Evaluation of evidence of prevention and management of facial pressure injuries in medical staff

Abstract Aim This systematic review evaluated the quality of evidence for the prevention and management of facial pressure injuries in medical staff. Design This review was presented in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines. Methods We retrieved the relevant studies from 19 databases. Using the literature evaluation standards and evidence grading system of the Australian Joanna Briggs Institute Evidence‐Based Health Care Center, we evaluated the quality of the literature encompassing different types of research and assessed their levels of evidence. Results A total of 13 studies were included, including seven expert consensuses, two recommended practices, one clinical decision, one best practice information booklet, one systematic review and one randomized controlled trial. In the end, 31 best evidence were summarized, including skin cleaning and care, PPE placement and movement, reasonable use of dressings, treatment measures and education and training.


| INTRODUC TI ON
In December 2019, the novel coronavirus pneumonia outbreak occurred. On 12 January 2020, the World Health Organization (WHO) named the disease coronavirus disease 2019 (COVID-19) (Jin et al., 2020), and on 30 January 2020, the WHO declared it a public health emergency of international concern (Cucinotta & Vanelli, 2020). Droplet transmission and contact transmission through the respiratory tract are currently believed to be the main transmission routes of COVID-19 (General Office of National Health Committee, 2021). In terms of the age distribution of the infected individuals, not all age groups showed COVID-19 resistance . Patients with COVID-19 and individuals in close contact with infected yet asymptomatic individuals are high-risk groups of COVID-19 (Epidemiology Working Group for NCIP Epidemic Response & Chinese Center for Disease Control and Prevention, 2020; Wang et al., 2020). The medical staff attending to COVID-19 patients are at high risk of infection as they have to closely interact with the patients (Kluytmans-van den Bergh et al., 2020;Kua et al., 2021). Therefore, to minimize the risk of contracting the virus, doctors, nurses and other medical professionals around the world need to use personal protective equipment (PPE; Mahmood et al., 2020).
In the process of using masks, goggles and face shields to avoid contracting the disease, the use of PPE entails close contact with the skin, and to prevent cross-infection, the removal of protective equipment needs to be minimized, which results in prolonged use of the protective equipment and sustained pressure on local skin (European Pressure Ulcer Advisory Panel (EPUAP), 2021). Under such a demanding work environment requiring prolonged use of face masks coupled with increased workload and mental stress, the skin sweats a lot, resulting in a humid environment inside the mask (Bischoff et al., 2019). The skin is the first line of defence against the environment, and it is repeatedly affected by physical factors (sustained pressure, tension and friction) and chemical factors (humidity), and because of these factors, the tolerance of the skin changes, causing skin resistance to decline (Sivamani et al., 2003). These factors are directly associated with the occurrence of pressure ulcers and friction injuries (Schwartz et al., 2018) and eventually lead to facial pressure injuries. The most commonly affected parts are the nasal bridge, cheeks, forehead and ears (Darlenski & Tsankov, 2020).
The results of a multi-centre cross-sectional survey conducted by Jiang et al. (2020) revealed that the overall incidence of skin damage caused by PPE among medical staff was 42.8%. Facial pressure injuries are usually regarded as mild irritation and are often overlooked (LeBlanc et al., 2021). However, it is worth noting that even minor skin irritation can increase the risk of infection in medical staff because skin irritation may cause people to subconsciously touch their faces when they are not wearing PPE (Gefen & Ousey, 2020a;Kantor, 2020). Furthermore, facial pressure injuries can also cause discomfort and local skin conditions, such as erythema, indentation, a stinging sensation, reduced facial comfort, localized warmth and weakened sense of touch, and occasional skin breakage (Wang & Parish, 2019).
Fortunately, if appropriate measures are taken, the facial and ear injuries that medical staff are most vulnerable to can also be prevented. A parallel double-arm randomized clinical trial compared the use of foam and extra-thin hydrocolloid dressing by medical staff working at the forefront of the fight against COVID-19 to prevent PPE-induced facial stress injuries (Gasparino et al., 2021). The results showed that foam and extra-thin hydrocolloid dressing could effectively prevent PPE-induced facial stress injuries (Gasparino et al., 2021). The National Pressure Injury Advisory Panel recommends that medical staff clean the skin before and after wearing PPE, treat PPE-related pressure injury and reduce the pressure caused by PPE (European Pressure Ulcer Advisory Panel (EPUAP), 2021). A self-controlled study showed that the use of a hydrocolloid dressing combined with 3 M Cavilon No-Sting Barrier Film for facial skincare can effectively reduce the incidence of facial pressure injuries in medical staff (Zhang et al., 2021). At present, several studies have reported on the prevention and management of facial pressure injuries in medical staff. However, uniform standards for the prevention and management of facial pressure injuries in medical staff remain to be established.
This study was aimed at (1) providing references for clinical practice in the prevention and management of facial pressure injuries in medical staff, (2) reducing the incidence of such injuries and (3) improving the facial comfort of medical staff. To this end, we collected and evaluated the evidence obtained through evidence-based nursing methods and summarized the findings.

| Establishment of the problem
We used the problem development tool of the population, intervention, professional, outcome, setting and type of evidence (PIPOST) model of the Joanna Briggs Institute (JBI) Evidence-Based Health Care Center to construct evidence-based problems. The first 'P' (population) is the target population for the application of evidence.
This includes medical personnel working at the frontlines in the fight against COVID-19, those working in fever clinics and those who go out for batch nucleic acid testing or to infection wards and need to wear protective equipment. 'I' (intervention) is the recommended intervention. This includes the prevention and management of facial pressure injuries caused by the use of protective equipment. The second 'P' (professional) is the implementer of evidence application, that is, medical staff. 'O' (outcome) is the outcome indicator, which is the alleviation of facial pressure injury and restoration of facial comfort. 'S' (setting) is the evidence application site, that is, anti-epidemic frontline, fever clinic, nucleic acid testing site or infection ward. 'T' (type of evidence) is the type of evidence resource, which includes guidelines, expert consensus, clinical decisions, recommended practices, best practice information booklets, systematic reviews, evidence summaries and randomized controlled trials (RCTs).
Disease 2019' and 'COVID-19 Pandemic'. The article types searched were guidelines, expert consensus, clinical decisions, recommended practices, best practice information booklets, systematic reviews, evidence summaries and RCTs. The retrieval period was from 1 January 2010 to 23 October 2021.

| Study inclusion and exclusion criteria
The inclusion criteria were as follows: (1) the research object was medical staff; (2) the research was aimed at the prevention and management of facial pressure injuries; (3) the types of reports included were guidelines, expert consensus, clinical decisions, practice recommendations, best practice information booklets, systematic reviews, evidence summaries and RCT findings; and (4) the language was Chinese or English.
The exclusion criteria were as follows: (1) duplicate publications; (2) studies with original text unavailable; and (3) studies with unqualified quality evaluation.

| Literature quality evaluation process
The literature quality evaluation was independently completed by two authors. For training, these authors had enrolled into and completed the Evidence-Based Nursing Practitioner Training Class of the JBI Evidence-based Nursing Cooperation Center of Fudan University. If there was a disagreement, the two authors tried to resolve their disagreements by discussion, and if they could still not reach a consensus, they invited evidence-based nursing experts and evidence-based training instructors to make a judgement after thorough discussion.

Evaluation of guidelines
The quality evaluation standard of guidelines adopted the Appraisal of Guidelines for Research and Evaluation Instrument (AGREE II; Brouwers et al., 2010) for quantitative scoring. The scale had a total of six areas, 23 main items, and two additional overall evaluation items. Each item was evaluated on a scale of 1 to 7 (1 = strong disagreement, 7 = strong agreement). According to the scores in each field, the recommended level was evaluated.

Evaluation of expert consensus
The quality evaluation standard of the expert consensus adopted the corresponding evaluation standard of the Australian JBI Evidencebased Health Care Center (2016; Tufanaru et al., 2020). There were six items in total. The evaluation results were divided into 'Yes', 'No', 'Unclear' and 'Not applicable'.

Evaluation of systematic reviews
The quality evaluation standard of systematic reviews adopted the corresponding evaluation standard of the Australian JBI Evidencebased Health Care Center (2016; Aromataris et al., 2015). There were 11 items in total. The evaluation options were divided into 'Yes', 'No', 'Unclear' and 'Not applicable'.

Evaluation of randomized controlled trials
The quality evaluation standard of randomized controlled trial reports adopted the corresponding evaluation standard of the Australian JBI Evidence-Based Health Care Center (2016;Tufanaru et al., 2020). There were 13 items in total, and the evaluation options were divided into 'Yes', 'No', 'Unclear' and 'Not applicable'.

Evaluation of clinical decisions, practice recommendations, best practice information booklets, recommended practices and evidence summaries
For the quality evaluation of these literatures, we traced the original literature on which these evidence were based and selected the corresponding evaluation standards of the Australian JBI Evidencebased Health Care Center (Tufanaru et al., 2020) according to the type of literature.

| Evidence summaries, classifications and recommendation levels
We screened the included studies individually, extracted the evidence based on the research question and then reviewed the evidence. We used the '2014 JBI Evidence Pre-grading and Evidence Recommendation Grade System' (Aromataris et al., 2015) to evaluate and grade the included evidence. According to the validity, feasibility, suitability and clinical significance of the evidence, the JBI evidence recommendation strength grading principle was combined to clarify the recommendation level of the evidence. The system divided the evidence level into five levels, and the recommendation opinions are divided into two recommendations (A and B).

| Included studies
The retrieval flow chart is shown in Figure 1. A total of 244 studies were retrieved and imported into the ENDNOTE v9.0 software.
After review, 58 duplicate studies were removed, and 127 studies were removed after reading the titles and abstracts and full texts were checked for the remaining 59 studies. After reading the full texts of these studies, we excluded studies that did not qualify on quality evaluation, studies that were inconsistent in research types and research objects and repeated publications. Finally, 13 studies were included, including seven expert consensuses, two recommended practices, one clinical decision, one best practice information booklet, one systematic review and one RCT. The general information of the included studies is shown in Table 1.

| Quality evaluation of expert consensus
Items were from the evaluation standard of the Australian JBI

| Quality evaluation of the systematic review
Items were from the evaluation standard of the Australian JBI Evidence-based Health Care Center (2016). We included one systematic review (Yu et al., 2021), for which item 8 (Were the methods used to combine studies appropriate?) was evaluated as 'Not applicable', and item 9 (Was the likelihood of publication bias assessed?) was evaluated as 'No'. The remaining items evaluated as 'Yes'.

| Quality evaluation of the RCT
Items were from the evaluation standard of the Australian JBI Evidence-based Health Care Center (2016). A randomized controlled study (Gasparino et al., 2021) was included, and for items 7 (Were treatment groups treated identically other than the intervention of interest?), 8 (Was follow up complete and if not, were differences between groups in terms of their follow up adequately described and analysed?), and 11 (Were outcomes measured in a reliable way?), the evaluation was 'Unclear'. For item 9 (Were participants analysed in the groups to which they were randomized?), the evaluation was 'No', and for the remaining items, the evaluation was 'Yes'.

| Quality evaluation of recommended practices, clinical decisions and best practice information booklets
Items were from the evaluation standard of the Australian JBI Evidence-based Health Care Center (2016). In this study, we included two recommended practices Padula et al., 2021), one clinical decision (Steven & Esther, 2021), and one best practice information booklet (Coyer et al., 2020), and upon tracing the original literature of evidence, we obtained one systematic review (Cai et al., 2019), one expert consensus (Edsberg et al., 2016), one RCT (Towfigh et al., 2008), one quasi-experimental study (Tomas et al., 2015) and one cohort study (Visscher et al., 2015). Regarding the quality evaluation results of the systematic review, item 9 (Was No full text (n = 10) Research object misfit (n = 9) Study design discrepancy (n = 27) Repeated publication in Chinese and English (n = 1) Duplicate publication (n = 1) Unqualified quality evaluation (n = 4)

| Synthesis of evidence
We summarized 31 pieces of evidence in five areas: skin cleaning and care, PPE placement and movement, reasonable use of dressings, treatment measures, and education and training. There were two pieces of Grade I evidence, 1 piece of Grade II evidence, 1 piece of Grade III evidence and 27 pieces of Grade V evidence. There were 16 strongly recommended items and 15 weakly recommended items. See Table 2  To evaluate the quality of the included studies and to ensure the scientific robustness of the evaluation results, we strictly followed the evaluation process and used the quality evaluation standards for various study types issued by the internationally recognized JBI Evidence-Based Health Care Center. The makers of this evidence comprise evidence-based methodology experts and medical staff with anti-epidemic experience. The evidence included kin cleaning and care, PPE placement and movement, reasonable use of dressings, treatment measures, education and training. In the process of formulating recommendations, we carefully considered the clinical experience and the feelings of medical staff, and good clinical applicability of the findings was accordingly ensured.  Clinical teams engaged in COVID-19 treatment continue to report facial skin tears and injuries due to prolonged use of protective masks (Rundle et al., 2020). Once the skin is damaged, it cannot serve as a natural barrier against infection. The area of skin damage can also become a channel for coronavirus and other bacterial, viral and fungal infections (Padula et al., 2021). Therefore, such personnel may staff. There are 31 pieces of best evidence. The best evidence extracted in this study was based on expert consensus issued by various societies, systematic reviews and randomized controlled trials, and the overall quality of the evidence in this study was relatively high.

| Prevention and management of facial pressure injuries in medical staff
The evidence derived from articles 1 to 14 emphasizes the importance of keeping the skin clean and moisturized. The use of a skin protectant is highly recommended. The following reasons may explain the high incidence of pressure injuries Jobanputra et al., 2021;Smart et al., 2020). First, the damage is directly attributed to the pressure and friction caused by PPE. Second, in medical work, medical staff need to compress the metal nose clip of the mask and tighten the elastic rope to ensure appropriate seal of the mask. Such materials with a small contact area and hard texture exert high pressure on local tissues. Third, medical staff wear PPE for a long time during high-intensity work; this leads to excessive sweating, which is not easy to resolve without removing the PPE.
This results in a wet skin surface, consequently resulting in pressure injuries. As medical staff have been under a lot of mental stress and workload during the COVID-19 pandemic, the skin sweats a lot, which increases the coefficient of friction (COF) between the facial skin and PPE (Gefen & Ousey, 2020a;Gefen & Ousey, 2020b).
Excessive moisture and friction cause local damage to the skin barrier function (Kleesz et al., 2012). A study showed that the skin response to respiratory protective equipment (RPE) was characterized by impaired skin barrier function. Excessive skin moisture will lead to sweating, and areas with increased sweating will have high transdermal water loss . The normal acidic pH of the stratum corneum plays an important role in the formation and maintenance of the permeability barrier and antibacterial defence . Therefore, keeping the skin clean and moist while using PPE can effectively prevent facial pressure injuries. It has been recommended that a pH-balanced and -insensitive skin cleanser should be used to clean the skin before and after wearing PPE. Scrubbing or massaging the skin with force should be avoided, and the skin should be kept clean and dry. The use of skin barrier protection products is recommended to protect the patient's skin and reduce the incidence of pressure injuries in medical staff.  (Padula et al., 2021), which increases the risk of infection and warrants immediate attention.
The evidence derived from article 31 illustrates the importance of providing education and training to medical staff. In fact, there are many sizes of N95 masks which vary with its manufacturer (Desai et al., 2020). PPE discomfort can increase skin damage on the face, ears and scalp and increase the exposure to COVID-19 (Gefen & Ousey, 2020a, 2020b. Therefore, it is extremely important for medical institutions to train medical staff on proper use of PPE for optimal prevention and management of pressure injuries.

| CON CLUS IONS
Due to the COVID-19 outbreak, facial pressure injuries in medical staff have become common. These injuries not only cause unpleasant emotional experiences but also increase the risk of infection.
Therefore, effective prevention and management of facial pressure injuries are extremely important. We systematically searched for articles on facial pressure injuries of medical workers and summarized 31 pieces of evidence into five aspects: skin cleaning and care, PPE placement and movement, reasonable use of dressings, treatment measures education and training. To ensure safety, clinical medical staff are requested to choose the method best suited to them to prevent facial pressure injuries.

| LI M ITATI O N S
Some limitations regarding the prevention and management of facial pressure injuries in medical staff need to be acknowledged. Thus far, no practical guidelines on facial pressure injuries in medical staff have been published, and there is a lack of high-quality randomized controlled studies. Therefore, our study is not included in the guidelines, and only one randomized controlled study is included.

ACK N OWLED G EM ENTS
We thank Medjaden Inc. for scientific editing of this manuscript.

FU N D I N G I N FO R M ATI O N
This study was supported by Fujian Science and Technology Planning Project (2020Y0080) and Military Biosafety Reach Special Project (20SWAQK48) in China.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.