key: cord-0693151-q4aqkr0w authors: Patel, Samir N.; Mahmoudzadeh, Raziyeh; Salabati, Mirataollah; Soares, Rebecca R.; Hinkle, John; Hsu, Jason; Garg, Sunir J.; Regillo, Carl D.; Ho, Allen C.; Cohen, Michael N.; Khan, M. Ali; Yonekawa, Yoshihiro; Chiang, Allen; Gupta, Omesh P.; Kuriyan, Ajay E. title: Bacterial dispersion associated with various patient face mask designs during simulated intravitreal injections date: 2020-10-28 journal: Am J Ophthalmol DOI: 10.1016/j.ajo.2020.10.017 sha: f1037de6f96e492e9b20554c7c30ad37af597986 doc_id: 693151 cord_uid: q4aqkr0w PURPOSE: To investigate bacterial dispersion with patient face mask use during simulated intravitreal injections. DESIGN: Prospective cross-sectional study METHODSSETTING: Single-center STUDY POPULATION: Fifteen healthy subjects were recruited INTERVENTION: Each participant was instructed not to speak for 2-minutes, simulating a “no-talking” policy, while in an ophthalmic examination chair with an blood agar plate secured to the forehead and wearing various face masks (no mask, loose fitting surgical mask, tight-fitting surgical mask without tape, tight-fitting surgical mask with adhesive tape securing the superior portion of the mask, N95 mask, and cloth mask). Each scenario was then repeated while reading a 2-minute script, simulating a talking patient. MAIN OUTCOME MEASURES: Number of colony-forming units (CFU) and microbial species. RESULTS: During the “no-talking” scenario, subjects wearing a tight-fitting surgical mask with tape developed fewer CFUs compared to subjects wearing the same mask without tape (difference, 0.93CFU; 95%CI, 0.32–1.55; P=.003). During the speech scenarios, subjects wearing a tight-fitting surgical mask with tape had significantly fewer CFUs compared to subjects without a face mask (difference, 1.07CFU; P=.001), subjects with a loose face mask (difference, 0.67; P=.034), and subjects with a tight face mask without tape (difference, 1.13; P<.001). There was no difference between those with a tight-fitting surgical mask with tape and an N95 mask in the “no-talking” (P>.99) and “speech” (P=.831) scenarios. No oral flora was isolated in “no-talking” scenarios, but was isolated in 8/75 (11%) cultures in speech scenarios (P=.02). CONCLUSION: Addition of tape to the superior portion of a patient’s face mask reduced bacterial dispersion during simulated intravitreal injections, and had no difference in bacterial dispersion compared to wearing N95 masks. Since the introduction of intravitreal anti-vascular endothelial growth factor (anti-VEGF) therapy, intravitreal injections have become one of the most commonly performed procedures in all of medicine. 1 Although these medications have excellent safety profiles, acute bacterial endophthalmitis remains an uncommon but visually devastating complication. 2 Multiple prior studies have evaluated potential risk factors associated with post-injection endophthalmitis. [3] [4] [5] [6] [7] In particular, prior studies have established that the dispersion of oral flora may be reduced by minimizing speaking during the procedure and thereby reduce the incidence of oral flora-associated endophthalmitis. 7 Prior experimental investigations involving simulated intravitreal injections suggest that face mask use by physicians may reduce bacterial dispersion associated with speech. 8, 9 However, it is unknown how patient face mask use may affect bacterial dispersion. These findings are of particular importance given that, during the COVID-19 pandemic, universal precautions have been established for patients and physicians to wear face masks in order to decrease potential exposure to coronavirus through respiratory secretions. 10, 11 However, there is concern that face mask use by patients during an intravitreal injection may result in increased bacterial dispersion toward the eye. Prior studies have suggested that various face mask designs may result in upward or downward bacterial dispersion. 12, 13 Furthermore, in response to the critical shortage of medical face masks resulting from the COVID-19 pandemic, many patients may be wearing homemade cloth masks as recommended by the Center for Disease Control and Prevention. 14 It is unknown how these types of face masks may affect the degree of bacterial dispersion. The purpose of this study is to investigate the amount of bacterial dispersion associated with various patient face mask designs during simulated intravitreal injections. This prospective, cross-sectional, single-center study was conducted in accordance with the tenets of the Declaration of Helsinki and conformed to the Health Insurance Portability and Accountability Act. The protocol was prospectively approved by Wills Eye Hospital institutional review board, and all participants provided written informed consent. Fifteen healthy subjects were recruited to participate in the study as previously described. 8 Inclusion criteria included subjects at least 18 years of age with the ability to read a standardized script for two minutes. All subjects had previously received fittesting for N95 face masks based on the United States Occupational Safety and Health Administration guidelines. Exclusion criteria included any subject with a history of upper respiratory symptoms, fever, cough, or chills within the past two weeks. Each subject was seated in an ophthalmologic examination chair in a standard examination lane at an outpatient ophthalmology office. The examination chair was reclined until the volunteer's face was approximately 45 degrees from to the ground. A J o u r n a l P r e -p r o o f standardized 100mm circular blood agar plate (BD BBL TSA II 5% Sheep blood; Becton, Dickinson and Company, Franklin Lakes, NJ) was then secured on the subject's forehead. Twelve scenarios were then tested. Six scenarios simulated a "no talking" policy in which the subjects were instructed to sit in silence for two minutes while breathing with their mouth closed. After completing the "no talking" scenarios, subjects were instructed to read a standardized script for two minutes for each of the six scenarios. For both the "no talking" and speaking scenarios, the face mask conditions included the following: (1) wearing no face mask, (2) wearing a loose fitting surgical face mask ( Figure 1A All blood agar plates were sealed and taken to Jefferson Clinical Microbiology Laboratory (Thomas Jefferson University, Philadelphia, PA) where they were incubated for 72 hours at 37°C in a 5% carbon dioxide-rich environment. The number of bacterial colonies per plate was counted, and bacteria were identified using standard laboratory techniques by microbiologists who were masked to the plate collection sequence. No organisms were excluded from analysis, and a culture was considered to be oral floraassociated when Enterococcus or Streptococcus species was grown on culture. All data were analyzed using statistical software (IBM SPSS 25 Statistics, Armonk, NY, USA). The mean difference of colony-forming units (CFUs) and 95% confidence intervals among the groups were analyzed using an analysis of variance with adjustment for multiple comparisons using a Bonferroni correction. For categorical variables, significant differences between groups were analyzed using a Pearson's Chisquared test or Fisher's exact test. Pair wise comparisons between the "no talking" scenario and speech scenarios were performed using a paired student's t-test with adjustment for unequal variances. Statistical significance was considered to be a 2sided P value < .05. Fifteen subjects were recruited for the study, and a total of 180 blood agar plates were successfully incubated. Overall, the mean (SD) [range] CFU was 0.56 (0.88) [0 -4] in the "no talking" scenarios compared to 0.83 (0.95) [0 -6] in the speech scenarios (P = .044). Bacterial dispersion during "no talking" conditions Figure 2A shows the CFUs under the "no talking" scenario among the different face mask conditions. During the "no talking" scenario, subjects with tight-fitting face mask without tape grew the most colonies with a total of 17 CFUs or a mean (SD) of 1.13 (1.60) per subject. In contrast, during the "no talking" scenario, subjects with a tight-fitting face mask with adhesive tape or those with an N95 mask both grew the fewest number of colonies with a total of 3 CFUs or a mean (SD) of 0.20 (0.56) and 0.20 (40) per subject, respectively. Notably, during the "no talking" scenario, subjects wearing an N95 mask grew fewer CFUs compared to subjects wearing a tight-fitting face mask without tape (mean difference, 0.93; 95% CI, 0.32 -1.55; P = .003). Furthermore, during the "no talking" scenario, subjects wearing a tight-fitting face mask with tape grew fewer CFUs compared to subjects wearing a tight-fitting face mask without tape (mean difference, 0.93; 95% CI, 0.32 -1.55; P = .003). However, during the "no talking" scenario, there was no difference between subjects wearing the tight-fitting face mask with tape and subjects wearing an N95 mask (difference, 0; 95% CI, -0.62 -0.62; P >.99). During the "no talking" scenario, there was no significant difference in mean CFUs between the no face mask and loose fitting face mask group (P = .831), between the no face mask and tight-fitting face mask (P = .089), between the no face mask and tight-fitting face mask with adhesive tape (P = .201), between the no face mask and N95 face mask (P = .201), and between the no face mask and cloth face mask (P = .831). Bacterial dispersion during speech conditions Figure 2B shows the mean CFUs under the speech scenario among the different face mask conditions. During the speech scenarios, subjects wearing a tight-fitting face mask without tape grew the most colonies with a total of 21 CFUs or a mean (SD) of 1.4 (0.82) per subject. Subjects wearing an N95 mask grew the fewest colonies with a total of 3 CFUs or a mean (SD) of 0.2 (0.40) per subject. Subjects wearing a tight-fitting face mask with adhesive tape during the speech scenario grew significantly fewer CFUs compared to the following groups: 1) subjects with no face mask (mean difference, 1.07; 95% CI, 0.45 -1.68; P = .001); 2) subjects with a loose face mask (mean difference, 0.67; 95% CI, 0.05 -1.28; P = .034); and 3) subjects with a tight face mask without tape (mean difference, 1.13; 95% CI, 0.52 -1.75; P < .001). Subjects wearing an N95 face mask during the speech scenario grew significantly fewer CFUs compared to the following groups: 1) subjects with no face mask (mean difference 1.13; 95% CI, 0.52 -1.75; P < .001); 2) subjects with a loose face mask (mean difference 0.73; 95% CI, 0.12 -1.35; P = .02); 3) subjects with a tightfitting face mask without tape (mean difference 1.20; 95% CI, 0.59 -1.82; P < .001); and 4) subjects with a cloth face mask (mean difference, 0.67; 95% CI, 0.05 -1.28; P = .034). During the speech scenario, there was no difference between subjects wearing the tight-fitting face mask with tape and subjects wearing an N95 mask (difference, 0.07; 95% CI, -0.55 -0.68; P = .831). Bacterial dispersion between the speech and "no talking" conditions Table 1 summarizes the mean (SD) CFUs for each face mask design during the "no talking" and speech scenarios. In the no face mask scenario, subjects had significantly more mean CFUs during the speech scenario when compared to the "no talking" scenario (mean difference, 0.73; 95% CI, 0.12 -1.35; P = .020). There was no J o u r n a l P r e -p r o o f difference in mean CFUs between the "no talking" and the speech scenarios with a loose fitting face mask (P = .201), tight-fitting face mask without tape (P = .393), tightfitting face mask with adhesive tape (P = .831), N95 face mask (P > 0.99), or cloth face mask (P = .521). A total of 125 CFUs were isolated with 50 isolated during the "no talking" scenario and 75 isolated in the speech scenario. Of the 50 CFU in the "no talking" group, 0/50 (0%) were from oral flora, whereas 8/75 (11%) of the CFU in the speech group were from oral flora (P=.02). The most common organism isolated in the "no talking" scenario were Staphylococcal species (32/50, 64%). In the speech scenarios, the most common organism isolated in the "no talking" scenario were Staphylococcal species (38/75, 51%). In the speech scenarios, 8 oral flora organisms were isolated with of 4 of the cases in the no face mask scenario, and 4 cases in the tight-fitting face mask without adhesive tape scenario. Causative oral flora organisms included 3 colonies of Streptococcus mitis, 3 colonies of Streptococcus viridans, and 2 colonies of undifferentiated Streptococcus. The COVID-19 pandemic has necessitated universal face mask protocols for infection control, which has resulted in significant interest in understanding face mask effectiveness from various disciplines, 11,15-17 including ophthalmology 18 . However, there are few data on how patient face mask use may alter bacterial dispersion during an intravitreal injection, which subsequently may affect the risk of post-injection endophthalmitis. In this experimental study of bacterial dispersion during a simulated intravitreal injection scenario with six different face masks, we found that subjects wearing tight-fitting face masks without adhesive tape covering the superior portion of the mask grew the most bacterial organisms under both "no talking" and speech scenarios. However, the introduction of adhesive tape to secure the superior portion of the same tight-fitting face mask significantly reduced the amount of bacterial dispersion toward the eyes. Furthermore, the introduction of adhesive tape to secure the superior portion of a tight-fitting face mask resulted in no statistically significant difference in bacterial dispersion compared to bacterial dispersion with an N95 face mask. We assessed the number of CFUs under "no talking" conditions to simulate the clinical practice of a "no talking" policy in which intravitreal injections are administered under a strict policy of silence such that the physician, patient, and others in the room do not speak during the injection procedure. Under the "no talking" condition, bacterial growth around the subject's eye was highest when wearing a tight-fitting face mask without adhesive tape, and the source of bacteria in these "no talking" scenarios may be from the subject's natural breathing. A known physiologic reaction to stress and anxiety is to hyperventilate, which some patients routinely do as they are anticipating an intravitreal injection, and wearing a mask may contribute to this phenomenon. However, bacterial growth was significantly reduced when subjects wore the same tight-fitting face mask but had adhesive tape attached over the entire superior portion of the face mask. The bacterial growth was reduced such that the addition of adhesive tape to the tightfitting face mask resulted in similar rates of bacterial growth compared to those wearing an N95 face mask. A prior study on face mask use by the injector during simulated intravitreal injections reported that subjects speaking without a face mask resulted in an increased proportion of bacterial colony growth. 8 The current study evaluated bacterial growth under speech scenarios where subjects were instructed to read a 2-minute standardized script while wearing various face masks. Regardless of the type of face mask worn, subjects in the speech scenarios had significantly greater bacterial growth compared to the "no talking" scenarios. Specifically, subjects wearing no face masks had greater bacterial growth in the speech scenario compared to the "no talking" scenario. These findings underscore the importance of a speech reduction policy when intravitreal injections are administered. Furthermore, within the various speech scenarios, subjects wearing N95 face masks and tight-fitting face masks with adhesive tape had significantly fewer CFUs compared to subjects wearing no face masks, loose face masks, and tight-fitting face masks without tape. Current guidelines from the Center for Disease Control and Prevention at the time of writing recommend cloth face covering, which may not adhere to the face as well. 14 Our study found similar rates of bacterial dispersion with cloth based face mask use in both scenarios. Although we did not assess the effect of adhesive tape for cloth face masks, future studies may be indicated to evaluate these specific scenarios given the findings of this study. In both the "no talking" and speech scenarios, subjects wearing a tight-fitting face mask without tape had the highest bacterial growth -similar to, or even higher than, not wearing any mask. It is possible that the tight-fitting face without tape may result in a greater amount of bacterial dispersion upward toward the subject's eye. 12, 17 Indeed, a recent study assessing respiratory droplet velocities with face mask use during simulated coughs reported that even with tight-fitting face masks, small openings can lead to leakage of droplets around the mask. 17 However, our study suggests that securing the superior portion of the same, tight-fitting face mask with adhesive tape significantly decreased the amount of bacterial growth, suggesting that securing the superior portion of the mask with tape may create an important barrier for upward bacterial dispersion. Although all forms of endophthalmitis are visually threatening, oral floraassociated endophthalmitis is associated with a particularly poor prognosis. [19] [20] [21] Prior clinical studies have established that oral flora-associated endophthalmitis may be reduced with the implementation of a strict "no talking" policy by the physician and patient during intravitreal injection administration. 7, 21 Refraining from speaking during an intravitreal injection is thought to minimize the potential to contaminate the uncapped needle or conjunctival surface with oral flora immediately before or during the injection. Indeed, in our study, there were no cases of oral flora isolated during the "no talking" scenarios, which further supports the efficacy of a speech reduction policy to reduce the risk of oral flora-associated endophthalmitis. In contrast, during the speech scenarios, oral flora were isolated when subjects were speaking without a face mask or speaking J o u r n a l P r e -p r o o f with a tight-fitting face mask without tape, which parallel a prior study showing high rates of oral flora growth during similar scenarios. 8 Our study has several limitations. Blood agar plates were used for bacterial quantification and identification. However, these plates do not precisely reproduce the target field of the ocular surface. Furthermore, we standardized the distance between the agar plate and the eye, the duration of speech, and the positioning of the face mask, but in real world clinical scenarios there may be significant variability in all of these clinical factors. For example, prior studies have reported that the tendency to wiggle one's face beneath a surgical mask 12 may increase bacterial dispersion and shedding, presumably from the beard and facial skin. Another limitation is that the speech scenarios had the subject read a script for two minutes, which is likely more time than any patient would spend speaking during an intravitreal injection. However, two minutes may be realistic when considering the preparation time for an intravitreal injection during which the patient may be speaking. Furthermore, prior studies have suggested that bacterial dispersal can occur even with shorter durations of speech. 22 Another limitation is that our study included participants who were previously fitted to wear N95 face masks, which may not be generalizable to the true patient population who wears N95s. Our findings may be underestimating the real world benefit of taping because many patients come in with suboptimal fitting masks, regardless of material or type. Finally, it is unknown if the statistically significant difference in colony counts among the groups is of clinical significance. This study cannot prove that the presence of additional bacterial colonies surrounding the injection site necessarily contributes to an increased risk of post-injection endophthalmitis. However, given the devastating visual prognosis of postinjection endophthalmitis, it is critical to minimize the potential risk of endophthalmitis with any measures available. In summary, these experiments replicate the specific conditions of an intravitreal injection when patients are wearing different types of face masks. Until now, there was minimal evidence in the literature to guide practitioners in the management of patient face mask use during an intravitreal injection. These in vitro experiments suggest that addition of adhesive tape to the superior portion of a patient's face mask during an intravitreal injection reduces bacterial dispersion, which may subsequently reduce the risk of post-injection endophthalmitis. No talking" scenarios in which subjects were instructed to sit in silence for two minutes. 2B) Speech scenarios in which subjects were instructed to read a script for two minutes. Error bars represent standard deviation. * indicates P < 0.05 and ** indicates P < 0.01, and *** indicates P < 0.001. CFU = colony-forming units. • Patient face mask use may alter bacterial dispersion around the eye during intravitreal injections. • There was significantly more bacterial dispersion when wearing a tight-fitting face mask without tape compared to wearing a tight-fitting mask with tape. • There was no difference in bacterial dispersion between tight-fitting surgical masks with tape and N95 masks. • Taping the superior portion of a patient's face mask may limit bacterial dispersion when performing intravitreal injections. This study evaluated if face masks worn by patients during intravitreal injections altered bacterial dispersion around the eye. In this cross-sectional study of 15 participants undergoing simulated intravitreal injections with six face mask designs, there was significantly more bacterial dispersion when wearing a tightfitting face mask without tape compared to wearing a tight-fitting mask with tape. Taping the superior portion of a patient's face mask may limit bacterial dispersion when performing intravitreal injections. Trends of anti-vascular endothelial growth factor use in ophthalmology among privately insured and medicare advantage patients International practice patterns for the management of acute postsurgical and postintravitreal injection endophthalmitis: European vitreo-retinal society endophthalmitis study report 1 Endophthalmitis following intravitreal injections performed in the office versus operating room setting Outcomes and risk factors associated with endophthalmitis after intravitreal injection of antivascular endothelial growth factor agents The role of topical antibiotic prophylaxis to prevent endophthalmitis after intravitreal injection The impact of prefilled syringes on endophthalmitis following intravitreal injection of ranibizumab Effect of a strict 'no-talking' policy during intravitreal injection on post-injection endophthalmitis Bacterial dispersal associated with speech in the setting of intravitreous injections Reducing oral flora contamination of intravitreal injections with face mask or silence Universal masking in hospitals in the covid-19 era Visualizing speech-generated oral fluid droplets with laser light scattering Mask wiggling as a potential cause of wound contamination Is a mask necessary in the operating theatre? Use of cloth face coverings to help slow the spread of covid-19 Physical distancing, face masks, and eye protection to prevent person-to-person transmission of sars-cov-2 and covid-19: A systematic review and meta-analysis Visualizing the effectiveness of face masks in obstructing respiratory jets On respiratory droplets and face masks Do slit-lamp shields and face masks protect ophthalmologists amidst covid-19? Microbial spectrum and outcomes of endophthalmitis after intravitreal injection versus pars plana vitrectomy Endophthalmitis after intravitreal injection: The importance of viridans streptococci Endophthalmitis following intravitreal injection The use of surgical facemasks during cataract surgery: Is it necessary?