Formulation, development, and assessment of a paediatric azithromycin nanosuspension

Abstract

Palatability of paediatric medications is a core determinant of adherence and therapeutic success. Azithromycin (AZI), a macrolide antibiotic widely used to treat upper and lower respiratory tract infections in children, poses formulation challenges due to its bitterness, poor aqueous solubility, and instability under acidic conditions. This study focused on designing and evaluating a stable, palatable nanosuspension of AZI suitable for paediatric oral administration. A validated reverse phase high performance liquid chromatography method was developed for quantification of AZI in formulation matrices. Critical method parameters including acetonitrile content, phosphate buffer pH, phosphate buffer molarity, and column temperature were optimised through a Central Composite Design, enabling evaluation of linear, interaction, and quadratic effects on retention time, resolution, and peak symmetry. The optimised method achieved excellent linearity (0.5–150 μg/mL; R² > 0.999), low limits of detection (0.9 μg/mL) and quantification (3 μg/mL), and inter/intra-day precision below 2% relative standard deviation. This ensured the method's suitability for analytical validation in line with ICH Q2(R1) guidelines and assisted in assessing the formulation. Pre-formulation studies provided an insight into the physicochemical properties of AZI. Fourier transform infrared spectroscopy and differential scanning calorimetry were employed to assess the compatibility of AZI with excipients, identifying Tween®80 and PVP K30 as ideal and compatible stabilisers. Tween®80 served as a non-ionic surfactant to reduce interfacial tension and enhance dispersion, while PVP K30 provided steric stabilization to maintain nanoparticle integrity. Ethanol was selected as the solvent based on its high solubilizing capacity for AZI (38.44 mg/mL at 37°C). This selection was based on quantitative solvent screening studies, which ruled out acetone, aqueous buffer, and deionized water due to insufficient solubilization. The nanosuspension was formulated via a liquid antisolvent precipitation method, employing bottom-up nanocrystallisation with ultrasonication. This method offers advantages in reducing particle size through controlled nucleation while minimising energy consumption compared to top-down techniques like high pressure homogenisation. A Box Behnken design was employed to statistically model the influence of formulation and process variables, which include, PVP K30 concentration, Tween®80 concentration, solvent: antisolvent ratio, and sonication time on critical quality attributes: particle size, polydispersity index, zeta potential, and initial drug release rate. Model adequacy was confirmed via analysis of variance tables, perturbation plots, and 3D surface analysis, which indicates significant quadratic and interaction terms in the formulation model. The optimised nanosuspension demonstrated a mean particle size of 134.3 ± 26.6 nm, a narrow polydispersity index of 0.298 ± 0.092, and a zeta potential of –28.3 ± 5.6 mV, indicating sufficient repulsion for colloidal stability. Transmission electron microscopy revealed discrete, spherical particles without agglomeration, while Fourier transform infrared spectroscopy and X-ray diffraction analyses confirmed the retention of chemical integrity of the API. In vitro dissolution testing revealed significantly faster release of azithromycin from the nanosuspension within the first 10 minutes compared to the commercial suspension, pinpointing the improved biopharmaceutical profile achieved through nanosizing. To address the challenge of palatability, taste masking efficiency was quantitatively assessed using the Alpha-MOS ASTREE electronic tongue. This advanced instrument mimics human gustatory perception using sensor arrays responsive to bitterness, sweetness, sourness, umami and saltiness. Nanosuspensions prepared at pH values ranging from 5 to 7.5 were compared to commercial AZI suspensions, pure drug, and placebo formulations. Principal component analysis, radar plots, and distance matrix analysis were used to interpret the sensor data. At pH 5, the optimised nanosuspension achieved the greatest sensory divergence from the pure drug and commercial formulation on bitterness sensor. The radar plots revealed a flattened response profile for the nanosuspension compared to the spiked sensor responses seen with the commercial AZI sample. This suggests that the physical entrapment of the azithromycin within a stabiliser matrix, along with pH buffering significantly diminishes the bitterness. This is further supported by the distance plots, where the optimised nanosuspension at pH 5 showed a greater than 95% discrimination. Stability studies (physical and chemical) were conducted at 25°C/60% RH and 40°C/75% RH over eight weeks. Particle size, zeta potential, and drug content were monitored to assess colloidal and chemical stability. The nanosuspension remained physically stable with minimal growth in particle size and no signs of sedimentation or aggregation. Drug content remained within 95–105% of initial values, indicating chemical stability. Zeta potential also remained negative, suggesting that the stabilisers retained their surface activity over time. This demonstrates the successful formulation and characterisation of an AZI nanosuspension with significant advantages in taste masking, dissolution enhancement, and predicted paediatric acceptability.