- Email: [email protected]
Determination and characterization of the interaction between magaldrate antacid and some drugs: Part II
PHARMACEUTICA ACTA HEIJETIAE ELSEVIER
Pharmaceutics Acta Helvetiae 70 (1995) 289-300
Determination and characterization of the interaction between magaldrate antacid and some drugs: Part II Omaimah M.N. Al Gohary King Saud University, College of Pharmacy, Department of Pharmaceutics,
*
P.O. Box 7633, Riyadh 11472, Saudi Arabia
Abstract The interaction between magaldrate antacid and some drugs was investigated. Adsorption studies showed that magaldrate exhibited the highest adsorptive capacity for propranolol hydrochloride, intermediate for acetylsalicylic acid and the lowest for indomethacin. The adsorption was in accordance to the Langmuir, BET, and Freundlich model, respectively. An endothermic process was exhibited for all systems. Adsorption was pH dependent and was suppressed as the concentration and valency of the electrolyte were increased. The volume of the test mixture had a negligible effect on the extent of adsorption when the drug concentration was kept constant, whereas, when a fixed weight of drug in different solvent volumes was used, the adsorption was decreased as the volume increased. The equilibrium concentration of test drugs decreased as the antacid concentration was increased. Elution studies show that the drug recovery from the antacid surface was dependent on the volume and pH of elution media. A competitive mechanism was involved in the process and elution followed the sequence: 0.1 M HCl > 0.01 M HCl > 0.1 M MgCl, > 0.01 M MgCl, > 0.1 M NaCl > 0.01 M NaCl. The data obtained from adsorption-desorption process revealed that physical adsorption as well as chemisorption were existing. Melting point determinations (MP), differential scanning calorimetry analysis (DSC) and infra-red spectroscopy (IR), suggested a clear interaction between magaldrate antacid and the test drugs which could be attributed to chemical changes in test drugs. Keywords:
Drug interaction; Magaldrate; Propranolol; Acetylsalicylic acid; Indomethacin; Adsorption
1. Introduction Antacids play a role in the management of gastrointestinal disorders which may occur as side effects during antihypertensive, antiarrhythmic or antirheumatic therapy [ 11. The potential interactions of non-prescription antacids with various prescription drugs were discussed [2,3]. The antacid-drug interactions were attributed to different mechanisms, adsorption being one of the most important mechanisms reported in literature [4-151. Effects of antacids on clinical pharmacokinetics of drugs were studied by many workers ([16-271). Previous in vitro investigations showed that the dissolution was adversely affected for propranolol . HCl from
* Corresponding author. Tel./Fax:
+966 1 4644283.
0031-6865/9.5/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDI 0031-6865(95)00033-X
inderal tablets, acetylsalicylic acid from aspirin tablets and indomethacin from indocid capsules when magaldrate was present in the dissolution media (part I of this study). The purpose of the present work was to determine and characterize the possible interaction between magaldrate and the previously mentioned drugs by utilization of adsorptiondesorption studies as well as MP, DSC, and IR determinations.
2. Experimental
procedures
2.1. Chemicals Magaldrate powder (KTT, Osaka, Japan), propranolol hydrochloride powder (ICI, Macclesfield, UK), acetylsalicylic acid powder (Riedel-de Ha&, Hannover, Germany)
290
O.M.N.AI Gohary / Pharmaceutics Acta Heluetiae 70 (1995) 289-300
and indomethacin powder (Scherer, Seelze, Germany). All other reagents were of pure analytical grade and were supplied by BDH (Poole, UK)
and the pH of filtrates was recorded. The amount adsorbed was measured and compared to that adsorbed from a similar initial drug concentration at the same temperature and pH but in the absence of electrolytes.
2.2. Methods 2.2.1. Adsorption studies In vitro adsorption studies were carried out by shaking 1.0 g of magaldrate antacid powder with 20 ml of a solution containing different concentrations of propranolol HCl in 0.1 M HCl (pH 1.2) acetylsalicylic acid in acetate buffer (pH 4.5) and indomethacin in phosphate buffer (pH 7.4) Suspensions were shaken in a thermostatically controlled water bath (Karl Kolb type D 6.72, Bonn, Germany) at 80 rpm and 37 k 05°C until equilibrium was attained (60, 90 and 30 min), respectively. Filtered in sintered glass filters grade 5. The pH values of solutions and suspensions were measured using a Tacussel electronic pH meter (PHN81, Villeurbanne, France) with an accuracy of kO.05 pH units, The drug concentration remaining in the filtrate was determined spectrophotometritally using a Pye Unicam SP8800 spectrophotometer (Cambridge, UK) at 290 nm, 265 nm and 318 nm, respectively. Blank experiments showed no interference. Three runs were obtained and the results averaged. 2.2.2. Effect of temperature on the extent of adsorption of test drugs onto magaldrate antacid powder Adsorption experiments were carried out under the same conditions mentioned above, at various temperatures: 25, 37, 44 and 50°C using a fixed known concentration of each drug, the pH was kept constant for each system in all experiments. The amount of drug adsorbed was calculated and the results were compared. 2.2.3. Effect of pH on the extent of adsorption of test drugs onto magaldrate antacid powder Adsorption experiments were carried out as previously mentioned, at a pH range 2.5-7.4, and 37 + 0.5”C, using a fixed known concentration of each drug . The amount of drug adsorbed was calculated and the results were compared. 2.2.4. Effect of the presence of monovalent and divalent electrolytes on the extent of adsorption of test drugs onto magaldrate antacid powder 1.0 g of antacid powder was suspended in an aliquot (5 ml) of 0.01, 0.05 and 0.1 M sodium chloride or magnesium chloride solutions, then a fixed known amount of test drug was dissolved in 15 ml of the specified solvent and was added to that suspension. Adsorption experiments were carried out as previously mentioned at 37 f 0.5”C
2.2.5. Effect of varying the volume of test solution on the extent of adsorption of test drugs onto magaldrate antacid powder Adsorption experiments were carried out under the same conditions mentioned above, at 37 + 0.5”C. A fixed weight of antacid (1.0 g) and a fixed concentration (% w/v) of each test drug obtained from different drugsolvent ratio were used, fixed weight in different solvent volumes was used as well. Then, the amount of drug adsorbed(mg g-‘) was calculated and the results were compared. 2.2.6. Effect of varying the concentration of magaldrate antacid powder (g% w / v) on the equilibrium concentration of drug in solution, (Ce mg%) Different weights of antacid powder (0.5, 1.0, 1.5 and 2.0 g) were suspended in 20 ml of the specified solvent of each test drug containing a fixed known concentration of the drug and equilibrated at 37 k 0.5”C as mentioned under adsorption studies. Then, the amount of drug remaining in solution at equilibrium (Ce mg%) was determined spectrophotometrically at the specified wavelength. The effect of the antacid powder concentration (2.5, 5.0, 7.5 and 10 g% w/v) on the extent of adsorption was shown by plotting the equilibrium concentration Ce against the antacid powder concentration. 2.2.7. Desorption studies The residue obtained from adsorption process was collected and dried at room temperature away from light, shaken at 80 rpm and 37 f 0.5”C for 10 min with 20 or 50 ml of different elution media, then treated as mentioned under adsorption studies. Successive washings were carried out until no drug was detected in the filtrate. 2.2.8. Characterization of the interaction between magaldrate antacid and the test drugs 1.0 g of antacid powder with 20 ml of the specified solvent containing 200 mg propranolol HCl, 150 mg acetylsalicylic acid and 40 mg indomethacin were treated as previously mentioned under adsorption studies. The residue obtained from adsorption experiments were collected, dried over calcium chloride in a desiccator for 24 h away from light, then subjected to the following tests: Melting point (MP) determination. Melting point apparatus (Mettler FP800, Mettler instruments AG, Griefensee, Switzerland) was used for MP determinations of the sam-
O.M.N. Al Gohary / Pharmaceutics
ples at a heating rate of 2”C/min, until the desirable temperature reached, then the rate was lowered to 0.2”C/min until complete melting occurred. The melting point recorded for each sample was the average of five readings. Differential scanning calorimetry (DSC) analysis. The DSC thermograms were recorded on a Perkin Elmer DSC-4 differential scanning calorimeter (Norwald, USA) calibrated with indium (99.999%). 2 mg of the samples were heated at a heating rate of 30”C/min over a temperature range of 30-350°C in closed aluminum pans under an argon purge. At least three determinations were used to calculate mean values and standard deviation. Injkared spectroscopy (IR) determination. IR spectra were taken on a Perkin Elmer IR spectrophotometer (IR783, Beacons field, UK) as KBr pellets. The samples:KBr ratio was 1:300 mg, and a pressure of 15 tons was used, spectra were recovered over a 4000-250 cm-’ range, with a scan speed of 300 cm-‘/min. Different samples of antacid dry powder alone or treated with the specified solvents, and powdered drugs alone or their physical mixture with antacid in ratio 1:l were also subjected to MP, DSC, and IR determinations and the results were compared.
3. Results and discussion 3.1. Adsorption
studies
Fig. 1 shows that the adsorption of propranolol . HCl onto magaldrate in 0.1 M HCl at pH 4.7 and 37°C was in agreement with Langmuir adsorption isotherm [28]: x/m
= abC,/l
100
+ bC,
200
300
(la)
400
500 600 Ce mg%
700
800
go0
1000
Fig. 1. Langmuir adsorption isotherm of propranolol hydrochloride onto magaldrate from 0.1 M HCl, pH 1.2 at 37°C and pH 4.7. (-0-j C, against x/m (-W-I C, against C, /(x/m).
Acta Heluetiae 70 (1995) 289-300
291
0 00
I
50
100
150
200
250
300 350 Ce mg %
400
Fig. 2. BET adsorption isotherm of acetylsalicylic from acetate buffer pH 4.5 at 37°C.
450
500
550
600
acid onto magaldrate
or its linear form C,/(
x/m)
= l/ah
+ C,/a
(lb)
Where the slope = l/a and the intercept (c) = l/ah, CC,) is the equilibrium concentration of solute mg%, (x/ m) is the mass of solute (x) adsorbed per gram (m) of the adsorbent mg g-l, (b) is the enthalpy constant and (a) is the monolayer adsorptive capacity of antacid for the drug which may be calculated by extrapolating the plateau region of the curve back to the y - axis, the value of (a) equals 20.6 mg gg’. The adsorption constants (a), (b) and correlation coefficient (r) were estimated by linear regression analysis where r = 0.999, the value of (a) equals to the reciprocal of the slope, therefore, a = 21.739 mg g-‘, the value of b = l/at = 1.259 1 g-’ which is the affinity constant of the adsorption process. The data obtained from the adsorption of acetylsalicylic acid onto magaldrate in acetate buffer at pH 4.5 and 37°C were in accordance with BET adsorption isotherm (Fig. 2). Since a sigmoidal or S-shaped isotherm was obtained, thus indicating multilayer formation. The initial direction of curvature shows that adsorption becomes easier as the concentration rises. This isotherm assumed that the solute molecule was adsorbed onto fixed site as a single molecule and there was no lateral interaction between the molecules and that the heat of adsorption of a molecule to any layer other than the first was equal to heat of condensation [29]. The S curve usually appears when the solute molecule has a fairly large hydrophobic residue > C, and moderate intermolecular attraction which causes it to pack vertically in regular array in the adsorbed layers, and meets strong competition for substrate sites from molecules of the solvent [29].
292
O.M.N. Al Gohary/Pharmaceutica
Acta Helvetiae 70 (1995) 289-300
I
0’
30
.2
.1
3
.4
.6
.6
.7
.S
c/c,
Fig. 3. Linear plot of BET adsorption isotherm of aceytylsalicylic onto magaldrate from acetate buffer pH 4.5 at 37°C.
acid
Fig. 3 shows the straight line form of the BET adsorption isotherm [30]:
c
1
b-l
=Y,b+-.-
(2)
C0
40
where C, is the solubility when the solution is saturated (in g/100 ml), C the equilibrium concentration (g%), Y the mass of solid adsorbed at concentration C, Y, the mass of solid adsorbed per gram at monolayer capacity (g g-r ), and b the constant proportional to the difference between the heat of adsorption of solid in the first layer and the latent heat of condensation of the solid in successive layers. The adsorption constants (b), (Y,) and correlation coefficient (r), were estimated by linear regression analysis where r = 0.999, Slope = (b - l)/Y,b = 63.962, intercept = l/Y,b = 8.870, the adsorption constant b = 8.21, therefore, Y, = 0.0137 (g g-l). The double log plot for the adsorption isotherm of indomethacin onto magaldrate in phosphate buffer at pH
‘.*/
Fig. 5. Effect of temperature magaldrate antacid powder.
7.4 and 37°C (Fig. following Freundlich =
1.7
1.8
1.9 log Ce
2.0
2.1
66
4) was expressed model:
I 2.2
Fig. 4. Linear form of Freundlich adsorption isotherm of indomethacin onto magaldrate from phosphate buffer pH 7.4 at 37°C.
of test drugs onto
according
KC’/”
to the
(3a)
or its linear form: log ( x/m) = log k + (l/n)log
C
(3b)
Where (x/m> is the mass of solute (x) adsorbed per gram (m), of the adsorbent (mg g-l), (C) is the equilibrium solute concentration in the bulk (mg%), (k) and (l/n) are the Freundlich constants. K represents the drug amount adsorbed per unit weight of adsorbent at a unit drug concentration. The value of l/n is always less than unity and represents the drug amount adsorbed for a given concentration change. Larger l/n values reflect a greater fraction of adsorbed drug to unbound drug. This value, in turn, reflects the ratio of the rate of adsorption to desorption and the relative strength of the bond between the drug and the binding site [31]. The value of l/n is a dimensionless parameter, and decreases with increasing intensity of drug adsorption [32]. The adsorption constants and correlation coefficient (r) were estimated by linear regression analysis, where r = 0.999, K = 0.763 mg gg’, l/n = 0.515. Fig. 5 shows that varying the temperature over the
PH
1.6
60
on the extent of adsorption
Table 1 Effect of pH on the extent of adsorption antacid powder, at 37°C
0’
46
+ Pmpr.“QIOI HCI -+ *O~lyl..lloylloACM 4% Indomauoln
x/m
C
Y,b
36
Temperature%
I
0’
Y(C,-C)
25
of test drugs onto magaldrate
Amount of drug adsorbed (mg g-‘) A
B
C
4.5
17.5
20.2
_
5.2 6.5 7.4
18.3 20.7 21.9
15.1 11.7
13.9 11.3 10.3
A = propranolol
HCl, B = acetylsalicylic
acid, C = indomethacin.
O.M.N. Al Gohary /Pharmaceutics Table 2 Effect of electrolyte Electrolyte
concentration
cont. (M)
and valency
on the adsorption
Acta Heluetiae 70 (1995) 289-300
of test drugs onto magaldrate
powder 5.0% w/v,
0
0.01 0.05 0.1
MgCI,
A
B
C
A
B
C
10.5 10.0 9.1 7.2
11.7 10.5 9.9 9.7
10.3 8.9 7.0 6.1
10.5 9.8 8.1 6.0
11.7 8.6 6.3 4.3
10.3 8.3 5.9 3.4
4.7
4.5
7.2
4.7
4.5
7.2
pH of filtrate
HCl, B = acetylsalicylic
acid, C = indomethacin.
range 25-50°C had almost little effect on the extent of adsorption for the concentration studied, which revealed that the reaction is of van der Waals type. A trend indicating a decrease in the amount of drug adsorbed with increase in temperature was exhibited in all systems; indicating an exothermic process. The effect of pH over a range from 4.5-7.4 on the extent of adsorption of test drugs onto magaldrate powder was studied (Table 1). The adsorption of propranolol *HCl was increased as the pH increased. A competitive mechanism may exist involving hydronium ions, H30f, and the protonated amine. Since propranolol is a weak base, and exists predominantly as the cations below its pK, (9.231, as the pH rises, their concentration increased with respect to [H,O+]. At high pH values, the non-protonated forms of the drug which have low water solubility increase, hence, adsorption increases. Generally, pH and solubility effects act in concert (Lundelius’ rule). On the other hand, the equilibrium between the protonated and non-protonated forms of the drug is dynamic. At low pH, [H,O+] increases, causing dissolution of the antacid, thus, adsorption is suppressed due to reduction of active sites on the adsorbent. In addition, the system is flocculated at low pH, hence, adsorption decreases. Conversely, the adsorption of acetylsalicylic acid and indomethacin onto magaldrate was
Table 3 Effect of varying
the volume of test mixture on the extent of adsorption
Volume
A, 20 30 40 50
of test drugs onto magaldrate
Amount adsorbed
(mg)
(mg g-‘1
B1
50 37.5 100 75.0 150 112.5 200 150.0 250 187.5 pH of filtrate
A,,, = propranolol.
decreased as the pH increased. Since acetylsalicylic acid is a weak acid (pK, = 3.49), therefore, at pH 4.5, the ratio of the concentration of unionised acetylsalicylic acid to acetylsalicylicylate anion is 1.259 X 10P4:1. Indomethacin, a weak acid (pK, = 4.5) will be mainly ionized at higher PH.. The same manner of interpretation of results for propranolol . HCl, applies for both acetylsalicylic acid and indomethacin. The results (Table 2) show that the amount of drug adsorbed was suppressed as the concentration and valency of the electrolyte was increased. This could be attributed to the neutralisation of a portion of the negative charge on antacid surface by adsorption of sodium or magnesium ions onto the surface, hence, the system flocculated. Increasing the concentration of electrolyte favours its preferential adsorption, in addition, the surface potential of the antacid would be decreased, leading to compression of the electrical double layer, consequently, the zeta potential would also be decreased, thus, adsorption decreased. Moreover, competition between sodium or magnesium ions with the test drugs for the available adsorption sites onto the antacid or for positions in the electrical double layer around the particles may suppress adsorption. Divalent magnesium ions caused reduction in adsorption than monovalent sodium ions at the same pH studied. Since
Drug amount
(ml)
10
at 37°C
Amount of drug adsorbed (mg g-’ ) NaCl
A = propranolol
293
Drug amount
A,
%
C,
A2
15 30 45 60 75
19.0 19.1 19.0 18.7 18.8 4.7
19.9 20.0 20.1 19.9 19.7 4.5
8.6 8.8 8.9 8.5 8.7 7.2
100 75 100 75 100 75 100 75 100 75 pH of filtrate
acid, C,,, = indomefhacin
Amount adsorbed
(mg)
C,
HCl, B,,, = acetysalicylic
powder (1.0 g), at 37°C
B2
(mg 8-l) c2
A2
B2
c2
30 30 30 30 30
20.3 18.2 14.3 13.1 10.4 4.7
23.8 20.1 16.2 13.3 10.4 4.5
11.2 8.7 7.5 5.3 4.9 7.2
294
O.M.N. AI Gohary/Pharmaceutica
Acta Heluetiae 70 (1995) 289-300
magnesium ions neutralised more of the negative charge on the antacid surface than sodium ions (Schulze-Hardy rule), consequently, the charge on the antacid is lowered and hence adsorption decreased. The sedimentation volume, which is the ratio between the ultimate volume of sediment and the initial volume of suspension, equals 1.1 and 1.0 for 0.1 M magnesium chloride and 0.1 M sodium chloride, respectively, which indicates that the system was best flocculated by magnesium ions than by sodium ions. This suppression could be also ascribed to the insulation of the groups on both the drug and the adsorbent by the oppositely charged ions of the electrolyte [ll]. Table 3 shows that the volume of the test mixture had a negligible effect on the extent of adsorption in case if the drug concentration was kept constant regardless of the volume of test mixture. On the other hand, when a fixed weight (mg) in different solvent volumes was used, the amount adsorbed was decreased as the volume of test mixture increased, this may be due to that the solvent molecules compete with the drug particles for the active sites available for adsorption onto the adsorbent surface. In addition, the extent of adsorption is dependent on the concentration of adsorbate, as shown in Figs. (l-4). Taking into account that the adsorbent concentration had a negligible effect on adsorption, since in all cases, 1.0 g was used regardless of the volume of test mixture. The effect of varying the concentration of the antacid on the extent of adsorption of test drugs was studied. Fig. 6 shows that as the antacid concentration was increased, the equilibrium concentration of test drugs decreased. This is due to the fact that as the adsorbent concentration increased, more active sites are made available for adsorption of a fixed amount of drug, hence, the equilibrium concentration becomes progressively less. This implies that the equilibrium concentration is dependent on the adsorbent concentration in the system. Regression analysis of data showed a straight line relationship. The coefficient of
Table 4 Desorption
of test drugs (mg%) from magaldrate
Drug (amount adsorbed before elution mg g- ‘1 Propranolol (18.22) Acetylsalicylic (19.2) Indomethacin (8.7)
Washes C
HCI 3
b a C
acid 3
b a
5
b a
C
Solvent 99.1 99.8 4.7-1.2 52.3 58.1 4.5 20.2 26.1 7.2
5.0% w/v,
loo-
*
60-m 0
2.6
1 6
I 7.6
10
12.6
Magaldrate concentration g% w/v - Pmpr.nolol nc, + Ao*lyl.‘lloyllo Mid + Indom.thwln Fig. 6. Effect of magaldrate antacid concentration of test drugs, at 37°C.
concentration
on the equilibrium
variation (r) is 0.997, 0.994 and 0.991 for propranolol . HCl, acetylsalicylic acid and indomethacin, respectively. Table 4 shows that the elution of the adsorbed drugs onto antacid surface by different elution media followed the sequence: 0.1 M HCl > 0.01 M HCl > 0.1 M MgCl, > 0.01 M MgCl, > 0.1 M NaCl > 0.01 M NaCl. In all cases, the amount recovered by 50 ml eluent was more than that by 20 ml. This may suggest a competitive mechanism between the elution media and the adsorbed drugs for the active sites available for adsorption onto the antacid surface. When 0.1 M HCl was the eluent, recovery of test drugs was greater than that by other media, this may be due to that desorption is pH dependent, on the other hand competitive mechanism may exist. In addition, the dissolution of antacid particles in acidic pH favoured desorption. Whilst indomethacin showed negligible amounts due to its poor solubility in acidic medium. When electrolytes were used as eluents, magnesium chloride was more efficient than sodium chloride in displacing drugs from the negatively charged antacid surface. In all cases the amount recovered represented the amount that was
by different elution media at 37°C
Washes
3
4
0.01 M HCI
0.1 M
0.01 M
MgCl,
MgCl,
0.1 M NaCl
0.01 M NaCl
99.1 99.8 4.7-1.2 55.0 59.4 4.5-1.2 _
82.2 85.5 4.7-2.5 49.3 51.3 4.5-2.5 _
7.2-1.2
7.2-2.5
24.0 26.0 4.7-6.3 19.3 20.5 4.5-6.3 37.4 39.5 7.1
23.1 25.1 4.7-6.3 16.5 17.6 4.5-6.3 32.2 34.3 7.1
22.0 24.1 4.7-6.2 18.1 18.7 4.5-6.2 32.1 33.5 7.1
16.0 17.9 4.7-6.2 15.2 15.9 4.5-6.2 30.5 31.9 7.1
0.1 M HCl
3
a pH of filtrate, b eluent 50 ml, ’ eluent 20 ml, solvent, i.e. specified
solvent of each drug.
O.M.N. Al Gohary/Pharmaceutica
Acta Helvetiae 70 (1995) 289-300
295
adsorbed through weak van der Waals attraction forces or electrostatic hydrogen bonds. Whilst the amount retained onto the antacid surface unrecovered could be attributed to forces of ionic type, moreover, the weak basic antacid neutralised the acidic acetylsalicylic acid or indomethacin partially. Propranolol . HCl being a weak base, its desorption from different adsorbents was reported to be pH dependent and via competitive mechanism [33]. 3.2. Characterization of the interaction drate antacid and the test drugs
between
(1)
magal-
Melting point determinations showed that pure dry powders of propranolol HCl, acetylsalicylic acid, indomethacin and magaldrate antacid, melted at 164.1, 137.1, 160.5 and above 300°C respectively, whereas the physical mixtures or the residues obtained from adsorption studies showed melting points above 300°C for all systems. This is indicative of an interaction between test drugs and magaldrate antacid. Study on the effect of antacid on the test drugs was performed using DSC analysis to compare the effect of antacid on the thermal behaviour of the drugs. Fig. 7 shows that propranolol . HCl dry powder thermogram has an endotherm at 163.86 (trace 3). When 1.0 gram of the antacid powder was equilibrated for 1 h at 37°C with 200 mg of drug in 20 ml 0.1 M HCl, the residue obtained (trace 5) showed clear disappearance of the above reference peak, and the peak at 72.1 of trace 2 as well, instead, a small lump is reported. There is a clear interaction of the physical mixture (trace 41, a clear lowering and shift of propranolol . HCl endotherm to 153.48 instead of 163.86, is reported. Fig. 8 shows that the endotherm of acetylsalicylic acid dry powder which was clear at 143.11 (trace 3) became smaller and shorter in trace 4 and shifted to 137.65 which suggests an interaction due to physical mixture. The equilibration for 1.5 h at 37°C of 1.0 g of the antacid with 20 ml of acetate buffer, pH 4.5 containing 150 mg of the drug, resulted in complete disappearance of the drug reference peak (trace 5). Fig. 9 shows that indomethacin has a clear endotherm at 162.24 (trace 3).In the physical mixture (trace 4), a clear lowering of the intensity of the peak is shown and its value is reported to be 159.92. There was a complete disappearence of the drug reference peak (trace 5) after the treatment of 1.0 g of antacid with 40 mg of the drug in 20 ml of phosphate buffer (pH 7.4) and equilibration for 0.5 h at 37°C. The results obtained from Figs. 7-9 are suggestive of clear interaction between the antacid magaldrate and the test drugs.
(3)
(4)
(5)
OQ 40
70
100
I30
I6a
190
220
250
280
310
340
TemperatureoC Fig. 7. DSC thermograms of: 1. Antacid dry powder alone; 2. Antacid powder in 0.1 M HCl; 3. Propranolol HCl dry powder alone; 4. Physical mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 and 2.
This interaction is further investigated using IR spectroscopy. Fig. 10 revealed that in the reference propranolol HCl (trace 3), the -NH- stretching at 3000 cm-’ and the
296
O.M.N. AI Gohary/Pharmaceutica
-OH peaks at 3300 cm-’ are very clear and could be taken as reference in comparing any change that may occur. The fingerprint region is also indicative for each
Acta Helvetiae 70 (1995) 289-300
compound. In (trace 41, the -OH peak is clearly broadened which may be due to interaction in physical mixture because the fingerprint region is not altered. The amino
(1)
(1)
(2’
(2)
,,__/
(3)
(3)
(4) (4)
(5) 04
40
70
100
130
lb0
I90
no
1m
180
310
340
Temperature'C Fig. 8. DSC thermograms of: 1. Antacid dry powder alone; 2. Antacid powder in acetate buffer (pH 4.5); 3. Acetylsalicylic acid dry powder alone; 4. Physical mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 and 2.
m
40
70
100
130
160
190
:I0
293
so
310
3;
Tempereture'C Fig. 9. DSC thermograms oE 1. Antacid dry powder alone; 2. Antacid powder in phosphate buffer (pH 7.4); 3. Indomethacin dry powder alone; 4. Physical mixture of 1 and 2, ratio 1:l; 5. Residue obtained from adsorption of 3 and 2.
O.M.N. Al Gohary/Pharmaceutica
4m
,500
am
mm
leD0
1605
297
Acta Helvetiae 70 (1995) 289-300
%Lcm
Wavenumber
WN
600
bou
281
cm
Fig. 10. IR spectra of: 1. Antacid dry powder alone; 2. Antacid powder in 0.1 M HCl; 3. Propranolol 3, ratio 1:l; 5. Residue obtained from adsorption of 3 onto 2.
HCl dry powder alone; 4. Physical mixture of 1 and
O.M.N. Al Gohary / Pharmaceutics
298
Acta Helvetiae 70 (1995) 289-300
-’ of carbonyl of acid and O(C = O)-CH,, respectively Tt%ze 3). In the dry physical mixture (trace 4 ),;he -OH indicapeak is shifted to higher frequency at 3500 cm tive of some interaction. The possibility of interaction is illustrated by the disappearance of the clear-COOH and C = 0 peaks (trace 5). In Fig. 12, the chart of indomethacin alone (trace 3) and
and -OH peaks are shifted and broadened (trace 5). This could be attributed to chemical change or interaction. The fingerprint is completely altered due to change in the final product. The IR spectra of acetylsalicylic acid (Fig. 11) showed a clear absorbance at the region of 3000 cm-’ indicative of presence of -OH. Also two clear peaks at 1700, 1750
(1)
(2)
(3)
(4)
(5)
4000
3500
3000
2500
2000
1800
1600
1400
I200
wavenumber
1000
cm
HOO
600
400
a
200
-1
Fig. 11. IR spectra of: 1. Antacid dry powder alone; 2. Antacid powder in acetate buffer (pH 4.5); 3. Acetylsalicylic mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 onto 2.
acid dry powder alone; 4. Physical
O.M.N. Al Gohary /Pharmaceutics
Acta Heluetiae 70 (1995) 289-300
299
r (1)
)i
(2)
(3)
(4)
(5)
am
nm
Sam
1yD
ma
lam
ym
Km
Wavenumber
%xu
cm
Fig. 12. IR spectra of: 1. Antacid dry powder alone; 2. Antacid dry powder in phosphate mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 onto 2.
-T,,
IQI m"
buffer (pH 7.4); 3. Indomethacin
dry powder alone; 4. Physical
300
O.M.N. AI Gohary / Pharmaceutics Acta Helvetiae 70 (1995) 289-300
with antacid (trace 4) gave combination of both of them. The acid absorps at 1700 - -1720 cm-‘. There is another sharp peak at 1750 cm- ‘. The finger print is also a combination. Trace 5 showed clear interaction and disappearance of all peaks. In conclusion, until the clinical value of these results has been determined, it would be best to avoid administering magaldrate antacid simultaneously with test drugs where possible.
References [l] Lanza, F.L. and Sibley, C.M. (1987) Role of antacids in the management of disorders of the upper gastrointestinal tract review of clinical experience. Am. J. Gastrol. 82, 1223-1241. [2] Duric, Z., Jovanovic, M. and Razumenic, R. (1985) Dosage forms of antacids and interactions with other drugs. Arh. Farm. 35(6), 305318. [3] Karsh, J. (1990) Adverse reactions and interactions with aspirin. Considerations in the treatment of the elderly patient. Drug Saf. 5(5), 317-327. [4] Tuncel, M. and Bergisadi, N. (1992) In vitro adsorption of ciprofloxacin hydrochloride on various antacids. Pharmazie 47, 304-305. [5] Ismail, F.A., Khalafallah, N. and Khalil, S.A. (1987) Adsorption of ketoprofen and bumadizone calcium on aluminum-containing antacids and its effect on ketoprofen bioavailability in man. Int. J. Pharm. 34, 189-196. [6] El-Gamal, S., Boraie, N., and Naggar, V. (1986) An in vitro study of flurazepam on some antacids and excepients. Pharm. Ind. 48, 12071210. [7] Thoma, K. and Lieb, H. (1985) Untersuchungen zur Adsorption von kationischen amphilen Arzneistoffen an Antacida und Adsorbentien. Pharm. Acta. Helv. 60, 66-70. [8] Calis, S., Sumnu, M. and Hincal, A.A. (1986) In vitro adsorption of propranolol hydrochloride by various antacids. Drug. Dev. Ind. Pharm. 12(13), 1833-1845. [9] Naggar, V. (1985) In vitro interaction of antacids and antidiarrheals with allupurinol and probenecid. Ind. J. Pharm. Sci. 47, 151-153. [lo] Khalil, S., Mortada, L. and El-Khawas, M. (1984) Decreased bioavailability of ampicillin and amoxycillin in presence of kaolin. Int. J. Pharm. 19, 233-238. [ll] Khalil, S., Naggar, V., Zaghloul, I. and Ismail, A. (1984) In vitro anticoagulant-antacid interactions Int. J. Pharm. 19, 307-321. [12] Naggar, V. and Khalil, S. (1980) In vitro antidiabetics-antacid interactions. Pharmazie 35 (7), 412-416. [13] Singh, A. and Mital, H. (1982) Desorption of atropine sulphate and hyoscyamine hydrobromide by various antacids in vitro. Pharmazie 37, 31-32. [14] Thoma, K., and Lieb, H. (1984) Untersuchungen zur Adsorption von kationischen amphilen Arzneistoffen an Antacida und Adsorbentien. Pharm. Acta Helv. 59, 322-328. [15] Iwuagwu, M.A. and Aloko, KS. (1992) Adsorption of paracetamol and chloroquine phosphate by some antacids. J. Pharm. Pharmcol. 44 (8), 655-658.
[I61 Guglar,
R. and Allgayer, H. (1990) Effects of antacids on the clinical pharmacokinetics of drugs. Clin. Pharmacokinet. 18, 210219. ]171 Thorpe, J.E., Baker, N. and Bromet-Petit, M. (1990) Effect of oral antacid administration on the kinetics of fluconazole. Antimicrob. Agents Chemother. 34, 2032-2033. [181Frost, R., Lasseter, K., Noe, A., Shamblen, E. and Lettieri, J. (1992) Effects of aluminum hydroxide and calcium carbonate antacids on the bioavailability of ciprofloxacin. Antimicrob. Agents Chemother. 36, 830-832. [191Flor, S., Guay, D., Opsahl, J., Tack, K. and Matzke, G. (1990) Effects of magnesium-aluminum hydroxide and calcium carbonate antacids on bioavailability of ofloxacin. Antimicrob. Agents Chemother. 34, 2436-2438. ml Caille, G., Du-Souich, P., Vezina. M., Pollock, S.R. and Stalker. D.J. (1989) Pharmacokinetic interaction between flurbiprofen and antacids in healty volunteers. Biopharm. Drug Dispos. 10, 607-615. ]211 Crema, F., Parini, J., Visconti, M., and Perucca, E. (1987) Effect0 degliantiacidi sulla biodisponibilita de1 carbenoxolone. Farmaco. [Pratt.] 42 (12), 357-364. ]221 Day, R., Lam, S., Paull, P., and Wade, D. (1988) Effect of food and various antacids on the absorption of tenoxicam. Br. J. Clin. Pharmacol. 24, 323-328. ]23] Cacek, A.T. (1986) Review of alterations in oral phenytoin bioavailability associated with formulation, antacids and food. Ther. Drug. Monit. 8(2), 166-171. of oral 500mg ]24] Ifan, A. and Welling, P.G. (1986) Pharmacokinetics penicillamine: effect of antacids on absorption. Biopharm. Drug Dispos. 7, 401-405. 1251 . _ Itthipanichpong, C., Sirivongs, P., Wittayalertpunya, S. and Chaiyos, N. (1992) The effect of antacid on aspirin pharmacokinetics in healthy Thai volunteers. Drug Metabol. Drug interaction. 1@3), 213-228. [Xl Miners, J. (1989) Drug interactions involving aspirin (acetylsalicylic acid) and salicylic acid. Clin. Pharmacokinet. 17(S), 327-344. [271Gaspari, F., Vigano, G., Tocatelli, M. and Remuzzi, G. (1988) Influence of antacid administration on aspirin absorption in patients with chronic renal failure on maintenance hemodialysis. Am. J. Kidney Dis. 11, 338-342. 1281Langmuir, I. (1918) The adsorption of gases on plane surfaces of glass, mica and platinum. J. Am. Chem.. Sot. 40, 1361-1403. [291Giles, C., Mac Ewans, T. Nakhwa, S. and Smith, D. (1960) Studies in adsorption. Part XI. A system of classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in measurment of specific surface areas of solids. J. Chem. Sot. 3973-3993. ]301 Brunauer, S., Emmet, P.H., and Teller, E. (1938) Adsorption of gases in multimolecular layers. J. Am. Chem. Sot. 60, 309-319. [31] Gangian, F., Cutie, A. and Jochsberger T. (1980) In vitro adsorption studies of cimetidine. J. Pharm. Sci. 69, 352-353. 1321 Abe, I., Kamaya, H. and Ueda, I. (1990) Adsorption of local anesthetics on activated carbon: Freundlich adsorption isotherm. J. Pharm. Sci. 79, 354-358. [33] Al Gohary, O., Lyall, J. and Murray, J.B. (1987) Adsorption of antihypertensives by suspensoids. Part I. The adsorption of propranolo1 hydrochloride by attapulgite, charcoal, kaolin and magnesium trisilicate. Pharm. Acta Helv. 62, 66-72.
Pharmaceutics Acta Helvetiae 70 (1995) 289-300
Determination and characterization of the interaction between magaldrate antacid and some drugs: Part II Omaimah M.N. Al Gohary King Saud University, College of Pharmacy, Department of Pharmaceutics,
*
P.O. Box 7633, Riyadh 11472, Saudi Arabia
Abstract The interaction between magaldrate antacid and some drugs was investigated. Adsorption studies showed that magaldrate exhibited the highest adsorptive capacity for propranolol hydrochloride, intermediate for acetylsalicylic acid and the lowest for indomethacin. The adsorption was in accordance to the Langmuir, BET, and Freundlich model, respectively. An endothermic process was exhibited for all systems. Adsorption was pH dependent and was suppressed as the concentration and valency of the electrolyte were increased. The volume of the test mixture had a negligible effect on the extent of adsorption when the drug concentration was kept constant, whereas, when a fixed weight of drug in different solvent volumes was used, the adsorption was decreased as the volume increased. The equilibrium concentration of test drugs decreased as the antacid concentration was increased. Elution studies show that the drug recovery from the antacid surface was dependent on the volume and pH of elution media. A competitive mechanism was involved in the process and elution followed the sequence: 0.1 M HCl > 0.01 M HCl > 0.1 M MgCl, > 0.01 M MgCl, > 0.1 M NaCl > 0.01 M NaCl. The data obtained from adsorption-desorption process revealed that physical adsorption as well as chemisorption were existing. Melting point determinations (MP), differential scanning calorimetry analysis (DSC) and infra-red spectroscopy (IR), suggested a clear interaction between magaldrate antacid and the test drugs which could be attributed to chemical changes in test drugs. Keywords:
Drug interaction; Magaldrate; Propranolol; Acetylsalicylic acid; Indomethacin; Adsorption
1. Introduction Antacids play a role in the management of gastrointestinal disorders which may occur as side effects during antihypertensive, antiarrhythmic or antirheumatic therapy [ 11. The potential interactions of non-prescription antacids with various prescription drugs were discussed [2,3]. The antacid-drug interactions were attributed to different mechanisms, adsorption being one of the most important mechanisms reported in literature [4-151. Effects of antacids on clinical pharmacokinetics of drugs were studied by many workers ([16-271). Previous in vitro investigations showed that the dissolution was adversely affected for propranolol . HCl from
* Corresponding author. Tel./Fax:
+966 1 4644283.
0031-6865/9.5/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDI 0031-6865(95)00033-X
inderal tablets, acetylsalicylic acid from aspirin tablets and indomethacin from indocid capsules when magaldrate was present in the dissolution media (part I of this study). The purpose of the present work was to determine and characterize the possible interaction between magaldrate and the previously mentioned drugs by utilization of adsorptiondesorption studies as well as MP, DSC, and IR determinations.
2. Experimental
procedures
2.1. Chemicals Magaldrate powder (KTT, Osaka, Japan), propranolol hydrochloride powder (ICI, Macclesfield, UK), acetylsalicylic acid powder (Riedel-de Ha&, Hannover, Germany)
290
O.M.N.AI Gohary / Pharmaceutics Acta Heluetiae 70 (1995) 289-300
and indomethacin powder (Scherer, Seelze, Germany). All other reagents were of pure analytical grade and were supplied by BDH (Poole, UK)
and the pH of filtrates was recorded. The amount adsorbed was measured and compared to that adsorbed from a similar initial drug concentration at the same temperature and pH but in the absence of electrolytes.
2.2. Methods 2.2.1. Adsorption studies In vitro adsorption studies were carried out by shaking 1.0 g of magaldrate antacid powder with 20 ml of a solution containing different concentrations of propranolol HCl in 0.1 M HCl (pH 1.2) acetylsalicylic acid in acetate buffer (pH 4.5) and indomethacin in phosphate buffer (pH 7.4) Suspensions were shaken in a thermostatically controlled water bath (Karl Kolb type D 6.72, Bonn, Germany) at 80 rpm and 37 k 05°C until equilibrium was attained (60, 90 and 30 min), respectively. Filtered in sintered glass filters grade 5. The pH values of solutions and suspensions were measured using a Tacussel electronic pH meter (PHN81, Villeurbanne, France) with an accuracy of kO.05 pH units, The drug concentration remaining in the filtrate was determined spectrophotometritally using a Pye Unicam SP8800 spectrophotometer (Cambridge, UK) at 290 nm, 265 nm and 318 nm, respectively. Blank experiments showed no interference. Three runs were obtained and the results averaged. 2.2.2. Effect of temperature on the extent of adsorption of test drugs onto magaldrate antacid powder Adsorption experiments were carried out under the same conditions mentioned above, at various temperatures: 25, 37, 44 and 50°C using a fixed known concentration of each drug, the pH was kept constant for each system in all experiments. The amount of drug adsorbed was calculated and the results were compared. 2.2.3. Effect of pH on the extent of adsorption of test drugs onto magaldrate antacid powder Adsorption experiments were carried out as previously mentioned, at a pH range 2.5-7.4, and 37 + 0.5”C, using a fixed known concentration of each drug . The amount of drug adsorbed was calculated and the results were compared. 2.2.4. Effect of the presence of monovalent and divalent electrolytes on the extent of adsorption of test drugs onto magaldrate antacid powder 1.0 g of antacid powder was suspended in an aliquot (5 ml) of 0.01, 0.05 and 0.1 M sodium chloride or magnesium chloride solutions, then a fixed known amount of test drug was dissolved in 15 ml of the specified solvent and was added to that suspension. Adsorption experiments were carried out as previously mentioned at 37 f 0.5”C
2.2.5. Effect of varying the volume of test solution on the extent of adsorption of test drugs onto magaldrate antacid powder Adsorption experiments were carried out under the same conditions mentioned above, at 37 + 0.5”C. A fixed weight of antacid (1.0 g) and a fixed concentration (% w/v) of each test drug obtained from different drugsolvent ratio were used, fixed weight in different solvent volumes was used as well. Then, the amount of drug adsorbed(mg g-‘) was calculated and the results were compared. 2.2.6. Effect of varying the concentration of magaldrate antacid powder (g% w / v) on the equilibrium concentration of drug in solution, (Ce mg%) Different weights of antacid powder (0.5, 1.0, 1.5 and 2.0 g) were suspended in 20 ml of the specified solvent of each test drug containing a fixed known concentration of the drug and equilibrated at 37 k 0.5”C as mentioned under adsorption studies. Then, the amount of drug remaining in solution at equilibrium (Ce mg%) was determined spectrophotometrically at the specified wavelength. The effect of the antacid powder concentration (2.5, 5.0, 7.5 and 10 g% w/v) on the extent of adsorption was shown by plotting the equilibrium concentration Ce against the antacid powder concentration. 2.2.7. Desorption studies The residue obtained from adsorption process was collected and dried at room temperature away from light, shaken at 80 rpm and 37 f 0.5”C for 10 min with 20 or 50 ml of different elution media, then treated as mentioned under adsorption studies. Successive washings were carried out until no drug was detected in the filtrate. 2.2.8. Characterization of the interaction between magaldrate antacid and the test drugs 1.0 g of antacid powder with 20 ml of the specified solvent containing 200 mg propranolol HCl, 150 mg acetylsalicylic acid and 40 mg indomethacin were treated as previously mentioned under adsorption studies. The residue obtained from adsorption experiments were collected, dried over calcium chloride in a desiccator for 24 h away from light, then subjected to the following tests: Melting point (MP) determination. Melting point apparatus (Mettler FP800, Mettler instruments AG, Griefensee, Switzerland) was used for MP determinations of the sam-
O.M.N. Al Gohary / Pharmaceutics
ples at a heating rate of 2”C/min, until the desirable temperature reached, then the rate was lowered to 0.2”C/min until complete melting occurred. The melting point recorded for each sample was the average of five readings. Differential scanning calorimetry (DSC) analysis. The DSC thermograms were recorded on a Perkin Elmer DSC-4 differential scanning calorimeter (Norwald, USA) calibrated with indium (99.999%). 2 mg of the samples were heated at a heating rate of 30”C/min over a temperature range of 30-350°C in closed aluminum pans under an argon purge. At least three determinations were used to calculate mean values and standard deviation. Injkared spectroscopy (IR) determination. IR spectra were taken on a Perkin Elmer IR spectrophotometer (IR783, Beacons field, UK) as KBr pellets. The samples:KBr ratio was 1:300 mg, and a pressure of 15 tons was used, spectra were recovered over a 4000-250 cm-’ range, with a scan speed of 300 cm-‘/min. Different samples of antacid dry powder alone or treated with the specified solvents, and powdered drugs alone or their physical mixture with antacid in ratio 1:l were also subjected to MP, DSC, and IR determinations and the results were compared.
3. Results and discussion 3.1. Adsorption
studies
Fig. 1 shows that the adsorption of propranolol . HCl onto magaldrate in 0.1 M HCl at pH 4.7 and 37°C was in agreement with Langmuir adsorption isotherm [28]: x/m
= abC,/l
100
+ bC,
200
300
(la)
400
500 600 Ce mg%
700
800
go0
1000
Fig. 1. Langmuir adsorption isotherm of propranolol hydrochloride onto magaldrate from 0.1 M HCl, pH 1.2 at 37°C and pH 4.7. (-0-j C, against x/m (-W-I C, against C, /(x/m).
Acta Heluetiae 70 (1995) 289-300
291
0 00
I
50
100
150
200
250
300 350 Ce mg %
400
Fig. 2. BET adsorption isotherm of acetylsalicylic from acetate buffer pH 4.5 at 37°C.
450
500
550
600
acid onto magaldrate
or its linear form C,/(
x/m)
= l/ah
+ C,/a
(lb)
Where the slope = l/a and the intercept (c) = l/ah, CC,) is the equilibrium concentration of solute mg%, (x/ m) is the mass of solute (x) adsorbed per gram (m) of the adsorbent mg g-l, (b) is the enthalpy constant and (a) is the monolayer adsorptive capacity of antacid for the drug which may be calculated by extrapolating the plateau region of the curve back to the y - axis, the value of (a) equals 20.6 mg gg’. The adsorption constants (a), (b) and correlation coefficient (r) were estimated by linear regression analysis where r = 0.999, the value of (a) equals to the reciprocal of the slope, therefore, a = 21.739 mg g-‘, the value of b = l/at = 1.259 1 g-’ which is the affinity constant of the adsorption process. The data obtained from the adsorption of acetylsalicylic acid onto magaldrate in acetate buffer at pH 4.5 and 37°C were in accordance with BET adsorption isotherm (Fig. 2). Since a sigmoidal or S-shaped isotherm was obtained, thus indicating multilayer formation. The initial direction of curvature shows that adsorption becomes easier as the concentration rises. This isotherm assumed that the solute molecule was adsorbed onto fixed site as a single molecule and there was no lateral interaction between the molecules and that the heat of adsorption of a molecule to any layer other than the first was equal to heat of condensation [29]. The S curve usually appears when the solute molecule has a fairly large hydrophobic residue > C, and moderate intermolecular attraction which causes it to pack vertically in regular array in the adsorbed layers, and meets strong competition for substrate sites from molecules of the solvent [29].
292
O.M.N. Al Gohary/Pharmaceutica
Acta Helvetiae 70 (1995) 289-300
I
0’
30
.2
.1
3
.4
.6
.6
.7
.S
c/c,
Fig. 3. Linear plot of BET adsorption isotherm of aceytylsalicylic onto magaldrate from acetate buffer pH 4.5 at 37°C.
acid
Fig. 3 shows the straight line form of the BET adsorption isotherm [30]:
c
1
b-l
=Y,b+-.-
(2)
C0
40
where C, is the solubility when the solution is saturated (in g/100 ml), C the equilibrium concentration (g%), Y the mass of solid adsorbed at concentration C, Y, the mass of solid adsorbed per gram at monolayer capacity (g g-r ), and b the constant proportional to the difference between the heat of adsorption of solid in the first layer and the latent heat of condensation of the solid in successive layers. The adsorption constants (b), (Y,) and correlation coefficient (r), were estimated by linear regression analysis where r = 0.999, Slope = (b - l)/Y,b = 63.962, intercept = l/Y,b = 8.870, the adsorption constant b = 8.21, therefore, Y, = 0.0137 (g g-l). The double log plot for the adsorption isotherm of indomethacin onto magaldrate in phosphate buffer at pH
‘.*/
Fig. 5. Effect of temperature magaldrate antacid powder.
7.4 and 37°C (Fig. following Freundlich =
1.7
1.8
1.9 log Ce
2.0
2.1
66
4) was expressed model:
I 2.2
Fig. 4. Linear form of Freundlich adsorption isotherm of indomethacin onto magaldrate from phosphate buffer pH 7.4 at 37°C.
of test drugs onto
according
KC’/”
to the
(3a)
or its linear form: log ( x/m) = log k + (l/n)log
C
(3b)
Where (x/m> is the mass of solute (x) adsorbed per gram (m), of the adsorbent (mg g-l), (C) is the equilibrium solute concentration in the bulk (mg%), (k) and (l/n) are the Freundlich constants. K represents the drug amount adsorbed per unit weight of adsorbent at a unit drug concentration. The value of l/n is always less than unity and represents the drug amount adsorbed for a given concentration change. Larger l/n values reflect a greater fraction of adsorbed drug to unbound drug. This value, in turn, reflects the ratio of the rate of adsorption to desorption and the relative strength of the bond between the drug and the binding site [31]. The value of l/n is a dimensionless parameter, and decreases with increasing intensity of drug adsorption [32]. The adsorption constants and correlation coefficient (r) were estimated by linear regression analysis, where r = 0.999, K = 0.763 mg gg’, l/n = 0.515. Fig. 5 shows that varying the temperature over the
PH
1.6
60
on the extent of adsorption
Table 1 Effect of pH on the extent of adsorption antacid powder, at 37°C
0’
46
+ Pmpr.“QIOI HCI -+ *O~lyl..lloylloACM 4% Indomauoln
x/m
C
Y,b
36
Temperature%
I
0’
Y(C,-C)
25
of test drugs onto magaldrate
Amount of drug adsorbed (mg g-‘) A
B
C
4.5
17.5
20.2
_
5.2 6.5 7.4
18.3 20.7 21.9
15.1 11.7
13.9 11.3 10.3
A = propranolol
HCl, B = acetylsalicylic
acid, C = indomethacin.
O.M.N. Al Gohary /Pharmaceutics Table 2 Effect of electrolyte Electrolyte
concentration
cont. (M)
and valency
on the adsorption
Acta Heluetiae 70 (1995) 289-300
of test drugs onto magaldrate
powder 5.0% w/v,
0
0.01 0.05 0.1
MgCI,
A
B
C
A
B
C
10.5 10.0 9.1 7.2
11.7 10.5 9.9 9.7
10.3 8.9 7.0 6.1
10.5 9.8 8.1 6.0
11.7 8.6 6.3 4.3
10.3 8.3 5.9 3.4
4.7
4.5
7.2
4.7
4.5
7.2
pH of filtrate
HCl, B = acetylsalicylic
acid, C = indomethacin.
range 25-50°C had almost little effect on the extent of adsorption for the concentration studied, which revealed that the reaction is of van der Waals type. A trend indicating a decrease in the amount of drug adsorbed with increase in temperature was exhibited in all systems; indicating an exothermic process. The effect of pH over a range from 4.5-7.4 on the extent of adsorption of test drugs onto magaldrate powder was studied (Table 1). The adsorption of propranolol *HCl was increased as the pH increased. A competitive mechanism may exist involving hydronium ions, H30f, and the protonated amine. Since propranolol is a weak base, and exists predominantly as the cations below its pK, (9.231, as the pH rises, their concentration increased with respect to [H,O+]. At high pH values, the non-protonated forms of the drug which have low water solubility increase, hence, adsorption increases. Generally, pH and solubility effects act in concert (Lundelius’ rule). On the other hand, the equilibrium between the protonated and non-protonated forms of the drug is dynamic. At low pH, [H,O+] increases, causing dissolution of the antacid, thus, adsorption is suppressed due to reduction of active sites on the adsorbent. In addition, the system is flocculated at low pH, hence, adsorption decreases. Conversely, the adsorption of acetylsalicylic acid and indomethacin onto magaldrate was
Table 3 Effect of varying
the volume of test mixture on the extent of adsorption
Volume
A, 20 30 40 50
of test drugs onto magaldrate
Amount adsorbed
(mg)
(mg g-‘1
B1
50 37.5 100 75.0 150 112.5 200 150.0 250 187.5 pH of filtrate
A,,, = propranolol.
decreased as the pH increased. Since acetylsalicylic acid is a weak acid (pK, = 3.49), therefore, at pH 4.5, the ratio of the concentration of unionised acetylsalicylic acid to acetylsalicylicylate anion is 1.259 X 10P4:1. Indomethacin, a weak acid (pK, = 4.5) will be mainly ionized at higher PH.. The same manner of interpretation of results for propranolol . HCl, applies for both acetylsalicylic acid and indomethacin. The results (Table 2) show that the amount of drug adsorbed was suppressed as the concentration and valency of the electrolyte was increased. This could be attributed to the neutralisation of a portion of the negative charge on antacid surface by adsorption of sodium or magnesium ions onto the surface, hence, the system flocculated. Increasing the concentration of electrolyte favours its preferential adsorption, in addition, the surface potential of the antacid would be decreased, leading to compression of the electrical double layer, consequently, the zeta potential would also be decreased, thus, adsorption decreased. Moreover, competition between sodium or magnesium ions with the test drugs for the available adsorption sites onto the antacid or for positions in the electrical double layer around the particles may suppress adsorption. Divalent magnesium ions caused reduction in adsorption than monovalent sodium ions at the same pH studied. Since
Drug amount
(ml)
10
at 37°C
Amount of drug adsorbed (mg g-’ ) NaCl
A = propranolol
293
Drug amount
A,
%
C,
A2
15 30 45 60 75
19.0 19.1 19.0 18.7 18.8 4.7
19.9 20.0 20.1 19.9 19.7 4.5
8.6 8.8 8.9 8.5 8.7 7.2
100 75 100 75 100 75 100 75 100 75 pH of filtrate
acid, C,,, = indomefhacin
Amount adsorbed
(mg)
C,
HCl, B,,, = acetysalicylic
powder (1.0 g), at 37°C
B2
(mg 8-l) c2
A2
B2
c2
30 30 30 30 30
20.3 18.2 14.3 13.1 10.4 4.7
23.8 20.1 16.2 13.3 10.4 4.5
11.2 8.7 7.5 5.3 4.9 7.2
294
O.M.N. AI Gohary/Pharmaceutica
Acta Heluetiae 70 (1995) 289-300
magnesium ions neutralised more of the negative charge on the antacid surface than sodium ions (Schulze-Hardy rule), consequently, the charge on the antacid is lowered and hence adsorption decreased. The sedimentation volume, which is the ratio between the ultimate volume of sediment and the initial volume of suspension, equals 1.1 and 1.0 for 0.1 M magnesium chloride and 0.1 M sodium chloride, respectively, which indicates that the system was best flocculated by magnesium ions than by sodium ions. This suppression could be also ascribed to the insulation of the groups on both the drug and the adsorbent by the oppositely charged ions of the electrolyte [ll]. Table 3 shows that the volume of the test mixture had a negligible effect on the extent of adsorption in case if the drug concentration was kept constant regardless of the volume of test mixture. On the other hand, when a fixed weight (mg) in different solvent volumes was used, the amount adsorbed was decreased as the volume of test mixture increased, this may be due to that the solvent molecules compete with the drug particles for the active sites available for adsorption onto the adsorbent surface. In addition, the extent of adsorption is dependent on the concentration of adsorbate, as shown in Figs. (l-4). Taking into account that the adsorbent concentration had a negligible effect on adsorption, since in all cases, 1.0 g was used regardless of the volume of test mixture. The effect of varying the concentration of the antacid on the extent of adsorption of test drugs was studied. Fig. 6 shows that as the antacid concentration was increased, the equilibrium concentration of test drugs decreased. This is due to the fact that as the adsorbent concentration increased, more active sites are made available for adsorption of a fixed amount of drug, hence, the equilibrium concentration becomes progressively less. This implies that the equilibrium concentration is dependent on the adsorbent concentration in the system. Regression analysis of data showed a straight line relationship. The coefficient of
Table 4 Desorption
of test drugs (mg%) from magaldrate
Drug (amount adsorbed before elution mg g- ‘1 Propranolol (18.22) Acetylsalicylic (19.2) Indomethacin (8.7)
Washes C
HCI 3
b a C
acid 3
b a
5
b a
C
Solvent 99.1 99.8 4.7-1.2 52.3 58.1 4.5 20.2 26.1 7.2
5.0% w/v,
loo-
*
60-m 0
2.6
1 6
I 7.6
10
12.6
Magaldrate concentration g% w/v - Pmpr.nolol nc, + Ao*lyl.‘lloyllo Mid + Indom.thwln Fig. 6. Effect of magaldrate antacid concentration of test drugs, at 37°C.
concentration
on the equilibrium
variation (r) is 0.997, 0.994 and 0.991 for propranolol . HCl, acetylsalicylic acid and indomethacin, respectively. Table 4 shows that the elution of the adsorbed drugs onto antacid surface by different elution media followed the sequence: 0.1 M HCl > 0.01 M HCl > 0.1 M MgCl, > 0.01 M MgCl, > 0.1 M NaCl > 0.01 M NaCl. In all cases, the amount recovered by 50 ml eluent was more than that by 20 ml. This may suggest a competitive mechanism between the elution media and the adsorbed drugs for the active sites available for adsorption onto the antacid surface. When 0.1 M HCl was the eluent, recovery of test drugs was greater than that by other media, this may be due to that desorption is pH dependent, on the other hand competitive mechanism may exist. In addition, the dissolution of antacid particles in acidic pH favoured desorption. Whilst indomethacin showed negligible amounts due to its poor solubility in acidic medium. When electrolytes were used as eluents, magnesium chloride was more efficient than sodium chloride in displacing drugs from the negatively charged antacid surface. In all cases the amount recovered represented the amount that was
by different elution media at 37°C
Washes
3
4
0.01 M HCI
0.1 M
0.01 M
MgCl,
MgCl,
0.1 M NaCl
0.01 M NaCl
99.1 99.8 4.7-1.2 55.0 59.4 4.5-1.2 _
82.2 85.5 4.7-2.5 49.3 51.3 4.5-2.5 _
7.2-1.2
7.2-2.5
24.0 26.0 4.7-6.3 19.3 20.5 4.5-6.3 37.4 39.5 7.1
23.1 25.1 4.7-6.3 16.5 17.6 4.5-6.3 32.2 34.3 7.1
22.0 24.1 4.7-6.2 18.1 18.7 4.5-6.2 32.1 33.5 7.1
16.0 17.9 4.7-6.2 15.2 15.9 4.5-6.2 30.5 31.9 7.1
0.1 M HCl
3
a pH of filtrate, b eluent 50 ml, ’ eluent 20 ml, solvent, i.e. specified
solvent of each drug.
O.M.N. Al Gohary/Pharmaceutica
Acta Helvetiae 70 (1995) 289-300
295
adsorbed through weak van der Waals attraction forces or electrostatic hydrogen bonds. Whilst the amount retained onto the antacid surface unrecovered could be attributed to forces of ionic type, moreover, the weak basic antacid neutralised the acidic acetylsalicylic acid or indomethacin partially. Propranolol . HCl being a weak base, its desorption from different adsorbents was reported to be pH dependent and via competitive mechanism [33]. 3.2. Characterization of the interaction drate antacid and the test drugs
between
(1)
magal-
Melting point determinations showed that pure dry powders of propranolol HCl, acetylsalicylic acid, indomethacin and magaldrate antacid, melted at 164.1, 137.1, 160.5 and above 300°C respectively, whereas the physical mixtures or the residues obtained from adsorption studies showed melting points above 300°C for all systems. This is indicative of an interaction between test drugs and magaldrate antacid. Study on the effect of antacid on the test drugs was performed using DSC analysis to compare the effect of antacid on the thermal behaviour of the drugs. Fig. 7 shows that propranolol . HCl dry powder thermogram has an endotherm at 163.86 (trace 3). When 1.0 gram of the antacid powder was equilibrated for 1 h at 37°C with 200 mg of drug in 20 ml 0.1 M HCl, the residue obtained (trace 5) showed clear disappearance of the above reference peak, and the peak at 72.1 of trace 2 as well, instead, a small lump is reported. There is a clear interaction of the physical mixture (trace 41, a clear lowering and shift of propranolol . HCl endotherm to 153.48 instead of 163.86, is reported. Fig. 8 shows that the endotherm of acetylsalicylic acid dry powder which was clear at 143.11 (trace 3) became smaller and shorter in trace 4 and shifted to 137.65 which suggests an interaction due to physical mixture. The equilibration for 1.5 h at 37°C of 1.0 g of the antacid with 20 ml of acetate buffer, pH 4.5 containing 150 mg of the drug, resulted in complete disappearance of the drug reference peak (trace 5). Fig. 9 shows that indomethacin has a clear endotherm at 162.24 (trace 3).In the physical mixture (trace 4), a clear lowering of the intensity of the peak is shown and its value is reported to be 159.92. There was a complete disappearence of the drug reference peak (trace 5) after the treatment of 1.0 g of antacid with 40 mg of the drug in 20 ml of phosphate buffer (pH 7.4) and equilibration for 0.5 h at 37°C. The results obtained from Figs. 7-9 are suggestive of clear interaction between the antacid magaldrate and the test drugs.
(3)
(4)
(5)
OQ 40
70
100
I30
I6a
190
220
250
280
310
340
TemperatureoC Fig. 7. DSC thermograms of: 1. Antacid dry powder alone; 2. Antacid powder in 0.1 M HCl; 3. Propranolol HCl dry powder alone; 4. Physical mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 and 2.
This interaction is further investigated using IR spectroscopy. Fig. 10 revealed that in the reference propranolol HCl (trace 3), the -NH- stretching at 3000 cm-’ and the
296
O.M.N. AI Gohary/Pharmaceutica
-OH peaks at 3300 cm-’ are very clear and could be taken as reference in comparing any change that may occur. The fingerprint region is also indicative for each
Acta Helvetiae 70 (1995) 289-300
compound. In (trace 41, the -OH peak is clearly broadened which may be due to interaction in physical mixture because the fingerprint region is not altered. The amino
(1)
(1)
(2’
(2)
,,__/
(3)
(3)
(4) (4)
(5) 04
40
70
100
130
lb0
I90
no
1m
180
310
340
Temperature'C Fig. 8. DSC thermograms of: 1. Antacid dry powder alone; 2. Antacid powder in acetate buffer (pH 4.5); 3. Acetylsalicylic acid dry powder alone; 4. Physical mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 and 2.
m
40
70
100
130
160
190
:I0
293
so
310
3;
Tempereture'C Fig. 9. DSC thermograms oE 1. Antacid dry powder alone; 2. Antacid powder in phosphate buffer (pH 7.4); 3. Indomethacin dry powder alone; 4. Physical mixture of 1 and 2, ratio 1:l; 5. Residue obtained from adsorption of 3 and 2.
O.M.N. Al Gohary/Pharmaceutica
4m
,500
am
mm
leD0
1605
297
Acta Helvetiae 70 (1995) 289-300
%Lcm
Wavenumber
WN
600
bou
281
cm
Fig. 10. IR spectra of: 1. Antacid dry powder alone; 2. Antacid powder in 0.1 M HCl; 3. Propranolol 3, ratio 1:l; 5. Residue obtained from adsorption of 3 onto 2.
HCl dry powder alone; 4. Physical mixture of 1 and
O.M.N. Al Gohary / Pharmaceutics
298
Acta Helvetiae 70 (1995) 289-300
-’ of carbonyl of acid and O(C = O)-CH,, respectively Tt%ze 3). In the dry physical mixture (trace 4 ),;he -OH indicapeak is shifted to higher frequency at 3500 cm tive of some interaction. The possibility of interaction is illustrated by the disappearance of the clear-COOH and C = 0 peaks (trace 5). In Fig. 12, the chart of indomethacin alone (trace 3) and
and -OH peaks are shifted and broadened (trace 5). This could be attributed to chemical change or interaction. The fingerprint is completely altered due to change in the final product. The IR spectra of acetylsalicylic acid (Fig. 11) showed a clear absorbance at the region of 3000 cm-’ indicative of presence of -OH. Also two clear peaks at 1700, 1750
(1)
(2)
(3)
(4)
(5)
4000
3500
3000
2500
2000
1800
1600
1400
I200
wavenumber
1000
cm
HOO
600
400
a
200
-1
Fig. 11. IR spectra of: 1. Antacid dry powder alone; 2. Antacid powder in acetate buffer (pH 4.5); 3. Acetylsalicylic mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 onto 2.
acid dry powder alone; 4. Physical
O.M.N. Al Gohary /Pharmaceutics
Acta Heluetiae 70 (1995) 289-300
299
r (1)
)i
(2)
(3)
(4)
(5)
am
nm
Sam
1yD
ma
lam
ym
Km
Wavenumber
%xu
cm
Fig. 12. IR spectra of: 1. Antacid dry powder alone; 2. Antacid dry powder in phosphate mixture of 1 and 3, ratio 1:l; 5. Residue obtained from adsorption of 3 onto 2.
-T,,
IQI m"
buffer (pH 7.4); 3. Indomethacin
dry powder alone; 4. Physical
300
O.M.N. AI Gohary / Pharmaceutics Acta Helvetiae 70 (1995) 289-300
with antacid (trace 4) gave combination of both of them. The acid absorps at 1700 - -1720 cm-‘. There is another sharp peak at 1750 cm- ‘. The finger print is also a combination. Trace 5 showed clear interaction and disappearance of all peaks. In conclusion, until the clinical value of these results has been determined, it would be best to avoid administering magaldrate antacid simultaneously with test drugs where possible.
References [l] Lanza, F.L. and Sibley, C.M. (1987) Role of antacids in the management of disorders of the upper gastrointestinal tract review of clinical experience. Am. J. Gastrol. 82, 1223-1241. [2] Duric, Z., Jovanovic, M. and Razumenic, R. (1985) Dosage forms of antacids and interactions with other drugs. Arh. Farm. 35(6), 305318. [3] Karsh, J. (1990) Adverse reactions and interactions with aspirin. Considerations in the treatment of the elderly patient. Drug Saf. 5(5), 317-327. [4] Tuncel, M. and Bergisadi, N. (1992) In vitro adsorption of ciprofloxacin hydrochloride on various antacids. Pharmazie 47, 304-305. [5] Ismail, F.A., Khalafallah, N. and Khalil, S.A. (1987) Adsorption of ketoprofen and bumadizone calcium on aluminum-containing antacids and its effect on ketoprofen bioavailability in man. Int. J. Pharm. 34, 189-196. [6] El-Gamal, S., Boraie, N., and Naggar, V. (1986) An in vitro study of flurazepam on some antacids and excepients. Pharm. Ind. 48, 12071210. [7] Thoma, K. and Lieb, H. (1985) Untersuchungen zur Adsorption von kationischen amphilen Arzneistoffen an Antacida und Adsorbentien. Pharm. Acta. Helv. 60, 66-70. [8] Calis, S., Sumnu, M. and Hincal, A.A. (1986) In vitro adsorption of propranolol hydrochloride by various antacids. Drug. Dev. Ind. Pharm. 12(13), 1833-1845. [9] Naggar, V. (1985) In vitro interaction of antacids and antidiarrheals with allupurinol and probenecid. Ind. J. Pharm. Sci. 47, 151-153. [lo] Khalil, S., Mortada, L. and El-Khawas, M. (1984) Decreased bioavailability of ampicillin and amoxycillin in presence of kaolin. Int. J. Pharm. 19, 233-238. [ll] Khalil, S., Naggar, V., Zaghloul, I. and Ismail, A. (1984) In vitro anticoagulant-antacid interactions Int. J. Pharm. 19, 307-321. [12] Naggar, V. and Khalil, S. (1980) In vitro antidiabetics-antacid interactions. Pharmazie 35 (7), 412-416. [13] Singh, A. and Mital, H. (1982) Desorption of atropine sulphate and hyoscyamine hydrobromide by various antacids in vitro. Pharmazie 37, 31-32. [14] Thoma, K., and Lieb, H. (1984) Untersuchungen zur Adsorption von kationischen amphilen Arzneistoffen an Antacida und Adsorbentien. Pharm. Acta Helv. 59, 322-328. [15] Iwuagwu, M.A. and Aloko, KS. (1992) Adsorption of paracetamol and chloroquine phosphate by some antacids. J. Pharm. Pharmcol. 44 (8), 655-658.
[I61 Guglar,
R. and Allgayer, H. (1990) Effects of antacids on the clinical pharmacokinetics of drugs. Clin. Pharmacokinet. 18, 210219. ]171 Thorpe, J.E., Baker, N. and Bromet-Petit, M. (1990) Effect of oral antacid administration on the kinetics of fluconazole. Antimicrob. Agents Chemother. 34, 2032-2033. [181Frost, R., Lasseter, K., Noe, A., Shamblen, E. and Lettieri, J. (1992) Effects of aluminum hydroxide and calcium carbonate antacids on the bioavailability of ciprofloxacin. Antimicrob. Agents Chemother. 36, 830-832. [191Flor, S., Guay, D., Opsahl, J., Tack, K. and Matzke, G. (1990) Effects of magnesium-aluminum hydroxide and calcium carbonate antacids on bioavailability of ofloxacin. Antimicrob. Agents Chemother. 34, 2436-2438. ml Caille, G., Du-Souich, P., Vezina. M., Pollock, S.R. and Stalker. D.J. (1989) Pharmacokinetic interaction between flurbiprofen and antacids in healty volunteers. Biopharm. Drug Dispos. 10, 607-615. ]211 Crema, F., Parini, J., Visconti, M., and Perucca, E. (1987) Effect0 degliantiacidi sulla biodisponibilita de1 carbenoxolone. Farmaco. [Pratt.] 42 (12), 357-364. ]221 Day, R., Lam, S., Paull, P., and Wade, D. (1988) Effect of food and various antacids on the absorption of tenoxicam. Br. J. Clin. Pharmacol. 24, 323-328. ]23] Cacek, A.T. (1986) Review of alterations in oral phenytoin bioavailability associated with formulation, antacids and food. Ther. Drug. Monit. 8(2), 166-171. of oral 500mg ]24] Ifan, A. and Welling, P.G. (1986) Pharmacokinetics penicillamine: effect of antacids on absorption. Biopharm. Drug Dispos. 7, 401-405. 1251 . _ Itthipanichpong, C., Sirivongs, P., Wittayalertpunya, S. and Chaiyos, N. (1992) The effect of antacid on aspirin pharmacokinetics in healthy Thai volunteers. Drug Metabol. Drug interaction. 1@3), 213-228. [Xl Miners, J. (1989) Drug interactions involving aspirin (acetylsalicylic acid) and salicylic acid. Clin. Pharmacokinet. 17(S), 327-344. [271Gaspari, F., Vigano, G., Tocatelli, M. and Remuzzi, G. (1988) Influence of antacid administration on aspirin absorption in patients with chronic renal failure on maintenance hemodialysis. Am. J. Kidney Dis. 11, 338-342. 1281Langmuir, I. (1918) The adsorption of gases on plane surfaces of glass, mica and platinum. J. Am. Chem.. Sot. 40, 1361-1403. [291Giles, C., Mac Ewans, T. Nakhwa, S. and Smith, D. (1960) Studies in adsorption. Part XI. A system of classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in measurment of specific surface areas of solids. J. Chem. Sot. 3973-3993. ]301 Brunauer, S., Emmet, P.H., and Teller, E. (1938) Adsorption of gases in multimolecular layers. J. Am. Chem. Sot. 60, 309-319. [31] Gangian, F., Cutie, A. and Jochsberger T. (1980) In vitro adsorption studies of cimetidine. J. Pharm. Sci. 69, 352-353. 1321 Abe, I., Kamaya, H. and Ueda, I. (1990) Adsorption of local anesthetics on activated carbon: Freundlich adsorption isotherm. J. Pharm. Sci. 79, 354-358. [33] Al Gohary, O., Lyall, J. and Murray, J.B. (1987) Adsorption of antihypertensives by suspensoids. Part I. The adsorption of propranolo1 hydrochloride by attapulgite, charcoal, kaolin and magnesium trisilicate. Pharm. Acta Helv. 62, 66-72.