Solid-State Characterization and Dissolution Properties of Lovastatin Hydroxypropyl-β-Cyclodextrin Inclusion Complex

February 2, 2007
Pharmaceutical Technology
Volume 31, Issue 2

The objectives of this study were to prepare and characterize inclusion complexes of lovastatin with hydroxypropyl-β-cyclodextrin (HPβ-CD) and to study the effect of the complexes on the dissolution rate of lovastatin (LVS). The findings suggest that LVS's poor dissolution profile can be overcome by preparing its inclusion complex with HPβ-CD.

Lovastatin (LVS) is a well-known compound for lowering plasma cholesterol levels. After oral administration, the inactive parent lactone is hydrolyzed to the corresponding hydroxyacid form. The hydroxyacid is the principal metabolite and a potent inhibitor of 3-hydroxy-3-methylglutaryl-coenzyme A (HMG CoA) reductase. This enzyme catalyzes the conversion of hydroxymethylglutarate to mevalonate, which is an early and rate-limiting step in cholesterol biosynthesis (1, 2).

LVS is white crystalline powder that is insoluble in water (0.4 g/mL). At room temperature, the partition coefficient of LVS in n-octanol/water system is approximately: Ko/w = 1.2 × 104 (3). Low aqueous solubility of LVS leads to inadequate dissolution in gastrointestinal fluids and, hence, poor absorption, distribution, and targeted organ delivery. The improvement of aqueous solubility in such a case is a valuable goal to improve therapeutic efficacy.

Cyclodextrin (CD) is a cyclic (α-1, 4)-linked oligosaccharide made of α-D-gluco-pyranose units (see Figure 1). Hydroxypropyl-β-cyclodextrin (HPβ-CD) is more water soluble than the parent molecule and has hydroxypropylester groups attached to the hydroxyl groups in position 2. The molecule has a cone-like configuration with a hydrophilic surface and a lipophilic cavity. In this cavity, hydrophobic molecules interact with lipophilic molecules without forming any covalent bonds and can produce so-called "inclusion complexes," which increase the water solubility and stability of the drug substance (4–6). Complexation with CDs has been reported to enhance the solubility, dissolution rate, and bioavailability of poorly water-soluble drugs. CDs first gained attention in marketed products as drug delivery technologies that enabled the development of various prostaglandins (7). The inclusion complex of rofecoxib/HPβ-CD (1:1 molar ratio) was prepared by Baboota et al. using a kneading method with a subsequent improvement in dissolution caused by amorphization (8). Several other drugs such as ganciclovir, nimesulide, itraconazole, and tolbutamide have been tested for CD inclusion to enhance solubility (9–12).

Figure 1: The structure formula for the β-cyclodextrin molecule.

β-CD has ideal dimensions to complex a range of commonly used drugs. Unfortunately, it has the limitation of a high affinity for cholesterol, which may lead to crystallization of a poorly water-soluble β-CD–cholesterol complex in the kidney. This complex could cause nephrotoxicity. HPβ-CD, a chemical derivative of β-CD, similarly improves the aqueous solubility of many drugs, but it is more hydrophilic than β-CD, forms a less-stable complex with cholesterol, and, therefore, is less toxic (13).

In this study, the authors compare the similarities of in vitro dissolution profiles of LVS from complexes, physical mixture, and pure LVS. Dissolution profiles can be compared by calculating a similarity factor (f2) and the mean dissolution time (MDT). The method for calculating the similarity factor was first reported by Moore and Flanner (14). It also has been adopted by the US Food and Drug Administration's Center for Drug Evaluation and Research (15) and by the Human Medicines Evaluation Unit of the European Medicines Agency (16) as a criterion for assessing the similarity of two dissolution profiles (17, 18). A similarity factor of 100% suggests that the test and reference profiles are identical. Values between 50 and 100 indicate that the dissolution profiles are similar, whereas smaller values imply an increase in dissimilarity between release profiles (14). MDT reflects the time for the drug to dissolve and is the first statistical moment for the cumulative dissolution process that provides an accurate drug-release rate (15). A higher MDT value indicates greater drug-retarding ability (16).

The present study was intended to improve the aqueous solubility and dissolution rate of LVS by preparing its complexes with HPβ-CD using various methods such as kneading, coevaporation, and physical mixing. The study further aimed to characterize the interaction between LVS and HPβ-CD.

Materials and methods

Materials. HPβ-CD was a gift sample from Roquette Frères, (Lestrem, France). LVS was received as a gift sample from Lincoln Pharmaceuticals Ltd. (Ahmedabad, India). The samples of sodium lauryl sulfate (SLS) were purchased from S.D. Fine Chemicals, (Vadodara, India). Directly compressible lactose, maize starch, sodium starch glycolate, colloidal silicon dioxide, and magnesium stearate were received as gift samples from Maan Pharmaceuticals Ltd. (Ahmedabad, India). All chemicals and solvents used in this study were of analytical reagent grade. Freshly distilled water was used throughout the work.

Phase-solubility study. Phase-solubility studies were performed according to the method reported by Higuchi and Connors (19). LVS, in amounts that exceeded its solubility, was transferred to screw-capped vials containing 25 mL of an aqueous solution of HPβ-CD (molecular weight = 1500 g/mol) in various molar concentrations (0, 2.0, 4.0, 6.0, 8.0, 10.0, 12.0, and 14.0 mM/L). The contents were stirred with an electromagnetic stirrer (Remi, Mumbai, India) for 36 h at 37 °C ± 0.1 °C and 350 rpm (this duration was previously tested to be sufficient to reach equilibrium). After reaching equilibrium, samples were filtered through a 0.22-μm membrane filter, suitably diluted, and analyzed spectrophotometrically for drug content at the wavelength of 238.2 nm using a spectrophotometer (Shimadzu-1601, ultraviolet-vis spectrophotometer, Shimadzu Corp., Kyoto, Japan). Solubility studies were performed in triplicate (n = 3). The apparent stability constant (Kc), according to the hypothesis of 1:1 stoichiometric ratio of complexes, was calculated from the phase-solubility diagrams using the following equation:

in which the slope is obtained from the initial straight-line portion of the plot of LVS concentration against HPβ-CD concentration, and S0 is the equilibrium solubility of LVS in water.

Preparation of inclusion complexes. Complexes of HPβ-CD and LVS were prepared in the molar ratio of 1:1 (on the basis of the phase solubility study) by various methods such as physical mixture, coevaporation, and kneading.

Physical mixture. A physical mixture of HPβ-CD and LVS was prepared by mixing the powders with a spatula for 15 minutes.

Coevaporation method.Methanol and water were used as solvents to prepare the complex by a coevaporation method. The required quantities of LVS and HPβ-CD were dissolved in methanol and water, respectively. Both the solutions were mixed and solvents were evaporated by controlled heating at 45–50 °C. The resultant solid was pulverized and then put through a 120 # sieve.

Kneading method. The required quantities of HPβ-CD and distilled water were mixed together in a motor to obtain a homogeneous paste. Then, LVS was added slowly; while grinding, a small quantity of methanol was added to assist the dissolution of LVS. The mixture was then ground for 1 h. During this process, an appropriate quantity of water was added to the mixture to maintain a suitable consistency. The paste was dried in an oven at 45–50 °C for 24 h. The dried complex was pulverized and then put through a 120 # sieve.

Drug content. The complexes prepared by kneading, coevaporation, and physical mixture were assayed for LVS content by dissolving a specific amount of the complex in methanol and analyzing for the LVS content spectrophotometrically at 238.2 nm on spectrophotometer (ultraviolet-vis spectrophotometer, Shimazdu-1601). The final moisture contents of all samples were measured with an electronic moisture balance (Sartorius, model MA-45, Goettingen, Germany).

Characterization of complexes. Infrared (IR) spectroscopic analysis. The IR spectra of moisture-free powdered samples of LVS, HPβ-CD, the physical mixture, and the complex prepared by the coevaporation and the kneading methods were obtained using a spectrometer (FTIR-8300, Shimadzu Co., Kyoto, Japan) with a potassium bromide (KBr) pellet method.

Powder X-ray diffraction (PXRD) analysis. PXRD patterns of LVS, HPβ-CD, the physical mixture, and complexes prepared by the coevaporation and the kneading methods were determined using a scanner (Phillips PW 3710) and a generator (IW 1830) with a CuK α anode at 40 kV and 30 mA at a scan rate of 1° min–1 from 2θ range from 1° to 40°.

Differential scanning calorimetry (DSC) analysis. DSC scans of the powdered sample of LVS, β-CD, the physical mixture, and complexes prepared by the coevaporation and the kneading methods were recorded using a DSC instrument (Shimadzu 60) with TDA trend-line software. The samples (6–7 mg) were weighed accurately in crimped aluminum pans and heated from 50 °C to 300 °C at a scanning rate of 10 °C /min under dry nitrogen flow (100 mL/min).

Wettability and dissolution studies. The wettability study was performed using open tubes containing LVS, the physical mixture, and complexes prepared by the coevaporation and the kneading methods. The tubes were placed with their lower capillary ends dipped into colored water (0.01% eosin in water). The upward migration of the colored front was registered as a function of time. The porosity of all samples also was measured using a mercury porosimeter (PoreMaster 60, Quantachrome Instruments, Boynton Beach, FL). Porosity is defined as the percentage of void space in a solid. Mercury density (ρHg) and helium density (PHe) values often are used to evaluate % porosity (€) (20). Each test was repeated four times and the mean was calculated.

Dissolution studies of LVS in powder form, the physical mixture, and complexes prepared by the coevaporation and the kneading methods were performed to evaluate the in vitro drug-release profile. Dissolution studies were carried out using a USP dissolution apparatus type II with a 500-mL dissolution medium at 37 °C ± 0.5 °C and 50 rpm for 3 h. In addition, 0.1 N HCl and a phosphate buffer (pH 6.8) containing 0.25% (weight/volune [w/v]) of sodium lauryl sulfate were used as other dissolution mediums. At fixed time intervals, 5-mL aliquots were withdrawn, filtered, suitably diluted, and assayed for LVS content by measuring the absorbance at 238.2 nm using a spectrophotometer. An equal volume of fresh medium at the same temperature was placed into the dissolution medium after each sampling to maintain its constant volume throughout the test.

Table I: Tablet formulations of plain lovastatin (LVS), the physical mixture (PM), and complexes made with the coevaporation (CPC) and the kneading methods (CPK).

Pure drug, the physical mixture, and complexes prepared by the coevaporation and the kneading methods were evaluated in dissolution-rate studies. Dissolution studies were performed in triplicate (n = 3) and the calculated mean values of cumulative drug release were used while plotting the release curves. MDT values were calculated to compare the extent of improvement in the dissolution rate of the physical mixture, and the complexes prepared by the coevaporation and the kneading methods. Preliminary tests demonstrated that there was no change in the λmax of LVS because of the presence of HPβ-CD dissolved in the dissolution medium.

Table II: Gibbs free energy of transfer (ΔGtr°) for the solubilization process of lovastatin in aqueous solutions of hydroxypropyl-β-cyclodextrin (HP-βCD) at 37 8C.

Formulation studies. Formulation excipients were selected on the basis of preliminary tests that demonstrated the excipients did not interfere with the λmax of LVS. The % compressibility and the angle of repose were measured to assess the tableting ability of the kneading method.

The % compressibility was calculated using the following equation:

in which V0 is the powder volume before tapping and Vf is powder volume after tapping infinitely. The powder volume was calculated from Kawakita's equation (21) using data from 200 tapping points obtained with a tapping density analyzer (bulk density test apparatus, DBK Instruments, Mumbai, India) equipped with a 20-mL cylinder.

For the angle of repose measurement, a pile of powder was carefully made by dropping the powder material through a funnel tip from a height of 2 cm (22). The angle of repose was calculated by tangentially inverting the ratio of the formed pile's height and radius.

Tablets containing 10 mg of LVS were made by direct compression using various formulation excipients such as directly compressible lactose, colloidal silicon dioxide, and magnesium stearate (see Table I). Tablets equivalent to 10 mg LVS were made similarly using the kneading method, but using less lactose. The blend was compressed on an eight-station, single-rotary machine (Cadmach, India) using round-shaped, flat punches to obtain tablets of 4–7 kg/cm2 hardness and 3.5–3.7 mm thickness (see Table I). For the assay, three tablets were crushed and a blend equivalent to 10 mg of LVS was weighed and dissolved in dissolution mediums. The tablets were studied in triplicate (n = 3) for the drug-release profile using the same methodology described for the in vitro dissolution studies.

Figure 2: The phase solubility curve of lovastatin in an aqueous solution of hydroxypropyl-β-cyclodextrin (HP-βCD) at 37 8C.

Statistical analysis. A model-independent, mathematical approach proposed by Moore and Flanner (14) for calculating f2 was used for comparison among the dissolution profiles of various samples. The f2 is a measure of similarity in the percentage dissolution between two dissolution curves and is defined by following equation:

in which n is the number of withdrawal points, Rt is the percentage dissolved of reference at the time point t, and Tt is the percentage dissolved of test at the time point t (14).

A value of 100% for f2 suggests that the test and reference profiles are identical. Values between 50 and 100 indicate that the dissolution profiles are similar, whereas smaller values imply an increase in dissimilarity between release profiles (14).

Results and discussion

Phase-solubility study. Phase-solubility analysis has been among the preliminary requirements for the optimization of the development into inclusion complexes of the drugs as it permits an evaluation of the affinity between β-CD and drug molecule in water. This process has been used by many researchers to determine the exact molar ratios in which the drugs could make complexes with β-CD (23, 24).

The solubility of LVS in water at 25 °C is 0.4 g/mL; therefore, LVS can be considered a water-insoluble drug. The phase-solubility diagram of LVS in the presence of HPβ-CD was obtained by plotting the apparent equilibrium concentration of LVS against various molar concentrations of HPβ-CD (see Figure 2). This representation provides direct information about the complexation efficiency. The apparent solubility of LVS increased linearly as a function of HPβ-CD concentration over the entire concentration range studied. The solubility of LVS is increased 17.2-fold at a 14 mM/L concentration of HPβ-CD. Increased solubility may be caused by improved dissolution of LVS particles in water by HPβ-CD.

Linearity was characteristic of the AL-type system (19) and suggested that water-soluble complexes formed in the solution. Furthermore, the slope value (0.0011) was lower than 1.0, thus indicating that inclusion complexes in a molar ratio of 1:1 formed between the guest (LVS) and host (HPβ-CD) molecules. The stability constant (Kc) for the complexes at 37 °C (assuming a 1:1 stoichiometry) calculated from the slope of the initial straight portion of the solubility diagram was 917.67 M–1 , which indicated a suitable and stable complex formation. It is reported that cyclodextrin-drug complexes with Kc values in the range of 200–5000 M–1 show improved dissolution properties and, hence, better bioavailabilities (19).

The values of Gibbs free energy change (ΔGtr°) were calculated to understand transfer process of LVS from pure water to aqueous solution of HPβ-CD. The ΔGtr° values of LVS from pure water to aqueous solutions with various concentrations of HPβ-CD at 37 °C were calculated using the following equation (25):

in which So/Ss is the ratio of molar solubility of LVS in an aqueous solution of HPβ-CD to that of the pure water. The obtained values of Gibbs free energy are shown in Table II. The ΔGtr° values provide information about whether the reaction condition is favorable or unfavorable for drug solubilization in the aqueous carrier solution. Negative Gibbs free energy values indicate favorable conditions. The ΔGtr° values were all negative for HPβ-CD at various concentrations, thus indicating the spontaneous nature of LVS solubilization. These values decreased with increased concentration of HPβ-CD, thereby demonstrating that the reaction became more favorable as the concentration of HPβ-CD increased.

Drug content. The drug content (%w/w) of the complexes prepared by the coevaporation and the kneading methods and the physical mixture were found to be 21.02 (±0.23)%, 20.78 (±0.39)%, and 14.74 (±5.35)%, respectively, which corresponds approximately with the stoichiometric ratio of the complex and indicates chemical stability and content uniformity of LVS in its complex form. The content uniformity for the physical mixture is lower than the complexes prepared by the coevaporation and the kneading methods. This effect may be attributed to insufficient mixing caused by simple mixing of powders without applying pressure (HPβ-CD and LVS). The final moisture content of the pure drug, the physical mixture, and the complex prepared by the coevaporation and the kneading methods were 0.23%, 0.89%, 1.09%, and 1.14%, respectively.

Characterization of complexes. IR spectroscopic analysis. Fourier transform infrared spectroscopy (FT IR) has been used to assess the interaction between β-CD and guest molecules in the solid state. Upon complexion, the peak band of the guest shifts in the absorption spectrum. Nonetheless, some of the changes are very subtle and require careful interpretation of the spectrum (26).

The IR spectra of LVS, HPβ-CD, the physical mixture, and complexes prepared by the coevaporation and the kneading methods are presented in Figure 3. The spectrum of pure LVS presented characteristic peaks at 3510 cm–1 (alcohol O–H stretching vibration); 3016 cm–1 (olefinic C–H stretching vibration); 2970, 2930, and 2876 cm–1 (methyl and methylene C–H stretching vibration); 1725, 1713, and 1690 cm–1 (lactone and ester carbonyl stretch [hydrogen bonded for 1711 and 1700 cm–1 ]); 1430, 1378, and 1350 cm–1 (methyl and methylene bending vibration); 1275, 1228, 1080, and 1050 cm–1 (lactone and ester C–O–C bending vibration); 972 cm–1 (alcohol C–OH stretch); and 873 cm–1 (trisubstituted olefinic C–H wag).

Figure 3: Fourier transform infrared spectrograms of (a) lovastatin, (B) hydroxypropyl-β-cyclodextrin, (c) the physical mixture, and complexes prepared by (d) the coevaporation method and (e) the kneading method.

The IR spectrum of the HPβ-CD is characterized by intense bands at 3300–3500 cm–1 because of O–H stretching vibrations. The vibration of the –CH and CH2 groups appears in the 280–3000-cm–1 region. The spectrum patterns of the physical mixture correspond with the superposition of the IR spectra of the two components. The absence of characteristic peaks of LVS and the presence of characteristic peaks caused by HPβ-CD in the IR spectra of complexes prepared by the coevaporation and the kneading methods indicate that LVS is inside the cavity of HPβ-CD. The IR spectra of the physical mixture and complexes prepared by the coevaporation method and the kneading methods showed most of the characteristic peaks were similar to that of HPβ-CD, except one peak at 1700 cm–1 for lactone and ester carbonyl stretching vibration (hydrogen bonded for 1711 and 1700 cm–1 ), which is characteristic of LVS. This effect indicates that the pyrol part of LVS remains outside the HPβ-CD, whereas the remaining part fits inside the HPβ-CD cavity.

DSC analysis. DSC is a method that confirms the formation of a complex in the solid state. The disappearance of thermal events of guest molecules when they are examined as CD complexes is generally taken as a proof of real inclusion (27, 28). The DSC scans for pure LVS, HPβ-CD, the physical mixture, and complexes prepared by the coevaporation and the kneading methods are presented in Figure 4. The LVS showed a melting endotherm at 173.7 °C with enthalpy of fusion (ΔH) 104.282 J/g. In the thermogram of the HPβ-CD, the endothermic peak near 100 °C was caused by loss of water from HPβ-CD molecules. In the thermogram of the physical mixture, a sharp endotherm was observed at the same position as that of LVS, thus indicating the presence of untrapped LVS. A characteristic sharp endothermic peak of LVS in the range of 171 °C to 176 °C was absent in the thermograms of complexes prepared by the coevaporation and the kneading methods, indicating partial amorphization of the drug and trapping of LVS inside the HPβ-CD cavity.

Figure 4: Differential scanning calorimetry thermograms of (a) lovastatin, (b) hydroxypropyl-β-cyclodextrin, (c) the physical mixture, and complexes prepared by (d) the coevaporation method and (e) the kneading method.

PXRD analysis. PXRD can provide useful information about the inclusion complex of CDs and other polymers. Their structure can be classified as cage-type or column-type (29). The PXRD patterns of pure LVS, the physical mixture, and complexes prepared by the coevaporation and the kneading methods are presented in Figure 5. The PXRD patterns of pure LVS showed numerous sharp peaks, which are the characteristic of a crystalline compound. In contrast, PXRD patterns of HPβ-CD lacked crystalline peaks, which is the characteristic of an amorphous compound. Some drug crystallinity peaks were still detectable in the physical mixture shown in Figure 5c. The PXRD pattern consisted mostly of the HPβ-CD character, but some of the LVS characteristics remained. Compared with the PXRD patterns of pure LVS and HPβ-CD, the PXRD patterns of complexes prepared by the coevaporation and the kneading methods (Figures 5d and 5e) were more related to that of amorphous HPβ-CD (Figure 5b). Moreover, no other peaks than those that could be assigned to the pure HPβ-CD and LVS were detected in the complexes, thus indicating the absence of chemical interactions in the solid state between the two entities. These results confirm that LVS is no longer present as a crystalline material and its HPβ-CD solid complexes exhibit amorphous nature.

Figure 5: X-ray diffractograms of (a) lovastatin, (b) hydroxypropyl-β-cyclodextrin, (c) the physical mixture, and complexes prepared by (d) the coevaporation method (e) and the kneading method.

Wettability and dissolution studies. The solubility of LVS in water was 40 g (±0.544)/100 mL. The kneading method, the coevaporation method, and the physical mixture improved the solubility of LVS to 720 g (±36.01)/100 mL, 569 g (±24.49)/100 mL, and 374 g (±11.155)/100 mL, respectively. Thus, upon complexation of LVS with HPβ-CD by kneading, coevaporation, and physical mixing, the solubility of LVS improved 18-fold, 14.23-fold, and 9.35-fold, respectively.

The improvement in wettability of LVS by the physical mixture, the kneading method, and the coevaporation method is shown in Figure 6. The kneading method and the coevaporation method showed highest wettability in water (99.82% and 90.04%, respectively), as compared with plain LVS (31.88%). Even the physical mixture of HPβ-CD with LVS enhances the wettability of LVS in water significantly. The % porosity of LVS, the physical mixture, and complexes prepared by the coevaporation and the kneading methods were 0.54 (±0.02), 0.72 (±0.08), 0.98 (±0.06), and 1.03 (±0.07), which ruled out a possible improvement in wettability because of variation in the samples' porosity.

Figure 8: In vitro dissolution profiles of pure lovastatin and its physical mixture and complexes in dissolution media-B (phosphate buffer [pH 6.8]) (tests were done in triplicate). CPK is the kneading method, CPC is the coevaporation method, PM is the physical mixure, and LVS is lovastatin.

It is generally accepted that the dissolution media are not completely representative of gastrointestinal conditions, yet it is proposed in guidelines that a good method will use a dissolution media that is physiologically meaningful or closely mimics in vivo conditions (30, 31). It has been suggested that the inclusion of surface active agents in dissolution media is important for poorly soluble compounds because the lack of surface tension-lowering agents would result in poorer wetting and in vitro dissolution rates that are not representative of in vivo rates (32). FDA has promoted the use of surfactants in media for conducting dissolution studies of poorly soluble compounds (33, 34).

Figure 9: Comparative drug-release profiles of conventional tablets containing lovastatin (LVS) and tablets containing a complex made with the kneading method (CPK) in dissolution media-A (0.1 N HCl) and dissolution media-B (phosphate buffer [pH 6.8]) (tests performed in triplicate).

Dissolution studies of pure LVS and all other prepared systems (complexes and physical mixture) were carried out in dissolution media (0.1 N HCl and phosphate buffer pH 6.8) containing aqueous sodium lauryl sulfate solution (0.25% w/v) because sodium lauryl sulfate showed minimal surface tension at 0.2% with no significant change at higher concentrations (35, 36). When the LVS was dispersed on the surface of the aqueous surfactant solution, LVS rapidly left the surface and was dispersed in the bulk of solution, which indicates wetting of LVS, unlike pure water.

The DP30 min (percent drug dissolved within 30 min) and t50% (time to dissolve 50% drug) values in 0.1 N HCl and the phosphate buffer (pH 6.8) are reported in Table III. (For ease in discussion, hereafter, abbreviations for 0.1 N HCl-dissolution media A [DM-A] and for phosphate buffer [pH 6.8]-dissolution media B [DM-B] are used.) From these data, it is evident that the onset of dissolution of pure LVS is very low in both dissolution media (DP30 min value 10.62% in DM-A and 13.75% in DM-B). Even t50% values in both dissolution media are much higher (>>3 h). The kneading and the coevaporation methods considerably enhanced DP30 min and lowered t50% values compared with pure LVS and the physical mixture. Figures 7 and 8 show the dissolution profiles of pure LVS, the physical mixture, and the complexes prepared by the coevaporation and the kneading methods in DM-A and DM-B, respectively, over a period of 3 h. It is evident that the dissolution rate of pure LVS is very low in both DM-A and DM-B: approximately 36.12% and 34.00% of the drug was dissolved in 3 h, respectively. The kneading method and the coevaporation method enhanced the dissolution rate of LVS significantly (90–100% in both dissolution media) within 3 h. Hence, the faster dissolution of LVS from the kneading method and the coevaporation method is attributed to the solubilizing effect of the carrier (HPβ-CD). In addition, other factors such as particle-size reduction, the absence of aggregation and agglomeration between hydrophobic drug particles, good wettability, and dispersibility of the dispersed drug (37) also might have contributed to the observed increase in the dissolution rate of LVS from complexes.

Table IV: Mean dissolution time (MDT) values for pure lovastatin (LVS), the physical mixture (PM), and complexes prepared by the coevaporation (CPC) and kneading (CPK) methods in dissolution media-A (0.1 N HCl) (DM-A) and dissolution media-B (phosphate buffer [pH 6.8]) (DM-B).

The dissolution rate of LVS from the physical mixture is significantly higher (74.94% in DM-A and 78.37 % in DM-B) than that of pure LVS within 3 h. Physical mixing of LVS with HPβ-CD brings the drug in close contact with HPβ-CD. The increased dissolution rate observed in the case of the physical mixture may be caused by one or more of the factors mentioned previously.

Figure 6: Wettability study of pure lovastatin, the physical mixture, and complexes prepared by the coevaporation and the kneading methods in water (tests performed in triplicate). CPK is the kneading method, CPC is the coevaporation method, PM is the physical mixure, and LVS is lovastatin.

The MDT value of LVS from pure LVS, the physical mixture, and complexes prepared by the coevaporation and the kneading methods were calculated (n = 3) using the equation:

in which i cr, n is the number of dissolution times, tmid is the time at the midpoint between times ti and ti–1, and ΔM is the amount of LVS dissolved (μg) between times ti and t i–1. The obtained values of MDT for pure LVS, the physical mixture, and complexes prepared by the coevaporation and the kneading method are presented in Table IV. MDT reflects the time for the drug to dissolve and is the first statistical moment for the cumulative dissolution process that provides an accurate drug-release rate (17). It is an accurate expression for drug-release rate. A higher MDT value indicates greater drug-retarding ability (18). The MDTs of LVS are 64.92 min in DM-A and 66.13 min in DM-B. These values decreased a greater extent after the preparation of a complex of LVS with HPβ-CD (e.g., 22.00 min in DM-A and 30.56 min in DM-B for the kneading method and 33.79 min in DM-A and 41.08 min in DM-B for the coevaporation method). The physical mixture also shows sufficiently lower MDT value compared with pure LVS in both dissolution media. The kneading method, which exhibited the best dissolution profile, was used for the formulation studies.

Figure 7: In vitro dissolution profiles of pure lovastatin and its physical mixture and complexes in dissolution media-A (0.1 N HCl) (tests performed in triplicate). CPK is the kneading method, CPC is the co-evaporation method, PM is the physical mixure, and LVS is lovastatin.

A value of 100% for the similarity factor (f2) suggests that the test and reference profiles are identical. Values between 50 and 100 indicate that the dissolution profiles are similar, whereas smaller values imply an increase in dissimilarity between release profiles (14). Calculated f2 values are presented in Table V. From this table, it is evident that the release profile of LVS prepared by the kneading method is very different from pure LVS in both dissolution media (f2 values 10.55 in DM-A and 11.64 in DM-B). Even release profiles of LVS prepared by the coevaporation method and the physical mixture also are significantly different from pure LVS in both dissolution media.

Table III: The % drug dissolved within 30 minutes (DP30 min) and the time to dissolve 50% of the drug (t50%) from pure lovastatin (LVS), the physical mixture (PM), and complexes made with the coevaporation (CPC) and the kneading methods (CPK) in DM-A* and DM-B**.

Formulation studies. The kneading method was studied for physical properties to assess its tableting ability. The % compressibility and angle of repose of the complex prepared by the kneading method were 11.92% and 26.2°, respectively, which indicate good compressibility and flow properties, thereby making the complex suitable for tableting. In general, compressibility index values up to 15% and an angle of repose between 25° and 30° results in good to excellent flow properties (38).

Table V: Similarity factor for the release profiles of lovastatin in dissolution media-A (0.1 N HCl) (DM-A) and dissolution media-B (phosphate buffer [pH 6.8]) (DM-B).

The release profile of LVS from tablets containing LVS without HPβ-CD and tablets containing LVS with HPβ-CD in both dissolution media is shown in Figure 9. During in vitro dissolution studies, the complex prepared by the kneading method exhibited approximately 95% and 85% drug release within 60 min from DM-A and DM-B, respectively. Tablets prepared by compressing the kneading method provided the same drug release within approximately 120 min. The t50% values of LVS also are significantly lowered in both the dissolution media.

The release of LVS from tablets made with the kneading method was faster and higher as compared with the tablets containing pure LVS and no HPβ-CD. Moreover, the hardness of all formulations is approximately the same, which indicates that improvement in the dissolution of LVS from tablets containing the complex made with the kneading method is not dependent on hardness. This confirmed the advantage of improved aqueous solubility of LVS in its complex form, which can be formulated as tablets with better dissolution characteristics. The release profiles of LVS from conventional tablets containing LVS are significantly different from tablets containing the complex made with the kneading method; the f2 values are 24.55 and 24.64 in DM-A and DM-B, respectively. The MDT values of LVS from tablets containing the complex made with kneading method in both dissolution media (46.02 min in DM-A and 55.64 min in DM-B) are significantly lower than that of conventional tablets containing only LVS and no HPβ-CD (87.96 min in DM-A and 92.96 min in DM-B).

Conclusion

This study showed a significant, linear increase in the aqueous solubility of lovastatin (LVS) with increasing concentrations of hydroxypropyl-β-cyclodextrin (HPβ-CD). The maximum studied concentration of HPβ-CD (14 mM/L) resulted in a 17.2-fold improvement in the saturation solubility of LVS. An inclusion complex of LVS and HPβ-CD in a molar ratio of 1:1 was prepared successfully with the kneading and coevaporation methods. The prepared complexes and the physical mixture of LVS and HPβ-CD were characterized by Fourier transform infrared, differential scanning calorimetry, and X-ray diffraction spectroscopy. When compared with the pure drug, the dissolution profile of the LVS and HPβ-CD complex is dramatically improved, which proved its suitability to develop an oral form. The inclusion complex prepared by the kneading method showed the highest improvement in in vitro drug release because of the presence of entrapped drug inside the HPβ-CD cavity. The absence of unentrapped drug also was well characterized by Fourier transform infrared, differential scanning calorimetry, and X-ray studies. The in vitro drug release of the physical mixture improved, too, but to a lesser extent compared with complexes prepared by the kneading and coevaporation methods. Tablets containing the complex prepared by the kneading method showed a better dissolution profile compared with tablets prepared using LVS without HPβ-CD. These findings suggested that LVS's poor dissolution profile can be overcome by preparing its inclusion complex with HPβ-CD.

Acknowlegment

The authors thank Roquette Frères, France for a generous gift of HPβ-CD. The authors thank Lincoln Pharmaceuticals Ltd. and Maan Pharmaceuticals Ltd., India for providing lovastatin and various formulation excipients, respectively as gift samples. The authors are grateful to the Department of Pharmacy, M.S. University, India for conducting differential scanning calorimetry studies and to Sun Pharmaceutical Advance Research Center, India for conducting X-ray diffraction studies of the samples.

Rakesh P. Patel* is an assistant professor and Madhabhai M. Patel is a professor and the head of the department of pharmaceutics, S K Patel College of Pharmaceutical Education and Research, Ganpat University, Ganpat vidyanagar, Kherva 382711, Mehsana, Gujarat, India, raka_77us@yahoo.com

*To whom all correspondence should be addressed.

Submitted: July 10, 2006. Accepted: Oct. 11, 2006.

Keywords: lovastatin, hydroxypropyl-β-cyclodextrin, inclusion complexation, in vitro dissolution studies.

References

1. J.S. MacDonald et al., "Preclinical Evaluation of Lovastatin," Am. J. Cardiol. 62 (15), 16J–27J (1998).

2. E.E. Slater and J.S. MacDonald, "Mechanism of Action and Biological Profile of HMG CoA Reductase Inhibitors: A New Therapeutic Alternative," Drugs 36 (Suppl. 3), 72–82 (1998).

3. S.B. Gerald, K.E Dean, and J.K. Micheal, "Lovastatin," in Analytical Profiles of Drug Substances and Excipients, Vol. 21, G.B. Harry, Ed., (Academic Press, Inc., New York, NY, 1992), pp. 277–315.

4. D.O. Thompson, "Cyclodextrins-Enabling Excipients: Their Present and Future Use in Pharmaceuticals," Crit. Rev. Ther. Drug Carrier Syst. 14 (1), 1–104 (1997).

5. C.M. Fernandes, V.M. Teresa, and F.J. Veiga, "Physicochemical Characterization and In Vitro Dissolution Behavior of Nicardipine-Cyclodextrins Inclusion Compounds," Eur. J. Pharm. Sci. 15 (11), 79–88 (2002).

6. M. Kamada et al., "Cyclodextrin Conjugate-Based Controlled Release System: Repeated- and Prolonged-Releases of Ketoprofen after Oral Administration in Rats," J. Control. Release 82 (2-32-3), 407–416 (2002).

7. K. Uekama and M. Otagiri, "Cyclodextrins in Drug Carrier Systems," Crit. Rev. Ther. Drug Carrier Syst. 3 (1), 1–40 (1987).

8. S. Baboota, M. Dhaliwal, and K. Kohli, "Physicochemical Characterization, In Vitro Dissolution Behavior, and Pharmacodynamic Studies of Rofecoxib-cyclodextrin Inclusion Compounds," AAPS PharmSciTech . 6 (1), E83–E90 (2005).

9. G. S. Tirucherai and A. K. Mitra, "Effect of Hydroxypropyl Beta Cyclodextrin Complexation on Aqueous Solubility, Stability, and Corneal Permeation of Acyl Ester Prodrugs of Ganciclovir," AAPS PharmSciTech. 4 (3), E45 (2003).

10. B. N. Nalluri et al.,"Physicochemical Characterization and Dissolution Properties of Nimesulide-Cyclodextrin Binary Systems," AAPS PharmSciTechnol . 4 (1), E2 (2003).

11. J. Peeters et al., "Characterization of the Interaction of 2-Hydroxypropyl-Betacyclodextrin with Itraconazole at pH 2, 4, and 7," J. Pharm. Sci . 91 (66), 1414–1422 (2002).

12. F. J. Veiga et al., "Molecular Modeling and 1H-NMR: Ultimate Tools for the Investigation of Tolbutamide: β-Cyclodextrin and Tolbutamide: Hydroxypropyl-β-Cyclodextrin Complexes," Chem. Pharm. Bull.(Tokyo). 49 (10), 1251–1256 (2001).

13. J. Szejtli, "Medicinal Applications of Cyclodextrins," Med. Res. Rev. 14 (3), 353–386 (1994).

14. J.W. Moore and H. Flanner, "Mathematical Comparison of Dissolution Profiles," Pharm. Technol. 20 (6), 64–74 (1996).

15. US Food and Drug Administration, Guidance for Industry SUPAC-MR: Modified Release Solid Oral Dosage Forms Scale-Up and Postapproval Changes: Chemistry, Manufacturing, and Controls; In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation (FDA, Rockville, MD, 1997).

16. Human Medicines Evaluation Unit, European Agency for the Evaluation of Medicinal Products (EMEA), Notes for Guidance on Quality of Modified-Release Products; A Oral Dosage Forms; B Transdermal Dosage Forms, Section 1 (Quality) (EMEA, London, UK, 1999).

17. C. Reppas and E. Nicolaides, "Analysis of Drug Dissolution Data," in Oral Drug Absorption Prediction and Assessment, J.B. Dressman and H. Lennernäs, Eds. (Marcel Dekker, Inc., New York, NY, 2000), pp. 229–254.

18. M.L. Vueba et al.,"Influence of Cellulose Ether Polymers on Ketoprofen Release from Hydrophilic Matrix Tablets," Eur. J. Pharm. and Biopharm. 58 (1), 51–59 (2004).

19. T. Higuchi and K. Connors, "Phase Solubility Techniques," in Advances in Analytical Chemistry and Instrumention, C.N. Reilly, Ed. (John Wiley, New York, NY, 1965) 4, 117–212.

20. S. Mattsson and C. Nyström, "The Use of Mercury Porosimetry in Assessing the Effect of Different Binders on the Pore Structure and Bonding Properties of Tablets," Eur. J. Pharm. Biopharm. 52 (2), 237–247 (2001).

21. K. Kawakita, "Compression of Powder," Kagaku . 26, 149–150 (1956).

22. R.L. Carr, "Evaluating Flow Properties of Solids," Chem. Eng. 72 (2), 163–168 (1965).

23. D. Peri et al.,"Inclusion Complexes of Tolnaftate with β-Cyclodextrin and Hydroxy-β-Cyclodextrin," Drug Dev. Ind. Pharm. 20 (88), 1401–1410 (1994).

24. J.J.T. Labenderia et al., "Glibornuride-β-Cyclodextrin Inclusion Complexes: Preparation, Structural Characterization, and In Vitro Dissolution Behavior," Eur. J. Pharm. and Biopharm. 39 (6), 255–259 (1993).

25. L. Chengsheng, L. Chenguang, and H.D. Kashappa, "Enhancement of Dissolution Rate of Valdecoxib Using Solid Dispersions with Polyethylene Glycol 4000," Drug Dev. Ind. Pharm. 31 (1), 1–10 (2005).

26. A.R. Hedges. "Industrial Applications of Cyclodextrins," Chemical Reviews 98 (5), 2035–2044 (1998).

27. T. Pralhad and K. Rajendrakumar, "Study of Freeze-Dried Quercetincyclodextrin Binary Systems by DSC, FT-IR, X-ray Diffraction and SEM Analysis," J. Pharm. Biomed. Anal. 34 (2), 333–339 (2004).

28. R.O. Williams, V. Mahaguna, and M. Sriwongjanya, "Characterization of an Inclusion Complex of Cholesterol and Hydroxypropyl-bcyclodextrin," Eur. J. Pharm. Biopharm. 46 (3), 355–360 (1998).

29. R.K. McMullan et al., "Topography of Cyclodextrin Inclusion Complexes: Part I. Classification of Crystallographic Data of α-Cyclodextrin Inclusion Complexes,"Carbohydr. Res. 31 (1), 37–46 (1973).

30. D. Horter and J. B. Dressman, "Influence of Physiochemical Properties on Dissolution of Drugs in the Gastrointestinal Tract," Advanced Drug Delivery Reviews 25 (1), 3–14 (1997).

31. J. P. Skelly et al., "Scale Up of Immediate Release Oral Solid Dosage Forms," J. Parenter. Sci. Technol. 47 (2), 52–56 (1993).

32. E. Galia, J. Horton, and J.B. Dressman, "Albendazole Generics—A Comparative In Vitro Study," Pharm. Res. 16 (12), 1871–1875 (1999).

33. C. Noory et al., "Rethinking the Use of Water as a Dissolution Medium," Dissolution Technol. 6 (4), 6–7 (1999).

34. V. P. Shah et al., "In Vitro Dissolution of Sparingly Water-Soluble Drug Dosage Forms," Int. J. of Pharm. 125 (11), 99–106 (1995).

35. M. Barzegar-Jalali et al., "Enhancement of Dissolution Rate and Anti-Inflammatory Effects of Piroxicam Using Solvent Deposition Technique," Drug Dev. Ind. Pharm. 28 (6), 681–686 (2002).

36. L. Tang, S.U. Khan, and N.A. Muhmmad, "Evaluation and Selection of Bio-relevant Dissolution Media for a Poorly Water Soluble New Chemical Entity," Pharm. Develop. Technol. 6 (44), 531–540 (2001).

37. H. Vromans, A.C. Eisson, and F.L. Coenraad, "Mechanism of Dissolution of Drug-Cyclodextrin Complexes," Drug Dev. Ind. Pharm. 15, 250–255 (1989).

38. M.E. Aulton, "Pharmaceutical Technology," in: Pharmaceutics: The Science of Dosage Form Design (London, UK: Churchill Livingstone, 1988), pp. 600–616.