In the present study, we first performed a Monte Carlo simulations in conjunction with pharmacodynamic profiles to compare different total daily dose of vancomycin among different newborn subgroups and treatment efficacy was then evaluated in two dosing regimen.We describe detailed pharmacological target attainment of thirteen total daily doses in six categories subgroups. The simulation results show that different Gram positive bacteria, gestational age, postnatal age, postmenstrual age, birth weight and serum creatinine may have implications for dose adjustment. Cases were then grouped according to whether they met criteria for the lowest recommended doses. The distribution of trough concentration and clinical efficacy between two groups show significance difference(P<0.05), yet regimens achieving target trough concentration between-group analyses showed no significant difference(P>0.05).
For many drugs, there are important PK differences among adults, children, infants and neonates[18].There are also many factors that influence vancomycin PK in neonates population[19, 20]. In our simulated data, we found that neonates with PMA ≥ 42w and birth weight > 4000g display a large variation of simulated curves in terms of dose-effect correspondence. Although the organs of these neonates are more mature than those of preterm infants, as body weight and gestational age increased, and so does various factors affecting the metabolism of vancomycin in vivo, resulting in increased variation in the individual response to different doses of vancomycin concentrations.It is suggested that this group of neonates are closer to the younger infants population and that the neonatal PK parameters we referred to are not necessarily applicable to this group or are not applicable to the neonatal population in our study center.
The recommended pharmacokinetic-pharmacodynamic target AUC/MIC ratio for vancomycin is ≥ 400 in current guidelines[15] yet there are no relevant guideline recommendations for targets of other gram-positive cocci. It is reported that exposure targets for vancomycin in the treatment of neonatal CoNS infection were far below 400[21].That means the clinically effective dosage may be lower than the recommended dosage for neonates especially with non-MRSA infection. Some studies have found that altered plasma albumin affect vancomycin dosing in neonates[22, 23].For neonates with normal albumin levels, a higher vancomycin protein-binding rate can lead to an insufficient total dosage. While for neonates with hypoproteinemia, normal dosages may result in unexpectedly high plasma concentrations, causing dose-dependent side effects such as acute kidney injury.Comparisons between actual daily dose and simulated Recommended dosage regimens indicates that these neonates were treated by using a lower dose yet a satisfactory clinical outcome was gained.This might be explained by a lower levels of albumin of our patients.
Actually, the relationship between vancomycin trough concentration and area under the concentration-time curve is not clear in Neonates[24, 25].The majority considered that high trough concentration of vancomycin was related to incidence of nephrotoxicity[26]. And AUC/MIC is the best predictor of vancomycin efficacy[27]. A comparisons for values of different dosage regimens of concentrations was conducted in this study. We found that Recommended dosage regimen group presented higher values (P<0.05)in trough concentration than in not-recommended groups, yet the target concentration ( 5-15mg/l ) compliance rate indicates no significant differences (P>0.05).Subsequent efficacy study showed that Recommended dosage neonate group presented higher clinical response rate (P<0.05༉than in not-recommended groups.Our results further demonstrate that AUC/MIC is not necessarily related to trough concentration, and clinicians should be cautious about the predictive efficacy of trough concentration for efficacy. Efforts should be made to obtain as much information as possible about AUC/MIC parameters as well as concentrations to prevent adverse effects such as AKI caused by vancomycin.
To date, as the diagnostic criteria for neonatal AKI are not uniform and the incidence of AKI in neonates included in studies varies widely, the incidence of neonatal AKI has been inconsistently reported in the literature, ranging from 18–40%[28].Due to the unique pathophysiological characteristics of neonates with AKI, they differ significantly from adults and the currently improving and maturing clinical criteria for the diagnosis of AKI in adults are not applicable to the diagnosis of AKI in neonates[29].The use of traditional criteria for predicting the occurrence of AKI does not truly reflect the occurrence of AKI in newborns.On the one hand, AKI caused by non-pharmacological factors such as prematurity combined with hypoxemia, sepsis, necrotizing small bowel colitis and arteriovenous catheterization may be included[30]. On the other hand, the poor consistency of all criteria for diagnosing AKI may lead to clinical underdiagnosis of some AKI in neonates[31].We did not evaluate the occurrence of AKI in neonates in this study. It is expected that the combined detection of multiple biomarkers to predict AKI of vancomycin in newborns will provide a realistic evaluation of its nephrotoxicity in our further study.
At present, there are many related studies on Continuous infusion versus intermittent administration of vancomycin[32–35].Previous studies[34, 35] have shown that continuous vancomycin infusion can maintain blood levels at stable levels, thereby reducing acute kidney injury due to high concentrations, yet the comparative efficacy of the two regimens remains controversial[32, 33].One study constructing population PK models for intermittent and continuous infusion of vancomycin in neonates and infants respectively showed that continuous dosing was more advantageous than intermittent dosing, yet the risk of nephrotoxicity is unknown for both regimens[34].In this study, our clinicians were given an interval dosing regimen considering the ease of administration and drug stability factors which Monte Carlo simulations of the two dosing regimens are needed to compare the efficacy and safety in the future.
This study had some limitations.Firstly, this study is a single-center data with a small sample size, and the recommended dose obtained by establishing Monte Carlo simulation has limited significance in guiding clinical use.Secondly, a multivariate analysis for evaluating confounding factors was not performed which may suggest a potentially overestimate of efficacy of vancomycin in this retrospective study. Finally, in the analysis of adjusted dosing regimens, we did not analyze specific reasons regarding the dose adjustments made by physicians and did not conclude which dosing regimen was preferred for clinical use.Perhaps which dose modifications were required depends on numerous factors including patient-related factors, drug-related and pathogen-related factors.