Vancomycin use in neonates and children: evidence-based practice is needed
- 1Academic Division of Child Health, University of Nottingham, Derbyshire Children's Hospital, Derby, UK
- 2Derbyshire Children's Hospital, Derby, UK
- Correspondence to Dr Helen M Sammons, Academic Division of Child Health, Clinical, The Medical School, Derbyshire Children's Hospital, Uttoxeter Road, Derby, Derby DE22 3DT, UK;
- Received 13 December 2012
- Revised 15 March 2013
- Accepted 18 March 2013
- Published Online First 16 April 2013
Vancomycin is an antibiotic that has been used in clinical practice for many years, but we have a situation where our current dosage regimen has been found to be insufficient to produce the recommended therapeutic levels. Timely studies, such as the one by Zhao et al,1 are working towards providing a firm evidence base for future recommendations. Vancomycin is the treatment of choice for methicillin resistant Staphylococcus aureus (MRSA). In Europe, it is used for the treatment of late-onset sepsis in neonates and in children, ventricular shunt or central venous catheter-associated infections. However, despite nearly 50 years of use there are still challenges around its dosage in both the neonatal and childhood populations.
The paper by Zhao et al1 represents a logical investigation into the issues surrounding vancomycin dosage in neonates. Recent work has shown large variability in practice across units in dose, frequency and timing of therapeutic drug monitoring levels (TDM).2 In the neonatal population, vancomycin is used in an intensive care setting, and so the use of a continuous infusion is an appropriate choice for maintaining steady therapeutic levels. The paper takes a comprehensive approach to the development of a dosing regimen. They start with the collection of TDM levels from three units in France, each using a different dosing regime. This showed that less than half the neonates had serum vancomycin concentrations within the recommended therapeutic range (15–25 mg/l). By using population pharmacokinetics analysis, they were able to use sparse sampling from a large cohort of babies, to create an optimised regime and recommendations for TDM. Their final step was to clinically validate their dosing regimen prospectively. This increased the numbers with a therapeutic level to just under three quarters, and all after some dose adjustments.
Their recommendations are for a patient-tailored optimised dosing regimen with continuing TDM at 6–12 h due to the large inter-individual variability. The variables found to be of significance were current weight, birth weight, postnatal age and serum creatinine. Using these identified variables, however, leads to complex calculations for both loading and maintenance dose. This cannot be done by hand, and an excel calculator was used in the study. Future work on the vancomycin dosing regimen must look at the prescribing risks around using such a tool and development of robust safety netting around standard ranges for dosing. A 10-fold variation could potentially exist in the weight of a term baby (4 kg) and the most premature infants (0.4 kg). Previous work has suggested that the use of a consistent regimen of a loading dose of 7 mg/kg, followed by a constant continuous dose of 30 mg/kg/day irrespective of age, gestational age or creatinine level, lead to less prescription errors.3 All these risks should be examined in the proposed future randomised controlled trial of continuous versus intermittent administrations suggested by the paper, along with providing further evidence of efficacy and safety.
The current situation in children also represents a challenge. A recent systematic review in adults has not shown a difference in mortality when comparing continuous to intermittent infusions, but there was a significantly lower risk of nephrotoxicity with continuous infusions.4 Continuous infusions in children in the intensive care setting, therefore, should be considered, but the children most commonly treated are those with neurosurgical shunts or central venous catheter-associated infections. The practicality of treating with a continuous infusion in these children becomes an issue, and intermittent dosing is most often used.
Vancomycin's effectiveness of killing bacteria depends upon the amount of time that its concentration remains higher than the minimum inhibitory concentration of the bacteria (MIC). The breakpoint MIC of vancomycin for susceptible pathogens is reported as <2–4 mg/l depending on the organism. There have been reports in the literature of increasing clinical failure rates of vancomycin, and this has led to a change in practice for adults. The recommendation to increase the target therapeutic levels for vancomycin in children and neonates followed. The predose trough level was increased to 10–15 mg/l (15–20 mg/l in MRSA), but there was no accompanying change in the dosage regimen of 15 mg/kg every 8 h. This has led to confusion among clinicians and many children never attaining a sufficient level of vancomycin during their antibiotic treatment.
Recent US guidance recommended 60 mg/kg/day (15 mg/kg 6 hourly) in children with severe MRSA infections. Following this, Frymoyer et al5 assessed this dose change from 45 to 60 mg/kg/day. They showed a significant increase in initial trough levels with the mean±SD initial vancomycin trough level increasing from 7±5 μg/ml to 9±5 μg/ml, but still the higher dose did not consistently produce therapeutic levels of 15 mg–20 mg/l. The study did not examine clinical outcome, and presently, there is no evidence on improved clinical outcome with an increase in dose or the change in therapeutic levels.
Vancomycin may have been in clinical use for many years, but there still remain many questions around its safe and effective use in neonates and children. Further work is needed to confirm the optimal dose, frequency of dosing and TDM schedule. This should be done in studies that can assess both the safety and efficacy of vancomycin for future practice.
Competing interests None.
Provenance and peer review Commissioned; externally peer reviewed.