key: cord-0836538-kt92w0ez
authors: van der Vossen, Jos M.B.M.; Heerikhuisen, Margreet; Traversari, Roberto A.A.L.; van Wuijckhuijse, Arjan L.; Montijn, Roy C.
title: Heat sterilisation dramatically reduces filter efficiency of the majority of FFP2 and KN-95 respirators
date: 2020-10-22
journal: J Hosp Infect
DOI: 10.1016/j.jhin.2020.10.012
sha: b07f45dd81e28954450daff22c9460603218767c
doc_id: 836538
cord_uid: kt92w0ez

BACKGROUND: Because of the enormous demand for personal protective equipment and especially respiratory protective devices (respirators) during the initial phase of the corona pandemic shortages arose. Sterilisation of used respirators can reduce these shortages. In our study, respirator testing was carried out after only one sterilisation cycle. AIM: To determine if steam sterilisation and reuse could be safely applied for used respirators. METHODS: In a cabinet an aqueous solution of NaCl (0.02% w/v) was nebulized and passed through a sample of the material of a respirator. Passing particle concentrations were measured directly from the cabinet and via the filter material of the respirator for particles ≥ 0.3 μm, ≥ 0.5 μm and ≥ 1.0 μm. FINDINGS: only three out of ten steam sterilised respirators met the requirements of 94% filtration efficiency. CONCLUSION: The results prove that heat sterilisation cannot be generically applied for reuse of respirators safely.

The Corona pandemic crisis quickly resulted in an enormous demand for personal protective equipment and especially respiratory protective devices (respirators). Initially, China faced a huge demand for respirators [1] . The resulting shortage in respirators spread to other countries as national guidelines made recommendations for use by respirators by healthcare workers and the public [2, 3] .

Possible solutions for the shortage in respirators were exploring the most promising decontamination procedures that allow for safe reuse. Rubio-Romero et al. [4] reviewed the options for reuse. They concluded that autoclaving or high temperature steam treatment (121 °C) is not fully recommended.

Despite this information, several health care institutions, many of them hospitals, have continued to explore the effect of heat sterilisation on the filter efficiency of respirators. Insight into these effects might allow strained health care institutions to re-use these protective respirators if the filter properties of the respirators are not negatively affected by the sterilisation process.

Given the enormous demand of particularly FFP2 and comparable non-CE-marked KN95, KP95, N95, P95, R95 respirators in intensive care units with Covid-19 patients, The Netherlands Organisation for Applied Scientific Research (TNO) decided to modify their test system for testing respirators. The system was previously developed for testing the protective properties of packaging material to protect medical equipment after heat sterilisation for penetration of contaminants, known as the "final pack test" [5] . This test determines the ability to keep particles with 1 µm diameter size, representing the average size of bacteria, out of the packaging after cooling down from sterilisation. In the modified system with test set-up for respirator materials, a range of particles is used with sizes between 0.3 µm and 1 µm, thus including also smaller particles. 

In a conditioned cabinet of about 1 m 3 , two test units with a circular dimension of 50 mm diameter were placed, one of which was covered by respirator material with a clamp to prevent leakage, while the other remains completely opened. In this cabinet with test units, an aqueous solution of NaCl (0.02% w/v) was nebulized. Subsequently, the resulting particles were sampled with a flow of 28 L/minute (1 ft 3 /minute) via both units, with and without respirator filter material, through equal -length tubing of Tygon® in the cabinet connected to polyurethane tubing outside the cabinet and subsequently through a Lighthouse Solair particle counter 3100 with respirator material and Lighthouse Solair particle counter 1100 without respirator material in each channel. These particle counters detect and quantify particles in three diameter sizes: 0.3, 0.5 and 1.0 µm during a period of 10 minutes. The setup of the test system is schematically presented in figure 1.

The higher the filter efficiency of the respirator material, the more particles are retained. From the detected number of particles in the surrounding air which went through the unit in the channel without respirator material and the number of particles that penetrate through the respirator, the filter efficiency of the sample of the material of the sterilised respirators was calculated and compared to the filter efficiency of a new reference respirator (non-sterilised) of the same type and brand. Analyses of the respirators were performed in triplicate from different respirators. For each particle size, the filter efficiency was calculated by equation 1. New respirators that meet the above-mentioned criteria set for this study were included for subsequent heat sterilisation 15 minutes 121°C in an autoclave. The respirators included in the test are listed in Table 1 . Testing of the filter material was performed at least in triplicate. According to the European standard for performance criteria for respiratory protective FFP respirators should not only being judged on the filter efficiency but also be assessed on other aspects e.g. face seal leakage, J o u r n a l P r e -p r o o f leakage of the exhalation valve leakage if fitted, breathing resistance, field of vision, compatibility with skin. Because these tests are performed on the faces of users that are not uniform, these tests therefore require a relative large number of replicates. In the filter efficiency experiments the unit with mounted respirator was uniform and did not allow any leakage because of the use of a tightly closed clamp around the respirator material. Therefore, the passage of particles through the respirator material was guaranteed so triplicate measurements of the efficiency were assumed sufficient.

If the disinfected respirators perform comparably as the new reference respirators (not disinfected) and meet the criteria set above, then the disinfection method was considered acceptable for the specific disinfection method applied, which was in this case steam sterilisation during 15 minutes at 121°C. If disinfected respirators show a reduced filter efficiency, these were judged unsafe for use by health care professionals nursing COVID-19 patients.

Determination of the filter efficiency of respirator materials showed large differences based on baseline analyses of the unused new respirators. Best performing respirators were FFP2-2, FFP2-5 and KN95-4 as can be observed in figure 2 . In this figure, the results of single fold sterilised respirators were also included except for those respirators that did not comply with the pre-set criteria as described in "Comparative testing of filter efficiency of respirator material". Therefore, respirators KN95-1 and KN95-2 were not included for sterilisation testing. The results shown in Figure 2 , indicate the robustness of the respirators FFP2-2. These respirators only show a marginal reduction in filter efficiency after sterilisation while the other respirators show a huge drop in filter efficiency. However, apart from FFP2-2 also KN95-3 and FFP2-5 meet the minimum requirements described in "Comparative testing of filter efficiency of respirator material".

While the studied FFP2 respirators in the particle filter test meet the pre-set minimal criteria, non-CEmarked respirators also may be as effective in filtering particles as CE-market FFP2 respirators based on the observations in this study with the KN95-3, KN95-4 and KN95-5 respirators. A similar J o u r n a l P r e -p r o o f observation was described by Wezel et al. [7] . They concluded that a shortage of CE-approved can be abolished by using non-CE-market KN95 respirators after applying simple in house testing. The testing of non-CE-market, Conformité Européenne, not to be confused with CE, China Exported, is also advised by the authors based on their test experiences, since a lot of the supplied certificates appear to be forged. Apart from using these alternative non-CE-marked respirators, heat sterilisation was also suggested as an alternative in the fight against respirator shortages [8] . That appeared to be a valid option for some brands of respirators but certainly not for the majority of respirators as demonstrated in this study. The majority of respirators tested lose their minimal required filter capacity. Although this study involved only a limited number of FFP2 and KN95 respirators, this study demonstrates that not all brands can resist heat sterilisation. However, simple in-house testing may provide the essential insight whether heat sterilisation can be used as an alternative for a particular respirator brand.

The results prove that heat sterilisation cannot be generically applied for reuse of respirators safely. J o u r n a l P r e -p r o o f minutes. The absence of coloured bars mean efficiency score is below 90% and respirators KN95-1 and KN95-2 were not tested for sterilisation resistance because of underperformance based on preset criteria. Therefore, post-sterilisation filter performance of these masks is not included in the figure.

J o u r n a l P r e -p r o o f 0,3µ before 0,5µ before 1µ before 0,3µ after 0,5µ after 1µ after

He Z. Ming W-K. Facemask shortage and the coronavirus disease (COVID-19) outbreak: Reflection on public health measures

Advice on the use of masks in the context of COVID-19: interim guidance

Disposable masks: Disinfection and sterilization for reuse, and non-certified manufacturing, in the face of shortages during the COVID-19 pandemic

Einfluss des Prionen-Sterilisationszyklus auf die

Respiratory protective devices -Filtering half masks to protect against particles -Requirements, testing, marking

In-hospital verification of non-CE-marked respiratory protective devices to ensure safety of healthcare staff during the COVID-19 outbreak

Sterilization of disposable face masks by means of standardized dry and steam sterilization processes; an alternative in the fight against mask shortages due to COVID-19

We thank Daniel van Rijswijk and Volmar Hatt for their technical assistance and modification of the test rig.

J o u r n a l P r e -p r o o f

Filter efficiency = Number of particles collected without filtering -Number particles through filter X 100% Number of particles collected without filtering