In situ implants have been developed as an alternative to traditional preformed implant systems (Packhaeuser et al., 2004; Graham et al., 1999; Dunn et al., 1990). These systems are liquids, injected subcutaneously or intramus-cularly, and deform in contact with aqueous fluids in the body to semisolid or solid matrices, releasing their con-tent in a controlled manner. Because of their injectable nature, their placement is less invasive and painful for patients, hence improving compliance. In situ implants may be present as thermoplastic pastes, thermally gelling systems, in situ cross-linked polymer systems, and poly-mer precipitation systems (Lu et al., 2007; Packhaeuser et al., 2004).
Thermoplastic pastes are injected as melt, so polymer that is used should have a low melting point, but this
requires a high injection temperature, which leads to pain and tissue necrosis that may inhibit drug release (Chitkara et al., 2006). Thermally gelling systems are formed of a polymer showing abrupt change in solubil-ity upon changing temperature. They form a gel in body temperature. Pluronics in high concentration (>15%) are considered implant-forming polymers; however, it was previously stated that these concentrations cause significant cytotoxicity (Pluta and Karolewicz, 2006). Poly(ethylene glycol)/poly-DL-lactide/poly(ethylene glycol) (PEG-PDL-PEG) triblock copolymer is another example, but its synthesis complexity hinders its wide use (Li et al., 2007). In situ implants, based on cross-linking of polymers, were studied in many previous researches. Photocatalyzed cross-linking of PEG-oligo-glycolyl acry-late could only be used in surgical sites accessible to a
ReseaRch aRtIcle
A novel injectable in situ forming poly-DL-lactide and DL-lactide/glycolide implant containing lipospheres for controlled drug delivery
Soad A. Yehia, Ahmed H. Elshafeey, and Ibrahim Elsayed
Department of Pharmaceutics and Industrial Pharmacy, Faculty of Pharmacy, Cairo University, Cairo, Egypt
abstractOne of the greatest challenges in in situ forming implant (ISFI) systems by polymer precipitation is the large burst release during the first 1–24 hours after implant injection. The aim of this study was to decrease the burst-release effect of a water-soluble model drug, donepezil HCl, with a molecular weight of 415.96 Da, from in situ forming implants using a novel in situ implant containing lipospheres (ISILs). In situ implant suspensions were prepared by dispersing cetyl alcohol and glyceryl stearate lipospheres in a solution of poly-DL-lactide (PDL) or DL-lactide/glycolide copolymer (PDLG). Also, in situ implant solutions were prepared using different concentrations of PDL or PDLG solutions in N-methyl-2-pyrrolidone (NMP). Triacetin and Pluronic L121 were used to modify the release pattern of donepezil from the in situ implant solutions. In vitro release, rheological measurement, and injectability measurement were used to evaluate the prepared in situ implant formulae. It was found that ISIL decreased the burst effect as well as the rate and extent of drug release, compared to lipospheres, PDL, and PDLG in situ implant. The amount of drug released in the first day was 37.75, 34.99, 48.57, 76.3, and 84.82% for ISIL in 20% PDL (IL-1), ISIL in 20% PDLG (IL-2), lipospheres (L), 20% PDL ISFI (I5), and 20% PDLG ISFI (I8), respectively. The prepared systems showed Newtonian flow behavior. ISIL (IL-1 and IL-2) had a flow rate of 1.94 and 1.40 mL/min, respectively. This study shows the potential of using in situ implants containing lipospheres in controlling the burst effect of ISFI.
Keywords: PDLG, injectable, in situ implant, lipospheres, burst release
Address for Correspondence: Ahmed H. Elshafeey, Department of Pharmaceutics and Industrial Pharmacy, Faculty of Pharmacy, Cairo University, Kasr El-Aini Street, Cairo 11562, Egypt; Fax: +202 25081440; E-mail: [email protected]
(Received 01 September 2011; revised 30 September 2011; accepted 08 October 2011)
light source (Yoon et al., 2007). Thiolated chitosan does not require initiation for cross-linking, but gelation takes several hours (Bernkop-Schnurch et al., 2003). PEG-based copolymer, containing multiple thiol (–SH) groups, was previously stated to solidify rapidly, but, being nonbio-degradable, is considered a great disadvantage. The Ca2+/alginate system undergoes ion-mediated gelation upon injection (Packhaeuser et al., 2004); however, Ca2+ leak-age during shelf storage and immunological response upon injection are considered major drawbacks of this system (Suzuki et al., 1998).
In case of polymer precipitation systems, a biodegrad-able polymer is dissolved in water-miscible solvents, such as N-methyl-2-pyrrolidone (NMP) and dimethyl sulfox-ide (DMSO). Upon injection of the drug-containing poly-mer solution into an aqueous medium or body tissue, solvent diffuses into the aqueous surrounding, leaving polymer that precipitates by forming a solid or semisolid mass from which drug release occurs slowly (Cho et al., 2007). One of the greatest challenges in in situ forming implant (ISFI) systems by polymer precipitation is the relatively large burst release during the first 1–24 hours after implant injection. It was found that the drug release from implant is directly related to the rate of phase inver-sion. Variables that affect the rate of phase inversion are hydrophilicity of the solvent, molecular weight (MW) of the polymer, polymer concentration, excipient additives (DesNoyer and McHugh, 2003), and nonsolvent composi-tion (Luan and Bodmeier, 2006; Brodbeck et al., 1999). In a previous study, a relatively hydrophobic solvent, such as ethyl benzoate, was used in the preparation of ISFI, which reduced the rate of water influx into the systems and slowed the phase inversion rate (Brodbeck et al., 1999). Several formulations exhibited zero-order drug release; however the injectability of these implants was greatly reduced as a result of their increased viscosity (Brodbeck et al., 1999). Different strategies have been employed to reduce burst release without increasing solution viscos-ity, one of which is the inclusion of surfactant excipients, such as Pluronic coblock polymers, in ISFI formulations (Patel et al., 2010a; DesNoyer and McHugh, 2003). Patel et al. has also shown that adjusting the polymer MW used in ISFI polymer formulations could help modulate burst drug release (Patel et al., 2010a). Other strategies include using in situ forming microparticle-based systems, which are comprised of an emulsion of polymer solu-tion and oil (Ahmed et al., 2008; Kranz et al., 2008). The aim of this study was to decrease the burst release effect of a water-soluble model drug, donepezil HCl, a revers-ible, noncompetitive cholinesterase inhibitor, with a MW 415.96 Da, from ISFIs. Novel in situ implant containing lipospheres (ISILs) were prepared, aiming at decreasing the burst effect; further, the effect of different variables, including polymer type and MW, polymer concentration, and added excipients on burst release effect were also compared. The viscosity and the syringeabilty of the pre-pared systems were determined. Lipospheres were used because they are well-known carrier systems developed
for the parenteral delivery of bioactive compounds (Nasr et al., 2008). Moreover, they have some advantages over other delivery systems, such as good physical stabil-ity, low cost of ingredients, and ease of preparation and scale-up (Nasr et al., 2008; Esposito et al., 2005).
Methods
MaterialsDonepezil was kindly supplied from Hikma Pharma Company (Cairo, Egypt) (batch no.: 842491). Cetyl alco-hol (CA) was purchased from Al-Gomhoria Co. (Cairo, Egypt). Purasorb PDL 04 (poly-DL-lactide), 0.4 dL/g (MW, 18,000-28,000), and PDLG 7502 (75/25 DL-lactide/glycolide copolymer), 0.2 dL/g (MW, 4,000-15,000), were a gift from Purac Biomaterials Co. (Gorinchem, The Netherlands). Glyceryl stearate (GTS), triacetin, Pluronic L121, NMP, and DMSO were purchased from Sigma-Aldrich (St. Louis, Missouri, USA). All other reagents and buffer components were of analytical grade.
MethodsPreparation of lipospheresLipospheres were prepared using the melt-dispersion technique. An amount of 1 g of lipid (CA) and GTS (ratio, 1:1) was melted in a thermostatically controlled water bath (GFL; Gesellschatt Laboratories, Berlin, Germany) at 80°C. Then, 50 mg of drug was dispersed in the mol-ten lipid with continuous stirring. The suspension was emulsified into previously heated 0.25% polyvinyl alco-hol (PVA) aqueous solution as an external phase (pH 12, using trisodium phosphate) at 80°C, with continuous stir-ring at 500 rpm on a magnetic stirrer. The emulsion was kept at 80°C for 3 minutes, then rapidly cooled to 20°C by immersing in an ice bath without stopping the stir-ring, to obtain uniform lipospheres (Barakat and Yassin, 2006; Cortesi et al., 2002). The prepared lipospheres were isolated by filtration on filter paper, washed with distilled water, and air-dried at room temperature for 48 hours.
Preparation of different in situ implant delivery systemsIn situ implant solutionsFour different groups of in situ implant formulations, containing 15 mg/mL of donepezil, were prepared. The first group was prepared using 25, 50, 100, and 150% (w/v) CA dissolved in NMP formulae (I1-I4); the second group was prepared using 20, 33, and 40% (w/w) PDL or PDLG dissolved in NMP (I5-I10); the third group was prepared using 20 and 33% (w/w) PDL or 20, 33, and 40% (w/w) PDLG dissolved in a mixture of NMP and triacetin (I11-I15); and the fourth group was prepared using a mixture of 20% (w/w) PDL or PDLG with 13 and 20% (w/w) Pluronic dissolved in NMP (I16-18). Briefly, the lipid, polymer, or polymer mixture, in addition to drug, were added to the solvent (NMP, alone or in a mixture with triacetin) and dissolved in a vial using a probe sonicator (Hielscher, Teltow, Germany) for 60 seconds in an ice bath (Kranz
Jour
nal o
f L
ipos
ome
Res
earc
h D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Sydn
ey o
n 03
/14/
13Fo
r pe
rson
al u
se o
nly.
130 S. A. Yehia et al.
Journal of Liposome Research
et al., 2008). Table 1 shows the composition of different in situ implant formulationss.
ISILsISILs were prepared by suspending the previously pre-pared lipospheres formula in 20% (w/w) PDL (IL-1) or PDLG (IL-2) solution. PDL or PDLG were added to DMSO and dissolved using a probe sonicator for 60 sec-onds, then the prepared lipospheres were suspended in the formed solution by shaking, using a vortex mixer.
In vitro evaluation of the prepared in situ implant formulaeScanning electron microscope (SEM) imagingSurface characteristics of the representative donepe-zil in situ implant formulation were observed using an SEM (JXA-840A; JEOL Ltd., Tokyo, Japan). Samples were injected into phosphate-buffered saline (PBS; pH 7.4) to solidify. After 1 day, the sample was taken and dried for 2 days in a dessicator to allow complete diffusion of NMP; then, this sample was coated with gold and visualized using an SEM.
In vitro releaseThe prepared solution or suspension was injected slowly, using a 21-gauge needle, into an overnight-soaked dialysis bag containing 1.5 mL of PBS (pH 7.4) to form an implant (Graves et al., 2007). The dialysis bag was immersed in 20 mL of PBS at 37 ± 0.5°C in a horizontal shaker (GFL; Gesellschatt Laboratories) operated at 100 rpm for 30 days (Jain et al., 2000a). Medium was replaced every sampling time with fresh medium. Sampling intervals were 2 hours, then 1, 2, 3, 5, 7, 10, 13, 17, 21, 25, and 30 days. Samples were assayed spectrophotometrically for drug content at 315 nm. The obtained results were the mean of three runs.
To determine the release model, the in vitro release data were analyzed with respect to a zero-order, first-order, and diffusion-controlled mechanism with respect to the simplified Higuchi model (Higuchi, 1963). Further, the following Korsmeyer-Peppas equation was also used to characterize drug release (Ritger and Peppas, 1987):
M
MKtt n
∞
=
where M Mt / ∞ represents the drug dissolved fraction at time t, K is a kinetic constant, and n is the diffusional exponent. It depends on the release mechanism and the shape of the drug-delivery device. An exponent, n ≤ 0.5, corresponds to a Fickian diffusion release (case I diffu-sional), 0.5 < n < 1 to an anomalous (non-Fickian) trans-port, n = 1 to a relaxation (case II) release kinetics, and n > 1 to a super case II transport (Gad et al., 2008; Vernon et al., 2004; Peppas and Sahlin, 1989).
Rheological evaluationHigh viscosity of the prepared in situ implant was a chal-lenging parameter for their injection. So, 1 mL of each formula was used to measure its rheological properties using a cone and plate rheometer with spindle 40 (Ito et al., 2007). Viscosity and shear stress were recorded at different rates of shear, ranging from 1 to 500 sec−1. Results were recorded only when the torque was within the acceptable range (10–100%) (Brookfield, 2010). Several models were used to analyze the rheological behavior of the prepared formulae.
Injectability evaluationThe injectability test was done using handmade equip-ment modified from a similar piece of equipment previ-ously used by Leroux et al. and Lacout et al. (Lacout et al.,
Table 1. Compositions of the different in situ implant solutions.
2005; Leroux et al., 1999). A sample (1 mL) of each in situ implant formula was filled into a 5-mL syringe attached to a 19-gauge needle. The syringe was fitted to a rubber tube that ended with an air pump. Pressure exerted on the solution surface, by the forced air, was measured by a sphygmometer in mmHg units. The presence of a valve allowed for maintaining constant pressure (70 mmHg). The time taken to eject 1 mL of each formula was recorded. Flow-rate values (mL/min) were calculated to give an indication for the injectability of the formulae (Paul et al., 1998).
Statistical analysisRelease half-lives (t
50%) of the formulae were statistically
analyzed using one-way analysis of variance to test the significance of difference at P ≤ 0.05. A subsequent Tukey honestly significant difference test was also performed. SPSS 15.0 software (SPSS, Inc., Chicago, Illinois, USA) was used to carry out these statistical analyses.
Results and discussion
For the preparation of the in situ forming drug-delivery systems, the polymer was dissolved in NMP. Upon injec-tion of the polymer solutions into the release medium, the polymer solidified as the solvent diffused into the buffer (pH 7.4) and formed implants. The implants, which were collected 24 hours after exposure to the aqueous medium, had a smooth surface with no observed pores (Figure 1).
In vitro releaseIn vitro release of in situ implant formulae containing CAFigure 2 shows the release of donepezil from CA in situ implant formulae. It is obvious that increasing CA con-centration led to a decrease in burst release effect and the release rate of drug from the formed implants. Formula I1 (containing 25% CA; w/v) and I2 (containing 50% CA; w/v) had nearly the same burst release, where the per-centage drug release was 80.57 and 81.43%, respectively, on day 1; however, formula I3 (containing 100% CA) and I4 (containing 150% CA) showed lower burst release (60.48 and 46.52%, respectively). Although the extent of drug release in all formulae was 100%, the rate of drug release decreased by increasing the amount of CA where
formulae I1 and I2 released 100% drug after 2 and 10 days, respectively, and I3 as well as I4 released the drug after 13 days. This might have been the result of CA hydropho-bicity, which hindered water diffusion and drug release (Paradkar et al., 2003).
By fitting the release data in a zero-order, first-order, and diffusion-controlled mechanism according to sim-plified Higuchi models (Table 2), it was found that all formulae followed a diffusion-controlled mechanism; further, according to the Korsmeyer-Peppas model, anomalous release was observed in I4. However, data in the case of I1, I2, and I3 were not enough to precede Korsmeyer-Peppas calculations.
Table 2 shows the t50%
, which was calculated from the equation fitted to the release mechanism (i.e., diffu-sion) of the different formulae. The formulae could be arranged in a decreasing order of I4 > I3 > I2 > I1, with respective values of 2.1, 0.88, 0.64, and 0.60 days. The t
50% of all formulae were significantly different from each
other (P < 0.001), except for I1and I2.
In vitro release of in situ implant formulae containing PDL or PDLGFigure 3 shows the release of donepezil from PDL and PDLG in situ implant formulae. It is clear, from Figure 3, that increasing polymer concentration in both PDL and PDLG caused a decrease in burst drug release, where
Figure 1. Scanning electron micrograph (magnification ×1,000) of I9 (33% PDLG).
Figure 2. Release profile of donepezil from in situ implant formulae containing CA.
Jour
nal o
f L
ipos
ome
Res
earc
h D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Sydn
ey o
n 03
/14/
13Fo
r pe
rson
al u
se o
nly.
132 S. A. Yehia et al.
Journal of Liposome Research
the percentage of drug release on day 1 was 76.3%, and 64.96% for 20 and 33% polymer concentrations in the case of PDL and 84.82, 17.30, and 14.41% for 20, 33, and 40% concentrations in the case of PDLG. This decrease in burst effect resulting from an increase in polymer con-centration might have been caused by the hydrophobic character of the polymers used. So, in situ implant for-mula containing a high polymer concentration tended to form more dense mass upon contact with the dissolution medium. This dense mass might hinder water diffusion, drug release, and, also, polymer degradation (Gad et al., 2008; Cui et al., 2005). A cake was formed during the preparation of I7 (containing 40% PDL); thus, no release data were found for it.
It is worth noting that the release pattern of both poly-mers (PDL and PDLG) was almost the same in the case
of the 20% polymer concentration; however, in the case of the 33% concentration, PDLG gave slower drug release than PDL, as shown in the decrease of burst effect from 64.93 to 17.3% on day 1 in PDL and PDLG, respectively. Further, 100% of the drug was released after 17 and 25 days from PDL and PDLG, respectively. This might have been the result of the low MW of PDLG, compared to PDL. These results are in accord with previously estab-lished findings, whereby lower MW PLGA ISFIs released drug slower than higher MW ones (Patel et al., 2010a). Previous studies by Graham et al. have shown that in vitro implant formation correlated with drug release from different ISFIs, with slower forming implants releas-ing drug slower than faster forming implants (Graham et al., 1999). The reason for this effect, as hypothesized by McHugh et al., is that the mobility of the drug is reduced in the more viscous interior nonphase inverted core than in the porous, water-rich precipitated matrix shell (McHugh, 2005). Therefore, as solvent efflux increases and implant precipitation increases, drug dissolution and release would thereby be increased. Low MW PLGA implants showed greater swelling, resulting in a greater volume expansion over time, compared to higher MW PLGA implants. This in vitro swelling effect led to greater water influx than solvent efflux, and lower MW PLGA implants may retain a greater percent of their solvent and drug than higher MW implants (Solorio et al., 2010). Moreover, previous studies have also shown that, in vitro, higher MW PLGA polymers have a lower solubility than lower MW PLGA polymers, and a lower critical water concentration threshold is required to induce their pre-cipitation out of solution (Patel et al., 2010a). The ability of low MW PLGA implants to retain their solvent (solubi-lized polymer) and drug in vitro may lead to a persistent
Table 2. Kinetic parameters of donepezil release from in situ implant formulae.
osmotic effect that results in implant swelling (Patel et al., 2010b). It has been reported that implant formulations comprised of high MW polymers in solvents with high water miscibility will undergo a rapid phase inversion, resulting in a characteristic honeycomb-like structure of diffusion pores for drug and water to travel through (Raman and McHugh, 2005; Brodbeck et al., 1999). When NMP is used as the solvent, the critical concentration of water required to induce phase inversion decreases as the MW of the polymer increases because of the change in affinity of the solvent for the polymer (Solorio et al., 2010; Patel et al., 2010a; McHugh, 2005).
It is worth noting that a triphasic release pattern was observed in I10; first, an initial burst release followed by slow diffusion of the drug from pores present in the matrix, then polymer erosion, which allowed a higher rate of drug release (Ramchandani and Robinson, 1998). Water diffusion through the concentrated polymer chain (40%) led very slowly to polymer hydrolysis, yielding lactic or glycolic acid or both. These produced acids decreased the microenvironmental pH, which enhanced acid-catalyzed hydrolysis of other polymer chains, rap-idly leading to accelerated polymer erosion and drug release (Jiang et al., 2002).
The study of the release kinetics mechanisms outlined in Table 2 shows that all formulae followed a diffusion-controlled mechanism, except I10, which followed a zero-order release mechanism.
The t50%
of the different formulae (Table 2) could be arranged in a decreasing order of I10 > I9 > I8 > I6 > I5, with respective values of 5.57, 3.75, 1.03, 0.96, and 0.91 days. The t
50% of all formulae were significantly different
from each other (P < 0.001), except for I5 and I6.
Effect of partial replacement of NMP with triacetin on the in vitro releaseFigure 4 shows the release of donepezil from in situ implant formulae containing PDL and PDLG dissolved in a mixture of NMP and triacetin (ratio, 9:1). Triacetin is a short-chain triglyceride with an excellent low-toxicity profile (Jain, 2000; Jain et al., 2000b). Upon mixing with NMP in a ratio of 9:1 (I11-I15), a slower rate and extent of drug release was observed, in comparison to formu-lae dissolved in NMP only (I5-I10). The burst effect was decreased on day 1 from 76.3 to 46.7% and from 64.96 to 41.84% in the case of 20 and 33% PDL, respectively. Also, a decrease in burst effect from 84.82 to 63.08%, from 17.3 to 10.36%, and from 11.41 to 9.99% in the case of 20, 33, and 40% PDLG, respectively, was observed. This might have been the result of the hydrophobic properties of triacetin, which decreased the polarity of NMP. The triacetin-NMP mixture diffused slowly upon injection into dissolution medium, with decreased water influx, drug escape, and burst release (Matschke et al., 2002). Further, the addi-tion of triacetin might have decreased system porosity (Graham et al., 1999).
Further inspection of Figure 4 reveals that a tripha-sic release pattern was observed in both I14 and I15
(containing 33 and 40% PDLG, respectively). However, in the absence of triacetin (Figure 3), triphasic release was observed only in I10 (containing 40% PDLG). Triacetin is chemically formed of triester of glycerol and acetic acid; upon hydrolysis, it yields acetic acid, which might cause a further decrease in microenvironmental pH (SIDS, 2002), hence increasing the chance of polymer erosion in the lower polymer concentration. It is worth noting that, although I15 had a higher polymer concentration (40%), it showed a higher drug release rate than I14 (33% poly-mer concentration) after day 3. This might have been the result of the higher percentage of polymer, in I15, yield-ing a greater amount of acid, which led to a higher rate of self acid catalysis of polymer hydrolysis.
Table 2 shows that all formulae containing triacetin followed a diffusion release mechanism, except I14 and I15, which followed a zero-order release mechanism. The t
50% of the release mechanism (Table 2) could be arranged
in the following descending order of I12 > I14 > I11 > I13 > I15, with respective values of 13.41, 5.33, 3.84, 3.82, and 3.61 days. The t
50% of all formulae were significantly differ-
ent from each other (P < 0.001), except for I11 and I13.
Effect of partial replacement of PDL or PDLG with Pluronic on in vitro releaseFigure 5 shows the release of donepezil from in situ implant formulae containing 33% PDL, 33 and 40% PDLG, as well as its release after partial replacement of PDL and PDLG with Pluronic L121 in each for-mula. Pluronics were used in many previous studies as release modifiers by enhancing or sustaining drug release, depending on the system polymer/pluronic mixture characters (Carrasquillo et al., 2001; Blanco and Alonso, 1998). Pluronic L121 is a fairly hydropho-bic polymer (DesNoyer and McHugh, 2003).
It is obvious, from Figure 5, that partial replacement of 33% PDL with 13% pluronic in NMP (I16) decreased burst release as well as the rate and extent of drug release, in comparison with I6 (containing 33% PDL). This indicates
Figure 4. Release profile of donepezil from in situ implant formulae containing PDL or PDLG, each dissolved in a mixture of NMP and triacetin (9:1).
Jour
nal o
f L
ipos
ome
Res
earc
h D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Sydn
ey o
n 03
/14/
13Fo
r pe
rson
al u
se o
nly.
134 S. A. Yehia et al.
Journal of Liposome Research
that the PDL-Pluronic mixture had a higher sustainment efficiency than PDL itself (McHugh and Desnoyer, 2007; DesNoyer and McHugh, 2003).
On the other hand, I17 (containing 20% PDLG and 13% pluronic) showed higher burst release and lower release rate and extent than I9 (containing 33% PDLG), where the amount released on day 1 was 32.92 and 17.3% for I17 and I9, respectively. However, the extent of drug release from both formulae was 79.4% after 30 days and 100% after 25 days, respectively. Also, I18 (containing 20% each of PDLG and Pluronic) showed a burst release of 25.83%, and the extent of drug release was 70.6% after 30 days. However, the burst release from I10 (containing 40% PDLG) was 11.41%, and the extent of drug release was 100% after 17 days. This indicates that the PDLG-Pluronic mixture had a higher sustainment efficiency than PDLG itself, but it did not decrease the burst effect.
By fitting the release data into the different kinetic release models, it was found that all formulae showed a diffusion release mechanism.
Table 2 shows that partial replacement of 33% PDL with 13% Pluronic in NMP (I16) significantly increased the t
50% from 0.96 to 16.21 days (P < 0.001). Further, partial
replacement of 33% PDLG with 13% Pluronic in NMP (I17) significantly increased the t
50% from 3.75 to 11.84
days (P < 0.001). Also, partial replacement of 40% PDLG with 20% Pluronic in NMP (I18) significantly increased the t
50% from 5.75 to 14.45 days (P < 0.001).
ISILsFigure 6 shows the release of donepezil from lipo-spheres prepared with the ratio of 1:10:10 of drug to GTS to CA, I5 (20% PDL), I8 (20% PDLG), as well as ISILs. It is obvious, from Figure 6, that the presence of lipospheres within the in situ implant decreased the burst effect as well as the rate and extent of drug release, compared to lipospheres, PDL, and PDLG in situ implant. The amount of drug released on day 1 was 37.75, 34.99, 48.57, 76.3, and 84.82% for lipospheres
containing implant in 20% PDL (IL-1), lipospheres containing implant in 20% PDLG (IL-2), lipospheres (L), 20% PDL ISFI (I5), and 20% PDLG ISFI (I8), respec-tively. Further, the extent of drug release was 79.32 and 69.14% for IL-1 and IL-2, respectively. However, it was 100% after 5, 13, and 17 days for L, I5, and I8, respec-tively. This indicates the potential of using lipospheres containing in situ implants in controlling the burst effect of solution ISFI.
By studying the release kinetics mechanism (Table 2), it was found that both lipospheres containing ISFI formulae showed a diffusion-controlled mechanism. The t
50% of 20% PDL and 20% PDLG (0.91 and 1.03 days,
respectively) was significantly increased after combining with the lipospheres to 3.69 and 4.65 days, respectively (P < 0.001).
Rheological evaluationRheological data obtained from a viscometer were shear stress and viscosity at different rates of shear val-ues. These data were fitted to the power law model to study the rheological behavior, as shown in the follow-ing equation:
τ = γ K n
where τ is the shear stress, γ is the rate of shear, K is the consistency index (seconds), and n is the flow index (dimensionless). For a shear thinning fluid, n lies between zero and 1, whereas it approaches 1 in the case of a Newtonian system and exceeds 1 in a dilatant system (Gad et al., 2008).
Table 3 shows viscosity in poise (η), measured at 50 sec−1 rate of shear, consistency index (K), flow index (n), and regression coefficient (r2) of each formula. It can be observed, from Table 3, that viscosity at 50 sec−1 increased with increasing CA or polymer concentration, whatever the polymer type. On the other hand, the use of triacetin in some formulae (I11-I15) increased viscos-ity, in comparison to formulae without it (I6-I10) (Perova et al., 1998).
Figure 5. Release profile of donepezil from in situ implant formulae containing PDL or PDLG, each in combination with Pluronic.
Figure 6. Release profile of donepezil from in situ implant formulae containing lipospheres.
Jour
nal o
f L
ipos
ome
Res
earc
h D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y U
nive
rsity
of
Sydn
ey o
n 03
/14/
13Fo
r pe
rson
al u
se o
nly.
Injectable lipospheres for controlled drug delivery 135
The flow index (n) values in the case of CA contain-ing formulae were smaller than, and far away from, 1. It indicated the non-Newtonian and shear thinning behavior of these formulae, according to the power law equation. It was previously reported that CA showed a pseudoplastic behavior when formulated as oil-in-water emulsion (Ribeiro et al., 2004; Sepulveda et al., 2003).
On the other hand, all other formulae showed a flow index ranging from 0.8 to 1.2, indicating closeness to the Newtonian behavior.
Several equations (e.g., Bingham, Casson, and Carreau equations) were used to describe the non-Newtonian system. Comparison of their regression coefficients could verify the type of non-Newtonian system.
The Bingham equation was used to describe linear plastic system:
τ = τ + γ0 K
where τ0 is the yield value.
However, the nonlinear plastic system was described by the Casson equation:
τ τ + γ1/201/2 1/2 1/2 = K
On the other hand, Carreau’s model was used for the description of pseudoplastic shear thinning systems:
n n
n n K m
−−
=+ γ
∞
∞02 /2
1
(1 ( ) )
where η0 and η
∞ refer to viscosity values at lowest and
highest rate of shear values, respectively, K is a constant parameter with the dimension of time (1/K is the rate
of shear at which viscosity begins to decrease), and m is a dimensionless constant indicating degree of pseu-doplasticity (Barnes et al., 1993). Closeness of m values to zero indicate Newtonian flow, whereas higher values indicate closeness of pseudoplastic behavior (Moncalvo and Friedel, 2010). η
0, η
∞, and K could be determined
from the shear rate/viscosity curve (Vlad and Oprea, 2001).
Table 4 shows the rheological data fitting for CA containing formulae (I1-I4) to the previously explained equations. The calculated regression coefficient values indicated that the pseudoplastic behavior predominated. This was also confirmed with the closeness of the Carreau constant (m) to 1, as can be observed in Table 5.
Injectability measurementThe injectability test was done, using a 19-gauge syringe needle, which is acceptable for intramuscular injection (19-22 gauge) (Boylan and Nail, 2009). Constant pressure was applied (70 mmHg). Upon using Pascal’s law, the applied pressure to 5 mL syringe (radius = 6 mm) could be generated from a force of 0.9 N (Rungseevijitprapa and Bodmeier, 2009). This force is nearly equivalent to the previously stated pinch force required to press a remote button with a diameter of 13.6 mm (near the syringe-plunger diameter) (Smaby et al., 2004). Mean flow rate was taken as an injectability indicator for the prepared formulae (Paul et al., 1998). Table 6 shows the time taken and mean flow rate for the prepared formulae.
It can also be observed, from Table 6, that upon increasing polymer concentration, flow time increased and flow rate decreased under the same conditions. It might have been the result of the previously observed and discussed increase in viscosity resulting from higher polymer concentration (Ito et al., 2007; You et al., 2006).
Table 3. Rheological parameters of the prepared in situ implants formulae.
Formulae containing CA (I1-I4) had, relatively, the high-est flow rate. On the other hand, formulae containing tria-cetin (I11-I15) or Pluronic (I16-I18) showed a lower flow rate than that without it. It might have been caused by the high viscosity of the incorporated liquids (DesNoyer and McHugh, 2003; Perova et al., 1998). Formulae of ISILs (IL-1 and IL-2) had a flow rate of 1.94 and 1.40 mL/min, respectively, which were relatively higher than most of the triacetin and Pluronic containing formulae.
Effect of viscosity on in vitro release and injectabilityIt can be observed, from Figure 7, that formulae with high viscosity showed a slower release rate and higher t
50% values. These results are in accord with the previously
reported by Kranz et al., who studied the relation between viscosity of in situ implant formulae and their release rate and found the inverse proportionality between them (Kranz et al., 2008).
Figure 8 shows the correlation between mean flow rate as injectability indicator and viscosity measured at the 50 sec−1 rate of shear. Increasing viscosity led to lower flow rate and injectability and vice versa.
conclusion
Burst release is one of the major drawbacks that hin-der the wide use of in situ implant. The effect of several techniques was studied to overcome this burst release and decrease the drug release rate. Although triacetin decreased burst drug release from PDL and PDLG in situ implant formulae, it showed a disadvantage of its hydro-lysis into acetic acid, which accelerated the PDL or PDLG hydrolysis, resulting in a triphasic release. Triacetin also increased solution viscosities and decreased their inject-ability. On the other hand, Pluronic L121 failed to over-come the burst release of PDLG in situ implant and, on the contrary, burst release increased. Finally, lipospheres succeeded at decreasing the drug burst release for both PDL and PDLG in situ implant formulae (better than Pluronic L121) without affecting the hydrolysis rate of lac-tide polymers and drug release pattern from them (better than triacetin).
acknowldgement
The authors would like to thank all colleagues and staff members in Pharmaceutics Department, College of Pharmacy, Cairo University.
Table 4. Regression coefficient values yielded from Bingham, Casson, and Carreau equations.
The authors report no conflict of interest. The authors alone are responsible for the content and writing of this paper.
ReferencesAhmed, A. R., Dashevsky, A., Bodmeier, R. (2008). Reduction in
burst release of PLGA microparticles by incorporation into cubic phase-forming systems. Eur J Pharm Biopharm 70: 765–769.
Barakat, N. S., Yassin, A. E. (2006). In vitro characterization of carbamazepine-loaded precifac lipospheres. Drug Deliv 13:95–104.
Barnes, H. A., Hutton, J. E., Walters, K. (1993). An introduction to rheology. Amsterdam: Elsevier.
Bernkop-Schnurch, A., Hornof, M., Zoidl, T. (2003). Thiolated polymers—thiomers: synthesis and in vitro evaluation of chitosan-2-iminothiolane conjugates. Int J Pharm 260:229–237.
Blanco, D., Alonso, M. J. (1998). Protein encapsulation and release from poly(lactide-co-glycolide) microspheres: effect of the protein and polymer properties and of the co-encapsulation of surfactants. Eur J Pharm Biopharm 45:285–294.
Boylan, J. C., Nail, S. L. (2009). Parenteral products. In: Florence, A. T., Siepmann, J. (Eds.), Modern pharmaceutics: basic principles and systems, 5th ed. (pp. 565–609). New York: Informa.
Brodbeck, K. J., DesNoyer, J. R., McHugh, A. J. (1999). Phase inversion dynamics of PLGA solutions related to drug delivery. Part II. The role of solution thermodynamics and bath-side mass transfer. J Control Release 62:333–344.
Brookfield. (2010). Laboratory viscometer application data sheet. Brookfield Engineering Laboratories. Available at: http://www.brookfieldengineering.com/education/applications/laboratory-dishwasher-detergent.asp. Accessed on April 2, 2011.
Carrasquillo, K. G., Stanley, A. M., Aponte-Carro, J. C., De Jesus, P., Costantino, H. R., Bosques, C. J., et al. (2001). Non-aqueous encapsulation of excipient-stabilized spray-freeze dried BSA into poly(lactide-co-glycolide) microspheres results in release of native protein. J Control Release 76:199–208.
Chitkara, D., Shikanov, A., Kumar, N., Domb, A. J. (2006). Biodegradable injectable in situ depot-forming drug delivery systems. Macromol Biosci 6:977–990.
Cho, K. Y., Lee, S., Park, J. K. (2007). In situ micro-sized gel-forming injectable implant using biodegradable amphiphilic graft copolymer. Macromol Biosci 7:784–788.
Cortesi, R., Esposjto, E., Luca, G., Nastruzzi, C. (2002). Production of lipospheres as carriers for bioactive compounds. Biomaterials 23:2283–2294.
Cui, F., Cun, D., Tao, A., Yang, M., Shi, K., Zhao, M., et al. (2005). Preparation and characterization of melittin-loaded poly (DL-lactic acid) or poly (DL-lactic-co-glycolic acid) microspheres made by the double emulsion method. J Control Release 107:310–319.
DesNoyer, J. R., McHugh, A. J. (2003). The effect of Pluronic on the protein release kinetics of an injectable drug delivery system. J Control Release 86:15–24.
Dunn, R. L., English, J. P., Cowsar, D. R., Vanderbilt, D. P. (1990). Biodegradable in situ forming implants and methods of producing the same. US Patent 4938763.
Esposito, E., Cortesi, R., Nastruzzi, C. (2005). Production of lipospheres for bioactive compound delivery. In: Nastruzzi, C. (Ed.), Lipospheres in drug targets and delivery: approaches, methods, and applications (pp. 23–40). Boca Raton, Florida, USA: CRC.
Gad, H. A., El-Nabarawi, M. A., Abd El-Hady, S. S. (2008). Formulation and evaluation of PLA and PLGA in situ implants containing
secnidazole and/or doxycycline for treatment of periodontitis. AAPS PharmSciTech 9:878–884.
Graham, P. D., Brodbeck, K. J., McHugh, A. J. (1999). Phase inversion dynamics of PLGA solutions related to drug delivery. J Control Release 58:233–245.
Graves, R. A., Freeman, T., Mandal, T. K. (2007). In vitro dissolution method for evaluation of buprenorphine in situ gel formulation: a technical note. AAPS PharmSciTech 8:E62.
Higuchi, T. (1963). Mechanism of sustained-action medication. Theoretical analysis of rate of release of solid drugs dispersed in solid matrices. J Pharm Sci 52:1145–1149.
Ito, F., Fujimori, H., Makino, K. (2007). Incorporation of water-soluble drugs in PLGA microspheres. Coll Surf B Biointerf 54:173–178.
Jain, R. A. (2000). The manufacturing techniques of various drug loaded biodegradable poly(lactide-co-glycolide) (PLGA) devices. Biomaterials 21:2475–2490.
Jain, R. A., Rhodes, C. T., Railkar, A. M., Malick, A. W., Shah, N. H. (2000a). Controlled delivery of drugs from a novel injectable in situ formed biodegradable PLGA microsphere system. J Microencapsul 17:343–362.
Jain, R. A., Rhodes, C. T., Railkar, A. M., Malick, A. W., Shah, N. H. (2000b). Controlled release of drugs from injectable in situ formed biodegradable PLGA microspheres: effect of various formulation variables. Eur J Pharm Biopharm 50:257–262.
Jiang, G., Woo, B. H., Kang, F., Singh, J., DeLuca, P. P. (2002). Assessment of protein release kinetics, stability, and protein polymer interaction of lysozyme encapsulated poly(D,L-lactide-co-glycolide) microspheres. J Control Release 79:137–145.
Kranz, H., Yilmaz, E., Brazeau, G. A., Bodmeier, R. (2008). In vitro and in vivo drug release from a novel in situ forming drug delivery system. Pharm Res 25:1347–1354.
Lacout, J.-L., Freche, M., Goncalves, S., Rodriguez, F. (2005). Method for preparing a calcium phosphate pasty material for injection. US Patent 6923989.
Leroux, L., Hatim, Z., Frèche, M., Lacout, J. L. (1999). Effects of various adjuvants (lactic acid, glycerol, and chitosan) on the injectability of a calcium phosphate cement. Bone 25:31S–34S.
Li, F., Li, S., Ghzaoui, A. E., Nouailhas, H., Zhuo, R. (2007). Synthesis and gelation properties of PEG-PLA-PEG triblock copolymers obtained by coupling monohydroxylated PEG-PLA with adipoyl chloride. Langmuir 23:2778–2783.
Lu, Y., Yu, Y., Tang, X. (2007). Sucrose acetate isobutyrate as an in situ forming system for sustained risperidone release. J Pharm Sci 96:3252–3262.
Luan, X., Bodmeier, R. (2006). Influence of the poly(lactide-co-glycolide) type on the leuprolide release from in situ forming microparticle systems. J Control Release 110:266–272.
Matschke, C., Isele, U., van Hoogevest, P., Fahr, A. (2002). Sustained-release injectables formed in situ and their potential use for veterinary products. J Control Release 85:1–15.
McHugh, A. J. (2005). The role of polymer membrane formation in sustained release drug delivery systems. J Control Release 109:211–221.
McHugh, A. J., Desnoyer J. R. (2007). Injectable system for controlled drug delivery.
Moncalvo, D., Friedel, L. (2010). A viscosity correction factor for shear-thinning liquid flows in safety valves. J Loss Prevention Process Ind 23:289–293.
Nasr, M., Mansour, S., Mortada, N. D., El Shamy, A. A. (2008). Lipospheres as carriers for topical delivery of aceclofenac: preparation, characterization, and in vivo evaluation. AAPS PharmSciTech 9:154–162.
Packhaeuser, C. B., Schnieders, J., Oster, C. G., Kissel, T. (2004). In situ forming parenteral drug delivery systems: an overview. Eur J Pharm Biopharm 58:445–455.
Paradkar, A., Maheshwari, M., Tyagi, A. K., Chauhan, B., Kadam, S. S. (2003). Preparation and characterization of flurbiprofen beads by melt solidification technique. AAPS PharmSciTech 4:E65.
Patel, R. B., Carlson, A. N., Solorio, L., Exner, A. A. (2010a). Characterization of formulation parameters affecting low
molecular weight drug release from in situ forming drug delivery systems. J Biomed Mater Res A 94:476–484.
Patel, R. B., Solorio, L., Wu, H., Krupka, T., Exner, A. A. (2010b). Effect of injection site on in situ implant formation and drug release in vivo. J Control Release 147:350–358.
Paul, S., Egbert, J. E., Walsh, A. W., Hoey, M. F. (1998). Pressure measurements during injection of corticosteroids. Med Biol Eng Comput 36:729–733.
Peppas, N. A., Sahlin, J. J. (1989). A simple equation for the description of solute release. III. Coupling of diffusion and relaxation. Int J Pharm 57:169–172.
Perova, T. S., Christensen, D. H., Rasmussen, U., Vij, J. K., Nielsen, O. F. (1998). Far-infrared spectra of highly viscous liquids: glycerol and triacetin (glycerol triacetate). Vibrational Spectroscopy 18: 149–156.
Pluta, J., Karolewicz, B. (2006). In vitro studies of the properties of thermosensitive systems prepared on Pluronic F-127 as vehicles for methotrexate for delivery to solid tumours. Polim Med 36:37–53.
Raman, C., McHugh, A. J. (2005). A model for drug release from fast phase inverting injectable solutions. J Control Release 102: 145–157.
Ramchandani, M., Robinson, D. (1998). In vitro and in vivo release of ciprofloxacin from PLGA 50:50 implants. J Control Release 54:167–175.
Ribeiro, H. M., Morais, J. A., Eccleston, G. M. (2004). Structure and rheology of semisolid o/w creams containing cetyl alcohol/non-ionic surfactant mixed emulsifier and different polymers. Int J Cosmet Sci 26:47–59.
Ritger, P. L., Peppas, N. A. (1987). A simple equation for description of solute release II. fickian and anomalous release from swellable devices. J Control Release 5:37–42.
Rungseevijitprapa, W., Bodmeier, R. (2009). Injectability of biodegradable in situ forming microparticle systems (ISM). Eur J Pharm Sci 36:524–531.
Sepulveda, E., Kildsig, D. O., Ghaly, E. S. (2003). Relationship between internal phase volume and emulsion stability: the cetyl alcohol/stearyl alcohol system. Pharm Dev Technol 8:263–275.
SIDS. (2002). Triacetin. UNEP. Available at: http://www.inchem.org/documents/sids/sids/102761.pdf. Accessed on March 11, 2011.
Smaby, N., Johanson, M. E., Baker, B., Kenney, D. E., Murray, W. M., Hentz, V. R. (2004). Identification of key pinch forces required to complete functional tasks. J Rehabil Res Dev 41: 215–224.
Solorio, L., Babin, B. M., Patel, R. B., Mach, J., Azar, N., Exner, A. A. (2010). Noninvasive characterization of in situ forming implants using diagnostic ultrasound. J Control Release 143: 183–190.
Suzuki, Y., Nishimura, Y., Tanihara, M., Suzuki, K., Nakamura, T., Shimizu, Y., et al. (1998). Evaluation of a novel alginate gel dressing: cytotoxicity to fibroblasts in vitro and foreign-body reaction in pig skin in vivo. J Biomed Mater Res 39:317–322.
Vernon, B. L., Fusaro, F., Borden, B., Roy, K. H. (2004). Partition-controlled progesterone release from waterborne, in situ-gelling materials. Int J Pharm 274:191–200.
Vlad, S., Oprea, S. (2001). Evlauation of rheological behaviour of some thermoplastic polyurethane solutions. Eur Pol J 37:2461–2464.
Yoon, J. J., Chung, H. J., Park, T. G. (2007). Photo-crosslinkable and biodegradable Pluronic/heparin hydrogels for local and sustained delivery of angiogenic growth factor. J Biomed Mater Res A 83:597–605.
You, Y., Lee, S. J., Min, B.-M., Park, W. H. (2006). Effect of solution properties on nanofibrous structure of electrospun poly(lactic-co-glycolic acid). J Appl Polym Sci 99:1214–1221.
