Abstract
Introduction
Oxidant agents used for top-decontamination in cytotoxic reconstitution units may spread into IV bags. The amount of hydrogen peroxide passing through the bags and its release have been evaluated using wrapped or unwrapped infusion bags made of three layers of polyolefin (Freeflex®, Fresenius).
Methods
3 experiments were carried out using 2 packages of 30 bags of polyolefin with (package A) or without (package B) overwrapping. A 43 min decontamination cycle with H2O2 has been used for each series of the experiment. A quantitative analysis has been performed by spectrophotometry at 562 nm, measuring a colored adduct formed by ferric iron and xylenol orange. To evaluate the release of H2O2, a sampling was performed from each bag at t0, t1, t2, t7, t21 days.
Results
A linear calibration curve over the concentration range of 0.00–2.70 ppm was used prior to each analysis. The concentration of H2O2 in overwrapped bags (A) was non-significantly different to 0 (p>0.05 Wilcoxon test) until 21 days. At t0 there was no significant difference between package A and B, but from day 2 H2O2 was detected in non-overwrapped bags (B) (First series, at t2 p=6.18×10–10 Wilcoxon test). The higher concentration was reached after 21 days (0.17 ppm).
Conclusion/Discussion
These results allow removing overwrapping for a lean production system, with an extemporaneous use. However, the graduate release of H2O2 and the absence of any toxicity threshold make it impossible to remove the overwrapping for a pre-empted production.
Introduction
Vaporized oxidant agents are used in order to guarantee sterility in the isolators and top-decontamination of medical devices and raw materials used for cytotoxic reconstitution. Two oxidant agents are frequently used: peracetic acid is being used since 1980 whereas hydrogen peroxide has been developed more recently [1].
Unlike peracetic acid, hydrogen peroxide is odorless and slightly corrosive with the same efficiency and almost the same decontamination time. However it remains irritating to skin, mucous membranes and respiratory tract [1].
Firstly hydrogen peroxide (H2O2) turns into water and oxygen through an exothermic reaction [2]. Secondly oxygen leads to the formation of free radicals which attack the lipid membrane and the DNA of microorganisms [3].
During the decontamination process hydrogen peroxide is in contact with infusion bags. Then, it may pass through the bag and affect its integrity, its pH value or oxidize anticancer drugs in which it will be injected in.
Oxidant agents can pass through the bag whatever top-decontamination method is used [2].
The aim of this study is to assess the amount of hydrogen peroxide passing through the bags and its release rate over time.
We have compared H2O2 diffusion during decontamination with or without overwrapping.
In case of a non-significant amount of H2O2 in bags, it will be possible to consider removing the overwrapping before decontamination. This removal will reduce the amount of waste in a rapidly congested environment.
Materials and methods
Materials
Solvents and Reagents:
Freeflex® infusion bags made of three layers of polyolefin, with a polypropylene overwrap, 50 ml of NaCl 0.9 % have been provided by Fresenius Kabi (Sèvres, France) [3].
Hydrogen peroxide solution 30 % has been purchased from Eurobioconcept (Bonneuil-sur-Marne, France)
PierceTM Quantitative Peroxide Assay Kit, Thermo Scientific (Illkirch Cedex, France).
Equipment:
Eppendorf BioSpectrometer® (Montesson, France).
Bags decontamination
Three series of measurement were carried out using two packages of thirty bags; package A composed of 30 overwrapped bags and package B composed of 30 non-overwrapped bags.
Each series has undergone a decontamination cycle with H2O2. The cycle lasted 43 min and included an injection phase of hydrogen peroxide (25 min), a contact phase (3 min) and an aeration phase (15 min).
In order to reflect the storage conditions in the unit, bags were kept in a refrigerator and away from light. Samples were taken at t0, t1, t2, t7, t21 days. The volume collected in each bag and at each sampling time was 1.5 ml. The sampling has been performed after homogenization.
Spectrophotometry assay
This is an indirect method based on the oxidation of ferrous iron (Fe2+) into ferric iron (Fe3+) by hydrogen peroxide [4, 5]: Fenton’s reagent. Then, after an incubation period of 20 min, ferric iron forms a complex with xylenol orange (XO). The complex is detected by a spectrophotometer at 562 nm (Figure 1).
Iron oxidation and complexation reaction.
The analysis was conducted according to the instructions of the kit. First, a solution of ferrous iron and xylenol orange is extemporaneously prepared. Then increasing concentrations of hydrogen peroxide (from 0.00 to 2.70 ppm) are added to the mixture. The reaction has been incubated.
A calibration curve was made prior to each analysis in sodium chloride 0.9 %, to avoid any matrix effect. The calibration range was linear from 0.00 to 2.70 ppm, each R2 value was superior or equal to 0.99 (Figure 2).
Spectroscopy calibration range.
Statistical analysis
Analysis of the results was carried out with a nonparametric Wilcoxon test. This test allows a comparison of the average concentration of hydrogen peroxide in bags of packages A and B with a confidence interval of 95 % (p<0.05).
Results
At day one, no H2O2 concentration has been observed just after the end of decontamination process at t0.
In package A (overwrapped bags), a very slight increase of H2O2 concentration is observed. The average concentration and standard deviation measured is 0 ppm after 21 days for series n°1, 0.014±0.023 ppm after 21 days for series n°2 and 0.004±0.017 ppm after 21 days for series n° 3.
In package B (non-overwrapped bags), there is a gradual increase of H2O2 concentration over time. The maximum concentration measured is 0.17±0.04 ppm after 21 days for series n°1, 0.09±0.03 ppm after 21 days for series n°2 and 0.05±0.04 ppm after 21 days for series n°3 (Figure 3).
Average concentrations of H2O2 with standard deviation at 0, 1, 2, 7 and 21 days – Series n°1 (a), n°2 (b) and n°3 (c).
No significant difference is found between overwrapped and non-overwrapped bags at t0 and t1 whereas at t2, there is a significant difference between bags from package A and bags from package B (p<0.05). Moreover the difference gets more significant with time.
All p values are listed in Table 1.
p values of series 1, 2 and 3.
| Series n°1 | Series n°2 | Series n°3 | |
|---|---|---|---|
| t1 | 1.71E-10 | 3.34E-01 | 4.19E-02 |
| t2 | 6.18E-10 | 2.78E-03 | 5.58E-03 |
| t7 | 5.18E-12 | 5.72E-03 | 1.53E-03 |
| t21 | 5.18E-12 | 5.07E-11 | 5.87E-06 |
Moreover no hydrogen peroxide concentration difference is observed depending on the bags position in the decontamination airlock. We have a homogeneous distribution of oxidant agent.
Discussion
In this paper, for the first time we have demonstrated that a gradual increase of H2O2 concentration over time has been observed in non-overwrapped bags. Plastics can absorb and release oxidizing agents.
However, for the first series there is a peak at t1 with an average H2O2 concentration of about 0.11±0.07 ppm for package B and 0.006±0.023 ppm for package A. This unexpected increase may suggest a contamination of samples during handling of bags and samples assay.
The main issue of this study remains in the absence of a toxicity threshold for hydrogen peroxide [6, 7]. Because of this absence, and due to our observation of a significant H2O2 release over time, we decided to keep the overwrapping of bags according to the precautionary principle.
The lack of hydrogen peroxide at t0 in overwrapped or non-overwrapped bags allows to consider withdrawing the overwrapping for a lean production with an extemporaneous used. However we decided to consider that the release of H2O2 in non-overwrapped bags makes it impossible to remove the overwrapping for a pre-empted production.
Hydrogen peroxide assay by this spectroscopy method has several benefits. Firstly we used a ready-to-use kit with an easy implementation, and secondly the chemical reaction was fast (around twenty minutes).
Perspectives
Another method using high performance liquid chromatography [8] (HPLC) based on the oxidation of triphenylphosphine by hydrogen peroxide could be used for this assay in order to confirm the results of our experiments.
Moreover, it could be interesting to study the impact of oxidation on anti-cancer drugs and monoclonal antibodies, in order to assess if very low quantities of hydrogen peroxide can affect their structure or mechanism of action.
Finally, it may be interesting to compare the importance of absorption and release of oxidizing agents with other plastics.
About the authors
Camille Gérard is a PharmD candidate, currently working as a resident at the Institut Curie (France). She is a former student of the Lille II University, School of Pharmacy (France).
Samuel Huguet is an analytical chemist at the Institut Curie (France) since 2013. His domains of interest are analytical chemistry particularly chromatography and mass spectrometry.
Laurence Escalup is a pharmacist at the Institut Curie (France) since 2000. Her domains of interest are clinical trials and pharmaceuticals technology of oncology.
Isabelle Ferry is a pharmacist at the Institut Curie (France) since 1999. Her domains of interest are clinical trials and oncology.
Marion Lafay is a pharmacist at the Institut Curie (France) since 2009. Her domains of interest are pharmaceuticals technology of oncology.
Julien Fouque is a radiopharmacist at the Institut Curie (France) since 2013. His domains of interest are analytical, oncology, imaging, and radiopharmaceutical.
Olivier Madar is radiopharmacist at the Institut Curie (France) since 2005. His domains of interest are oncology, imaging, and radiopharmaceutical.
Keyvan Rezai is a pharmacologist at the Institut Curie (France) since 2001. His domains of interest are PK/PD and oncology.
Caroline Giard is a pharmacist at the Institut Curie (France) since 2008. Her domains of interest are pharmaceuticals technology of oncology and quality control of chemotherapy preparations.
Conflict of interest statement: Authors state no conflict of interest. All authors have read the journal’s Publication ethics and publication malpractice statement available at the journal’s website and hereby confirm that they comply with all its parts applicable to the present scientific work.
References
1. Rutala WA, Weber DJ. Healthcare Infection Control Practices Advisory Committee (HICPAC). Guideline for Disinfection and Sterilization in Healthcare Facilities, 2008.Search in Google Scholar
2. Muller HJ, Sommermeyer K, Cech F. Permeation of gaseous hydrogen peroxide and peracetic acid into IV bags during their surface sterilization. Hosp Pharmacist 2003;10:84–86.Search in Google Scholar
3. Summary of Product Characteristics for Sodium Chloride 0.9 % Intravenous Infusion. Freeflex Fresenius Kabi.Search in Google Scholar
4. Edet S, Vazquez R, Guerrault-Moro MN, Brossard D, Crauste-Manciet S. Permeation of the sterilizing agent in bags: A simplified routine evaluation method. Group for Evaluation and Research in Areas under Control; Gerpac 2010 Oct 6-8; Giens Peninsula, France.Search in Google Scholar
5. Duval S, André V, Tournamille JF, Rouleau A. Etude de perméabilité des poches de solvant Freeflex au peroxyde d’hydrogène au sein d’une unité de pharmacie clinique oncologique; Group for Evaluation and Research in Areas under Control; Gerpac 2006 Oct 4-6; Giens Peninsula, France.Search in Google Scholar
6. Institut National de Recherche et de Sécurité (viewed on 30/12/2016). Peroxyde d’hydrogène et solutions aqueuses -fiche toxicologique n° 123. [On line]. http://www.inrs.fr/publications/bdd/doc/fichetox.html?refINRS=FT%20123.Search in Google Scholar
7. Substitution-CMR (viewed on 30/12/2016). A tool to support the substitution of CMR substances. [On line]. http://www.substitution-cmr.fr/index.php?id=111&L=0&tx_kleesubstitution_pi3%5BCMR%5D=15&tx_kleesubstitution_pi3%5Bfiche%5D=61&tx_kleesubstitution_pi3%5Bdonnee%5D=177&tx_kleesubstitution_pi3%5Bonglet%5D=4&cHash=e3650e5ff2.Search in Google Scholar
8. Pinkernell U, Effkemann S, Karst U. Simultaneous HPLC determination of peroxyacetic acid and hydrogen peroxide. Anal Chem 1997;69(17):3623–3627.10.1021/ac9701750Search in Google Scholar PubMed
© 2017 Walter de Gruyter GmbH, Berlin/Boston
