-made granules as appeared from the SEM was the cause for higher values of cars index. Changing the mixing speed of the MTR (rpm) was of high impact on the results as the maximum torque was in direct proportion with the rpm. Increased binder volume during wet massing led to decreased pellets particle size and decreased friability.
ACKNOWLEDGMENT
This study was financially supported by KACST grant number; P-S-10-0034 and by SABIC; grant number MED-30-46.
The authors would like to thank the FMC Biopolymer for their generous donation of their respective products and thanks to Mr. Maher Shoumali (FMC biopolymer) for his great efforts.