Antimicrobial dosing in neonates

KEYWORDS:

• Neonates
• antibiotics
• pharmacokinetics
• pharmacodynamics
• clinical trials

Neonatal sepsis still remains a significant cause of morbidity and mortality, prolonged hospitalization, and increased medical costs. Early- and late-onset sepsis often presents with subtle and nonspecific clinical signs and, if not promptly treated, may have detrimental consequences including survival with neurodevelopmental deficits or even death [1–3]. Εmpiric antimicrobial therapy is an undeniable part of treatment in neonates with suspected sepsis. Beta-lactams, aminoglycosides and glycopeptides are among the most commonly prescribed drugs in neonates, while antifungal and antiviral agents are used much less frequently.

Recently, data on antibiotic prescription in pediatric population from the Antibiotic Resistance and Prescribing in European Children project were published (www.arpecproject.eu). In the latter study involving 240 neonatal intensive care units (NICU) from 41 countries, antibiotics were prescribed in 89.2% out of the 3515 neonates evaluated followed by antifungals (8.8%) and antivirals (1.9%). Of note, gentamicin combined with ampicillin/amoxicillin or benzylpenicillin were by far the most frequently used regimens [4]. Moreover, therapeutic administration was recorded in 75% of the antibiotics prescribed in neonates, for reasons like sepsis, central line-associated blood stream infections, central nervous system and skin/soft tissue infections, surgical disease, whereas prophylactic administration (25%) was given for maternal and newborn risk factors, for medical problems and surgical disease [4]. Treatment strategies for neonatal sepsis significantly vary in terms of antibiotic choice and therapy duration among neonatal centers. As documented in a cross-sectional survey in 19 European countries, 20 different antibiotic combinations were recorded, with median therapy duration for suspected sepsis of 7 (IQR 3–10) days [5]. A wide variation in antibiotic prescribing practices among NICUs was also revealed in a large antibiotic use cohort study in California (127 NICUs, 52,061 infants and 746,051 patient-days) [6].

1. Off-label use of antibiotics

Lack of evidence-based data is the main cause for the off-label/unlicensed antibiotic use in neonates. Limited knowledge exists in (a) pharmacokinetics (PK) and safety, (b) route (continuous or intermittent), dose and duration of administration, and (c) guidelines of using specific antibiotic(s). A recent prospective cross-sectional study conducted in 36 Italian NICUs confirmed the high prevalence of off-label/unlicensed drug use in neonates. Anti-infective agents were prescribed off-label in 75% of cases and the reasons of off-label use were age, dose, dosing frequency, and formulation [7]. In a Portuguese NICU, although ampicillin and gentamicin were the most frequently prescribed antibiotics, drug doses and/or frequencies used were clearly different from those specified in the Summary of Product Characteristics [8]. This finding is in line with the results of a previous survey involving 3 European neonatal centers in which an unregistered dose use of antibiotics was recorded in 37.8-51.7% of the prescriptions [8,9].

2. High variability in antibiotic dosing

Due to the lack of evidence-based data, the dosing of commonly used antibiotics exhibits high inter-NICU variability. Guidelines are mainly based on small studies and personal opinion rather than on randomized clinical trials (RCTs), leading to broad variability in dosing recommendations among hospitals. Dosing guidelines are inconsistent and incomplete resulting in inappropriate prescription of the majority of antibiotics given to critically ill neonates. High variability in dosing of the most frequently prescribed antibiotics has been documented in 92% of the prescriptions of 89 NICUs in 21 European countries [10]. In a recent French national survey, a significant inter-NICU variability was also identified as a total of 444 dosage regimens were administered for 41 antibiotics [11]. Slow and not generalized implementation of evidence-based data in the various neonatal centers is a crucial reason of the high variability in dosing.

This heterogeneity in clinical practice is also depicted in dosage recommendations by reference books commonly used by neonatologists, such as Neofax, Red Book, Blue Book, and British National Formulary for children (BNFC). For example, BNFC 2015–2016 recommends that amikacin be administered in neonates in a multiple daily regimen (loading dose 10 mg/kg, then 7.5 mg/kg every 12 h) or in extended interval dose regimen (15 mg/kg every 24 h) [12]; in contrast, Neofax recommends that daily dosage and intervals to be assigned according to postmenstrual and postnatal age, whereas minimum dosing interval is 24 h [13]. Yet, according to the multicenter European point prevalence study of neonatal Exposure to Excipients, a multiple daily dosing regimen of aminoglycosides was only used in 14% of cases [10].

Due to lack of evidence-based data in neonates, many antimicrobial drugs are given with doses derived from adults or older children. This is suboptimal as neonates have functional and anatomical differences compared to older patients, and consequently distinct PK and toxicity associated with the antimicrobial agents. Infants are not just small children or even more small adults. Term and especially preterm neonates are a unique population; one could say that a neonate is a rapidly evolving biological system [14,15]. Growth, maturation, and specific pathophysiological characteristics (e.g. body composition, protein binding capacity, renal function and structure) influence PK behavior of antibiotics (absorption, distribution, metabolism, elimination) and result in significant interindividual variability in drug exposure. Moreover, disease characteristics (sepsis, hemodynamic changes, renal impairment, perinatal asphyxia, intrauterine growth restriction), comedications, other treatment modalities like extracorporeal membrane oxygenation, therapeutic hypothermia, and other covariates including pharmacogenetics have a great impact on PK and pharmacodynamic (PD) behaviors of many drugs in neonates, further aggravating the variability within the specific population [14,15]. Such scientific concerns along with regulatory and ethical issues make clinical trials in neonates difficult to perform and study design challenging.

During recent years, quantitative PK approaches for the optimization of drug dosing regimens in neonates have been developed and gained wide acceptance. PK modeling using compartmental analysis can predict drug concentration at any time, offers a more flexible timetable for PK samples collection, allows less blood sampling while quantifies the effect of several factors (covariates) on PK parameters explaining intra- and intersubject PK variability [16]. Pharmacometric modeling and simulation, tailored to term/preterm neonates, are valuable tools for understanding PK behavior of antibiotics, providing the basis for dose optimization and treatment individualization in this vulnerable population [16]. Population PK-PD modeling and simulation using nonlinear mixed effect modeling is a well-established approach of great value and similar studies encouraged by regulatory agencies are constantly increasing [17]. Studies using newer methodologies adapted to the specific demands of neonatal population, such as ultra-low-volume drug concentration assays, opportunistic protocols, dried matrix spot samples, physiologically based PK modeling, microdosing are emerging with promising results, although some of these approaches/techniques need further validation [18].

The US federal legislation and the European pediatric regulation significantly contributed to the increase of pediatric studies resulting in a rise in drug-labeling changes. However, few of these refer to neonates, which remain one of the last therapeutic orphans [19]. Design and reporting of RCTs in neonates seem to be of poor quality without any substantial improvement during the past years, as identified by Kaguelidou et al. in a systematic review of RCTs of antibiotics for neonatal infections [20]. A survey in the US FDA databases identified that among the 43 drugs studied in neonates between 1998 and 2014, only 20 were approved and just 2 of them were antibiotics (meropenem and linezolid) resulting in labeling with dose recommendations for preterm neonates [21].

As optimal dosing in neonates, infants, and children is a matter of public health and a prerequisite for high-quality care, there is a growing body of studies on pharmacometric modeling and simulation trying to evaluate and optimize dosing strategies of antibiotics prescribed in term/preterm neonates. A sixfold increase in the number of neonatal drug trials, registered in the clinicaltrials.gov database was noted from 1999 to 2012, whereas systemic anti-infectives were among the 3 most commonly studied drug groups [22]. An excellent systematic review by Wilbaux et al. on population PK approaches for the antibiotics primarily eliminated through kidneys in neonates identified 69 models for 13 antibiotics with vancomycin and gentamicin being the most studied, although there is no consensus on the optimal dosing for preterm or term neonates [23]. Unwillingness on behalf of the pharmaceutical industry to conduct clinical trials in neonates [22] has shifted this responsibility to the academic society and publicly funded bodies such as the European Commission, the UK National Institute for Health Research and the US National Institute of Child Health and Human Development to sponsor pediatric and neonatal clinical trials. Global Research in Pediatrics (GRiP), another initiative that focuses on pediatric clinical pharmacology training, stimulates and facilitates drug development and safe use of medicines in children, as well [24].

In recent years, several studies on antimicrobial dosing in preterm/term neonates have been or are conducted. As part of the TINN (Treat Infection in NeoNates) project funded by the European Commission, a prospective, open-label population PK study of ciprofloxacin for neonates and infants <3 months old was conducted. Doses of 7.5 mg/kg and 12.5 mg/kg given twice daily were suggested for neonates with PMA <34 and ≥34-week gestation, respectively. As, however, external validation was not performed, this regimen needs to be evaluated in clinical practice to confirm its benefits [25]. Ampicillin, despite being one of the most commonly used antibiotics in the NICU setting, lacks FDA labeling on dosing for preterm/term neonates, and it has been evaluated only recently by the Pediatric Trials Network (PTN) in the Pharmacokinetics of Understudied Drugs Administered to Children Per Standard of Care (POPS) trial. The latter aimed to identify appropriate dosing of several understudied drugs in children that are given as part of the standard medical care. Investigators proposed ampicillin doses of 50–75 mg/kg every 8–12 h according to gestational and postnatal age, and were able to achieve therapeutic targets (trough serum concentrations ≥8 mg/L) in over 90% of the studied subjects [26].

Another study by PTN determined meropenem PK in preterm and term infants ≤90 days of age with suspected or complicated intra-abdominal infections. In this investigation, in which dosing strategy was modified according to gestational and postnatal age, meropenem was proved to be sufficient for clinical and microbiological cure [27]. FDA recently changed the label for meropenem to reflect these findings (available at www.pediatrictrials.org). The results of a European multicenter RCT (NEOMERO project) aiming to evaluate PK, safety, and efficacy of meropenem in neonatal late-onset sepsis and meningitis in infants aged <3 months are under publication [28].

The Antibiotic Safety in Infants With Complicated Intra-Abdominal Infections (SCAMP) trial (clinicaltrials.gov: NCT01994993) is an ongoing partially randomized, multicenter, open-label safety study that aims to determine the safety of clindamycin, ampicillin, metronidazole, and piperacillin-tazobactam in infants with the above type of infections. Secondary objectives of the study include efficacy, PK, identification of biomarkers related to disease severity and antibiotic exposure, and diversity or shift of intestinal microbiota. An antistaph-trio study (clinicaltrials.gov: NCT01728363) aims to determine pharmacokinetics and safety of three antistaphylococcal antibiotics (rifampin, clindamycin, and ticarcillin-clavulanate) in term and premature infants. In order to better use teicoplanin in preterm neonates, we conduct a study on its PK involving neonates of different gestational and postnatal age. In addition, NEOVANC RCT (clinicaltrials.gov: NCT02790996) will compare two vancomycin-dosing regimens, a short one with loading dose and a standard 10 day regimen in neonates with sepsis due to coagulase-negative staphylococci. Among Pediatric Investigation Plans submitted to European Medicines Agency, only very few are on neonatal infections.

Neonatologists face the everyday problem of using many off-label antimicrobial agents and receive a high volume of information regarding newer data of PK and safety studies. Moreover, pharmacologists or infectious diseases experts are lacking from many NICUs. These make decision-making difficult. To overcome these issues, reference books used in everyday clinical practice should be revised and updated according to the existing data (Table 1). It is imperative that evidence-based guidelines are structured and disseminated in the neonatal centers via scientific societies or regulatory initiatives, so that to provide the optimal antimicrobial dosing and health care to the sick infants.

Table 1. Algorithm for improving the current situation about dosing of antibiotics in the neonate.

PK and toxicity studies of available antibiotics should be funded by international or national public organizations to be performed in neonates

PK and toxicity studies of new antibiotics should be funded by industry to be performed in neonates

Evidence-based guidelines are structured and disseminated in the neonatal centers via scientific societies or regulatory initiatives

Neonatal reference books used in everyday clinical practice should be revised and updated according to the existing data

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.