Neonatal epilepsy: Development and validation of an HPLC-UV method for carbamazepine dosage

Objectives
In January 2026, the ANSM issued an alert regarding the currently marketed oral form of carbamazepine. The alert concerns a restriction on its use in term newborns younger than 4 weeks or preterm babies younger than 44 weeks, due to the concentration of a known active ingredient (propylene glycol) exceeding the recommended threshold. In this context, our laboratory is working on a safer oral liquid formulation of carbamazepine for our neonatal patients. The objective of this work is to develop and validate an analytical method using high-performance liquid chromatography coupled with an ultraviolet detector (HPLC-UV) to test this preparation.

Methods
The HPLC-UV system is configured with a Symetry Shield C18 5 µm reversed-phase column maintained at 25°C. The mobile phase is an isocratic mixture of H₂O and methanol (50/50 v/v), with a flow rate of 1.2 mL/min, an injection volume of 10 µL, and a run time of 8 min. A forced degradation study was conducted to demonstrate the method’s suitability as a stability indicator under this conditions: basic (0.66M NaOH, 17h, 85°C), acidic (0.66M HCl, 17h, 85°C), oxidative (1.5% H₂O₂, 21h, 85°C), and photolysis (UV, 17 h, 25°C). The carbamazepine impurity A described in the pharmacopoeia, as well as oxcarbamazepine, were analyzed under the same conditions. The specificity, linearity, accuracy, and trueness of the method were evaluated according to ICH(Q2) guidelines.

Résults
An 8-minute isocratic analysis separates the degradation products, impurity A (tr = 7.3 min), oxcarbazepine (tr = 1.4 min), and the excipients in the preparation without interfering with the carbamazepine peak (tr = 6.4 min). Significant degradation is observed, primarily under acidic conditions. The method is linear between 10 and 100 µg/mL with an r² > 0.996. The mean recovery ranges from 98.2% to 99.3%, with no significant bias detected by statistical tests (Cochran and ANOVA). The coefficients of variation for repeatability and intermediate precision are 0.83% and 0.1%, respectively. The accuracy profile shows that the tolerance intervals are within the acceptance limits at the three concentrations tested: 25, 50, and 75 µg/mL.

Discussion - conclusion
The HPLC-UV method developed has been validated in accordance with ICH (Q2) guidelines, with appropriate values for linearity, accuracy, and precision. It is appropriate for the determination of future carbamazepine formulations and helps ensure the quality, safety, and compliance of pediatric formulations administered in clinical practice.

See the poster