Use of powered air-purifying respirator (PAPR) by healthcare workers for preventing highly infectious viral diseases—a systematic review of evidence

High infectivity combined with high case fatality rate during the COVID-19 pandemic has placed an emphasis on healthcare worker (HCW) protection both from a personal as well as a societal perspective. Several other outbreaks of virulent highly infectious diseases have occurred in recent decades including the Ebola crisis in 2014–2016, Middle East respiratory syndrome coronavirus (MERS-CoV), and severe acute respiratory syndrome (SARS, due to SARS-CoV-1) epidemic [1, 2]. Teasing out the true infection risk in the HCW group is difficult. This is due to the high rates of community infection, HCW travel and under-reporting of non-HCW populations, and the lack of phylogenetic viral analysis. Personal protective equipment (PPE) and infection control guidelines from the WHO are based on the assumption that the primary mechanism of transmission of SARS-CoV-2 is direct and indirect droplet spread as well as fomite transmission [3]. The WHO advises that airborne transmission can occur, but only when aerosol-generating procedures (AGPs) are performed in patients infected with SARS-CoV-2 [4]. Aerosol-generating procedures result in the generation of small aerosolized particles through disruption of surface tension of the alveolar lining [5]. Aerosolized particle clouds can travel up to 8 m [6]. A detailed list of AGPs is provided in Table 1 [7]. The degree of airborne spread in the coronavirus group is contentious [8, 9]. Recently, the stability of SARS-Cov-2 and SARS-Cov-1 was evaluated under different experimental conditions [10]. SARS-CoV-2 and SARS-CoV-1 remained viable in aerosols throughout the 3-h duration of the experiment with a reduction in infectious titer [10]. However, the clinical relevance of this experimental model has been questioned [11]. Establishing with certainty whether SARS-CoV-2 is infectious through airborne transmission may be methodologically challenging.

In this review, we consider the implication for HCWs of Ebola in addition to the coronaviruses. Ebola virus can be transmitted by direct contact with blood, bodily fluids, or skin of Ebola patients or individuals who have died of the disease. Development of Ebola disease results in a high case fatality rate, as high as 50%. Recommendations for respiratory protective equipment are therefore similar [12].

Our findings have been reported according to the standards for the Preferred Reporting Items for Systematic Reviews and Meta-Analysis [28]. The protocol for this review was prospectively registered with the International Register of Systematic Reviews identification number CRD42020184724.

Recently published field studies of HCWs managing patients with COVID-19 demonstrated equivalent rates of healthcare provider infection in cohorts utilizing PAPR versus other appropriate respiratory protection. We identified a trend towards a lower level of cross-contamination in participants using PAPR technology compared to alternative respiratory protection in low-quality simulation studies. We identified moderate quality of evidence towards improved healthcare worker comfort (heat tolerance and visibility) with PAPR technology compared with alternative respirators. PAPR users scored the technology lower with on mobility, dexterity, audibility, and communication. We identified moderate quality of evidence towards improved healthcare worker comfort (audibility and mobility) with APR (airflow-powered respirator) technology compared with PAPR.

There appears to be no reported difference in observed infection rates in participants utilizing PAPR or other appropriate respiratory protection. The preferred use of PAPR for respiratory protection may be due to perceived logistical advantages by institutional policy makers. A prospective international multicenter cohort study found no difference in infection rates between cohorts utilizing varied respiratory protection [50]. A series published recently found no airway proceduralist infections in the cohort utilizing PAPR versus a cohort equipped with more routine respiratory protection in addition to usual PPE [41]. This study was performed retrospectively in Wuhan during the outbreak of SARS-CoV-2 [41]. Differences in the airway proceduralist’s COVID outcomes appear distinct: 10.7% in the El-Boghdadly et al. study versus 0% reported in Yao et al. [41, 50]. These findings may have been confounded by a well-designed enhanced respiratory and contact protective system in the study with no provider infections.

We observed a trend towards lower contamination rates in simulation studies in participants utilizing PAPR [40, 42]. These observations are counterintuitive towards an assumption that due to the complexity of technology, cross-contamination during doffing with PAPR is more likely. The results of our review demonstrate a trend towards lower HCW contamination rates and decreased doffing violations whilst utilizing PAPR.

We found no studies which assessed compliance with donning or doffing protocols for equipment utilizing PAPR.

In line with subjective reports that PAPR may be more effective in decreasing the effort needed to maintain the work of breathing compared to a more conventional filtering facepiece, we identified moderate quality of evidence for this outcome [7, 43]. We identified a moderate quality of evidence towards improved healthcare worker comfort with regard to heat tolerance and visibility with PAPR technology. It is thought that through the positive airflow, PAPR’s eliminated the heat build-up [52]. A decrease in audibility and communication difficulties can be anticipated due to increased weight of the equipment and noise generated by positive airflow. In observational studies, we identified a trend towards a greater level of self-reported comfort amongst the PAPR wearers [42, 44]. Powell et al. noted a lower temperature measurement in subjects using PAPR [43]. Prior reports have outlined the potential for claustrophobia in healthcare providers with field use of PAPR [53]. During the tuberculosis outbreaks, the use of PAPR’s had a low institutional uptake. This occurred due to a number of factors, including concerns that doctors would appear frightening to their patients and that the motor’s hissing noise would interfere with patient communication [54]. The greater acceptance of PAPR by HCWS during both the SARS-Cov-1 pandemic and Ebola may be influenced by HCW perception of relative risk. Khoo et al. published a survey illustrating that PAPR as opposed to N95 was more comfortable for HCWs during an Ebola outbreak in Singapore [55].

We identified no studies exploring the costs of the resource use of PAPR versus any other filtration pieces. The costs of maintenance of PAPR equipment which require disinfection, cleaning, self-storage, battery maintenance, and a requirement for education of a significant proportion of the HCW workforce have not been considered in evidence-based literature. These costs are juxtaposed against more wears per piece of PAPR compared to disposable face-filtering pieces. The PAPR use may be a resource utilization prepared strategy for times of a greater need for N95/FFP2. It has been noted that there have been fewer equipment shortages for PAPR than N95 [56].

We identified a single simulation randomized controlled trial which demonstrated a trend towards a lower risk of contamination when the PAPR use was incorporated with a teaching program. During the SARS-CoV-1 outbreak, recent training in infection control increased the likelihood of workers’ adherence to recommended barrier precautions [57]. Whilst the initial focus was on the use of more stringent respiratory PPE components, further studies found that SARS-CoV-1 transmission was not supported if more standardized PPE was used. Critical system factors protecting the HCWs included compliance with N95 mask application and ongoing use, as well as complementary respiratory protection protocols [25].

Current reports of the choice of protective respiratory technology during the SARS-CoV-2 pandemic are disparate. In a recently published experience of intubation and ventilation of critically ill patients in Wuhan, Meng et al. illustrated the use of a positive pressure ventilation system for anesthesiologists dealing with COVID-19-positive patients [58]. There have been three separate descriptive reports from Singapore on the routine use of PAPR in their protocols for anesthesia in suspected or confirmed COVID-19 patients [59, 60] [27]. Recommendations from the Joint Task Force of the Chinese Society of Anesthesiology and the Chinese Association of Anesthesiologists center on the N95 use for proceduralists. These recommendations do not specifically mention the use of a PAPR device. Although some authors make recommendations for the use of PAPR for critical care of COVID-19 patients, they acknowledge that there is no conclusive evidence to show that this advanced respiratory technology decreases the likelihood of viral airborne transmission [61].

The utilization of PAPR with high filtration efficiency may represent an example of a “precautionary principle” wherein the action taken to reduce the risk is guided by logistical advantages of the PAPR system. With a higher APF factor than N95 masks, it is scientifically plausible that the PAPR use may result in a long-term lower HCW infection rates. There is however limited literature supporting the PAPR use during epidemics/pandemics of SARS-CoV-1, SARS-CoV-2, MERS, and Ebola. Given the lack of demonstrable efficacy, institutional decision makers may be applying a pragmatic choice to use PAPR on the basis of the precautionary principle.

Current PAPR certification standards have been developed primarily for industrial applications. There is a need for respirator standards to better expand to suit the requirements of healthcare workers [62]. In terms of the laboratory research, industrial, radioactive, or biological particles behave in a similar manner with regard to a filtration standard. The quantification of the infectious dose with this emerging viral disease has not occurred. Therefore, it remains a challenge to determine the optimum respiratory protection under individual circumstances. Future developments include adjusting the testing standards to activities to which the user (HCW) is engaged.

Our systematic review has been limited by a number of available studies graded as low evidence. A recently published study by El-Boghdadly et al. had only 28.8% of laboratory-confirmed infections. The remainder were diagnosed through self-isolation and hospitalization without confirmed laboratory testing [50]. In addition, in the absence of phylogenetic analysis, it is not possible to conclude the source of infection, be it patient contact or community-acquired. The comparison of infection rates with HCWs not wearing the PAPR technology may be biased by other PPE protection factors such as the utility of system-related compliance measures [63]. Despite the theoretical advantages of PAPR, there have to date been no controlled clinical trials on the efficacy of this technology during the SARS-Cov-1, SARS-Cov-2, EBOLA, or MERS pandemics in comparison with other high-level respiratory protection [64]. At present, the minimal infective dose for SARS-CoV-2 pathogen is unknown for any of the transmission modes [65]. Higher viral load shedding may be more readily associated with greater disease severity [66]. Whether a higher PAPR filtration factor translates to decreased infection rates of HCWs remains to be elucidated. True randomized controlled studies may not be ethically feasible due to the higher filtration factor of PAPR. Pragmatic observational studies, as published recently in well-resourced areas may be both more ethical and feasible [41]. Most of the studies included have been performed using simulation. Despite the simulation being designed to simulate exposure to highly contagious diseases, they are performed in a safe setting without true haste [46]. This may introduce systematic bias to the studies themselves and the review. We graded the risk of bias in observational on-field studies as high. This is due to a number of factors including the observational nature of SARS-CoV-2 infection rate assessment and potential for confounding due to attendant infection control processes.

Equivalent rates of healthcare provider infection have been demonstrated in cohorts utilizing PAPR versus other appropriate respiratory protection. There have been no field studies reporting the effectiveness and utility of PAPR in protecting the healthcare workers from cross-infection due to other highly virulent viral diseases including SARS-CoV-1, Ebola, or MERS. Evidence base of low quality indicates greater wearer protection in HCWs using PAPR compared to alternative respiratory devices, from cross-contamination and during doffing in simulation studies. Provider satisfaction appears higher with regard to thermal comfort; however, lower in relation to audibility and mobility with PAPR technology. Precautionary principles may be guiding the institutional risk management strategies of HCW protection during epidemics and pandemics.

The closure of this knowledge gap with regard to optimal respiratory protection during pre-defined highly virulent pandemics needs further prospectively collected field data.