Preparation of risperidone nanosuspension
Risperidone nanosuspension was prepared by acid-alkali neutralization technique. In this regard, various types of surface-active agents or polymeric stabilizers were analyzed in nanoparticle formation and stabilization to obtain the nanosuspension with favorable particle size and size distribution. According to
Table 1, various types of stabilizers such as ionic surfactant (NaCMC), nonionic surfactants (polysorbate80, Cremophor EL) and semi-synthetic polymers (PVP, HPMC, HPMC E
15) were investigated in various ratios.
As shown in
Figure 1, various types of stabilizers with different ratios were employed to produce risperidone suspension in the nano-sized range. Among them, ionic NaCMC and nonionic tween 80 surfactants were more efficient in producing smaller nanoparticles (135 ± 1.073 nm and 145 ± 1.073 nm, respectively). However, the application of nonionic polymeric stabilizers such as HPMC K4M and HPMC E
15 resulted in larger particles with sizes of 209 ± 1.073 nm and 328 ± 1.073 nm, respectively. In the same way, the use of 1:1 PVP/drug as a polymeric stabilizer led to the formation of suspensions with large agglomerated particles (1724 ± 1.073 nm). On the other hand, Cremophor El, as a non-ionic surfactant, in the ratio of 0.04:1, reduced the size of the nanoparticles to 214 ± 1.073 nm.
Evaluation of the particle stability utilizing particle size analyzer as well as microscopic observation showed that risperidone nanoparticles were prone to particle growth some minute after preparation, as the risperidone particle size grew to 1 µm in formulations containing nonionic and anionic stabilizers because of susceptibility to Ostwald ripening and crystal growth. However simultaneous usage of HPMC E
15 and Cremophor EL (with the ratio of 3:0.04:1 to the drug) was suitable to produce the nanoparticles with the size of 188 ± 1.511. It strongly stabilized the nanosuspension without significant size increase after SFD (
Figure 2). In general, zeta potential of particles should be at least −20 mV for sterically stabilized systems to obtain a physically stable nanosuspension (28). The zeta potential of optimized nanosuspension was found to be −8.43 ± 0.00 mV, indicating that the prepared nanosuspension do not suffer from instability problems.
In general, the process of nanoparticle stabilization is dependent on several factors, including the interaction between drug and stabilizer, the viscosity of medium and the amount of drug surface energy. On the other hand, the nature of surfactants plays a significant role in forming efficient and stable nanoparticles (
10,
29). Therefore, there is not a unique method to select the optimal stabiliser for nanosuspension preparation. Surfactants can stabilize nanoparticles by reducing surface activation energy (
30). This process involves two major mechanisms: steric stability (often for nonionic surfactants and polymers) and electrostatic stability (for anionic or cationic surfactants) (
29). As a hydrophilic polymer, HPMC E
15 can prevent particles from being aggregated by creating a steric barrier around the particles. Moreover, adsorption of HPMC E
15 on the surface of the particles will reduce the surface interfacial energy of nanoparticle and will stabilize them. Since this method does not involve drug encapsulation, the drug loading is 100% (
9). The results also showed that particles’ size decreased with the increase of surfactant concentration up to 0.6% (w/v). However, further increase in HPMC concentration showed the opposite results (
Figure 3). As demonstrated by other studies, freeze-drying can induce aggregation in nanoparticles according to the generation of new surface such as ice crystal surface in the medium (
31). Therefore, the migration of surfactant to the new surface can decrease their efficacy and induce particle aggregation especially in a lower concentration of surfactant. In a higher amount of HPMC E
15, the nanoparticles’ size increased because of enhanced nanosuspension viscosity (
32).
There are a few reports on the development of drug nanoparticles by using Cremophor EL. This surfactant is prepared by reacting castor oil with ethylene oxide (
33). Previous studies have shown that nonionic nonpolymeric surfactant such as Cremophor EL, has a higher adsorption potential on the particles’ surface compared with the polymer of the same length (
10). The steric barrier of Cremophor EL and interaction with a lipophilic part of risperidone will cover the nuclei more efficiently and create an inhibitory effect. It will also reduce the inter nuclei interaction and prevent particle growth (
Figure 4). Further increase in the concentration of Cremophor EL from 0.006 to 0.008% (w/v) has a beneficial effect on particle size reduction (
Figure 1). Moreover, the application of Cremophor EL and HPMC E
15 (formulation A
12) was appropriate to increase the stability in the size of particles for further studies.
Particle size of nanoparticles after SFD
As suggested by
Figure 2, the size of the nanoparticles was not influenced after SFD. This means that the application of sugars as the soluble carriers can preserve the nanoparticles from aggregation and maintain the physical stability of particles. As shown in
Figure 2, the nanoparticle preservation ability of lactose and maltodextrin was higher than mannitol. However, the formulation containing mannitol as a matrix agent exhibited less redispersibility with a larger mean particle size. In addition, the results showed that PDI was between 0.1-0.3 reflecting mid-range polydispersity in all the formulations (data not shown) (
11).
Generally, the transformation of nanoparticles to dry powder can reduce their physical, chemical, and microbial instability and will make them suitable for the preparation of solid dosage forms (
34,
35). Spray freeze-drying is a well-known method to increase the stability of sensitive nanoparticles (
15,
17,
36 and
37). Although, spray drying is the more common method to prepare the powder some disadvantages such as high temperature and instability of labile drugs have limited its application (
6,
15 and
18). Two important factors determine the particle size and morphology of nanoparticles in the spray drying: evaporation rate and solute diffusion. The high evaporation rate of the droplet and low diffusion coefficient of nanosuspension leads to the accumulation of particles in the droplet surface and its instability. The use of SFD will reduce the aggregation of the nanoparticles by the rapid freezing of the substances and preventing further diffusion of nanoparticles and formation of irreversible aggregated. In addition, due to the low temperature of drying, the physical and chemical damage to nanoparticles will be minimum (
3,
18 and
38).
For this purpose, we used mannitol, lactose, and maltodextrin as matrix agents and cryoprotectants to preserve nanosuspension. In General, cryoprotectants prevent irreversible aggregation of the nanoparticle by filling the nanoparticle gaps created during water sublimation at the drying stage (
35,
38). Maintaining the size of nanoparticles was achieved by sugar and its high redispersibility which carried out via different mechanisms including; maintaining the glassy state of sugars during freezing and prevention of ice nucleation and mechanical stress caused by ice crystals (which can lead to the formation of irreversible coalescences in nanoparticles) and coating the nanoparticles with sugars and prevention of direct contact during SFD (
15,
36).
SEM test
Scanning electron microscopy of SFD microparticles showed the feather-like particles with porous structure for all of the formulations with no evidence of particle agglomeration. Porous sugars scaffold prevented nanoparticles from aggregation as shown in
Figure 5. It was assumed that multiple pores were formed by sublimation of ice crystals. The particles were formed with smooth and shapeless sheet structures. Formulations that contained lactose or maltodextrin as a matrix agent showed uniform particles with no crystals on the surface. However, according to
Figure 5c, sugar crystals can be found in formulations containing mannitol.
Differential scanning calorimetry and FTIR
As shown in
Figure 6, the pure risperidone exhibited a sharp endothermic peak at 171 ºC corresponding to its melting point which confirmed its crystalline nature. However, thermograms of L1 and MD1 microparticles did not show any distinct peak, which could be due to the amorphous state of risperidone incorporated in particles. The endothermic peak at 152 ºC is related to the crystallization of mannitol in the particles. As mentioned in previous studies, mannitol will form three polymorphisms including α, β, and δ with melting points of 166, 166.5 and 155 ºC, respectively (
39). The presence of surfactants and the use of the SFD process led to the formation of δ polymorph (
40). The crystal structure was formed when the ratio of mannitol increased. It repeated in spray drying of formulations containing antibody when the mannitol content of formulations exceeded 30% (
41). On the other hand, the addition of lactose and maltodextrin did not show any distinct peak due to the formation of amorphous structure in the matrix of the particles. Previous studies showed that the amorphous matrix is appropriate for including extra particles while more crystal regions will push the material out of the matrix (
42,
43). Therefore, increasing the amount of mannitol in the formulation will result in the deposition of nanoparticles in the external surface of the microparticles. The results of the DSC tests and SEM images confirmed the crystallinity of mannitol. The mechanical stress caused by the growth of mannitol crystals increased the possibility of nanoparticle aggregation.
Crystallized mannitol can also lead to separation of nanoparticle suspensions and decrease their stability (in comparison to other formulations) (
15). Moreover, the formation of mannitol crystals during SFD creates new surfaces. Therefore, surfactant particles can be distributed on new surfaces leading to decreased concentration of surfactants around the nanoparticles and so the probability of nanoparticle growth will increase (
31). In parallel, Packhaeuser CB
et al. showed that mannitol crystallization during freezing drying is responsible for less mannitol protective effect compared to other sugars (
44).
The IR spectrum of risperidone was characterized by stretching and bending bands according to
Table 3. Risperidone has three stable polymorphs A, B and E. The presence of amide carbonyl-stretching band (at 1650cm
−1) as a single peak, indicates the presence of the most stable polymorph A in all of the samples (as indicated in DSC results) (
45).
The appearance of characteristic peaks of risperidone in all of the formulations indicates the absence of chemical interaction between the risperidone and the other excipients, confirming the stability of the drug in nanoparticles (
Figure 7).
Micromeritic properties of microparticles
The flow properties of microparticles, pure risperidone and nanoparticles were determined by calculating the Carr̕s index and Hausner̕s ratio. As depicted in
Table 4, Carr̕s index of pure risperidone was classified as “very poor” flowability because of inter particulate attraction of risperidone coarse particles. Meanwhile, the formulation of microparticles with mannitol at the ratio of 1:1 improved the flowability of powder by decreasing the compressibility index to 29. However, the increase of the mannitol ratio to 5:1 and 10:1 worsen the powder flowability. These results were repeated when lactose and maltodextrin were used in incremental ratio. However, comparing Carr’s indexes showed that the ratio of 1:1 of lactose to risperidone resulted in “good” powder flowability and other formulations showed poor and very poor flowability. It should be noted that the general flowability of porous powder is low, especially in products of SFD with high porosity (
3,
46). The proper flowability of the powder will result in the uniformity of the dosage form. Various factors, including morphology of particle, particle size, bulk and tap densities, moisture content, and processing, will affect the flowability of a powder (
47). Formulation A with the sticky nanoparticles and lacking matrix agents showed the sticky point as nanoparticles covering with surfactants caused particle adhesion and reduced flowability(
14). Matrix agents will increase the flowability by decreasing the adhesion between nanoparticles. Therefore, the increase in the flowability of nanoparticles will be expected in the formulation of microparticles containing matrix agents. Our results indicated that the use of matrix agents (at the ratio of 1:1 to the drug) can improve the flowability of nanoparticles.
Dissolution study of microparticles and FDT
The dissolution properties of microparticles were evaluated in 20 mL SSF at 37 ± 0.5 °C. According to
Figure 8A, the dissolution rate of the non-processed pure drug was very slow and only 62% of the drug was dissolved after 30 min; however, the selected microparticle (10 L) was dissolved rapidly after 2 min (f2 = 20). The rapid absorption of risperidone in the oral cavity depends on the rapid dissolution of the drug in saliva. However, the data showed that the addition of the carrier to the formulation affected the risperidone dissolution rate. As shown in
Figure 8B, the addition of mannitol to the nanosuspension led to a slight retard in the risperidone dissolution. This effect was increased when the mannitol ratio was increased from the ratio of 1:1 to 10:1 (Man/Ris) (f2 = 42.2) and was reversed when lactose was applied as the carrier. As depicted in
Figure 8C, the addition of lactose to the microparticles resulted in a higher dissolution rate with a direct relation (f2 = 51.9). However, the addition of maltodextrin showed no predictable effect on the risperidone dissolution rate in all the applied ratios (
Figure 8D).
To study the dissolution behavior of risperidone nanoparticles in tablet form, the microparticles were formulated in fast disintegration tablets. Various fast disintegration tablets were developed (data are not shown), and the tablet with including 83.2% MCC, 9.5% SSG and 6% of microparticles (equivalent to 1 mg risperidone) was selected as microparticles carrier with a disintegration time of 13 s. According to the FDA guidance, fast disintegration tablets should be disintegrated within 30 s or less (Food and Administration 2008).
As seen in
Figure 8E, the risperidone dissolution of the FDT is slightly slower than SFD risperidone microparticles. However, risperidone was dissolved very quickly from the fast dissolution tablet (10 min) in comparison with the commercial tablet (f2 = 28.5), which was suitable for delivery of risperidone in the oral cavity. According to the Noyes Whitney equation, any reduction in particle size causes an increase in the surface area which has a linear relation with the dissolution rate (
48). Also, a decrease in the thickness of the diffusion layer around the particles and an increase of the concentration gradient between the surface of the particle and its surrounding solution can be regarded as other reason (
49). On the other hand, the presence of surfactants (with bigger effect) and matrix agents (with lesser effect) can help to increase the dissolution rate. Freeze-drying of nanoparticles which results in the production of highly porous particles with high water penetration efficiency, makes the condition more suitable for hydration. Some articles showed that the use of lactose, as well as other low molecular weight sugars such as trehalose in the tablet matrix, can act as a disintegration agent (
50,
51). Therefore, soluble matrix agents such as lactose and maltodextrin will increase the hydration of microparticles and enhance the dissolution rate. However, the results showed that the mannitol (by creating crystals) would worsen the particle agglomeration and lead to a lower dissolution rate compared with lactose and maltodextrin. Indeed, the higher dissolution rate was observed by an increase in the lactose concentration. Higher lactose content will induce a greater reduction in the contact between nanoparticles and hence increase the redispersibility of nanoparticles leading to faster particle hydration and dissolution (
15) and this was repeated by using maltodextrin as an oligosaccharide.
particle size of nanoparticles using different surfactant or polymeric stabilizers in 1:1 ratio (surfactant/RIS).
Particle size of nanoparticles before and after SFD
Particle size of nanoparticles using HPMC E15 alone and in combination with Cremophore EL
Chemical structure of Cremophor EL and HPMC E15
SEM images of spray freeze-dried microparticles: (a) MD1 formulation, (b) L1 formulation, (c) M1 formulation
DSC thermograms from top to down is: (A) M1 formulation, (B) pure risperidone, (C) L1 formulation, (D) MD1 formulation
FTIR spectrum from top to down is: (A) 10M formulation, (B) 10MD formulation, (C) pure risperidone and (D) 10L formulation
In-vitro drug release of: (A) risperidone nanoparticles (10MD, 10M, 10L formula), nanoparticle without matrix former and pure risperidone in SSF, (B) risperidone nanoparticles containing mannitol as a matrix former, (C) risperidone nanoparticles containing lactose as a matrix former, (D) risperidone nanoparticles containing maltodextrin as a matrix former, (E) lyophilized powder of 5L formula, Fast disintegration tab of 5L formula and Marketed tab
| A13 | A12 | A11 | A10 | A9 | A8 | A7 | A6 | A5 | A4 | A3 | A2 | A1 | Ingredients (w/v%) |
|---|
| 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | 0.2 | Risperidone |
| | | | | | | | | | | | 0.2 | Pvp |
| | | | | | | | | | | 0.2 | | Hpmc K4M |
| | | | | | | | 0.2 | | | | | CMC |
| | | | | | | 0.2 | | | | | | Tween80 |
| 0.8 | 0.6 | 0.4 | 0.8 | 0.6 | 0.4 | 0.2 | | | | | | | Hpmc E15 |
| 0.008 | 0.008 | 0.008 | | | | | | | 0. 008 | 0.006 | | | Cremophor EL |
| Formulation | Ingredients (w/v%) |
|---|
| Risperidone | Hpmc E15 | Cremophor EL | Mannitol | Lactose | Maltodextrin |
|---|
| A | 0. 2 | 0. 6 | 0. 008 | - | - | - |
| M1 | 0. 2 | 0. 6 | 0. 008 | 0. 2 | - | - |
| M5 | 0. 2 | 0. 6 | 0. 008 | 1 | - | - |
| M10 | 0. 2 | 0. 6 | 0. 008 | 2 | - | - |
| L1 | 0. 2 | 0. 6 | 0. 008 | - | 0. 2 | - |
| L5 | 0. 2 | 0. 6 | 0. 008 | - | 1 | - |
| L10 | 0. 2 | 0. 6 | 0. 008 | - | 2 | - |
| MD1 | 0. 2 | 0. 6 | 0. 008 | - | - | 0. 2 |
| MD5 | 0. 2 | 0. 6 | 0. 008 | - | - | 1 |
| MD10 | 0. 2 | 0. 6 | 0. 008 | - | - | 2 |
| Absorption peak (cm−1) |
|---|
| 1000-1200 | 1129 | 1352 | 1535 and 3060 | 1650 | 3064 |
|---|
| TYPE OF BOND | C–Nstretching of the tertiary amine | C-FStretching of the aryl fluoride | C-N stretching of the oxazole ring | C=C stretching of an arene ring | C=Ostretching vibration of the -lactam ring | C-Hasymmetric stretching of aromatic ring |
| 1M | 5M | 10M | 1L | 5L | 10L | 1MD | 5MD | 10MD | A | Pure risperidone |
|---|
| balk density (g/mL) | 0. 071 | 0. 09 | 0. 06 | 0. 022 | 0. 083 | 0. 05 | 0. 06 | 0. 05 | 0. 05 | 0. 033 | 0. 127 |
| tap density (g/mL) | 0. 1 | 0. 15 | 0. 11 | 0. 025 | 0. 1 | 0. 083 | 0. 085 | 0. 1 | 0. 075 | 0. 044 | 0. 212 |
| Carr’s Index | 29 | 40 | 45. 4 | 12 | 17 | 39. 75 | 29. 4 | 50 | 33. 3 | 25 | 40 |
| Hausner’s Ratio | 1. 4 | 1. 66 | 1. 83 | 1. 13 | 1. 2 | 1. 6 | 1. 4 | 2 | 1. 5 | 1. 3 | 1. 66 |
| Flowability | poor | very very poor | very very poor | good | fair | very very poor | poor | very very poor | very poor | passable | very very poor |