Gastric Emptying of Tablets and Granules in Humans, Dogs, Pigs, and Stomach-Emptying-Controlled Rabbits NOBUOAOYAGI“, HIROYASU OGATA*,NAHOKO KANIWA*, MITSURU UCHIYAMA*, YUKIO YASUDA~,AND YOSHIKUNI TANIOKA~ Received November 13, 1991 from the ‘Nafional Institute of Hygienic Sciences, 7-78-7 Kamiyoga, Sefagaya-ku, Tokyo 7 5 8 , Japan, the *Meii College of Pharmacy, 7-22-7 Yato-cho, Tanashi-shi, Tokyo 788, Japan, and. the SCenfral Insfitufe of Experimenfal Animals, 7480 Nogawa, Accepted for publication March 30,1992. Takafsu-ku, Kawasaki-shi, Kanagawa 273, Japan. Abstract Rates of gastric emptying of nondgestible tablets and granules in humans were compared with those in three animal models: dogs, minipigs, and stomach-emptying-controlledrabbits. The rates of gastric emptying of both dosage forms in dogs tended to be faster than or similar to those in humans, whereas the rates in pigs were slower. In stomach-emptying-controlled rabbits, no tablets were emptied from the stomach because of their large size. The rate of gastric emptying of granules in rabbits was slow and variable. Fooddelayed gastric emptymg in dogs, especially for tablets. In rabbits, the rate of gastric emptying of granules was faster when the granules were given before feeding, in comparison with that after feeding or under fasting conditions. We concluded that the dog is a better animal model for bioavailabilitystudies under fasting conditions than the pig and the rabbit.
Bioavailabilities of drug formulations for human use are often estimated in animals, because of the ethical and economic difficultiesof human studies. Dogs are extensively used as an animal model, but the in vivo data for oral dosage forms in dogs are often not comparable with those in humans.1-4 Satisfactory results have also not been obtained in minipigs5 and stomach-emptying-controlled rabbits6 with griseofulvin tablets. The discrepancies in bioavailabilities between humans and animals are basically caused by differences in gastrointestinal physiology. Among various physiological factors, gastric emptying is one of the key factors affecting dissolution and absorption of oral drugs. Our previous studies of griseofulvin tablets suggested fast gastric emptying in dogs,134 slow emptying in minipigs: and variable gastric emptying in stomach-emptying-controlled rabbits.6 To avoid the misinterpretation of animal data, the differences in gastric emptying between humans and animals should be understood. However, such differenceshave not been clarified well, despite several studies comparing gastric emptying for humans and dogs.7.8 The present study was conducted to clarify the differences in gastric emptying between humans and animals by use of nondigestible tablets and granules. As animal models, dogs, minipigs, and stomach-emptying-controlled rabbits were used. These animals were chosen, because dogs are widely used for bioavailability studies, minipigs are considered similar in physiology to humans,9 and stomach-emptyingcontrolled rabbits are reported to show improved gastric emptying.lc”n The effects of food on gastric emptying in dogs and rabbits were also investigated.
Experimental Section Preparations-Four preparations were used in this study: two kinds of spherical tablet, containing 250 mg of aspirin and 5.7% (wlw) barium sulfate, and two kinds of cylindrical granule, containing 50% (w/w) aspirin and 318 (w/w) barium sulfate. The tablet and granule forms of aspirin are commercially marketed by Shionogi Pharma1170 I Journal of Phannaceufical Sciences Vol. 87, No. 12, December 7992
ceutical Co. Ltd.(Osaka,Japan) as EA Tablet Shionogi and Minimax, respectively. All of the preparations were enteric coated, except for the barium sulfate granule, which was coated with ethyl cellulose. The sizes and densities of these formulations are shown in Table I. Humans and Animals-Healthy, male human volunteers (n = 12) ranging from 22 to 54 years of age (mean, 31 years) participated in the study with written informed consent. Male beagle dogs (9-13kg), female a t t i n g e n minipigs (16-25 kg), and male albino rabbits (2.54.5kg) were used. Stomach-emptying-controlled rabbits were prepared as described by Maeda et a1.12 f i r gastric lavage 2 days before dosing, the rabbits with a cangue were fed 200 g of a special soft diet prepared to contain two parts solid diet CR-S(Nihon Clea Co. Ltd.,Tokyo, Japan) and three p a w water. Gastric Emptying of Aspirin Tablets and G r a n u l e g O n e tablet (250mg) of aspirin was administered to humans (n = 121,dogs (n = 12),minipigs (n = 5),and stomachemptyingcontrolled rabbits (n = 8)after an overnight fast, together with 200,50,100, and 10 ml of water, respectively. For the granule study, a 200-mg granule containing 100 mg of aspirin was administered to dogs (n = 6) and humans (n = 5) aged 22-35 years (mean, 29 years), and a 150-mg granule containing 75 mg of aspirin was given to minipigs (n = 5)and stomach-emptyingcontmlled rabbits (n = 8). The pigs and rabbits administered the granules were the same animals that were given the tablets, and a 1-week interval was allowed between administrations. The pigs and rabbits were also given a commercial plain tablet of 500 mg of aspirin (Bayer Aspirin, Bayer AG, Osaka, Japan) to determine the rate of elimination of salicylic acid. All human subjects were prohibited from taking food until 4 h after dosing. No food was given to dogs or pigs until the end of the test, but 50 g of the special soft diet was given to the rabbits just after dosing, as in previous bioavailability studies.6-10Blood samples from the animals and urine samples from the human subjects were taken at different times for up to 24 h, and the plasma, serum, and urine samples were stored frozen until assay. All salicylates excreted in human urine were determined colorimetrically.~~ Concentrations of salicylic acid in plasma of dogs were determined by HPLC, as described previously8 with an octadecylsilane column (TSK GEL LS-410;5 pm; Tosoh Co. Ltd., Tokyo, Japan). Concentrations of salicylic acid in serum of pigs and rabbits were determined fluorometrically.~4The differences in gastric emptying rates among species were estimated from the difference in the observed lag time and time to the peak concentration in blood or urinary excretion rate-time curves for salicylic acid. Table Cslzes and Densltles of Tablets and Granulw of Asplrln and Barlum Sulfate Formulation Diameter
Size, cm Thickness
Aspirin Tablet Granule
Barium Sulfate Tablet Granule
0022-3549/924200-7 7 70$02..50/0 0 7992, American Phannaceuflcal Association
Gastric Emptying of Barium Sulfate Tablets and GranulesFive tablets and 50 granules of barium sulfate were administered to dogs (n = 4), minipigs (n = 3), and humans (n = 3; aged 34, 37, and 54 years), according to a two-way crossover design with a 1-week interval. To stomach-emptying-controlledrabbits (n = 51, two tablets and 50 granules were administered. The dosing conditions were the same as those in the aspirin study. The numbers of tablets and granules remaining in the stomach were determined by periodic roentgenography of unanesthetized humans and animals, except for pipe, which were lightly anesthetized by intramuscular injection of 1-2 mg of droperidol per head before roentgenography to prevent their movement. The rates of gastric emptying of barium tablets and granules in dogs and rabbits after postprandial administration were also investigated;the formulationswere administered to rabbits and dogs at 30 min after feeding with 50 g of the special soft diet and 250 g of commercial solid diet CD-1 (Nihon Clea Co. Ltd.,Tokyo, Japan), respectively. The gastric emptying rates of rabbits under fasting conditions were also investigated. None of the animals were given any food until 24 h after dosing. Statistical Tests-The differences in gastric emptying between humans and animals were examined statistically by Dunnett's multipleamparison test after examination by analysis of variance.
Results and Discussion Drugs in entericcoated formulations are not released and absorbed until the formulations reach the intestine; therefore, the slower the gastric emptying of enteric formulations, of the the longer are the resulting lag time16 and t, concentration in blood-time curves. Therefore, the differences in gastric emptying between humans and animals can be estimated from these parameters, although t,, cannot be used for estimation if elimination of a drug from humans differs significantly from elimination from animals.16 Using these characteristics of entericcoated formulations, we first estimated the differences in gastric emptying between humans and animals from the lag time and ,t of salicylic acid after oral administration of enteric-coated-tablet and granular forms of aspirin. The elimination rate constants for salicylic acid after dosing with aspirin in humans (n = 12), dogs (n = lo),rabbits (n = 81, and minipigs (n = 5)were 0.247 f 0.074(mean f SD), 0.208 2 0.083,0.204f 0.135,and 0.382 & 0.177h-', respectively, determined with an enteric-coated tablet of aspirin in humans and dogs and a plain tablet of aspirin in pigs and rabbits. The differencesin elimination rate constants between humans and these animals were not large enough to produce significant differencesbetween ,t values for humans and for animalsI6 after administration of aspirin, which was absorbed rapidly (absorption rate constant, 3-4 h-lI.17 This result indicates that, if t, values after administration of enteric-coated aspirin formulations to humans and animals, differ significantly, these differences are mostly ascribable to differences in gastric emptying rates. Gastric Emptying of Tablets-Figures 1 and 2 show the lag times and t, values for individual humans and animals after oral administration of entericcoated tablets of aspirin under fasting conditions, except for rabbits, which were provided with a special soft diet just after dosing, as is the usual procedure.6JO The mean lag times for the tablets were short in dogs (0.8 h) and humans (1.2h) but long in pigs and rabbits. Lag times of >24 h were observed in some pigs and rabbits; such long lag times indicate that the enteric-coated tablets still remained in their stomachs at >24 h after dosing. The t, values were similar to the lag times (Figure 2);the mean t,, was reduced in the following order: dogs (2.9h) < humans (5.3h) < pigs and rabbits. These findings indicate that gastric emptying of aspirin tablets in dogs was faster than or similar to that in humans, whereas that in pigs and rabbits was slower. To confirm the above-described findings, we determined the
888 Human Dog Pig Rabbit
Flgure 1-Individual lag times for the urinary excretion rate or concentration in blood-time curves for salicylic acid after oral administration of enteric-coated tablets (0)and granules (0)of aspirin to humans, dogs, minipigs, and stomach-emptying-controlled rabbits under fasting conditions, except for rabbits, which were fed after dosing. The bars show mean values. The differences in the mean lag times for the tablets between humans and pigs and between humans and rabbits were found significant by Dunnett's test (p = 0.05). No significant differences were observed for granules (p = 0.05). In this statistical test, lag times exceeding 24 h were assigned a value of 24 h for convenience of calculation.
o m 0
0 0 0
Human Pig Human Pig Dog Rabbit Dog Rabbit Flgure 2-Individual Lax values for the urinary excretion rate or concentration in blood-time curves for salicylic acid after oral administration of enteric-coated tablets (0) and granules (0) of aspirin to humans, dogs, minipigs, and stomach-emptying-controlled rabbits under fasting conditions, except for rabbits, which were fed after dosing. The bars show mean values. The differences in the mean f,* values for the tablets between humans and pigs and between humans and rabbitswere found signlficant by Dunnett's test (p = 0.05). No significant differences were observed for granules (p = 0.05). In this statisticaltest, 5,values exceeding 24 h were assigned a value of 24 h for convenience of calculation.
gastric emptying rates by a direct method with barium sulfate tablets similar in size and density to the aspirin tablets (Table I). The barium sulfate tablets were administered to humans and animals under the same dosing conditions as those used for the aspirin tablets. Figure 3 shows the fractions of barium sulfate tablets remaining in individual stomachs. The tablets were completely ejected from the stomach of dogs within 1.5 h but not of humans, one of whom retained 80%of the tablets in the stomach even at 2 h after dosing. Considering that the gastric emptying of large particles closely follows the phasic gastric contraction of the interdigestive migrating motor complex,ls phasic contraction might not have occurred in this subject in the 2 h after dosing, although it usually occurs in a cycle of 1.5-2 h,19 or the contraction might not have been strong enough to expel the tablets completely from the stomach. The barium sulfate tablets were emptied very slowly from Journal of Pharmamrn'cal Sciences I 1171 Vol. 81, No. 12, December 1992
L 2 10
5 ' 2 4
Flgure SFractions of enteric-coated tablets of barium sulfate sulfate remaining in the stomachs of individual humans (n = 3), dogs (n = 4). minipigs (n = 3), and stomach-emptying-controlled rabbits (n = 5) under fasting conditions, except for rabbits, which were fed after dosing. Five tablets were given to humans, dogs, and pigs, and two tablets were given to rabbits. Dunnett's test (p = 0.05) for the mean fractions remaining in the stomachs at 2 h revealed significant differences between humans and pigs and between humans and rabbits.
the stomach of pigs, a result indicating that the interdigestive contraction in pigs20 is not forceful enough to expel the tablets. In stomach-emptying-controlledrabbits, all tablets remained in the stomach even at 24 h after dosing, the tablets may have been too large to pass through the gastric pylorus, as indicated by Takahashi et al.,21 who revealed that 7.7mm-diameter tablets were not expelled from the stomach of rabbits. Gastric Emptying of Granules-Figure 1 shows the lag times for enteric-coated granules of aspirin in individual humans and animals. The lag times were all within a narrow range of 1 h, and no significant differences could be found among the four species. On the other hand, the differences in ,t values among the species were more distinguishable than those in lag times, as shown in Figure 2. The mean t,, (3.4 h) for dogs was shorter than that (4.5 h) for humans, but those for pigs (6.6 h) and rabbits (7.3 h) were longer. The t , for stomach-emptying-controlledrabbits showed large interindividual variability. These findings indicate that gastric emptying in dogs is faster than or similar to that in humans and that gastric emptying in rabbits and pigs is variable and slow, respectively. To confirm these findings, we determined the movement of granules from the stomach by using barium sulfate granules similar in size and density to the aspirin granules (Table I) (Figure 4). However, the granules were not always clearly visible by roentgenography because of their small size. Therefore, some error might have been introduced when the numbers of granules on the roentgenograms were counted. Even in dogs, the counting of granules was often difficult because of the movement of the animals when the roentgenograms were taken. Therefore, in dogs, the time when no granules were evident in the stomach was determined, and a line was drawn from this point to the zero-time point (100%) (Figure 4). The granules were completely expelled from the stomach of dogs by 2 h after dosing. Among humans, one subject still retained -90% of the granules in the stomach at 1.5 h after dosing. In this subject, the same one who showed slow gastric emptying of the barium sulfate tablets (Figure 3), the granules appeared to lie along the gastric wall, as if stuck 1172 I Journal of Pharmaceutical Sciences Vol. 81. No. 12, December 1992
Flgure 4-Fractions of ethyl cellulose-coated granules of barium sulfate remaining in the stomachs of individual humans (n = 3), dogs (n = 4), minipigs (n = a),and stomach-emptying-controlled rabbits (n = 5) under fasting conditions. except for rabbits, which were fed after dosing. F i granules were given to humans and animals. Dunnett's test (p = 0.05) for the mean fractions remaining in the stomachs at 1.5 h revealed significant differences between humans and pigs and between humans and rabbits when interpolated values were used in the absence of experimental values at 1.5 h.
to the mucus. Such adhesion might have caused the slow gastric emptying. Also, the phasic contraction promoting gastric emptying22 may not have occurred in this subject within 1.6 h after dosing. In stomach-emptying-controlled rabbits, the barium sulfate granules slowly moved away from the stomach, in agreement with the findings of Goto et al.,23 who found that one-fourth of microcapsules still remained in the stomach of rabbits at 24 h after administration. The rates of gastric emptying of granules in rabbits differed markedly among individuals, probably closely corresponding to the large variability in t , values observed for aspirin granules (Figure 2). This large variability in gastric emptying indicates that it is difficult to control gastric emptying sufficiently by means of a n artificial method; emptying may be influenced by various factors, including the skill of gastric lavage during the preparation of stomach-emptyingcontrolledrabbits and the stress induced by forced drug administration. The gastric emptying of granules observed in rabbits and pigs explains well the absorption behaviors of griseofulvin products in our previous studies6.6; the considerable variability in gastric emptying in rabbits accounts for the large variability in t, after the oral administration of griseofulvin tablets,'J and the slow gastric emptying in pigs accounts for the long t,, (6.6 h) of a rapidly absorbed product compared with that (3.9 h) in humans.6 These findings suggest that griseofulvin, consisting of fine drug particles, should be excluded from the stomach at a rate similar to that of the 1-mm granules, a suggestion that is highly probable judging from the h d i n g s of Meyer et al.,24 who reported that particles of 0.015 and 1 mm were emptied from the stomach at similar rates in dogs. Effects of Food on Gastric Emptying-Food delays that gastric emptying of tablets16 and granules26 in humans, a fact that has generated interest in how food a€Fects gastric emptying rates in animals. Thus, the effects of food on gastric emptying in dogs and rabbits were investigated. Figure 5 shows the gastric emptying of barium sulfate tablets in dogs. The gastric emptying was markedly delayed by feeding, and
30 min after feeding 1
FlgureH r a c t i o n s of enteric-coated tablets of bariumsulfate remaining in the stomach of individual dogs (n = 4) under fasting (- - - -) and conditions. Five tablets were administered to each dog nonfasting (-) except for one fed dog, which was given four tablets. The paired ttest (p = 0.01) revealed a significantdifference between fasting and nonfasting conditions in the time required for 50% of the tablets to be emptied.
some tablets were still retained in the stomach of two of the dogs even at 20 h after dosing, the tablets appearingto adhere strongly to the gastric wall. This extreme delay in gastric emptyingcaused by feeding was not unexpected. Our previous study of enteric-coated tablets of pyridoxal phosphate also revealed very slow gastric emptying, >10 h, in fed dogs, although the emptying in fed humans was fast (3.3 h on average).S The gastric emptying time of >10 h in fed dogs corresponds to the disruption time of the interdigestive migrating motor complex after feeding,26 indicating that gastric emptying in dogs is completely dependent on this complex. The long disruption time (>10 h) in dogs compared with that (-3.5 h) in humans16 shows that fed dogs cannot be used for bioavailability studies of large-size dosage forms, including enteric-coated tablets, unless the disruption time after feeding can be improved to make it similar to that in humans. This improvement could be achieved by controlling calorie intake and food composition.27 The effectsof food on the gastric emptying of barium sulfate granules in dogs were also investigated (Figure 6). Food delayed gastric emptying, but the retarding effect on barium sulfate granules was much lower than that on barium sulfate tablets, in agreement with our previous findings (less of an effect of food on the gastric emptying of aspirin granules than of pyridoxal phosphate tabletsl.8 This result suggests that, under nonfasting conditions, the expulsion of granules from the stomach should depend less on interdigestive gastric contraction than should that of tablets, although the emptying of granules may depend considerably on interdigestive gastric contraction under fasting conditions.22 Figure 7 shows the gastric emptying in rabbits of barium sulfate granules given 30 min after feeding and under fasting conditions. Gastric emptying rates under both conditions were slower than when granules were administered before 100
.t I a 0 C .-
I-it-iP L -
6 ' 1 0 24 31
Time (h) Figure 7-Fractions of ethyl cellulose-coated granules of barium sulfate remaining in the stomach of individual rabbits after administration at 30 min after feeding (n = 6) and under fasting conditions (n = 3).
feeding (Figure 41, a result giving rise to speculation that granules given before feeding might be expelled from the stomach by food. These findings show that, to improve gastric emptying in rabbits, which is extremely slow in conventional rabbits, drug administration before feeding is most favorable. The effects of food on the gastric emptying of barium sulfate tablets in rabbits were also investigated (data not shown). However, no movement out of the stomach occurred even at 24 h after administration, regardless of feeding conditions. In conclusion, the dog is a better animal model for bioavailability studies under fasting conditions than pigs and stomach-emptying-controlled rabbits, judging from the similarity in gastric emptying to humans, although dogs may not be suitable for studies in a fed condition, especially at large-size dosage forms.
References and Notes
1. Aoyagi N.; 0 ata, H.; Kaniwa, N.; Koibuchi, M.; Shibazaki, T.; Ejima, A. J. fhurm. Sci. 1982, 71, 1169-1172. 2. Ogata, H.; Aoyagi, N.; Kaniwa, N.; Koibuchi, M.; Shibazaki, T.;
Time (h) Figure &Fractions of ethyl cellulose-coated granules of barium sulfate remaining in the stomach of individual dogs (n = 4) under fasting (- - - -) and nonfasting (-) conditions. The paired t test (p = 0.05) revealed a significant difference between fasting and nonfastingconditions in the time required for 50% of the granules to be emptied.
Ejima, A.; Shimamoto, T.; Yashiki, T.; Ogawa, Y.; Niehida, Y. Int. J. Clin. Pharmacol. Ther. Toxicol. 1982,20, 576-581. Ogata, H.; Aoyagi, N.; Kaniwa, N.; Shibazaki, T.; Ejima, A.; Takasugi, N.; Mafune, E.; Hayashi, H.; Suwa, K. Znt. J. Clin. Pharmacol. Ther. Taxicol. 1984,22, 240-245. Aoyagi, N.; Ogata, H.; Kaniwa, N.; Ejima, A.; Nakata, H.; Tsutsumi, J.; Fujita, T.; Amada, I. Int. J. Clin. Pharmacol. Ther. Toxicol. 1985,23, 57%504. Aoyagi, N.; Ogata, H.; Kaniwa, N.; Ejima, A.; Yasuda, Y.; Tanioka, Y. J. Pharmacobio-Dyn. 1984, 7, 7-14. Aoyagi, N.; Ogata, H.; Kaniwa, N.; Ejima, A. J. PharmacobioDyn. 1984, 7, 630-640. Chaumeil, J. C.; Piton, Y. Ann. Pharm. Fr. 1973,31,691-704. Journal of Pharmaceutical Sciences I 1173 Vol. 81, No. 12, December 1992
8. Kaniwa, N.; Aoyagi, N.; Ogata, H.; Ejima, A. J. PharmclcobioDvn. 1988,11, 563-570. 9. Withey, R; J. Rev. Can. Bwl. 1973,32,Suppl., 21-30. 10. Maeda, T.; Takenaka, H.; Yamahira, Y.; Noguchi, T. J. Pharm. Sci. 1977,66,69-73. 11. Maeda, T.; Takenaka, H.; Yamahira, Y.; Noguchi, T. J. Pharm. Sci. 1979,68,12861289. 12. Maeda, T.; Takenaka, H.; Yamahira, Y.; Noguchi, T. Chem. Pharm. Bull. 1979.27.30663072. 13. Kaniwa, N.; Ogata, H:; Aoyagi, N.; Ejima, A. J. Pharmacobioe n . 1981,4,860-864. 14. Shibazaki, T.; Nishimura, T.; Hara, M.; Iijima, T. Chem. Pharm. Bull. 1978,26,1737-1745. 15. Mojaverian, P.; Rocci, M. L.; Conner, D. P.; Abrams, W. B.; V l a ~ ~ eP. s ,H. Clin. Pharmucol. Ther. 1987,41,11-17. 16. Kaniwa, N.; Ogata, H.; Aoyagi, N.;Takeda, Y.; Uchiyama, M. J. Pharm. Sci. 1989,78,1020-1024. 17. Mason, W. D.;Winer, N. J. Pharm. Sci. 1981, 70,262-265. 18. Mroz, C.T.; Kelly, K. A. Surg. Gynecol. Obstet. 1977, 145, 369-3 77. 19. Gupta, P. K.;Robinson, J. R. Int. J. Pharm. 1988,43,45-52. 20. Ruckebusch, T.; Bueno, L. Br. J. Nutr. 1976,35,397-405. 21. Takahashi, T.;Shirai, Y.; Nakamura, Y.; Uezono, Y.; Makita, H.;
1174 I Journal of Pharmaceutical Sciences Vol. 87, No, 72, December 7992
Nakanishi, Y.; Imaaato, Y. Chem. Pharm. Bull. 1985,33,54955502. 22. Gruber, P.; Rubinstein, A.; Li, V. H. K.; Bas, P.; Robinson, J. R. J. Pharm. Sci. 1987,76,117-122. 23. Goto, S.;Uchida, T.; Aoyama, T. J. Pharmacobio-Dyn. 1985,8, 270-277. 24. Meyer, J. H.; Dressman, J.;Fink, A.; Amidon, G. Gastroenterology 1985,89,805413. 25. Maekawa, H.; Takagishi, Y.; Doi, Y. Yakuzazgaku 1970, 30, 102-110. 26. Itoh, Z.;Aizawa, I.; Takeuchi, S.; Takayanagi, R. Dig.Dis. 1977, 22. 117-124. 27. Wever, I. D.; Eeckhout, C.; Vantrappen, G.; Hellemans, J.Am. J. Physiol. 1978,235,E661-E665.
This study was sup rted in part by a grant from the Japan Health grateful1 acknowledge the help of Dr. K. Science Foundation. I sa and Dr. K. Miwa with beagc dog studies. We thank Shionogi Co. Ltd. for preparing the barium sulfate formulaPy-aceutical tions.