The increased use of vancomycin in neonatal and paediatric patients has prompted numerous pharmacokinetic studies and the development of many empirical administration methods. The majority of dosage guidelines use the relationship between pharmacokinetic parameters and patient variables such as chronological age, bodyweight, and/or measures of renal function. Currently, those dosage guidelines which are based upon postconceptional age and bodyweight seem to provide the best options for empirical administration in neonates and infants. In addition, serum creatinine may prove to be a useful guide to the empirical administration of vancomycin in neonates older than 7 to 14 days. Several investigators have reported the individualisation of dosage regimens based on pharmacokinetic-based administration methods. The most common techniques employed have been Sawchuk-Zaske method and Bayesian forecasting. However, only a limited number of studies have evaluated either empirical administration or individualised administration techniques in patient populations outside those of the original reports; this makes choosing between the methods difficult. Pharmacokinetic data and administration recommendations have gradually become available in special paediatric patient populations. The majority of studies have focused on patients requiring cardiopulmonary bypass surgery or with burns, cancer or central nervous system infections. However, a limited amount of information is available regarding vancomycin disposition in children older than 1 year of age with and without end-stage renal failure. The monitoring of serum vancomycin concentrations may be useful in selected neonatal and paediatric patient populations, especially where large interpatient variability occurs and administration guidelines are not clearly established. Similar to the literature on adults, the lack of conclusive evidence concerning the relationship between serum vancomycin concentrations and therapeutic responses leaves this topic open to debate.