Physicochemical Validation of the Sterile Filtration Process for Hospital-Compounded Gentamicin Eye Drops

Background
During routine analytical quality control of hospital-compounded gentamicin eye drops (15 mg/mL), out-of-specification results were detected in the first sampled unit. However, analysis of a second unit from the same batch showed compliance with specifications. Given the recurrent nature of this discrepancy, a physicochemical validation of the sterile filtration process was deemed necessary.

Objective
This study aimed to investigate the cause of these repeated analytical deviations and to assess the potential impact of the sterile filtration process on gentamicin concentration. A physicochemical validation of the filtration process was conducted to ensure the robustness and reproducibility of the preparation method.

Materials and Methods
A batch of gentamicin eye drops was compounded under a microbiological safety cabinet (SafeFast Elite, Class II). The stock solution used was gentamicin (160 mg/2 mL, Panpharma),^® freshly diluted with 0.9% (w/v) sodium chloride. The resulting solution was sterilized by filtration using various 0.2 µm filters (Minisart^®): cellulose acetate, polyethersulfone, and modified cellulose. Sequential 10 mL fractions were collected for each filter type, numbered, and stored for analysis. The active ingredient concentration was determined using UV-Raman spectrometry (QCrX^®, Icone Service).

Results
UV-Raman analysis showed that concentrations were within expected specifications for the modified cellulose and polyethersulfone filters, with mean values of 14.90 mg/mL and 14.61 mg/mL, respectively. In contrast, the cellulose acetate filter led to a marked decrease in concentration in the first 10 mL fraction, measured at 9.94 mg/mL. Subsequent fractions from this filter showed concentrations within specifications.

Discussion–Conclusion
Physicochemical validation demonstrated that polyethersulfone and modified cellulose filters ensure consistent and compliant gentamicin concentrations. However, cellulose acetate caused a significant initial drop in concentration, likely due to adsorption phenomena during the first filtration phase.

See the poster