- Email: [email protected]
Evaluation of Lincomycin as a Cholesterol Gallstone Dissolution Rate Accelerator
Evaluation of Lincornycin as a Cholesterol Gallstone Dissolution Rate Accelerator BOB D. RUSH*’, MARY J. RIJWART‘,BRADLEY D. ANDERSON*”
AND WILLIAM
I. HIGUCHI”
Received June 20, 1984, from the * Diabetes and Gastrointestinal Diseases Research and *Pharmacy Research, The Upjohn Company, Kalamazoo, Present Accepted for publication March 1, 1985. MI 49001 and the ’College of Pharmacy, University of Michigan, Ann Arbor, MI 48109. address: 6University of Utah, College of Pharmacy; Salt Lake City, UT 84112. Abstract 0 These studies were undertaken to test the hypothesis that interfacial resistance may be an important rate-limiting factor in cholesterol gallstone dissoluton. The addition of lincomycin hydrochloride to the gallbladder bile of dogs in an in vitro bath system resulted in an acceleration in the rate of dissolution of a compressed cholesterol
monohydrate pellet incubating in the bile. However,the constant infusion of lincomycin for 13 d directly into the gallbladders of conscious, unrestrained dogs, which resulted in biliary lincomycin concentrations comparable to that of the in vitro tests, did not alter the dissolution rate of a compressed cholesterol monohydrate pellet which had been surgically placed into the gallbladder. We therefore conclude that the interfacial resistance between the cholesterol monohydrate pellet and the bile may be reduced by the addition of lincomycin to the gallbladder bile which, in the in vitro environment,results in an acceleration in the rate of dissolution of compressed cholesterol pellets. However, the ineffectiveness of lincomycin in accelerating the dissolution of cholesterol pellets in vivo suggests that interfacial resistance is not the only rate-limiting factor in gallstone dissolution. Other factor!;, such as mixing, may also be critical. Cholesterol is a normal constituent of vertebrate bile. Although it is virtually insoluble in water, cholesterol is dispersed in bile by the formation of bile acid-lecithin micelles. Hence, the degree of cholesterol saturation in gallbladder bile is a critical factor in the formation or dissolution of cholesterol gallstones. Gallstone dissolution studies have shown that the administration of the bile acids chenodeoxycholic’.’ or ursodeoxycholic acid3s4 will result in some degree of cholesterol stone dissolution. These bile acids alter the composition of bile so that the bile becomes undersaturated with respect to cholesterol and, thus, ultimately leads to the resolubilization of the gallstone. However, the dissolution of the stones progresses slowly; therefore, therapy must be continued for months or even years. The work of Higuchi indicates that the dissolution of cholesterol gallstones proceeds at a much slower rate than predicted from diffusion rate models5 which could indicate that the dissolution is controlled by interfacial resistance?-” Higuchi et al.5 defined interfacial resistance barriers as “barriers to dissolution which arise from intrinsically slow crystal-surface solution equilibria or from ahorbed or deposited substances on the stones surfaces.” Accondingly, he and his co-workers hypothesized that the addition of agents to the bile which reduce the interfacial barrier might “have therapeutic value in decreasing the time required for total gallstone dissolution in patients receiving bile acid therapy.” Utilizing in vitro models, these investigators have demonstrated that benzalkonium chloride, cetylpyridinium and, more generally, primary, secondary, and tertiary amines, quaternary ammonium salts, as well as cationic surfactants, but not anionic surfactants,” are effective gallstone dissolution rate accelerators. The purpose of this study was to test in the in vivo model the concept that the rate of dissolution of cholesterol gallstones in cholesterol-undersaturated bile may be accelerated by the addition of compounds to the bile which reduce the interfacial
0022-3549/85/0600-062l$O 1.OO/O 0 1985, American
Pharmaceutical Association
resistance. Moreover, this study represents one of the first known attempts to test this concept in an animal model which mimics the conditions encountered in humans with cholelithiasis undergoing bile acid dissolution therapy. Lincomycin hydrochloride was chosen because preliminary tests showed it to be better tolerated in animals than many of the other compounds which had been shown to accelerate the rate of dissolution of cholesterol gallstones in the in vitro test system.
Experimental Section In V i t r o Studies-Mongrel dogs were fasted overnight and then anesthetized with sodium pentobarbital, 30 mg/kg iv. The abdominal cavity was opened and bile was aspirated from the gallbladder into a sterile syringe. Each dog served as a bile donor only once. The sample of bile was mixed, divided into two sterile containers, and the appropriate quantity of lincomycin hydrochloride (0 or 45 mmol/L) was added. The samples were frozen for storage. The quantity of bile from each dog was sufficient for duplicates of both control and lincomycin-treated samples. The dosage of 45 mmol/L was determined in a series of pilot studies in dogs to approximate the maximum concentration of lincomycin tolerated when administered into the gallbladders of these animals. Using these bile samples, the rate of dissolution per unit area ( J / A ) of a compressed cholesterol monohydrate pellet and the equilibrium solubility (Cs) of cholesterol was experimentally determined by the methods of Higuchi et al. and Molokhia et al.6.7311The total resistance to dissolution ( R) was calculated using the following equation J / A = Cs/R. Solubility (Cs) Determination-The equilibrium solubility was determined by introducing an excess quantity of radioactive cholesterol monohydrate into the bile.6*7.”The flask was simultaneously flushed with nitrogen and shaken by a wristaction shaker in a 37°C water bath. Aliquots were taken daily and, after filtering through a 0.22-pm filter (also a t 37°C in a controlled atmosphere), the radioactivity was determined by liquid scintillation counting. The cholesterol concentration of the filtrate was determined when the radioactivity reached a constant level. Dissolution R a t e Determination (J/A)-Compressed pellets of radioactive cholesterol were secured to the bottom of water-jacketed chambers containing the bile ~ p e c i m e n . ~ .The ~’ temperature was maintained a t 37°C and the bile was continuously stirred at the rate of 150 rpm. The amount of cholesterol dissolved in the bile was analyzed a t various periods of time. In Vivo Studies-Fasting beagle dogs were anesthetized with sodium pentobarbital (30 mg/kg iv). The abdominal cavity was opened using a sterile technique and the gallbladder was exposed. The bile was aspirated by needle puncture from the gallbladder and frozen. A weighed compressed cholesterol monohydrate pellet (diameter of -0.5 cm, weighing -200 mg) was inserted into the gallbladder through a small incision. A
Journal of Pharmaceutical Sciences / 621 Vol. 74, No. 6, June 1985
polyethylene infusion cannula was inserted through the same incision and secured into place with a purse string suture of 5 0 silk. The cannula was then tunneled subcutaneously to an area between the shoulders where it was exteriorized through a 2 in. (5.0 cm) length of Silastic tubing secured to a Velcro disk [0.5 in. (1.2 cm) diameter]. After awakening, the dogs were placed into harnesses and the polyethylene cannulas were connected to swivel apparatuses. A constant infusion of either normal saline or lincomycin solution (9.03 pmol/kg/h in normal saline) was started at the rate of 0.84 mL/h. We had determined in preliminary tests that this dosage was near the maximum tolerated dosage of lincomycin for dogs. Patency of the cannulae were checked daily by the aspiration of the bile out and back into the gallbladder. Dogs were fed only once daily in order to minimize gallbladder emptying so that the gallbladder concentration of lincomycin could he kept a t the highest possible level throughout the day. After 13 d the dogs were fasted overnight and then sacrificed by barbiturate overdose and intravenous KCl injection. The bile was aspirated from the gallbladder and frozen. The remains of the cholesterol pellets, which were still intact, were recovered from the gallbladders and washed thoroughly with water. After drying a t room temperature for 72 h, changes in weight of the pellets were calculated. The bile which was aspirated from the gallbladder during the initial surgical preparation, as well as the bile which was collected from the gallbladder at the termination of the infusion, was compared in the in vitro dissolution tests as described above. before the insertion Analysis of Bile Composition-Just of the gallbladder cannulas during the initial surgical preparation of the animals, gallbladder bile was collected by needle aspiration from the, thus far, naive dogs. This bile was stored frozen for the later analysis of bile salt and cholesterol content. A t the termination of the study, fasting gallbladder bile was again aspirated from these same animals and the cholesterol and bile salt concentration of this bile was determined and compared with the earlier sample from the same animal. The cholesterol concentration of the gallbladder bile was determined using the cholesterol esterase-oxidase peroxidasephenol 4-aminophenazone system,':' which is supplied commercially by Hiodynamics, Inc. The bile acid concentration was measured enzymatically with 3a-hydroxysteroid dehydrogenase which was obtained from Accurate Chem. and Scientific, Inc.I4 The lincomycin concentration of the gallbladder bile was determined using a modification of a method which was reported by Tsuji."
Results In Vitro-Table I shows that the addition of 45 mmol of lincomycin to the bile significantly lowered the resistance ( R ) by 34% while increasing the rate of dissolution ( J I A )by 48%. As expected, the equilibrium solubility (Cs) was not changed by lincomycin. In Vivo-Table I1 shows the effects of lincomycin on the dissolution rate ( J I A ) ,cholesterol solubility ( C s ) , and the interfacial resistance ( R ) of the gallbladder bile of dogs which had received a constant infusion of lincomycin for 13 d a t 0 or 9.03 fimol/kg/h. Gallbladder bile which was aspirated from these animals at the termination of the infusion was compared to the bile which had been aspirated from the same dogs during the initial surgical preparation. The J / A , as determined by in vitro procedures, decreased significantly in control dogs due to a significant increase in resistance ( R ) over this 13-d period. However, dogs which received lincomycin did not show any alterations in J j A or R when postinfusion bile was compared with their own bile, which had been removed from these animals before the infusion was begun. In spite of these relative differences between control and lincomycin-treated dogs in the in vitro determined dissolution rate parameters, the infusion
622 / Journal of Pharmaceutical Sciences Vol. 74, No. 6, June 1985
Table I-Effects of Lincomycin on the Dissolution of Cholesterol Monohydrate Pellets In Vitro
Sample
No. 1 2 3 4 5 6
7 8 9 10 '1 Mean kSDM a
JIA x lo4
R x 10-3
Cs,mg/mL
Control Lincomvcin Control Lincomvcin Control Lincomvcin
0.77 1.23 6.5 0.77 0.88 7.2 0.77 1.10 7.5 0.77 0.99 7.9 1.01 0.77 5.7 0.64 0.98 5.1 0.65 0.46 3.0 0.74 0.52 3.6 0.82 0.46 3.4 1.17 0.60 6.0 0.99 0.64 6.3 __ 0.96" 5.64 0.65 k0.51 k0.04 k0.05
5.9 6.4 6.1 5.1 4.8 4.8 2.8 2.9 2.9 5.6 __ 6.6 4.89 k0.44
84.4 93.9 97.5 102.1 74.4 79.8 65.0 68.3 72.9 99.0 98.0 85.03 24.1
47.9 73.1 55.0 51.7 47.5 48.8 42.3 39.2 34.8 47.O 67.0 50.4" +3.4
Significant difference, p 5 0.01 by the paired t test.
of lincomycin did not affect the rate of dissolution of compressed cholesterol monohydrate pellets which had been placed into the gallbladders of these animals (Table 11). Bile Composition-The bile acid and cholesterol concentrations of the aspirated bile were determined in a representative number of these dogs as an indicator of changes in bile composition. Table I11 shows that neither the bile acid nor the cholesterol concentration of bile taken from lincomycin-treated dogs was different from that of saline-infused control animals. However, all dogs, both those receiving the saline infusion and those treated with lincomycin, had increased biliary concentrations of cholesterol when compared with their own bile (which had been aspirated from the gallbladders during the initial surgical preparation). Presumably, this increase in biliary cholesterol was due to the partial dissolution of the implanted cholesterol pellet. Table 111 also shows that the constant infusion of lincomycin at the rate of 9.03 pmol/kg/h resulted in a biliary concentration of lincomycin of 41.8 mM, approximating the 45 mM concentration of lincomycin which was used in the in vitro tests.
Discussion These experiments were designed to test the concept that decreasing the interfacial resistance of cholesterol gallstones will accelerate gallstone dissolution in cholelithiasis patients who are undergoing chenodeoxycholic acid dissolution therapy. Dogs were chosen for this study because their bile is normally undersaturated with respect to cholesterol, as are subjects receiving bile acid therapy. The test compound was infused directly in the gallbladder in order to attain the high concentration of lincomycin which was determined by in vitro testing to be necessary for increasing J / A . Direct delivery to the gallbladder also ensured that variations in the absorption, metabolism, or excretion profile of the test compound were minimized. The choice of lincomycin as the test compound was made after examining a number of other possible candidates. Toxicity to the gallbladder or animal at the concentrations necessary to influence dissolution were major factors in the selection of lincomycin, which although less potent, was also less toxic than many of the compounds previously examined by Higuchi et a1.6~~ The addition of lincomycin in vitro to gallbladder bile obtained from fasting, naive dogs resulted in an increase in the rate of cholesterol dissolution ( J / A ) of cholesterol pellets incubating in this bile (Table I). This was a consequence of a decrease in the interfacial resistance ( R) resulting from the
Table Il-Effects
of Lincomycin Infusion on Gallstone Dissoiution Rate: In Vivo and In Vitro
Dose of Lincomycin, Fmol/E;g/h
Dog No.
1 2 3 4 5 6
0 0 0 0 0 0
Mean +SDM 7 8 9 10 11 12 13 14 15
9.03 9.03 9.03 9.03 9.a3 91 .13 9.03 9.a3 9.a3
Mean +SDM
J/A x
Cholesterol Solubility,
lo4
R x 10-3
mg/mL Before After Infusion Infusion
Before Infusion
~ f Infusion
Dissolution of Artificial t Gallstone, ~ ~ %
Before Infusion
After Infusion
1.12 1.18 0.87 0.96 1.01 1.51
1.03 0.55 0.59 0.78 0.76 1.oo
8.55 7.64 5.57 6.46 6.13 6.75
8.03 6.41 8.00 6.83 7.66 6.44
76.6 64.8 64.4 67.2 60.8 40.7
77.7 115.7 135.3 87.7 101.4 64.6
58.6 54.9 48.0 40.4 49.4 51.2
1.10 20.09
0.79' kO.08
6.83 2.44
7.23 20.31
62.4 k4.9
97.1 k10.6
50.4 k2.5
0.99 0.99 0.86 1.09 1.06 1.06 1.26 0.71 1.26
0.99 1.17 1.15 0.81 1.18 0.64 1.26 0.60 1.29
8.85 9.35 9.08 8.27 6.50 5.61 6.75 6.86 8.58
9.82 8.68 9.04 10.65 6.99 5.66 4.98 4.74 11.56
89.5 94.9 105.5 75.9 61.1 52.9 53.6 96.6 68.1
99.0 74.0 78.6 130.1 59.5 88.3 38.6 79.0 89.6
47.2 33.2 56.9 36.8 55.5 61.1 42.2 30.4 43.0
1.03 20.06
1.01 kO.09
7.76 k0.45
8.01 20.84
77.6 k6.6
81.9 45.1 k8.5 k3.5 a Significant difference,p 5 0.05, from before-infusionsample; paired t test. Significant difference, p 5 0.01, from before-infusion sample; paired t test.
Table Ill-Effects
Lincomycin Infusion Rate, mol/kg/h
0
of Lincomvcin llnfusion on Gallbladder Bile Composition'
Bile Acid Concentration, mmol/L Before Infusion 320.5 k15.9 (n = 5)
After Infusion 289.3 224.4 (n =5)
Cholesterol Concentration, mmol/L (or mg/mL)
Lincomycin Concentration, mmol/L
Before Infusion
After Infusion
1.09 (or 0.42 mg/mL) 20.17 (or 0.07 rng/mL) (n = 3)
1.55 (or 0.60 rng/mL) k0.27 (or 0.10 mg/mL) (n = 5)
0
1.03 (or 0.40 mg/mL) 1.60 (or 0.61 mg/mL) 41.8 326.2 30.4 20.27 (or 0.10 mg/mL) k7.4 221.2 238.0 kO.10 (or 0.04 mg/mL) tn = 6) (n = 6) (n = 4) (n = 4) (n = 6) a Values are presented as the mean 2 SDM; n = indicates the number of samples assayed; cholesterolconcentrationsare presented in both mmol/ L as well as mg/mL of bile for ease in comparison to Tables I and II. 9.03
addition of lincomycin. However, the 13-d lincomycin infusion directly into the gallbladder of conscious, unrestrained dogs did not alter the dissolution parameters ( J / A or R ) as determined by in vitro procedures when compared with bile which had been aspirated from these same animals during the initial surgical preparation. On the other hand, J / A in control dogs was decreased after 13 d of saline infusion. This decrease in the in vitro determined cholesterol iIlissolution rate of bile from control dogs may have been related to factors such as dilution of the gallbladder bile with the perfusate (normal saline) or an increase in the cholesterol concentration of the gallbladder bile due to the partial dissolution of the implanted cholesterol pellet. Also, surgical or postsurgical stress may have caused transient alterations in bile composition. However, although lincomycintreated animals encountered identical experimental conditons, which would have presumably resulted in the same negative effects on these in vitro determined parameters, the J / A in these animals were not different after 13 d of lincomycin
treatment when compared with their own control bile. These observations may indicate that factors which resulted in the decreased J / A in control animals might have been negated by lincomycin treatment. To the extent that these in vivo experiments represent a valid test system, it then appears that interfacial resistance was not the only rate-limiting factor in the dissolution of the implanted cholesterol pellet. This observation suggests that another factor, such as mixing, may also be rate limiting in a system in which bile is undersaturated with respect to cholesterol. This outcome is not unexpected, since gallbladder stagnation has been postulated as a possible cause of stone formation from supersaturated bile. Furthermore, the in vitro test system utilized rapid mixing whereas in the in vivo environment agitation is expected to be minimal. Poor mixing will result in a cholesterol-saturated layer of bile surrounding the pellet which could effectively halt dissolution. This theory may be consistent with the recent findings of Bugliosi et a1.I6 who
Journal of Pharmaceutical Sciences / 623 Vol. 74, No. 6, June 1985
showed in an in vivo dog study that gallstone dissolution which was accelerated by methyl tert-butyl ether was further increased by mixing with the aid of an infusion aspiration pump. Furthermore, the low solubility of methyl tert-butyl ether would presumably make mixing even more necessary in order to assure contact between the added solvent and the surface of the gallstone. On the other hand, the importance of mixing may perhaps be minimized by the use of compounds which are more effective reducers of interfacial resistance. However, in preliminary tests in our laboratory, we have found such compounds to be much too toxic. Similar problems were also encountered with higher doses of lincomycin. Our conclusion must be tempered by the knowledge that because of the experimental design we were unable to control several variables. First of all, although the bile taken from the dogs after 13 d of infusion showed high concentrations of lincomycin and favorable in vitro dissolution parameters compared to control animals, we cannot ascertain the length of time each day that these conditions existed. Accordingly, there may have been transient alterations in bile composition due to the surgical or postsurgical stress or other factors which were beyond our control. In summary, these data suggest that interfacial resistance may not be the critical rate-limiting factor in cholesterol gallstone dissolution in vivo. Mixing may be a n equally important rate-limiting factor. However, it must be noted that cholesterol dissolution in bile was accelerated by only 50% under the ideal conditions which were provided in the in vitro studies. A compound which could decrease interfacial resistance much more effectively than lincomycin might have minimized the inferred importance of mixing.
624 / Journal of Pharmaceutical Sciences Vol. 74, No. 6, June 1985
References and Notes 1. Thistle, J. L.; Schoenfield, L. J. J. Lab. Clin. Med. 1969, 74, 1020. 2. Danzinger, R. G.; Hofmann, A. F.; Schoenfield, L. J.; Thistle, J. L. New Engl. J. Med 1972,286, 1. 3. Nakagawa, S.; Makino, I.; Ishizaki, T.; Dohi, I. Lancet 1977, 2, 367. 4. Makiro, I.; Shinzaki, K.; Yoshiro, K.; Nakagawa, S. Jpn. J. Gastroenterol. 1975. 72. 690. 5. Higuchi, W.; Sjuib, F.; Mufson, D.; Simonelli, A.; Hofmann, A. J. Pharrn. Sci. 1973,62, 942. 6. Higuchi, W.; Prakongpan, S.; Surpuriya, V.; Young, F. Science 1972. 178. 633. 7. Higuchc W.; Prakongpan, S.; Young, F. J. Pharm. Sci. 1973,62, 945. 8. Higuchi, W.; Prakongpan, S.; Young, F. J. Pharm. Sci. 1973,62, * nnn
ILU I .
9. Prakongpan, S.; Higuchi, W.; Kwan, K.; Molokhia, A. J. Pharm. Sci. 1977., 6.5. - - , 686. --10. KwanrK.; Higuchi, W.; Molokhia, A,; Hofmann, A. J. Pharm. Sci. 1977,66, 1094. 11. Molokhia, A.; Hofmann, A.; Higuchi, W.; Tuchinda, W.; Feld, K.; Prakongpan, S.; Danzinger, R. J . Pharm. Sci. 1977,66, 1101. 12. Kwan, K.; Higuchi, W.; Molokhia, A.; Hofmann, A. J. Pharrn. Sci. 1977.66. , - - ,1105. 13. Roda, A.; Festi, D.; Sama, C.; Mazzella, G.; Aldini, R.; Roda, E.; Barbara, L. Clin. Chin. Acta 1975,64, 337. 14. Domingo, N.; Annie, J.; Hauton, J. Clin. Chin. Acta 1972,37,399. 15. Tsuji, K . “GLC and HPLC Determination of Therapeutic Agents,” Dart 11: Marcel Dekker: New York. 1978 D 253. 16. Bugliosi, T. F.; Allen, M. J.; Borody, T.’).; Stryker, S.; LaRusso, N. F.; Thistle, J. L. Gastroenterology 1984, 86, 1350. 17. Patel, D. C.; Higuchi, W. I. J. Colloid Interface Sci. 1980, 74, 211. ~~
Acknowledgments The authors would like to thank R. A. Conradi for invaluable technical assistance.
AND WILLIAM
I. HIGUCHI”
Received June 20, 1984, from the * Diabetes and Gastrointestinal Diseases Research and *Pharmacy Research, The Upjohn Company, Kalamazoo, Present Accepted for publication March 1, 1985. MI 49001 and the ’College of Pharmacy, University of Michigan, Ann Arbor, MI 48109. address: 6University of Utah, College of Pharmacy; Salt Lake City, UT 84112. Abstract 0 These studies were undertaken to test the hypothesis that interfacial resistance may be an important rate-limiting factor in cholesterol gallstone dissoluton. The addition of lincomycin hydrochloride to the gallbladder bile of dogs in an in vitro bath system resulted in an acceleration in the rate of dissolution of a compressed cholesterol
monohydrate pellet incubating in the bile. However,the constant infusion of lincomycin for 13 d directly into the gallbladders of conscious, unrestrained dogs, which resulted in biliary lincomycin concentrations comparable to that of the in vitro tests, did not alter the dissolution rate of a compressed cholesterol monohydrate pellet which had been surgically placed into the gallbladder. We therefore conclude that the interfacial resistance between the cholesterol monohydrate pellet and the bile may be reduced by the addition of lincomycin to the gallbladder bile which, in the in vitro environment,results in an acceleration in the rate of dissolution of compressed cholesterol pellets. However, the ineffectiveness of lincomycin in accelerating the dissolution of cholesterol pellets in vivo suggests that interfacial resistance is not the only rate-limiting factor in gallstone dissolution. Other factor!;, such as mixing, may also be critical. Cholesterol is a normal constituent of vertebrate bile. Although it is virtually insoluble in water, cholesterol is dispersed in bile by the formation of bile acid-lecithin micelles. Hence, the degree of cholesterol saturation in gallbladder bile is a critical factor in the formation or dissolution of cholesterol gallstones. Gallstone dissolution studies have shown that the administration of the bile acids chenodeoxycholic’.’ or ursodeoxycholic acid3s4 will result in some degree of cholesterol stone dissolution. These bile acids alter the composition of bile so that the bile becomes undersaturated with respect to cholesterol and, thus, ultimately leads to the resolubilization of the gallstone. However, the dissolution of the stones progresses slowly; therefore, therapy must be continued for months or even years. The work of Higuchi indicates that the dissolution of cholesterol gallstones proceeds at a much slower rate than predicted from diffusion rate models5 which could indicate that the dissolution is controlled by interfacial resistance?-” Higuchi et al.5 defined interfacial resistance barriers as “barriers to dissolution which arise from intrinsically slow crystal-surface solution equilibria or from ahorbed or deposited substances on the stones surfaces.” Accondingly, he and his co-workers hypothesized that the addition of agents to the bile which reduce the interfacial barrier might “have therapeutic value in decreasing the time required for total gallstone dissolution in patients receiving bile acid therapy.” Utilizing in vitro models, these investigators have demonstrated that benzalkonium chloride, cetylpyridinium and, more generally, primary, secondary, and tertiary amines, quaternary ammonium salts, as well as cationic surfactants, but not anionic surfactants,” are effective gallstone dissolution rate accelerators. The purpose of this study was to test in the in vivo model the concept that the rate of dissolution of cholesterol gallstones in cholesterol-undersaturated bile may be accelerated by the addition of compounds to the bile which reduce the interfacial
0022-3549/85/0600-062l$O 1.OO/O 0 1985, American
Pharmaceutical Association
resistance. Moreover, this study represents one of the first known attempts to test this concept in an animal model which mimics the conditions encountered in humans with cholelithiasis undergoing bile acid dissolution therapy. Lincomycin hydrochloride was chosen because preliminary tests showed it to be better tolerated in animals than many of the other compounds which had been shown to accelerate the rate of dissolution of cholesterol gallstones in the in vitro test system.
Experimental Section In V i t r o Studies-Mongrel dogs were fasted overnight and then anesthetized with sodium pentobarbital, 30 mg/kg iv. The abdominal cavity was opened and bile was aspirated from the gallbladder into a sterile syringe. Each dog served as a bile donor only once. The sample of bile was mixed, divided into two sterile containers, and the appropriate quantity of lincomycin hydrochloride (0 or 45 mmol/L) was added. The samples were frozen for storage. The quantity of bile from each dog was sufficient for duplicates of both control and lincomycin-treated samples. The dosage of 45 mmol/L was determined in a series of pilot studies in dogs to approximate the maximum concentration of lincomycin tolerated when administered into the gallbladders of these animals. Using these bile samples, the rate of dissolution per unit area ( J / A ) of a compressed cholesterol monohydrate pellet and the equilibrium solubility (Cs) of cholesterol was experimentally determined by the methods of Higuchi et al. and Molokhia et al.6.7311The total resistance to dissolution ( R) was calculated using the following equation J / A = Cs/R. Solubility (Cs) Determination-The equilibrium solubility was determined by introducing an excess quantity of radioactive cholesterol monohydrate into the bile.6*7.”The flask was simultaneously flushed with nitrogen and shaken by a wristaction shaker in a 37°C water bath. Aliquots were taken daily and, after filtering through a 0.22-pm filter (also a t 37°C in a controlled atmosphere), the radioactivity was determined by liquid scintillation counting. The cholesterol concentration of the filtrate was determined when the radioactivity reached a constant level. Dissolution R a t e Determination (J/A)-Compressed pellets of radioactive cholesterol were secured to the bottom of water-jacketed chambers containing the bile ~ p e c i m e n . ~ .The ~’ temperature was maintained a t 37°C and the bile was continuously stirred at the rate of 150 rpm. The amount of cholesterol dissolved in the bile was analyzed a t various periods of time. In Vivo Studies-Fasting beagle dogs were anesthetized with sodium pentobarbital (30 mg/kg iv). The abdominal cavity was opened using a sterile technique and the gallbladder was exposed. The bile was aspirated by needle puncture from the gallbladder and frozen. A weighed compressed cholesterol monohydrate pellet (diameter of -0.5 cm, weighing -200 mg) was inserted into the gallbladder through a small incision. A
Journal of Pharmaceutical Sciences / 621 Vol. 74, No. 6, June 1985
polyethylene infusion cannula was inserted through the same incision and secured into place with a purse string suture of 5 0 silk. The cannula was then tunneled subcutaneously to an area between the shoulders where it was exteriorized through a 2 in. (5.0 cm) length of Silastic tubing secured to a Velcro disk [0.5 in. (1.2 cm) diameter]. After awakening, the dogs were placed into harnesses and the polyethylene cannulas were connected to swivel apparatuses. A constant infusion of either normal saline or lincomycin solution (9.03 pmol/kg/h in normal saline) was started at the rate of 0.84 mL/h. We had determined in preliminary tests that this dosage was near the maximum tolerated dosage of lincomycin for dogs. Patency of the cannulae were checked daily by the aspiration of the bile out and back into the gallbladder. Dogs were fed only once daily in order to minimize gallbladder emptying so that the gallbladder concentration of lincomycin could he kept a t the highest possible level throughout the day. After 13 d the dogs were fasted overnight and then sacrificed by barbiturate overdose and intravenous KCl injection. The bile was aspirated from the gallbladder and frozen. The remains of the cholesterol pellets, which were still intact, were recovered from the gallbladders and washed thoroughly with water. After drying a t room temperature for 72 h, changes in weight of the pellets were calculated. The bile which was aspirated from the gallbladder during the initial surgical preparation, as well as the bile which was collected from the gallbladder at the termination of the infusion, was compared in the in vitro dissolution tests as described above. before the insertion Analysis of Bile Composition-Just of the gallbladder cannulas during the initial surgical preparation of the animals, gallbladder bile was collected by needle aspiration from the, thus far, naive dogs. This bile was stored frozen for the later analysis of bile salt and cholesterol content. A t the termination of the study, fasting gallbladder bile was again aspirated from these same animals and the cholesterol and bile salt concentration of this bile was determined and compared with the earlier sample from the same animal. The cholesterol concentration of the gallbladder bile was determined using the cholesterol esterase-oxidase peroxidasephenol 4-aminophenazone system,':' which is supplied commercially by Hiodynamics, Inc. The bile acid concentration was measured enzymatically with 3a-hydroxysteroid dehydrogenase which was obtained from Accurate Chem. and Scientific, Inc.I4 The lincomycin concentration of the gallbladder bile was determined using a modification of a method which was reported by Tsuji."
Results In Vitro-Table I shows that the addition of 45 mmol of lincomycin to the bile significantly lowered the resistance ( R ) by 34% while increasing the rate of dissolution ( J I A )by 48%. As expected, the equilibrium solubility (Cs) was not changed by lincomycin. In Vivo-Table I1 shows the effects of lincomycin on the dissolution rate ( J I A ) ,cholesterol solubility ( C s ) , and the interfacial resistance ( R ) of the gallbladder bile of dogs which had received a constant infusion of lincomycin for 13 d a t 0 or 9.03 fimol/kg/h. Gallbladder bile which was aspirated from these animals at the termination of the infusion was compared to the bile which had been aspirated from the same dogs during the initial surgical preparation. The J / A , as determined by in vitro procedures, decreased significantly in control dogs due to a significant increase in resistance ( R ) over this 13-d period. However, dogs which received lincomycin did not show any alterations in J j A or R when postinfusion bile was compared with their own bile, which had been removed from these animals before the infusion was begun. In spite of these relative differences between control and lincomycin-treated dogs in the in vitro determined dissolution rate parameters, the infusion
622 / Journal of Pharmaceutical Sciences Vol. 74, No. 6, June 1985
Table I-Effects of Lincomycin on the Dissolution of Cholesterol Monohydrate Pellets In Vitro
Sample
No. 1 2 3 4 5 6
7 8 9 10 '1 Mean kSDM a
JIA x lo4
R x 10-3
Cs,mg/mL
Control Lincomvcin Control Lincomvcin Control Lincomvcin
0.77 1.23 6.5 0.77 0.88 7.2 0.77 1.10 7.5 0.77 0.99 7.9 1.01 0.77 5.7 0.64 0.98 5.1 0.65 0.46 3.0 0.74 0.52 3.6 0.82 0.46 3.4 1.17 0.60 6.0 0.99 0.64 6.3 __ 0.96" 5.64 0.65 k0.51 k0.04 k0.05
5.9 6.4 6.1 5.1 4.8 4.8 2.8 2.9 2.9 5.6 __ 6.6 4.89 k0.44
84.4 93.9 97.5 102.1 74.4 79.8 65.0 68.3 72.9 99.0 98.0 85.03 24.1
47.9 73.1 55.0 51.7 47.5 48.8 42.3 39.2 34.8 47.O 67.0 50.4" +3.4
Significant difference, p 5 0.01 by the paired t test.
of lincomycin did not affect the rate of dissolution of compressed cholesterol monohydrate pellets which had been placed into the gallbladders of these animals (Table 11). Bile Composition-The bile acid and cholesterol concentrations of the aspirated bile were determined in a representative number of these dogs as an indicator of changes in bile composition. Table I11 shows that neither the bile acid nor the cholesterol concentration of bile taken from lincomycin-treated dogs was different from that of saline-infused control animals. However, all dogs, both those receiving the saline infusion and those treated with lincomycin, had increased biliary concentrations of cholesterol when compared with their own bile (which had been aspirated from the gallbladders during the initial surgical preparation). Presumably, this increase in biliary cholesterol was due to the partial dissolution of the implanted cholesterol pellet. Table 111 also shows that the constant infusion of lincomycin at the rate of 9.03 pmol/kg/h resulted in a biliary concentration of lincomycin of 41.8 mM, approximating the 45 mM concentration of lincomycin which was used in the in vitro tests.
Discussion These experiments were designed to test the concept that decreasing the interfacial resistance of cholesterol gallstones will accelerate gallstone dissolution in cholelithiasis patients who are undergoing chenodeoxycholic acid dissolution therapy. Dogs were chosen for this study because their bile is normally undersaturated with respect to cholesterol, as are subjects receiving bile acid therapy. The test compound was infused directly in the gallbladder in order to attain the high concentration of lincomycin which was determined by in vitro testing to be necessary for increasing J / A . Direct delivery to the gallbladder also ensured that variations in the absorption, metabolism, or excretion profile of the test compound were minimized. The choice of lincomycin as the test compound was made after examining a number of other possible candidates. Toxicity to the gallbladder or animal at the concentrations necessary to influence dissolution were major factors in the selection of lincomycin, which although less potent, was also less toxic than many of the compounds previously examined by Higuchi et a1.6~~ The addition of lincomycin in vitro to gallbladder bile obtained from fasting, naive dogs resulted in an increase in the rate of cholesterol dissolution ( J / A ) of cholesterol pellets incubating in this bile (Table I). This was a consequence of a decrease in the interfacial resistance ( R) resulting from the
Table Il-Effects
of Lincomycin Infusion on Gallstone Dissoiution Rate: In Vivo and In Vitro
Dose of Lincomycin, Fmol/E;g/h
Dog No.
1 2 3 4 5 6
0 0 0 0 0 0
Mean +SDM 7 8 9 10 11 12 13 14 15
9.03 9.03 9.03 9.03 9.a3 91 .13 9.03 9.a3 9.a3
Mean +SDM
J/A x
Cholesterol Solubility,
lo4
R x 10-3
mg/mL Before After Infusion Infusion
Before Infusion
~ f Infusion
Dissolution of Artificial t Gallstone, ~ ~ %
Before Infusion
After Infusion
1.12 1.18 0.87 0.96 1.01 1.51
1.03 0.55 0.59 0.78 0.76 1.oo
8.55 7.64 5.57 6.46 6.13 6.75
8.03 6.41 8.00 6.83 7.66 6.44
76.6 64.8 64.4 67.2 60.8 40.7
77.7 115.7 135.3 87.7 101.4 64.6
58.6 54.9 48.0 40.4 49.4 51.2
1.10 20.09
0.79' kO.08
6.83 2.44
7.23 20.31
62.4 k4.9
97.1 k10.6
50.4 k2.5
0.99 0.99 0.86 1.09 1.06 1.06 1.26 0.71 1.26
0.99 1.17 1.15 0.81 1.18 0.64 1.26 0.60 1.29
8.85 9.35 9.08 8.27 6.50 5.61 6.75 6.86 8.58
9.82 8.68 9.04 10.65 6.99 5.66 4.98 4.74 11.56
89.5 94.9 105.5 75.9 61.1 52.9 53.6 96.6 68.1
99.0 74.0 78.6 130.1 59.5 88.3 38.6 79.0 89.6
47.2 33.2 56.9 36.8 55.5 61.1 42.2 30.4 43.0
1.03 20.06
1.01 kO.09
7.76 k0.45
8.01 20.84
77.6 k6.6
81.9 45.1 k8.5 k3.5 a Significant difference,p 5 0.05, from before-infusionsample; paired t test. Significant difference, p 5 0.01, from before-infusion sample; paired t test.
Table Ill-Effects
Lincomycin Infusion Rate, mol/kg/h
0
of Lincomvcin llnfusion on Gallbladder Bile Composition'
Bile Acid Concentration, mmol/L Before Infusion 320.5 k15.9 (n = 5)
After Infusion 289.3 224.4 (n =5)
Cholesterol Concentration, mmol/L (or mg/mL)
Lincomycin Concentration, mmol/L
Before Infusion
After Infusion
1.09 (or 0.42 mg/mL) 20.17 (or 0.07 rng/mL) (n = 3)
1.55 (or 0.60 rng/mL) k0.27 (or 0.10 mg/mL) (n = 5)
0
1.03 (or 0.40 mg/mL) 1.60 (or 0.61 mg/mL) 41.8 326.2 30.4 20.27 (or 0.10 mg/mL) k7.4 221.2 238.0 kO.10 (or 0.04 mg/mL) tn = 6) (n = 6) (n = 4) (n = 4) (n = 6) a Values are presented as the mean 2 SDM; n = indicates the number of samples assayed; cholesterolconcentrationsare presented in both mmol/ L as well as mg/mL of bile for ease in comparison to Tables I and II. 9.03
addition of lincomycin. However, the 13-d lincomycin infusion directly into the gallbladder of conscious, unrestrained dogs did not alter the dissolution parameters ( J / A or R ) as determined by in vitro procedures when compared with bile which had been aspirated from these same animals during the initial surgical preparation. On the other hand, J / A in control dogs was decreased after 13 d of saline infusion. This decrease in the in vitro determined cholesterol iIlissolution rate of bile from control dogs may have been related to factors such as dilution of the gallbladder bile with the perfusate (normal saline) or an increase in the cholesterol concentration of the gallbladder bile due to the partial dissolution of the implanted cholesterol pellet. Also, surgical or postsurgical stress may have caused transient alterations in bile composition. However, although lincomycintreated animals encountered identical experimental conditons, which would have presumably resulted in the same negative effects on these in vitro determined parameters, the J / A in these animals were not different after 13 d of lincomycin
treatment when compared with their own control bile. These observations may indicate that factors which resulted in the decreased J / A in control animals might have been negated by lincomycin treatment. To the extent that these in vivo experiments represent a valid test system, it then appears that interfacial resistance was not the only rate-limiting factor in the dissolution of the implanted cholesterol pellet. This observation suggests that another factor, such as mixing, may also be rate limiting in a system in which bile is undersaturated with respect to cholesterol. This outcome is not unexpected, since gallbladder stagnation has been postulated as a possible cause of stone formation from supersaturated bile. Furthermore, the in vitro test system utilized rapid mixing whereas in the in vivo environment agitation is expected to be minimal. Poor mixing will result in a cholesterol-saturated layer of bile surrounding the pellet which could effectively halt dissolution. This theory may be consistent with the recent findings of Bugliosi et a1.I6 who
Journal of Pharmaceutical Sciences / 623 Vol. 74, No. 6, June 1985
showed in an in vivo dog study that gallstone dissolution which was accelerated by methyl tert-butyl ether was further increased by mixing with the aid of an infusion aspiration pump. Furthermore, the low solubility of methyl tert-butyl ether would presumably make mixing even more necessary in order to assure contact between the added solvent and the surface of the gallstone. On the other hand, the importance of mixing may perhaps be minimized by the use of compounds which are more effective reducers of interfacial resistance. However, in preliminary tests in our laboratory, we have found such compounds to be much too toxic. Similar problems were also encountered with higher doses of lincomycin. Our conclusion must be tempered by the knowledge that because of the experimental design we were unable to control several variables. First of all, although the bile taken from the dogs after 13 d of infusion showed high concentrations of lincomycin and favorable in vitro dissolution parameters compared to control animals, we cannot ascertain the length of time each day that these conditions existed. Accordingly, there may have been transient alterations in bile composition due to the surgical or postsurgical stress or other factors which were beyond our control. In summary, these data suggest that interfacial resistance may not be the critical rate-limiting factor in cholesterol gallstone dissolution in vivo. Mixing may be a n equally important rate-limiting factor. However, it must be noted that cholesterol dissolution in bile was accelerated by only 50% under the ideal conditions which were provided in the in vitro studies. A compound which could decrease interfacial resistance much more effectively than lincomycin might have minimized the inferred importance of mixing.
624 / Journal of Pharmaceutical Sciences Vol. 74, No. 6, June 1985
References and Notes 1. Thistle, J. L.; Schoenfield, L. J. J. Lab. Clin. Med. 1969, 74, 1020. 2. Danzinger, R. G.; Hofmann, A. F.; Schoenfield, L. J.; Thistle, J. L. New Engl. J. Med 1972,286, 1. 3. Nakagawa, S.; Makino, I.; Ishizaki, T.; Dohi, I. Lancet 1977, 2, 367. 4. Makiro, I.; Shinzaki, K.; Yoshiro, K.; Nakagawa, S. Jpn. J. Gastroenterol. 1975. 72. 690. 5. Higuchi, W.; Sjuib, F.; Mufson, D.; Simonelli, A.; Hofmann, A. J. Pharrn. Sci. 1973,62, 942. 6. Higuchi, W.; Prakongpan, S.; Surpuriya, V.; Young, F. Science 1972. 178. 633. 7. Higuchc W.; Prakongpan, S.; Young, F. J. Pharm. Sci. 1973,62, 945. 8. Higuchi, W.; Prakongpan, S.; Young, F. J. Pharm. Sci. 1973,62, * nnn
ILU I .
9. Prakongpan, S.; Higuchi, W.; Kwan, K.; Molokhia, A. J. Pharm. Sci. 1977., 6.5. - - , 686. --10. KwanrK.; Higuchi, W.; Molokhia, A,; Hofmann, A. J. Pharm. Sci. 1977,66, 1094. 11. Molokhia, A.; Hofmann, A.; Higuchi, W.; Tuchinda, W.; Feld, K.; Prakongpan, S.; Danzinger, R. J . Pharm. Sci. 1977,66, 1101. 12. Kwan, K.; Higuchi, W.; Molokhia, A.; Hofmann, A. J. Pharrn. Sci. 1977.66. , - - ,1105. 13. Roda, A.; Festi, D.; Sama, C.; Mazzella, G.; Aldini, R.; Roda, E.; Barbara, L. Clin. Chin. Acta 1975,64, 337. 14. Domingo, N.; Annie, J.; Hauton, J. Clin. Chin. Acta 1972,37,399. 15. Tsuji, K . “GLC and HPLC Determination of Therapeutic Agents,” Dart 11: Marcel Dekker: New York. 1978 D 253. 16. Bugliosi, T. F.; Allen, M. J.; Borody, T.’).; Stryker, S.; LaRusso, N. F.; Thistle, J. L. Gastroenterology 1984, 86, 1350. 17. Patel, D. C.; Higuchi, W. I. J. Colloid Interface Sci. 1980, 74, 211. ~~
Acknowledgments The authors would like to thank R. A. Conradi for invaluable technical assistance.