If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The pharmacokinetics of macrolide antibiotics — erythromycin (ER), clarithromycin (CL), roxithromycin (RO), azithromycin (AZ), dirithromycin (DI) and the concentrations achieved in polymorphonuclear neutrophils (PMNs) and saliva were investigated.
In a four-way crossover trial, 10 healthy volunteers received 1000 mg ER twice a day, 500 mg CL twice a day, 150 mg RO twice a day and 500 mg AZ every day over a period of 3 days. In a second trial, 10 healthy volunteers received 500 mg DI every day over a period of 5 days. Concentrations of these antibiotics were measured in serum, urine, saliva and PMNs by high-performance liquid chromatography (HPLC) on days 1 and 3 in the first trial and on days 1 and 5 in the second trial.
We found considerable differences in the pharmacokinetics, not only in serum, but also in PMNs and saliva. All substances except RO exhibited higher concentrations in PMNs than in serum, indicating excellent intraphagocytic distribution. In contrast, concentrations in saliva were lower than those measured in serum, with the exception of AZ. ER is characterized by low serum concentrations and moderate concentrations in saliva and PMNs. CL reached considerable concentrations in serum, saliva and PMNs. RO achieved the highest serum levels, but concentrations in saliva and in PMNs were below the detection limit. In contrast, AZ and DI yielded the lowest serum concentrations and the highest saliva and PMN concentrations.
Our data emphasize the importance of tissue distribution, in addition to serum kinetics, in evaluating the pharmacokinetic profiles of antibiotics.
Macrolide antibiotics have been available, and used clinically, since 1952, with erythromycin (ER) the most widely used agent of this class. Structurally, these antibiotics are characterized by a macrocyclic lactone ring linked glycosydically to amino glucose and nitrogen free glucose moieties. The newer compounds differ from ER in the size or substitution pattern of the lactone ring system, or in both these respects. They share a common antibacterial spectrum, with excellent activity against intracellular pathogens such as Legionella pneumophila, Mycopasma spp. and Chlamydia spp. [
]. Clarithromycin (CL), roxithromycin (RO), azithromycin (AZ) and dirithromycin (DI) have improved pharmacokinetic properties, especially high volumes of distribution and tissue concentrations, and improved acid stability [
The objective of the two studies was the comparison of pharmacokinetics and tissue concentrations of four new macrolides (CL, RO, AZ and DI) and ER after multiple-dose administration. Tissue concentrations were evaluated by measuring levels of antibiotics in polymorphonuclear neutrophils (PMNs) and saliva. PMNs are important cells of the human phagocytosis system and easy to isolate from the blood. Antibiotics accumulated in salivary gland tissue are excreted in saliva, which is easily collected. The levels of these substances in saliva reflect concentrations in the gland tissue [
The first study adopted a randomized four-way crossover multiple-dose design, and the second study a multiple-dose design.
Ten healthy subjects (five females, five males) without any known allergies to antibiotics or other medication participated in each of the trials. The mean age, body weight, body surface and creatinine clearance were similar for the volunteers in both studies (Table 1). Written informed consent was obtained. The studies were approved by the local ethics committee, according to German law. Female participants were tested for human chorionic gonadotropin (HCG) in urine to rule out pregnancy. All subjects had normal biochemical and hematologic profiles. None of the volunteers received any other antimicrobial agent during the 4 weeks preceding the study or during the study period. Furthermore, alcohol, tea, coffee and other beverages containing caffeine, as well as any kind of medication, were not allowed during the study period.
ER was given as the stearate preparation (Erythromycin 500 NEO®, lot number 44139VA, Abbott Ltd, Wiesbaden). RO was provided by Albert Roussel, Wiesbaden (lot number 165W0153), CL by Abbott Ltd, Wiesbaden (lot number 47051 TFB), AZ by Pfizer Ltd, Karlsruhe (lot number P 2705–04–001) and DI by Eli Lilly & Co, Indianapolis (lot number B0299-2B). During the first trial the subjects received 1000 mg ER twice a day, 150 mg RO twice a day, 500 mg CL twice a day and 500 mg AZ every day in four different courses, each over a period of 3 days, in a randomized order. ER and RO were given in the first and second courses, and CL and AZ in the third and fourth courses. The wash-out periods between the first three courses were 10 days; those between the third and the fourth courses 6 weeks. The prolonged period was chosen because of the longer half-life of AZ.
During the second trial the subjects were given 500 mg DI every day over a period of 5 days. 500 mg DI are equivalent to 440 mg erythromycylamine, which is the active compound. Concentrations are reported as DI.
Blood samples were taken on days 1 and 3 of dosing in the first study and on days 1 and 5 in the second study. Blood was collected before and 15, 30, 60 and 90 min and 2, 3, 4, 6, 8 and 12 h after dosage in both studies. In addition, blood samples were taken 24 h after the last administration of ER, CL and RO, and 24 and 48 h after the last administration of DI and AZ, on days 3 and 5 respectively. Serum was separated within 30 min after being taken and frozen at −80°C.
All volunteers provided pre-dose urine samples. After dosing, urine was collected on days 1 and 3 in the first trial and on days 1 and 5 in the second trial. The collection periods were between 0 and 6 and 6 and 12 h after administration of the antibiotics, except DI, with a third urine sampling between 12 and 24 h. On day 3 of the first study and on day 5 of the second study, additional 12–24- and 24–48-h samples were obtained. Samples for assay were frozen at −80°C.
Saliva and PMNs were collected on days 1 and 3 in the first trial and on days 1 and 5 in the second trial. Samples were taken for the first study before and 1 and 3 h after administration of ER and before and 4 and 8 h after administration of CL, RO and AZ. On day 3 an additional sample was obtained 24 h after administration of antibiotic except in the case of AZ, with two additional samples being taken after 24 and 48 h. During the second study, saliva and PMNs were sampled before and 4, 8 and 24 h after administration of DI. Samples for assay were frozen at −80°C. To produce saliva samples, the volunteers chewed a cotton wool swab for 60 s. Afterwards, approximately 0.7 mL saliva was recovered from the swab by centrifugation at 1560 g for 10 min.
PMNs were separated from heparinized blood by gradient centrifugation using a separation medium (Ficoll-Paque®). After lysis and washing out of erythrocytes from the cell pellet, the PMN count was adjusted by dilution with HBSS (Hanks’ balanced salts solution) to 2 × 107 cells/mL. The dilution volume was separated from the cell suspension by centrifugation. Lysis of PMNs was carried out with an ultrasonic homogenizer.
Serum, urine, saliva and PMNs were assayed for concentrations of the investigated antibiotics by high-performance liquid chromatography (HPLC). Since CL undergoes considerable first-pass metabolism, the main metabolite, 14-hydroxyclarithromycin (14-OH-CL), which also has antibacterial activity, was determined additionally.
HPLC assays were carried out using a modification of published methods and an internal Abbott protocol [
]. CL was used as internal standard. The organic phase was evaporated in a vacuum concentrator. The residue was dissolved in mobile phase. ER and CL were separated by reverse-phase chromatography and detected with an electrochemical detector.
HPLC assays were performed by using a modification of a method described by Demotes-Mainairde et al. [
Pharmacokinetic calculations and statistical evaluations
Pharmacokinetic parameters were calculated for the first day assuming an open one-compartment model, and for the third day of the first trial and the fifth day of the second trial assuming an open two-compartment model. The decision to use a specific (mathematical) model was based on the Schwarz criterion. The models have been used to calculate Tmax, Cmax and t1/2 only; the majority of the pharmacokinetic parameters have been calculated ‘non-compartmentally’, i.e. the results do not depend on a particular assumption. All results were normalized to 70 kg body weight; clearance values were calculated for a body surface of 1.73 m2. Wilcoxon's rank test was used for statistical analysis. P values <0.05 were considered significant.
The pharmacokinetic results are summarized in Table 3. Comparison of the data on day 1 and at steady state (day 3 in study 1, day 5 in study 2) revealed some significant differences (p < 0.05). After the last dosing a rise in urinary recovery (Urec)and half-life (t1/2) was observed for all substances (for DI a common half-life was calculated analyzing all the data from all days). An increase of the area under the curve (AUC) for CL, AZ, DI, and of the peak serum concentration (Cmax) for DI, occurred after multiple dosing. An explanation of these changes could be the longer period of urine and serum collection at steady state, compared with the first day of dosing.
Table 3Pharmacokinetic results: arithmetic means and standard deviations
The pharmacokinetics at steady state were as follows: Cmax and AUC were highest for RO with 4.3 mg/L and 47.3 mgXh/L respectively, followed by CL with 2.3 mg/L and 27.9 mgXh/L respectively (and 0.51 mg/L and 9.1 mgXh/L for the metabolite 14-OH-CL). ER was characterized by low Cmax and AUC values (1.3 mg/L and 5.17 mgXh/L respectively), while AZ and DI exhibited by far the lowest values for these parameters (0.59 mg/L and 2.59 mgXh/L, and 0.5 mg/L and 3.2 mgXh/L, respectively). The terminal half-life of DI showed the highest value (19.2 h), followed by RO (11.7 h), AZ (10.7 h) and CL and its metabolite (7.8 and 10.5 h, respectively). The T1/2 of ER was calculated as 7.9 h. DI and AZ were characterized by the largest volumes of distribution (Vd), at 48.6 and 36.9 L/kg, respectively, indicating high tissue concentrations. In contrast, the lowest Vd of 0.87 L/kg was calculated for RO. CL and ER yielded the moderate Vd values of 3.2 and 10.8 L/kg, while a Vd of 19.3 L/kg was measured for 14-OH-CL. Urinary recoveries (0–48 h) were highest for CL and 14-OH-CL (52% and 26%, respectively, of the given dose), followed by RO (15%), AZ (10.9%). DI (5.35%) and ER (4.2%). Renal clearance (clren) varied considerably among the investigated macrolides. RO exhibited the lowest and AZ the highest value, at 10.7 and 296.6 mL/min/1.73 m3, respectively.
Macrolide concentrations in PMNs
Macrolide concentrations in PMNs are shown in Table 4. Accumulation factors corresponding to the ratio of the cellular concentration in PMNs to the extracellular concentration in serum (= C/E) are given in Table 5.
Table 4Macrolide concentrations in PMNs (mg/L): Arithmetic means and standard deviations
Concentrations for all substances except RO exceeded by far those in serum; therefore, the accumulation factors were considerably greater than 1 and increased significantly at steady state (day 3 in study 1, day 5 in study 2). This increase could be due to tissue accumulation after multiple dosing. AZ exhibited the highest concentrations in PMNs: 15.9 and 13.1 mg/L after 4 and 8 h of application on day 1, and 39.3, 41.6, 28.7 and 24.8 mg/L after 4, 8, 24 and 48 h on day 3, respectively. Levels of DI were 3.81, 3.81 and 2.85 mg/L after 4, 8 and 24 h on day 1 and 13.31, 13.38 and 10.47 mg/l after 4, 8 and 24 h on day 5. CL achieved PMN concentrations of 5.78 and 4.82 mg/L on day 1 and 13.9 and 9.42 mg/L on day 3 after 4 and 8 h of administration. Values for 14-OH-CL were 3.98 and 3.39 mg/L on day 1, and 7.99 and 6.28 mg/L on day 3, after 4 and 8 h. Concentrations of CL and 14-OH-CL 24 h after dosing on day 3 were below the detection limit of the HPLC method. Levels of ER in PMNs were 2.9 and 1.9 mg/L after 1 and 3 h on day 1, and 6.07, 6.58 and 3.99 mg/L after 1, 3 and 24 h on day 3. In contrast the concentrations of RO were below the detection limit of the HPLC method in all PMN samples. The highest accumulation factor was calculated for AZ at 859, followed by DI at 106, 14-OH-CL at 10.2, CL at 6.5 and ER at 6.2, at steady state.
Macrolide concentrations in saliva
Macrolide concentrations in saliva are shown in Table 6. The levels of all substances except AZ were below those measured in serum. An increase in these values was observed at steady state; only the metabolite 14-OH-CL and the 1-h value for ER did not show any differences after first and last dosings.
Table 6Macrolide concentrations in saliva (mg/L): Arithmetic means and standard deviations
The mean concentrations of CL were 1.06 and 0.48 mg/L 4 and 8 h after administration on day 1, and 1.87, 0.83 and 0.25 mg/L after 4, 8 and 24 h on day 3. Values for 14-OH-CL were 0.94 and 0.52 mg/L after 4 and 8 h on day 1, and 0.84, 0.45 and 0.19 mg/L after 4, 8 and 24 h at steady state. AZ achieved saliva concentrations of 0.59 and 0.31 mg/L after 4 and 8 h on day 1, and 1.61, 0.67 and 0.42 mg/L after 4, 8 and 24 h on day 3. The levels of ER were 0.20 and 0.10 mg/L and 0.31 and 0.19 mg/L after 1 and 3 h of administration on days 1 and 3, respectively. On day 3, concentrations of ER were below the detection limit 24 h after application. DI was characterized by low values with 0.09, 0.05 and 0.03 mg/L on day 1 and 0.31, 0.17 and 0.10 mg/L on day 5, after 4, 8 and 24 h of application. Levels of RO were below the detection limit of the HPLC method in almost all saliva samples. The maximal matrolide concentrations in serum, PMNs and saliva are shown in Figure 1.
Macrolide antibiotics are of significant interest and importance since intracellular infections are increasing with higher numbers of immunocompromised hosts. In these two studies the pharmacokinetics of ER, CL, RO, AZ and DI and the concentrations achieved in PMNs and saliva were analyzed.
ER showed low serum concentrations and moderate concentrations in PMNs and saliva. Since ER is characterized by poor water solubility and rapid inactivation, by gastric acid, the bioavailability of this compound is unpredictable. In the literature Cmax values after dosing with 500 mg ER stearate have been given as 1.23 [
]. Concentrations of ER in PMNs exceeded those in serum, with a peak level of 6.58 mg/L 3 h after application at steady state, corresponding to an accumulation factor of 6.2. Studies in vitro with 14C-labeled ER demonstrated accumulation factors of 9.7 [
] in PMNs compared with extracellular fluid. Studies in vivo in volunteers are not available. Concentrations in saliva of ER were below those measured in serum. A maximal value of 0.31 mg/L was detected at steady state 1 h after administration, and an accumulation factor of 0.17 was calculated. These results are in good agreement with other studies. Touminen found an accumulation factor for saliva between 0.12 and 0.20 compared with serum levels after administration of 500 mg ER acistrate [
CL and 14-OH-CL were characterized by considerable concentrations in serum and in PMNs, while saliva levels were moderate. Peak serum concentrations after the first dose of 500 mg CL were 2.1 and 0.81 mg/L for the original substance and the metabolite, respectively. Between days 1 and 3 there was no significant change in this parameter. In contrast, AUC values rose significantly from 15.3 mgXh/L after the first dosing to 27.9 mgXh/L after the last dosing. An explanation of this increase could be the longer serum collection period at day 3. Pharmacokinetic data published in the literature are in good agreement with these results. Neu measured Cmax values of 2.41 and 0.66 mg/L after administration of 500 mg CL and 14-OH-CL respectively [
]. After the first dose peak serum concentrations varied between 1.77 and 1.89 mg/L for CL and 0.67 and 0.8 mg/L for the metabolite; at steady state these values were between 2.4 and 3.5 mg/L and 0.7 and 0.8 mg/L, respectively. This significant increase in serum levels after multiple dosing was in contrast to our data.
Levels of CL and its metabolite in PMNs exceeded those in serum. In PMNs the highest concentrations of CL and its metabolite were measured at steady state after 4 h of dosing at 13.9 and 6.28 mg/L. The highest accumulation factors were 6.45 for CL and 10.2 for 14-OH-CL after 8 and 4 h of administration, respectively. No other data on in vivo studies have been published. PMN concentrations of CL were investigated in vitro by Ishiguro et al [
]. After incubation of PMNs for 30 min with 10 mg/L CL, the intracellular accumulation factor reached 16.4. Saliva concentrations of CL and its metabolite were highest at steady state 4 h after dosing with 1.87 and 0.84 mg/L, respectively. The accumulation factor reached maximal values of 0.89 and 0.88 at steady state 24 h after administration, indicating lower saliva than serum levels. The results of another study included an accumulation factor for CL in saliva of 0.5 and 1.9 after single doses of 150 and 400 mg, respectively [
]. In serum and saliva the metabolite 14-OH-CL was not reported.
With RO the highest serum concentrations of all were found. Cmax values reached 5.3 and 4.3 mg/L after first and last dosing, respectively. The data are in agreement with other studies. Nilson found maximal serum concentrations between 5.4 and 7.9 mg/L after giving a single dose of 150 mg RO [
]. In contrast to our results, there was a rise of Cmax from 7.9 to 9.3 mg/L after the first and the last dose, respectively. Levels of RO in PMNs and saliva were below the detection limits of the HPLC method. The low volume of distribution derived from our pharmacokinetic calculations also indicated relatively low tissue concentrations of this compound.
These data are in contrast to results obtained in several in vitro studies, where PMN concentrations were investigated with 14C-labeled RO. Carlier et al. found an accumulation factor of 14 after 30 min of incubation of human PMNs with 14C-labeled RO [
]. After a single dose of 150 mg RO to four volunteers, the highest concentration was 0.3 mg/L.
AZ and DI are different from the other macrolides in being characterized by the lowest serum levels and longest terminal half-lives. At steady state, peak serum concentrations reached 0.59 and 0.50 mg/L for AZ and DI, respectively; half-lives were calculated as 10.8 h for AZ and 19.2 h for DI. Published pharmacokinetic studies have reported similar serum levels but longer half-lives. After administration of 500 mg AZ, Cmax values of 0.41 and 0.40 mg/L were achieved in two other studies [
]. In contrast to their low serum levels, these antibiotics are characterized by high tissue concentrations. The highest concentrations in PMNs were found 8 h after dosage at steady state, at 41.6 mg/L for AZ and 13.4 mg/L for DI. Maximal accumulation factors of 859 (AZ) and 106 (DI) were calculated at steady state, 48 and 24 h after dosing. In several in vitro studies PMNs were incubated with 14C-labeled AZ and DI. After an incubation period of 40 min with 4 mg/L of AZ, an accumulation factor of 298 was calculated [
]. Concentrations of erythromycylamine (the active metabolite of DI) in human lung tissue were about 20 to 40 times greater than the simultaneous values in plasma 12 or 24 h after administration of 250- or 500-mg tablets [
Saliva concentrations reached peak values at steady state 4 h after dosing with 0.67 mg/L for AZ and 0.31 mg/L for DI. The highest accumulation factor of AZ was calculated 24 h after dosing at steady state, with a value of 6.38. DI exhibited lower concentrations in saliva than in serum, corresponding to accumulation factors below 1. There are no other published data available on concentrations of these macrolides in saliva.
In conclusion, we found considerable differences both in pharmacokinetics and in PMN and saliva concentrations with the investigated macrolides. These differences may be an expression of different lipid solubilities, usage of different channels or carrier systems to enter into the cells or different forms of trapping inside cell organelles; in addition, different mechanisms of efflux from the cells have been reported [
We acknowledge the technical assistance of E. Borner, H. Hartwig, M. Rau and J. Voeckler. This study was supported by E. Lilly, Bad Homburg, Germany; Abbott, Wiesbaden, Germany and Pfizer, Karlsruhe, Germany.
Roxithromycin; a review of its antibacterial activity, pharmacokinetic properties and clinical efficacy.