Sterility testing of aqueous and oily eye drops using the Bactec^® automated system: feasibility and validation

INTRODUCTION
Our Hospital Pharmacy (HoP) produces compounded eye drops, based on aqueous or oily excipients. These eye drops are used immediately after manufacture on the basis of a parametric release, the sterility test prescribed by the European Pharmacopoeia (EP) requiring 14 days of incubation. The aim of this study was to validate a rapid sterility control technique for these eye drops, using Bactec^® technology based on the detection of CO₂ production method (DCM) with colorimetric changeover in the event of growth.

MATERIAL AND METHODS
Two series (n=6) of eye drops (9.9 mL/vial) were prepared under aseptic conditions, without active ingredients: aqueous eye drops (AED) based on 0.9% NaCl and oily eye drops (OED) based on castor oil. 6 ATCC reference microorganisms (Bioball^®, 550CFU/Unit): S. aureus (G1), P. aeruginosa (G2), C. sporogenes (G3), B. subtilis (G4), A. brasiliensis (G5) and C. albicans (G6)), recommended by the EP, reconstituted in NaCl 0.9% solution, were used for validation studies. Preparations were enriched with 100µL of each germ (final concentration 1CFU/mL). Aliquots of 3mL eye drops (3CFU) were introduced into 3 different DCM media: BACTEC™ Peds (B1, aerobic medium), BACTEC™ Plus Anaerobic (B2) and BACTEC Plastic Mycosis IC/F (B3, filamentous fungi), then cultured in a BD Bactec™ Fx40 incubator. Identical aliquots were cultured in the media prescribed by the EP: Tryptone Soy (TSB) and Thioglycolate (TB) broths. For OED, 3mL aliquots were filtered on a 0.22µm membrane (FM) and then transferred to Count-tact^® agar. 2 parameters were evaluated: detection or non-detection of the germ (minimum 5-day reading) and time to a positive result (Time to detection or TTD).

RESULTS AND DISCUSSION
Media B1 and B3 detected all germs except G3 (anaerobic). G3 was difficult to detect on TB and FM, but was detected by medium B2. With DCM, G5 was detected with both B1 and B3. Thus, B1 medium was sufficient to detect aerobic germs and filamentous fungi with both AED and OED, and B2 for anaerobic germs. Detection of all germs was faster with DCM, in all cases <24h whatever the matrix, with TTD ranging from 7h02’±18’ for G5 in aqueous media to 18h21’±29’ for G3 in oily media. With PE reference techniques, detection required between 3 and 5 days of incubation to obtain a positive result.

CONCLUSION
The DCM technique enabled rapid detection of reference microorganisms present in both aqueous and oily eye drops, with a low detection limit (1 CFU/mL) and a low total number of incubated CFUs (3 CFUs). The technique outperforms reference techniques: rapid <24h, repeatable, sensitive and covers all the germs used, enabling compounded eye drop to be released within 24 hours.