Since the global outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, over 400 million patients have been infected. Due to the high-risk of human-to-human transmission via aerosols and droplets, upper gastrointestinal endoscopy was initially withheld because of the potential to release aerosols from the oropharynx by the gag reflex or coughing or from body fluids, especially at less than 1 m from the patient. Thus, preprocedural polymerase chain reaction (PCR) for the coronavirus disease 2019 (COVID-19) virus was performed in patients who required an endoscopy. Personal protective equipment (PPE) with an N95 respirator was mandated in all procedures on patients with the infection. The endoscopic unit was prepped before the procedure and the disinfection of instruments was required after completion. Non-urgent cases were deferred to reserve necessary resources. This caused a delay in diagnosis and management. The endoscopic case volumes decreased by 50% to 80%.1 The estimated reduction in the detection of intraluminal gastrointestinal malignancies was over 20%, which may lead to cancer-related crises in the future. Our review aims to assess the possibility of aerosol generation during upper gastrointestinal endoscopy. Appropriate peri-procedural management is being explored to mitigate further infection to healthcare workers (HCW), and to reduce the delay in endoscopic diagnosis and treatment for patients in the pandemic.

SARS-CoV-2 can be detected in oropharyngeal swab specimens, blood, saliva, urine, and feces. Previous studies on isolated viral RNA by real-time PCR have confirmed the involvement of multiple organs in the infection. The infection is transmitted frequently via droplets, airborne carriers, and fomites.2-4 After coughing or talking, droplets from the respiratory tract containing viable viruses can infect other individuals. Large droplets of over 5 μm can deposit by falling on surfaces and cause propagation through fomites. During normal breathing, aerosol particles smaller than 5 μm can disperse farther by airborne transmission than by droplets.5 According to the World Health Organization, aerosol-generating procedures (AGPs) includes open airway suction, sputum induction, cardiopulmonary resuscitation, endotracheal intubation and extubation, non-invasive ventilation such as bilevel positive airway pressure and continuous positive airway pressure, bronchoscopy, and manual ventilation. Nebulizer administration and high-flow oxygen delivery are also considered AGPs, but there are limited data available on these.6-8 Several factors are associated with a greater risk of dissemination. Forced air over the respiratory mucosa, such as during coughing and positive-pressure ventilation, can emit virus-laden particles. Severe and symptomatic patients in close-range contact are likely to transmit the infection to others. In poorly ventilated spaces, infected particles may accumulate and be transmitted over a wider area. A prolonged exposure time also results in a higher rate of infection.7

Endoscopy was deemed as an additional aerosolization procedure in described lists.8 Some numerical statistical studies were conducted to prove this theory.9-12 Aerosols from patients undergoing upper gastrointestinal endoscopy were measured by a handheld optical particle counter before, during, and after the procedure. Particle sizes were reported to be in the range of 0.3 to 10 μm.10-12 Sagami et al.9 compared conscious sedated upper gastrointestinal endoscopy with non-endoscopy as a control group, in the same plastic enclosure and positive-pressure environment. They reported a significant increase in the number of particles during and after the procedure in the endoscopy group. A higher body mass index and more episodes of burping might have induced more aerosol generation, according to their multivariate analysis.9 Chan et al.10 also described a significant increase in particles of all sizes during upper gastrointestinal endoscopy, when measured at 10 cm from the patient’s mouth. Application of a dental sucker may reduce the number of counted particles. Gregson et al.11 evaluated the surgical assessment endoscopy of bariatric patients and found that endoscopic-evoked coughs generated more aerosols than the baseline tidal breathing and volitional cough. From these observations, it can be concluded that coughing is a dominant source of aerosols during upper gastrointestinal endoscopy. Compared to trans-nasal endoscopy and lower gastrointestinal endoscopy, per-oral gastroscopy seemed to generate the greatest number of aerosols. The largest particle-generating events were associated with the application of a local anesthetic throat spray, which produced 10 times the particles generated from cough. The second largest event was extubation.12 Considering these prospective quantitative studies, upper gastrointestinal endoscopy should be determined as an AGP. Periprocedural management should follow the AGP protocols to prevent the transmission of infected particles.

Various articles have proposed new inventions of protective devices for use during endoscopic procedures to reduce the spread of aerosols from patients to endoscopic staff. Lai13 developed an aerosol box made from acrylic or transparent polycarbonate sheet, to assist in tracheal intubation during the early COVID-19 outbreak. Two holes were created to fit the insertion of both hands. During intubation, the box was placed to cover the head of the patient. An extra plastic drape could be attached to the uncovered side of the box. This minimized droplet scattering to others. It was reusable after cleaning with alcohol or bleach, and affordable. In the past year, several designs were adapted, resulting in an endoscopic box shield being used in upper gastrointestinal endoscopy. Two holes were added for both hands, and smaller holes were made on the other side of the box for endoscopic shaft introduction (Fig. 1A). Prior studies have proven its protective efficacy by reducing the extent of ultraviolet visualization of fluorescent dye scattered after bursting from a rubber balloon placed in a mannequin’s hypopharynx. Contamination was decreased on the endoscopist’s clothing and the floor of the room.14-17

Another shielding instrument with a more compact feature is the acrylic face shield developed by the Colombia University endoscopic unit.18 They created an open-and-close system orifice for the patient’s mouth. The face shield is worn while waiting and during the procedure. Endoscopy can be performed as usual by opening the orifice (Fig. 1B). Acrylic face shields are intended to be reusable and can be sterilized before they are used on different patients. Moreover, their mass production could reduce manufacturing costs.

Transparent plastic sheets can be modified to be used as less rigid endoscopic barriers. Sabbagh et al.19 invented this by folding a 75×75-cm size plastic sheet with adhesive tape placed away from the edge of the fold to form a slit, and lubricated the opening to facilitate the introduction of the scope. After the oxygen nasal cannula and oral protector are given to the patient, the plastic sheet is fixed to cover the patients head, and endoscopy can be performed by insertion via the created channel (Fig. 1C).

Devices with an entire facial cover may cause patient discomfort. With increased availability and low cost, a double surgical mask with a slit was designed by Lazaridis et al.20 Two surgical masks were taped at the upper and lower borders of the masks, with a 1.2-cm slit cut through both masks using scissors, for insertion of the scope (Fig. 1D). They also applied this clinical practice for antegrade double-ballon enteroscopy.20

Recently, Fujifilm released two endoscopic accessories to reduce the risk of droplet dissemination. One of them is a droplet-shielding mouthpiece called “B1”, which consists of a drape shield to cover the patient’s face to protect HCWs and is attached with double-layer sponges with slits to reduce the gap between the scope and the mouthpiece (Fig. 1E). They have claimed that B1 could eliminate over 99% of droplets 5 μm in size and larger, compared to the currently existing mouthpiece. The second is an endoscope cover called “P1” which covers the forceps valve. The drape sheet covers the entire working portion of the endoscope. Installing P1 to the port of the working channel may minimize secretion spillage when the instrument is pulled out (Fig. 1F).21

In collaboration with Olympus Medical System and an otolaryngologist, Kikuchi et al.22 proposed the shielding device for endoscopic procedures (STEP) as a barrier device. It consists of a mask with a suction port and a valve-like hole for endoscopy, which connects to the vinyl tubular plastic drape taped to the base of the endoscope. During the procedure, the endoscope is stored in the drape and inserted through the mask so that the endoscopist would not be in direct contact with the patient’s aerodigestive fluids (Fig. 1G). An analysis of contact, droplet, and aerosol infections was performed by comparing STEP and non-STEP groups using simulation models. A significantly lower spread was found in all three models with similar procedural times and visual analog scale scores for endoscopic image evaluation from the pharynx to the second part of the duodenum. Therefore, the STEP may be useful for infection control during endoscopy. The summarized information on the tools is shown in Table 1.13-22

PPE for endoscopy personnel should be adjusted according to patient risk stratification. A full PPE is essential for high-risk or confirmed positive patients. The PPE includes a filtering facepiece 2 or 3 mask, double gloves, and other coverings, as mentioned above. In low-resource settings, reusable respirators, face shields, goggles, and boots may be acceptable after appropriate sterilization and decontamination methods.26 The staff should be trained in the correct sequence of donning and doffing the PPE, as it is critical. A minimal number of workers should be in the room and the team should not be switched during procedures to minimize PPE usage and decrease the contamination risk. A bedside endoscopy can be performed for critically ill patients in the intensive care unit. Patients with persistent cough should remain in the room until it subsides. After completion of the procedure, the patient must be transported to the recovery area as soon as deemed safe. Procedure-related staff should properly remove the PPE and wash their hands. As much as possible, all required documentation should be done outside the room.

After a COVID-19 patient is shifted out of the procedure room, the room should be maintained under negative pressure for at least 30 minutes, and in the absence of negative pressure, for 60 minutes, before the next patient, to allow for the dispersion of virus-containing aerosols. Standard cleaning with a hospital-grade disinfectant solution should be done by personnel wearing PPE. Reprocessing staff should also wear PPE due to the risk of exposure to contaminated devices. All patients should be traced and contacted after 7 and 14 days, to enquire about new COVID-19 diagnoses or symptoms.23,28

An upper gastrointestinal endoscopy protocol in the current COVID-19 pandemic should be provided, as the procedure is considered an AGP. Periprocedural preparation and protective devices are necessary to prevent the aerosol dispersion of infection, which would be beneficial for healthcare personnel and patients.