1.Evaluation of Predictive Accuracy between Two Types of Vancomycin TDM Analysis Software
Shungo Imai ; Takehiro Yamada ; Ayako Nishimura ; Hiromitsu Oki ; Masayuki Kumai ; Takenori Miyamoto ; Kumiko Kasashi ; Ken Iseki
Japanese Journal of Drug Informatics 2015;16(4):169-178
Objective: To attain optimal blood concentration rapidly, it is needed to perform initial dose setting appropriately when vancomycin (VCM) used. In order to design initial dose settings of VCM more currently, we compared the predictive performance of two types of VCM therapeutic drug monitoring (TDM) analysis software retrospectively.
Method: We utilized two TDM analysis software, SHIONOGI-VCM-TDM ver.2009 (VCM-TDM) and “Vancomycin MEEK TDM analysis software Ver. 2.0” (MEEK), based on patient’s background. 112 patients who received VCM and performed TDM were analyzed during the period from October 2011 through September 2012 and compared the actual trough level with the predictive trough level. The predictive performance was evaluated by calculating ME (mean prediction error), MAE (mean absolute prediction error), and RMSE (root mean squared error). Age, gender, and a renal function were evaluated as patient’s background.
Results: VCM-TDM gave good predictive performance for patients overall. When classified patient’s background complexly (sex, age, and renal function), as for male patients, VCM-TDM showed good predictive performance except for the group over 65 years old and CCr over 85 mL/min. For female patients, the difference of predictive performance was not accepted by all groups.
Conclusion: These results suggest, for male patients, we should use VCM-TDM for initial dose settings except for the group over 65 years old and over CCr 85 mL/min. For the other patients, we consider that both of software can be used. These new findings seem to contribute to proper dosage settings of VCM.