Objective
The objective of the study was to explore the maternal-fetal pharmacokinetics of intraamniotic (IA), intravenous (IV), or intramuscular (IM) administration of erythromycin or azithromycin in a pregnant sheep model.
Study Design
Pregnant ewes of 115-121 days’ gestation received a single maternal IV infusion (5 mg/kg over 60 min), a single IM injection, or a single IA injection (3.2 mg/kg fetal weight) of either erythromycin lactobionate or azithromycin. Maternal/fetal blood and amniotic fluid (AF) samples were collected across 48 h for macrolide assay by liquid chromatography and tandem mass spectrometry.
Results
Maternal administration achieved therapeutic maternal plasma macrolide concentrations (≥0.5 μg/mL) with low concentrations in AF equivalent to less than 7% transfer; fetal plasma levels were even lower (<1.5% transfer). The IA administration achieved therapeutic concentrations in AF and sustained for 48 h, with poor maternal-fetal transfer (<1% maternal, <0.3% fetal). Modest pharmacokinetic differences were evident between erythromycin and azithromycin.
Conclusion
Maternal macrolide administration achieves subtherapeutic concentrations in AF or fetal plasma, whereas a single IA injection achieves therapeutic concentrations in AF but not in maternal-fetal circulations. Combined maternal and single IA administration of macrolides may be a more effective regimen for treatment of intrauterine, but not fetal, infection.
Macrolide antibiotics such as erythromycin and azithromycin are widely prescribed during pregnancy for the treatment of a variety of microbial infections; these drugs are perceived to be effective, well tolerated, and free of serious maternal and fetal side effects. Erythromycin is a first-line antibiotic prescribed for the treatment of Chlamydia trochomatis and Group B streptococci infections and is the most frequently administered antibiotic for treatment of premature prelabor rupture of membranes (PPROM) based on the findings of the ORACLE I trial.
Azithromycin, a second-generation macrolide, is less widely used in pregnancy, although it has a longer duration of action and half-life, greater tissue penetration, reduced adverse effects, and a wider microbial coverage. Azithromycin is effective against many organisms of significance in pregnancy such as Staphylococci , Streptococci , Chlamydia , Haemophilus , Legionella , Listeria , Nisseriae , Toxoplasma , Mycoplasma , and Plasmodium spp. Because of its preferential accumulation in tissues, it has improved efficacy against intracellular pathogens such as Mycoplasma and Ureaplasma spp.
In light of the widely accepted role of intrauterine infection and inflammation in the aetiology of preterm labor and PPROM, antimicrobial therapy in pregnancy has been evaluated with the goal of reducing rates of preterm birth and neonatal morbidity and mortality. A recent metaanalysis of maternal macrolide administration following PPROM found statistically significant reductions in chorioamnionitis, preterm birth, and neonatal morbidity. However, antibiotic treatment has been shown to have, at best, only modest benefits on preterm delivery rates and neonatal outcomes when administered to women at risk of preterm birth or women presenting with preterm birth.
This relative lack of success is consistent with strong evidence that systemic maternal erythromycin administration is largely ineffective in eradicating intrauterine Ureaplasma infection, although reports of more successful outcomes do exist. Ureaplasmas and Mycoplasmas are the microorganisms most frequently isolated from human amniotic fluid (AF) and the placenta and are strongly associated with an intrauterine inflammatory response and preterm birth. Accordingly, their eradication from the amniotic cavity is likely an essential prerequisite for development of an effective antimicrobial regimen. The efficacy of an intraamniotic (IA) route of antibiotic administration has not yet been evaluated and reported.
In our experimental sheep model of intrauterine Ureaplasma colonization during pregnancy, we have recently shown that maternal intramuscular (IM) erythromycin administration does not eradicate Ureaplasma infection from the amniotic fluid, chorioamnion, umbilical cord, or the fetal lung. However, the biodistribution of macrolide antibiotics administered maternally in this model has not yet been described. The inability of maternal systemic erythromycin administration to combat intrauterine Ureaplasma colonization is likely to be attributable to poor transplacental passage.
In the perfused human placenta model, the transfer rate of macrolides is 2-4%. Sampling and analysis of maternal and fetal plasma at the time of therapeutic abortion or elective cesarean section at term suggests that the degree of erythromycin passage to the fetus is both low and inconsistent. Transplacental passage of erythromycin may be limited by the presence of drug efflux pumps in the placenta and fetal membranes. However, a small study performed in pregnant women at term reported that the transfer of macrolides from maternal circulation to amniotic fluid was considerable (reaching 30-50% of maternal plasma concentrations) and that the placenta accumulated antibiotics and maintained high concentrations for several days after administration.
The aim of the present study, therefore, was to compare the pharmacokinetics of maternal vs IA administration of erythromycin and azithromycin in the pregnant ovine model to determine the best antibiotic and route of administration for pharmacological treatment of intrauterine Ureaplasma infection.
Materials and Methods
Surgical procedures and antibiotic administration
All experimental procedures described in this study were approved by the Animal Ethics Committee of the University of Western Australia. Pregnant ewes bearing single fetuses underwent aseptic surgery at 110 days’ gestation (term = 150 days) for maternal and fetal catheterization. Ewes were premedicated with IM acetylpromazine (0.02 mg/kg) and buprenorphine (0.01 mg/kg, IM) approximately 30 minutes prior to induction of anesthesia with intravenous (IV) midazolam (0.25 mg/kg) and ketamine (5 mg/kg).
After induction of anesthesia, the trachea was intubated, and anesthesia was maintained with 1.5-2% isoflurane in 100% O 2 delivered with intermittent positive pressure ventilation. Using aseptic techniques, a paramedian abdominal incision was made, the pregnant uterus was identified, and the fetal head palpated and then delivered through a uterine incision. Polyvinyl catheters containing heparinized saline (50 IU/mL) were inserted into a fetal carotid artery and amniotic fluid cavity. The fetus was administered antibiotics (100 mg engemycin [oxytetracyclin], IM) and returned to the uterus. After closure of the uterine incision, catheters were exteriorised through a small incision in the ewe’s flank and the abdominal incision was closed in 2 layers.
Polyvinyl catheters containing heparinized saline (50 IU/mL) were then inserted into the maternal carotid artery and jugular vein. The catheters were secured to the maternal skin in 3 places and positioned behind the ewe’s neck. The ewe was administered antibiotics (400 mg engemycin, IM) and a fentanyl patch (Duragesic 75 μg/h; Alza Corp., Mountain View, CA) was applied to the maternal skin for postsurgery analgesia. Ewes were allowed to recover in individual floor pens with free access to food and water and were monitored daily for at least 5 days before experimentation.
At 115 days’ gestation, chronically catheterized pregnant ewes (53.5-64.4 kg) were randomly selected to receive one of the following: (1) a single maternal IV infusion of erythromycin lactobionate (5 mg/kg maternal weight; n = 5) over 60 minutes in 100 mL of saline; (2) single maternal IV infusion of azithromycin (5 mg/kg maternal weight, n = 5); (3) a single maternal IM injection of erythromycin lactobionate (5 mg/kg maternal weight, n = 5) (note: an IM azithromycin administration arm was not incorporated because clinically this antibiotic is not given IM); or a single IA injection of (4) erythromycin lactobionate or (5) azithromycin (3.2 mg/kg estimated fetal weight, n = 5).
At this gestational age, fetal weight was estimated to be 2.4 kg and amniotic fluid volume 250 mL. Maternal (5 mL) and fetal (2 mL) arterial blood and amniotic fluid (2 mL) samples were collected into heparinized tubes 15 minutes before and immediately prior to the administration of the macrolide antibiotics as described in the preceding text; after the infusion/injections, samples were taken at 1, 2, 4, 8, 12, 24, and 48 hours. Arterial PO 2 (PaO 2 ), PCO 2 (PaCO 2 ), O 2 saturation (SaO 2 ), pH (pHa), and electrolytes were measured with a RapidLab 1265 blood gas analyzer (Siemens, Munich, Germany). Fetal and maternal plasma and amniotic fluid samples were collected and stored at −80°C prior to quantification of the macrolide antibiotic concentrations.
Sample extraction and analysis
Calibration standards, consisting of a series of 5 10-fold serial dilutions of erythromycin and azithromycin antibiotics (4000 to 0.04 ng/mL), ovine plasma/amniotic fluid samples, and quality controls (all 0.1 mL) were extracted with 300 μL methanol by vortexing for 30 seconds and mixing on an orbital shaker for 10 minutes after addition of internal standard (0.5 ng roxithromycin). Quality control samples (200 ng/mL in plasma or amniotic fluid) were run with each assay batch.
After centrifugation (3200 × g , 20 minutes), extracts were transferred to glass tubes and dried with gentle heating (30°C) under nitrogen. Dried extracts were resuspended in 100 μL methanol and filtered through 0.22 μm cellulose acetate centrifuge tube filters (Corning Inc, Corning, NY); 0.5 μL was injected into an Agilent 1200-series capillary liquid chromatography system coupled to an Agilent 6340 Ion Trap mass spectrometer (Agilent Technologies, Palo Alto, CA). Methanol blanks were injected after every 11 sample injections. Separation was achieved using reverse-phase chromatography on a 3.5 μm Agilent Eclipse XDB-C18 column (2.1 × 150 mm); a gradient of 0-75% acetonitrile in 0.1% formic acid over 7 minutes was employed, followed by a 1 minute recovery wash at 100 % acetonitrile.
Macrolide MRM transitions were monitored in positive mode as follows: azithromycin (748.9-591.0), erythromycin (733.9-576.6), and roxithromycin (837.0-679.0); elution times were 2.9, 5.1, and 5.9 minutes, respectively. All mass to charge ratio peaks were manually integrated using 6300 Series Ion Trap liquid chromatography and mass spectrometry software 6.1.
Integrated signals from samples, controls, and standards were blank corrected, expressed as a ratio of the internal standard, and log transformed. A linear curve fit was applied to the calibration curve and this used to derive the concentration of the samples. Extraction efficiency for each sample was calculated by reference to the signal from a given amount of roxithromycin internal standard extracted and processed as per standards and samples. The limit of quantitation was defined as 5 SD above the background signal.
Statistical and pharmacokinetic analysis
Macrolide concentration data from 5 animals per time point were grouped and the mean and SDs calculated. These data were subject to pharmacokinetic analysis using the Microsoft Excel add-in programme PKSolver. Maternal pharmacokinetic data were fitted using a 2 compartment model, whereas the data from IA administration were best described by a noncompartmental model. Plasma and amniotic fluid control values from 4-6 assays were determined to derive the interassay coefficient of variation (CV).
Results
The azithromycin and erythromycin assays had a limit of quantitation of less than 0.1 and less than 0.3 μg/L, respectively. Mean extraction efficiency was 70.75%. Interassay CVs for erythromycin in maternal plasma and AF control samples were 14.7% and 19.4%, respectively (n = 6). Azithromycin interassay CVs for plasma and AF samples (n = 4) were 6.2% and 5.7%, respectively.
The biodistribution profiles of the macrolide antibiotics in the 3 compartments (maternal plasma, AF, and fetal plasma) following IV, IM, and IA administration are shown in Figures 1 and 2 using a log concentration axis to allow data from all compartments to be displayed on the same graph. Figure 3 compares the profiles achieved of the 2 antibiotics on a linear axis. The pharmacokinetic parameters for IM and IV maternal macrolide administration are listed in Table 1 , whereas the parameters for IA administration are shown in Table 2 .
| Parameter | Unit | Erythromycin | Erythromycin | Azithromycin |
|---|---|---|---|---|
| IM | IV | IV | ||
| A | μg/mL | 1.151 | 3.075 | 1.218 |
| Alpha | 1/hour | 0.259 | 1.964 | 2.284 |
| B | μg/mL | 0.042 | 0.247 | 0.154 |
| Beta | 1/hour | 0.259 | 0.374 | 0.0269 |
| Dose | mg | 284.9 | 282.9 | 273.5 |
| k 10 | 1/hour | 0.2593 | 1.4916 | 0.2187 |
| k 12 | 1/hour | 0.0000 | 0.3540 | 1.8112 |
| k 21 | 1/hour | 0.2592 | 0.4921 | 0.2806 |
| t ½α | Hour | 2.6736 | 0.3529 | 0.3035 |
| t ½α | Hour | 2.6742 | 1.8546 | 25.797 |
| V/F | (mg)/(μg/mL) | 276.8 | 85.2 | 199.3 |
| CL/F | (mg)/(μg/mL) per hour | 71.8 | 127.0 | 43.6 |
| V2/F | (mg)/(μg/mL) | 0.0 | 61.2 | 1286.2 |
| CL2/F | (mg)/(μg/mL) per hour | 0.0 | 30.1 | 360.9 |
| T max | Hour | 1.218 | 1.0 | |
| C max | μg/mL | 0.751 | 1.552 | 0.631 |
| AUC 0-t | μg/mL per hour | 3.970 | 2.227 | 4.674 |
| AUC 0-inf | μg/mL per hour | 3.970 | 2.227 | 6.277 |
| AUMC | μg/mL per hour | 17.413 | 2.567 | 213,990 |
| MRT | Hour | 4.39 | 1.15 | 34.1 |
| Vss | μg/(ng/mL) | 146.4 | 1485.5 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
