We recruited 25 people with CF and chronic P. aeruginosa infection (6) from the Adult Cystic Fibrosis Centre, The Prince Charles Hospital, Brisbane, Australia. Ten healthy volunteers were recruited from hospital and research staff to assess mask comfort and mask weight change. All participants performed up to five randomly ordered tests in a validated cough system (7): uncovered cough, coughing with surgical mask worn for 10 minutes, coughing with surgical mask worn for 20 minutes, coughing with surgical mask worn for 40 minutes, and coughing with N95 mask worn for 20 minutes (3, 7). The N95 test was an optional test based on the poor comfort ratings observed in our earlier mask study (3).

The duration of the mask wear tests were selected on the basis of observation of patients moving around communal areas of the hospital described here. Two types of masks were tested: surgical mask (catalog # 47107; Halyard FLUIDSHIELD Level 3 Fog-Free Procedure Mask [∆P < 2.5]) and N95 mask (catalogue #46,827 [small] or 46,727 [regular], Halyard FLUIDSHIELD N95 Particulate Filter Respirator and Surgical Mask). New masks were used for each test. The total wear time of the masks included 1-minute positioning of the participant into the rig, 2 minutes of tidal breathing with air filtered by a high-efficiency particulate air filter, and a 5-minute cough period, followed by another 2 minutes of tidal breathing. Cough aerosol collection, sputum processing, and P. aeruginosa genotyping were performed as previously described (1, 3, 8). All participants rated their comfort levels after each test (3, 9). All masks were weighed before and after each test.

SPSS version 25 was used for statistical analysis. Participants with CF were stratified by the amount of aerosol colony-forming units (CFU) produced during the uncovered cough test: high producer (total count ≥ 10 CFU) or no/low producer (total count < 10 CFU) (3). Categorical variables were examined using Pearson chi-squared test or Fisher’s exact test. Continuous variables were examined using a Student’s t test or Mann-Whitney U test. CFU counts were log transformed, and the paired t test examined changes over time. The McNemar-Bowker test was used to examine comfort scores over time. The Wilcoxon signed-rank test was used to examine the change in mask weight over time.

P. aeruginosa was cultured from the sputum of 25/25 participants with CF and was cultured in cough aerosols during the uncovered cough test of 20/25 participants (Table 1). P. aeruginosa was cultured from cough aerosols of 9/20 participants during any of the surgical mask tests (10-min, 20-min, and 40-min total wear times) and 4/20 participants during the N95 mask test of 20 minutes of total wear. The CFU counts were significantly reduced for the surgical mask tests compared with the uncovered cough test (P < 0.001). Between mask tests, the CFU count remained similar as the duration of surgical mask wear increased, as well as between mask types (Table 1). The P. aeruginosa strain types found in the cough aerosols were genetically indistinguishable from the paired sputum sample of each participant.

Participants with CF rated surgical masks less comfortable than healthy volunteers for all test durations (surgical mask: 10 min, P = 0.001; 20 min, P = 0.007; 40 min, P = 0.023; N95: 20 min, P = 0.018; Table 2). Participants with CF were more tolerant of surgical mask wear (good comfort) after 10 and 20 minutes of total wear time if they had higher lung function, yet this difference was lost after 40 minutes of surgical mask wear. N95 masks were rated less comfortable in participants with CF and healthy volunteers (Table 2).

The change in mask weight for each test ranged from no weight change to a maximum weight change of 0.02 g and was comparable between participants with CF and healthy volunteers (Table 2). There was a minor increase in surgical mask weight (median change, 0.01 g) after 40 minutes compared with 10 minutes of wear (P = 0.031; Table 2). No statistical differences in mask weight change were seen in other time or mask type comparisons (Table 2).

Our study demonstrates that face masks worn for clinically relevant periods are effective at reducing the release of potentially infectious aerosols during coughing in people with CF. These results expand on our earlier observations that demonstrated surgical masks and N95 masks were both effective at reducing the release of infectious cough aerosols when the mask wear was of shorter duration (3). The outcomes of our studies demonstrate that surgical masks are effective and tolerable as source control (3) and support Cystic Fibrosis Foundation recommendations for surgical mask wear to reduce the risk of CF pathogen transmission in the hospital setting (2).

Surgical masks were the preferred mask type for source control in terms of comfort, which is similar to our short-term wear mask study findings (3). Healthy volunteers tolerated the surgical masks better than those with CF, and participants with CF who had higher lung function tolerated surgical masks better. When the comfort of surgical masks was assessed after extended wear in this cohort, a major finding was that the comfort ratings remained unchanged regardless of wear time for both people with and without CF. Therefore, surgical masks are not only effective but are also well tolerated by participants after 40 minutes of total wear.

An accompanying editorial of our recent mask study (3) questioned whether mask dampness may affect the ability of the mask to function as source control after prolonged wear times (5). The CF infection control guidelines indicate that masks being used as source control should be replaced when damp (2), and excessive moisture accumulation was a common reason for surgical mask replacement in people with tuberculosis using surgical masks as source control (10). Our data indicate that although there was evidence of surgical mask moisture accumulation after 40 minutes of total wear (estimated by increased weight), the surgical mask continued to function effectively as source control, mitigating this concern.

There are several limitations to this study. First, the infectious dose of P. aeruginosa is unknown, and therefore the infection risk cannot be determined. Second, participants remained in view of staff while wearing the masks, and this may have modified the extent to which participants interfered with the mask, leading to an incorrect estimation on the mask’s protective effects. Third, participants were seated during the cough testing, and this may have affected the participant’s ability to cough freely. Fourth, although some participants experienced episodes of spontaneous cough during testing, we were unable to differentiate between spontaneous and voluntary cough, and therefore the protective effects of the masks may be overestimated. Fifth, the effectiveness and tolerability of masks is reported in adults only, and these characteristics need to be studied in children. Sixth, our study had a maximum wear time of 40 minutes, and the effectiveness of masks worn for longer periods is unknown. Finally, we did not assess inward protection provided by masks, but this has been highlighted as an understudied field of research (11).

Our study confirms the effectiveness of surgical masks at reducing the release of P. aeruginosa cough aerosols in people with CF and provides evidence of patient tolerability and functionality of these masks as source control after 40 minutes of total wear.