- Email: [email protected]
Bioavailability Studies of Theophylline Ethylcellulose Microcapsules Prepared by Using Ethylene-Vinyl Acetate Copolymer as a Coacervation-Inducing Agent
BioavailabiIity Studies of Theophy IIine Ethylcellulose Microcapsules Prepared by Using Ethylene-Vinyl Acetate Copolymer as a Coacetvatlon-Inducing Agent %AN-YANG
LIN' AND JUEI-CHYI YANG
Received June 11, 1986, from the Biophannaceutics La&ratory, De rtment of Medical Research, Veterans General Hospital, Shih-Pai, Taipei, Accepted for publication January 6, &7. Tahvan, Republlc of China.
Abstract 0 The Moavailability of theophylline microcapsules prepared by using ethylene-vinyl acetate (EVA) copolymer a8 a coacervationlndudng agent was studied in rats. The dissolution rate of the microcap suies was determined by the rotating-basket and rotating-bottle methods. The higher the concentration of EVA copolymer used, the more sustained was the release of theophylline from the microcapsules. The mean maximum serum levels (&) and time to maximum serum levels (t-) were not significantly different for theophylline mlcrocapsuies prepared by a lower concentration of EVA copolymer (0 and 0.83%, respectively), compared with those for theophylline powder; whereas a significant difference was found when the higher concentration of EVA copolymer was used (>1.7%). With regards to the area-under-the-curve (AUC) value, there was no significant difference between the theophylline powder and theophylline microcapsules. The elimination kinetics and the corresponding half-life (ten) were significantly different when the concentration of EVA copolymer was >3.3%.From the above results, it is evident that theophylline microcapsules prepared by using 3.3 and 5.0% EVA copolymer as the coacervation-indudngagent may act as sustained-release dosage forms. The correlation between the dissolution rate in vitro and the bioavaiiabilrty In rats for theophylline microcapsules was investigated. The mean C,, and -t axrelated well with the time taken to release 75% of the drug in vitro (b6%);however, the mean AUC showed no valid correlation with b6%.This implies that the dissdution rate correlated better with the rate of absorptlon (C-, 4-) than with the extent of absorption (AUC). Moreover, a llnear relationship was aim found between the time taken to release 50% of the microcap sules in vitro (&) and the time taken to absorb 50% of drug In vlvo indicating the presence of an in vitro:in vlvo correlation. The Weibull function (&) was also used to evaluate the In vttro:in vlvo correlation. It obvlously indicates that & significantly conelated wlth t, (In vIyo) (rotating-basket method: r = 0.9922; rotating-bottle method: r = 0.9929). Thus, the results suggest that the theophylllne microcapsules prepared by udng EVA copolymer a8 a coacervatlonindudng agent not only possess a sustained-release behavior, but also give a good conelation between in vttro dissolutlon rates and in vlvo absorption.
(e),
Theophylline has been popularly used as a bronchodilator for the treatment of chronic asthma and chronic obstructive lung disease.' A narrow therapeutic range of serum theophylline concentration (10-20 ccg/mL) often has resulted in many adverse side effects such as gastrointestinal (GI) irritation (>20 pg/mL), central nervous system (CNS)effecta (>20 pg/mL), focal and generalized seizure (>40 ccg/mL), and cardiovascular toxic effects (>40 pghdd.1 Sustained-release theophylline preparations offer the advantage of less frequent dosing with decreased fluctuation in the serum theophylline level during the dosing interval; thus, these preparations have been widely used in clinical therapy.We have previously shown that theophylline release behavior from ethylcellulose microcapsules, prepared by using ethylene-vinyl acetate (EVA)copolymer as a coacervationinducing agent, was dependent on the concentrations of the coacervation-inducing agent."7 The release kinetics of theophylline from ethylcellulose microcapsules agree with both 00~-3549/87/~-0219$0 .&YO 1 0 1987. American Pharmaceutical Associetion
first-order and square-root-of-time equations. But, after the application of the differential rate treatment, it was found that kinetic data actually conformed with the first-order equati0n.B-g In the present study, the bioavailability parameters of theophylline microcapsules prepared by using various concentrations of a coacervation-inducing agent were examined. We also attempted to clarify the relationship between the in vitro dissolution rates determined by the rotating-basket and rotating-bottle dissolution methods, and the in vivo absorption kinetics for theophylline microcapsules.
Experimental Section MaterialeTheophylline (Wako Pure Chemical Industry, Tokyo), ethylene-vinyl acetate copolymer (EVA; vinyl acetate content 28%; Toyo Soda Manufacturing Co., Tokyo), and ethylcellulose (100cps; h w Chemical Co., MI) were used. Other materials were of analytical reagent grade. Preparation of Theophylline Microcapaules-Theophylline microcapsules were prepared by the phase-separation method, using EVA copolymer aa a coacervation-inducing agent, aa described in previous studies."" Three hundred millilitera of cyclohexane solution, containing from 0 to 5% of EVA copolymer, were respectively placed in a 1000-mL, three-neck, round-bottomed flask equipped with a stirrer, a thermometer, and a reflex condemr. Theophylline (3 g) and ethylcellulose (3 g) were added to the stirred cyclohexaneEVA copolymer solution (150 rpm) at mom temperature. The system waa then heated to 82 "C to form a homogeneom suspenaion. With continuous stirring, the system was allowed to cool to 24 "C, and waa then stirred for another 10 min. During the cooling proceee, a pheee separation occurred. The microcapsules were separated from the solution by decantation, rinsed with n-heme, and dried at 40 "C in a vacuum drier for 24 h. Theophylline microcapsulea (40-60 meah; 250-450 pn) were used for the experiment. Dissolution Studies-The dieeolution rate of theophylline from microcapsules in a pH 1.2 eolution was determined by using the previously reported rotating-basket (100 rpm) and rotating-bottle methode (20 rpm) at 37 'C.7.11 Samples of the microcapsules (equivalent to 100 mg of theophylline for use in the rotating-baaket method and 30 mg of theophylline for use in the rotating-bottle method) were taken for the in vitro dieeolution studiee. The volume of diesolution medium used waa 500 mL for the rotating-basket method and 50 mL for the rotating-bottle method. These two dissolution methoda were carried out under sink conditiona (solubility of theophylline in pH 1.2 solution was -11.68 mg/mL). Aliquota of the diesolution medium were aaaayed for theophylline by ultraviolet spectrophotometry at 270 nm (UVIKON 810, Kontron, Switzerland). Bioavailability Studie-tudies were conducted on male Sprague-Dawley rate weighing 300350 g. The animals were fasted from the night before and throughout the experiment. Each rat waa given a dose of -40 m g k g of theophylline powder and an equivalent doee in microcapsule form (250450 pn). The test sample waa administrated orally with 10 mL of water by a feeding tube. Serial blood samplee (0.7 mL) were taken a t 0.25,0.5,1.0,1.5,2.0,2.5,3.0, 4.0,5.0,6.0,8.0,10.0,12.0,24.0,26.0, and 27.0 h after zero time. The samples were centrifuged, and the serum waa removed and stored at - 20 "C until the time of analysis. The serum theophylline concentraJournal of Phannaceutkal Sdences / 21Q Vol. 76, No. 3, March 1987
Results and Discussion
tion was assayed utilizing a homogenous enzyme immunoaasay technique (EMIT system, Syva). Bioavailability and Pharmacokinetic AnalysigTheophylline absorption from the various dosage forms was characterized by a single-compartment, first-order pharmacokinetic model.13 Although a multicompartment model may be more accurate, a single-compartment model is sufficientlyaccurate in the case of theophylline and is more easily understood for clinical purposes. Thus, the serum concentration of theophylline, C,at any time, t , after a single oral dose, D, is deacribed by14
In Vitro Dissolution of Theophylline From Ethylcellulose Microcapsules-The dissolution rate profiles of theophylline microcapsules determined by two different dissolution methods are shown in Figure 1. Each point represents the average of three experiments. As shown in Figure 1, the rate of dissolution of theophylline from ethylcellulose microcapsules was related to the concentration of the coacervationinducing agent. The higher the concentration of EVA copolymer used, the slower the release rate of the theophylline microcapsules. This might be attributed to the lower porosity and thicker wall of the microcapsules that were prepared by using the higher concentration of EVA ~opolymer.~ The rotating-bottle method has been recommended for testing sustained-release oral dosage forms because it is reasonably simple in practical use and satisfactorily parallels physiological conditions.18 Figure 1also indicates that the release rate of theophylline as determined by the rotating-bottle dissolution method was slightly faster than that determined by the rotating-basket dissolution method, although the rotating speed of the rotating-basket method was higher than that of the rotating-bottle method (100 rpm versus 20 rpm). The different dissolution rates might therefore be due to the different motion patterns in the dissolution medium. The microcapsules studied with the rotating-bottle dissolution method were completely immersed and moved in the dissolution bottle by means of air moving rather than by laminar motion, as was the case of the aggregated microcapsules in the rotating basket, leading to a fast release. Serum Concentration and Pharmacokinetic Parameters of Theophylline after Oral Administration of Theophylline Microcapsules-The rats used in this study tolerated the procedures well and remained clearly conscious throughout the experiment, although several blood samples were taken. The serum concentration versus time curves of theophylline in rats corresponded to a one-compartment open model, as shown in Figure 2. The absorption of theophylline from theophylline powder was very rapid, whereas the microcapsules prepared with the higher concentrations of EVA copolymer produced a more gradual rise and a more sustained plateau of serum levels of theophylline. Figure 2 indicates that the higher the concentration of EVA copolymer used, the more sustained the release behavior of the theophylline microcapsules. Moreover, the theophylline microcapsules prepared by using a higher concentration of EVA copolymer gave a transient steady-state plateau level of theophylline in the serum after administration without producing a sharp peak of serum concentration.
where F is the fraction of the drug absorbed, Vd is the apparent volume of distribution, and k, and kel are the pseudo first-order rate constants for absorption and elimination, respectively. The log plasma theophylline concentration-time curve was plotted. From the terminal linear portion of the curve, the apparent elimination rate (k,,) was obtained from the slope of linear portion by least squares regression analysis. The half-lives of theophylline microcapsules were calculated by dividing 0.693 by the elimination rate constant. The C , , , and tmnxvalues were recorded for each formulation, where , ,C was chosen as the highest observed serum theophylline concentration after oral administration of the dose and t,,, was the time of the maximum measured serum concentration. Another parameter, AUC, from zero to the final sampling time (AUC,,), was calculated using the trapezoidal rule. Moreover, AUC+, is the serum AUC from time zero to infinity calculated fromI6
where C,is the final serum theophylline concentration obtained, and kel is the elimination rate constant obtained from eq 1. The rate and extent of the theophylline absorption from test samples were examined from the Wagner-Nelson equationt6 Fab =
Ct + kel
[AUCo-tl [AUCo-scl
(3)
where Febis the fraction of dose absorbed at time t . The relative bioavailability value (F,,) for the single-dose study was determined by17 FreI =
AUCo,, (theophylline microcapsule) AUCo,, (theophylline powder)
X
Dose (theophylline powder) Dose (theophylline microcapsule)
(4)
100
Flgure 1-Effect of EVA copolymer concentration on the release of theophylline from microcapsules in pH 1.2 solution. Key: solid points: rotating-basket dissolution method; open points: rotating-bottle dissolution method; concentration of EVA wpolymer: (0, +) 0%; ( 0 ,V) 0.83%; (0, ). 1.7%;(0, 0 ) 3.3%; (A, A) 5.0%.
80
’0 0)
!
4
.A’
60
t
U
$
40
U 0)
a
20
t
0 0
2
4
6
8
0
Ti.me,
220 /Journal of Pharmaceutical Sciences Vol. 76, No. 3, March 1987
2
4
6
8
Figure 2-Serum theophylline concentration levels after oral administration of theophylline powder and theophylline microcapsules. Key: concentration of EVA copolymer: (A) theophylline powder; (6) 0%; (C)0.83%; (D) 7.7%; ’ (E) 3.3%; (F) 5.0%. The bars indicate the standard error of the mean (SEM).
:
o
0
2 6 $ 60 g 2 50
B
A
5 0 h
24
16
:IA
C
D
F
Time, h
The bioavailability and pharmacokinetic parameters were estimated from the serum concentration of theophylline after4 administration of either theophylline powder and theophylline microcapsules (Table I). The mean values of the maximum serum level (Cmu) and the time to maximum serum level (tmm) were not significantly different (p > 0.05) for theophylline microcapsules prepared by using the lower concentrations of EVA copolymer (0% and 0.83%)compared with the theophylline powder group. The higher C,,, value is consistent with the shorter t,,, value. However, a significant difference was found in theophylline microcapsule groups prepared by using higher concentrations of EVA copolymer (>1.7%). Furthermore, the serum level of theophylline slowly decreased for a long time for theophylline microcapsules prepared by using 3.3 and 5.0% of EVA copolymer. This indicates that theophylline microcapsules prepared by using higher concentrations of EVA copolymer exhibited a better sustained-release behavior than theophylline microcapsules prepared by using a lower concentration of EVA copolymer. The AUC values for both theophylline powder and theophylline microcapsules were not significantly different (p > 0.05). The AUC values of theophylline microcapsules prepared by using 5% of EVA copolymer were larger than any one of the AUC values. This may be due to
the higher value of final serum theophylline concentration and the lower value of the elimination rate (eq 2). Theophylline elimination kinetics and the corresponding t1/, are also tabulated in Table I. The elimination data and the corresponding t E values are in agreement with the results of C,,, and t,,,. There was no significant difference in the elimination rate and half-life values between theophylline powder and theophylline microcapsules prepared by using EVA copolymer as a coacewation-inducing agent from 0 to 1.7%. By contrast, there was a significant difference between theophylline microcapsules prepared by using 3.3-5.0% of EVA copolymer and theophylline powder. Assuming that the theophylline powder was completely absorbed’e and its relative bioavailability (Frel).probably approximated 1008, then the relative bioavailability of the five types of theophylline microcapsules ranged from 77.6 to 105.2%. The higher the concentration of EVA copolymer used, the larger the F,,l value of theophylline microcapsules. The results of the bioavailability data and pharmacokinetic parameters indicate that theophylline microcapsules prepared by using 3.3 and 5.0%of EVA copolymer as a coacervation-inducing agent possess sustained-release action. In Vitro:In Vivo Relationship-The relationship of tT5%, the time taken to release 75% of drug in vitro to , , C tmax,
Table CPharmacoklnetlc Parameters of Theophylllne Mlcrocapruler after Oral Admlnlstratlon In Rats
~
0
0.83
6 6
42.23 2 4.65’
1.93 2 0.09
333.72 2 72.89
0.119 2 0.01
5.89 t 0.68
NS
NS
NS
NS
NS
41.53 t 4.65
2.08 2 0.29
0.128 2 0.03
2.75 2 0.26 p < 0.05 5.34 2 0.57
393.18 2 56.73 NS 407.82 2 21.11 NS 388.14 2 29.76
0.120 2 0.02
6.04 2 1.05 NS 6.12 2 0.83
NS
NS
NS
6.43 t 0.71 p < 0.05 1.62 2 0.21
450.32 2 80.72 NS 429.76 2 30.63
0.066 2 0.04 p < 0.05 0.052 5 0.01 p < 0.05 0.134 2 0.02
9.50 2 0.69
p < 0.05
NS 1.7
6
37.53 2 1.40
p < 0.05‘ 3.3
8
20.18 2 5.25
p < 0.05 5.0
TheoDhvllined
8 6
17.95 2 0.78 p < 0.05 51.78 2 1.69
NS
NS
77.6 91.5 95.1 90.1
p < 0.05 15.64 2 3.84
105.2
p < 0.05 6.01 2 0.79
100
a Mean 2 SEM. No significant difference compared with theophylline group. =Significantdifference compared with theophylline group. “Actual dose per rat was 40 mg/kg.
Journal of Pharmaceutical Sciences / 221 Vol. 76, No. 3, March 7987
agent. Moreover, a parallel relationship was shown for the rotating-basket dissolution method and the rotating-bottle dissolution method, as shown in Figure 5.This demonstrates that the rotating-basket dissolution method was in close agreement with the rotating-bottle dissolution method. In conclusion, this study confirms that our theophylline microcapsules, prepared by using EVA copolymer as a coacervation-inducing agent, exhibited sustained-release be-
and AUCb, in vivo, was investigated (Figure 3). To compare each correlation, we used the coefficient of determination r2. It is obvious that there were statistically significant correlations between t754tand, ,C (r = 0.9921),and t75% and t,,, (r = 0.9837);whereas, t759band AUC-, (r = 0.3359) showed no correlation between in vitro and in vivo results. This shows that the in vitro dissolution rate (both dissolution methods) correlated better with , C and t,,, which correspond to the rate of absorption, than with AUCh,, which reflects the extent of absorption. The poorer correlation between dissolution rate and AUCk, may be due to the fact that the extent of absorption was affected to a greater extent by various physiological factors in the GI tract, compared with the rate of absorption. This result has also been reported by several investigators.20 In another attempt to evaluate the absorption kinetics of microcapsules, the serum profiles of Figure 2 were transformed using the Wagner-Nelson equation (eq 3). Plots for tSo8,the time taken to release 50% of drug from microcapsules in vitro, and F,6tQ,the time taken to absorb 50%of drug in vivo, are presented in Figure 4. The correlations between the bioavailability data and the dissolution variables were statistically significant, and the correlation coefficient (r) in no case was <0.9. Drug release from dosage forms has been suggested by using the Weibull distribution function.21-2s The Weibull analysis of dissolution data can play an important function in product development and may lead to more rational quality control. We used the Weibull distribution function to describe the dissolution curves of theophylline microcapsules:
log [-ln (W/W,,)] = b log ( t - Ti)- log a
- 00
3
1
5
7
t50%. h
(5)
of in vitro t-
Flgure &Correlation
and in vivo =%. Key: see Figure
1.
where W is the quantity of drug remaining in the microcapsules, W,is the initial quantity of drug in the microcapsules, a and b are the scale parameter and shape factor, respectively, Ti is the location parameter, and td represents the time interval necessary to release 63.2% of the drug. The Weibull function was also used for interpreting the profiles of the drug absorbed versus time.26 By this approximation, we focused on t d (in v,tro) and td (in viva) of the Weibull functions. The evaluation of Weibull distribution parameters was performed by means of a linear regression analysis. From the shape factor, b, of the Weibull function, all the microcapsules appear to release their content in a first-order process, since the b value ranged from 0 to 1. Figure 5 shows the linear relationship between td (in vitm) and td (in ,,iv0) of theophylline microcapsules (rotating-basket method: r = 0.9922;rotatingbottle method: r = 0.9929).This also indicates that a good correlation between in vitro dissolution and in vivo absorption was found in our developed theophylline microcapsules that were prepared by using EVA as a coacervation-inducing
2
0
4
8
6
10
birvibo,h
regression between b fin vit,,,) theophylline microcapsules. Key: see Figure 1.
Figure +Linear
and &
(in V
10-
5 4
8 -
Flgure 3-Correlation
-\&,
of
t,5% with C-, t,,, and AUC,,. Key: s8e Figure 7 ; the bars indicate SEM.
52 '4
u~
0
2
4
6
t75r.
222 1Journal of Pharmaceutical Sciences Vol. 76, No. 3, March 1987
8 h
1
0
0
2
4
6
t75X.
8 h
1
0
0
2
4
6 t75%. h
8
1
0
~
of ~
J
havior both in vitro and in vivo. The in vitro:in vivo mrrelation has been established. Further studies need to be performed to examine the bioavailability of theophylline microcapsules in humans.
References and Notes 1. Hendeles, L.; Weinber er, M. Pharmacothera 1983,3,2-43. 2. Baptista, R.; Driscoll, F. A m . Pharm. 19861yNS24,285-293. 3. Weinber er, M. Pharmacothemm 1984,4,181-198. 4. Cummisfe J M . Po a, V. J. Asthma 1984,21,243-257. 5. M e r i c h , A.; h e & - , S. J.; Green, E. R. J. Allergy Clin. Zmmunol. 1981,61,465-467. 6. Lin, S. Y.; Yang, J. C.; Jiang, S. S. J. Taiwan Phurm. Sci. 1985, 37, 1-10. 7. Lin, S.Y.; Yang, J. C. J. Microencapsulation 1985,2,315-325. 8. Lin, S.Y.; Yan ,J. C. J. Taiwan Pharm. Sci. 1986,38,160-165. 9. Benita, S. Ap Biochem. Biotech. 1984,10,255-258. 10. Lin, S.Y. J. kicmencapsulution 1985,2,91-101. 11. Lin, S . Y.; Yan ,J. C. J. Controlled Release 1986,3,221-228. 12. Lin, S.Y.; Ho, T.; Chiou, H. L. Bwmater. Med. Dev. Art. Org. 1986,13,187-201. 13. Shen, D.D.;Fixley, M.; Azarnoff, D.L. J.Pharm. Sci. 1978,67, 916-919. 14. Lesko, L. J.; Canada, A. T.; Eastwood, G.; Walker, D.;Brous*an, D. R. J. Pharm. Sci. 1979,68,1392-1394.
6.
k.
&. f.
15. Gibaldi, M.; Perrier, D. Pharmacokinetics, Vol. I; Marcel Dekker: New York, 1975;pp 293-296. 16. Wagner, J. G. Fundamentals of Clinical Phurmacokinetics, 1st ed.; Dru Intelli ence: Hamilton, IL, 1975;pp 174-176. 17. Green, D.; laccar, C. L.; Helsel, C.L.; Niehls, M. E.; McGeady, S. J.;Mansmaun, H. C. J.Asthma 1985,22,159-163. 18. Souder, J. C.; Ellenbogen, W. E. Drug Standards 1958,26,77-
1.
79
19. Hendeles, L.; Weinberger, M.; Bighley, L. Am. J. Hosp. Phurm. 1977,34,525-527. 20. Aoyagi, N.; Ogata, H.; Kaniwa, N.; E’ima, A. Znt. J. Clin. Pharmacol. Ther. Toxic. 1985,23,529-5!34 21. Wagner, J. G. J. Pharm. Sci. 1969,58,1253-1257. 22. Langenbucher, F. J. Pharm. Phurmacol. 1972,24,979-981. 23. Lin, S . Y.;Kawashima, Y. Phurm. Res. in press. 24. Emi, W.; Eckert, M. Acta Pharm. Technol. 1983,29,281-286. 25. Riegelman, S.; U ton, R. A. in Drug Absor tion, F’rescott, L. F.; Nimmo, W. S., EBB.,ADIS: New York, 19&; pp 297-312.
Acknowledgments This paper is Part X of “Studies on Microencapsulation”. This work was su rted in art by National Science Council (NSC-74-0412-B075-frNSC-74-8412-8075-36, NSC-74-0412-BO7538),Taiwan, Republic of China.
Journal of fharmaceufical Sciences / 223 Vol. 76, No. 3, March 1987
LIN' AND JUEI-CHYI YANG
Received June 11, 1986, from the Biophannaceutics La&ratory, De rtment of Medical Research, Veterans General Hospital, Shih-Pai, Taipei, Accepted for publication January 6, &7. Tahvan, Republlc of China.
Abstract 0 The Moavailability of theophylline microcapsules prepared by using ethylene-vinyl acetate (EVA) copolymer a8 a coacervationlndudng agent was studied in rats. The dissolution rate of the microcap suies was determined by the rotating-basket and rotating-bottle methods. The higher the concentration of EVA copolymer used, the more sustained was the release of theophylline from the microcapsules. The mean maximum serum levels (&) and time to maximum serum levels (t-) were not significantly different for theophylline mlcrocapsuies prepared by a lower concentration of EVA copolymer (0 and 0.83%, respectively), compared with those for theophylline powder; whereas a significant difference was found when the higher concentration of EVA copolymer was used (>1.7%). With regards to the area-under-the-curve (AUC) value, there was no significant difference between the theophylline powder and theophylline microcapsules. The elimination kinetics and the corresponding half-life (ten) were significantly different when the concentration of EVA copolymer was >3.3%.From the above results, it is evident that theophylline microcapsules prepared by using 3.3 and 5.0% EVA copolymer as the coacervation-indudngagent may act as sustained-release dosage forms. The correlation between the dissolution rate in vitro and the bioavaiiabilrty In rats for theophylline microcapsules was investigated. The mean C,, and -t axrelated well with the time taken to release 75% of the drug in vitro (b6%);however, the mean AUC showed no valid correlation with b6%.This implies that the dissdution rate correlated better with the rate of absorptlon (C-, 4-) than with the extent of absorption (AUC). Moreover, a llnear relationship was aim found between the time taken to release 50% of the microcap sules in vitro (&) and the time taken to absorb 50% of drug In vlvo indicating the presence of an in vitro:in vlvo correlation. The Weibull function (&) was also used to evaluate the In vttro:in vlvo correlation. It obvlously indicates that & significantly conelated wlth t, (In vIyo) (rotating-basket method: r = 0.9922; rotating-bottle method: r = 0.9929). Thus, the results suggest that the theophylllne microcapsules prepared by udng EVA copolymer a8 a coacervatlonindudng agent not only possess a sustained-release behavior, but also give a good conelation between in vttro dissolutlon rates and in vlvo absorption.
(e),
Theophylline has been popularly used as a bronchodilator for the treatment of chronic asthma and chronic obstructive lung disease.' A narrow therapeutic range of serum theophylline concentration (10-20 ccg/mL) often has resulted in many adverse side effects such as gastrointestinal (GI) irritation (>20 pg/mL), central nervous system (CNS)effecta (>20 pg/mL), focal and generalized seizure (>40 ccg/mL), and cardiovascular toxic effects (>40 pghdd.1 Sustained-release theophylline preparations offer the advantage of less frequent dosing with decreased fluctuation in the serum theophylline level during the dosing interval; thus, these preparations have been widely used in clinical therapy.We have previously shown that theophylline release behavior from ethylcellulose microcapsules, prepared by using ethylene-vinyl acetate (EVA)copolymer as a coacervationinducing agent, was dependent on the concentrations of the coacervation-inducing agent."7 The release kinetics of theophylline from ethylcellulose microcapsules agree with both 00~-3549/87/~-0219$0 .&YO 1 0 1987. American Pharmaceutical Associetion
first-order and square-root-of-time equations. But, after the application of the differential rate treatment, it was found that kinetic data actually conformed with the first-order equati0n.B-g In the present study, the bioavailability parameters of theophylline microcapsules prepared by using various concentrations of a coacervation-inducing agent were examined. We also attempted to clarify the relationship between the in vitro dissolution rates determined by the rotating-basket and rotating-bottle dissolution methods, and the in vivo absorption kinetics for theophylline microcapsules.
Experimental Section MaterialeTheophylline (Wako Pure Chemical Industry, Tokyo), ethylene-vinyl acetate copolymer (EVA; vinyl acetate content 28%; Toyo Soda Manufacturing Co., Tokyo), and ethylcellulose (100cps; h w Chemical Co., MI) were used. Other materials were of analytical reagent grade. Preparation of Theophylline Microcapaules-Theophylline microcapsules were prepared by the phase-separation method, using EVA copolymer aa a coacervation-inducing agent, aa described in previous studies."" Three hundred millilitera of cyclohexane solution, containing from 0 to 5% of EVA copolymer, were respectively placed in a 1000-mL, three-neck, round-bottomed flask equipped with a stirrer, a thermometer, and a reflex condemr. Theophylline (3 g) and ethylcellulose (3 g) were added to the stirred cyclohexaneEVA copolymer solution (150 rpm) at mom temperature. The system waa then heated to 82 "C to form a homogeneom suspenaion. With continuous stirring, the system was allowed to cool to 24 "C, and waa then stirred for another 10 min. During the cooling proceee, a pheee separation occurred. The microcapsules were separated from the solution by decantation, rinsed with n-heme, and dried at 40 "C in a vacuum drier for 24 h. Theophylline microcapsulea (40-60 meah; 250-450 pn) were used for the experiment. Dissolution Studies-The dieeolution rate of theophylline from microcapsules in a pH 1.2 eolution was determined by using the previously reported rotating-basket (100 rpm) and rotating-bottle methode (20 rpm) at 37 'C.7.11 Samples of the microcapsules (equivalent to 100 mg of theophylline for use in the rotating-baaket method and 30 mg of theophylline for use in the rotating-bottle method) were taken for the in vitro dieeolution studiee. The volume of diesolution medium used waa 500 mL for the rotating-basket method and 50 mL for the rotating-bottle method. These two dissolution methoda were carried out under sink conditiona (solubility of theophylline in pH 1.2 solution was -11.68 mg/mL). Aliquota of the diesolution medium were aaaayed for theophylline by ultraviolet spectrophotometry at 270 nm (UVIKON 810, Kontron, Switzerland). Bioavailability Studie-tudies were conducted on male Sprague-Dawley rate weighing 300350 g. The animals were fasted from the night before and throughout the experiment. Each rat waa given a dose of -40 m g k g of theophylline powder and an equivalent doee in microcapsule form (250450 pn). The test sample waa administrated orally with 10 mL of water by a feeding tube. Serial blood samplee (0.7 mL) were taken a t 0.25,0.5,1.0,1.5,2.0,2.5,3.0, 4.0,5.0,6.0,8.0,10.0,12.0,24.0,26.0, and 27.0 h after zero time. The samples were centrifuged, and the serum waa removed and stored at - 20 "C until the time of analysis. The serum theophylline concentraJournal of Phannaceutkal Sdences / 21Q Vol. 76, No. 3, March 1987
Results and Discussion
tion was assayed utilizing a homogenous enzyme immunoaasay technique (EMIT system, Syva). Bioavailability and Pharmacokinetic AnalysigTheophylline absorption from the various dosage forms was characterized by a single-compartment, first-order pharmacokinetic model.13 Although a multicompartment model may be more accurate, a single-compartment model is sufficientlyaccurate in the case of theophylline and is more easily understood for clinical purposes. Thus, the serum concentration of theophylline, C,at any time, t , after a single oral dose, D, is deacribed by14
In Vitro Dissolution of Theophylline From Ethylcellulose Microcapsules-The dissolution rate profiles of theophylline microcapsules determined by two different dissolution methods are shown in Figure 1. Each point represents the average of three experiments. As shown in Figure 1, the rate of dissolution of theophylline from ethylcellulose microcapsules was related to the concentration of the coacervationinducing agent. The higher the concentration of EVA copolymer used, the slower the release rate of the theophylline microcapsules. This might be attributed to the lower porosity and thicker wall of the microcapsules that were prepared by using the higher concentration of EVA ~opolymer.~ The rotating-bottle method has been recommended for testing sustained-release oral dosage forms because it is reasonably simple in practical use and satisfactorily parallels physiological conditions.18 Figure 1also indicates that the release rate of theophylline as determined by the rotating-bottle dissolution method was slightly faster than that determined by the rotating-basket dissolution method, although the rotating speed of the rotating-basket method was higher than that of the rotating-bottle method (100 rpm versus 20 rpm). The different dissolution rates might therefore be due to the different motion patterns in the dissolution medium. The microcapsules studied with the rotating-bottle dissolution method were completely immersed and moved in the dissolution bottle by means of air moving rather than by laminar motion, as was the case of the aggregated microcapsules in the rotating basket, leading to a fast release. Serum Concentration and Pharmacokinetic Parameters of Theophylline after Oral Administration of Theophylline Microcapsules-The rats used in this study tolerated the procedures well and remained clearly conscious throughout the experiment, although several blood samples were taken. The serum concentration versus time curves of theophylline in rats corresponded to a one-compartment open model, as shown in Figure 2. The absorption of theophylline from theophylline powder was very rapid, whereas the microcapsules prepared with the higher concentrations of EVA copolymer produced a more gradual rise and a more sustained plateau of serum levels of theophylline. Figure 2 indicates that the higher the concentration of EVA copolymer used, the more sustained the release behavior of the theophylline microcapsules. Moreover, the theophylline microcapsules prepared by using a higher concentration of EVA copolymer gave a transient steady-state plateau level of theophylline in the serum after administration without producing a sharp peak of serum concentration.
where F is the fraction of the drug absorbed, Vd is the apparent volume of distribution, and k, and kel are the pseudo first-order rate constants for absorption and elimination, respectively. The log plasma theophylline concentration-time curve was plotted. From the terminal linear portion of the curve, the apparent elimination rate (k,,) was obtained from the slope of linear portion by least squares regression analysis. The half-lives of theophylline microcapsules were calculated by dividing 0.693 by the elimination rate constant. The C , , , and tmnxvalues were recorded for each formulation, where , ,C was chosen as the highest observed serum theophylline concentration after oral administration of the dose and t,,, was the time of the maximum measured serum concentration. Another parameter, AUC, from zero to the final sampling time (AUC,,), was calculated using the trapezoidal rule. Moreover, AUC+, is the serum AUC from time zero to infinity calculated fromI6
where C,is the final serum theophylline concentration obtained, and kel is the elimination rate constant obtained from eq 1. The rate and extent of the theophylline absorption from test samples were examined from the Wagner-Nelson equationt6 Fab =
Ct + kel
[AUCo-tl [AUCo-scl
(3)
where Febis the fraction of dose absorbed at time t . The relative bioavailability value (F,,) for the single-dose study was determined by17 FreI =
AUCo,, (theophylline microcapsule) AUCo,, (theophylline powder)
X
Dose (theophylline powder) Dose (theophylline microcapsule)
(4)
100
Flgure 1-Effect of EVA copolymer concentration on the release of theophylline from microcapsules in pH 1.2 solution. Key: solid points: rotating-basket dissolution method; open points: rotating-bottle dissolution method; concentration of EVA wpolymer: (0, +) 0%; ( 0 ,V) 0.83%; (0, ). 1.7%;(0, 0 ) 3.3%; (A, A) 5.0%.
80
’0 0)
!
4
.A’
60
t
U
$
40
U 0)
a
20
t
0 0
2
4
6
8
0
Ti.me,
220 /Journal of Pharmaceutical Sciences Vol. 76, No. 3, March 1987
2
4
6
8
Figure 2-Serum theophylline concentration levels after oral administration of theophylline powder and theophylline microcapsules. Key: concentration of EVA copolymer: (A) theophylline powder; (6) 0%; (C)0.83%; (D) 7.7%; ’ (E) 3.3%; (F) 5.0%. The bars indicate the standard error of the mean (SEM).
:
o
0
2 6 $ 60 g 2 50
B
A
5 0 h
24
16
:IA
C
D
F
Time, h
The bioavailability and pharmacokinetic parameters were estimated from the serum concentration of theophylline after4 administration of either theophylline powder and theophylline microcapsules (Table I). The mean values of the maximum serum level (Cmu) and the time to maximum serum level (tmm) were not significantly different (p > 0.05) for theophylline microcapsules prepared by using the lower concentrations of EVA copolymer (0% and 0.83%)compared with the theophylline powder group. The higher C,,, value is consistent with the shorter t,,, value. However, a significant difference was found in theophylline microcapsule groups prepared by using higher concentrations of EVA copolymer (>1.7%). Furthermore, the serum level of theophylline slowly decreased for a long time for theophylline microcapsules prepared by using 3.3 and 5.0% of EVA copolymer. This indicates that theophylline microcapsules prepared by using higher concentrations of EVA copolymer exhibited a better sustained-release behavior than theophylline microcapsules prepared by using a lower concentration of EVA copolymer. The AUC values for both theophylline powder and theophylline microcapsules were not significantly different (p > 0.05). The AUC values of theophylline microcapsules prepared by using 5% of EVA copolymer were larger than any one of the AUC values. This may be due to
the higher value of final serum theophylline concentration and the lower value of the elimination rate (eq 2). Theophylline elimination kinetics and the corresponding t1/, are also tabulated in Table I. The elimination data and the corresponding t E values are in agreement with the results of C,,, and t,,,. There was no significant difference in the elimination rate and half-life values between theophylline powder and theophylline microcapsules prepared by using EVA copolymer as a coacewation-inducing agent from 0 to 1.7%. By contrast, there was a significant difference between theophylline microcapsules prepared by using 3.3-5.0% of EVA copolymer and theophylline powder. Assuming that the theophylline powder was completely absorbed’e and its relative bioavailability (Frel).probably approximated 1008, then the relative bioavailability of the five types of theophylline microcapsules ranged from 77.6 to 105.2%. The higher the concentration of EVA copolymer used, the larger the F,,l value of theophylline microcapsules. The results of the bioavailability data and pharmacokinetic parameters indicate that theophylline microcapsules prepared by using 3.3 and 5.0%of EVA copolymer as a coacervation-inducing agent possess sustained-release action. In Vitro:In Vivo Relationship-The relationship of tT5%, the time taken to release 75% of drug in vitro to , , C tmax,
Table CPharmacoklnetlc Parameters of Theophylllne Mlcrocapruler after Oral Admlnlstratlon In Rats
~
0
0.83
6 6
42.23 2 4.65’
1.93 2 0.09
333.72 2 72.89
0.119 2 0.01
5.89 t 0.68
NS
NS
NS
NS
NS
41.53 t 4.65
2.08 2 0.29
0.128 2 0.03
2.75 2 0.26 p < 0.05 5.34 2 0.57
393.18 2 56.73 NS 407.82 2 21.11 NS 388.14 2 29.76
0.120 2 0.02
6.04 2 1.05 NS 6.12 2 0.83
NS
NS
NS
6.43 t 0.71 p < 0.05 1.62 2 0.21
450.32 2 80.72 NS 429.76 2 30.63
0.066 2 0.04 p < 0.05 0.052 5 0.01 p < 0.05 0.134 2 0.02
9.50 2 0.69
p < 0.05
NS 1.7
6
37.53 2 1.40
p < 0.05‘ 3.3
8
20.18 2 5.25
p < 0.05 5.0
TheoDhvllined
8 6
17.95 2 0.78 p < 0.05 51.78 2 1.69
NS
NS
77.6 91.5 95.1 90.1
p < 0.05 15.64 2 3.84
105.2
p < 0.05 6.01 2 0.79
100
a Mean 2 SEM. No significant difference compared with theophylline group. =Significantdifference compared with theophylline group. “Actual dose per rat was 40 mg/kg.
Journal of Pharmaceutical Sciences / 221 Vol. 76, No. 3, March 7987
agent. Moreover, a parallel relationship was shown for the rotating-basket dissolution method and the rotating-bottle dissolution method, as shown in Figure 5.This demonstrates that the rotating-basket dissolution method was in close agreement with the rotating-bottle dissolution method. In conclusion, this study confirms that our theophylline microcapsules, prepared by using EVA copolymer as a coacervation-inducing agent, exhibited sustained-release be-
and AUCb, in vivo, was investigated (Figure 3). To compare each correlation, we used the coefficient of determination r2. It is obvious that there were statistically significant correlations between t754tand, ,C (r = 0.9921),and t75% and t,,, (r = 0.9837);whereas, t759band AUC-, (r = 0.3359) showed no correlation between in vitro and in vivo results. This shows that the in vitro dissolution rate (both dissolution methods) correlated better with , C and t,,, which correspond to the rate of absorption, than with AUCh,, which reflects the extent of absorption. The poorer correlation between dissolution rate and AUCk, may be due to the fact that the extent of absorption was affected to a greater extent by various physiological factors in the GI tract, compared with the rate of absorption. This result has also been reported by several investigators.20 In another attempt to evaluate the absorption kinetics of microcapsules, the serum profiles of Figure 2 were transformed using the Wagner-Nelson equation (eq 3). Plots for tSo8,the time taken to release 50% of drug from microcapsules in vitro, and F,6tQ,the time taken to absorb 50%of drug in vivo, are presented in Figure 4. The correlations between the bioavailability data and the dissolution variables were statistically significant, and the correlation coefficient (r) in no case was <0.9. Drug release from dosage forms has been suggested by using the Weibull distribution function.21-2s The Weibull analysis of dissolution data can play an important function in product development and may lead to more rational quality control. We used the Weibull distribution function to describe the dissolution curves of theophylline microcapsules:
log [-ln (W/W,,)] = b log ( t - Ti)- log a
- 00
3
1
5
7
t50%. h
(5)
of in vitro t-
Flgure &Correlation
and in vivo =%. Key: see Figure
1.
where W is the quantity of drug remaining in the microcapsules, W,is the initial quantity of drug in the microcapsules, a and b are the scale parameter and shape factor, respectively, Ti is the location parameter, and td represents the time interval necessary to release 63.2% of the drug. The Weibull function was also used for interpreting the profiles of the drug absorbed versus time.26 By this approximation, we focused on t d (in v,tro) and td (in viva) of the Weibull functions. The evaluation of Weibull distribution parameters was performed by means of a linear regression analysis. From the shape factor, b, of the Weibull function, all the microcapsules appear to release their content in a first-order process, since the b value ranged from 0 to 1. Figure 5 shows the linear relationship between td (in vitm) and td (in ,,iv0) of theophylline microcapsules (rotating-basket method: r = 0.9922;rotatingbottle method: r = 0.9929).This also indicates that a good correlation between in vitro dissolution and in vivo absorption was found in our developed theophylline microcapsules that were prepared by using EVA as a coacervation-inducing
2
0
4
8
6
10
birvibo,h
regression between b fin vit,,,) theophylline microcapsules. Key: see Figure 1.
Figure +Linear
and &
(in V
10-
5 4
8 -
Flgure 3-Correlation
-\&,
of
t,5% with C-, t,,, and AUC,,. Key: s8e Figure 7 ; the bars indicate SEM.
52 '4
u~
0
2
4
6
t75r.
222 1Journal of Pharmaceutical Sciences Vol. 76, No. 3, March 1987
8 h
1
0
0
2
4
6
t75X.
8 h
1
0
0
2
4
6 t75%. h
8
1
0
~
of ~
J
havior both in vitro and in vivo. The in vitro:in vivo mrrelation has been established. Further studies need to be performed to examine the bioavailability of theophylline microcapsules in humans.
References and Notes 1. Hendeles, L.; Weinber er, M. Pharmacothera 1983,3,2-43. 2. Baptista, R.; Driscoll, F. A m . Pharm. 19861yNS24,285-293. 3. Weinber er, M. Pharmacothemm 1984,4,181-198. 4. Cummisfe J M . Po a, V. J. Asthma 1984,21,243-257. 5. M e r i c h , A.; h e & - , S. J.; Green, E. R. J. Allergy Clin. Zmmunol. 1981,61,465-467. 6. Lin, S. Y.; Yang, J. C.; Jiang, S. S. J. Taiwan Phurm. Sci. 1985, 37, 1-10. 7. Lin, S.Y.; Yang, J. C. J. Microencapsulation 1985,2,315-325. 8. Lin, S.Y.; Yan ,J. C. J. Taiwan Pharm. Sci. 1986,38,160-165. 9. Benita, S. Ap Biochem. Biotech. 1984,10,255-258. 10. Lin, S.Y. J. kicmencapsulution 1985,2,91-101. 11. Lin, S . Y.; Yan ,J. C. J. Controlled Release 1986,3,221-228. 12. Lin, S.Y.; Ho, T.; Chiou, H. L. Bwmater. Med. Dev. Art. Org. 1986,13,187-201. 13. Shen, D.D.;Fixley, M.; Azarnoff, D.L. J.Pharm. Sci. 1978,67, 916-919. 14. Lesko, L. J.; Canada, A. T.; Eastwood, G.; Walker, D.;Brous*an, D. R. J. Pharm. Sci. 1979,68,1392-1394.
6.
k.
&. f.
15. Gibaldi, M.; Perrier, D. Pharmacokinetics, Vol. I; Marcel Dekker: New York, 1975;pp 293-296. 16. Wagner, J. G. Fundamentals of Clinical Phurmacokinetics, 1st ed.; Dru Intelli ence: Hamilton, IL, 1975;pp 174-176. 17. Green, D.; laccar, C. L.; Helsel, C.L.; Niehls, M. E.; McGeady, S. J.;Mansmaun, H. C. J.Asthma 1985,22,159-163. 18. Souder, J. C.; Ellenbogen, W. E. Drug Standards 1958,26,77-
1.
79
19. Hendeles, L.; Weinberger, M.; Bighley, L. Am. J. Hosp. Phurm. 1977,34,525-527. 20. Aoyagi, N.; Ogata, H.; Kaniwa, N.; E’ima, A. Znt. J. Clin. Pharmacol. Ther. Toxic. 1985,23,529-5!34 21. Wagner, J. G. J. Pharm. Sci. 1969,58,1253-1257. 22. Langenbucher, F. J. Pharm. Phurmacol. 1972,24,979-981. 23. Lin, S . Y.;Kawashima, Y. Phurm. Res. in press. 24. Emi, W.; Eckert, M. Acta Pharm. Technol. 1983,29,281-286. 25. Riegelman, S.; U ton, R. A. in Drug Absor tion, F’rescott, L. F.; Nimmo, W. S., EBB.,ADIS: New York, 19&; pp 297-312.
Acknowledgments This paper is Part X of “Studies on Microencapsulation”. This work was su rted in art by National Science Council (NSC-74-0412-B075-frNSC-74-8412-8075-36, NSC-74-0412-BO7538),Taiwan, Republic of China.
Journal of fharmaceufical Sciences / 223 Vol. 76, No. 3, March 1987