Mean human serum, tear fluid, aqueous humor, and conjunctival tissue azithromycin levels after administration of a single oral dose of azithromycin.
Khalid F. Tabbara, Soliman A. Al-Kharashi, Samir M. Al-Mansouri, Othman M. Al-Omar, Hendrik Cooper, Ahmed M. Abu El-Asrar, George Foulds. Ocular Levels of Azithromycin. Arch Ophthalmol. 1998;116(12):1625–1628. doi:10.1001/archopht.116.12.1625
To assess azithromycin levels in human serum, aqueous humor, tear fluid, and conjunctival tissue specimens after administration of a single 1-g oral dose of azithromycin.
Sixty patients undergoing cataract surgery were included in this analysis. Serum, aqueous, and tear specimens were collected 3, 6, and 12 hours and 1, 2, 3, and 4 days after azithromycin administration. Conjunctival tissue biopsy specimens were collected 1, 2, 3, 4, 6, 8, 10, 12, and 14 days after azithromycin administration. All specimens were subjected to analysis by high-performance liquid chromatography–mass spectrometry.
Azithromycin concentration ranges during the specified sampling times were as follows: serum, 21 to 974 ng/mL; tear, 82 to 2892 ng/mL; aqueous, 10 to 69 ng/mL; and conjunctival, 0.7 to 32 µg/g. Levels above the 90% minimal inhibitory concentration (MIC90) for Chlamydia trachomatis were detected after 4 days in all tear samples and after 14 days in all conjunctival tissue specimens following oral azithromycin administration.
We demonstrated prolonged high levels of azithromycin in drug-targeted ocular tissue. Prolonged high concentrations of azithromycin in conjunctival tissue make this drug suitable for treatment of conjunctivitis caused by chlamydiae and other susceptible organisms.
AZITHROMYCIN IS a member of a class of antibiotics called azalides, which are macrolide antibiotics.1 These antibiotics include natural members, prodrugs, and semisynthetic derivatives. The chemical structures of macrolides are characterized by a large lactone ring containing between 12 and 16 atoms to which 1 or more sugars are attached via glycosidic bonds. The lactone ring is substituted by hydroxyl or alkyl groups. Many such members have been synthesized and investigated.2 Azithromycin is a macrolide derivative and was the first member of the 15-member ring azalide class of antimicrobial drugs described.
One problem with current therapeutic regimens for trachoma is the lack of compliance. Advances in the microbial study of trachoma has had little direct impact on public health control measures. Several antimicrobial agents are effective against trachoma. Results of recent prospective clinical trials3- 5 show that azithromycin therapy is as effective as a 6-week course of topical tetracycline hydrochloride ointment in children with active trachoma. To our knowledge, conjunctival tissue levels of azithromycin have not been studied in humans. The main objective of this study was to assess azithromycin concentration in human serum, conjunctival tissue, tear fluid, and aqueous humor after administration of a single oral dose of azithromycin in patients undergoing cataract surgery.
We enrolled 60 patients (48 men and 12 women) undergoing cataract surgery (age range, 43-78 years; mean age, 66 years). The patients' weights ranged from 51 to 86 kg (mean, 61 kg). All patients were admitted to the hospital and scheduled to undergo cataract extraction. Patients were given 1 g of azithromycin at 1 of the following times before surgery: 3, 6, or 12 hours or 1, 2, 3, 4, 6, 8, 10, 12, or 14 days. Five patients were enrolled at each time point. Serum, tear, and aqueous samples were collected 3, 6, and 12 hours and 1, 2, 3, and 4 days (a total of 35 collections) after administration of a single dose of azithromycin. Conjunctival biopsy specimens were collected 1, 2, 3, 4, 6, 8, 10, 12, and 14 days after administration of a single dose of azithromycin. Inclusion criteria required a documented diagnosis of cataract and an age of 18 years or older. Patients were excluded if they took other systemic medications during the study, were allergic to azithromycin or related macrolides, or were pregnant or lactating during the study. Ethics committee approval was obtained, and patients were informed of the operative procedures for conjunctival biopsy. Patients gave consent in accordance with the recommendation of the revised declaration of Helsinki.
All patients were given a single oral dose of azithromycin 2 hours after eating a light meal at a specific time before surgery. A member of the staff observed the patients consume the dose. All patients were informed to avoid food intake for 2 hours after taking the dose and were instructed to abstain from ingestion of other drugs. All study samples were collected before cataract extraction to minimize any surgical effect on drug disposition. Tear samples (30-100 µL) were collected into a capillary tube by capillary action at the lid margin. Serum and tear samples were immediately frozen in cryogenic vials at −86°C. The aqueous sample (100-200 µL) was obtained by anterior chamber paracentesis. Conjunctival biopsy specimens were obtained from the superior bulbar conjunctiva. Conjunctival biopsy specimens measured between 2 and 4 mm2; they were blotted dry, placed in a cryogenic vial, and immediately frozen. All samples were stored at −86°C until analysis. Separate conjunctival biopsy control samples were obtained from a drug-free source.
Azithromycin quantitation in serum and tissue samples was carried out using high-performance liquid chromatography with electrochemical detection6 and bioassay.7 We used high-performance liquid chromatography with detection by tandem mass spectrometry, using deuterated azithromycin as an internal standard as a specific and sensitive monitor in sample processing and drug calibration. Primary standard consisted of an analytically certified sample of azithromycin (Pfizer, Reference Standards Laboratory, Quality Control Division, Groton, Conn).
Tear and aqueous samples were defrosted and centrifuged at 3000 rpm for 3 minutes. For tear samples, 20 or 25 µL were transferred to a fresh tube and mixed with an equal volume of isotonic saline solution before analysis. Aqueous samples (50 µL) were processed without dilution. All samples had 50 ng of internal standard (ie, pure azithromycin) added and were injected as such. Serum samples (50 µL) were added to 50 µL of buffer (0.06 mol/L potassium carbonate) in a 15-mL glass centrifuge tube. Fifty nanograms of isotonic saline solution and 200 µL of deionized water were added and vortexed for 20 seconds. Samples were then extracted with 3 mL of methyl-tert-butyl ether and centrifuged (5 minutes at 3300 rpm).
Conjunctival biopsy specimens were analyzed, using the same instrumentation described above, and processed. Wet samples were weighed and had 500 ng of isotonic saline solution added to each sample. The samples were homogenized using a homogenizing probe on ice until a uniform suspension was obtained (approximately 45 seconds). Two milliliters of acetonitrile was added and the samples were again homogenized. After centrifugation (5 minutes at 3300 rpm), the supernatant was transferred to a 15-mL glass centrifuge tube and evaporated to dryness; 500 µL of 0.06 mol/L potassium carbonate was added; and the residue was vortexed for 30 seconds. Samples were then extracted with 3 mL of methyl-tert-butyl ether and centrifuged (5 minutes at 3300 rpm).
Chromatograms of azithromycin and deuterated azithromycin from blank tissue samples were processed immediately after high-concentration quality control samples and displayed absence of interfering response and no appreciable sign of sample cross-contamination. All samples outside the upper calibration level were serially diluted and reassayed. Quality control consisted of the running of unknowns that typically had an accuracy off the target value of ±10%. Calibration reruns at the end of the assay were typically within 5% of the initial calibration. Correlation coefficients exceeded 1.00 for all accepted calibrations. The limit of detection for the method for serum, tear, and aqueous samples was approximately 10 ng/mL, whereas the limit of detection for processed conjunctival biopsy specimens was approximately 50 ng/mg. Similar bioassays for the determination of azithromycin in tissue samples have been described previously.8
Serum, tear, and aqueous azithromycin concentrations are reported in nanograms per milliliter. Biopsy specimens are reported in nanograms per milligram of wet tissue weight. Table 1 shows the azithromycin levels in serum, tear, aqueous, and conjunctival specimens for all study times. The mean serum, tear, and aqueous concentrations with SDs and coefficients of variation are displayed for the sampling points of 3, 6, and 12 hours and 1, 2, 3, and 4 days. Occasional high SD and coefficient of variation values reflect a sample mean in which the number of samples comprising the mean was small (n=3). Concentrations are expressed in nanograms per milliliter for serum, tear, and aqueous samples. Time point statistics are derived from 5 patient samples, except for day 1 tear samples, which could be obtained from only 4 patients, and day 2 and day 3 tear samples, which could be collected from only 3 patients. For aqueous samples, day 3 and day 4 samples could be collected from only 4 patients. Mean biopsy specimen concentrations, with SD and coefficient of variation values, are represented for the sampling points of 1, 2, 3, 4, 6, 8, 10, 12, and 14 days. Concentrations are expressed in nanograms per gram for conjunctival biopsy specimens. Time point statistics are derived from 5 patient samples, except for days 1 and 3, on which 4 conjunctival specimens were collected. Mean accuracy for all quality control samples tested in conjunction with the patient sample run was more than ±10% off their expected target result. Figure 1 displays azithromycin pharmacokinetic profiles in serum, tear, aqueous, and conjunctival specimens. All samples are expressed using a common concentration axis of log micrograms per milliliter vs time in days. The ranges of azithromycin concentration for serum, tear, aqueous, and conjunctival specimens were 21 to 974 ng/mL, 82 to 2892 ng/mL, 10 to 69 ng/mL, and 0.7 to 32 µg/g, respectively, over the sampling time specified. The terminal elimination phase rate constant (Kel) of azithromycin in serum was estimated using least squares methods as the slope of the logarithm of the concentration against the time curve. All samples between 24 and 96 hours were used in its calculation. The apparent rate constant for the elimination in serum was approximately 0.02 per hour, corresponding to a half-life of 44 hours (half-life=log natural/Kel). Table 2 shows ratios of tear fluid, aqueous humor, and conjunctival tissue relative to serum values obtained for the time point and patient value of the ratio. The values in Table 2 are expressed as the mean ratio of the other component (tear, aqueous, or conjunctival) value relative to the time points concentration in serum. Mean ratios were greater in conjunctival tissue samples than in tear, serum, or aqueous samples. Mean concentrations of azithromycin in conjunctival tissue to mean serum concentration ratios were 288, 57, 174, and 257 at 1, 2, 3, and 4 days after administration of azithromycin, respectively.
In this study, serum concentrations declined relatively rapidly between 3 and 12 hours after administration of azithromycin, followed by a slow rate of decline between 24 and 96 hours. Our estimated half-life of 44 hours was similar to values reported by other investigators.9 Tear levels showed a rapid equilibration and a slower relative decline during the 96-hour sampling period; the tear pool represented the highest concentration observed of all the fluid compartments investigated. Tear fluid concentration over time may be affected by the circulating pool and dynamic redistribution of azithromycin from the tissue pool. The volume of distribution reported for azithromycin and its polyphasic metabolism profile further suggest that such a redistribution process may occur. We found that aqueous levels were predominantly lower than serum or tear levels, suggesting that minimal amounts of the drug partitioned into this compartment. It has been proposed10 that pigmented structures such as the iris, ciliary body, and choroid-retina may absorb the drug from this compartment and that the ocular tissue concentration of azithromycin may be paralleled by tissue vascularity.10 Results of studies8,11- 13 in humans and animals identify many cell types and tissue sites that have a strong affinity for azithromycin. Conjunctival biopsy specimens investigated during the 14-day sampling period displayed persistent high tissue levels relative to other ocular compartments. Conjunctival tissue affinity for azithromycin was apparent throughout the sampling range and displayed a gradual decline during the 14-day sampling interval. For comparison purposes, tissue levels of azithromycin in the prostate, cervix, lung, and tonsils remained high for several days.8,11- 13 Concentrations of azithromycin were between 1 and 9 mg/kg in most tissue samples obtained between 12 hours and 3 days after administration of azithromycin, 500 mg.8 The high tissue levels persisted for several days. For prostate and tonsil tissues, azithromycin levels were greater than 1 mg/kg for several days.8 Azithromycin tissue levels were lower in samples of fat, muscle, bone, and gastric mucosa than in the prostate and tonsils.8
Table 2 shows a significant redistribution of ingested azithromycin to the tear pool with respect to serum, especially after the 6-hour point. The aqueous humor–serum ratio was consistently less than 1, except for at the last point (96 hours), indicating that equilibration from circulation to this compartment was low. Conjunctival biopsy–serum ratios were significantly greater than circulating serum values, with serum decrease being paralleled by concurrent and rapid conjunctival uptake. These phenomena suggest that extensive absorption or binding mechanisms exist in ocular tissue, leading to this tissue's accumulation of the drug. The ocular pharmacokinetic profile displayed by azithromycin and its increased serum and conjunctival tissue half-life allow for simplified short-term therapeutic schedules. In addition, its gastric tolerability and diminished incidence of adverse effects make the compound suitable for the treatment of conjunctivitis caused by susceptible organisms such as Chlamydia trachomatis. This study was performed on patients undergoing cataract surgery with no clinical evidence of inflammation. Patients with trachoma or other infectious conjunctivitis have marked conjunctival inflammation. In such patients, there is higher conjunctival and tear levels of azithromycin.
Azithromycin is stable at gastric pH and has been found to have an absolute bioavailability of approximately 37% after oral administration. It is cleared primarily by biliary and fecal routes, with a serum half-life greater than 60 hours. Results of an open crossover study investigating the pharmacokinetic profiles of 3- and 5-day 500-mg oral regimens of azithromycin show that the drug is absorbed and distributed rapidly and evenly after administration of each daily dose, with a polyphasic decline of plasma concentrations showing average half-lives of 27.9 hours (3-day regimen) and 35.8 hours (5-day regimen). The observed pharmacokinetic profiles of azithromycin with both regimens conformed with previously established models for the drug.13 No significant drug interactions have been discovered to date. Use of azithromycin does not inactivate cytochrome P450 and does not induce the metabolism of other compounds.14
Metabolism is predominantly hepatic, with biliary excretion of inactive metabolites. Drug elimination is biphasic, with a terminal half-life of up to 5 days. Azithromycin has been approved for use in pediatric patients, and its pharmacokinetic profile is not significantly altered in elderly patients or in those with mild to moderate renal or hepatic insufficiency.8,9
Azithromycin has a large apparent volume of distribution of 25 to 35 L/kg, and plasma protein binding is low.15 This is a consequence of extensive and rapid tissue distribution and intracellular accumulation from plasma to tissue sites. Tissue concentrations of azithromycin above the MIC90 for many infective organisms have been correlated in animal models of infection. Short treatment protocols consisting of a single 1-g oral dose have been shown to provide adequate concentrations in urogenital tissues, above the MIC90 for C trachomatis for 10 days16,17 and in its serovars involved in induction and transmission of trachoma.3- 5 Although C trachomatis has an MIC90 of 0.25 µg/mL,18 levels of 10.0 µg/mL are required to completely inhibit inclusion formation.19 High levels are essential to inhibit proliferation of chlamydia and other organisms that proliferate intracellularly. Furthermore, because the chlamydial life cycle is 48 hours, an azithromycin tissue level of greater than the MIC90 of Chlamydia 14 days after administration of a single oral dose may help eradicate trachoma in endemic areas and control other susceptible organisms causing conjunctival infections.
Accepted for publication June 26, 1998.
This study was supported in part by the Abdul-Rahman Saad Al-Rashed Fund for Research in Ophthalmology, Riyadh, Saudi Arabia.
Reprints: Khalid F. Tabbara, MD, The Eye Center, PO Box 55307, Riyadh 11534, Saudi Arabia.