Teicoplanin 24-h loading dose regimen using a decision tree model to target serum trough concentration of 15-30 μg/mL: A retrospective study.

Introduction: Teicoplanin (TEIC) is typically administered as a loading dose over 36-48 h. Achieving an effective concentration quickly is expected to treat severe infections, such as sepsis and methicillin-resistant Staphylococcus aureus infections. We aimed to identify the TEIC loading dose to be completed within 24 h, targeting the concentration of 15-30 μg/mL and factors affecting the loading dose by utilizing the decision tree (DT) model.

Methods: Patients treated with TEIC between January 2017 and December 2022 who met the 24-h loading dose regimen were enrolled. A LD_22.5 (corrected TEIC loading dose targeting concentration of 22.5 μg/mL) was determined, factors affecting the concentration were extracted, and a DT model was constructed. The validity of the DT was assessed using the coefficient of determination (R²) in the DT and population pharmacokinetics (PopPK) models for LD_22.5.

Result: A total of 149 patients were divided into training (n = 104, 70 %) and test groups (n = 45, 30 %). We indicated an average of 14.5 mg/kg for LD_22.5 and extracted four factors (estimated glomerular filtration rate, age, albumin, and C-reactive protein) from the DT model. The R² values were 0.724, 0.695, 0.681, and 0.653 for the DT models (training and test groups) and two PopPK models, respectively.

Conclusion: We established a 24-h loading dose regimen targeting the TEIC concentration of 15-30 μg/mL and identified four factors affecting the loading dose by using DT. By following the indicated DT algorithm flowchat, optimal decisions regarding the loading dose could be made for TEIC therapy.