Comparison of the Dissolution Rate of Ceftriaxone Sodium Preparations for Injection

Abstract

This study aimed to compare the dissolution rate of eight different formulations of ceftriaxone sodium preparation for injection, the original product and seven generic versions. The dissolution time was measured precisely as the point at which the freeze-dried ceftriaxone sodium preparation became a transparent solution on the addition of 0.9% sodium chloride solution. To investigate whether differences in the crystalline structure may explain the differences in dissolution rates, the eight products were subjected to X-ray diffraction (XRD) and differential scanning calorimetry (DSC). Powder surface characteristics were examined, including surface area, amount of water adsorbed, water interactions and morphology. The measurement of near-infrared spectroscopy of powder preparations was conducted, and we predicted dissolution time by partial least squares (PLS). The dissolution time of the eight products were different. There were no differences in XRD and DSC findings between the original and generic products, surface characteristics, i.e., surface area, morphology etc., were different between preparations. On near-infrared (NIR) spectroscopy, a good relationship was demonstrated between the actual and predicted dissolution time for each ceftriaxone preparation. The difference in dissolution time between the eight products was due to differences in powder surface characteristics, such as water interaction and crystal shape. Finally, it was shown that the dissolution rates of the products could be predicted by NIR analysis.

Dissolution properties are important in the formulation design of pharmaceutical preparations, and improvement in dissolution rate is a critical issue in drug development for non-parenteral preparations. There are a number of mathematical models for dissolution profile.¹⁾ Dissolution speed is represented by the Noyes–Whitney equation. It is clear from this equation that particle diameter affects dissolution. Therefore, in general, particle size is reduced and soluble materials are added to improve dissolution rate.²⁾ Crystal polymorphism, crystal habit and differences in crystallinity are also known to have an effect on dissolution rate. Sarkar et al. reported that amorphous forms of a drug which improve it’s dissolution rate and crystal habit changes can help to develop better solid dosage forms of drug.³⁾ Analytical methods used to evaluate dissolution properties during physicochemical evaluation are X-ray diffraction (XRD), differential scanning calorimetry (DSC), surface area measurement, particle size measurement and infrared (IR) spectrometry. In recent years, near-infrared (NIR) spectroscopy has attracted attention, primarily because it entails less sample preparation and is non-destructive. The prediction of product-specific properties such as particle size, water content, tablet hardness, etc. is performed using NIR spectroscopy,^4–7) as are studies of interactions with water.^8–10)

Dissolution rate is a particularly an important property in clinical practice. Previous studies have reported differences in dissolution rate between the original product and generic products, including injection preparations.^11–13) The dissolution rate of the preparation affects its efficacy, and there is a risk of injected products not being fully dissolved. Information on dissolution rate is important in product selection in a busy clinical setting.^13,14) Particularly in the preparation of products for intravenous injection, dissolution rate is critical, as these products must be administered as solutions.

Ceftriaxone has a prolonged biological half-life and is suitable for once-daily dosing. It has a wide spectrum of antimicrobial activity and superior tissue penetration, making it a useful and commonly prescribed antibiotic.¹⁵⁾ Ceftriaxone sodium is water-soluble,¹⁶⁾ but takes some time to dissolve fully, and it is expected that this property affects its use in a clinical setting. Dissolution time is thought to be similar to between preparations, because ceftriaxone preparations as no additives are included. In the present study, the dissolution time of eight freeze-dried ceftriaxone sodium preparations, the original product and seven generic versions, were precisely measured to the point that the product becomes a transparent solution after the addition of 0.9% sodium chloride solution. To clarify differences in the dissolution rate of these products, XRD, DSC, water adsorption, specific surface area, morphology and the state of the powder during the dissolution process were examined. An attempt was made to predict the rate of dissolution of the eight products using NIR spectroscopy.

Experimental

Materials

The original ceftriaxone sodium product (A) and seven generic versions (B to H) used in the study are summarized in Table 1 and shown in Fig. 1. Commercially available isotonic sodium chloride solution (Terumo Co., Ltd., Tokyo, Japan) was used as diluent.

Fig. 1. Ceftriaxone Sodium Preparations of Packing Photographs

Table 1. Preparations Used in the Present Study

Preparations		Company	Lot No.
A	ROCEPHIN® intravenous 1 g	Chugai Pharmaceutical Co., Ltd., Tokyo, Japan	10I020A	10J030A	12I030A
B	SEFIROM®	Nichi-iko Pharmaceutical Co., Ltd., Toyama, Japan	0F143	1F149
	Ceftriaxone sodium hydrate				KI1100
C	Ceftriaxone Na for intravenous injection 1 g [SANDOZ]	Sandoz Co., Ltd., Yamagata, Japan	0M330	1F331	2N331
D	LIASOPHIN for intravenous injection 1 g	Chemix Inc., Kanagawa, Japan	LV0N12	LV1705	3202
E	CEFXONE	Shiono Chemical Co., Ltd., Tokyo, Japan	YW01	ZM01	AP01
F	ROZECLART	TAIYO Pharmaceutical Industry Co., Ltd., Aichi, Japan	A62241	AB0011	B32345
G	Ceftriaxone Na for intravenous injection 1 g “Mylan”	Mylan Inc., Osaka, Japan	059AOQ	061AOQ	M052AUC
H	CERONEED®	Sawai Pharmaceutical Co., Ltd., Osaka, Japan	11101	11103
	CEFTRIAXONE Na				12802

SEFIROM® and CERONEED® became the designation of ceftriaxone sodium hydrate and CEFTRIAXONE Na, respectively.

The Measurement of Dissolution Time of Ceftriaxone Sodium Preparations in Isotonic Sodium Chloride

Isotonic sodium chloride solution (10 mL) was injected into the ceftriaxone sodium (1 g) preparation vial. Dissolution time (time until contents completely dissolved) was measured after repeatedly inverting the glass vial at intervals of one second. The time that contents were no visible to the human eye was criteria of ceftriaxone completely dissolved. Experiments was performed at a constant temperature of 25°C. Three batches of ceftriaxone sodium preparation were used in this study and three times measurement were performed in each batches. There was no difference in dissolution time among three different batches of a particular product.

The Characterization of Ceftriaxone Sodium Preparations

X-Ray Diffraction: X-Ray diffraction patterns of ceftriaxone sodium preparations were measured with RINT-Ultimate III (Rigaku Corporation, Tokyo, Japan) with CuKα radiation at 40 kV. The diffractometer was run at a scanning speed of 2.0°/min. The samples were scanned through the range 5–45° (2θ).

DSC: DSC was performed with a Thermoplus DSC 8230 instrument (Rigaku Corporation, Tokyo, Japan). Samples weighing 5 mg were heated in open aluminum pans at a scan rate of 5°C/min under a nitrogen gas flow of 50 mL/min.

Morphology: Powder morphology was observed using a stereomicroscope (SZX10, Olympus Co., Ltd., Tokyo, Japan) and scanning electron microscope (JSM-6510LV, JEOL Ltd., Tokyo, Japan).

Particle Size: The particle sizes of powder were measured with sieve analysis in accordance with the 16th edition of the Japanese Pharmacopoeia (2011) (range 45–500 µm).¹⁷⁾

The sieve used were 45, 150, 355 and 500 µm. A 1 g sample was shaken for 40 min.

Specific Surface Area: A 300-mg sample of ceftriaxone sodium powder was loaded into a Quantachrome powder cell and outgassed for 1 h at 130°C. Surface area was measured using automatic surface area analyzer (Monosorb Surface Area Analyzer, Quantachrome Instruments, Florida, U.S.A.). Nitrogen was used for adsorption.

Amount of Water Adsorbed: A 250-mg sample of ceftriaxone sodium powder was dried at 130°C to constant mass (difference ≤0.3 mg). A 200-mg sample of this dried powder was placed into a well-closed container with a supersaturated solution of potassium in water at a relative humidity of 65% and 45°C until the sample could adsorb no more water. The weight of adsorbed water on the surface of the powder was measured. Before the sample was measured, we confirmed in advance that the formulation property was not influenced by the drying process, as dissolution time was not different among undried sample and dried sample.

NIR Spectroscopy: NIR spectra were recorded using a NIR spectrometer (MPA Multi Purpose FT-NIR Analyzer, Bruker Optics Inc., Massachusetts, U.S.A.).

In the water interaction experiment, NIR spectra of sample preparations of ceftriaxone powder to which water had been added at room temperature were measured. The samples were scanned 60 times at 10 s intervals through the range 4000–9500 cm⁻¹.

In the evaluation of powder properties and prediction of dissolution time, NIR spectra were recorded through the sealed glass vials. The samples were scanned 10 times through the range 4000–12500 cm⁻¹. A chemometric analysis was performed using the partial least squares (PLS) method associated with the Pirouette software (Infometrix corporation, Woodinville, U.S.A.). The data were pretreated by multiplicative scatter correction (MSC). To build the calibration model, spectra were divided into calibration set and prediction set. Calibration model was created based on NIR spectra and dissolution time. The best conditions were determined the standard error of validation (SEV) by the leave-out-one method. Multivariate analysis using the PLS method was carried out on the NIR spectra, evaluating the correlation between actual and predicted values of powder properties, i.e., dissolution times.

Statistical Analysis

The dissolution times of the original and generic products were compared using Tukey test; statistical significance was accepted at the p<0.01 level.

Results

Dissolution Rate of Product in Isotonic Sodium Chloride Solution

Figure 2 shows the dissolution time of freeze-dried ceftriaxone sodium preparations. The dissolution time of the eight products were different; the original product (product A) demonstrated the most prompt dissolution. Dissolution time for products A and C (generic) were relatively rapid, but were much longer for the other products. In particular, the dissolution of products B, D, G and H required more than double time of products A and C. The preparations that dissolution rate was prompt, the undissolved powder was dispersed in solution, while the other preparation that dissolution rate was not prompt, the undissolved powder was aggregated at the bottom of the vials, when isotonic sodium chloride solution was injected into the ceftriaxone sodium.

Fig. 2. Dissolution Time of Ceftriaxone Sodium Preparations

Dissolution time of freeze-dried ceftriaxone sodium preparations in 0.9% sodium chloride solution. Each value is mean±S.D. of three. Tukey test; ** p<0.01, versus preparation A, ## p<0.01, versus preparation C.

The Characterization of Ceftriaxone Sodium Preparations

X-Ray Diffraction: The XRD patterns of the ceftriaxone sodium powders were similar, as shown in Fig. 3. All preparations, the crystalline peaks were confirmed at diffraction angles (2θ) 11–13°, 18–25° and 28°. Crystal polymorphism was not present and there were no differences in peak intensity between preparations. Crystallinity appeared to be equivalent in all preparations.

Fig. 3. XRD Diffractograms of Ceftriaxone Sodium Preparations

XRD patterns of ceftriaxone powder preparations at room temperature. Measurement conditions were: CuKα target; 40 kV voltage; 20-mA current; and 2.0°/min scanning speed.

DSC: A similar peak was observed on DSC for all eight preparations as shown in Fig. 4. The DSC curve of eight freeze-dried ceftriaxone sodium preparations showed two endothermic peaks at 75°C and 145°C, and a larger exothermic peak at 260°C. The endothermic peaks were probably due to dehydration process, and the exothermic peak was probably due to the melting/decomposition transition. There were no difference between eight preparations. These results support the results of XRD study.

Fig. 4. DSC Thermograms of Ceftriaxone Sodium Preparations

Samples weighing 5 mg were heated in open aluminum pans at a scan rate of 5°C/min under nitrogen gas flow (50 mL/min). Samples were heated from room temperature to 300°C.

Morphology: Stereomicroscope photographs of the ceftriaxone sodium powder are shown in Fig. 5a. The shape of the crystals can be seen to vary in the different preparations. In preparations B, D, G and H, the shape of the crystals was confirmed spherical form, particularly in product H, the particle with 100 µm or greater diameter was occupied with large part of powder particle size distribution. In preparations dissolved easily, there were little particle with 100 µm or greater diameter observed. These differences were confirmed on scanning electron microscopy (SEM) of ceftriaxone sodium powder, as shown in Fig. 5b. Crystals of the original ceftriaxone sodium product consisted of thin plates which were characteristic morphology and SEM image of the other preparations, their morphology is not observed. In preparation C, whose dissolution was also relatively quick, the crystal shape was fine and needle-like, while in preparations D and H, whose dissolution was slow, the crystals were thick and aggregated.

Fig. 5. Microscopic Photographs of Ceftriaxone Sodium Preparations

Pictures of dried ceftriaxone sodium preparations: (a) stereomicroscope photographs; (b) SEM photographs.

Particle Size: As shown in Fig. 6, particle size distributions for 8 preparations, were quite different among preparations. Preparation A which shows the particle size of ≤45 µm was mainly distributed, the most rapid dissolution rate. Preparation C was comprised of a relatively wide range of particle size distribution, dissolution rate was relatively rapid. Preparations B and G were made of particle with the range of 45 to 150 µm, with >90%, but dissolution rate was not rapid. Preparations D, E and F were largely composed of particle size of 45 to 355 µm. The particle size of preparation H was consisted of large particle (≥150 µm), especially the large part of the particle size of 355 to 500 µm. It was showed that the dissolution rate was not dependent on the particle size distribution in all preparations.

Fig. 6. Particle Size Distribution

Particle size distribution of ceftriaxone sodium powder was measured using sieving analysis. Each value is mean±S.D. of three.

Specific Surface Area: The specific surface area by Brunauer Emmett Teller (BET) is shown in Fig. 7. Differences in specific surface area were confirmed between products. In product C, the specific surface area was the largest in the all preparations. Products which dissolved more quickly tended to have a larger surface area, whereas poorly soluble preparations had smaller surface areas. However, no significant correlation was found between specific surface area and dissolution time, as relationship between the surface area and dissolution rate was not consistent with this trend in preparation A.

Fig. 7. Specific Surface Area of Ceftriaxone Sodium Preparation (m²/g)

Surface area was measured using an automatic surface area analyzer, and calculated by measuring nitrogen adsorption onto the surface. Each value is mean±S.D. of three.

Amount of Water Adsorbed: The amount of water adsorbed was different between preparations. The correlation between the amount of water adsorbed per 1 g of ceftriaxone sodium powder and dissolution time is shown Fig. 8. A significant correlation was found between the amount of water adsorbed and the dissolution time, such that the dissolution time was inversely proportional to amount of moisture adsorbed. It appeared that rapidly dissolved preparations more adsorbed water than slowly dissolved preparations.

Fig. 8. Correlation between Dissolution Time and Amount of Moisture Adsorbed

The line indicates the correlation between the amount of water adsorbed and the dissolution time, where the amount of moisture adsorbed is measured under supersaturation conditions after drying of the ceftriaxone powders.

NIR Spectroscopy: In the study of water interaction, the difference spectra were used for analysis. For all preparations, the NIR spectra changed shortly after the addition of water. The difference spectra were obtained by subtraction of the starting spectra from the spectra recorded at 10 min.

Figure 9 shows the difference spectra between 10 min after the drop of water was added from the spectra 1 min after the drop of water. In products A and D, the spectra had peak changes at 5260 cm⁻¹ and 7000 cm⁻¹. The band around 5260 cm⁻¹ is associated with the combination band of the OH stretching and bending (OH st+OH def) which better reflects the molecular state of water. The absorption maxima depend strongly on the state of the hydrogen bonding. The band around 7000 cm⁻¹ corresponds to the first overtone of the OH stretching (OH 1st overtone) which is related to the intermolecular hydrogen bond. In other words, both bands can be explained by the interactions of the water molecule. In particular, preparation A showed strong band. These results suggest that the water interaction was strongest in preparations A and D. In contrast, spectrum peak changes could not be clearly identified in the other products.

Fig. 9. NIR Spectra for Eight Ceftriaxone Sodium Products

The difference spectra of ceftriaxone sodium on the addition of water. The difference spectra were obtained by subtraction of the starting spectra from the spectra recorded at 10 min.

Figure 10 shows NIR spectra of ceftriaxone powder samples taken through the sealed glass vials. The NIR spectra of the eight products were different, reflecting differences in the particle size of the various powders. These results support the results of morphology study.

Fig. 10. NIR Spectra of Ceftriaxone Preparations

NIR spectra of ceftriaxone powder samples taken through the bottom of glass vials. The baselines of the NIR spectra were different between preparations, reflecting differences in particle size.

Prediction of Dissolution Rate of Ceftriaxone Sodium Preparation Using NIR Chemoinformetrics

The relationship between actual and predicted dissolution rate of ceftriaxone sodium by PLS method is shown in Fig. 11. To find a potential fit between the spectral information and the dissolution time, the PLS regression algorithm was calculated. MSC was performed before prediction using the PLS method due to the removal of particle size. The predicted and actual NIR values showed good correlation using the PLS method. The linear plot shows a slope of 0.9688, an intercept of 7.4411, and a correlation coefficient constant, γ, of 0.9933. The relationship between SEV and the number of factors for evaluation of dissolution rate suggested the follow: the SEV values was almost constant after Factor 6, the calibration model was calculated based on a 6-latent variable model.

Fig. 11. Prediction of Dissolution Time of Ceftriaxone Sodium Preparations with NIR Spectroscopy

Correlation plot for the measured and predicted dissolution times with MSC. Correlation analysis of PLS modeling using samples in the prediction set.

Figure 12 shows the loading vectors. For the loading vector for Factor 1, the positive peak at around 4640 cm⁻¹ was described the surface water. The loading vector of Factor 2, there were positive peaks at 4500 cm⁻¹ and 7000 cm⁻¹ which could be assigned the hydrate water. The loading vector of Factor 3, there were positive peaks around 8000 cm⁻¹ which could be assigned the water with two OHs H-bonded.

Fig. 12. Loading Vectors of PLS Analysis for Prediction of Dissolution Time on NIR Spectra

(A) Factor 1 loading; (B) Factor 2 loading; (C) Factor 3 loading.

Regression vector of PLS analysis for prediction of dissolution time on NIR spectra was shown Fig. 13. The regression vector indicate important factors for the dissolution time in the NIR spectra. The positive peaks at around 4600 cm⁻¹ and 5000 cm⁻¹. Those spectra might be due to hydroxyl group. The regression vector meaning that the peaks attributable to water interaction was associated with dissolution time.

Fig. 13. The Regression Vector of the Calibration Model for Dissolution Time of Ceftriaxone Sodium Preparations

Discussion

The dissolution of ceftriaxone sodium preparations in 0.9% sodium chloride solution varied considerably in the eight different preparations studied. The products could basically be divided into two types: the dissolution times of products A (the original) and C (generic) were relatively short, while the other six generic drugs needed more than twice the time for complete dissolution. These differences in dissolution time undoubtedly affect the preparation time for the different products.

The differences in dissolution time cannot be due to differences in the additives in these preparations as no additives are included. Accordingly, it is possible that the differences in dissolution are due to differences in crystal properties, itself the result of differences in the manufacturing process of the ceftriaxone sodium powders. In general then, differences in crystal polymorphism, crystal habit and particle diameter can be attributed to differences in the crystallization process. However, crystal polymorphism was not confirmed on XRD or DSC, and no significant differences in crystallinity were demonstrated between the eight ceftriaxone sodium preparations. The surface properties of the powder must also be considered. The particle diameters of the different products were confirmed by particle size study and microscopically, with products whose particles were spherical tending to have poorer dissolution. The particle size distribution of preparation A which shows the fastest dissolution rate was consists of small particles. Preparations B and G that dissolution was not prompt was comprised of a small particle. The particle size of preparation C that dissolution rate was relatively short was not small. In preparation B, D, G and H, spherical form were confirmed. These results suggested that the dissolution rate of slowly soluble drugs are related to spherical particle shape more than particle size distribution. This may be due to difficulties in penetration into powder inside of the solvent.¹⁸⁾ As previously reported, dissolution rate can be improved by crushing powders,^19,20) as the particle diameter decreases and the surface area increases. The morphological analysis of surface of powder using SEM and the specific surface areas supports this. For product C, which had the largest specific surface area, it was confirmed that multiple pores were formed due to needle-shaped crystal clusters in the particle. Product A has a thin plate-layer crystal structure, with low porosity and a small specific surface area. The other six generic preparations also had small specific surface areas, in this case due to prism formation in the crystal structure, resulting in no pores on the surface. Also, during the dissolution process in 0.9% sodium chloride solution, rapidly dissolved ceftriaxone sodium preparations are homogeneously dispersed in solution, while slowly dissolved preparations clump together at the bottom of the vial, and the effective surface area where the powder touches the solution is therefore reduced. It is probable that this mechanism causes the long dissolution times. These results indicated that dissolution time was negative linear relation to specific surface area with the exception of preparation A. The specific surface area differences have been shown to major factor influencing dissolution rate of ceftriaxone sodium preparation. However, preparation A has different dissolution mechanism which was disintegrated by contact with solutions, so dissolution rate of preparation A was not dependent on surface area.

The water adsorption was different in the eight ceftriaxone sodium preparations. In slowly dissolving products, water adsorption was significantly lower than in rapidly dissolving products. This result suggested about the influence of wettability of powder on dissolution rate. Unlike in the case of specific surface area, correlation was obtained between the amount of water adsorbed and dissolution time. Nitrogen adsorption range (result of specific surface area) was the whole surface. Whereas, water adsorption result was obtained the surface hydrophilicity information, which is water adsorption range was only hydrophilic surface.^21–23) Accordingly, it was suggested that differences in affinity for water of the powder surface between products.

NIR spectroscopy confirmed that water interactions were strongest in products A and D. This may be related to the fact that dissolution is quickest in product A. The difference in dissolution rate between products A and D suggests that there may be another interaction. The specific surface area of preparation A was small, but dissolution rate was rapidly because it was thought powder interact readily with water. In contrast, product D took a little longer to solubilize, because it was probable that only the powder’s surface get wet and solutions penetration inside of powder was difficult. However, this could not be confirmed by NIR spectroscopy. Further studies will be needed to determine the reason for this difference in dissolution rate.

It was conformed that the baselines of the eight ceftriaxone sodium powder preparations in the NIR spectra were different. This is due to scattering and absorbance, and depends on powder properties such as particle size.²⁴⁾ The Kubelka–Munk equation is used for determination of diffuse reflectance spectra and prediction of particle size.^25–28) This indicated that the particle size of the eight preparations was different. This result agrees with the findings on powder morphology and particle size study. Our experimental results indicated that the prediction of dissolution time using PLS models correlated significantly with actual dissolution time. It has previously been reported that NIR spectroscopy can predict dissolution rate.^29–31) Therefore, it is suggested that the measurement of NIR spectra provides a good prediction of the dissolution rate properties of ceftriaxone sodium powders. The loading vector shows which variable is combined by the formation of the coordinate axis of the factor, and which phenomenon is attributed to calibration model. The results of the present study show that water interaction was concerned with prediction of dissolution time. The regression vector explained which explanatory variable is important when modeling of objective variable. From the results of regression vector, the peaks were related to hydrophilic group. The present result suggested that hydrophilic might be related to the dissolution time.

In the present study, it is suggested that the dissolution rate of ceftriaxone sodium preparations for injection varies according to differences in particle size and wettability. Analysis of crystal properties (i.e., XRD and DSC) did not reveal the cause of the differences in dissolution rate, but analysis of powder surface properties (i.e., SEM, amount of water adsorption and NIR spectroscopy) did. In the future, it may be possible to develop an injection preparation with better dissolution rate, superior to both the current original preparation and generic versions, by improving the relevant properties of the powder.

Conclusion

The result of this dissolution study showed that the dissolution rate of ceftriaxone sodium preparations for injection varied considerably among eight different products tested. These differences in dissolution time may affect work efficiency in clinical practice. The difference in dissolution times between preparations was due to differences in powder surface characteristics, such as water interaction and crystal shape. However it does not seem to be possible to predict the dissolution rate in all preparations on the basis of any single factor. Factors affect dissolution rate was varied according to preparations. NIR chemometric method can predict the water interactions of ceftriaxone powder, and is therefore an indicator of dissolution rate. The chemometric method is simple and nondestructive, the prediction of dissolution rate using NIR spectra is a useful index in the evaluation of dissolution rate.

References

• 1)  Costa P., Sousa Lobo J. M., Eur. J. Pharm. Sci., 13, 123–133 (2001).
• 2)  El-Badry M., Fetih G., Fathy M., Saudi Pharm. J., 17, 217–225 (2009).
• 3)  Sarkar M., Perumal O. P., Panchagnula R., Indian J. Pharm. Sci., 70, 619–630 (2008).
• 4)  Otsuka M., Yamane I., J. Pharm. Sci., 95, 1425–1433 (2006).
• 5)  Otsuka M., Yamane I., J. Pharm. Sci., 98, 4296–4305 (2009).
• 6)  Ebube N. K., Thosar S. S., Roberts R. A., Kemper M. S., Rubinovitz R., Martin D. L., Reier G. E., Wheatley T. A., Shukla A. J., Pharm. Dev. Technol., 4, 19–26 (1999).
• 7)  Maltesen M. J., van de Weert M., Grohganz H., AAPS PharmSciTech, 13, 747–755 (2012).
• 8)  Zhou G. X., Ge Z., Dorwart J., Izzo B., Kukura J., Bicker G., Wyvratt J., J. Pharm. Sci., 92, 1058–1065 (2003).
• 9)  Cao W., Mao C., Chen W., Lin H., Krishnan S., Cauchon N., J. Pharm. Sci., 95, 2077–2086 (2006).
• 10)  Giangiacomo R., Food Chem., 96, 371–379 (2006).
• 11)  Liebenberg W., de Villiers M. M., Wurster D. E., Swanepoel E., Dekker T. G., Lötter A. P., Drug Dev. Ind. Pharm., 25, 1027–1033 (1999).
• 12)  Henwood S. Q., de Villiers M. M., Liebenberg W., Lötter A. P., Drug Dev. Ind. Pharm., 26, 403–408 (2000).
• 13)  Yamasaki H., Miyano N., Oie S., Kamiya A., Jpn. J. Pharm. Health Care Sci., 34, 860–864 (2008).
• 14)  Maeda S., Yamagata M., Fujii H., Inoue T., Masuda K., Oohashi M., Naotsuka M., Kanamori T., Yakuzaigaku, 64, 186–191 (2004).
• 15)  Chugai Pharmaceuticals. “Drug interview form ‘Rocephine® for injection 1 g,’” 2009
• 16)  “Japanese Pharmacopoeia,” 16th ed., Hirokawa Book Co., Tokyo, 2011, p. 821.
• 17)  “Japanese Pharmacopoeia,” 16th ed. Instruction manual, Hirokawa Book Co., Tokyo, 2011, pp. B414–B429
• 18)  Sabin G. P., Breitkreitz M. C., de Souza A. M., da Fonseca P., Calefe L., Moffa M., Poppi R. J., Anal. Chim. Acta, 706, 113–119 (2011).
• 19)  Matsumoto K., Kimura T., Nakai Y., Yonemochi E., Oguchi T., Yamamoto Y., Yakuzaigaku, 57, 1–7 (1997).
• 20)  Lin C. W., Cham T. M., Int. J. Pharm., 127, 261–272 (1996).
• 21)  Esumi K., Powder Soc. Eng., 21, 78–88 (1989).
• 22)  Naono H., J. Soc. Powder Technol. Japan, 28, 34–39 (1991).
• 23)  Karanfil T., Dastgheib S. A., Environ. Sci. Technol., 38, 5834–5841 (2004).
• 24)  Nieuwmeyer F. J., Damen M., Gerich A., Rusmini F., van der Voort Maarschalk K., Vromans H., Pharm. Res., 24, 1854–1861 (2007).
• 25)  Cairós C., Coello J., Maspoch S., Appl. Spectrosc., 62, 1363–1369 (2008).
• 26)  O’Neil A. J., Jee R. D., Moffat A. C., Analyst (Lond.), 123, 2297–2302 (1998).
• 27)  Szalay A., Antal I., Zsigmond Z., Marton S., Erös I., Regdon G. Jr., Pintye-Hódi K., Part Syst. Charact., 22, 219–222 (2005).
• 28)  Otsuka M., Powder Technol., 141, 244–250 (2004).
• 29)  Tabasi S. H., Moolchandani V., Fahmy R. M., Hoag S. W., Int. J. Pharm., 382, 1–6 (2009).
• 30)  Gendre C., Boiret M., Genty M., Chaminade P., Pean J. M., Int. J. Pharm., 421, 237–243 (2011).
• 31)  de Oliveira Neves A. C., Soares G. M., de Morais S. C., da Costa F. S., Porto D. L., de Lima K. M., J. Pharm. Biomed. Anal., 57, 115–119 (2012).