Study of the passage of hydrogen peroxide through eye-drops bottles intended for sterility testing during their decontamination in an isolator

Context
Sterility testing of eye-drops produced in our unit is conducted in a sterile isolator. Hydrogen peroxide (H2O2) is the oxidizing agent responsible for the decontamination process. The bottles used to package the eye-drops are all made of low density polyethylene (LDPE), but with different types of caps depending on the preparations. As the possible passage of the H2O2 through LDPE has not been studied, the bottles are protected by overwrapping bags during decontamination. The aim of this work is to study the passage and release of H2O2 in eye-drops vials intended for sterility testing without overwrapping.

Methods
Three batches of 7 non-overwrapped vials were decontaminated : 1 batch with Novelia^® Soft PF1500 vials, 1 batch with Novelia^® PF200 and 1 batch with conventional insert vials. The decontaminated vials were pre-filled with 8 mL of water. Decontamination cycles included 8 minutes of H2O2 injection (2g/min), 4 minutes of contact and 13 minutes of aeration. H2O2 measurement was performed at T0 and T24h by spectrophotometry (EvolutionOne^®, Thermofischer) at 560 nm after sample pre-treatment using the Thermo Scientific Pierce quantitative peroxide assay kit (water-compatible formulation). This kit detects and measures H2O2 levels in samples using an iron-based reagent and xylenol orange (XO). This assay method has been validated in accordance with ICH criteria.

Results
The H2O2 assay was validated for concentrations ranging from 0 to 60 μmol/L. Linearity (r = 0.9997), repeatability (CV < 2%), intermediate precision (CV < 5%) and accuracy (100.05% ± 1.96%) were also demonstrated. The limit of detection is 0.32 μmol/L and the limit of quantification is 0.97 μmol/L. For all eye-drops, H2O2 is detectable but not quantifiable at T0. At T24h, H2O2 levels averages 3.75 ± 0.29 μmol/L in Novelia® Soft PF1500 bottles, 4.11 ± 1.96 μmol/L in Novelia® PF200 bottles and 1.95 ± 0.34 μmol/L in conventional bottles.

Discussion-conclusion
The quantities of H2O2 found at T0 are negligible. This result means that the overwrap can be removed for a sterility testing carried out immediately after decontamination. However, the increase in H2O2 concentrations in the vials after 24 hours indicated absorption and release of the decontaminant. In the absence of a threshold value, the impact of this passage on sterility control will have to be assessed in order to decide on the decontamination of vials without overwrapping.

See the poster