Sterilisation and PFE test data of the Aura 1862+ (3M, Saint Paul, Minnesota, USA) face mask indicate that this type of face mask shows good performance after multiple sterilisation cycles.11,12,13 From the 18,166 reprocessed medical, single use FFP2 face masks sterilised after use in a medical autoclave at the sterilisation department of CSA Services, located in De Meern – Utrecht, the Netherlands, the majority (n=7,993) were Aura 1862+ (3M, Saint Paul, Minnesota, USA). Therefore, this particular type of face mask was chosen for the LCA.
The packaging materials were disposed of in the hospital where the face masks are used primarily. After first use, face masks were transported to the sterilisation department. All masks were manually checked before reprocessing by personnel wearing PPE. Of all Aura 1862+ facemasks that entered the CSA, approximately 10% were discarded. To remain conservative, the total rejection rate was set at 20%.
Face masks were placed in a sterilization bag that contained up to five masks. A total of 1,000 masks were placed into an autoclave (Getinge, GSS6713H-E, Sweden) per cycle. After sterilization, the masks were transported to the hospital. Masks were reprocessed a maximum of five times before final disposal.11,12
The assessment of environmental impact is performed following as closely as possible the internationally accepted life cycle assessment (LCA) method following the ISO 14040 and 14044 standards.16,17 The LCA examines all the phases of the product's life cycle from raw material extraction to production, packaging, transport, use and reprocessing until final disposal.18 The LCA was modelled using SimaPro 9.1.0.7 (PRé Sustainability, Amersfoort, The Netherlands). Life cycle impact analysis data were retrieved from the Ecoinvent database (Ecoinvent version 3.6, Zürich, Switzerland).19
In the LCA, the ‘functional unit’ defines the primary function that is fulfilled by the investigational products and indicates how much of this function is considered.20 In this study, we pragmatically chose ‘protection of 100 healthcare workers against airborne viruses using an FFP2 certified face mask’.
To make a valid comparison between the disposable and reprocessing face masks, the system boundaries should be equal in both scenarios. The system boundaries in this study consisted of the production, use and eventually disposal and waste treatment of the masks. For the reprocessed face masks, the lifecycle is extended due to the sterilisation process (figure 1). The production of machinery for the manufacturing of face masks and autoclaves was not included in this study.
Life cycle inventory analysis (LCIA)
The inventory data include all phases from production (including material production and part production), transport, sterilisation to end-of-life of the life cycle of the single use and reprocessed face masks. We disassembled one face mask to obtain the weight of each individual component on a precision scale (Fit Evolve, Bangosa Digital, Groningen, the Netherlands) with a calibrated inaccuracy of 1.5%. Component information and materials were obtained from the data fact sheet provided by the manufacturer. We conducted a separate validation experiment to establish the material composition in the filtering fabric (supplement file 1).
The production facility for the face masks is located in Shanghai, China.21,22 Further distribution took place from Bracknell, UK, to Neuss, Germany, and the final destination was set in Rotterdam, the Netherlands.
The information gathered in the life cycle impact assessment (LCI) is stated in the ReCiPe midpoints by using SimaPro23. The kg CO2 eq was chosen as the primary unit in the impact category.
Uncertainty analysis
The final LCA model contains several uncertainties based on assumptions and measurement inaccuracies.24 To quantify the effect of the uncertainties on the primary endpoint, we performed a Monte Carlo analysis in which LCA system values were sampled from their probability function rather than assumed as a point estimate.23 Thus, the probability function for all primary and secondary outcome measures was determined based on nominal value, a spread around that value and a probability distribution. An unpaired, two-sided T-test was conducted to identify differences between reprocessing and disposable scenarios. P<0.05 was considered to be statistically significant (supplemental file 2).
Sensitivity analysis
A sensitivity analysis was conducted to check the sensitivity of the outcome measures to variation in the input parameters. To determine which parameters are interesting to investigate, two aspects were considered: the variations of the assumption made for the parameter and the relative contribution of the parameters to the total impact. The following three parameter variations were chosen for the sensitivity analysis:
- Rejection percentage. The rejection rate was defined based on experiences from the participating sterilisation department and studies that show that sterilisation of the face masks up to 5 times is possible. Masks were used 5 times, and approximately 10% were discarded during the total life cycle. Out of this experience and to remain conservative, the total rejection rate was set at 20%. Therefore, it is interesting to investigate whether variation in PFE testing outcomes or differences in user protocols influence the outcomes. This should indicate whether masks of higher or lower quality can also be suitable candidates for reprocessing.
- Autoclave capacity largely depends on the loading of the autoclave. To mimic different loads of the autoclave, it is interesting to know the influence of sterilising fewer masks per run on the model.
- As it is likely that many hospitals have a Central Sterilisation Services Department (CSSD), it is interesting to know the effect of having zero transportation. Moreover, in case hospitals are not willing to change the routing in their CSSD, it is interesting to observe how outcomes are influenced if transportation is set on the maximal realistic value of 200 km.
The parameters were varied with 250 and 500 face masks per sterilisation batch. A rate varying with 10% and 30% of the face masks being rejected due to quality reasons and variation in transport kilometres of 0 km – 200 km.
Cost price comparison
A cost analysis was performed to provide insight into costing from a procurement perspective. The cost analysis is based upon five face masks in a permeable laminate bag, Halyard type CLFP150X300WI-S20, and includes the expenses of energy, depreciation, water consumption, cost of personnel, and overhead and is compared to the prices for a new disposable 3M Aura face mask during the first and second Corona waves.