http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, Early Online: 1–5 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.976548
ORIGINAL ARTICLE
Concentration of amoxicillin in maternal serum, cord blood, amniotic fluid and the placenta after vaginal administration
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
Julia Zare˛ba-Szczudlik1, Ewa Romejko-Wolniewicz1, Zbigniew Lewandowski2, Hanna Ro´z_ an´ska3, and Krzysztof Czajkowski1 1
2nd Department of Obstetrics and Gynecology, 2Department of Epidemiology, Medical University of Warsaw, Warsaw, Poland, and 3Department of Hygiene of Food of Animal Origin of National Veterinary Research Institute in Puł awy, Poland
Abstract
Keywords
Objective: The aim of this study was to assess the amoxicillin concentration in maternal serum, cord blood, amniotic fluid and the placenta, 2 h following vaginal administration and the factors influencing the drug level. Methods: Twenty-eight full-term pregnant women who qualified for elective cesarean delivery were included in the study. Vaginal suppositories containing 250 mg of amoxicillin were administered 2 h prior to the operation. Amoxicillin levels were determined using the diffusion microbial assay. Results: The amoxicillin level in amniotic fluid was significantly higher in comparison to that of maternal serum, cord blood or the placenta. Maternal age positively and gestational weight gain negatively correlated with the amoxicillin concentration in maternal serum. The maternal serum hemoglobin level and red blood cell count were positively correlated with amoxicillin concentration in the amniotic fluid. Neonatal birth weight was positively correlated with maternal serum and cord blood amoxicillin levels. Hypertensive women had significantly higher amoxicillin concentrations in amniotic fluid, and women with thrombocytopenia presented significantly higher cord blood amoxicillin concentrations. Conclusions: Amoxicillin presented poor concentration in maternal–fetal compartments after vaginal administration, but the factors influencing the drug level in different compartments require further investigation.
Amoxicillin penetration, antibiotic vaginally, pregnancy
Introduction Amoxicillin is a beta-lactam antibiotic characterized by a broad spectrum of activity and a low toxicity [1,2]. For these reasons, it is widely used in obstetrics. According to the United States Food and Drug Administration (FDA), amoxicillin is a pregnancy category B drug. Most often, it is administered to pregnant women orally or intravenously. Reports of intravaginal administration of amoxicillin to treat vaginal or cervical infections with susceptible bacteria are rare and controversial. No studies have been published regarding the assessment of amoxicillin levels in fetal and afterbirth tissues following this route of administration. The profiles of patients in our hospital – women with highrisk pregnancy – are the reason we check genital tracts for the presence of many bacteria and fungi. According to cervical and vaginal culture, amoxicillin in the dose of 250 mg is the most frequently used intravaginal antibiotic in pregnant women in our hospital to treat local infections. Address for correspondence: Ewa Romejko-Wolniewicz, MD, PhD, 2nd Department of Obstetrics and Gynecology, Medical University of Warsaw, ul.Karowa 2, 00-315 Warsaw, Poland. Tel: +48507097063. Fax: +48225966487. E-mail: [email protected]
History Received 6 June 2014 Revised 15 September 2014 Accepted 10 October 2014 Published online 11 November 2014
The aim of this study was to assess the amoxicillin concentration in maternal serum, cord blood, amniotic fluid and the placenta 2 h after vaginal administration and to determine the factors that might affect this concentration. Goldberger et al. [3] estimated the concentration of penicillin after vaginal administration in 10 non-pregnant women and revealed that maximum level occurred 1 h after the application, but the therapeutic levels of penicillin were maintained for up to 3 h. Abel et al. [4] measured the levels of penicillin after vaginal administration at four time points and concluded that the degree of penicillin absorption from the vaginas of non-pregnant women was completely inconsistent and unpredictable. Rock et al. [5] concluded that patients who were near term absorbed little or no penicillin; the authors examined four patients and showed no absorption in two of them who were near term, little absorption in one near term (the level of penicillin was higher after 4 h than 1 h after application) and good absorption, similar to that of nonpregnant women, in one patient who was 2 months before term. Considering the results of the above studies and the maximum serum levels detected 2 h after oral and intramuscular administration [3–5], we decided to measure the amoxicillin levels 2 h after vaginal administration.
2
J. Zare˛ba-Szczudlik et al.
Materials and methods Twenty-eight full-term pregnant women (Weeks 37–41) who qualified for elective cesarean delivery were included in the study. Vaginal suppositories containing 250 mg of amoxicillin in a base of cocoa butter were administered 2 h prior to the cesarean section.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
Study protocol After admission, informed consent was obtained from each patient and the patient’s medical history was collected, a physical and obstetrical examination was performed and information from the pregnancy card was obtained. Each patient underwent a cervical swab for a bacterial culture assay. During the cesarean section, 5 mL of amniotic fluid was collected from each patient via aspiration for a bacterial culture assay. The following samples were collected during the cesarean section: 20 mL of amniotic fluid, 100 g of placenta, 9 mL of cord blood and 9 mL of maternal blood. The amniotic fluid and placenta samples were placed in plastic containers. After clotting, the maternal and umbilical blood samples were centrifuged, and the serum was filtered off and placed in plastic containers. The material was frozen at 20 C. Samples were sent monthly to the National Veterinary Research Institute in Puławy, where the antibiotic levels were determined. The study protocol was approved by the Ethics Committee of the Medical University of Warsaw.
J Matern Fetal Neonatal Med, Early Online: 1–5
Table 1. Maternal and neonatal characteristics.
Variable (unit)
Min
Max
SD, standard deviation; Min, the minimal value of variable; Max, the maximal value of variable; SGA, small for gestational age; LGA, large for gestational age. Table 2. Correlation of the amoxicillin concentration after vaginal administration with quantitative maternal and neonatal parameters.
Parameters
Amoxicillin levels were determined with the diffusion microbial assay described by Kundrat [6] using Kundrat agar and Bacillus stearothermophilus ATCC 7953 (SigmaAldrich Company Ltd. Poznan, Poland) as the test strain. The medium composition was as follows: peptone 17.0 g/L, NaCl 3.0 g/L, D-(+)-glucose 3.0 g/L, starch 3.0 g/L, gelatin 2.5 g/L, bromocresol purple 0.016 g/L, and agar 10.0 g/L, pH 6.8. The principle of the assay is growth inhibition of a spore suspension of the test strain B. stearothermophilus in the presence of antibiotics in the test sample. Tissue samples were placed in wells cut in the agar (10 mm in diameter) and incubated for 3 h at 64 C. The antibiotic level was determined based on the diameter of the test strain’s growth inhibition zones, with reference to the standard curve prepared using antibiotic-free material. The standard curve for determining the amoxicillin levels in the test samples was prepared using the reference substance from Sigma (cat. no. A8523, Sigma-Aldrich Company Ltd. Poznan, Poland) with an amoxicillin content of 900 mg/g (FDA-compliant). Each measurement was repeated four to five times depending on the quantity of available material. Arithmetic means were calculated from the obtained results.
Maternal age
Quantitative variables were presented as the mean values and standard deviations and as medians, quartiles and range. Qualitative variables were presented as frequency and percentage (Table 1). Comparisons of amoxicillin levels
Median
Age (years) 30.4 ± 4.2 31.0 22.0 38.0 Pre-pregnancy weight (kg) 65.7 ± 12.1 64.5 45.0 97.0 24.0 ± 4.4 22.9 17.6 33.3 Pre-pregnancy BMI (kg/m2) Gestational weight gain (kg) 14.1 ± 5.1 14.0 5.0 27.0 Multiparas 14 (50.0) History of cesarean section 13 (46.4) Pre-existing hypertension 2 (7.1) Pregnancy-induced hypertension 2 (7.1) Pre-existing diabetes 3 (10.7) Gestational diabetes 8 (28.6) Thrombocytopenia 6 (21.4) Serum hemoglobin level (g/dL) 11.9 ± 0.9 11.8 9.7 13.5 Red blood cell count (mln/mL) 3.9 ± 0.3 3.9 3.2 4.5 Bacteria in the cervix 7 (25.0) Gestational age upon delivery 38.3 ± 1.2 38.0 37.0 41.0 (weeks) Birth weight (g) 3534.6 ± 559.7 3550.0 2050.0 4480.0 Body length (cm) 54.8 ± 3.1 55.5 49.0 60.0 Boys 16 (57.1) SGA 2 (7.1) LGA 11 (39.3)
Drug level determination method
Statistical analysis
Mean ± SD or n (%)
Maternal Cord Amniotic serum blood Placenta fluid r r r R p p p p
0.54 0.10 0.16 0.003 ns ns Maternal pre-pregnancy weight 0.03 0.15 0.06 ns ns ns Maternal pre-pregnancy BMI 0.15 0.21 0.06 ns ns ns Gestational weight gain 0.41 0.06 0.27 0.029 ns ns Maternal serum hemoglobin level 0.05 0.14 0.05 ns ns ns Maternal red blood count 0.10 0.08 0.16 ns ns ns Neonatal birth weight 0.40 0.38 0.13 0.034 0.045 ns
0.17 ns 0.01 ns 0.02 ns 0.01 ns 0.41 0.035 0.46 0.017 0.31 ns
r, correlation coefficient; p, significance level (bold values).
between compartments were performed using the signed-rank test, and the amoxicillin levels in patient’s compartments were standardized for use in statistical analyses using the following coefficients: a serum specific gravity of 1.026 g/mL and an amniotic fluid specific gravity of 1.006 g/mL. Associations between quantitative variables were tested using correlation analysis (Spearman’s correlation coefficient; Table 2). Amoxicillin levels between subgroups distinguished according to the considered maternal parameters were compared using a two-tailed Mann–Whitney test for independent groups (Table 3). For revealed qualitative predictors of amoxicillin concentration, the effect was tested for possible confounding factors with Wilcoxon two-sample test.
Vaginal amoxicillin in pregnancy
DOI: 10.3109/14767058.2014.976548
3
Table 3. Amoxicillin concentration after vaginal administration in the different compartments related to qualitative maternal parameters. Maternal serum Parameters
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
Diabetes Yes No Hypertension Yes No Thrombocytopenia Yes No Cervical culture Yes No
Cord blood
Placenta
Amniotic fluid
Med (Q1–Q3)
p
Med (Q1–Q3)
p
Med (Q1–Q3)
p
Med (Q1–Q3)
p
9.0 (6.0–12.0) 9.0 (8.0–10.0)
ns
10.0 (9.0–12.0) 8.0 (8.0–11.0)
ns
11.0 (10.0–12.0) 10.0 (9.0–12.0)
ns
37.0 (25.0–39.0) 28.0 (12.5–38.0)
ns
8.0 (7.0–10.0) 9.0 (7.5–10.5)
ns
10.5 (8.5–12.5) 9.0 (8.0–11.0)
ns
10.0 (7.5–11.0) 10.0 (9.0–12.0)
ns
39.0 (38.0–45.5) 29.0 (16.3–38.0)
0.038
10.0 (7.0–12.5) 8.5 (8.0–10.0)
ns
12.0 (10.0–12.0) 9.0 (8.0–10.0)
0.050
11.0 (9.0–12.0) 10.0 (9.0–12.0)
ns
31.0 (12.5–38.0) 33.0 (22.0–39.0)
Ns
9.0 (7.0–11.0) 9.0 (8.0–10.0)
ns
10.0 (9.0–12.0) 9.0 (8.0–11.0)
ns
10.0 (9.0–12.0) 10.0 (9.0–12.0)
ns
38.0 (12.5–39.0) 32.0 (25.0–38.0)
Ns
Med, median; Q1, lower quartile; Q3, upper quartile; p, significance level (bold values).
The statistical analyses were performed according to manuals by L. D. Fisher and Gerald van Belle [7] and the SAS system [8].
Results Maternal and neonatal characteristics The maternal and neonatal characteristics are included in Table 1. Seven women were diagnosed with bacteria in the cervix, while one was diagnosed with amniotic fluid infection. No child was diagnosed with an infection. The Apgar scores at 3 and 5 min were between 8 and 10 points for all newborns. Amoxicillin concentration The concentrations of amoxicillin in different compartments 2 h after vaginal application were the following: 29.9 ± 12.70.013 ng/g in the amniotic fluid, 10.6 ± 9.9 ng/g in maternal serum, 10.2 ± 1.8 ng/g in the placenta and 9.3 ± 2.3 ng/g in the cord blood. The level of amoxicillin in the amniotic fluid was significantly higher than that in the other three tissues (p50.05). There were no significant differences among the drug levels in maternal serum cord blood and the placenta. Maternal age positively correlated with the amoxicillin concentration in maternal serum (r ¼ 0.54, p ¼ 0.033; Table 2). Gestational weight gain but not maternal prepregnancy weight and BMI was negatively associated with maternal serum levels of amoxicillin (r ¼ 0.41, p ¼ 0.029; Table 2). Maternal serum hemoglobin level and red blood count were positively correlated with amoxicillin concentration in amniotic fluid (r ¼ 0.41, p ¼ 0.035 and r ¼ 0.46, p ¼ 0.017, respectively; Table 2). Neonatal birth weight was positively correlated with maternal serum and cord blood amoxicillin levels (r ¼ 0.40, p ¼ 0.034 and r ¼ 0.38, p ¼ 0.045, respectively; Table 2). Diabetes had no influence on the amoxicillin concentrations in any of the examined compartments. Hypertensive women had significantly higher amoxicillin concentrations in amniotic fluid (p ¼ 0.038; Table 3). This effect was tested for
three parameters: maternal age, pre-pregnancy BMI, gestational weight gain and only BMI was associated with hypertension (BMI in hypertensive group was significantly higher: 26.9 versus 23.5 kg/m2, respectively, p ¼ 0.02). Women with thrombocytopenia (platelet count below 150 103/mL) presented significantly higher cord blood amoxicillin concentrations (p ¼ 0.05; Table 3). This effect was tested for five parameters: maternal age, pre-pregnancy BMI, gestational and weight gain, serum hemoglobin level, gestational age and none of these was associated with thrombocytopenia. No differences in amoxicillin levels were identified between patients with negative cervical swab cultures (21 cases) and patients with bacteria in the cervical canal (7 cases; Table 3). No significant differences were seen based on parity, gestational age at delivery and the gender of the newborn.
Discussion The aim of our study was also to recognize factors that might affect the concentrations of amoxicillin in different compartments. There are studies assessing amoxicillin levels in different tissues at different time points after administration to different patients. Based on these studies, a curve was drawn for tissue antibiotic levels as a function of time. However, results from different patients do not allow unambiguous conclusions or an evaluation of factors that might affect the antibiotic levels. In our study, we examined amoxicillin levels in four compartments of every patient at one time point following vaginal administration. We had only one opportunity to measure the level of amoxicillin in the amniotic fluid and the placenta during the cesarean section, so we decided to measure the levels in the other compartments at the same time. Small amounts of intravaginally administered amoxicillin were detected in the tested samples, which is not surprising considering that amoxicillin is a hydrophilic substance that passes through the cell membrane poorly. Therefore, one may conclude that amoxicillin, when administered vaginally, penetrates into maternal and fetal tissues poorly, although its presence in the amniotic fluid suggests its capability to
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
4
J. Zare˛ba-Szczudlik et al.
penetrate into this compartment. Most likely, vaginal administration of higher doses would result in higher concentrations in the amniotic fluid; further studies are required to confirm this hypothesis. Unfortunately, it is impossible to compare our results with other results as there are no reports on the vaginal use of amoxicillin in the literature. Our study revealed no significant differences in the amoxicillin levels in the maternal serum, cord blood, placenta and amniotic fluid 2 h after vaginal administration, depending on parity, maternal prepregnancy weight, maternal prepregnancy BMI, diabetes during pregnancy, gestational age at delivery and the gender of the newborn. Muller et al. [9] also observed no effect of the gestational age, maternal body weight and BMI, maternal blood pressure, heart rate and temperature, and the presence of edema on the pharmacokinetics of amoxicillin but following intravenous administration to pregnant women. Our study revealed that maternal age correlated positively with the amoxicillin concentration in maternal serum (Table 2). This observation may be related to differences in the amount of fat tissue depending on age or blood vessel status. We found that gestational weight gain was negatively associated with maternal serum levels of amoxicillin (Table 2). According to the literature, hydrophilic drugs, such as amoxicillin, penetrate adipose tissue poorly; this phenomenon is more common among individuals with a higher BMI or higher body weight gain [10–12]. In addition, tissue perfusion may be lower in individuals with excessive body weight [10]. Unfortunately, none of these studies were related to amoxicillin. In our study, we also found no direct relationship between the maternal BMI and the amoxicillin levels in the individual compartments. It is likely that in women with a high gestational weight gain, the hydrophilic drug showed poorer penetration properties because of the increased adipose tissue. In our study, maternal serum hemoglobin level and red blood counts correlated positively with the amoxicillin concentration in the amniotic fluid (Table 2). Women with thrombocytopenia presented significantly higher cord blood amoxicillin concentrations, and that effect was not associated with any maternal anthropometric and rheological parameters or maternal age (Table 3). No relationship was identified between the maternal blood leukocyte counts and the amoxicillin levels in the tested compartments (results not enclosed). The only study referring to this topic is related to the pharmacokinetics of the aminoglycoside antibiotic amikacin. The authors compared the results in 10 immuno-compromised patients with hematological disorders and one patient with aplastic anemia with earlier studies on the pharmacokinetics of amikacin in healthy volunteers [13]. They observed that the volume of distribution and the clearance of amikacin were increased in patients with hematological disorders, including aplastic anemia, although the mechanism of this phenomenon remains unknown. The authors suggested a possible relationship with changes in the creatinine clearance, hemoglobin levels, temperature and serum albumin levels in patients with hematological diseases. In pregnant women, the distribution volume and drug clearance are
J Matern Fetal Neonatal Med, Early Online: 1–5
increased, and in cases of reduced erythrocyte parameters, these values are increased even more, which might explain our results. Neonatal birth weight was found to correlate positively with maternal serum and cord blood amoxicillin levels (Table 2). There are no reports in the literature regarding the administration of amoxicillin to pregnant women with high or low birth weight babies. The factors responsible for the tissue penetration of drugs may be the same as the factors affecting the birth weight, SGA or LGA, although further studies are required to confirm this hypothesis. We found significantly higher concentrations of amoxicillin in women with arterial hypertension compared with women with normal blood pressure (Table 3). Those women presented higher pre-pregnancy BMI, that parameter was not associated with amoxicillin concentration 2 h after vaginal administration. Unfortunately, the literature lacks reports on the effect of hypertension on the pharmacokinetics of beta-lactam antibiotics. Hypertension leads to vascular lesions, resulting in poorer drug penetration. However, it should be added that half of the hypertensive patients had pregnancy-induced hypertension, while the others had pre-existing hypertension, a condition observed in relatively young patients who had most likely not yet developed vascular lesions. Arterial hypertension in the course of pregnancy is often accompanied by edemas. Thus, amoxicillin, being a hydrophilic compound, may penetrate into tissues more easily, which would explain our results. No differences in amoxicillin levels were identified between patients with negative cervical swab cultures and patients with a bacterial presence in the cervical canal in our study. The available data suggest that infection leads to a better penetration of amoxicillin [2,14]. In conclusion, amoxicillin presented poor concentration in maternal–fetal compartments after vaginal administration. The highest concentrations of amoxicillin 2 h after vaginal administration were observed in the amniotic fluid; however, we suggest further investigations to examine different time points. We found higher amoxicillin levels in the amniotic fluid of hypertensive women and those with higher hemoglobin levels; however, this finding also needs to be examined further. The fact that maternal serum concentrations correlated positively with neonatal birth weight and negatively with gestational weight gain 2 h after vaginal administration warrant further investigations.
Declaration of interest This study was supported by Medical University of Warsaw, constitutional work. The authors report no declarations of interests.
References 1. Sykes R. Penicillin: from discovery to product. Bull World Health Organ 2001;79:778–79. 2. Miller EL. The penicillins: a review and update. J Midwifery Women’s Health 2002;47:426–33. 3. Goldberger MA, Walter RI, Lapid LS. Absorption of penicillin from the vagina. Am J Obstet Gynecol 1947;53:529–31.
DOI: 10.3109/14767058.2014.976548
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
4. Abel S, Farmer CJ, Doucette J. Observations on vaginal absorption of penicillin. Am J Obstet Gynecol 1948;55:461–68. 5. Rock J, Barker RH, Bacon WB. Vaginal absorption of penicillin. Science 1947;105:13. 6. Kundrat W. Methoden zur Bestimmung von AntibiotikaRuckstanden in tierischen Produkten. Zeitschr Anal Chem 1968; 243:624–30. 7. Belle G, Fisher LD, Heagerty PJ, Lumley T. Biostatistics. A methodology for the health sciences. New York: John Wiley & Sons, Inc.; 1993. 8. SAS Institute Inc. SAS/STATÕ 9.3 user’s guide: the MIXED procedure (chapter). Cary, NC: SAS Institute Inc.; 2011. 9. Muller AE, DeJongh J, Oostvogel PM, et al. Amoxicillin pharmacokinetics in pregnant women with preterm premature rupture of the membranes. Am J Obstet Gynecol 2008;198:108.e1–108.e6.
Vaginal amoxicillin in pregnancy
5
10. Bearden DT, Rodvold KA. Dosage adjustments for antibacterials in obese patients: applying clinical pharmacokinetics. Clin Pharmacokinet 2000;38:415–26. 11. Hanley MJ, Abernethy DR, Greenblatt DJ. Effect of obesity on the pharmacokinetics of drugs in humans. Clin Pharmacokinet 2010; 49:71–87. 12. Falagas ME, Karageorgopoulos DE. Adjustment of dosing of antimicrobial agents for bodyweight in adults. Lancet 2010;375: 248–51. 13. Kaojarern S, Maoleekoonpairoj S, Atichartakarn V. Pharmacokinetics of amikacin in hematologic malignancies. Antimicrob Agents Chemother 1989;33:1406–8. 14. Canafax DM, Yuan Z, Chonmaitree T, et al. Amoxicillin middle ear fluid penetration and pharmacokinetics in children with acute otitis media. Pediatr Infect Dis J 1998;17:149–56.
ORIGINAL ARTICLE
Concentration of amoxicillin in maternal serum, cord blood, amniotic fluid and the placenta after vaginal administration
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
Julia Zare˛ba-Szczudlik1, Ewa Romejko-Wolniewicz1, Zbigniew Lewandowski2, Hanna Ro´z_ an´ska3, and Krzysztof Czajkowski1 1
2nd Department of Obstetrics and Gynecology, 2Department of Epidemiology, Medical University of Warsaw, Warsaw, Poland, and 3Department of Hygiene of Food of Animal Origin of National Veterinary Research Institute in Puł awy, Poland
Abstract
Keywords
Objective: The aim of this study was to assess the amoxicillin concentration in maternal serum, cord blood, amniotic fluid and the placenta, 2 h following vaginal administration and the factors influencing the drug level. Methods: Twenty-eight full-term pregnant women who qualified for elective cesarean delivery were included in the study. Vaginal suppositories containing 250 mg of amoxicillin were administered 2 h prior to the operation. Amoxicillin levels were determined using the diffusion microbial assay. Results: The amoxicillin level in amniotic fluid was significantly higher in comparison to that of maternal serum, cord blood or the placenta. Maternal age positively and gestational weight gain negatively correlated with the amoxicillin concentration in maternal serum. The maternal serum hemoglobin level and red blood cell count were positively correlated with amoxicillin concentration in the amniotic fluid. Neonatal birth weight was positively correlated with maternal serum and cord blood amoxicillin levels. Hypertensive women had significantly higher amoxicillin concentrations in amniotic fluid, and women with thrombocytopenia presented significantly higher cord blood amoxicillin concentrations. Conclusions: Amoxicillin presented poor concentration in maternal–fetal compartments after vaginal administration, but the factors influencing the drug level in different compartments require further investigation.
Amoxicillin penetration, antibiotic vaginally, pregnancy
Introduction Amoxicillin is a beta-lactam antibiotic characterized by a broad spectrum of activity and a low toxicity [1,2]. For these reasons, it is widely used in obstetrics. According to the United States Food and Drug Administration (FDA), amoxicillin is a pregnancy category B drug. Most often, it is administered to pregnant women orally or intravenously. Reports of intravaginal administration of amoxicillin to treat vaginal or cervical infections with susceptible bacteria are rare and controversial. No studies have been published regarding the assessment of amoxicillin levels in fetal and afterbirth tissues following this route of administration. The profiles of patients in our hospital – women with highrisk pregnancy – are the reason we check genital tracts for the presence of many bacteria and fungi. According to cervical and vaginal culture, amoxicillin in the dose of 250 mg is the most frequently used intravaginal antibiotic in pregnant women in our hospital to treat local infections. Address for correspondence: Ewa Romejko-Wolniewicz, MD, PhD, 2nd Department of Obstetrics and Gynecology, Medical University of Warsaw, ul.Karowa 2, 00-315 Warsaw, Poland. Tel: +48507097063. Fax: +48225966487. E-mail: [email protected]
History Received 6 June 2014 Revised 15 September 2014 Accepted 10 October 2014 Published online 11 November 2014
The aim of this study was to assess the amoxicillin concentration in maternal serum, cord blood, amniotic fluid and the placenta 2 h after vaginal administration and to determine the factors that might affect this concentration. Goldberger et al. [3] estimated the concentration of penicillin after vaginal administration in 10 non-pregnant women and revealed that maximum level occurred 1 h after the application, but the therapeutic levels of penicillin were maintained for up to 3 h. Abel et al. [4] measured the levels of penicillin after vaginal administration at four time points and concluded that the degree of penicillin absorption from the vaginas of non-pregnant women was completely inconsistent and unpredictable. Rock et al. [5] concluded that patients who were near term absorbed little or no penicillin; the authors examined four patients and showed no absorption in two of them who were near term, little absorption in one near term (the level of penicillin was higher after 4 h than 1 h after application) and good absorption, similar to that of nonpregnant women, in one patient who was 2 months before term. Considering the results of the above studies and the maximum serum levels detected 2 h after oral and intramuscular administration [3–5], we decided to measure the amoxicillin levels 2 h after vaginal administration.
2
J. Zare˛ba-Szczudlik et al.
Materials and methods Twenty-eight full-term pregnant women (Weeks 37–41) who qualified for elective cesarean delivery were included in the study. Vaginal suppositories containing 250 mg of amoxicillin in a base of cocoa butter were administered 2 h prior to the cesarean section.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
Study protocol After admission, informed consent was obtained from each patient and the patient’s medical history was collected, a physical and obstetrical examination was performed and information from the pregnancy card was obtained. Each patient underwent a cervical swab for a bacterial culture assay. During the cesarean section, 5 mL of amniotic fluid was collected from each patient via aspiration for a bacterial culture assay. The following samples were collected during the cesarean section: 20 mL of amniotic fluid, 100 g of placenta, 9 mL of cord blood and 9 mL of maternal blood. The amniotic fluid and placenta samples were placed in plastic containers. After clotting, the maternal and umbilical blood samples were centrifuged, and the serum was filtered off and placed in plastic containers. The material was frozen at 20 C. Samples were sent monthly to the National Veterinary Research Institute in Puławy, where the antibiotic levels were determined. The study protocol was approved by the Ethics Committee of the Medical University of Warsaw.
J Matern Fetal Neonatal Med, Early Online: 1–5
Table 1. Maternal and neonatal characteristics.
Variable (unit)
Min
Max
SD, standard deviation; Min, the minimal value of variable; Max, the maximal value of variable; SGA, small for gestational age; LGA, large for gestational age. Table 2. Correlation of the amoxicillin concentration after vaginal administration with quantitative maternal and neonatal parameters.
Parameters
Amoxicillin levels were determined with the diffusion microbial assay described by Kundrat [6] using Kundrat agar and Bacillus stearothermophilus ATCC 7953 (SigmaAldrich Company Ltd. Poznan, Poland) as the test strain. The medium composition was as follows: peptone 17.0 g/L, NaCl 3.0 g/L, D-(+)-glucose 3.0 g/L, starch 3.0 g/L, gelatin 2.5 g/L, bromocresol purple 0.016 g/L, and agar 10.0 g/L, pH 6.8. The principle of the assay is growth inhibition of a spore suspension of the test strain B. stearothermophilus in the presence of antibiotics in the test sample. Tissue samples were placed in wells cut in the agar (10 mm in diameter) and incubated for 3 h at 64 C. The antibiotic level was determined based on the diameter of the test strain’s growth inhibition zones, with reference to the standard curve prepared using antibiotic-free material. The standard curve for determining the amoxicillin levels in the test samples was prepared using the reference substance from Sigma (cat. no. A8523, Sigma-Aldrich Company Ltd. Poznan, Poland) with an amoxicillin content of 900 mg/g (FDA-compliant). Each measurement was repeated four to five times depending on the quantity of available material. Arithmetic means were calculated from the obtained results.
Maternal age
Quantitative variables were presented as the mean values and standard deviations and as medians, quartiles and range. Qualitative variables were presented as frequency and percentage (Table 1). Comparisons of amoxicillin levels
Median
Age (years) 30.4 ± 4.2 31.0 22.0 38.0 Pre-pregnancy weight (kg) 65.7 ± 12.1 64.5 45.0 97.0 24.0 ± 4.4 22.9 17.6 33.3 Pre-pregnancy BMI (kg/m2) Gestational weight gain (kg) 14.1 ± 5.1 14.0 5.0 27.0 Multiparas 14 (50.0) History of cesarean section 13 (46.4) Pre-existing hypertension 2 (7.1) Pregnancy-induced hypertension 2 (7.1) Pre-existing diabetes 3 (10.7) Gestational diabetes 8 (28.6) Thrombocytopenia 6 (21.4) Serum hemoglobin level (g/dL) 11.9 ± 0.9 11.8 9.7 13.5 Red blood cell count (mln/mL) 3.9 ± 0.3 3.9 3.2 4.5 Bacteria in the cervix 7 (25.0) Gestational age upon delivery 38.3 ± 1.2 38.0 37.0 41.0 (weeks) Birth weight (g) 3534.6 ± 559.7 3550.0 2050.0 4480.0 Body length (cm) 54.8 ± 3.1 55.5 49.0 60.0 Boys 16 (57.1) SGA 2 (7.1) LGA 11 (39.3)
Drug level determination method
Statistical analysis
Mean ± SD or n (%)
Maternal Cord Amniotic serum blood Placenta fluid r r r R p p p p
0.54 0.10 0.16 0.003 ns ns Maternal pre-pregnancy weight 0.03 0.15 0.06 ns ns ns Maternal pre-pregnancy BMI 0.15 0.21 0.06 ns ns ns Gestational weight gain 0.41 0.06 0.27 0.029 ns ns Maternal serum hemoglobin level 0.05 0.14 0.05 ns ns ns Maternal red blood count 0.10 0.08 0.16 ns ns ns Neonatal birth weight 0.40 0.38 0.13 0.034 0.045 ns
0.17 ns 0.01 ns 0.02 ns 0.01 ns 0.41 0.035 0.46 0.017 0.31 ns
r, correlation coefficient; p, significance level (bold values).
between compartments were performed using the signed-rank test, and the amoxicillin levels in patient’s compartments were standardized for use in statistical analyses using the following coefficients: a serum specific gravity of 1.026 g/mL and an amniotic fluid specific gravity of 1.006 g/mL. Associations between quantitative variables were tested using correlation analysis (Spearman’s correlation coefficient; Table 2). Amoxicillin levels between subgroups distinguished according to the considered maternal parameters were compared using a two-tailed Mann–Whitney test for independent groups (Table 3). For revealed qualitative predictors of amoxicillin concentration, the effect was tested for possible confounding factors with Wilcoxon two-sample test.
Vaginal amoxicillin in pregnancy
DOI: 10.3109/14767058.2014.976548
3
Table 3. Amoxicillin concentration after vaginal administration in the different compartments related to qualitative maternal parameters. Maternal serum Parameters
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
Diabetes Yes No Hypertension Yes No Thrombocytopenia Yes No Cervical culture Yes No
Cord blood
Placenta
Amniotic fluid
Med (Q1–Q3)
p
Med (Q1–Q3)
p
Med (Q1–Q3)
p
Med (Q1–Q3)
p
9.0 (6.0–12.0) 9.0 (8.0–10.0)
ns
10.0 (9.0–12.0) 8.0 (8.0–11.0)
ns
11.0 (10.0–12.0) 10.0 (9.0–12.0)
ns
37.0 (25.0–39.0) 28.0 (12.5–38.0)
ns
8.0 (7.0–10.0) 9.0 (7.5–10.5)
ns
10.5 (8.5–12.5) 9.0 (8.0–11.0)
ns
10.0 (7.5–11.0) 10.0 (9.0–12.0)
ns
39.0 (38.0–45.5) 29.0 (16.3–38.0)
0.038
10.0 (7.0–12.5) 8.5 (8.0–10.0)
ns
12.0 (10.0–12.0) 9.0 (8.0–10.0)
0.050
11.0 (9.0–12.0) 10.0 (9.0–12.0)
ns
31.0 (12.5–38.0) 33.0 (22.0–39.0)
Ns
9.0 (7.0–11.0) 9.0 (8.0–10.0)
ns
10.0 (9.0–12.0) 9.0 (8.0–11.0)
ns
10.0 (9.0–12.0) 10.0 (9.0–12.0)
ns
38.0 (12.5–39.0) 32.0 (25.0–38.0)
Ns
Med, median; Q1, lower quartile; Q3, upper quartile; p, significance level (bold values).
The statistical analyses were performed according to manuals by L. D. Fisher and Gerald van Belle [7] and the SAS system [8].
Results Maternal and neonatal characteristics The maternal and neonatal characteristics are included in Table 1. Seven women were diagnosed with bacteria in the cervix, while one was diagnosed with amniotic fluid infection. No child was diagnosed with an infection. The Apgar scores at 3 and 5 min were between 8 and 10 points for all newborns. Amoxicillin concentration The concentrations of amoxicillin in different compartments 2 h after vaginal application were the following: 29.9 ± 12.70.013 ng/g in the amniotic fluid, 10.6 ± 9.9 ng/g in maternal serum, 10.2 ± 1.8 ng/g in the placenta and 9.3 ± 2.3 ng/g in the cord blood. The level of amoxicillin in the amniotic fluid was significantly higher than that in the other three tissues (p50.05). There were no significant differences among the drug levels in maternal serum cord blood and the placenta. Maternal age positively correlated with the amoxicillin concentration in maternal serum (r ¼ 0.54, p ¼ 0.033; Table 2). Gestational weight gain but not maternal prepregnancy weight and BMI was negatively associated with maternal serum levels of amoxicillin (r ¼ 0.41, p ¼ 0.029; Table 2). Maternal serum hemoglobin level and red blood count were positively correlated with amoxicillin concentration in amniotic fluid (r ¼ 0.41, p ¼ 0.035 and r ¼ 0.46, p ¼ 0.017, respectively; Table 2). Neonatal birth weight was positively correlated with maternal serum and cord blood amoxicillin levels (r ¼ 0.40, p ¼ 0.034 and r ¼ 0.38, p ¼ 0.045, respectively; Table 2). Diabetes had no influence on the amoxicillin concentrations in any of the examined compartments. Hypertensive women had significantly higher amoxicillin concentrations in amniotic fluid (p ¼ 0.038; Table 3). This effect was tested for
three parameters: maternal age, pre-pregnancy BMI, gestational weight gain and only BMI was associated with hypertension (BMI in hypertensive group was significantly higher: 26.9 versus 23.5 kg/m2, respectively, p ¼ 0.02). Women with thrombocytopenia (platelet count below 150 103/mL) presented significantly higher cord blood amoxicillin concentrations (p ¼ 0.05; Table 3). This effect was tested for five parameters: maternal age, pre-pregnancy BMI, gestational and weight gain, serum hemoglobin level, gestational age and none of these was associated with thrombocytopenia. No differences in amoxicillin levels were identified between patients with negative cervical swab cultures (21 cases) and patients with bacteria in the cervical canal (7 cases; Table 3). No significant differences were seen based on parity, gestational age at delivery and the gender of the newborn.
Discussion The aim of our study was also to recognize factors that might affect the concentrations of amoxicillin in different compartments. There are studies assessing amoxicillin levels in different tissues at different time points after administration to different patients. Based on these studies, a curve was drawn for tissue antibiotic levels as a function of time. However, results from different patients do not allow unambiguous conclusions or an evaluation of factors that might affect the antibiotic levels. In our study, we examined amoxicillin levels in four compartments of every patient at one time point following vaginal administration. We had only one opportunity to measure the level of amoxicillin in the amniotic fluid and the placenta during the cesarean section, so we decided to measure the levels in the other compartments at the same time. Small amounts of intravaginally administered amoxicillin were detected in the tested samples, which is not surprising considering that amoxicillin is a hydrophilic substance that passes through the cell membrane poorly. Therefore, one may conclude that amoxicillin, when administered vaginally, penetrates into maternal and fetal tissues poorly, although its presence in the amniotic fluid suggests its capability to
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
4
J. Zare˛ba-Szczudlik et al.
penetrate into this compartment. Most likely, vaginal administration of higher doses would result in higher concentrations in the amniotic fluid; further studies are required to confirm this hypothesis. Unfortunately, it is impossible to compare our results with other results as there are no reports on the vaginal use of amoxicillin in the literature. Our study revealed no significant differences in the amoxicillin levels in the maternal serum, cord blood, placenta and amniotic fluid 2 h after vaginal administration, depending on parity, maternal prepregnancy weight, maternal prepregnancy BMI, diabetes during pregnancy, gestational age at delivery and the gender of the newborn. Muller et al. [9] also observed no effect of the gestational age, maternal body weight and BMI, maternal blood pressure, heart rate and temperature, and the presence of edema on the pharmacokinetics of amoxicillin but following intravenous administration to pregnant women. Our study revealed that maternal age correlated positively with the amoxicillin concentration in maternal serum (Table 2). This observation may be related to differences in the amount of fat tissue depending on age or blood vessel status. We found that gestational weight gain was negatively associated with maternal serum levels of amoxicillin (Table 2). According to the literature, hydrophilic drugs, such as amoxicillin, penetrate adipose tissue poorly; this phenomenon is more common among individuals with a higher BMI or higher body weight gain [10–12]. In addition, tissue perfusion may be lower in individuals with excessive body weight [10]. Unfortunately, none of these studies were related to amoxicillin. In our study, we also found no direct relationship between the maternal BMI and the amoxicillin levels in the individual compartments. It is likely that in women with a high gestational weight gain, the hydrophilic drug showed poorer penetration properties because of the increased adipose tissue. In our study, maternal serum hemoglobin level and red blood counts correlated positively with the amoxicillin concentration in the amniotic fluid (Table 2). Women with thrombocytopenia presented significantly higher cord blood amoxicillin concentrations, and that effect was not associated with any maternal anthropometric and rheological parameters or maternal age (Table 3). No relationship was identified between the maternal blood leukocyte counts and the amoxicillin levels in the tested compartments (results not enclosed). The only study referring to this topic is related to the pharmacokinetics of the aminoglycoside antibiotic amikacin. The authors compared the results in 10 immuno-compromised patients with hematological disorders and one patient with aplastic anemia with earlier studies on the pharmacokinetics of amikacin in healthy volunteers [13]. They observed that the volume of distribution and the clearance of amikacin were increased in patients with hematological disorders, including aplastic anemia, although the mechanism of this phenomenon remains unknown. The authors suggested a possible relationship with changes in the creatinine clearance, hemoglobin levels, temperature and serum albumin levels in patients with hematological diseases. In pregnant women, the distribution volume and drug clearance are
J Matern Fetal Neonatal Med, Early Online: 1–5
increased, and in cases of reduced erythrocyte parameters, these values are increased even more, which might explain our results. Neonatal birth weight was found to correlate positively with maternal serum and cord blood amoxicillin levels (Table 2). There are no reports in the literature regarding the administration of amoxicillin to pregnant women with high or low birth weight babies. The factors responsible for the tissue penetration of drugs may be the same as the factors affecting the birth weight, SGA or LGA, although further studies are required to confirm this hypothesis. We found significantly higher concentrations of amoxicillin in women with arterial hypertension compared with women with normal blood pressure (Table 3). Those women presented higher pre-pregnancy BMI, that parameter was not associated with amoxicillin concentration 2 h after vaginal administration. Unfortunately, the literature lacks reports on the effect of hypertension on the pharmacokinetics of beta-lactam antibiotics. Hypertension leads to vascular lesions, resulting in poorer drug penetration. However, it should be added that half of the hypertensive patients had pregnancy-induced hypertension, while the others had pre-existing hypertension, a condition observed in relatively young patients who had most likely not yet developed vascular lesions. Arterial hypertension in the course of pregnancy is often accompanied by edemas. Thus, amoxicillin, being a hydrophilic compound, may penetrate into tissues more easily, which would explain our results. No differences in amoxicillin levels were identified between patients with negative cervical swab cultures and patients with a bacterial presence in the cervical canal in our study. The available data suggest that infection leads to a better penetration of amoxicillin [2,14]. In conclusion, amoxicillin presented poor concentration in maternal–fetal compartments after vaginal administration. The highest concentrations of amoxicillin 2 h after vaginal administration were observed in the amniotic fluid; however, we suggest further investigations to examine different time points. We found higher amoxicillin levels in the amniotic fluid of hypertensive women and those with higher hemoglobin levels; however, this finding also needs to be examined further. The fact that maternal serum concentrations correlated positively with neonatal birth weight and negatively with gestational weight gain 2 h after vaginal administration warrant further investigations.
Declaration of interest This study was supported by Medical University of Warsaw, constitutional work. The authors report no declarations of interests.
References 1. Sykes R. Penicillin: from discovery to product. Bull World Health Organ 2001;79:778–79. 2. Miller EL. The penicillins: a review and update. J Midwifery Women’s Health 2002;47:426–33. 3. Goldberger MA, Walter RI, Lapid LS. Absorption of penicillin from the vagina. Am J Obstet Gynecol 1947;53:529–31.
DOI: 10.3109/14767058.2014.976548
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Universitat de Girona on 12/19/14 For personal use only.
4. Abel S, Farmer CJ, Doucette J. Observations on vaginal absorption of penicillin. Am J Obstet Gynecol 1948;55:461–68. 5. Rock J, Barker RH, Bacon WB. Vaginal absorption of penicillin. Science 1947;105:13. 6. Kundrat W. Methoden zur Bestimmung von AntibiotikaRuckstanden in tierischen Produkten. Zeitschr Anal Chem 1968; 243:624–30. 7. Belle G, Fisher LD, Heagerty PJ, Lumley T. Biostatistics. A methodology for the health sciences. New York: John Wiley & Sons, Inc.; 1993. 8. SAS Institute Inc. SAS/STATÕ 9.3 user’s guide: the MIXED procedure (chapter). Cary, NC: SAS Institute Inc.; 2011. 9. Muller AE, DeJongh J, Oostvogel PM, et al. Amoxicillin pharmacokinetics in pregnant women with preterm premature rupture of the membranes. Am J Obstet Gynecol 2008;198:108.e1–108.e6.
Vaginal amoxicillin in pregnancy
5
10. Bearden DT, Rodvold KA. Dosage adjustments for antibacterials in obese patients: applying clinical pharmacokinetics. Clin Pharmacokinet 2000;38:415–26. 11. Hanley MJ, Abernethy DR, Greenblatt DJ. Effect of obesity on the pharmacokinetics of drugs in humans. Clin Pharmacokinet 2010; 49:71–87. 12. Falagas ME, Karageorgopoulos DE. Adjustment of dosing of antimicrobial agents for bodyweight in adults. Lancet 2010;375: 248–51. 13. Kaojarern S, Maoleekoonpairoj S, Atichartakarn V. Pharmacokinetics of amikacin in hematologic malignancies. Antimicrob Agents Chemother 1989;33:1406–8. 14. Canafax DM, Yuan Z, Chonmaitree T, et al. Amoxicillin middle ear fluid penetration and pharmacokinetics in children with acute otitis media. Pediatr Infect Dis J 1998;17:149–56.
