Lung Function Decline and Other Physiological Changes in Health Workers after Working in COVID-19 Isolation Wards: An Observational Study

Background: Global healthcare systems have been under huge pressure since Coronavirus Disease 2019 (COVID-19) was declared a pandemic. It is critical to prevent further spread of COVID-19 and to protect health workers from infection. This study aims to gure out short-term physiological impact on health workers induced by working in isolation wards for hours with personal protective equipment (PPE), so as to provide insights on reducing physiological impact and infection risk of health workers.

to protect health workers against COVID-19. Although the transmitting way of SARS-CoV-2 is not fully understood yet, contact, respiratory droplet and fomites have been proposed by some scholars. [8,9] Thus, personal protective equipment (PPE) including N95 ltering facepiece respirator (FFR) or medical mask, gown, goggle and gloves are highly recommended for health workers who work in isolation wards. [4] However, overloading working schedule and side effect of PPE such as skin damage also exerted negative physiological impact on health workers. [10] The exhalation and inhalation resistance of N95 FFR and high-intensity work might contribute to changes of lung function, consequently decreasing working e ciency. [11] Keeping physical function in good status when working in isolation wards might play an important role in preventing health workers from infection. To our best knowledge, there is no similar study concerning the changes of lung function, oxygen saturation, heart rate, and weight of health workers after working in isolation wards. This research aims to show short-term physiological impact of PPE on health workers working under the COVID-19 pandemic and provide reference on formulating work shifting system scienti cally.

Methods
This observational study was conducted from February to March 2020 in Wuhan, China. The inclusion criteria were as follows: 1) licensed doctors and nurses who have received intense training to manage COVID-19 patients; 2) working in isolation wards with PPE over four hours; 3) not smoking; 4) not pregnant; 5) without any chronic respiratory diseases history including chronic obstructive pulmonary disease (COPD) and asthma; 6) showing no respiratory tract infection symptom within last two weeks. All subjects involved in this study had worked in isolation wards for at least ve weeks. The information of any subject infected by COVID-19 within 28 days after they ended working in isolation wards were recorded. All subjects were informed about the procedure and details of this study and signed consents. This study was approved by the Institutional Review Board (IRB) of authors' hospital.
Measurement was conducted under the supervision of a pulmonologist. Lung function parameters including forced vital capacity (FVC) % predicted, forced expiratory volume in one second (FEV 1 ) % predicted, ratio of FEV 1 to FVC (FEV 1 /FVC), peak expiratory ow (PEF) % predicted and mean forced expiratory ow, mid-expiratory phrase (FEF  ) % predicted were measured and calculated by Ubreath Spirometer System PF680 (e-LinkCare Meditech Co.,Ltd., Hangzhou, China). Blood oxygen saturation measured via pulse oximetry (SpO 2 ) was obtained by Pulse Oximeter CMS50DL (Heal Force Bio-meditech Holdings Group, Shanghai, China). The information of all subjects including sex, age, height, baseline weight was collected and body mass index (BMI) was calculated. Before these health workers entered and after they left isolation wards, their heart rate, SpO 2 , weight, and lung function were measured in less than ve minutes without wearing PPE.
When working in isolation wards, all these health workers were equipped with PPE consisting of protective coveralls (AAMI PB70 level 4) covered by surgical gowns, N95 FFR (In line with the standard of GB/T32610-2016 KN95; NIOSH N95; FFP2) covered by surgical mask (In line with the standard of GB/T32610-2016 KN90; FFP1),, protective goggles, gloves (double layers), long shoe covers (double layers), and surgical cap. Their work mainly contained managing patients with COVID-19, establishing venous access, sputum induction, ventilators management, nursing, etc. Working durations in isolation wards of all subjects were over four hours. The actual working durations could vary, but were recorded and analyzed. Actual working time for four to six hours was de ned as short working-duration while more than six hours but less than nine hours was de ned as long working-duration.
All the statistics were analyzed using Stata/SE 12.0 (College station, TX) and GraphPad Prism 8 (San Diego, CA). Normally distributed parameters were analyzed using Student's t test and simple linear regression analysis. Non-normally distributed parameters were normalized before analyzing. P < 0.05 was regarded signi cant for all analyses. Normally distributed parameters displayed in this study were as mean ± standard deviation and non-normally distributed parameters were shown as median (25 th percentile, 75 th percentile).

Results
There were 36 healthy health workers, 34 females and 2 males in one isolation ward involved in this study. The demographic data were shown in predicted were also lower than before-working means, but no signi cant difference was observed (P = 0.13, P = 0.85, P = 0.23, P = 0.25, respectively). Comparison was shown in Table 2. The before-working weights of two health workers (one male and one female) were measured in incorrect method (wearing PPE) and were consequently excluded from weight change analysis. After working in isolation wards, the weights of other health workers (n = 34) signi cantly decreased (54.79 (48.50, 61.00) versus 55.50 (48.90, 61.00) P<0.01). The decreasing level was 0.43±0.71 kg.  *Parameters of two health workers were measured in incorrect method and were excluded. Therefore, n = 34 for this characteristic.
A signi cant negative correlation between BMI and the decline levels of FEV 1 /FVC (β = 2.87, P = 0.02) was observed through linear regression (shown in Figure 1). The correlation between BMI and decline levels of FEF 25-75 % predicted (β = 1.80, P<0.05) also showed signi cance (shown in Figure 2). However, there was no signi cant correlation between age and changing levels of any measured parameter.  Table 3). None of these recruited health workers were con rmed infected by COVID-19 within 28 days after they ended working in isolation wards. *Parameters of two health workers were measured in incorrect method and were excluded. Therefore, n = 20 for these characteristics.

Discussion
This is the rst study focusing on short-term physiological changes (especially lung function) in health workers who worked in COVID-19 isolation wards for over four hours with PPE including N95 FFR, trying to provide reference on formulating scienti c working shift system to protect health workers. SARS-CoV-2 is a new pathogen which is extremely contagious. There were more than 3,000 health workers in China and at least 9,000 in US reported infected. [12,13] More than 40,000 health workers were infected in Europe according to local media. The importance of protecting health worker from high infection risk has been emphasized by many authors. [14,15] Since droplet mode is one of the main spreading ways, respiratory protection such as medical mask and respirator plays a role in protecting health workers. However, respiratory protection may also bring discomfort. [16] Spirometry test is widely applied in evaluating patients' respiratory function. FVC, FEV 1 , and FEV 1 /FVC are indicators of obstructive ventilation disorder, [17] usually seen decreased among patients with COPD and bronchial asthma. PEF re ects the function of respiratory muscle and airway resistance while FEF  acts as an indicator of small airway function. [18,19] All these parameters can re ect lung function objectively. The result of this study showed signi cant short-term increase of heart rate and decline of FVC % predicted, which indicated that continuously working for hours in isolation wards was not an easy task. Slight obstructive respiratory was observed. Exhalation and inhalation resistance from respiratory protection and fatigue of respiratory muscles after long-duration working might be the contributing factors. Roberge et al conducted a trial on physiological impact of N95 FFR on health workers working for an hour (average wearing time when there was no pandemic), nding no signi cant difference. [11] However, N95 FFR was not the only equipment in uencing lung function in our study while the second layer of surgical mask as well as protective coveralls may also exert impact on lung function. Moreover, health workers in our study had a much longer working duration (averaging 6.5 hours) and higher working intensity, which was just the actual condition under COVID-19 pandemic. These could lead to more signi cant short-term lung function decline and increase of heart rate. Mental stress from COVID-19 outbreak and facing heavily illness quarantined patients might also facilitate the changes. [20] Currently available evidence showed negative correlations between lung volume (especially functional residual capacity and expiratory reserve volume) and BMI among both normal and overweight people. [21] Therefore, this study assessed the association between changing levels of respiratory functions and BMI.
Decline of FEV 1 /FVC enlarged as BMI increased among subjects were observed in the result. Similar condition was also recorded between changing levels of FEF 25-75 % predicted and BMI. These nding indicated that health workers with higher BMI tended to suffer larger decline of lung function. A possible explanation by Littleton et al was that some airways start to closed as the BMI increase. [22] But this explanation was mainly based on lung function decline among obese people while the mechanism in normal or mild-overweight people remains further research. Considering all subjects in this study generally have a lower BMI, the decline of lung function among health workers with higher BMI might be more signi cant.
Other lung function parameters measured in this study such as FEV 1 % predicted, FEV 1 /FVC, PEF % predicted, FEF 25-75 % predicted all showed lower means but the differences were not signi cant. It could be in uenced by sample size while a larger sample might bring differences. There was no signi cant difference in SpO 2 between before-working and after-working results, which indicated that changes of oxygen diffusion function induced by working in isolation wards with PPE less than nine hours could be compensated or the oxygen diffusion function could not be in uenced. However, for those who work in isolation wards repeatedly, whether the acute decline of lung function would exert long-term impact is still unknown. The possibility of conversion from acute to chronic decline should not be ignored. Long-term decline of lung function could be associated with many diseases such as cardiovascular events, restrictive respiratory disorder, COPD, and lead to chronic hypoxemia. [23] The weight loss of health workers after working in COVID-19 isolation wards was the consequence of dehydration because they could not drink water or urinate during working. The PPE including double layers of protective clothing (protective coveralls covered by surgical gown), double layers of respiratory protection (N95 FFR covered by surgical mask) and double layers of shoe covers would quickly made health workers sweaty and tired. Severe dehydration could lead to various health events such as falls, delirium, renal failure and even deaths. [24] Respiratory infection is also strongly associated with dehydrations. [24] In this study, no health workers showed any symptom of dehydration except for fatigue.
No infection of COVID-19 was reported afterwards. Hence, the short-term dehydration induced by working in isolation wards could be compensated but long-term impact on those who worked in isolation wards repeatedly is still unknown.
There was no signi cant difference between age and changing levels of each parameter. It was deduced that young health workers (ranging from 22 to 39) involved in this study show same impact on physiological function. It was hypothesized that longer working duration would induce more signi cant changes both in lung function and heart rate. Huang et al recommended a working shift system of continuously working for six hours because working for longer duration made workers feel exhausted and even dizzy. [12] Thus, health workers involved in this study were divided into two groups according to different working duration (less than or over six hours) in isolation wards. No signi cant difference was observed between these two groups. Therefore, continuous working in isolation wards for six to nine hours would not exacerbate the short-term lung function decline or exert more severe physiological impact comparing to working for four to six hours.
Considering the lung function decline and other physiological impact on health workers and the valid using duration of PPE, the balance between protecting health workers and helping more patients should be kept. First, the application scope of respiratory protection should be carefully evaluated. N95 FFR is recommended consistently in aerosol generating procedure such as endotracheal intubation and collecting respiratory specimens while it is still con icting whether N95 FFR is necessary in non-aerosol generating procedure. [25][26][27] Surgical mask is recommended to be worn all times in healthcare facilities. [25] Second, the constant working duration of PPE required limitation. There is no consistent recommendation for constant working duration with PPE in isolation wards. The valid using duration of N95 FFR used to be recommended for eight hours by Centers for Disease Control and Prevention of United States while WHO suggested that using a respirator for more than four hours should be avoided.
[27, 28] Working with medical mask or N95 FFR constantly could increase infection risk and induce discomfort such as fascial heat and pain. [16] In this study, no health worker was con rmed infection afterwards, which was a testament of the e cacy of protection when working in isolation wards within nine hours. But six-hours duration might be recommended considering working fatigue.
There are limitations in this study. The imbalance of sex proportion could contribute bias, although it was literally independent when recruiting health workers. There might be more persuasive result if more male health workers could be recruited, because sex distribution was reported in association between lung function and obesity. [22] Age distribution is another disadvantage. Although age in this study was normally distributed, there was no health workers older than 40 years old recruited. The impact of working with PPE in isolation wards within nine hours on elder health workers were still unknown. Moreover, because of the maximum working stress of COVID-19 and the shortage of rest, only 36 health workers in one isolation ward could spent their precious resting time to complete the lung function test in authors' institution. Large sample study is literally better for more convincing result, but prospective and empty controlled group was not feasible considering the infection risk. After working in COVID-19 isolation wards for certain period such as two or more months, signi cant decline of lung function might be observed. Mid-term and long-term follow up were absence and some potential impact might be consequently ignored.

Conclusions
This study revealed the short-term lung function declined of health workers after working in COVID-19 isolation wards with PPE for four to nine hours. The decline levels were associated with BMI. Heart rate increase and weight loss were reported. No harmful impact on young health workers under 40 years old was found in this short-term observational study. Six-hour working duration could be recommended considering the physiological impact on health workers and the valid using duration of PPE when formulating working shift system. In order to reveal further impact of working in COVID-19 isolation wards on health workers, long-term follow up is required.

Declarations
Ethic approval and consent to participate The Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology (IORG No: IORG0003571) gave a nal APPROVAL for this study. All subjects signed written informed consent for participate.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. datasupplement.csv