Introduction

Top Abstract Introduction Methods Results Discussion References

Theophylline (1,3-dimethylxanthine) is a bronchodilator frequently used in the treatment of asthma and chronic obstructive pulmonary disease. In humans, its metabolism involves C8 hydroxylation to form 1,3DMU and N-demethylation at the 1 and 3 positions to yield 3MX and 1MX, respectively (Grygiel and Birkett, 1980; Lelo et al., 1986). The demethylation pathways are catalyzed primarily by the CYP1A2 isozyme, whereas the formation of 1,3DMU is mediated by multiple CYP isozymes, including CYP1A2 and CYP2E1 (Gu et al., 1992; Sarkar et al., 1992). 1MX undergoes subsequent conversion to 1MU catalyzed by the cytosolic enzyme xanthine oxidase (Birkett et al., 1983). Because of these characteristics of metabolism, theophylline is a commonly used substrate probe of the CYP enzyme system. Many biological and environmental factors that modulate the expression and activities of CYP isozymes responsible for theophylline metabolism can alter the biotransformation of theophylline (Jusko et al., 1979), which has a narrow therapeutic index with potential serious toxicity.

Age and gender are important biological determinants of the pharmacokinetics of selected drugs that undergo hepatic oxidative metabolism mediated by CYP isozymes (Bonate 1991; Durnas et al., 1990). A previous study investigating the disposition of theophylline in relation to gender showed that the plasma clearance of theophylline and the formation clearance of theophylline metabolites are similar in young male and female nonsmokers (Loi et al., 1991). The volume of distribution of theophylline exhibits a modest gender-related difference. Other studies also reported that the elimination of theophylline does not differ between men and women (Jusko et al., 1979; Powell et al., 1977). Previous studies from our laboratory have demonstrated that in healthy male nonsmokers, age is associated with a reduction in the basal rate of theophylline metabolism (Crowley et al., 1988; Vestal et al., 1987). The plasma clearance of theophylline is ~30% lower in elderly male subjects than in young male subjects. Likewise, Jackson et al. (1989) found that in healthy subjects and in patients with asthma, there is a fall in theophylline clearance during late adult life (seventh through ninth decades). This age-dependent decrease in theophylline elimination is attributed to a decline in the formation of all theophylline metabolites (Crowley et al., 1988; Vestal et al., 1987). However, it is not known whether this age-related reduction in theophylline metabolism is influenced by gender because there has been no systematic investigation on the effects of both age and gender on theophylline biotransformation.

In addition to age and gender, drug interactions play an important role in modulating the metabolism of theophylline (Upton, 1991). It is well recognized that some commonly prescribed medications, such as cimetidine (Cusack et al., 1985; Reitberg et al., 1981; Vestal et al., 1983b) and ciprofloxacin (Nix et al., 1987; Schwartz et al., 1988; Wijnands et al., 1986) can impair the elimination of theophylline. Despite an extensive body of knowledge concerning the effect of a single inhibitor on theophylline metabolism, there is a paucity of data on the effect of the combined administration of two inhibitors on theophylline disposition. With cimetidine and ciprofloxacin used as model compounds, two studies in a small number of young subjects have shown disparate results concerning the combined inhibitory effects of these two drugs on theophylline elimination. Coadministration of therapeutic doses of cimetidine (400 mg every 12 hr) and ciprofloxacin (500 mg every 12 hr) produced a further decrease in inhibition of theophylline metabolism compared with the use of each agent alone (Loi et al., 1993), whereas the administration of a therapeutic dose of ciprofloxacin (500 mg every 12 hr) combined with a maximally inhibiting dose of cimetidine (600 mg every 6 hr) yielded an additional inhibitory effect on theophylline clearance compared with the use of ciprofloxacin alone but not compared with the use of cimetidine alone (Davis et al., 1992). This disparity may be attributed to differences in dosing regimens of the inhibitors used in these studies. However, there has not been a systematic investigation of the combined inhibition of theophylline metabolism by two agents in relation to age and gender.

Consumption of multiple medications is common among elderly patients, and this could predispose the elderly to an increased risk of drug interactions involving more than one inhibitor. A reduction in clearance of theophylline by enzyme inhibition could result in a higher plasma concentration, predisposing elderly patients in particular to an increased risk of theophylline toxicity. Shannon (1993) reported that age is an independent risk factor for chronic theophylline overmedication, and Shannon and Lovejoy (1990) reported that the elderly have a greater risk of serious adverse effects from theophylline intoxication than younger patients. Accordingly, the objectives of this study were to examine the disposition of theophylline in relation to age and gender and to investigate the individual and combined effects of cimetidine and ciprofloxacin, administered at therapeutic doses, on theophylline elimination and on the metabolic pathways of theophylline biotransformation in young and elderly healthy male and female nonsmokers.

	Methods

Top Abstract Introduction Methods Results Discussion References

Subjects. The subjects were eight young male nonsmokers aged 23 to 40 years, eight young female nonsmokers aged 22 to 33 years, eight elderly male nonsmokers aged 65 to 76 years and eight elderly female nonsmokers aged 65 to 78 years. All subjects were determined to be healthy on the basis of medical history, physical examination and laboratory screen that included a complete blood count, 12-channel chemistry screen, urinalysis and electrocardiogram. Women of childbearing potential were screened for pregnancy with a serum -HCG test and were required to use a reliable barrier contraceptive method throughout the study; young female volunteers were not taking oral contraceptives. Two subjects with idiopathic hypothyroidism received treatment with L-thyroxine, which rendered them euthyroid. Subjects abstained from the use of over-the-counter medications for the duration of the study.

Study protocol. The study protocol was approved by the Human Subjects Review Committee of the University of Washington and by the Research and Development Committee of the Boise VA Medical Center. All subjects gave written informed consent. The study lasted a total of 25 days. The study subjects were required to abstain from ingesting alcohol, dietary methylxanthines, charcoal-broiled meat and cruciferous vegetables, including Brussels sprouts, broccoli, cabbage and cauliflower, beginning 5 days before and continuing throughout the study period. Single-dose studies of theophylline pharmacokinetics were performed on days 1, 8, 15 and 22. After an overnight fast, subjects arrived at the clinic at ~7:00 A.M. After the subject assumed a recumbent position, an indwelling Teflon catheter with an obturator (Deseret Medical, Sandy, UT) was inserted in a forearm or antecubital vein for blood sampling. Theophylline (5 mg/kg) was administered intravenously over 30 min into a contralateral forearm vein. Blood samples (10 ml each) were obtained before the start of the infusion and at 10, 20, 30, 35 and 50 min and 1, 1.5, 3, 4, 6, 9, 12, 23, 24 and 48 hr after the start of the infusion. Samples were immediately centrifuged at 2000 rpm and 4°C for 10 min, and the separated plasma was stored at 20°C until analysis. An additional blood sample was obtained at 1 hr to determine plasma protein binding of theophylline. Subjects remained recumbent for 3 hr after the start of the infusion. Only minimal activity was allowed for the remainder of the pharmacokinetic study. Urine excreted between 0 and 72 hr was collected and acidified with boric acid to maintain pH 5.5. The total volume of the urine samples was measured, and aliquots were stored at 20°C until analysis.

Analytical methods. Plasma theophylline concentrations were measured by HPLC as previously described (Vestal et al., 1983a). The urinary concentrations of theophylline and its metabolites were determined by HPLC using the ion-pair gradient elution method of Muir et al. (1980) with -hydroxyethyl theophylline as the internal standard. As reported previously (Loi et al., 1993), cimetidine and ciprofloxacin do not interfere with the plasma and urine assays for theophylline and its metabolites. Although metabolites of cimetidine and ciprofloxacin were not available to allow evaluation of possible interference with the HPLC assay, base-line plasma and urine samples taken before each theophylline kinetic study did not show chromatographic peaks that interfered with the analytes of interest. Plasma protein binding of theophylline was determined through liquid scintillation spectrometry after equilibrium dialysis according to the method of Cusack et al. (1985). Plasma cimetidine concentrations were measured by HPLC according to the methods described by Mihaly et al. (1982) and Ziemniak et al. (1981). Plasma ciprofloxacin concentrations were determined by HPLC with fluorescence detection (Awni et al., 1987). The excitation and emission wavelengths were 278 and 470 nm, respectively. Separation was performed using a C₁₈ RadialPak column (5 mm I.D.). Difloxacin (Abbott Laboratory, Chicago, IL) was used as the internal standard. Assay sensitivities were 0.1, 0.05 and 0.05 µg/ml for plasma theophylline, cimetidine and ciprofloxacin, respectively, and 5 µg/ml for urine theophylline and its metabolites.

Instrumentation. The HPLC system (Millipore-Waters, Milford, MA) for analysis of plasma theophylline consisted of a WISP model 710B autosampler, a series 440 absorbance detector set at 280 nm and a model 510 solvent delivery system fitted with an RCM-100 module containing a C₁₈ RadialPak cartridge (5 mm I.D.). Data were recorded on a HP model 3390A integrator (Hewlett-Packard, Avondale, PA). Urinary theophylline and its metabolites were measured using two model 510 solvent delivery systems, a model 680 gradient controller, a model 440 detector set at 254 nm and a model 715 WISP autosampler. Separation was performed using an Ultrasphere ODS column (4.6 mm × 25 cm; Beckman Instruments, Inc., Fullerton, CA). The HPLC system for determining the plasma cimetidine concentrations included a model 710B WISP autosampler, a model 490 programmable multiwavelength detector set at 228 nm and a model 590 programmable solvent delivery module containing a Zorbax Sil column (4.6 mm I.D. × 25 cm; DuPont Instruments, Wilmington, DE). Data were recorded on a HP model 3390A integrator. Plasma ciprofloxacin concentrations were measured using a model 710B WISP autosampler and a model 510 solvent delivery system fitted with a RCM-100 module containing a C₁₈ RadialPak cartridge (5 mm I.D.). Data were recorded on a Waters model 740 integrator. Detection was made using a Kratos FS 970 fluorescence detector. Radioactivity was counted in a Beckman model LS7500 scintillation counter.

Drugs and reagents. Drugs used in the study were theophylline in 5% w/v dextrose (Kendall McGaw Laboratories, Inc., Irvine, CA), cimetidine (Tagamet) and ciprofloxacin. -Hydroxyethyl theophylline was obtained from Sigma Chemical Co. (St. Louis, MO). Difloxacin was a gift of Abbott Laboratory. SKF-92374, the internal standard for the cimetidine assay, was a gift from SmithKline and Beecham. All solvents were spectroscopy grade and were obtained from Burdick and Jackson (Muskegan, MI) or J.T. Baker Chemical Co. (Phillipsburg, NJ). Other chemicals were obtained from Sigma Chemical Co.

Pharmacokinetic analysis. Pharmacokinetic parameters of theophylline were calculated using standard model-independent methods. The terminal elimination rate constant () of theophylline was determined from nonlinear least-squares regression of the log plasma theophylline concentration for the terminal elimination phase. The plasma elimination half-life (t1/2) was calculated as: t1/2 = 0.693/_z. The AUC from time zero with extrapolation to infinity (last measured concentration divided by _z) [AUC(0)] was determined by the trapezoidal method. Total plasma clearance of theophylline (CL) was calculated according to the following relationship: CL = dose/AUC(0). The apparent volume of distribution at steady state (V_ss) for theophylline was calculated from the following: where k_o is the rate of drug infusion, and T is the infusion time. The AUC(012) values for cimetidine and ciprofloxacin after multiple-dose administration were determined by trapezoidal rule from time zero to 12 hr after dosing on the days of pharmacokinetic studies. In addition, for some analyses, the AUC(012) values for cimetidine and ciprofloxacin were normalized by the administered dose expressed as mg/kg of b.wt.: dose-normalized AUC(012) = AUC(012)/(dose/body weight). The formation clearance (CL_m) to theophylline metabolites (3MX, 1MU and 1,3DMU) was calculated from the equation CL_m = f_m·CL, where f_m is the molar fraction of individual theophylline metabolites excreted in urine. The renal clearance (CL_R) of theophylline was calculated from CL_R = f_e·CL, where f_e is the molar fraction of theophylline recovered in urine.

Statistical analysis. Data are expressed as mean ± S.E. Descriptive statistics, linear regression analysis and two-factor analysis of variance (age and gender) were performed to compare the demographic data and the pharmacokinetic parameters among groups. Where significant relationships were found, specific comparisons were made with Tukey's Studentized range test as appropriate. Comparison of percentage change in theophylline plasma clearance and elimination half-life and in formation clearance of theophylline metabolites within each group was made with two-way analysis of variance with Tukey's Studentized range test. A value of P  .05 was considered to be significant, and values of P > .10 are identified in the tables as nonsignificant (NS).

	Results

Top Abstract Introduction Methods Results Discussion References

The characteristics of the study subjects are described in table 1. Although there was no age difference in total body weight or calculated body surface area, both parameters were higher in the male than in the female subjects (P = .01 and P = .004, respectively). Serum albumin concentrations, which were clinically within normal limits in all groups, were lower in the elderly than in the young subjects (P = .002). This age-related difference was apparent only in the male group. Creatinine clearance, determined from 24-hr urinary excretion of creatinine, exhibited a significant age difference that was independent of gender when adjusted for body surface area. Elderly subjects had lower values of creatinine clearance compared with the young group (P = .001).

After 5 days of treatment with cimetidine and ciprofloxacin, individually and in combination, there were both age and gender differences in systemic exposure to these agents (table 2). When administered alone, the AUC(012) values of cimetidine and ciprofloxacin were 54% (P = .011) and 41% (P = .003) higher in elderly than in young subjects, respectively. In the female group, respective AUC(012) values of cimetidine and ciprofloxacin were 32% (P = .04) and 36% (P = .001) higher than in the male group. During the combined administration phase, systemic exposure to cimetidine and ciprofloxacin displayed similar age and gender differences compared with those observed when each drug was administered alone (table 2). However, when the AUC(012) values were normalized for the administered dose expressed on a mg/kg b.wt. basis, there was a significant effect of age, but not gender, on systemic exposure to cimetidine and ciprofloxacin. The AUC(012) values of cimetidine and ciprofloxacin during the combined administration phase were similar to those obtained when each drug was administered alone. The relationship between creatinine clearance and the AUC(012) values for each inhibitor is illustrated in figure 1. There was a significant negative correlation between renal function and systemic exposure to cimetidine and ciprofloxacin. A similar relationship was observed between creatinine clearance and dose-normalized AUC(012) values of cimetidine and ciprofloxacin.

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Relationship between observed creatinine clearance and inhibitor AUC (012) for (A) ciprofloxacin, (B) cimetidine, and (C and D) combined therapy in all patient groups.

The effects of age and gender on base-line theophylline pharmacokinetics and metabolic pathways are shown in tables 3 and 4, respectively. The elderly subjects had a slower rate of theophylline metabolism than the young subjects. Total plasma theophylline clearance in the elderly male and female groups was 20% and 24% lower, respectively, than the corresponding young subjects (P = .007). There was no significant gender difference in the base-line plasma clearance of theophylline. The volume of distribution of theophylline showed a significant gender difference. The young and elderly female subjects had a 10% and 22% lower volume of distribution, respectively, than in the corresponding male groups (P = .007). There was no age-related difference in the volume of distribution of theophylline. The elimination half-life of theophylline displayed both gender and age differences. In male subjects, the elimination half-life of theophylline was 16% longer than that in the female subjects (P = .047), whereas in the elderly, the elimination half-life was 26% longer than that in the young subjects (P = .004). Plasma protein binding of theophylline was lower in the elderly than in the young subjects, but there was no gender difference in this parameter. Consistent with the age-related decline in plasma clearance of theophylline, the base-line formation clearance to 3MX, 1MU and 1,3DMU in the elderly was 68% (P = .008), 75% (P = .05) and 83% (P = .012), respectively, of the mean base-line values observed in the young subjects. Gender differences did not influence the formation clearance to theophylline metabolites. The base-line renal clearance of theophylline was not affected by age or gender (table 4).

Treatment with cimetidine and ciprofloxacin, administered individually and in combination, significantly decreased the elimination of theophylline in all groups as indicated by a reduction in the plasma theophylline clearance and an increase in its elimination half-life (table 3). The combined treatment phase was associated with a further lowering of the plasma clearance and an enhanced prolongation of the elimination half-life of theophylline compared with either agent administered individually (table 3). The volume of distribution and plasma free fraction of theophylline were not altered by any of the treatment regimens in young female and elderly male groups. However, small but statistically significant changes occurred in the volume of distribution of theophylline in the young male group and in the volume of distribution and free fraction of theophylline in elderly women with the addition of enzyme inhibitors (table 3). The administration of cimetidine alone and ciprofloxacin alone produced a similar proportionate change in plasma theophylline clearance in both the young women and elderly men. In the young men and elderly women, the proportionate change in theophylline clearance was slightly greater after treatment with ciprofloxacin alone compared with cimetidine alone (fig. 2). In all groups, coadministration of cimetidine and ciprofloxacin resulted in a greater proportionate decline in the plasma clearance and a higher proportionate increase in the elimination half-life of theophylline compared with each inhibitor alone (table 5). However, there were no age or gender differences in the response to inhibition of theophylline elimination by cimetidine and ciprofloxacin given individually or in combination. There was no correlation between either the AUC or the dose-normalized AUC of cimetidine or ciprofloxacin and the percentage change in theophylline clearance.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Proportionate change in plasma theophylline elimination (A) half-life and (B) clearance after treatment with cimetidine and ciprofloxacin. YM, young males; YF, young females; EM, elderly males; EF, elderly females.

The effects of cimetidine and ciprofloxacin on the renal clearance of theophylline and on the formation clearance to theophylline metabolites are summarized in table 4. The mean urinary recovery of theophylline and its metabolites was 82.4%. In all groups, administration of cimetidine alone and ciprofloxacin alone reduced the formation of all theophylline metabolites compared with base line. Concomitant administration of cimetidine and ciprofloxacin further decreased formation clearance to 3MX, 1MU and 1,3DMU compared with each agent alone. In the young male, elderly male and elderly female subjects, after treatment with cimetidine and ciprofloxacin, the renal clearance of theophylline tended to be lower than the base-line values, although the difference did not reach statistical significance. In the young female group, the renal clearance of theophylline was decreased by all treatments.

The effects of cimetidine and ciprofloxacin on the proportionate change (expressed as a percentage of base line) in individual pathways of theophylline metabolism are shown in table 5 and figure 3. Treatment with cimetidine alone produced a similar percentage reduction in the formation clearance to 3MX, 1MU and 1,3DMU in the young male group, whereas in the other three groups of subjects, the proportionate decline in formation clearance to 3MX was greater than that to 1,3DMU (fig. 3). The percentage decrease in formation clearance to 1MU was greater than that to 1,3DMU only in the elderly female group. The effect of cimetidine on the formation clearance to the two demethylated metabolites 3MX and 1MU was different only in the elderly male group (P = .05). There was no correlation between the proportionate decrease in formation clearance to each metabolite and the AUC of cimetidine.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Proportionate change in formation clearance to theophylline metabolites after treatment with (A) cimetidine, (B) ciprofloxacin, and (C) combined therapy. YM, young males; YF, young females; EM, elderly males; EF, elderly females.

The administration of ciprofloxacin alone was associated with a greater inhibition of the demethylation pathways than the hydroxylation pathway in the young female, elderly male and elderly female groups, as indicated by a greater proportionate decline in formation clearance to 3MX and 1MU compared with that of 1,3DMU (fig. 3). In the young male group, the proportionate change in formation clearance to all three metabolites was similar. However, this discrepancy was due to the data from one subject, which showed that the percentage decrease in formation clearance to 1MU (4%) was substantially lower than that observed from the other subjects in the group (range, 32-57%). When the data from this subject were excluded from analysis, the mean proportionate decrease in formation clearance to 3MX and 1MU was 48% and 46%, respectively, which was greater than the mean reduction of 27% for 1,3DMU (P < .05). Inhibition of formation of theophylline metabolites associated with ciprofloxacin therapy was unaffected by age or gender (table 5). The proportionate change in formation clearance to all three metabolites did not correlate with the AUC of ciprofloxacin.

In all groups, coadministration of cimetidine and ciprofloxacin produced a further proportionate decline in the formation clearance of all theophylline metabolites compared with each agent alone (fig. 3). The combined regimen exerted a greater impairment on the formation clearance to 3MX and 1MU than that to 1,3DMU. Age and gender did not alter the response to inhibition by the combined regimen.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The present study confirms our previous observations of an age-related reduction in theophylline plasma clearance in healthy male nonsmokers (Crowley et al., 1988; Vestal et al., 1987). In addition, this study extends the observation of an age-associated decrease in the metabolism of theophylline to healthy female nonsmokers. The base-line plasma clearance of theophylline was 24% lower in elderly female nonsmokers than in young women. In both gender groups, the decrease in theophylline clearance with age can be attributed to a reduction in the formation clearance to its metabolites. Furthermore, the extent of reduction in metabolic clearance of theophylline and its metabolites was similar in male and female groups. This suggests that the age-related decline in the biotransformation of theophylline is not influenced by gender differences.

Although the present and other studies (Antal et al., 1981; Crowley et al., 1988; Vestal et al., 1987) have demonstrated that age is associated with a decrease in oxidative metabolism of theophylline, the mechanism of this age difference has not been determined. Recent studies in human liver microsomes suggest that there are no marked age-dependent differences in the expression of hepatic CYP isozymes (Schmucker et al., 1990; Shimada et al., 1994). The in vitro activity of selected individual CYP isozymes such as CYP1A2, CYP2E1 and CYP3A4 (Hunt et al., 1990, 1992; Shimada et al., 1994) did not correlate with age, indicating that the ability of CYP isozymes to mediate the oxidative metabolism of xenobiotics is preserved in normal aging. Because CYP1A2 and CYP2E1 play a major role in catalyzing the biotransformation of theophylline, it is unlikely that the age-related decrease in theophylline metabolism in humans can be attributed to an intrinsic alteration in the catalytic property of the enzymes involved. An alternate explanation for this observation may be that age is associated with a 20% to 40% decline in liver volume and blood flow (Wynne et al., 1989). It is possible that the loss in liver mass may result in lower total CYP content available for metabolic conversion of theophylline to its metabolites, contributing to the decreased plasma clearance of theophylline and the decreased formation clearance to its metabolites in the elderly subjects. The nonselective age-related change in theophylline metabolic pathways observed in this as well as other studies (Crowley et al., 1988; Vestal et al., 1987) is consistent with this hypothesis.

Gender is an important contributing factor to the variability in pharmacokinetics of some drugs, although not all substrates that undergo hepatic oxidative metabolism exhibit gender-related differences in their elimination (Bonate, 1991). The results of this study, which demonstrate a lack of gender difference in metabolic clearance of theophylline in adults, is consistent with earlier reports (Jusko et al., 1979; Loi et al., 1991; Powell et al., 1977) and further illustrate that the effect of gender on in vivo drug metabolism is substrate specific. In contrast to the finding of a gender-independent effect on plasma clearance of theophylline, in our study, the volume of distribution of this drug was lower in female than in male subjects, and this difference was present in both young and elderly groups. A previous study in young nonsmokers reported a similar gender difference in volume of distribution of theophylline (Loi et al., 1991), and the present investigation demonstrates that this also occurs in elderly subjects. This difference likely is attributed to differences in body composition between the two gender groups. Furthermore, the smaller volume of distribution in women may account for the shorter elimination half-life of theophylline observed in the female group. Thus, in nonsmokers, gender does not influence the plasma clearance and metabolic pathways of theophylline but is associated with alteration in the volume of distribution and elimination half-life of theophylline. These gender-related effects are independent of age.

The magnitude of inhibition of theophylline metabolism by cimetidine (Cusack et al., 1985; Reitberg et al., 1981; Vestal et al., 1983a) and ciprofloxacin (Nix et al., 1987; Schwartz et al., 1988; Wijnands et al., 1986) reported previously is similar to that observed in this study. In male nonsmokers, cimetidine causes similar inhibition in theophylline elimination and metabolite formation in young and elderly groups (Adebayo and Coker, 1987; Vestal et al., 1987). Our results are in agreement with these observations. Furthermore, in healthy female nonsmokers, the proportionate change in theophylline clearance, elimination half-life and formation clearance to theophylline metabolites associated with cimetidine treatment was similar to that in men. This provides evidence that the response to inhibition of theophylline biotransformation by cimetidine is not influenced by age or gender. Likewise, the inhibitory effect of ciprofloxacin on theophylline metabolism was similar among the four groups of subjects. This uniformity of response to inhibition of theophylline metabolism by cimetidine and ciprofloxacin indicates that in healthy nonsmokers, age and gender do not appear to influence the extent of inhibition of theophylline metabolism by these two agents.

Previous in vivo and in vitro studies have shown that ciprofloxacin alters the demethylation pathways of theophylline biotransformation to a greater extent than the hydroxylation pathway (Robson et al., 1990; Sarkar et al., 1990). This study confirms that ciprofloxacin causes a greater proportionate impairment of the formation clearance to 3MX and 1MU compared with that of 1,3DMU. Furthermore, our data indicate that this preferential inhibitory effect is not influenced by age and gender. Because CYP1A2 is the major isozyme responsible for the demethylation reactions of theophylline metabolism (Gu et al., 1992; Sarkar et al., 1992), it can be inferred that neither age nor gender alters the adaptive response of this CYP isozyme to inhibition by ciprofloxacin. The effect of cimetidine on theophylline metabolic pathways is less clear. In the young male group, the proportionate change in formation clearance of the three metabolites was similar, suggesting that cimetidine was a nonselective inhibitor of theophylline metabolic pathways. This is similar to the results from a previous study (Vestal et al., 1987). Consistent with this finding, Knodell et al. (1991) reported that in human liver microsomes cimetidine is a moderate inhibitor of the activity of CYP1A2 and CYP2E1, which are the major CYP isozymes involved in formation of theophylline metabolites. However, although the proportionate decrease in formation clearance to 3MX was greater than that to 1,3DMU in the other three groups of subjects, differences between 1MU and 1,3DMU did not reveal a discernible trend. In vitro studies with human liver tissues indicate that there is a wide intersubject variation in expression and activity of individual CYP isozymes, including CYP1A2 and CYP2E1 (Forrester et al., 1992; Hunt et al., 1990; Shimada et al., 1994). It is possible that the apparent disparity in response to treatment with cimetidine may be attributed to intersubject variation in composition of individual CYP isozymes and the relative contribution of each CYP isozyme to the formation of theophylline metabolites.

The present study confirms in a larger number of subjects our previous observation (Loi et al., 1993) that combined administration of therapeutic doses of cimetidine and ciprofloxacin resulted in a greater decline in theophylline clearance and a further prolongation in its elimination half-life compared with each individual inhibitor. This suggests that the enhanced inhibitory effect on theophylline metabolism from coadministration of these two agents is observed when less than maximal inhibitory doses of each individual agent are used. Furthermore, age and gender differences did not influence the magnitude of changes in theophylline elimination associated with coadministration of cimetidine and ciprofloxacin. This further supports the conclusion that age and gender do not play a prominent role in determining the response to inhibitory drug interactions involving theophylline as the target substrate.

In all groups of subjects, after concomitant administration of cimetidine and ciprofloxacin, the formation clearances to 3MX and 1MU were impaired to a greater extent than that to 1,3DMU. This indicates that the combined regimen exerts a preferential inhibitory effect on the CYP1A2-mediated N-demethylation pathways of theophylline metabolism. In addition, the proportionate decrease in formation clearance to 3MX, 1MU and 1,3DMU associated with coadministration of cimetidine and ciprofloxacin was greater than that observed after administration of each inhibitor alone. This indicates that the combined regimen of cimetidine and ciprofloxacin produces a further decline in theophylline elimination through an enhanced inhibition of the formation of all theophylline metabolites. Of interest, a recent in vitro study in rat liver microsomes indicated that the additivity of the combined inhibitory effect on CYP isozyme activity from incubation with two inhibitors declined as the concentrations of the inhibitors exceeded their K_i values (Wei et al., 1995). This may explain the difference between the results of our study and those of Davis et al. (1992). Thus, concomitant administration of two inhibitors of CYP isozymes yields additional inhibitory effects on the metabolic conversion of the selected substrates only when less than maximally inhibiting doses are used.

A potential confounding factor in the interpretation of the results from this study arises from differences in plasma concentrations of inhibitors among the four groups of subjects. Systemic exposure to cimetidine and ciprofloxacin showed both age- and gender-related differences. The age-related increase in AUC of ciprofloxacin and cimetidine observed in this study was similar to those reported previously (LeBel et al., 1986; Ljungberg and Nilsson-Ehle, 1989; Vestal et al., 1987). Elimination of both drugs primarily involves renal excretion (Gladziwa and Klotz, 1993; Vance-Bryan et al., 1990). Because creatinine clearance among the four groups of subjects was different, this accounts, at least in part, for the observed age and gender differences in AUC values of both inhibitors. Despite a higher systemic exposure to cimetidine and ciprofloxacin in the older individuals, the magnitude of inhibition of theophylline metabolism was similar among the four groups of subjects. Although this may suggest a potential alteration in concentration-response relationship to inhibition of theophylline metabolism with age, the present study was not designed specifically to evaluate this relationship. Alternately, a recent study in rats demonstrates a nonlinear relationship between reduction in theophylline clearance and steady state concentrations of ciprofloxacin (Davis et al., 1994). The plasma clearance of theophylline declines in a hyperbolic fashion as the plasma concentration of ciprofloxacin increases. At high plasma concentrations of ciprofloxacin, the incremental reduction in theophylline clearance was smaller than the proportionate increase in inhibitor concentration. Thus, an increase in inhibitor concentration may not yield a significantly greater degree of inhibition. Further studies are needed to characterize the possible influence of age and gender on concentration-effect relationship of cimetidine and ciprofloxacin on inhibition of theophylline metabolism.

The findings of this study may have significant clinical implications for the use of theophylline in the elderly population. Shannon (1993) reported that age provides the best predictor of major toxicity in cases of chronic theophylline overmedication and that elderly patients have a significantly greater risk of serious adverse effects from theophylline toxicity than younger patients, even at similar plasma theophylline concentrations (Shannon and Lovejoy, 1990). Our data suggest that even though cimetidine and ciprofloxacin exert similar proportionate inhibition on theophylline metabolism in both age groups, because of a lower base-line theophylline clearance, the rise in plasma theophylline concentration as a consequence of the drug interaction may be greater in elderly than in young patients. As theophylline has a narrow therapeutic index, this may predispose the elderly to a greater risk of theophylline toxicity, particularly in elderly patients who are frail, have multiple medical problems and take multiple chronic medications. When adding a metabolic inhibitor like cimetidine or ciprofloxacin, careful monitoring of plasma theophylline concentrations and appropriate dose adjustment based on the expected change in theophylline elimination should be instituted to avoid excessively high theophylline levels.

In conclusion, this study demonstrates that in healthy male and female nonsmokers, the basal oxidative metabolism of theophylline is reduced with age but is unaltered by gender. Administration of cimetidine and ciprofloxacin impairs the elimination of theophylline. At therapeutic doses, concomitant administration of both drugs exerts a greater inhibition on theophylline clearance and metabolite formation than occurs with each agent alone. Neither age nor gender influences the response to inhibition of theophylline metabolism by cimetidine and ciprofloxacin.