Cardiopulmonary Exercise and the Risk of Aerosol Generation While Wearing a Surgical Mask.

To the Editor:

Cardiopulmonary exercise training is associated with improvements in health-related quality-of-life, physical function, and overall mortality rates in patients with cardiovascular or pulmonary disease. , Currently, many structured cardiopulmonary exercise training programs are running at limited operational capacity or are closed due to an increased concern of possible spread of severe acute respiratory syndrome coronavirus 2 through droplet and aerosol generation that could occur with exercise. , In the community, many gyms are reopening, given the lack of evidence of potential aerosol generation during exercise. This lack of evidence is a concern because of potential coronavirus disease 2 transmission from asymptomatic or presymptomatic individuals in these settings. Presently, we sought to quantify and characterize particle generation associated with healthy adults performing exercise of different intensities.

Methods

This was a single-center study conducted at Mayo Clinic Florida and was approved by the institutional review board. Four adult volunteers free of cardiorespiratory, musculoskeletal, and metabolic disease consented and performed staged exercise.

Testing was conducted in the cardiopulmonary gym (473.2 m3; 6.3 room air exchanges/h). With the use of a light-scattering particle counter (FLUKE 985), particle counts and size distribution (0.3, 0.5, 1, 2, 5, and 10 μm) were measured at different locations in the gym before, during, and after staged, incremental-intensity exercise. This type of particle counter is endorsed by the International Organization for Standardization. Aerosol and droplets were defined as a particle size <5 μm and 5 to 10 μm, respectively.

Ambient measurements were obtained (with treadmills running) at the center of a square formed by the equipment. This central point was 1.8 m from each exerciser. The volunteers simultaneously performed 30 min of treadmill (Quinton MedTrack CR60; n = 2) or recumbent cross-trainer (NuStep T5XR; n = 2) exercise while wearing type 2 procedural masks (Halyard), with a bacterial filtration efficiency at 3 μm of 98% and particulate filtration efficiency at 0.1 μm of 98%. The exercise consisted of 3- × 10-min stages: (1) light, (2) hard, and (3) very hard, based on the Borg 6 to 20 rating of perceived exertion scale (RPE). Air samples were obtained at 0.3, 0.9, and 1.8 m away from each volunteer, twice per stage. After exercise, the ambient room was measured in the same location as before exercise until it returned to baseline.

A nonparametric distribution of measurements was assumed. The Wilcoxon/Kruskal-Wallis test was used for multiple group comparisons and, if that was significant, a Steel test was performed to compare each group to a single control that was the ambient measurement. Data were displayed as a mean with SDs. A probability value <.05 was considered significant.

Results

Age, BMI, and resting heart rate for the volunteers were 36.3 ± 6.5 y, 24.0 ± 3.1 kg/m2, and 67 ± 9 beats per minute (bpm). For stage 1 (light), heart rate increased to 113 ± 18 bpm (61 ± 11% of predicted maximum [PM]) and RPE was 9.5 ± 1.3. For exercise stage 2 (hard), heart rate increased to 144 ± 15 bpm (79 ± 9% PM), and RPE was 13.0 ± 0. For the final stage (very hard), heart rate was 168 ± 15 bpm (91 ± 8% PM), and RPE was 18.0 ± 1.2.

Table 1 shows the particle counts per liter of sampled air with each stage of exercise. Small respirable particles in the 0.3 to 0.5 μm range were generated the most with stage 3 (very hard exercise). In stage 2 (hard exercise), there was a nonsignificant increase in 0.3 μm size particles at 1.8 m. Proximity to the exerciser was associated with increased particle detection during stage 3 (very hard exercise), because the counts for all sizes at 0.3 m nearly doubled the counts at 1.8 m. Larger droplet-sized particles were not increased significantly during any stage of exercise. After exercise, the half-life for room clearance of 0.3 μm was 15 min and for 0.5 μm was 16 min.

Table 1

Particle Quantity Per Liter of Sampled Air by Size for Each Stage of Exercise

Sample	Particle Quantity Per Liter of Sampled Air by Size
	0.3 μm	0.5 μm	1 μm	2 μm	5 μm	10 μm
Ambient	478.4 ± 67.7	181.6 ± 63.5	128.1 ± 50.4	82.3 ± 32.8	31.4 ± 13.4	16.7 ± 7.4
Stage 1 (light), m
0.3	454.6 ± 31.1 (1.0)	118.3 ± 16.0 (.28)	71 ± 10.7 (.13)	42.3 ± 6.3 (.13)	16.4 ± 4.0 (.10)	8.4 ± 2.6 (.27)
0.9	486.9 ± 35.4 (1.0)	136.3 ± 29.8 (.59)	83.4 ± 20.2 (.28)	53.75 ± 16.4 (.42)	23.5 ± 9.4 (.84)	12.5 ± 4.2 (.88)
1.8	477.1 ± 51.6 (1.0)	142.5 ± 43.2 (.93)	89.3 ± 28.3 (.50)	58.8 ± 19.6 (.72)	25.9 ± 10.3 (.99)	15 ± 7.6 (1.0)
Stage 2 (hard), m
0.3	703.9 ± 213.1 (.42)	251.3 ± 120.4 (.84)	117.6 ± 53.0 (1.0)	69.4 ± 30.8 (1.0)	26.9 ± 9.0 (1.0)	15.9 ± 5.1 (1.0)
0.9	702.8 ± 187.7 (.10)	247.5 ± 91.4 (.77)	124.6 ± 37.8 (1.0)	71.8 ± 21.9 (1.0)	29.3 ± 10.9 (1.0)	15.5 ± 6.1 (1.0)
1.8	649.3 ± 165.4 (.06)	235.4 ± 100.1 (.97)	136.9 ± 53.3 (1.0)	89.5 ± 35.1 (1.0)	40.1 ± 18.8 (.99)	25.9 ± 12.3 (.84)
Stage 3 (very hard), m
0.3	3,041.1 ± 624.6 (.01)	884 ± 584.7 (.04)	304.4 ± 203.8 (.34)	185 ± 114.9 (.50)	34.3 ± 14.2 (1.0)	11.5 ± 3.2 (.67)
0.9	1,522.8 ± 161.7 (.01)	376.6 ± 136.3 (.03)	153.4 ± 47.4 (.95)	86 ± 18.7 (1.0)	29.5 ± 7.0 (1.0)	17.5 ± 4.6 (1.0)
1.8	1,029.3 ± 47.8 (.01)	352.5 ± 47.2 (.01)	152.5±18.2 (.59)	85.5 ± 13.9 (.99)	34.3 ± 8.3 (1.0)	18 ± 8.2 (1.0)
After exercise
Immediate	1,174.8 ± 208.8 (.08)	477 ± 140.5 (.08)	225.3 ± 110.0 (.41)	108 ± 20.8 (1.0)	47 ± 30.3 (1.0)	26.5 ± 17.4 (1.0)
30 min	266.7 ± 25.0 (<.01)	29.8 ± 11.8 (<.01)	15.3 ± 8.1 (<.01)	9.6 ± 5.2 (<.01)	4 ± 2.5 (<.01)	2.4 ± 1.9 (<.01)
1 H	288.2 ± 14.5 (<.01)	29.75 ± 6.1 (<.01)	15.3 ± 4.3 (<.01)	9.7 ± 3.4 (<.01)	3.9 ± 2.9 (<.01)	2.5 ± 2.5 (<.01)

Data shown as mean ± SD (P value). Data were analyzed with a nonparametric Wilcoxon/Kruskal-Wallis test; if significant, a Steel test with the ambient as the control was used.

Discussion

This study suggests that, when wearing a type II procedural mask, light exercise is not associated with an increase in particulate generation, but there is a trend to increasing particles with hard exercise. Very hard exercise, similar to what is performed during clinical treadmill stress tests and many workouts performed at gyms, significantly increased particle generation. This is the first study to quantify this particle generation with exercise. The exercise component of most cardiopulmonary rehabilitation classes is generally performed at a light-to-hard exertion, during which particle generation is negligible. If strenuous exertion is to be performed, then proper room clearance measures would need to be in place before someone else enters the room or should not be performed at all. The only guidance for room clearance measures is from a Dutch building code, which recommends sporting facilities to have a minimum of 6 air exchanges/h. This type of symptom-limited exercise potentially could increase the transmission of airborne infectious agents.

This study provides valuable information for guiding infection control measures for health-care workers and participants during cardiopulmonary rehabilitation, interval training, and symptom-limited exercise testing. Because the volunteers were healthy, the amount of particle generation should be considered the minimum because patients with cough may have potential for increased aerosol generation. Patients with advanced cardiac and pulmonary disease theoretically could also have decreased particle generation because of decreased exercise tolerance with subsequent decreased minute ventilation. The composition or infectious potential of these particles was not assessed.

In conclusion, light-to-hard exercise while wearing a type 2 procedural mask does not generate any additional particles above a baseline room measurement. Very hard exercise generates a significant increase in particles and warrants appropriate personal protective equipment to mitigate possible infectious risk.