Evaluation of a Potential Interaction Between Erythromycin and Glyburide in Diabetic Volunteers Joseph C. Fleishaker, PhD, and J. Paul Phillips, BS The effects of erythromycin on the pharmacokinetics and pharmacodynamics of glyburide were evaluated in 12 patients with non-insulin-dependent diabetes mellitus (fasting blood glucose levels 140-280 mg/dL), who received 4 days of treatment with erythromycin base (333 mg administered orally every 8 hr] and a control treatment in a randomized crossover design; 5 mg glyburide was administered on day 4 of each study period. Serum glyburide concentrations were determined by high-performance liquid chromatography. Peak serum glyburide concentrations were increased by 18%, and mean time to peak glyburide concentrations (TmaJ decreased from 4.9 to 3.0 hours during erythromycin treatment; only the difference in Tmax was statistically signipcant. No signijiccnt effects on glyburide clearance were observed. No signljiccnt differences in glucose clearance after carbohydrate loads were observed between erythromycin + glyburide and glyburide treatments. These data show that oral erythromycin base treatment does not affect glyburide metabolism but does affect the rate of glyburide absorption. This effect may be mediated by the stimulation of gastric motility by erythromycin. The clinical slgnfjicnnce of the effects of erythromycin on glyburide kinetics appear to be minimal, based on the determinations of serum glucose concentrations.
lyburide, a second-generation sulfonylurea hyG poglycemic agent, is used for the treatment of non-insulin-dependent diabetes mellitus. Glyburide is eliminated primarily in humans by hepatic oxidation via the cytochrome P-450 system.' The primary metabolites are 4-trans-hydroxyglyburide and 3-cis-hydroxyglyburide; both compounds are apparently much less active than the parent compound." Erythromycin is a macrolide antibiotic used in the treatment of gram-positive infections. Erythromycin, as well as other macrolide antibiotics, inhibits drug metabolism via cytochrome P-450.3 •4 Examples of compounds that are affected are triazolam," carbamazepine," and theophylline," The mechanism for this inhibition appears to be enzyme induction followed by irreversible binding of a reactive metabolite to the enzyme." Since erythromycin inhibits metabolism by the cytochrome P-450 system, concomitant administration of erythromycin and glyburide may result in decreased glyburide clearance levels, leading to higher serum glyburide concentrations and exagFrom the Clinical Pharmacokinetics Unit. The Upjohn ComPany. Kalamazoo, Michigan. Address for reprints: Joseph C. Fleishaker. PhD. Clinical Pharmacokinetics Unit. 7215-24-2, The Upjohn Company, Kalamazoo. MI49007.
J elln Pharmacol 1991;31:259-262
gerated hypoglycemic effects. This study assessed the effects of oral erythromycin on glyburide pharmacokinetics and pharmacodynamics in the population where this interaction will most likely be of clinical importance-adult subjects with non-insulin-dependent diabetes mellitus. METHODS
The clinical portion of the study was conducted at the Arkansas Research Medical Testing Center in Little Rock, Arkansas. Twelve otherwise healthy diabetic patients (4 men, 8 women], between the ages of 36 and 65 were selected. 'Patients weighed between 63 and 95 kg. All subjects underwent physical examinations and provided complete medicalliistories. Routine clinical chemistry analysis, hematology tests, and urinalysis tests Were given; results of these tests were within the range expected for diabetic subjects. Patients who had been receiving hypoglycemic agents had their medication withdrawn 14 days before the start of the study and during the study. Fasting blood glucose levels (off antidiabetic medications) were between 140 and 280 mg/dll, Patients received no known enzyme inducers 30 lpays before the start of the study and no alcohol 7 ,\:lays before the study. The study was approved by the I
259
FLEISHAKER AND PHILLIPS
local institutional review board; subjects provided written informed consent before they were enrolled in the study. Subjects were randomly assigned to treatment sequences. Subjects in group 1 received 4 days of erythromycin base treatment (E-MYCIN tablets, The Upjohn Company, Kalamazoo, MI) administered as 333-mg tablets at 8:00 AM, 1:00 PM, and 8:00 PM, during the first treatment phase; group 2 patients received erythromycin treatment during the second treatment phase. Treatment phases were separated by 7 days. All subjects received carbohydrate tolerance tests at 8:00 AM on days 3 and 4 of each study phase. The carbohydrate load consisted of 12 oz of SUSTACAL liquid (Mead Johnson, Evansville, IN). Subjects received single 5-mg oral doses of glyburide (MICRONASE tablets, The Upjohn Company, Kalamazoo, MI) at 7:00 AM on day 4 of each phase, after an overnight fast. Subjects remained in the clinic for the duration of the study. Blood samples (3 mL) for glucose analysis were collected on days 3 and 4 of each phase before the SUSTACAL liquid administration and at 0.5,1,2,3, and 4 hours after glucose loading. Samples were allowed to clot, and serum was harvested and frozen at -20°C until analyzed. Glucose concentrations were determined by the hexokinase method (Centrifichem, Baker Instruments Corporation, Allentown, PAl. Blood samples (7 mL) for glyburide analysis were collected before dosing on day 4 of each phase and at 1.0, 1.5, 2.0, 3, 4, 5, 6, 8, 10, 12, 14, 16, 20, and 24 hours after glyburide administration. Serum was harvested and stored frozen at -20°C until analyzed. Glyburide concentrations were determined by a specific high-performance liquid chromatographic method.v'" Gylyburide standard curves were linear over the range of 2 to 20 and 20 to 100 ng/rnl., The between-day coefficient of variation for quality control samples was less than 6.9°A». Glyburide pharmacokinetic parameters were determined by noncompartmental techniques." The terminal elimination rate constant (Kel) was determined by simple linear regression of the terminal portion of the log-concentration time profile. Halflife (t'l2) was calculated as 0.693/Kel • Area under the curve (AUC) was determined by trapezoidal rule up to the last time point at which a measurable concentration was observed and extrapolated to infinity. Apparent oral clearance (Clo) was calculated as dose/AUC. Apparent volume of distribution was described as Clo/K el • Maximum glyburide serum concentration (CmaJ and the times at which they occurred (T maJ were determined graphically. Analysis of variance was done with group (sequence of erythromycin administration), subject nested within group, glyburide or erythromycin + glybur260 • J elln Pharmacol 1991;31:259-262
ide treatment, and study phase used as model parameters. Serum glucose AUC from 0 to 4 hours (AUCQ-4) was determined by trapezoidal rule. Serum glucose levels Cmax, and Tmax were determined graphically. Analysis of variance, using the above model, was used to assess the effect of erythromycin treatment on glucose tolerance tests in the absence and presence of glyburide treatment. RESULTS
Mean glyburide serum concentration-time curves are shown in Figure 1; glyburide pharmacokinetic parameters are summarized in Table I. Glyburide oral clearance was lower after erythromycin pretreatment, but the decrease was less than 10% and was not significant (P = .19). Slight increases in Cmax and t'l2 were also observed after erythromycin treatment. Tmax decreased from 4.9 to 3.0 hours after erythromycin pretreatment; this difference was significant (P = .0187). Glucose concentrations after glyburide treatment and carbohydrate challenge are shown in Figure 2. Glucose disposition parameters are shown in Table II. Plasma glucose concentrations were significantly lower after the administration of glyburide (day 4 of the study phase) than they were on day 3 of each treatment phase. On day 3 of the treatment phase, a slight, but statistically significant, difference between erythromycin and control treatments in glucose AUC levels was observed, in the absence of glyburide. On day 4 (glyburide administration), there were no significant differences in glucose disposition parameters between erythromycin and control treatments.
200
gE
160
s
120
~ e
80
j
40
I
"
~
._A-A
A,
,,
...
'
....
....
......
0 0
4
8
12
16
20
24
Time (h) - e - ERY + Glyb
- A-
GlybLrjde
Figure 1. Serum glyburide concentrations after administration of glyburide during the erythromycin treatment (treatment A) or control period (treatment B).
ERYTHROMYCIN AND GLYBURIDE IN DIABETICS
TABLE I
TABLE II
Mean (±SD) Pharmacoklnetlc Parameters for Glyburlde Administered Alone (Treatment B) or Concomitantly With Erythromycin (Treatment A) In Diabetic Subjects
Mean (±SD) Glucose Disposition Parameters on Day 3 (No Glyburlde) and Study Day 4 (Glyburlde Treatment Day) of Each Study Phase During Erythromycin Treatment (Treatment A) and During Control Treatment (Treatment B)
B
A
Treatment
1213 (520) 4.9 (2.0) 179 (82.8) 4.9· (2.5) 0.11 (0.07) 8.3 (3.9) 50.3 (16.5)
1330 (542) 4.4 (2.0) 211 (62.8) 3.0· (1.7) 0.09 (0.04) 9.2 (4.0) 50.9 (14.8)
AUC (ng hr/rnt) Clo (L/hr) Cmu (ng/mL)
Tmax (hr) Kel (hr- 1)
tl/2 (hr)a Vdarea/F(L)
• Significantly different at P < .05 by analysis of variance.
Treatment
A
AUC 0-4 (mg hr/dL)
1111··t (422) 332 (102) 1.54· (0.58)
B
Day 3
emax (mg/dt) Tmax (hr)
1183·'* (483) 362* (141) 1.05· (0.50) Day 4
AUC 0-4 (mg hr/dL) Cmu (mg/dL)
Tmax (hr)
895t (375) 299 (101) 1.33 (0.78)
983* (503) 301* (115) 0.92 (0.47)
• Significantly among treatments different at P < .05 by analysis of variance;
t Significantly different within erythromycin treatment between days 3 and 4 at P < .001 by analysis of variance; f.Significantly different within control treatment between 3 and 4 at P < .01 by analysis of variance.
DISCUSSION
The macrolide antibiotics have been observed to inhibit the metabolism of several compounds that are oxidized via the cytochrome P-450 system. Troleandomycin is uniform in its effects on drug metabolism; it decreases the clearance of theophylline, methylprednisolone, and antipyrine." Erythromycin decreases the clearance of triazolam, carbamazepine, theophylline, and methylprednisolone.v":" However, phenytoin clearance appeared to be unaf-
300
I
!
J
270 240 210 180 150 0
2
3
4
Time
lyburide, a second-generation sulfonylurea hyG poglycemic agent, is used for the treatment of non-insulin-dependent diabetes mellitus. Glyburide is eliminated primarily in humans by hepatic oxidation via the cytochrome P-450 system.' The primary metabolites are 4-trans-hydroxyglyburide and 3-cis-hydroxyglyburide; both compounds are apparently much less active than the parent compound." Erythromycin is a macrolide antibiotic used in the treatment of gram-positive infections. Erythromycin, as well as other macrolide antibiotics, inhibits drug metabolism via cytochrome P-450.3 •4 Examples of compounds that are affected are triazolam," carbamazepine," and theophylline," The mechanism for this inhibition appears to be enzyme induction followed by irreversible binding of a reactive metabolite to the enzyme." Since erythromycin inhibits metabolism by the cytochrome P-450 system, concomitant administration of erythromycin and glyburide may result in decreased glyburide clearance levels, leading to higher serum glyburide concentrations and exagFrom the Clinical Pharmacokinetics Unit. The Upjohn ComPany. Kalamazoo, Michigan. Address for reprints: Joseph C. Fleishaker. PhD. Clinical Pharmacokinetics Unit. 7215-24-2, The Upjohn Company, Kalamazoo. MI49007.
J elln Pharmacol 1991;31:259-262
gerated hypoglycemic effects. This study assessed the effects of oral erythromycin on glyburide pharmacokinetics and pharmacodynamics in the population where this interaction will most likely be of clinical importance-adult subjects with non-insulin-dependent diabetes mellitus. METHODS
The clinical portion of the study was conducted at the Arkansas Research Medical Testing Center in Little Rock, Arkansas. Twelve otherwise healthy diabetic patients (4 men, 8 women], between the ages of 36 and 65 were selected. 'Patients weighed between 63 and 95 kg. All subjects underwent physical examinations and provided complete medicalliistories. Routine clinical chemistry analysis, hematology tests, and urinalysis tests Were given; results of these tests were within the range expected for diabetic subjects. Patients who had been receiving hypoglycemic agents had their medication withdrawn 14 days before the start of the study and during the study. Fasting blood glucose levels (off antidiabetic medications) were between 140 and 280 mg/dll, Patients received no known enzyme inducers 30 lpays before the start of the study and no alcohol 7 ,\:lays before the study. The study was approved by the I
259
FLEISHAKER AND PHILLIPS
local institutional review board; subjects provided written informed consent before they were enrolled in the study. Subjects were randomly assigned to treatment sequences. Subjects in group 1 received 4 days of erythromycin base treatment (E-MYCIN tablets, The Upjohn Company, Kalamazoo, MI) administered as 333-mg tablets at 8:00 AM, 1:00 PM, and 8:00 PM, during the first treatment phase; group 2 patients received erythromycin treatment during the second treatment phase. Treatment phases were separated by 7 days. All subjects received carbohydrate tolerance tests at 8:00 AM on days 3 and 4 of each study phase. The carbohydrate load consisted of 12 oz of SUSTACAL liquid (Mead Johnson, Evansville, IN). Subjects received single 5-mg oral doses of glyburide (MICRONASE tablets, The Upjohn Company, Kalamazoo, MI) at 7:00 AM on day 4 of each phase, after an overnight fast. Subjects remained in the clinic for the duration of the study. Blood samples (3 mL) for glucose analysis were collected on days 3 and 4 of each phase before the SUSTACAL liquid administration and at 0.5,1,2,3, and 4 hours after glucose loading. Samples were allowed to clot, and serum was harvested and frozen at -20°C until analyzed. Glucose concentrations were determined by the hexokinase method (Centrifichem, Baker Instruments Corporation, Allentown, PAl. Blood samples (7 mL) for glyburide analysis were collected before dosing on day 4 of each phase and at 1.0, 1.5, 2.0, 3, 4, 5, 6, 8, 10, 12, 14, 16, 20, and 24 hours after glyburide administration. Serum was harvested and stored frozen at -20°C until analyzed. Glyburide concentrations were determined by a specific high-performance liquid chromatographic method.v'" Gylyburide standard curves were linear over the range of 2 to 20 and 20 to 100 ng/rnl., The between-day coefficient of variation for quality control samples was less than 6.9°A». Glyburide pharmacokinetic parameters were determined by noncompartmental techniques." The terminal elimination rate constant (Kel) was determined by simple linear regression of the terminal portion of the log-concentration time profile. Halflife (t'l2) was calculated as 0.693/Kel • Area under the curve (AUC) was determined by trapezoidal rule up to the last time point at which a measurable concentration was observed and extrapolated to infinity. Apparent oral clearance (Clo) was calculated as dose/AUC. Apparent volume of distribution was described as Clo/K el • Maximum glyburide serum concentration (CmaJ and the times at which they occurred (T maJ were determined graphically. Analysis of variance was done with group (sequence of erythromycin administration), subject nested within group, glyburide or erythromycin + glybur260 • J elln Pharmacol 1991;31:259-262
ide treatment, and study phase used as model parameters. Serum glucose AUC from 0 to 4 hours (AUCQ-4) was determined by trapezoidal rule. Serum glucose levels Cmax, and Tmax were determined graphically. Analysis of variance, using the above model, was used to assess the effect of erythromycin treatment on glucose tolerance tests in the absence and presence of glyburide treatment. RESULTS
Mean glyburide serum concentration-time curves are shown in Figure 1; glyburide pharmacokinetic parameters are summarized in Table I. Glyburide oral clearance was lower after erythromycin pretreatment, but the decrease was less than 10% and was not significant (P = .19). Slight increases in Cmax and t'l2 were also observed after erythromycin treatment. Tmax decreased from 4.9 to 3.0 hours after erythromycin pretreatment; this difference was significant (P = .0187). Glucose concentrations after glyburide treatment and carbohydrate challenge are shown in Figure 2. Glucose disposition parameters are shown in Table II. Plasma glucose concentrations were significantly lower after the administration of glyburide (day 4 of the study phase) than they were on day 3 of each treatment phase. On day 3 of the treatment phase, a slight, but statistically significant, difference between erythromycin and control treatments in glucose AUC levels was observed, in the absence of glyburide. On day 4 (glyburide administration), there were no significant differences in glucose disposition parameters between erythromycin and control treatments.
200
gE
160
s
120
~ e
80
j
40
I
"
~
._A-A
A,
,,
...
'
....
....
......
0 0
4
8
12
16
20
24
Time (h) - e - ERY + Glyb
- A-
GlybLrjde
Figure 1. Serum glyburide concentrations after administration of glyburide during the erythromycin treatment (treatment A) or control period (treatment B).
ERYTHROMYCIN AND GLYBURIDE IN DIABETICS
TABLE I
TABLE II
Mean (±SD) Pharmacoklnetlc Parameters for Glyburlde Administered Alone (Treatment B) or Concomitantly With Erythromycin (Treatment A) In Diabetic Subjects
Mean (±SD) Glucose Disposition Parameters on Day 3 (No Glyburlde) and Study Day 4 (Glyburlde Treatment Day) of Each Study Phase During Erythromycin Treatment (Treatment A) and During Control Treatment (Treatment B)
B
A
Treatment
1213 (520) 4.9 (2.0) 179 (82.8) 4.9· (2.5) 0.11 (0.07) 8.3 (3.9) 50.3 (16.5)
1330 (542) 4.4 (2.0) 211 (62.8) 3.0· (1.7) 0.09 (0.04) 9.2 (4.0) 50.9 (14.8)
AUC (ng hr/rnt) Clo (L/hr) Cmu (ng/mL)
Tmax (hr) Kel (hr- 1)
tl/2 (hr)a Vdarea/F(L)
• Significantly different at P < .05 by analysis of variance.
Treatment
A
AUC 0-4 (mg hr/dL)
1111··t (422) 332 (102) 1.54· (0.58)
B
Day 3
emax (mg/dt) Tmax (hr)
1183·'* (483) 362* (141) 1.05· (0.50) Day 4
AUC 0-4 (mg hr/dL) Cmu (mg/dL)
Tmax (hr)
895t (375) 299 (101) 1.33 (0.78)
983* (503) 301* (115) 0.92 (0.47)
• Significantly among treatments different at P < .05 by analysis of variance;
t Significantly different within erythromycin treatment between days 3 and 4 at P < .001 by analysis of variance; f.Significantly different within control treatment between 3 and 4 at P < .01 by analysis of variance.
DISCUSSION
The macrolide antibiotics have been observed to inhibit the metabolism of several compounds that are oxidized via the cytochrome P-450 system. Troleandomycin is uniform in its effects on drug metabolism; it decreases the clearance of theophylline, methylprednisolone, and antipyrine." Erythromycin decreases the clearance of triazolam, carbamazepine, theophylline, and methylprednisolone.v":" However, phenytoin clearance appeared to be unaf-
300
I
!
J
270 240 210 180 150 0
2
3
4
Time
