Eur J Cardiothorac Surg 2004;25:256-260
© 2004 Elsevier Science NL
Pharmacokinetics of intravenous flucloxacillin and amoxicillin in neonatal and infant cardiopulmonary bypass surgery
Manuel R. Adrianzén Vargasa,
Mark H. Dantona,
Sheikh M. Javaida,
Jim Grayb,
Caroline Tobinc,
William J. Brawna,
David J. Barrona*
a Department of Cardiac Surgery, Birmingham Children's Hospital, Steelhouse Lane, Birmingham B4 6NH, UK
b Department of Microbiology, Birmingham Children's Hospital, Steelhouse Lane, Birmingham B4 6NH, UK
c Bristol Centre for Antimicrobial Research and Evaluation, Department of Medical Microbiology, Southmead Hospital, Bristol, UK
Received 1 August 2003;
received in revised form 12 November 2003;
accepted 16 November 2003.
* Corresponding author. Tel.: +44-121-333-9437; fax: +44-121-333-9441
e-mail: david.barron{at}bhamchildrens.wmids.nhs.uk
 |
Abstract
|
|---|
Objective: To determine the blood and tissue concentrations of flucloxacillin and amoxicillin during cardiopulmonary bypass (CPB) in infants weighing less than 5 kg. Methods: Eleven patients aged between 3 and 60 days and weighing <5 kg. Intravenous flucloxacillin 30 mg kg-1 and amoxicillin 30 mg kg-1 were administered at time of anaesthesia. Blood and muscle samples were collected at four stages of the operation: immediately before commencement of CPB; before cross-clamping; after weaning of CPB; and at the time of skin closure. Concentrations, half-lives (t1/2), clearance and volume of distribution were calculated for both antibiotics in serum and tissue. Results: After connection to CPB the mean serum concentration of flucloxacillin decreased by 42.5% from 75.5 to 43.4 mg l-1 (P=0.003) and that of amoxicillin decreased by 36.2% from 73.3 to 46.7 mg l-1 (P=0.021). Serum concentrations of the two antibiotics decreased by a further 16.5 and 14.5% during the remainder of the surgery, but remained >15-fold above the expected minimum inhibitory concentration (MIC) for target pathogens. Muscle concentrations of both antibiotics reached MIC values by the time of the first sample and there was no decrease associated with connection to CPB. Levels remained >8-fold above MIC for target pathogens throughout the procedure. The t1/2 for flucloxacillin was 2.64(±0.23) h and for amoxicillin was 3.16(±0.29) h, both of which are more than double the values in normal adults. There was an equivalent reduction in clearance for both antibiotics. Conclusions: Single doses of flucloxacillin and amoxicillin at 30 mg kg-1 maintain serum and muscle concentrations well above the MIC throughout cardiac surgery. This is partly due to a prolonged t1/2 and reduced clearance of both antibiotics in infants.
Key Words: Antibiotics • Cardiopulmonary bypass • Neonates
|
1. Introduction
|
|---|
The prophylactic administration of antibiotics to patients undergoing cardiac surgery is an accepted practice. The commonest causes of infection after cardiac surgery are wound and respiratory infections [1]. The standard regimen for patients undergoing cardiac surgery in this institution is flucloxacillin and amoxicillin, each administered in a dose of 30 mg kg-1 IV on anaesthetic induction and then every 8 h. Flucloxacillin covers Staphylococcus aureus and amoxicillin is used primarily to cover common respiratory pathogens such as Streptococcus pneumoniae and Haemophilus influenzae.
Cardiopulmonary bypass (CPB) can affect antibiotic disposition in serum and tissues due to the physiological changes it causes, including hypotension, hypothermia, haemodilution and decreased hepatic and renal perfusion [2,3]. These changes can also lead to an altered protein binding and drug distribution and the result can be inadequate antibiotic concentrations [6–8]. It cannot be assumed that observations in adults undergoing surgery are pertinent to children. The haemodilution effect of CPB is greater in children than in adults because of the relatively large priming volume compared to their circulating volume, which in neonates can be up to three times their normal blood volume. Furthermore, drug handling in neonates and infants differs from the older age groups, and half-lives are often markedly prolonged [4].
To date there is very limited evidence of the pharmacokinetics of antibiotics in CPB in children. Hatzopoulus et al. [5] found a marked decrease (44.5%) in serum vancomycin levels at the initiation of CPB which they attributed to haemodilution. No reports about the pharmacokinetics of flucloxacillin or amoxicillin in paediatric CPB are available, although studies evaluating cloxacillin [6] and flucloxacillin [7,8] in adults have shown that the concentrations of these antibiotics decrease after CPB is initiated. Despite giving a second dose during CPB, the levels were inadequate by the end of the procedure in >60% of the patients [5,8].
The aim of this investigation was to determine whether a single dose of 30 mg kg-1 of flucloxacillin and amoxicillin given on anaesthetic induction provided adequate serum and tissue concentration during and immediately after CPB in small infants weighing <5 kg.
|
2. Patients and methods
|
|---|
Eleven patients under 5 kg aged between 3 and 60 days undergoing cardiac surgery for correction of congenital heart defects under CPB were recruited. Parental informed consent for the study was obtained. The South of Birmingham Local Research Ethics Committee approved the research protocol. The standard antibiotic protocol of this unit is to give amoxicillin and flucloxacillin 30 mg kg-1 at induction of anaesthesia and then 8 hourly for 48 h following surgery.
After the induction of anaesthesia, 30 mg kg-1 of flucloxacillin and 30 mg kg-1 of amoxicillin were administered as a bolus injection through a central vein catheter. The cardiopulmonary bypass circuit was primed with 30 ml kg-1 of crystalloid (Plasma-lyte® isotonic electrolyte solution, Baxter Healthcare, Norfolk, UK), 200 ml of fresh frozen plasma and 150 ml of red cells. Bypass was performed using non-pulsatile flow (Sarns 8000® pump system, Terumo, Tokyo, Japan). The flow was regulated accordingly with the temperature and the requirements of the surgeon. A short-term cardio circulatory arrest was used depending on surgical requirements. Modified ultrafiltration is not standard practice in this unit and was not used in any of the study patients.
Blood samples consisting of 1 ml of blood from the central vein catheter were taken as follows:
- Immediately before the CPB was commenced
- After CPB was started and before the aorta was cross-clamped
- After CPB was weaned and before any other volume was given
- At the time of skin closure.
Tissue samples of approximately 2 mm3 from the rectus abdominus muscle were taken at the same time. This muscle was chosen as it was immediately accessible from the sternotomy incision and was a large muscle mass with a rich, non-polar blood supply. The muscle samples were transported in a dry ice container, then frozen in liquid nitrogen and stored at -70 °C. The serum and tissue levels of flucloxacillin and amoxicillin were measured using high performance liquid chromatography (HPLC) [9]. The adequacy of the levels of antibiotics was established using the minimum inhibitory concentration (MIC) of flucloxacillin for Staphylococcus aureus (not methicillin-resistant) and the MIC of amoxicillin for Haemophilus influenzae as follows: flucloxacillin level more than 2 mg l-1 in serum and 1 mg kg-1 in muscle, amoxicillin level more than 1 mg l-1 in serum and 0.25 mg kg-1 in muscle [10].
Serum levels of 4x the MIC would generally be accepted as sufficient concentration to ensure tissue penetration [10].
Pharmacokinetic parameters for serum half-life (t1/2), volume of distribution (Vd), systemic clearance (CL) were calculated with a one-compartment model. These pharmacokinetic parameters have skewed distributions and so statistical assessment of changes within individuals was done using Wilcoxon signed rank tests on paired values with medians and ranges as summary statistics of groups. Statistical significance was accepted if P<0.05.
|
3. Results
|
|---|
Clinical data and bypass times are shown in Table 1. The mean age of the patients was 25.9 (±7.1) days and mean weight was 3.6 (±0.2) kg. The pharmacokinetic data are summarised in Table 2 together with standard reference values in adult populations. No patient had clinical evidence of sepsis in the early post-operative period and there was no growth on all cultures taken in the first 48 h post-operatively.
Mean serum concentrations of flucloxacillin and amoxicillin are shown in Fig. 1a and b
, respectively. Mean concentrations of both antibiotics fell markedly on institution of CPB (42.5% for flucloxacillin and 36.2% for amoxicillin) but values remained well above the MIC. Thereafter, serum values of flucloxacillin and amoxicillin decreased only by a further 16 and 14%, respectively, throughout the remaining period of the surgery. Mean antibiotic concentrations remained well in excess of the recommended MIC values for target pathogens, indeed, remaining more than 15 times over the MIC at all time points. Furthermore, no single value from any patient fell below the recommended level for either antibiotic.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1. (a) Serum concentrations of flucloxacillin. (b) Serum concentrations of amoxicillin. Phase 1, sample taken pre-CPB; phase 2, sample taken on CPB; phase 3, sample taken before weaning from CPB; phase 4, sample taken post-CPB.
|
|
The muscle concentrations of flucloxacillin and amoxicillin are shown in Fig. 2a and b
, respectively. Initial concentrations were 15–30% of serum concentrations but the onset of CPB has a much less marked effect on tissue concentrations with a decrease in tissue concentration of 11% for amoxicillin and 4% for flucloxacillin. Concentrations tended to gradually decrease during the procedure although the changes were only significant for flucloxacillin that decreased from 12.7 to 7.7 mg kg-1 by phase 4 (P=0.01). Mean concentrations for both antibiotics in the muscle remained at over 20 times the recommended MIC for target pathogens throughout the procedures. Again, not a single value in any patient strayed below the MIC level for muscle concentration.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 2. (a) Muscle concentrations of flucloxacillin. (b) Muscle concentrations of amoxicillin. Phase 1, sample taken pre-CPB; phase 2, sample taken on CPB; phase 3, sample taken before weaning from CPB; phase 4, sample taken post-CPB.
|
|
Pharmacokinetics of both antibiotics differed from the standard reference values in adults (Table 2). The t1/2 of both flucloxacillin and amoxicillin was prolonged to between 2 and 3 times expected, and the clearance was reduced by 25–50%. The volume of distribution was greater than reference values for both antibiotics, particularly with flucloxacillin which usually has a lower Vd than amoxicillin due to greater protein binding.
|
4. Discussion
|
|---|
Numerous studies have reported the effect of CPB surgery on the disposition of antibiotic drugs in adults [1–3,6–8] but there remains limited information regarding paediatric practice [5].
CPB has a complex effect on drug pharmacokinetics through haemodilution, hypothermia, diminished protein binding, reduced clearance and changes in volume of distribution. Many studies of ß-lactam antibiotics in adults undergoing CPB surgery have found that reduced clearance does not counterbalance the initial haemodilution effect, so that serum and tissue concentrations after a single dose prophylaxis administered before CPB are inadequate by the end of surgery [6,7,11]. Thus, administration of additional doses of prophylaxis has been recommended [6,8,11]. It cannot be assumed that the distribution of antibiotics in infants undergoing CPB surgery will be the same as in adults. Differences in body composition and organ function significantly affect antibiotic pharmacokinetics in young infants, tending to lead to decreased drug clearance and increased volume of distribution [4].
In this study the initial serum concentrations of flucloxacillin and amoxicillin were similar to those seen in adults after administration of high doses of these antibiotics [12]. The subsequent decrease in serum levels associated with CPB is consistent in magnitude with previous experience of ß-lactam antibiotics in adults. The relatively greater haemodilution effect in infants was expected to have resulted in a greater decrease than seen in adults and the similarity may reflect the longer half-life and reduced clearance.
We observed that 30 mg kg-1 doses of intravenous flucloxacillin and amoxicillin provided serum concentrations well in excess of the MIC values for target pathogens throughout CPB surgery. Even though the serum levels decreased abruptly for both antibiotics when CPB was commenced, the serum concentrations remained more than 15 times above the MIC at the end of the procedure. Serum concentrations that achieve anything greater than 4x the MIC are widely accepted as being adequate since these correspond to the break point concentrations for these antibiotics when performing antibiotic susceptibilities [10].
Muscle concentrations showed more variation, although they also remained well above the recommended MIC for the target pathogens. These levels of more than 8-fold the MIC were maintained throughout the procedures. This was encouraging in view of previous reports in adult patients that suggest that serum and tissue levels of flucloxacillin may be inadequate by the end of the operation [6–8]. The improvement in antibiotic concentrations seen in this study may be due to longer t1/2 in infants and also the relatively higher dose (30 mg kg-1) compared to the standard adult dose used in most studies (8–16 mg kg-1). The dosage chosen in this study is the standard recommended paediatric dose.
Pharmacokinetic parameters for both antibiotics are significantly different compared to those in normal adult subjects. Diminished renal perfusion and renal function caused by the combination of hypotension and hypothermia that occurred during CPB and the use of hypothermic circulatory arrest [2,3] are part of the explanation of these altered values. However, differences in neonatal drug handling may also be important. Flucloxacillin and amoxicillin are cleared by renal excretion [13] through a combination of glomerular filtration, active tubular secretion and passive tubular reabsorption. In healthy neonates and very young infants, glomerular filtration is the main renal mechanism of drug elimination [13]. It depends directly on renal blood flow and the amount of glomeruli. In neonates, glomerular filtration is just 30% of that observed in normal adults, due mainly to developmental immaturity [14,15]. Total renal blood flow can be significantly reduced during CPB, particularly during hypothermia when circulating cardiac output is reduced; this leads to decreased drug clearance and metabolism during CPB surgery [3,5,6]. These observations of prolonged t1/2 are similar to those reported in neonatal units [16,17] although reflecting the additive effects of CPB.
Antibiotic levels in the muscle were much less affected by the onset in CPB when compared to serum levels, although they then showed a gradual decline during the following 2 h. Since the diffusion of the drug to the tissue depends on the concentration of the unbound fraction of the drug in plasma, plasma protein binding may be an important factor [18]. Plasma protein binding may be reduced in the neonate since they have lower plasma albumen than adults and have saturation of their protein binding sites at lower drug concentrations than older infants and children. This reflects a quantitative difference in the number of binding sites, as well as a qualitative difference in receptor affinity [18]. In addition, CPB itself can reduce plasma protein binding [6,19]. The degree of protein binding is inversely related to the volume of distribution [19], thus the decreased protein binding may also partly explain the increased Vd for flucloxacillin of 0.45 l kg-1 (versus 0.13 l kg-1 in adults). The additional volume of the CPB circuit may also explain the increased Vd for flucloxacillin.
Much of drug dosing in neonates has been calculated by extrapolating data from adults. This concept is gradually changing as we learn more about neonatal physiology and how immaturity and developmental issues fundamentally alter the pharmacokinetics of drugs in neonates.
In summary, the pharmacokinetics of flucloxacillin and amoxicillin in babies less than 5 kg undergoing CPB surgery is altered. Mean half-life (t1/2) is increased and CL is diminished for both antibiotics. Using currently widely used dosages, the serum and muscle concentrations of both antibiotics remained well above the MIC for target pathogens throughout CPB.
|
Acknowledgments
|
|---|
This study was supported by grants from the Research and Development Committee from the Birmingham Children's Hospital. We acknowledge Dr P Davis, PhD, for his assistance with the statistical analysis of the data, Annette O'Hara and Lisa Chance for assistance with data collection.
|
References
|
|---|
- Matsumoto H., Yamashita M., Iguro Y., Miyazaki T. Postoperative infections in cardiovascular surgery: what is the appropriate prophylactic administration of antibiotics?. Kyobu Geka 1995;46(5):375-379.
- Buylaert W.A., Herregods L.L., Mortier E.P., Bogaert M.G. Cardiopulmonary bypass and the pharmacokinetics of drugs: an update. Clin Pharmacokinet 1989;17:10-26.[Medline]
- Holley F.O., Ponganis K.V., Stanski D.R. Effect of cardiopulmonary bypass on the pharmacokinetics of drugs. Clin Pharmacokinet 1982;7:234-251.[Medline]
- Paap C.M., Nahata M.C. Clinical pharmacokinetics of antibacterial drugs in neonates. Clin Pharmacokinet 1990;19:280-318.[Medline]
- Hatzopoulos F., Stile-Calligaro I., Rodvold K., Sullivan-Bolyai J., Del Nido P., Levitsky S. Pharmacokinetics of intravenous Vancomycin in pediatric cardiopulmonary bypass surgery. Pediatr Infect Dis J 1993;12:300-304.[Medline]
- Kullberg B.J., Mattie H., Huysmans H., Van Furth R. Evaluation of Cloxacillin concentrations in plasma and muscle tissue during cardiopulmonary bypass. Scand J Infect Dis 1991;23:223-238.
- Farrington M., Fenn A., Philips I. Flucloxacillin concentration in serum and wound exudates during open heart surgery. J Antimicrob Chemother 1985;16:253-259.[Abstract/Free Full Text]
- Wilson A., Taylor B., Treasure T., Gruneberg R., Patton K., Femingham D., Sturridge M. Antibiotic prophylaxis in cardiac surgery: serum and tissue levels of teicoplanin, flucloxacillin and tobramycin. J Antimicrob Chemother 1988;21:201-212.[Abstract/Free Full Text]
- White L.O., Lovering A.M. Non-microbiological assays. In: Reeves D.S., Wise R., Andrews J.M., White L.O., eds. Clinical antimicrobial assays. Oxford: Oxford University Press, 1999:45-63.
- The British Society for Antimicrobial Chemotherapy (BSAC). Standardised disc testing method. On-line version www.bsac.org.uk.
- Menges T., Sablotski A., Welters I., Wagner R.M., Zickmann B., Gronau S. Concentration of cefamandole in plasma and tissues of patients undergoing cardiac surgery: the influence of different cefamandole dosage. J Cardiothorac Vasc Anesth 1997;11:565-570.[CrossRef][Medline]
- Kucers N.M.K., Bennett R.J. The use of antibiotics: a comprehensive review with clinical emphasis. . Woburn, MA: Butterworth, 1987.
- Yaffe S., Aranda J. Pediatric pharmacology: therapeutic principles in practice. . 1992.
- Aranda J.V., Stern L. Clinical aspects of developmental pharmacology and toxicology. Pharmacol Ther 1983;20:1-15.[CrossRef][Medline]
- Robillard J.E. Maturational changes in the fetal glomerular filtration rate. Am J Obstet Gynecol 1975;122:601-606.[Medline]
- Charles B.G., Preechagoon Y., Lee T.C., Steer P.A., Flenady V.J., Debuse N. Population pharmacokinetics of intravenous amoxicillin in very low birth weight infants. J Pharm Sci 1997;86:1288-1292.[CrossRef][Medline]
- Hengren L., Ehrnebo M., Broberger U. Pharmacokinetics of free and total flucloxacillin in newborn infants. Eur J Clin Pharmacol 1987;32:403-409.[CrossRef][Medline]
- Heimann G. Basic pharmacokinetic principles. In: Polin R.A., Fox W.W., eds. Fetal and neonatal physiology. Philadelphia: W.B. Saunders. 1998:137-146.
- Chemtob S., Aranda J. Basic pharmacologic principles. In: Polin R.A., Fox W.W., eds. Fetal and neonatal physiology. Philadelphia: W.B. Saunders. 1998:125-137.
Top
Abstract
1. Introduction
