We thank Tse, Wong, and Sridhar for their question regarding the use of vancomycin serum levels that were obtained at the time of delivery in an attempt to assess a “therapeutic” drug level. The current guidelines on vancomycin monitoring do recommend the use of area under the concentration-time curve/minimum inhibitory concentration (AUC/MIC). However, the AUC/MIC value is recommended for use in the evaluation of vancomycin “treatment” effectiveness rather than prophylaxis. In using the AUC/MIC, one starts with a mathematic calculation that involves the determination of an estimated creatinine clearance, an estimate of the vancomycin clearance with conversion to liters per hour, the chosen vancomycin dosing regimen, followed by an estimate of the AUC/MIC. Once treatment is initiated, >1 serum level is then obtained to verify (through further mathematic computation) that the chosen dosing regimen is adequate. This AUC/MIC verification is not feasible in the evaluation of fetal levels at the time of birth because only 1 cord blood value can be obtained at delivery. Likewise, most pregnant women receive only 1 dose before delivery (29 of 55 patients in our study), thereby negating the possibility of obtaining multiple serum levels.

The MIC of the infecting organism is also required for the AUC/MIC to be accurate, and the MIC for group B streptococcus usually is not obtained or is unavailable, unless one were to assume that all isolates have an MIC of ≤1 μg/mL. However, 2 cases of vancomycin-resistant group B streptococcus were reported recently with MICs of 4 μg/mL. Additionally, the AUC/MIC evaluation also uses the best estimate of vancomycin volume of distribution, and the volume of distribution in pregnancy has never been addressed. This value would most likely be elevated in pregnancy because the antimicrobial was shown to cross the placenta and enter the fetal circulation, thereby adding to the volume of distribution.

The guidelines discuss several very important issues. To begin, an AUC/MIC of 400 μg/mL is the target for Staphylococcus aureus infections. The target AUC/MIC for group B streptococcus is unknown. Second, serum vancomycin trough levels are considered the most accurate and practical way to evaluate efficacy and are the recommended therapeutic drug monitoring parameter to use. Third, the guidelines state that “minimum serum vancomycin trough concentrations should always be maintained above 10 mg/L to avoid development of resistance” (milligrams per liter is equivalent to micrograms per milliliter). Last, to achieve the optimal trough levels, the recommended dosing regimen is 15-20 mg/kg (using actual body weight) administered every 8-12 hours in patients with normal renal function for a bacteria with an MIC of ≤1 μg/mL. In reference to this last statement, renal function (creatinine clearance) usually increases above normal in most pregnancies, which may explain the reason that our study found that 20 mg/kg intravenously every 8 hours was needed to obtain a cord blood level of >10 μg/mL in most patients. There actually has been some discussion of using higher vancomycin trough levels of 15-20 μg/mL in cases of serious infections (ie, sepsis, endocarditis, meningitis).

In conclusion, if vancomycin “treatment” is needed in a pregnant patient with a serious bacterial infection, then our study data would suggest the use of 20 mg/kg intravenously every 8 hours (maximum individual dose of 2 g) to obtain the greatest number of serum values of >10 μg/mL. At the present time, the AUC/MIC value is not likely to be obtained when treating pregnant patients. Regarding group B streptococcus “prophylaxis,” the standard regimen of 1 g intravenously every 12 hours may be adequate; however, one must wonder whether this protocol potentially could lead to the advancement of resistance in group B streptococcus and other significant bacteria because of a high frequency of serum levels below the recommended 10 μg/mL. We concur that further research on these topics is needed.