Downloaded from
`
`http://aac.asm.org/
`
` on March 4, 2014 by guest
`
`ANTiMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 1994, p. 319-325
`0066-4804/94/$04.00+0
`Copyright © 1994, American Society for Microbiology
`
`Vol. 38, No.2
`
`In Vitro Analysis of the Interaction between Sucralfate
`and Ketoconazole
`JAMES D. HOESCHELE,1 ANINDYA K. ROY, 1 VINCENT L. PECORARO/ AND PEGGY L. CARVER2*
`The Department of Chemistry1 and the College of Pharmacy, 2 The University of Michigan,
`Ann Arbor, Michigan 48109-1065
`
`Received 9 November 1993/Retumed for modification 21 February 1993/Accepted 30 November 1993
`
`In healthy volunteers, the bioavailability of ketoconazole is significantly decreased during simultaneous
`administration with sucralfate. In an elfort to address this problem, we examined the interaction between
`sucralfate and ketoconazole in aqueous solutions and in simulated gastric fluid (SGF) at various initial pHs
`(1, 2, 3, and 6) in the presence or absence of glutamic acid hydrochloride (GA). Samples from each solution
`were taken 30 min and 2 h after the addition of ketoconazole to evaluate the solubility of ketoconazole over the
`usual time period of maximal absorption of ketoconazole in humans. The addition of GA to SGF leads to an
`increase in solution acidity, while the pHs of SGF at a pH of 1, 2, or 3 are markedly increased by the addition
`of sucralfate. There is a net decrease in acidity from initial pHs for the pH 1, 2, and 3 solutions when GA and
`sucralfate are combined. The concentration of ketoconazole in SGF at pHs of 1, 2, 3, 4, and 6 was evaluated
`in order to assess the pH-dependent solubility properties of the drug in the absence of other interacting species.
`Regardless of the initial pH, combinations of GA plus ketoconazole showed high concentrations of ketoconazole
`(=100%) in solution. In contrast, significant decreases in the concentration of soluble ketoconazole were
`observed when sucralfate was mixed with ketoconazole, and, in some cases, soluble ketoconazole was not
`detectable. The addition of GA to a mixture of sucralfate and ketoconazole leads to a significant increase in the
`concentration of solubilized ketoconazole. Nonetheless, important sucralfate-ketoconazole interactions are still
`observed. After 2 h, =35% of the maximal ketoconazole concentration remained in solution. Comparison of the
`ketoconazole concentrations at dilferent pHs with the predicted concentrations of the three protonation species
`of ketoconazole [H2(ketoconazole)2+, H(ketoconazole)+, or ketoconazole] showed no correlation. Therefore,
`the decrease in ketoconazole solubility is not simply a reflection of pH perturbation associated with the
`dissolution of sucralfate. The observed data are most consistent with a model that has H2(ketoconazole)2+ or
`H(ketoconazole)+ forming an electrostatic interaction with the sucralfate polyanion. The findings of this study
`suggest that the coadministration of sucralfate with other azole antifungal agents should be investigated.
`
`Ketoconazole is a dibasic imidazole antifungal agent with a
`pK,.1 of 6.51 and a pK,.2 of 2.94 that is virtually insoluble in
`neutral or slightly acidic solutions (3, 13). Recent studies have
`demonstrated that the dissolution and subsequent absorption
`of ketoconazole in humans are dependent on the presence of
`low gastric pH (1, 3, 8, 11). Sucralfate is a basic aluminum salt
`of sucrose octasulfate that forms aluminum ions and sucrose
`sulfate anions in solution at low pHs (14). Preliminary in vitro
`studies in our laboratory demonstrated that the addition of
`sucralfate to 50-ml aqueous solutions of ketoconazole at pH 1
`or 2 results in a significant decrease in the amount of ketocon(cid:173)
`azole in solution. In a subsequent study with healthy volunteers
`(4), we demonstrated that the oral bioavailability of ketocon(cid:173)
`azole was significantly decreased during simultaneous admin(cid:173)
`istration with sucralfate.
`We hypothesized that sucralfate may interact with ketocon(cid:173)
`azole by adsorption of ketoconazole to the paste form of
`sucralfate (produced at low pH) by an electrostatic interaction
`between the mono- or divalently charged ketoconazole moi(cid:173)
`eties and the negatively charged sucrose octasulfate ions or by
`some combination of these two factors. However, since sucral(cid:173)
`fate acts as a buffer with a pK,. of approximately 4.5 (9, 14), a
`decrease in ketoconazole concentrations in solutions with an
`
`• Corresponding author. Mailing address: College of Pharmacy, The
`University of Michigan, 428 Church Street, Ann Arbor, MI 48109-
`1065. Phone: (313) 764-9384. Fax: (313) 763-2022.
`
`initial pH of <2 could also result from a sucralfate-induced
`increase in pH.
`The purpose of this in vitro study was to investigate the
`mechanism of this interaction by examining the influence of
`ketoconazole, glutamic acid hydrochloride (GA), sucralfate,
`and combinations of these three agents on pH and the
`solubility of ketoconazole in both simple aqueous solutions
`and in simulated gastric fluid (SGF).
`
`MATERIALS AND METHODS
`
`Materials. GA (340-mg capsules [Acidulin; Eli Lilly Co.,
`Indianapolis, Ind.; lot 1EX35A]), sucralfate (1-g tablets [Car(cid:173)
`afate; Marion Laboratories, Inc., Kansas City, Kans.; lot
`K00790]), and ketoconazole (200-mg tablets [Nizoral; Janssen
`Pharmaceutica, Inc., Piscataway, N.J.; lot 97J089]) were ob(cid:173)
`tained from their respective manufacturers. R,S-Ketoconazole
`analytical standard (99.92% pure), purchased from Janssen
`Life Sciences Products, was used to obtain the UV -visible
`absorption spectrum of ketoconazole. Pepsin was purchased
`from Sigma Chemical Company (St. Louis, Mo.). All other
`chemicals and solvents were reagent grade. Stock solutions of
`SGF were prepared according to the USP formulation (12).
`Instrumentation. UV -visible spectra were recorded on a
`Perkin-Elmer Lambda 9 UV-visible-near-infrared spectropho(cid:173)
`tometer equipped with a Perkin-Elmer 3600 data station.
`Titration data were obtained with a Mettler DL 40RC Memo
`Titrator equipped with a DG-111-SC combination pH elec(cid:173)
`trode and a Mettler GA 15 recorder calibrated to record pH
`
`319
`
`

`

`320
`
`HOESCHELE ET AL.
`
`ANTIMICROB. AGENTS CHEMOTHER.
`
`£
`
`ISO
`
`12S
`
`100
`
`1S
`
`so
`
`2S
`
`Downloaded from
`
`http://aac.asm.org/
`
` on March 4, 2014 by guest
`
`200
`
`2SO
`WAVELENGTH. nm
`FIG. 1. uy-visible spectrum of ketoconazole in aqueous solution
`at_p!"ll. Umts for. the extinction coefficient (e) are in milliliters per
`milligram per centimeter. Ketoconazole concentration, 12 IJ.g/ml.
`
`300
`
`ous solution or SGF differs from that previously reported in
`methanol-phosphate buffer mixtures (1). The difference stems
`from a change in solvent and also differs because the previously
`reported spectrum was collected from a sample that had an
`optical density far exceeding the linear range of the spectro(cid:173)
`photometer used for measurements. The true spectrum for
`~etoconazole is presented in Fig. 1. Under these assay condi(cid:173)
`tions, the SGF had a smallA220 that consistently accounted for
`7% of the maximal ketoconazole absorbance at this wave(cid:173)
`length. Corrections for sucralfate and GA collectively
`amounted to =8% of the maximal ketoconazole A 220• These
`values could conveniently be subtracted to obtain reliable
`determinations of ketoconazole concentrations.
`The ketoconazo_Ie ratios, K%s, are calculated by equation 1,
`where [keto]MAx IS the concentration of ketoconazole
`
`[keto]oss
`K%=--=--(cid:173)
`[keto]MAX
`
`(1)
`
`assuming that all added drug is soluble (i.e., that a 200-mg
`tablet would produce a 2-mglml solution), and [keto]088 is the
`observed concentration of ketoconazole. Ketoconazole has
`three protonation species: H2(ketoconazole )2 +, H(ketocon(cid:173)
`azole) +, and ketoconazole free base. The aqueous solubility of
`ketoconazole has been shown to be a function of solution pH
`(3). Therefore, the solubilities of these three protonation states
`may not be identical. The concentration of any protonation
`form can be calculated directly by using the reported K values
`(!C1 = 10- 2.94 and K2 = 10- 6 .51 ) for ketoconazole vi~ equa(cid:173)
`tions 2 to 4:
`
`[H2(ketoconazole )2+] =
`
`[H(ketoconazole)+] =
`
`[keto]MAX
`1 +KI/[H+]+KtKz/[H+f
`
`[keto]MAX
`1 + [H+]/KI +Kz/[H+]
`
`[ketoconazole] =
`
`[keto]MAX
`1 + [H+]/Kz+[H+f/K1K2
`
`(2)
`
`(3)
`
`(4)
`
`continuously. The pH electrode was calibrated by using a 3- or
`4~point calibration with Mallinckrodt buffer (pH 7.0), potas(cid:173)
`SIUm hydrogen phthalate (pH 4.0) and potassium hydrogen
`tartrate (pH 3.56) primary standard solutions (5). In some
`cases, a standardized solution of hydrochloric acid (1.101 N)
`was used as a calibrator to ensure the accuracy of low pH
`determinations.
`Dissolution and interaction studies with unbutrered aque(cid:173)
`ous solutions. The initial pH conditions were selected to
`simulate a typical range of gastric pHs for healthy subjects
`(pHs of 1, 2, and 3). A fourth pH (pH 6) was included to
`simulate elevated gastric pH as might be found in patients with
`achlorhydria or in those receiving H2-receptor antagonists.
`Ketoconazole tablets (200 mg) were placed in 50-, 100-, and
`5~0-ml unbuffered aqueous solutions of pHs 1, 2, 3, and 6 and
`stmed at a constant velocity. Aliquots (250 J.Ll) of each solution
`were obtained at time 0 (prior to the addition of ketocon(cid:173)
`azole ); at 1, 3, and 5 min; and then at 5-min intervals for 2 h.
`Samples were centrifuged at 1,000 X g for 10 min and filtered,
`and _th~n the supe?Iatant was assayed spectrophotometrically.
`Prehmmary expenments demonstrated negligible binding of
`ketoconazole to filters.
`In the interaction phase of the study, ketoconazole tablets
`were dissolved as described above for 1 h in order to achieve
`dissolution equilibrium. Following removal of a baseline (con(cid:173)
`trol) sample, 1.0 g of sucralfate was added to the solution, and
`samples were obtained over 2 h and assayed as described
`above.
`In order to examine the ability of each drug to alter pH,
`ketocon~ole. (200 mg), sucralfate (1.0 g), GA (1,360 mg), or
`the combmat1on of all three agents was added to 50-, 100-, and
`500-ml unbuffered aqueous solutions adjusted to pH 1, 2, 3, or
`6, and the pH was monitored for 2 h.
`Interaction studies with SGF. One or more entities (GA,
`sucralfate, or ketoconazole) were added to 100 ml of SGF
`w~ich had been adjusted to a starting pH of 1, 2, 3, or 6 just
`pnor to each experiment with either 1 or 2 N NaOH or HCl,
`as required. In drug combinations involving GA, two 340-mg
`cap~~les of GA were added at time zero, followed by an
`additional tw<;> 340-mg capsules and the remaining drug com(cid:173)
`ponents 10 mm later. In drug combinations not involving GA,
`all drugs were added at a time designated time zero. Solutions
`were. stirred via _a stirring bar-magnetic stirrer, and the pH was
`momtored contmuously for 2 h following addition of the first
`drug. Samples (1 ml each) of each mixture were taken at 30
`min and 2 h and were either filtered or centrifuged immedi(cid:173)
`ately to obtain a clear filtrate, diluted 200-fold in 0.1 N HCl,
`and assayed for ketoconazole spectrophotometrically at 220
`nm.
`In order to separate the effects of protons and glutamic acid
`added by GA, we adjusted the pH of GA-ketoconazole (AK)
`and sucralfate-ketoconazole (SK) solutions collected at 30 min
`and 2 h to the corresponding 30-min or 2-h pH obtained for the
`GA-SK (ASK) sample wit~ the same initial pH. For example,
`the pH (measured at 30 mm) of an AK solution with an initial
`pH of 1 was adjusted to match the pH measured at 30 min
`from the ASK solu!ion with an initial pH of 1. HCl (1 N) was
`used for all pH adjustments in order to minimize changes in
`volume and ketoconazole concentration. Ketoconazole con(cid:173)
`centrations are reported in Table 2 as K%ad· so that the acidity
`of the solutions is invariant, although SK ~nd ASK solutions
`differed in their GA content. AK samples served as the control.
`Measurement of ketoconazole concentrations. The concen(cid:173)
`tration of ketoconazole in aqueous solutions and SGF was
`monito~e1d spectrophot<;>metrically at 220 nm [e = 50 (mV
`mg)cm
`]. The absorption spectrum of ketoconazole in aque-
`
`

`

`VoL. 38, 1994
`
`IN VITRO INTERACTION OF SUCRALFATE WITH KETOCONAZOLE
`
`321
`
`Downloaded from
`
`http://aac.asm.org/
`
` on March 4, 2014 by guest
`
`TABLE 1. Ketoconazole concentrations in aqueous solutions 2 h
`after addition of ketoconazole
`
`Vol of
`water
`
`50 ml
`
`100 ml
`
`500ml
`
`pH
`
`Initial0
`
`Finalb
`
`K%c
`
`KH2Kd
`
`KHKd
`
`KKd
`
`1
`2
`3
`6
`1
`2
`3
`6
`1
`2
`3
`6
`
`1.04
`2.80
`3.62
`6.02
`1.17
`2.29
`3.49
`6.01
`0.99
`2.09
`3.15
`6.00
`
`101.6
`73.2
`8.9
`1.6
`99.6
`82.4
`17.5
`1.6
`98.3
`98.9
`58.2
`2.0
`
`98.8
`58.0
`17.3
`0.1
`98.3
`81.7
`22.0
`0.1
`98.9
`87.6
`38.1
`0.1
`
`1.2
`42.0
`82.6
`76.0
`1.7
`18.3
`78.0
`76.5
`1.1
`12.4
`61.8
`76.9
`
`0.0
`0.0
`0.1
`23.9
`0.0
`0.0
`0.10
`23.5
`0.0
`0.0
`0.0
`23.1
`
`a Initial pH of solution at the start of the experiment.
`b Final pH (2 h after the start of the experiment).
`c K%, ratio of observed to maximal ketoconazole concentration in water
`( e~uation 1) at 2 h.
`Predicted percentage of concentration of H2(ketoconazole f+ (KH2K), H(ke(cid:173)
`toconazole)+ (KHK), and ketoconazole (KK) at the final pH of the experiment
`(equations 5 to 7).
`
`GA to the SK regimen reduced the magnitude of the sucral(cid:173)
`fate-induced increase in pH to s0.5 pH units.
`pH dependence and solubility of ketoconazole in SGF.
`Figure 3 illustrates the relationship between the initial and
`final pH of solutions containing AK, SK, and ASK in SGF at 30
`min and 2 h after addition of each agent. The addition of
`ketoconazole alone had little effect on the pH of SGF,
`although unbuffered aqueous solutions (Table 1) exhibit = 1-
`pH-unit shifts to more basic values.
`In contrast to aqueous solutions, ketoconazole was >93%
`soluble in 100-ml SGF solutions maintained at pHs of 1 (99% ),
`2 (100% ), 3 (93% ), and 4 (94% ). At pH 6, a milky precipitate
`formed, which, after filtration, yielded only 5% of the maximal
`available ketoconazole concentration in solution. Therefore, in
`the range of pHs 1 to 4, ketoconazole is completely soluble in
`100 ml of SGF.
`AK. As expected, the addition of GA to SGF leads to an
`increase in solution acidity. The effect is minor at an initial pH
`of 1; however, it is marked at higher pHs. Equilibrium pH is
`rapidly achieved regardless of initial pH. Most important, the
`addition of GA ensures that the final solution pH will remain
`below 2 regardless of the initial solution acidity.
`Regardless of the initial pH, AK solutions showed the same
`high level of ketoconazole (=100%) in solution (Fig. 4). As
`seen in Table 2, the soluble ketoconazole levels in SGF do not
`directly correlate with any of the predicted distributions of
`individual protonation forms of ketoconazole.
`SK. The most dramatic changes in pH were observed
`following the addition of both sucralfate and ketoconazole to
`SGF solutions. The pHs of SGF solutions at initial pHs of 1, 2,
`and 3 are markedly increased by the addition of sucralfate.
`After only 30 min, a pH 1 starting solution is raised to pH 1.8
`and after 2 h has equilibrated at pH 2.7. Similarly, solutions
`with initial pHs of 2 and 3 exhibit final (2-h) pHs of 4.2 and
`4.35, respectively. Most interesting is the observation that
`addition of sucralfate to SGF at pH 6 results in an increased
`acidity as the pH drops to 5. This is co_nsistent with our
`observation that sucralfate effectively buffers solutions to a pH
`of=4.
`Appreciable decreases in the soluble ketoconazole levels
`were observed when sucralfate was mixed with ketoconazole.
`
`Ko/o
`
`100
`
`90
`
`80
`
`70
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`0
`
`2
`
`3
`
`5
`
`6
`
`7
`
`8
`
`4
`pH
`FIG. 2. Predicted percentage (equations 1 to 7) of ketoconazole
`species (K%) as a function of pH. Predicted (equations 5 to 7)
`percentages of concentrations of H2(ketoconazole )2+ ( * ), H(ketocon(cid:173)
`azole)+ (D), and ketoconazole (l>) are shown.
`
`The values of K" K• KHK• and KK correspond (at a specified
`pH) to the expected ratios of Hz(ketoconazole ), H(ketocon(cid:173)
`azole) and ketoconazole free base to [keto]MAX, respectively,
`as described by equations 5 to 7
`[H2(ketoconazole )2+]
`[keto]MAX
`
`X 100
`
`(5)
`
`KH2K =
`
`KHK =
`
`[H(ketoconazole +)]
`
`[keto]MAX
`
`X 100
`
`KK =
`
`[ketoconazole]
`
`[keto]MAX
`
`X 100
`
`(6)
`
`(7)
`
`and depicted in Fig. 2. The sum of K" K• KHK• and KK equals
`100.
`2
`
`RESULTS
`
`pH dependence and solubility of ketoconazole in aqueous
`solutions. With the exception of the pH 2 (50-ml) solution,
`dissolution of ketoconazole reached equilibrium within 20 min
`at all pHs regardless of the fluid volume. However, the extent
`of dissolution (assuming each tablet of ketoconazole contained
`200 mg) varied widely with both pH and the volume of solution
`employed (Table 1). Regardless of the volume studied, disso(cid:173)
`lution of ketoconazole in aqueous solutions at pH 3 was poor
`(8.9, 17.5, and 58.2% at 50, 100, and 500 ml, respectively).
`The addition of sucralfate to aqueous solutions of ketocon(cid:173)
`azole resulted in a significant reduction in the amount of
`ketoconazole in solution at pHs of 1 and 2. At pH 1, the
`amount of ketoconazole in solution decreased by approxi(cid:173)
`mately 48, 44, and 42% in 50-, 100-, and 500-ml solutions,
`respectively. At pH 2, the addition of sucralfate resulted in less
`dramatic but significant interactions (55, 43, and 18% reduc(cid:173)
`tions for 50, 100, and 500 ml, respectively). Addition of
`sucralfate to a 50-ml solution at pH 6 decreased the pH to -4.8
`and increased the amount of ketoconazole in solution from 0. 7
`to 12.2 mg (a 17-fold increase). With the exception of the 50-ml
`pH 6 solution, addition of sucralfate did not affect the amount
`of ketoconazole in solutions with an initial pH of 3 or 6.
`The addition of ketoconazole, sucralfate, or both agents to
`aqueous solutions with an initial pH of 2 or 3 resulted in an
`approximately 1- to 2-pH-unit increase in pH. The addition of
`
`

`

`322
`
`HOESCHELE ET AL.
`
`ANTIMICROB. AGENTS CHEMOTHER.
`
`..
`
`.c
`N
`'1:1
`c
`01
`c
`
`e
`
`0
`C')
`'Iii
`:c
`c:a.
`
`A
`
`8
`
`ll pH
`
`7
`
`6
`
`5
`
`4
`
`3
`
`2
`
`0
`0
`
`1.0
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`0.0
`
`.0.2
`
`Initial pH
`FIG. 3. The pH dependence of drug dissolution in SGF. (A) pH 30
`min (open symbols) and 2 h (solid symbols) after addition of drug. 6,
`A, AK; 0, e, ASK; 0, •, SK. (B) Difference between pHs at 30 min
`and 2 h after addition of drug. •, AK; ~.ASK; £3, SK.
`
`As little as 14% of the ketoconazole originally added to SGF
`remains in a soluble fraction 2 h after the addition of sucral(cid:173)
`fate, despite the fact that the solution is still rather acidic (pH
`2.72). In comparison, 96% of the available ketoconazole was
`soluble at pH 2.72 in the absence of sucralfate. Most interest(cid:173)
`ing, at pHs between 4 and 5, there was no detectable ketocon(cid:173)
`azole in solution, despite the fact that ketoconazole alone is
`still soluble in SGF at this pH (Table 2).
`As shown in Fig. 4, longer exposure times led to lower
`ketoconazole concentrations. For example, at an initial pH of
`1, SK solutions at 30 min versus those at 2 h yield K%s of 23
`versus 14, respectively. Although the pH of the SK solutions
`continued to rise between 30 min and 2 h, the increased
`basicity was not the sole cause of the lowered ketoconazole
`solubilities. This is clearly seen by comparing the K%s for SK
`at pH 1 (pH = 1.78, K% = 23) and SKat pH 2 (pH = 4.00,
`K% = 23) at 30 min. Furthermore, there is no correlation
`between the predicted ratios of H2(ketoconazole )2 +, H(keto(cid:173)
`conazole) +, or ketoconazole and the observed ketoconazole
`solubilities.
`ASK. As shown in Fig. 3, there is a net decrease in acidity
`from the initial values for the pH 1, 2, and 3 ASK solutions.
`The effect is minor, however, since the largest difference
`between AK and ASK for these three solutions is 0.6 pH units.
`However, once again, the pH 6 solution becomes markedly
`more acidic. The bottom graph of Fig. 3 illustrates that the
`
`1!!1
`
`•
`0
`A
`
`1!!1
`
`•
`0
`A
`
`•
`•
`D
`0
`A
`
`100
`
`80
`
`80
`
`40
`
`20
`
`•
`
`•
`i
`
`~
`
`~
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`Initial pH
`
`Downloaded from
`
`http://aac.asm.org/
`
` on March 4, 2014 by guest
`
`Initial pH
`FIG. 4. Comparison of the observed ketoconazole concentrations
`(K%) in AK, SK, and ASK 30 min and 2 h after addition of drug to
`SGF. The concentrations were measured at the equilibrium (final) pH
`of each solution. •, K% in AK after 30 min; £3, K% in SK after 30 min;
`~, K% in ASK after 30 min; Iii:!, K% in AK after 2 h; 0, K% in SK after
`2 h; ~. K% in ASK after 2 h. Final pHs are given in Table 2.
`
`ASK solutions undergo the slowest approach to pH equilib(cid:173)
`rium. In general, the pHs of the ASK solutions increased by 0.4
`pH units between 30 and 120 min.
`The addition of GA to SK leads to a significant increase in
`the concentration of solubilized ketoconazole, especially for
`solutions with initial pHs of 3 or greater (Fig. 4). For example,
`after 30 min, there is no detectable ketoconazole in SK at an
`initial pH of 3 or 6 (Table 2), while 46 to 60% of ketoconazole
`is still soluble in ASK solutions at an initial pH of 3 or 6.
`However, even with an initial pH of SGF at 1 or 2, the soluble
`ketoconazole concentrations nearly double in the presence of
`GA.
`Like those in SK solutions, ketoconazole concentrations in
`ASK solutions decreased substantially between 30 min and 2 h.
`Only at an initial pH of 1 was the ketoconazole concentration
`invariant between these two time points. The greatest effect
`was at an initial pH of 6, where the ketoconazole concentration
`dropped from 60% to 36%. There is a slight, but potentially
`important, difference in pH between the 30-min and 2-h ASK
`samples (Fig. 3). Utilizing data in Table 2, one can compare
`the observed ketoconazole concentrations with the predicted
`ketoconazole solubilities, assuming that the soluble form of the
`drug is Hz(ketoconazole ?+, H(ketoconazole ), or both species.
`Clearly there is no direct correlation between the protonation
`states of ketoconazole and the amount of soluble ketoconazole
`in SGF, demonstrating that a change in pH due to the addition
`of sucralfate is not the mechanism of drug interaction.
`The data presented in Fig. 4 represent the ketoconazole
`solubility that one would expect when utilizing various drug
`combinations in patients with gastric pHs ranging between 1
`and 6. However, this analysis does not distinguish pH effects
`that may be important for direct chemical interaction between
`ketoconazole and sucralfate (e.g., the protonation state of
`sucralfate varies through the pH range of 2 to 5, and the
`different forms of sucralfate could interact with ketoconazole
`differently). Furthermore, it does not identify the role other
`than alteration of pH that GA may play in enhancing the
`solubility of ketoconazole.
`The data in Table 2 demonstrate that ketoconazole is, as
`expected, essentially 100% available in the AK solutions. The
`addition of acid to SK solutions causes a dramatic increase in
`the solubility of ketoconazole, especially at the high pHs; the
`addition of acid to SK at pHs of 3 and 6 increased ketocon-
`
`

`

`Downloaded from
`
`http://aac.asm.org/
`
` on March 4, 2014 by guest
`
`VoL. 38, 1994
`
`IN VITRO INTERACfiON OF SUCRALFATE WITH KETOCONAZOLE
`
`323
`
`TABLE 2. Ketoconazole concentrations in SGF 30 min and 2 h after addition ofketoconazole
`
`Solution
`
`30 min postaddition
`AK
`AK
`AK
`AK
`SK
`SK
`SK
`SK
`ASK
`ASK
`ASK
`ASK
`
`2 h postaddition
`AK
`AK
`AK
`AK
`SK
`SK
`SK
`SK
`ASK
`ASK
`ASK
`ASK
`
`pH
`
`Initial"
`
`Finalb
`
`1
`2
`3
`6
`1
`2
`3
`6
`1
`2
`3
`6
`
`1
`2
`3
`6
`1
`2
`3
`6
`1
`2
`3
`6
`
`0.98
`1.71
`1.90
`1.93
`1.78
`4.00
`4.32
`4.89
`1.41
`2.29
`2.32
`2.28
`
`0.93
`1.70
`1.87
`1.93
`2.72
`4.07
`4.35
`4.89
`2.08
`2.72
`2.67
`2.70
`
`K%c
`
`104
`103
`98
`99
`23
`23
`0
`0
`37
`52
`46
`60
`
`101
`99
`102
`102
`14
`20
`0
`0
`35
`37
`37
`36
`
`K%ad{
`
`KH2Kd
`
`KHKd
`
`KKd
`
`96
`95
`100
`102
`25
`44
`48
`34
`
`96
`96
`100
`98
`18
`49
`33
`34
`
`98.9
`94.4
`91.6
`91.1
`93.5
`8.0
`4.0
`1.1
`97.1
`81.7
`80.7
`82.0
`
`99.0
`94.6
`92.2
`91.1
`62.4
`6.9
`3.7
`1.1
`87.9
`62.4
`65.1
`63.5
`
`1.1
`5.6
`8.4
`8.9
`6.5
`91.7
`95.4
`96.7
`2.9
`18.3
`19.3
`18.0
`
`1.0
`5.4
`7.8
`8.9
`37.6
`92.8
`95.6
`96.7
`12.1
`37.6
`34.9
`36.5
`
`0.0
`0.0
`0.0
`0.0
`0.0
`0.3
`0.6
`2.3
`0.0
`0.0
`0.0
`0.0
`
`0.0
`0.0
`0.0
`0.0
`0.0
`0.3
`0.6
`2.3
`0.0
`0.0
`0.0
`0.0
`
`• Initial pH of solution at the start of the experiment.
`b Final pH (2 h after the start of the experiment).
`c K%, ratio of obse!Ved to maximal ketoconazole concentration in SGF (equation 1) at 30 min or 2 h (prior to adjustment to final pH of ASK experiment); K%adi•
`ratio of obse!Ved to maximal ketoconazole concentration in SGF after adjustment to the final pH of the ASK experiment with the corresponding initial pH.
`d Predicted (equations 5 to 7) percentage of concentration of H2(ketoconazole)Z+ (KH2K), H(ketoconazole)+ (KHK), and ketoconazole (KK) at the final pH of the
`experiment.
`
`azole concentrations from 0 to 48% and from 0 to 34%,
`respectively. In fact, at these high pHs, the concentrations of
`ketoconazole in ASK and pH-adjusted SK samples are com(cid:173)
`parable. It appears that pH can influence ketoconazole solu(cid:173)
`bility, even though there is no direct correlation between the
`protonation state of ketoconazole and the amount of soluble
`ketoconazole present in SGF. We can also conclude that it is
`the presence of the protons and not the glutamic acid that is
`the important factor in increasing ketoconazole concentrations
`in the ASK samples.
`
`DISCUSSION
`The rationale for the specific study design of this in vitro
`analysis was the methodology employed in our previous study
`with human subjects that demonstrated a decreased bioavail(cid:173)
`ability of ketoconazole during simultaneous administration
`with sucralfate. In this study, subjects were administered two
`oral 680-mg doses of GA 10 min apart. A 400-mg oral dose of
`ketoconazole (with or without a 1-g oral dose of sucralfate) was
`administered with the second dose of GA. Simultaneous
`administration with sucralfate resulted in a 24% decrease in
`the bioavailability of ketoconazole ( 4 ). We sought to mimic the
`clinical conditions in vitro in order to understand the interac(cid:173)
`tion of sucralfate and ketoconazole on the molecular level. Our
`hypothesis was that ketoconazole and sucralfate undergo an
`electrostatic interaction, significantly decreasing the amount of
`ketoconazole available for transport across the gastrointestinal
`epithelium.
`
`Ketoconazole has a pK,.1 of 6.51 and a pK,.2 of 2.94 and is
`virtually insoluble in neutral or slightly acidic solutions (3).
`However, Carlson et a!. (3) reported that the in vitro dissolu(cid:173)
`tion of ketoconazole was rapid and virtually complete ( > 90%)
`in buffer solutions at pHs of 2, 3, and 4. At pH 5, only 37% of
`ketoconazole was in solution, while only 10% was available at
`pH 6. In contrast, we found that dissolution of ketoconazole
`was strongly dependent on the pH, medium, and (at pH 3) the
`volume of solution. Regardless of the volume studied, disso(cid:173)
`lution of ketoconazole in aqueous solutions at pH 3 was poor.
`In 100 ml of SGF, ketoconazole was >93% soluble in the pH
`range of 1 to 4. Since unbuffered solutions were utilized, the
`differences in ketoconazole solubility in aqueous solutions at
`pH 3 for 50 versus 500 ml are not surprising. As a weak base,
`the addition of ketoconazole increases pH; this effect is more
`pronounced in smaller fluid volumes.
`Because sucralfate and ketoconazole are weak bases and
`GA is a moderately strong acid, all three drugs can perturb
`gastric pH. Although the shift in pH is less in SGF than in
`water because pepsin and other proteins can act as. buffers,
`marked pH changes can still be observed. ln particular,
`sucralfate, which buffers solutions towards a pH of -4.5,
`dramatically increases the pH of 100-ml SGF solutions at
`initial pHs of less than 3 and decreases the pH of solutions that
`are initially at pH 6 (Table 2). Ketoconazole produces only
`minor effects on the pH of SGF (less than half a log unit).
`Addition of GA leads to very acidic solutions (pH $2)
`regardless of the initial pH of the SGF. Most important for
`
`

`

`324
`
`HOESCHELE ET AL.
`
`ANriMICROB. AGENTS CHEMOTHER.
`
`110
`
`100
`
`80
`
`a 60
`<
`"$.
`~
`
`40
`
`;~
`
`20
`
`0
`
`2
`
`3
`
`6
`
`6
`
`Downloaded from
`
`http://aac.asm.org/
`
` on March 4, 2014 by guest
`
`Initial pH
`FIG. 6. Comparison of the observed difference in ketoconazole
`concentrations in AK and SK solutions before (K%) and after (K%adj)
`adjustment to the same pH as the final pH for the ASK sample that
`had the identical initial pH value. The pHs prior to adjustment are
`given for each sample in Table 2 in the final pH column. The adjusted
`pHs are given in Table 2 in the rows for ASK entries. For example,
`after 2 h, SKat an initial pH of 6 had a final pH of 4.89. The pH of the
`SK solution was adjusted to 2.70, the value for ASK with an initial pH
`of 6 after 2 h. K%adj - K% (34 - 0) = 34. •. K%•di - K% for AK
`after 30 min; rzJ, K%adi - K% for SK after 30 min; m, K%adi - K%
`for AK after 2 h; lii!ll, K%adj - K% for SK after 2 h.
`
`soluble than Hiketoconazole )2+, this finding suggests two
`disparate roles for GA. At pH 4, =40% of GA is present as the
`glutamate monoanion. The monoanion rna~ compete with the
`sucralfate polymer for H2(ketoconazole) + or H(ketocon(cid:173)
`azole)+ via an electrostatic interaction. Alternatively, the
`glutamic acid portion of GA may be irrelevant and the role of
`GA may be simply as an acid. For example, it is possible that
`sucralfate protonation states differ in their ability to interact
`with ketoconazole. Therefore, one would observe a pH depen(cid:173)
`dence for the interaction, but this dependence would not
`correlate directly with ketoconazole protonation species.
`In order to differentiate these two possibilities, we compared
`the solubilities of ketoconazole in a variety of pH-adjusted
`sucralfate mixtures (Fig. 5 and 6). With the exception of SGF
`samples with an initial pH of 1, the amounts of soluble
`ketoconazole in the presence and absence of GA were equiv(cid:173)
`alent (within experimental error). At pH 1, the observed
`deviation is likely due to development of a different polymeric
`form of sucralfate. The gelatinous, intractable polymer of
`sucralfate that develops at pH 1 may trap ketoconazole within
`its matrix during the mixing process. The polymer precipitates
`with ketoconazole, making it kinetically inaccessible to further
`reactions with sucralfate. These data strongly suggest that the
`role of GA is not to interact directly with ketoconazole and are
`most reasonably attributed to the decreasing pH of the gastric
`fluid.
`The pH effects are probably the result of changes in the
`protonation states of sucralfate and/or
`the release of
`A12(0H)5 + from sucralfate as the solution becomes more
`acidic. If so, one would expect the increase in ketoconazole
`concentrations observed when SK solutions are adjusted to
`ASK pHs to show a bell-shaped curve peaking around pH 4.5,
`the pK of sucralfate. This exact trend is observed in Fig. 5.
`Similar trends have been observed by Hikal and colleagues ( 6),
`who reported maximal adsorption of furosemide (a weak base)
`at pH 3 (initial pH), and by Nagashima (9, 10), who found a
`similar trend with bile acids. Therefore, the pH dependence of
`
`Initial pH
`FIG. 5. Comparison of the observed ketoconazole concentrations
`in AK, SK, and ASK 30 min and 2 h after addition of drug to SGF. The
`concentrations were measured after the AK and SK solutions had been
`adjusted to the same pH (K%adj) as the final pH for the ASK sample
`that had the identical initial pH value. •. K%adi in AK after 30 min;[!],
`K%adj in SK after 30 min; lii!ll, K%adj in ASK after 30 min; lii!ll, K%adj in
`AK after 2 h; D, K%adj in SK after 2 h; !!!!, K%adj in ASK after 2 h.
`Adjusted pHs are given in Table 2 in the rows for ASK entries.
`
`clinical studies, the addition of GA to an SK mixture leads to
`a constant final pH ( =2. 7) for gastric pHs that initially ranged
`between 2 and 6. This supports the findings of our previous
`study of healthy subjects with elevated gastric pH following
`administration of a 300-mg oral dose of ranitidine, in whom
`administration of two 680-mg oral doses of GA decreased
`median gastric pH from 6.2 to 1.6 within 15 min (7).
`Decreased