Introduction Vancomycin protocols for newborns are not uniform. There is limited pharmacokinetic data and drug clearance varies widely in this population.

We reviewed our vancomycin policy after audit data suggested we were not achieving the new target range (trough of 10–15 mg/l). Only 50% achieved therapeutic levels, despite using higher doses in infants <29/40 than the BNF recommends. In addition to gestational age and weight, chronological age (CA) had a marked effect with subtherapeutic levels seen in 72% infants over 10 days of age. Our protocol was amended to include CA.

Objectives Does accounting for CA in a Vancomycin dosing schedule reduce subtherapeutic levels?

Methods Patients commenced on vancomycin were identified (July–December 2010). Demographics, vancomycin schedule, plasma creatinine and trough levels (pre 3rd dose) were compared with the previous protocol.

Results 56 prescriptions (53 patients) were analysed. Mean trough level increased from 11.6 mg/l to 12.9 mg/l (95% CI 11.2 to 14.6). Subtherapeutic levels dropped from 50% to 36%. The frequency of levels >20 mg/l did not change (12% vs 11%).

In infants >10 days old, mean levels rose significantly from 8.3 mg/l (95% CI 6.6–10 mg/l) to 14.3 mg/l (95% CI 11.9 to 16.8 mg/l).

Infants <36 weeks had fewer subtherapeutic levels (25% vs 57%, p=0.056). There was no difference in high levels in those with normal renal function.

Conclusion By including CA in the protocol more infants can achieve therapeutic levels.

The frequent occurrence of subtherapeutic levels highlights the need to review current national Vancomycin dosing schedules.