key: cord-0769860-v4xlt9ek
authors: Braga, F.; Espinosa, G.; Monteiro, A.; Marinho, B.; Drummond, E.
title: Physiological Effects of Exercising at Different Intensities Wearing TNT or Double-layer Cotton Facemasks Compared to Not Wearing a Mask
date: 2020-12-17
journal: nan
DOI: 10.1101/2020.12.11.20203224
sha: c365cf163aa050e9a31da8b92ed578ba993fb9ab
doc_id: 769860
cord_uid: v4xlt9ek

We compared the physiological differences between exercising wearing a TNT or a double-layer-cotton (DLC) facemask (FM) and not wearing a mask (NM). Sixteen volunteers underwent 4 sets (S) of 2 sequential bouts (B). B1 and B2 corresponded to light and moderate intensity cycling, respectively. FMs were used as follows: S1: NM; S2: TNT or DLC; S3: DLC or TNT; and S4: NM. Metabolic, pulmonary, and perceptual variables were collected. The main results are expressed as effect sizes and confidence intervals (ES [95%CI]) for TNT and DLC unless otherwise indicated. Compared to NM, FM increased the duty cycle (B1=1.11[0.58-1.61] and 1.53[0.81-2.18]; B2=1.27[0.63-1.84] and 1.93[0.97-2.68]) and decreased breath frequency (B1=0.59[0.23-0.94] and 1.43[0.79-2.07], B2=0.39[0.05-0.71] and 1.33[0.71-1.94]). Only B1 tidal volume increased (0.33[0.09-0.56] and 0.62[0.18-1.05]) enough to avoid a ventilation reduction with TNT but not with DLC (B1=0.52[0.23-0.79]; B2=0.84[0.44-1.22]). Both FMs reduced oxygen saturation in B1 (0.56 [0.07-1.03] and 0.69 [0.09-1.28]) but only DLC did so in B2 (0.66 [0.11-1.13]). Both end tidal CO2 (B1=0.23[0.05-0.4] and 0.71[0.38-1.02]; B2=0.56[0.2-0.9] and 1.20[0.65-1.68]) and mixed-expired-CO2 (B1=0.74[0.38-1.08] 1.71[1.03-2.37], B2=0.94[0.45-1.38] and 1.78[0.97-2.42]) increased with FMs. Ventilatory adaptations imposed during FM exercising influenced blood-lung gas exchange. Larger ESs were seen with DLC. No adverse changes to human health were observed

The CPET protocol consisted of 2 minutes of rest and 3 minutes of unload pedalling followed by a ramp 112 phase until exhaustion. A fixed comfortable cadence between 65 and 85 rpm was requested, and when the 113 participant failed to sustain a minimum of 60 rpm for more than 5 seconds despite verbal encouragement, 114 the test was interrupted. A five-minute passive recovery period followed. 115

First and second ventilatory thresholds (VT1 and VT2, respectively) were estimated based on standard 116 methodologies previously described (Wasserman et al. 1994; Lucia et al. 2000) . Peak values for 117 physiological variables were defined as the highest 30-s average value. 118

The second visit occurred 1 to 7 days after the first visit. Volunteers performed 4 sets of exercise with a 119 10-minute rest between them. Each set consisted of two six-minute bouts (B) of constant work exercise 120 (on the same CPET cycle ergometer) using 80% of the workload at VT1 for B1 and 80% of VT2 for B2. 121 S1 and S4 were performed with NM. For S2 and S3, a 10 x10 cm snippet of TNT (removed from a 122 surgical FM) or DLC (provided by a local mask manufacturer) was attached in front of the flow sensor 123 using a rubber band to cover it completely (figure 2-A, 2-B, 2-D and 2-E). For the first included 124 volunteer, DLC and TNT were used on S2 and S3, respectively. Thereafter, to reduce the effect of 125 exercise time on physiological response, this sequence was counterbalanced for the next volunteers. In 126 order to blind the volunteers from the mask they were using, a cloth frame was placed on the border of the 127 flow sensor (figure 2-C). 128

Throughout all sets of exercise, VO 2 , VCO 2 , V T , B f , V E , H R , SpO2, and FMMT were continuously 129 measured using the same devices previously described. RPE and STP were obtained every minute during 130 repeated measures ANOVA to achieve a power of 80% and a level of significance of 5% (two sided), 142 with an effect size (ES) of 0.4 and a correlation among measures of 0.6. Therefore, 15 volunteers were 143 needed. 144 Two last-minute average values from each B were used for analysis for all volunteers but one, who had a 145 drop in ventilation and oxygen uptake after the fourth minute of B2 wearing the DLC mask. In this case, 146 values between the second and fourth minutes were considered. NM values were considered as the 147 average of S1 and S4. To confirm the measure's stability regardless of time, we calculated the intraclass 148 correlation coefficient (ICC) and its 95% confidence interval (CI) for S1 and S4 based on the absolute 149 agreement of single measures in a two-way mixed-effects model between conditions (Table S1) 

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2020. ; https://doi.org/10.1101/2020.12.11.20203224 doi: medRxiv preprint Figure 3 shows the boxplot results for each variable. A detailed description follows below. 173 174 Oxygen uptake and carbon dioxide output, respiratory exchange rate and heart rate. 175 VO 2 was significantly different between FMs during B1 and B2 (figure 3-a), increasing 8.7 and 10.4% 176 during B1 and 6.8 and 6.3% during B2 for TNT and DLC, respectively (SES for all). No difference was 177 found between TNT and DLC during either bout. VCO 2 was also different between FMs (figure 3-b). 178

Compared to NM, VCO 2 increased 8.7 (SES) and 7.3% (SES) during B1 and 4.8 (VSES) and 1.9% (non-179 significant ES) during B2 for TNT and DLC, respectively. RER (figure 3-c) was significantly different 180 between FMs but in the opposite direction observed for VO 2 and VCO 2 , showing a non-proportional 181 change between these variables. RER was 3.1 (LES) and 5.7% (VLES) lower during B1 and 1.9 (MES) 182 and 4.3% (VLES) lower during B2 for TNT and DLC, respectively, than for NM. TNT RER was 2.7% 183 (MES) lower than DLC during B1, and no difference was found between them during B2. reductions in B f compared to NM and TNT, respectively. The net effect on V E (a product of B f and V T -195 figure 3-g) also showed significant differences between FM and NM exercise. However, V E was not 196 . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2020. ; https://doi.org/10.1101/2020.12.11.20203224 doi: medRxiv preprint T i /T TOT (figure 3-h) was different among FMs during both B1 and B2. Post hoc analysis shows 6.1 (LES) 199 and 10.4% (VLES) lower values of T i /T TOT for NM than for TNT and DLC, and no differences between 200 TNT and DLC during B1. During B2, T i /T TOT with NM was 5.1 and 9.8% (VLES for both) lower than 201 that with TNT and DLC, respectively. This time, the TNT value was 4.5% (MES) lower than that of the 202 DLC. SpO 2 (figure 3-i) was different between FMs during B1 and B2, but post hoc analysis was unable to 203 show significant differences within FM comparisons. However, ES showed significant differences for 204 pairwise analysis. DLC and TNT showed an MES reduction in SpO 2 compared to NM during B1. An 205 MES was seen during B2 for DLC relative to TNT and NM. All other comparisons were not significant. and B2, respectively, compared to NM. DLC also increased RPE over TNT during B1: 7.1% (MES). 223 ANOVA showed significant differences between FMs during B1 but not B2 for STP (figure 3-n). 224

However, no significant pairwise ES was observed during B1 or B2. 225 . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2020. ; https://doi.org/10.1101/2020.12.11.20203224 doi: medRxiv preprint for NM vs. TNT, TNT vs. DLC and NM vs. DLC, respectively. 231

The aim of this study was to investigate the effect of wearing TNT and DLC FMs during exercise at light 233 and moderate intensities on physiological responses in order to identify potential reasons for reported 234 malaise and reduced exercise performance, as well as any reason for concern regarding harmful effects on 235 gas exchange. To our knowledge, this is the first study considering different individualized exercise 236 intensities. 237

With regard to metabolic variables, both VO 2 and VCO 2 exhibited small increases with FM at the two 238 tested exercise intensities. The effects on VO 2 were larger than those on VCO 2 , as illustrated by a 239 reduction in RER. The absence of excess CO 2 indicates an increase mainly in aerobic rather than 240 anaerobic muscle metabolism. DLC produced a MES negative effect on RER that was twofold higher 241 than the effect of TNT compared to NM. What is the reason for this increase in aerobic metabolism? As 242 this phenomenon could be observed at higher exercise intensity (80% of VT2), it is unlikely that an 243 increase in energy demand on exercising muscles near the ventilatory compensation point did not 244 generate an excess of CO 2 . Therefore, we believe that another group of working muscles should have 245 accounted for the higher VO 2 fostered by FMs. were larger. Wider breathing amplitude to increase V T was another mechanism adopted to maintain V E ; 254 however, this response was observed only at lower intensity and not at higher intensity. This happened 255 . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2020. ; https://doi.org/10.1101/2020.12.11.20203224 doi: medRxiv preprint offset B f reduction. Thus, the net effect was a V E reduction for DLC at both intensities over NM and TNT. 257

The TNT FM flow restriction was not enough to reduce V E despite its impact on the frequency and time 258 components of ventilation. 259

There is a negative relationship between the duty cycle (T i /T TOT ) and diaphragmatic blood flow (Buchler 260 et al. 1985 ) and probably also between muscle performance and oxygen uptake. As muscle activation was 261 not measured, we hypothesized that efforts to keep V E should be attributed to accessory respiratory 262 muscles, which can also be an explanation for the increased VO 2 observed with FM. Another possible 263 contribution is active expiration as a result of the shorter expiratory time. 264

General concerns regarding the effects of FM use during exercise rely on their effect on gas exchange. In 265 this research, FM promoted some reduction over NM on SpO 2 at lower exercise intensity, but only DLC 266 maintained this effect at a higher intensity. Reduced V E is the main reason for dropping SpO 2 . However, 267 no clinically significant value has been seen (SpO 2 <90%). 268

Both FMs raised ETCO 2 and PECO 2 values. The effect on ETCO 2 increased with exercise intensity and 269 was more pronounced with DLC than with TNT. DLC also induced higher values of PECO 2 compared 270 with TNT, but the effects of both mask types were remarkably similar at both exercise intensities. 271

Moreover, the effect sizes observed for PECO 2 were larger than those for ETCO 2 . As a result, Difference between E T CO 2 and PECO 2 349 CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2020. ; https://doi.org/10.1101/2020.12.11.20203224 doi: medRxiv preprint . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted December 17, 2020. ; https://doi.org/10.1101/2020.12.11.20203224 doi: medRxiv preprint

Exercise with facemask; Are we handling a devil's sword?" -A 376 physiological hypothesis

The role of community-wide wearing of face mask 379 for control of coronavirus disease 2019 (COVID-19) epidemic due to SARS-CoV-2

Facemasks and hand hygiene to prevent influenza 382 transmission in households: A cluster randomized trial

Effect of inspiratory resistive loading on control of 385 ventilation during progressive exercise

387 predict nocturnal hypoventilation

Airborne transmission route of covid-19: Why 2 meters/6 438 feet of inter-personal distance could not be enough

Respiratory pump maintains cardiac stroke volume during 440 hypovolemia in young, healthy volunteers

Ventilatory efficiency during exercise in healthy subjects

Mask use during COVID-19: A risk adjusted strategy