Monitoring Serum Vancomycin Concentrations in Children: Is It Necessary?

This is the authors' abstract. We don't add key points for this paper.

The necessity for routine monitoring of serum vancomycin concentrations (SVCs) has recently come under intense review at many institutions, including our own, not just because of cost issues related to this frequently used antibiotic but because original concerns for nephrotoxicity have been largely alleviated. Many experts currently do not recommend measurement of SVCs in patients with normal renal function who are treated with the usual dosages of vancomycin,1-3 whereas others continue to firmly support routine monitoring.4 Early studies documented dosage and serum concentrations adequate for management of various infections in children, and adjustments were not required to assure cure in subsequent efficacy trials.5 The recommendation to monitor this antibiotic was introduced after anecdotal reports of nephrotoxicity usually in elderly adults who were also being treated with other potentially nephrotoxic agents such as aminoglycosides. It has never been established what concentration of vancomycin may actually be nephrotoxic under any circumstances. The purpose of this study was to review our experience at a large Children's Hospital monitoring SVCs, in an effort to determine how often concentrations might have initially been in a potentially subtherapeutic or toxic range. Materials and methods. The medical records at Children's Hospital in New Orleans were reviewed for all patients older than 60 days of age admitted to the general or subspecialty services who received intravenous vancomycin between July, 1994, and December, 1996. Because earlier reports had suggested that pharmacokinetic determinations for children with cancer may be different from those of other patients,6 this group was evaluated separately. We excluded neonates and infants younger than 60 days of age because serum measurements in this population are problematic and because the subject has been examined in detail previously.7, 8 The diagnosis, dosage of drug, initial serum creatinine, initial measurement of SVCs and any dosage adjustment as a response to this determination were recorded. The concomitant administration of other potentially nephrotoxic drugs was also noted. Patients with an elevated creatinine or other evidence of renal dysfunction were excluded. Also excluded were patients requiring longer infusion times because of adverse drug reactions ("red man" or "red neck" syndrome). Vancomycin was administered intravenously by syringe pump or Buretrol® infusion throughout 45 to 60 min in dosages of 7.5 to 10 mg/kg body weight. Serum samples were obtained by venipuncture, central venous catheter aspiration, heel stick or finger stick 45 to 60 min after flushing the infusion. SVCs were determined by fluorescence polarization immunoassay (AxSYM®; Abbott Laboratories, Diagnostic Division, Irving, TX). General guideline recommendations at our institution during this time period, similar to those previously published,5-7, 9 were that children with normal renal function receive 30 to 40 mg/kg/day of intravenous vancomycin divided every 6 h, that a serum creatinine be obtained within 24 h of the first dose, that vancomycin peak and trough serum concentrations be measured after the third dose (or the next daytime dose thereafter) and that dosage adjustments be made by the responsible physician if peak serum concentrations were not between 20 and 40 μg/ml. For purposes of data analysis peak SVCs between 15 and 40 μg/ml and trough values of 5 to 15 μg/ml were considered acceptable, based on our literature review.5, 7, 10 Results. During the study time period 167 infants and children without cancer and 42 cancer patients ages 3 months to 17.5 years were treated with vancomycin for various presumed or documented infections. Seventeen additional patients were excluded from our analysis, 12 with renal function impairment and 5 requiring long infusion times. Monitoring data for peak and trough SVCs are summarized in Table 1. In children without cancer peak SVCs were in an adequate therapeutic range for 93% of patients, with the highest determination being 55 μg/ml and the lowest 8 μg/ml. For children with cancer a higher percentage had peak SVCs below 10 μg/ml (10%), and trough values <5 μg/ml (21%). These patients required an increased dosage of 48 to 72 mg/kg/day to achieve peak concentrations of >20 μg/ml.TABLE 1In this series there were no treatment failures in the group with routine infections or in six patients with cancer and bacteremia, five caused by Staphylococcus epidermidis and one by Staphylococcus aureus. Discussion. When vancomycin was first approved by the Food and Drug Administration in 1958, nephrotoxicity and ototoxicity were included as potential adverse effects. This caution was based on a report of renal dysfunction in two patients in whom vancomycin serum concentrations had reached 80 to 100 μg/ml.11 It was also suggested that toxicity may have resulted from impurities in the original vancomycin product which was dark brown and was often referred to as "Mississippi Mud" because of its appearance. Subsequently, production of this antibiotic was revised so that preparations are now absolutely free of such impurities. Controlled investigations in animals, undertaken 20 years after the drug was released9, 12 failed to demonstrate either nephrotoxicity or ototoxicity attributable to vancomycin when administered without other nephrotoxic agents such as aminoglycosides. Moreover, post marketing reporting and surveys have rarely identified nephrotoxicity in patients with normal renal function who were not receiving concomitant nephrotoxic agents.1 Although >150 cases of nephrotoxicity and >50 cases of hearing loss associated with vancomycin use have been reported, the majority had additional confounding risk factors either from their underlying diseases or concomitant drug use. Of those without confounding variables where serum drug concentrations were measured, most had peak and trough concentrations within the recommended range. Still this is a very small number of reports considering the current extensive use of vancomycin. In spite of highly questionable causality, a recent review concluded that although literature is conflicting concerning synergistic nephrotoxicity for the combination of vancomycin and aminoglycosides, "periodic (at least weekly) evaluations of the serum creatinine level should be performed on the basis of a patient's risk for renal insufficiency secondary to disease states or cotreatment with nephrotoxic drugs."1 Original dosage recommendations for pediatric patients published in 1959 were guided by experience in adult patients and based on just 23 children, aged 1 month to 18 years who received dosage regimens between 40 and 180 mg/kg/day.13 However, additional data many years later supported the initial dosage recommendation of 40 mg/kg/day in infants and children with normal renal function, to achieve optimal peak SVCs; in that study SVCs >12 μg/ml or 8-fold greater than the MIC of the pathogen being treated generally correlated with clinical cure,7 although subsequent recommendations usually suggested a higher range of 20 to 40 μg/ml as ideal peak concentrations.11 The largest clinical trial including neonates and children used fixed doses of vancomycin from 30 to 60 mg/kg/day and found peak and trough concentrations ranging from 18 to 57 μg/ml and 3 to 19 μg/ml respectively.5 Clinical outcome for these staphylococcal infections was excellent. Unfortunately, there has never been any clear correlation between peak SVCs and clinical cure rates either in pediatric or adult efficacy trials. This lack of data makes it difficult to offer firm recommendations either for appropriate serum concentrations or dosage regimens. The minimal inhibitory concentration (MIC) of vancomycin for most susceptible organisms is <5 μg/ml and in almost all clinical trials, trough as well as peak levels have exceeded this value. There is no enhancement of bacterial killing with higher serum concentrations, suggesting that current standards for minimum serum concentrations may be high. More importantly the levels where toxicity might occur are at least 2-fold higher than those achieved with currently recommended dosages. With such a wide safety margin, it would appear prudent to dose this antibiotic in the higher range, i.e. 40 mg/kg/day and for children with cancer to begin with an even higher dosage of 60 mg/kg/day because they are more likely to have low peak concentrations. The present report confirms that concentrations of vancomycin are quite predictable in the majority of infants and young children with normal renal function. An exception appears to be children with cancer who for reasons incompletely understood exhibit increased metabolism and elimination of the drug, with resulting reduced serum concentrations.6 Our current data support this observation. On the basis of our experience and review of the literature, we instituted the following guidelines at our institution for monitoring SVCs in pediatric patients: Serum creatinine measurement is recommended within 24 h of beginning therapy to verify that renal function is normal and repeated at least weekly during therapy. An increase in serum creatinine of 0.5 mg/dl or more from baseline during therapy suggests a change in function and potential need to monitor SVCs. Peak and trough vancomycin serum concentration measurements are not generally necessary for children receiving dosages of 40 mg/kg/day or less who have normal renal function, are not receiving other potentially nephrotoxic medications and do not have cancer. Responsible physicians may, however, choose to individualize this approach based on other factors such as degree of illness, poor clinical response or persistent positive cultures. Monitoring vancomycin serum concentrations should be considered for patients with cancer, neonates, those receiving concurrent nephrotoxic drugs such as aminoglycosides, patients with abnormal renal function and children receiving dosages of vancomycin of >40 mg/kg/day. Mathew P. Thomas, Pharm.D. Russell W. Steele, M.D. Departments of Pharmacy (MPT) and Pediatrics (RWS); Children's Hospital Division of Infectious Diseases/Immunology; Department of Pediatrics; Louisiana State University School of Medicine (RWS)