Do Disinfectant Solutions during Gargling Reach the Pharynx?

Introduction: In times of COVID-19, gargling disinfectant is commonly used. Disinfectant solutions seem to decrease the infection’s symptoms. For disinfection, several techniques are reported. So far, there are no data about the regions in the upper airways achieved by gargled fluid. Methods: Ten healthy volunteers without any dysphagia were investigated with a high-sensitivity flexible endoscopic evaluation of swallowing (hsFEES®) during and after gargling colored water. One volunteer repeated the gargling process in fast and real-time MRI. Results: In all cases, no color accumulation was detected on the posterior pharyngeal wall, epi- or hypopharynx during gargling. The MRI scans confirmed the results. Conclusions: hsFEES® and fast MRI provide an insight into the gargling pattern. Data show that during gargling, the fluid covers the soft tissue in the oral cavity and the anterior part of the soft palate, but not the posterior pharyngeal wall nor the epi- and hypopharynx.


Introduction
The novel coronavirus (COVID- 19) is a member of the Coronaviridae family, which consists of single-stranded RNA viruses that cause respiratory infections in humans.In December 2019, COVID-19 virus broke out in Wuhan, China, and has since become a pandemic worldwide.Clinical courses vary from asymptomatic to severe respiratory distress syndrome with consecutive deaths [1].According to the Johns Hopkins Coronavirus Resource Center, 6,252,707 people worldwide have died as a result of coronavirus infection to date (https://coronavirus.jhu.edu).The virus is mainly transmitted via the air.Both symptomatic and asymptomatic persons are carriers of the infection [2].In particular, asymptomatic patients cause viral transmission via the virus present in saliva [3].Therefore, oropharyngeal disinfection to reduce the viral load in the mucosa is often discussed [4][5][6].Throat mucosa plays an important role for the viral replication [5].This is thought to reduce infection, as a low viral load is associated with a reduction in disease severity [7].In addition, a low viral load could minimize the risk of transmission [8].For decades, disinfection of the oropharynx has been mentioned to shorten the symptoms of respiratory infection.Zou et al. [5] showed higher viral load of SARS-CoV-2 in nasal swabs compared with throat swabs.Various techniques and antiseptic solutions are published for nasal and oropharyngeal disinfection [6,[9][10][11].In the literature, the reduction of SARS-CoV-2 in the mucosa by nasal irrigation and gargling is discussed [9][10][11][12][13][14][15].There are no literature data on the distribution and penetration depth of disinfectant solutions in the mouth and throat during gargling.With new diagnostic tools like the high-sensitivity flexible endoscopic evaluation of swallowing (high-sensitivity FEES ® ; hsFEES ® [16][17][18][19][20]) and fast MRI, the distribution of gargled fluid can be visualized.Furthermore, hsFEES ® is able to demarcate the affected areas following the gargling process with yellow-colored water [17,19,20].Fast and real-time MRI uses a high temporal resolution to visualize the gargling motion.In this study, the contact areas of gargled fluid to the oropharyngeal tissue were investigated.

Materials and Methods
The prospective study of hsFEES ® was approved by the responsible Ethical Committee (2021020302) and conducted according to the guidelines of the Declaration of Helsinki (Washington 2002) for conducting clinical studies involving humans.MRI was performed as a self-experiment by one of the authors.
High-Sensitivity FEES ® For hsFEES ® , a total of 10 healthy volunteers without any dysphagia were included.The group comprised 5 female and 5 male participants aged 22-42 years (median 31 years).Four of the volunteers have had a tonsillectomy in childhood.A chip on the tip of flexible endoscope (flexible video-endoscope ENT-VH, Olympus Medical Systems Corp., Tokyo, Japan) with white light and the image enhancement system "narrow-band imaging" (NBI) was used.The white balance was performed with a standard white balance tube.Videos were recorded using Digital Video Archive Software (DiVAS; Xion Medical, Berlin, Germany).
The NBI is an optical filter that allows narrow-band light with the wavelength of 415 nm blue light and 540 nm green light.Depending on the soft tissue, the NBI light is absorbed in different manners.This leads to a higher contrast of vessels and the surrounding mucosa.This technique is primarily used to diagnose pathological epithelial changes [21][22][23].Using the NBI technology in a flexible endoscopic evaluation of swallowing (FEES ® ) allows a higher detection rate of pathological findings [17].With NBI illumination, yellow and green color turns into bright red.This results in an enhanced contrast and is called "hsFEES ® " [17,19,20].Nienstedt et al. [17] showed a significantly higher detection rate using the NBI technique in FEES ® .It increases the overall reliability of FEES ® with a higher contrast of the bolus.Using the hsFEES ® , yellow color is able to identify micro-aspiration as well as tiny aspirated parts of the bolus in the trachea at low illumination.
To demonstrate that there are no lightning spots on the soft tissue in the oral cavity, the epipharynx, the pharynx, and the larynx, an investigation with white light and the NBI for the Does Gargling Reach the Pharynx for Disinfection?
hsFEES ® using the flexible endoscope was performed in the study group.After gargling for 10-15 s in a sitting position with a reclined head with 20 mL of yellow water, the flexible endoscopy of the oral cavity, the epipharynx, the pharynx, and the larynx with white light and NBI was repeated.The water was colored with commercially available yellow color particles (E 102).Previous studies show, that a dilution up to 1:10,000 is detectable with NBI in the oral cavity [17,20].Therefore, we decided to use a dilation of 1:1,000 in the gargling fluid.Finally, transnasal endoscopy was again performed during the gargling process.The images before gargling were compared with the images during and after gargling.Furthermore, one of the participants tried to gargle the fluid as deep as possible in the throat.

Magnetic Resonance Imaging
MRI investigations were performed on a clinical 3 Tesla (T) MR scanner (MAGNETOM Prismafit; Siemens Healthcare, Erlangen, Germany) using a 24 channel head/neck coil.A healthy volunteer was examined in supine position, and 20-30 mL of pineapple juice was ingested prior to each scan as a natural contrast agent via a flexible tube from a reservoir placed inside the bore of the magnet.
First, a series of sagittal images was acquired using a single-shot fast spin echo pulse sequence with Cartesian sampling, which yields strong T2-weighting and a temporal footprint of 304 ms for one 2D image.The series covers the entire process from ingestion of the juice to gargling for some seconds and finally the swallowing.The following MR parameters were used: TE = 91 ms, TR = 1,810 ms, flip angle = 150°, echo train length = 80, field of view FOV = 300 mm × 300 mm, spatial resolution = 1.2 mm × 1.2 mm, slice thickness = 10 mm.
Subsequently, a prototype spoiled gradient echo sequence equipped with a spiral read-out trajectory [24] was used to image the same maneuver as described above, however, at a considerably higher temporal resolution of 49.6 ms ("real-time").An automatic gradient impulse response function based on preemphasis [25] was used to minimize artifacts originating from inaccuracies of the gradient system.Further sequence parameters were as follows: sagittal slice orientation, TE = 0.84 ms, TR = 4.96 ms, flip angle = 20°, spatial resolution = 1.3 mm × 1.3 mm, slice thickness = 10 mm.Ten consecutive spiral arms, equidistantly spread across k-space (i.e., increment of 36°), were combined to build one undersampled frame.In consecutive frames, the k-space gaps were filled according to the golden ratio, i.e., the entire sampling pattern was rotated by 360°/(5*((sqrt(5))+1)) ≈ 22.25° [24,26].Images were reconstructed offline in MATLAB using a low-rank plus sparse model [27].

Results
High-Sensitivity FEES ® All of the participants were able to gargle the fluid for 10-15 s.No fluids could be detected in the epipharynx of the pharynx with the hsFEES.Only in the oral cavity (mainly the tongue) with the palatal arches, the soft tissues showed colored areas (Fig. 1).The 6 participants with tonsils showed in NBI light a slight moistening with the gargled fluid on the tonsils.In the 4 volunteers who underwent tonsillectomy in childhood, no colored soft tissue was found in the tonsil area using NBI light.
Endoscopy during the gargling process presented in every case a lowering of the soft palate without any penetration of the fluid into the pharynx.The posterior pharyngeal wall was not affected by gargling at any time during all endoscopies (Fig. 1).The results were the same for endoscopy after and during the gargling process.The trial of a "deep gargling" led to a leak into the valleculae and the pyriform sinus following an immediate trigger of the swallowing reflex.Moreover, this patient was highly focused on the "deep gargling process," that an aspiration occurred which instantly produced a coughing reflex (Fig. 2).

Magnetic Resonance Imaging
In both the fast spin echo as well as the real-time scan, the gargled bolus can be seen very well.The bolus of pineapple juice appears in a light white contrast.Furthermore, the structures, which are responsible for swallowing, are defined with high precision: the lips, the tongue with its base, the hard and soft palate with the uvula, and the pharynx particularly with the posterior wall.After applying the bolus of pineapple juice in the mouth, the typical tub-shape of the tongue is visible.The soft palate is lowered.There is no contact of the bolus to the posterior wall of the pharynx (Fig. 3, 4).
During the process of gargling, the soft palate remains in a low position with contact to the basal part of the tongue.The air runs like a staccato through this natural valve and produces the bubbling gargle.Later on, the swallowing reflex is triggered after contact to the swallowing trigger zones.Therefore, the soft palate occludes the epipharynx and the basal part of the tongue pushes the bolus into the pharynx.The larynx opens the esophagus via its anterior cranial movement.The pineapple juice is directed into the upper esophagus.Hightemporal-resolution real-time imaging shows that the soft palate and the basal part of the tongue are covered with the fluid like in a funnel (Fig. 4; online suppl.video 1; for all online suppl.material, see https://doi.org/10.1159/000533135).

Discussion
This is the first study with videoendoscopic and MRI documentation of gargling.The endoscopic techniques with hsFEES ® and fast MRI provides good insight into the gargling pattern.Data show, that the fluid does not reach the soft tissue in the posterior parts of the upper airways (pharynx, epipharynx, posterior pharyngeal wall).The velum and the soft palate are completely covered with the fluid.During the gargling process, no fluid penetrates into the pharynx as the velum has contact to the base of the tongue.This means, that the request to disinfect the pharynx and epipharynx via gargling does not reach the supposed target area [11][12][13].In a randomized, double-blind, placebocontrolled clinical trial, Mukherjee and colleagues investigated the effect of mouth rinses containing antimicrobial solutions to prevent upper respiratory tract infections.They showed that rinsing tended to protect against upper respiratory tract infections while reducing the severity and duration of cough and sore throat [28].However, as gargling is produced with an open airway while fluid is in the oral cavity under expiration, fluid in the pharynx is not expected.This is a natural protection of the upper airways to prevent aspiration.Trying to suppress this process and the airway-protection reflex with a "deep gargling" ended up in an aspiration with an immediate laryngeal clearing.The swallowing reflex was triggered.Looking at the swallowing pattern, this is the expected neurolaryngeal reaction.Moistening areas that elicit a swallowing or coughing reflex when gargling a water bolus is unphysiological as it carries a high risk of aspiration.Martinez and colleagues conducted mouth rinses of four SARS-CoV-2-infected patients for 1 min with 1% povidone-iodine PVP-I (PVP-1).They demonstrated that PVP-I resulted in a significant decrease in viral load in two of the four participants, albeit for only 3 h [29].The study by Gottsauner and colleagues had similar results.They analyzed the viral load in SARS-CoV-2-positive patients, before and after a mouth rinse with 20 mL 1% H 2 O 2 solution.The viral load initially decreased 2 h after mouth rinsing but increased again after 2-4 h [30].Our study may provide the answer to the lack of eradication.Gargling with disinfectants may moisten the soft tissue in the oral cavity and the soft plate.However, they do not reach the lower parts of the pharynx or the epipharynx.

Conclusion
Our data clearly demonstrate that gargling only addresses the oral cavity and the anterior part of the soft palate with the tonsils area.However, important areas such as the epipharynx, posterior pharyngeal wall, and hypopharynx are not reached.Thus, mouth rinsing with disinfectant solutions by gargling could only conditionally lead to virus eradication.

Statement of Ethics
All procedures performed in this study were in accordance with the ethical standards of the Institutional Research Committee (Ethikkommission, University of Wuerzburg; reference number 2021020302) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.Written informed consent was obtained from all participants to participate in the study.

Fig. 3 .Fig. 4 .
Fig. 3. Gargling with pineapple juice (+) in a timeline single-shot fast spin echo pulse sequence.T1: pineapple juice in the mouth (+) with lowered soft palate (*) and the basal part of the tongue in a high position without any contact to the posterior pharyngeal wall (→).T2: start of gargling.T3 and T4: gargling process.T5: position of the pineapple juice just before swallowing.T6: swallowing.