Personal Protective Equipment for Healthcare Workers during the COVID-19 Pandemic

1. Potential modes of respiratory virus transmission

In general, respiratory viruses can spread through multiple modes of transmission: contact, respiratory droplets, or aerosols [13]. Contact transmission can occur through direct physical contact with virus-laden respiratory secretions from infected individuals or indirectly through contact with inanimate objects or environments contaminated with the virus [13]. Conventionally, respiratory transmission is classified as either droplet or airborne transmission [4, 7]. It is generally accepted that droplet transmission occurs through deposition of large droplets (>5 µm in diameter) on the mucous membranes (eyes, nose, or mouth) of susceptible people. It occurs when a person is in close proximity to an infected person, as large droplets travel only short distances (<1 m). Airborne transmission occurs through inhalation of aerosols (≤5 µm in diameter) generated from the respiratory tract of an infected person. Aerosols remain suspended in the air for a prolonged period, allowing them to be transmitted over a long distance [4].

However, mode of transmission cannot be simply dichotomized. There is no clear cut-off to differentiate small and large droplets. Different cutoffs have been suggested based on the area of the respiratory tract where particles deposit (respirable particles <10 µm in diameter penetrating the lower respiratory tract or inspirable particles 10 - 100 µm in diameter depositing in the upper respiratory tract) [14] or based on how they behave (particles <10 µm in diameter suspended in the air or particles >20 µm in diameter that settle fast by gravity) [15]. Particle size is dynamic; it depends on the initial size and composition, the force and pressure at emission, environmental conditions (e.g. temperature, relative humidity and airflow), and the time spent airborne [4]. The distance traveled and the length of time particles remain suspended in the air is also determined by particle size, settling velocity, relative humidity, and airflow [16]. Large droplets settle faster due to gravity, contaminating the near vicinity; some of them can rapidly evaporate to form aerosol particles termed “droplet nuclei,” which behave as other aerosols. Settled droplets may facilitate fomite transmission and can be re-suspended in the air by diverse human activities. Large droplets can also move horizontally for more than 2 meters from the source during coughing or up to 8 meters during sneezing [17]. They can remain suspended for prolonged periods in certain environments, especially where turbulent airflow is abundant, such as in hospital settings where doors open constantly [15]. Particles of varying sizes (0.01 – 500 µm) are produced not only by medical procedures but also by respiratory activities such breathing, speaking, singing, coughing, or sneezing [18, 19, 20, 21]. The proportion of aerosol-size particles differs according to the respiratory activities and individuals [21]. As such, it is important to understand that the size of the particles and the resulting behavior follows a continuum; it may overlap either side of this cut-off [21].

However, being airborne does not in itself guarantee effective transmission through aerosols. The virus in aerosols must remain viable in a sufficient quantity to be inhaled by a susceptible host. The virus contained in droplets is subject to biological decay over time, which is affected by the initial metabolic state of the virus, genetic characteristics, and the environment [22, 23]. In this context, the relative contribution of different modes of transmission should be considered, albeit the possibility of airborne transmission does exist. Airborne transmission can be classified as obligate, preferential, or opportunistic. In obligate airborne transmission, transmission occurs only via inhalation of aerosols (e.g., in tuberculosis). Though transmission occurs through multiple routes in preferential airborne transmission, it predominately occurs through aerosols (e.g., in measles, varicella). In opportunistic airborne transmission, the virus is transmitted predominantly through other routes; however, the virus may be transmitted through aerosols under favorable circumstances where aerosols are generated by performing AGPs (e.g., in influenza, SARS-CoV-1 infection) [7, 22, 24].

2. Modes of SARS-CoV-2 transmission

The current consensus regarding the transmission of SARS-CoV-2 is that it is transmitted mainly through respiratory droplets and contact and that airborne transmission is possible during AGPs [9, 10, 11, 25]. Although no study has conclusively linked SARS-CoV-2 transmission to contaminated environmental surfaces, indirect contact with fomites is considered a possible route based on the evidence of heavy environmental contamination in healthcare settings, objects used by COVID-19 patients [26, 27], and the finding that the virus remains viable on plastic surfaces for as long as 3 days [28].

However, there has been controversy whether SARS-CoV-2 can become airborne when AGPs are not performed. Some studies have suggested the potential of airborne SARS-CoV-2 transmission. In one experimental study, viable SARS-CoV-2 was detected in the air for 3 hours when an aerosolized environment was created using a three-jet Collison nebulizer and a Goldberg drum [28]. However, though this experimental condition may simulate circumstances when AGPs are performed, it does not reflect real-life clinical settings. A study in Nebraska detected viral RNA in air samples collected in COVID-19 patient rooms more than 6 feet way from the source patient and in the hallway outside patient rooms, but failed to detect viable virus in air samples [29]. Guo et al. detected SARS-CoV-2 RNA in air samples collected in intensive care units and general wards at a hospital in Wuhan, but the viral RNA was not detected on face shields, in buffer rooms, or in doffing rooms [30]. Liu et al. also found a high concentration of viral RNA in air samples from patients' toilet areas and staff PPE removal areas in two hospitals in Wuhan, suggesting re-suspension of the virus from contaminated surfaces [31]. However, both studies in Wuhan did not investigate the infectivity of the virus in those air samples. The presence of viral RNA in the air does not necessarily indicate viable virus in sufficient amounts to cause infection, nor does it mean that the virus can effectively be transmitted through this route [11, 32]. Further studies are needed to determine whether it is possible to detect viable SARS-CoV-2 in air samples from patient rooms in which no AGPs are performed and what role it may play in transmission. More importantly, in the study by Liu et al., viral RNA was reduced to undetectable levels in staff PPE removal areas after implementation of rigorous disinfection procedures, which emphasizes the importance of environmental disinfection to prevent the spread of the virus in the perspectives of infection prevention and control. In contrast, other studies have shown that viral RNA was not detected in air samples collected from COVID-19 patient rooms [26], 10 cm away from the patient's chin [27], or 2-5 meters away from the patient [33]. Transmission did not occur among HCWs wearing surgical masks when they were exposed to a COVID-19 patient, even during endotracheal intubation [34, 35]. No instances of transmission were observed among HCWs caring for COVID-19 patients when they used surgical masks as part of PPE routine care [36].

Based on these findings, it is believed that SARS-CoV-2 is mainly transmitted through droplets and contact, and that airborne transmission is possible under certain circumstances when aerosols are generated during AGPs or support treatment [9, 11]. At the same time, the possibility of airborne transmission should carefully be considered as new evidence emerges.

1. Surgical mask

Most surgical masks are composed of three-layers: an outer fluid-repelling layer, a middle layer serving as a high filter, and an inner moisture-absorbing layer. Surgical masks without this three-layer feature cannot provide adequate protection [53]. In the US and Europe, surgical masks are classified as medical devices and regulated accordingly. In the US, five elements are tested to standardize their quality: fluid resistance to synthetic blood, particulate and bacterial filtration efficiency, breathing resistance (pressure drop), flammability, and biocompatibility [54, 55]. In Europe, similar standard requirements have been adopted [56]. Surgical masks are categorized into levels 1, 2, or 3 in the US and I, II, or IIR in Europe (Table 2).

Table 2
Comparison of the standard requirements for surgical masks in the US and Europe [55]

• Click for larger image
• Click for full table
• Download as Excel file

In Korea, however, there are no minimum standards or standardized testing methods to determine the filtering efficiency of surgical masks, and the efficiency of the filters in available surgical masks may vary widely. Fluid resistance to water is the only performance test required for surgical masks in Korea [57]. Fluid resistance reflects only one of the surgical mask's purposes: to minimize the amount of fluid that could transfer from the outer layers through to the inner layer in cases of splash or spray. However, the surface tension of water is greater than that of blood, and blood can penetrate through fabrics more readily than water [58, 59]. The lack of equivalent Korean standards makes it difficult for HCWs to choose appropriate surgical masks as recommended by the WHO. Also, it is difficult to uniformly recommend the use of any surgical mask during care for patients with COVID-19 in Korea unless reliable Korean standards for surgical masks are established. Healthcare facilities should cautiously check whether products meet the standard requirements when procuring surgical masks for HCWs.

2. Filtering facepiece respirators

FFRs are labeled according to their filtering efficiency and the national regulations defining the standard conditions. In the US, there are nine classes of FFRs according to filtration efficacy (95%, 99%, and 99.97%) and the filter's oil resistance (N, R, and P). N95 respirators filter 95% of airborne particles 0.3 microns in size and are not resistant to oil. They are regulated under NIOSH CFR Part 84 [60]. The European standard (EN149:2001) places FFRs into three classes: FFP1, FFP2, and FFP3 according to their filtering efficiency (80%, 94%, and 99%, respectively) [61, 62]. As the Korean standards follow the European standards, FFRs manufactured in Korea are classified similarly: KF80, KF94, and KF99 (Table 3) [62, 63]. FFP2/3 and KF94/99 respirators are used for HCWs. In addition to a filtering efficiency test, a breathing resistance test (pressure drop) is required. Pressure drop is an objective measure of breathability; a high pressure drop indicates more difficulty in breathing. KF94 and FFP2 respirators require ≤70 Pa at an airflow rate of 30 L/min, whereas N95 respirators require ≤343 Pa at 85 L/min. Since pressure drop increases with the flow rate, standard pressure drop requirements are similar, even though they appear different [64, 65]. In Korea and Europe, total inward leakage (TIL) is also tested on human subjects (Table 3) [62]. In the US, the TIL test is not performed. Instead, fit testing must be performed prior to working in the environment where wearing a respirator is required and be repeated annually under the Occupational Safety and Health Administration (OSHA) regulation 1910.134 [66]. Despite differences in test methods, it is generally considered that US N95, EU FFP2, and KF94 respirators are equivalent for filtering non-oil based airborne particles [64, 65, 67].

Table 3
Comparison of respirator approval standards for KF masks and FFP respirators [62, 63]

• Click for larger image
• Click for full table
• Download as Excel file

However, concerns have been raised because the fit test is not regularly performed in many Korean hospitals, despite the Korea OSHA recommending a fit test for wearers every year [68, 69]. Respirators must fit the face tightly for effective filtering of airborne particles. Noti et al. demonstrated that a poorly-fitting N95 respirator was not as effective as a tightly fitting respirator at blocking infectious viruses (66.5% vs. 99.6% blocked, respectively) and performed no better than unsealed surgical masks (66.5% vs. 56.6% blocked, respectively) in a simulation experiment [70]. In Korea, the TIL test is performed on ten human subjects doing five types of exercise [63]. This TIL test can eliminate respirators that are inherently poorly-fitting and that do not comply with this requirement or identify that the tested respirator is generally well-fitting. However, fitting is affected by a wearer's face shape and size, age, and gender, as well as the respirator design [71, 72]. Fit testing helps to select a respirator model that fits an individual's face well enough to provide at least the assigned protection factor of 10 [73]. Fit performance was also found to vary by respirator model, ranging from fitting less than 5% to those fitting 95% of the test subjects [71]. In addition to the model type, ear-loop designs appear to be less effective in achieving a proper fit than head-band designs [67]. This is worrisome, since most KF94 masks have ear loops. As AGPs may put HCWs at an increased risk for virus exposure and infection, the design of KF94 masks limits their use during AGPs. KF94 masks of various shapes and sizes and with elastic head-band designs should be offered to HCWs to improve the fitting of the masks. A recent study on the current status of fit testing in Korea showed that 82% of 52 HCWs failed to meet the criteria of fit factor 100, even when using N95 respirators [68]. Considering these findings, HCWs should be fit-tested for FFRs, regardless of their labels (KF94, N95, or FFP2) to ensure respiratory protection. Though it is challenging and laborious for hospitals to implement fit testing practices for all HCWs in the midst of the COVID-19 pandemic, protecting HCWs is of paramount importance.

Even so, fit testing alone does not guarantee respiratory protection [74]. Inappropriate donning and skipping the self-seal-check after donning an FFR were found to be frequent causes of improper fit [68, 74]. Since training on the proper use of FFRs can improve fitting of the respirators among HCWs [74, 75], training programs should be implemented along with fit testing.

The risk of exposure to blood or bodily fluids should also be considered when selecting the proper FFRs, because most FFRs are not water-resistant. To protect HCWs against the splash/spray of blood or bodily fluids as well as airborne particles (i.e., during an operation on a patient with COVID-19), surgical respirators with fluid resistance properties should be used [49]. A surgical N95 respirator, which is approved by the NIOSH as an FFR and the FDA as a surgical mask, is one example.

3. Powered air purifying respirator

PAPRs are increasingly used as an alternative to N95 respirators. PAPRs use a battery-powered fan to force air through a filter, cartridge, or canister to a tight-fitting facepiece or loose-fitting hood [49]. Loose-fitting PAPRs are commonly used in healthcare settings, as they have several advantages: higher respiratory protection with an assigned protection factor of 25 (as compared to 10 for N95 respirators), a barrier against splash, and less difficulty in breathing. They are also reusable, and do not require fit testing [50, 76].

However, there are disadvantages to PAPRs use: They are heavy, may impede HCWs' ability to care for patients, limit communication due to noise, require batteries to be recharged or replaced, and take up significant storage space [76]. Although a fit test is not required, they do need to be properly sized, as protection can decrease with oversized or stretched-out PAPRs [77]. Another disadvantage is that the wearer's exhaled air is unfiltered, which limits the use of PAPRs in close proximity to sterile fields [50, 76]. More importantly, risk of contamination during doffing procedures is high, requiring HCWs to receive special training and assistance in the doffing process. Cleaning and disinfection must be performed between uses. This process must be thorough and performed by trained individuals.

Loose-fitting PAPRs are suitable when AGPs are frequently performed (such as in intensive care unit settings), when HCWs are not able to wear tight-fitting FFRs, or when the fitting of a FFR may be compromised. For safe use, healthcare facilities should be aware of the advantages and disadvantages associated with PAPRs. They must also establish a robust maintenance program, including HCW training for proper PAPR use and the cleaning and disinfection process prior to the use of PAPRs [76, 78].

4. The comparative effectiveness of N95 respirators and surgical masks in preventing respiratory viral infections

Infectious aerosol particles are produced by diverse respiratory activities, including speaking and breathing [79, 80]. HCWs in close proximity to patients with COVID-19 are at risk of short-range airborne transmission as well as large-droplet transmission [81]. As such, there have been debates regarding the effectiveness of surgical masks against the virus in routine patient care, and use of N95 respirators or the equivalent is often advocated [82]. However, no clinical trial has compared the effectiveness of surgical masks and N95 respirators in preventing COVID-19 among HCWs. Based on the systematic review of five observational studies on HCWs, wearing any mask (surgical mask or N95 respirator) reduced the risk of developing respiratory infection (odds ratio [OR] for surgical masks, 0.13; 95% confidence interval [CI], 0.03 - 0.62 vs. OR for N95 respirators, 0.12; 95% CI, 0.05 - 0.26) [83]. A recent randomized clinical trial in the US demonstrated no significant difference in the incidence of laboratory confirmed influenza between outpatient HCWs wearing surgical masks and those wearing N95 respirators [84]. Two meta-analyses, which were separately performed by different research groups, reached the same conclusion: Surgical masks and N95 respirators offer similar protection against respiratory viral infection among HCWs during non-aerosol-generating care [85, 86]. Based on these findings, the Infectious Disease Society of America recommends that HCWs caring for patients with suspected or confirmed COVID-19 use either a surgical mask or N95 (or N99 or PRPR) respirator and that HCWs involved in AGPs use N95 or higher-level respirators [83]. Chu et al. investigated the effectiveness of face masks in preventing transmission of SARS, MERS, or COVID-19 in healthcare and non-healthcare settings by analyzing 44 observational studies. They found that the use of face masks (12 - 16-layer cotton masks, surgical masks, N95, or similar respirators) resulted in a large reduction of infection risk in healthcare settings (relative risk [RR], 0.30; 95% CI, 0.22 - 0.41). N95 or similar respirators had a stronger protective association (RR, 0.04; 95% CI, 0.004 - 0.30) than surgical masks or 12 - 16-layer cotton masks (RR, 0.33; 95% CI, 0.17 - 0.61), and both N95 and surgical masks had a strong association with protection when compared to single-layer masks [87]. The review, however, included only four studies comparing N95 or similar respirators with no mask, and two of them involved situations in which AGPs were performed. Based on this review alone, it is difficult to generalize that the use of N95 or similar respirators provides more protection during routine care for patients with COVID-19. Therefore, the use of N95, FFP2, or higher-level respirators such as PAPRs should be prioritized when AGPs are performed. It is also necessary to vigilantly monitor situations or procedures that may increase the possibility of aerosol transmission, because many of the characteristics of SARS-CoV-2 remain unknown.