Enprofylline Disposition in the Presence and Absence of Amoxycillin Or Erythromycin

Br. J. clin. Pharmac. (1987), 24, 57-61

Enprofylline disposition in the presence and absence of amoxycillin or erythromycin

DANIEL S. SITAR1'3 5, FRED Y. AOKI" 23'4'5, DARYL J. HOBAN4, KARL-GUNNAR HIDINGER6, PATRICK R. MONTGOMERY' 3'5 & PAUL A. MITENKO" 3'5 Sections of 'Clinical Pharmacology and 2Infectious Diseases, Departments of 3Medicine, 4Medical Microbiology and 5Pharmacology and Therapeutics, University of Manitoba, Winnipeg, Manitoba and 6Astra Pharmaceuticals Canada Ltd, Mississauga, Ontario, Canada

1 The kinetic disposition of a novel xanthine bronchodilator, enprofylline, was determined in young healthy male volunteers in the presence and absence of amoxycillin or erythromycin. These data were compared to those derived from a similar study of theophylline disposition in the presence and absence of erythromycin. 2 Erythromycin inhibited theophylline disposition only in those subjects in whom the control kinetic study was done after antibiotic ingestion, but the effect was modest. Erythromycin had no effect on enprofylline disposition. 3 Amoxycillin reduced the renal clearance of enprofylline, but the change was not statistically significant. Keywords enprofylline erythromycin theophylline amoxycillin drug interactions

Introduction Enprofylline, a novel xanthine drug whose enprofylline disposition is affected by concurrent mechanism of action is claimed not to involve antibiotic administration. inhibition of phosphodiesterase, is mainly excreted unmetabolized by the kidney (Borga et al., 1983). This is in contrast to theophylline Methods whose elimination is largely dependent on hepatic metabolism (Ogilvie, 1978). It has been reported Thirty-two young, healthy male volunteers par- that theophylline clearance is reduced during ticipated in this study after being informed of concurrent administration of macrolide anti- its nature and purpose and providing signed biotics such as erythromycin and troleandomycin. consent. This investigation was approved by the The duration of therapy appears to be important Faculty Committee on the Use of Human Sub- to demonstrate the interaction (Weinberger et jects in Research at the University of Manitoba. al., 1977; Pfeifer et al., 1979; Prince et al., 1981; Two subjects withdrew from the study before its Renton et al., 1981; Zarowitz et al., 1981). It has completion. One withdrew due to a rash which also been reported that renal clearance of enpro- we attributed to amoxycillin. The other with- fylline is impaired by concurrent ingestion of drew when he developed otitis media prior to probenecid (Borga et al., 1986). It is probable antibiotic treatment. Demographic characteristics that enprofylline and antibiotics will be pre- are presented in Table 1 for the thirty subjects scribed together. Therefore, the study described who completed the investigation. All subjects below was undertaken to determine whether had not smoked for at least 3 months, and twenty- Correspondence: Dr Daniel S. Sitar, Clinical Pharmacology Section, University of Manitoba, 770 Bannatyne Avenue, Winnipeg, Manitoba R3E 0W3, Canada 57 58 Daniel S. Sitar et al.

Table 1 Characteristics of young male volunteers for each of the three groups in the study Age Height Weight Treatment group (years) (cm) (kg) Enprofylline/amoxycillin 26 ± 4 174 ± 8 74.4 ± 10.0 Enprofylline/erythromycin 26 ± 3 173 ± 9 68.7 ± 9.8 Theophylline/erythromycin 24 ± 3 181 ± 6 74.7 ± 7.4 There were 10 subjects in each group. Data are presented as means ± s.d. five of them had never smoked. Good health was In order to assess compliance, blood samples confirmed by history, physical examination and were taken into heparinized tubes on the mornings the following tests: haemoglobin, total white of days 3 and 8 prior to ingestion of the first daily blood cell and platelet concentration, serum dose of antibiotic and a predose urine specimen albumin, total protein, lactic acid dehydrogenase, was requested on day 10 of antibiotic treatment glutamateoxaloacetate transaminase, alkaline prior to the xanthine infusion. As well, residual phosphatase, creatinine and bilirubin concen- drug doses were counted on the morning of day trations, and urinalysis and 24 h creatinine 10 of antibiotic ingestion. The urine specimens clearance. were frozen until assayed for antibiotic concen- Subjects who had ingested any drug during the tration. week preceding the first study day or who had Blood samples (14 ml) were taken from an a known hypersensitivity to xanthines, erythro- indwelling intravenous cannula in the opposite mycin, penicillins or cephalosporins were ex- forearm to the infusion site after discarding the cluded. No alcohol ingestion was allowed within first 3 ml portion. The cannula was kept patent 72 h of the first study day or throughout the with heparin in saline (10 u ml-). Samples were study. taken just before infusion (t = 0), at tho end of The study was an open parallel design with the infusion (20 min), and hourly thereafter until three groups of equal size (10 subjects per group), 6 h after the start of infusion for enprofylline, enprofylline plus amoxycillin, enprofylline plus and until 8 h after the start of the infusion for erythromycin, and theophylline plus erythro- theophylline. The samples were split into equal mycin. Baseline pharmacokinetics of theophyl- portions, half of which was placed in a tube line (10 subjects) (aminophylline, Squibb Canada containing heparin and the remainder into a Inc, 5 mg kg-1 as theophylline) or enprofylline tube containing potassium oxalate. Plasma was (20 subjects) (AB Draco, 1.5 mg kg-') were separated by centrifugation, frozen and stored at determined after a 20 min intravenous drug -20° C until analysed for drug content. Com- infusion. This test was performed either 1 week plete urine collections were taken from the sub- before antibiotic treatment or 1 week after dis- jects administered enprofylline in 2 h aliquots continuation of antibiotic treatment. Equal for drug analysis to determine its renal clearance. numbers of subjects were randomly assigned to All subjects were asked to void just before each both tim.e periods to avoid an order effect. In intravenous dose. Aliquots of all urine specimens those subjects receiving enprofylline as the test were frozen at -20° C until analysed for drug xanthine, antibiotic treatment was randomly concentration. assigned so that half of the subjects ingested Plasma samples were analysed for enprofylline amoxycillin 250 mg (Ayerst Laboratories) four and theophylline and urine samples were analysed times daily and the remainder ingested erythro- for enprofylline by high pressure liquid chroma- mycin base 250 mg (Abbott Laboratories) four tography (Borga et al., 1983). The lower limit of times daily. All subjects receiving theophylline detection for enprofylline and theophylline was as the test xanthine ingested erythromycin base 0.05 mg l-1. Assay reproducibility is indicated 250 mg four times daily as the antibiotic treat- by the following coefficients of variation: 5.0% ment. On the morning of the tenth day of anti- for 2.0 mg I-1 and 13.0% for 0.5 mg I-1 enprofyl- biotic ingestion, subjects received an intravenous line in plasma (n = 49), 5.5% for 200 mg 1-1 and infusion of their allocated xanthine to determine 11.5% for 50 mg 1-1 enprofylline in urine (n = its pharmacokinetic disposition. Antibiotic doses 8), and 3.8% for 8.0 mg I-1 and 5.1% for 4.0 mg were continued until the end of the kinetic study 1-1 theophylline in plasma (n = 21). Plasma and on the tenth day. urine samples were analysed for erythromycin Antibiotics and xanthine disposition 59 and amoxycillin by a bioassay method with B. four or eight out of a total of40 prescribed doses. subtilis as the microorganism (Grove & Randall, Over the 6 h study period for enprofylline dis- 1955; Arret et al., 1971). The lower limit of position, 4.5 ± 2.6% of a dose of erythromycin detection for the bioassays was 0.5 mg 1-1 for and 68 ± 14% of a dose of amoxycillin were both antibiotics. excreted in the urine. Erythromycin and amoxy- The kinetic disposition of enprofylline and cillin were not detected prior to the morning theophylline was determined by a model in- dose in the plasma of all subjects on days 3 and 8 dependent method. The terminal elimination of antibiotic treatment, but were present in the rate constant (Xi) was determined by least predose urine of all subjects able to void on the squares regression of the logarithm of plasma morning of the tenth day of antibiotic treatment. drug concentration vs time. Area under the Kinetic constants for enprofylline and theo- plasma concentration vs time curve (AUC) was phylline disposition in the presence and absence determined by the trapezoidal method. Residual of antibiotics are presented in Table 3. There area to infinity was determined by dividing the was no effect of either antibiotic treatment on last measured plasma drug concentration by enprofylline disposition, and the decreased renal XA. Plasma clearance (CLp) was calculated by clearance of enprofylline in the presence of dividing drug dose by AUC. Apparent volume amoxycillin only approached statistical signifi- of distribution (Varea) was calculated as CLp cance (P = 0.15). Although renal clearance of divided by Xz. Renal drug clearance (CLR) of this drug was high, it was less than plasma clear- enprofylline was calculated as amount of un- ance for all study days and for subjects in both changed drug in urine from 0 to 6 h divided by groups receiving enprofylline (P < 0.01). The the AUC for the same time period (Greenblatt effect of erythromycin on theophylline clearance & Koch-Weser, 1975). approached significance (P < 0.10). We found a Data are tested for significant differences by highly variable effect of erythromycin on theo- analysis of variance or by paired t-tests where phylline clearance (-23 to +36%; 11 ± 16%). appropriate. Data are presented as means ± s.d. No other kinetic constants for theophylline dis- A P value - 0.05 with two-tailed analysis was position were affected by erythromycin treat- considered significant, excepting comparisons ment. involving potential effects of erythromycin on When the data were examined for an order theophylline disposition, and amoxycillin on effect, theophylline clearance (45 ± 13 vs 40 ± CLR of enprofylline, where inhibition was ex- 9 ml kg-' h-1) was decreased and plasma half- pected and a one-tail analysis was used. life (7.6 ± 2.1 vs 8.2 ± 2.1 h) was increased in volunteers in whom the control dose occurred 7 Results days after discontinuation of antibiotic (P = 0.04). No such effect was observed in subjects The three groups of subjects in this study were receiving enprofylline as the test xanthine. not different (Table 1). Compliance with prescribed antibiotics was good as determined by Discussion plasma and urine analysis (Table 2) and by tablet counts. Four subjects missed one, two subjects Our data for the kinetic disposition of enprofylline missed two and one subject each missed three, are not different from those previously published

Table 2 Plasma and urine concentrations of erythromycin and amoxycillin for the male subjects studied Plasma antibiotic concentration Day 10 urine antibiotic concentration (mg 1-') (mg 11 Treatment group Day 3 Day 8 Erythromycin Enprofylline/erythromycin 1.5 ± 0.8 (6)a 1.2 ± 0.6 (9) 9.2 ± 9.5 (10) Theophylline/erythromycin 1.1 ± 0.5 (5) 1.0 ± 0.4 (5) 6.9 ± 5.6 (10) Amoxycillin Enprofylline/amoxycillin 2.0 ± 0.5 (9) 2.0 ± 1.4 (9) 226 ± 239 (9)b a Numbers in parentheses represent the number of subjects in whom antibiotic concentrations were greater than the detectable limit of 0.5 mg l-1. b One subject was unable to void before the intravenous dose of enprofylline. Data are presented as means ± s.d. 60 Daniel S. Sitar et al.

Table 3 Kinetic disposition for enprofylline alone and on the tenth day of therapy with either amoxycillin or erythromycin compared with the kinetic disposition of theophylline alone and on the tenth day of therapy with erythromycin Treatment Enprofylline Enprofylline/amoxycillin X, (h-1) 0.439 ± 0.114 0.386 ± 0.102 t½, (h) 1.68 ± 0.45 1.89 ± 0.46 CLp (ml kg-' h-1) 245 ± 68 225 ± 41 CLR (ml kg-1 h-1) 184 ± 44a 159 ± 37a area (I kg-) 0.563 ± 0.101 0.604 ± 0.154 Enprofylline Enprofyllinelerythromycin X, (h-1) 0.415 ± 0.067 0.418 ± 0.126 t½ (h) 1.71 ± 0.26 1.80 ± 0.55 CLp(ml kg- h1) 240±51 227 ± 83 CLR (ml kg-' h-1) 162 ± 61a 169 ± 75a VarIa (I kg-1) 0.582 ± 0.108 0.550 ± 0.141 Theophylline Theophylline/erythromycin X, (h-1) 0.088 ± 0.021 0.079 ± 0.019 t½ (h) 8.27 ± 1.97 9.21 ± 2.19 CLp (ml kg- h1) 41 ± 11 36 ± 9 Varea (I kg-) 0.465 ± 0.034 0.458 ± 0.030 a P < 0.01 CLR compared with CLp for the same dose. Data are presented as means ± s.d.

(Borga et al., 1983). Our finding t41at erythro- The pronounced variability in the inhibitory mycin did not adversely affect the kinetic dis- effect of erythromycin on theophylline meta- position of enprofylline, a xanthine which is bolism has been observed previously (Prince et mostly excreted unchanged by the kidney al., 1981; Zarowitz etal., 1981). Our observation (> 80%), is not unexpected and supports the of an order effect for the erythromycin-theo- hypothesis that the effect of erythromycin is phylline drug interaction has also been previously primarily on hepatic disposition, although the reported and is consistent with the data of Zaro- mechanism remains to be elucidated (Mao & witz et al. (1981). We also feel that this drug Tardrew, 1965; Wilson & van Boxtel, 1978; interaction is not likely to be clinically important Zarowitz et al., 1981). in otherwise healthy young adults. However, it Serum erythromycin and amoxycillin concen- should be remembered that theophylline clear- trations observed in this study are consistent ance is decreased in the presence of acute res- with published data (Lawson etal., 1974; Wilson piratory viral illness (Chang et al., 1978). It may & van Boxtel, 1978). The lower limit of detec- be that the inhibitory effect of erythromycin on tion of these antibiotics in the serum, 0.5 mg l-1, theophylline clearance will be augmented by meant that we are unlikely to detect the anti- concurrent respiratory tract infections. This biotics in the sera of all subjects prior to the could be clinically important, and has not been morning dose, and this was in fact the case explored for enprofylline. (Table 2). The use of multiple measures of com- residual We thank Mrs Erica Robotham and MrJames Sherwin pliance, including dose count on day 10 for technical assistance in the xanthine analyses, Ms of antibiotic ingestion and estimation of urinary Evelyn Witwicki for the antibiotic assays, and Mrs antibiotic content, makes it unlikely that our Judy McLeod, RN for nursing assistance during the observations resulted from noncompliance with xanthine kinetic studies. This study was supported by a the antibiotic regimens. grant from Astra Pharmaceuticals Canada Ltd. Antibiotics and xanthine disposition 61 References

Arret, B., Johnson, D. P. & Kirschbaum, A. (1971). Ogilvie, R. I. (1978). Clinical pharmacokinetics of Outline of details for microbiological assays of theophylline. Clin. Pharmacokin., 3, 267-293. antibiotics, second revision. J. pharm. Sci., 60, Pfeifer, H. J., Greenblatt, D. J. & Freidman, P. 1689-1694. (1979). Effects of three antibiotics on theophylline Borga, O., Andersson, K.-E., Edholm, L.-E., Fager- kinetics. Clin. Pharmac. Ther., 26, 36-40. strom, P.-E., Lunnell, E. & Persson, C. G. (1983). Prince, R. A., Wing, D. S., Weinberger, M. M., Enprofylline kinetics in healthy subjects after single Hendeles, L. S. & Riegelman, S. (1981). Effect of doses. Clin. Pharmac. Ther., 34, 799-804. erythromycin on theophylline kinetics. J. Allergy Borga, O., Larsson, R. & Lunell, E. (1986). Effects of clin. Immunol., 68, 427-431. probenecid on enprofylline kinetics in man. Eur. J. Renton, K. W., Gray, J. D. & Huna, 0. R. (1981). clin. Pharmac., 30, 221-223. Depression of theophylline elimination by erythro- Chang, K. C., Bell, T. D., Lauer, B. A. & Chai, H. mycin. Clin. Pharmac. Ther., 30, 422-426. (1978). Altered theophylline pharmacokinetics Weinberger, M., Hudgel, D., Spector, S. & Chidsey, during acute respiratory viral illness. Lancet, i, C. (1977). Inhibition of theophylline clearance by 1132-1134. troleandomycin. J. Allergy clin. Immunol., 59, Greenblatt, D. J. & Koch-Weser, J. (1975). Drug 228-231. therapy. Clinical pharmacokinetics, part 1. New Wilson, J. T. & van Boxtel, C. J. (1978). Pharmaco- Engl. J. Med., 293, 702-705. kinetics of erythromycin in man. Antibiot. Chemo- Grove, D. C. & Randall, W. A. (1955). Assay methods ther., 25, 181-203. of antibiotics. A laboratory manual. New York: Zarowitz, B. J. M., Szefler, S. J. & Lasezkay, G. M. Medical Encyclopedia Inc. (1981). Effect of erythromycin base on theophyl- Lawson, D. H., Henderson, A. K. & McGeachy, line kinetics. Clin. Pharmac. Ther., 29, 601-605. R. R. (1974). Amoxycillin pharmacokinetic studies in normal subjects, patients with pernicious anaemia (Received 20 October 1986, and those with renal failure. Postgrad. med. J., 50, accepted 12 February 1987) 500-503. Mao, J. C. H. & Tardrew, P. L. (1965). Demethylation of erythromycins by rabbit tissues in vitro. Bio- chem. Pharmac., 14, 1049-1058.