Gastine, S (1); Obiero,C (2); Williams, P (2,3); Kane, Z (1); Murunga, S (2); Thitiri, J (2); Omollo, R (2); Egondi, T (2); Nyaoke, B (2); Correia, E; Ellis (4), S; Berkley, JA (2,3); Sharland, M (5); Standing, JF (1,5)
(1) UCL Great Ormond Street Institute of Child Health, London, United Kingdom, (2) KEMRI, Welcome Trust Research Programm, Kilifi, Kenya, (3) University of Oxford, Center for Tropical Medicine & Global Health, Oxford, United Kingdom, (4) GARDP – Global Antibiotic Research & Development Partnership, Genève, Switzerland, (5) St George's University, London, United Kingdom
Objectives:
Neonatal sepsis remains a major cause of morbidity and mortality in the preterm and term neonatal population. Early and appropriate antibiotic treatment in sepsis therapy is crucial, with the combination of ampicillin and gentamicin being the acknowledged Standard of Care (SOC) during the first days of life, as by the WHO pocket book of hospital care for children, as well as by most local guidelines.
We analysed the pharmacokinetics of SOC ampicillin and gentamicin, obtained through back-up samples from the NeoFosfo Trial (NCT03453177), where fosfomycin was administered in addition to SOC treatment.
Methods:
SOC ampicillin and gentamicin therapy was initiated on the day of admission by the treating physician. According to local guidelines, ampicillin was dosed with 50 mg/kg/dose q12h for patients < 7 days and q8h if ≥ 7 days old. Gentamicin was dosed at 3 mg/kg q24h for patients < 2kg and 5 mg/kg if ≥ 2kg.
NeoFosfo study back-up samples, available for ampicillin and gentamicin pharmacokinetic analysis accounted for 474 measured concentrations from 59 patients.
Population pharmacokinetic (PopPK) modelling to derive a joint ampicillin and gentamicin model was performed using nonlinear mixed effects modelling (NONMEM 7.4, FOCE+I algorithm). In the first step, the PopPK models for the single substances were estimated. Subsequently, both single models were linked by introducing covariance between the clearance estimates.
As this was a study in preterm and term neonates on two predominantly renally cleared drugs, allometric scaling and renal maturation was included as fixed structural covariates [1]. Further covariate exploration was performed on the combined model parameters. Explored covariates included demographics, serum creatinine, transaminases, bilirubin and albumin, as well as electrolyte levels.
Results:
After data cleaning, 57 neonates with 356 (175/181) ampicillin and gentamicin samples were eligible for analysis. The patients median (range) gestational age was 40 wks (34.4 – 44) and post-natal age (PNA) 1 day (0 – 23) with time-varying documentation of PNA during the study.
Ampicillin pharmacokinetics were well described by a one compartment model in our study population (CL= 31.61 L/h/70kg; V=74.27 L/70kg; IIV CL=40%). For gentamicin a two-compartment model was found most appropriate (CL=4.91 L/h/70kg; Vc=19.00 L/70kg; Vp=14.41 L/70kg; Q=1.68 L/h/70kg; IIV CL=36%).
The joint modelling showed the two clearances had a 67% covariance.
Detected significant covariate effects included the effect of the individual’s serum creatinine by a power function on the ratio of the observed creatinine to the age-appropriate expected creatinine. Additionally, an effect of PNA on both clearances was detected during first days of life. This was implemented by estimating the fraction of adult clearance observed on the very first day of life, that was subsequently allowed to increase exponentially until adult clearance is reached. We estimated an initial neonatal clearance fraction of 33% compared to the adult value, for the first day of life.
Conclusions:
Combined ampicillin and gentamicin pharmacokinetics in SOC treatment of neonatal sepsis is well described by a combination of one- and two-compartment models. In addition to classical allometry and renal maturation functions, maturation during the first days of life impacts both drugs’ clearances. Relating the individual’s creatinine to the age-appropriate value aided in explaining the change in clearance associated with creatinine levels in this special paediatric population.
References:
[1] Germovsek et al. JAC 2018 Plasma and CSF pharmacokinetics of meropenem in neonates and young infants: results from the NeoMero studies
Reference: PAGE () Abstr 9450 [www.page-meeting.org/?abstract=9450]
Poster: Drug/Disease Modelling - Paediatrics