RESUMO
Background and Objectives: Teicoplanin (TEIC) is an effective drug for patients with febrile neutropenia (FN); however, it has been reported that these patients may have increased TEIC clearance compared with patients who do not have FN. The purpose of this study was to study therapeutic drug monitoring in patients with FN when the TEIC dosing design was based on the population mean method. Materials and Methods: Thirty-nine FN patients with hematological malignancy were included in the study. To calculate the predicted blood concentration of TEIC, we used the two population pharmacokinetic (population PK) parameters (parameters 1 and 2) reported by Nakayama et al. and parameter 3, which is a modification of the population PK of Nakayama et al. We calculated the mean prediction error (ME), an indicator of prediction bias, and the mean absolute prediction error (MAE), an indicator of accuracy. Furthermore, the percentage of predicted TEIC blood concentration within 25% and 50% of the measured TEIC blood concentration was calculated. Results: The ME values were -0.54, -0.25, and -0.30 and the MAE values were 2.29, 2.19, and 2.22 for parameters 1, 2, and 3, respectively. For all of the three parameters, the ME values were calculated as minus values, and the predicted concentrations tended to be biased toward smaller values relative to the measured concentrations. Patients with serum creatinine (Scr) < 0.6 mg/dL and neutrophil counts < 100/µL had greater ME and MAE values and a smaller percentage of predicted TEIC blood concentration within 25% of measured TEIC blood concentrations compared with other patients. Conclusions: In patients with FN, the accuracy of predicting TEIC blood concentrations was good, with no significant differences between each parameter. However, patients with a Scr < 0.6 mg/dL and a neutrophil count < 100/µL showed slightly inferior prediction accuracy.
