Study time and participants
The study was performed in February and March 2021 in the Department of Anesthesiology and Intensive Therapy of our university.
The study included a group of 21 volunteers (including 16 females (76%) and five males), with a median age of 23 years, with interquartile ranges (IQRs) of the participants (21–27), range of 20–59.
The study group consisted of intensive care registered nurses (RN, MSc or RN, BSc) (n = 3), nursing students (of second and third years of study) (n = 14), medical students (fifth year) (n = 3), and a senior consultant (MD, PhD) of the department. All of the students participating in the experiment were students in our university, and all were Caucasian.
All participants had valid medical examinations and were allowed to work by the recognized occupational doctor, and all were vaccinated twice with the COVID-19 vaccine (Pfizer/BioNTech). Persons with a history of pulmonary disease, anemia, vascular diseases, and other diseases that could limit respiratory function were excluded from the study.
Research Design
To evaluate the effects of wearing the respirators, we used a counterbalanced crossover design—a self-controlled trial.
We wanted to rule out the possibility of natural changes in the pulse oximetry (SpO2) and heart rate parameters during the normal functioning of the participants. Therefore, each subject served as his own control and performed the test two times: (1) without a face mask and PPE (control) for three hours and (2) wearing an FFP3 respirator with full PPE for a three-hour shift in the intensive care unit. The first measurements ("time point 0") of the SpO2 and pulse rate were recorded 2 minutes after fitting the FFP3/PPE to give participants time to adapt themselves.
The study was performed during the third peak of the COVID-19 pandemic. Although the department was not an isolation ward dedicated only to COVID-19 patients, all emergency admissions were made before obtaining the test for a SARS-CoV-2 infection. Therefore, the applicable unit protocol required the team to wear the PPE and masks.
During the test runs without masks and PPE, we encouraged the participants to perform each hour's same exercises for at least 10 minutes with the supervision of the study coordinator (IW).
Among the study participants, we had students; therefore, for safety and controlling reasons, we used the Nellcor PM10N monitoring system, which allows the study coordinator to assess the recorded parameters in real time (Fig. 1).
To avoid bias, each subject's sequence of interventions (test run or shift run) were randomly assigned. Strenuous physical activity was prohibited during the 24 hours preceding the test, and a night sleep of at least 6 hours was mandated [12]. We also asked the participants not to use any fingernail polish.
Equipment
Each volunteer was wearing the same type of respirators and PPE.
We used FFP3 masks – (Oxy-line respirators: X310 SV FFP3 RD, CE1437. EN149:2001 + A1:2009, PPE, made in Poland) with gowns (Disposal Medical Integrated Protective Coverall. RAEX. Hunan Xunzhuo Industry, made in China), and goggles (made in Poland).
Blood saturation and HR were measured in real time for the duration of the whole testing with a pulse oximeter (Nellcor - Portable SpO2 Patient Monitoring System. PM10N, Coviden. Mansfield USA, made in Korea) (Fig. 1).
The single-use SpO2 sensor (Nellcor. Neonatal-Adult SpO2 Sensor. MAXIN. Coviden. Mansfield USA, made in Mexico) was kept under the surgical glove on the fifth finger of the nondominant hand and connected under the gown with a PM10N device.
Additionally, every 30 minutes during the three-hour shift and during the control run, each subject completed a questionnaire concerning their well-being, with a score scale (ranging from 0 (no symptoms) to 6 points (unable to work)) for headache, shortness of breath, perspiration, fatigue, and thirst (Table 1).
Statistical methods
To analyze the results of the SpO2 and the HR in the HCWs wearing the FFP3/PPE, the patients were grouped according to sex and age, and the median values along with the rest of the quartiles were reported (Table 1). All pairwise comparisons were made using the Mann-Whitney U-test [13]. Due to the number of all the measurements over time, we expected statistically significant differences between the medians. To measure the differences between the groups, we calculated the 95% confidence intervals for the differences of the medians.
Table 2 and Table 3 present the results of the linear mixed-effects model (LMM) estimating the influence of working with the FFP3/PPE in the HCWs, the demographic factors and the levels of SpO2 [%]. LMM is a generalization of the standard linear model used, as the measurements are permitted to exhibit correlation and nonconstant variability between the individuals. LMM, therefore, provides the flexibility of modeling not only with the means but also with their variances and covariances. Moreover, LMM is recognized as a type of method that has some crucial advantages over repeated-measures ANOVA [14].
Table 4 presents a comparison of the different well-being markers between the 30-minute time points while working in the FFP3/PPE versus not working in the FFP3/PPE. In such cases, Friedman's one-way repeated measures analysis of variance by ranks (Friedman test) [15] was used to estimate the differences between consecutive groups (i.e., experiencing various ailments after 30, 60, 90, 120, 150, and 180 minutes).
Statistical analysis was performed using RStudio (Version 1.3.1073).
Ethics approval
The study was conducted according to the Helsinki Declaration (Seventh revision) and Good Clinical Practice Guidelines [16]. The study protocol was accepted by the Bioethics Committee of Poznań University of Medical Sciences, Poznań, Poland (Ref. no. : 123/21 on 04th Feb. 2021) and each participant gave informed consent. Participants were informed of the anonymous data extraction and analysis from the study files.