Adult patients referred and diagnosed with OETD were offered to be enrolled in this study. 1.

Diagnosed with unilateral or bilateral OETD for at least 3 consecutive months. Diagnosis was established by an average Eustachian tube dysfunction questionnaire-7 (ETDQ-7) score ≥3.

After chronic OETD was diagnosed and informed consent was obtained, patients were screened for baseline performance in terms of otoscopy, nasal endoscopy, tympanometry, audiometry and ETDQ-7. ETDQ-7 was established by McCoul et al. on 2012 (36). Since then, it was validated to many languages (33, 37–42) and became the primary tool for ETD diagnosis and follow up. Available CT scans were reviewed for intranasal structure abnormalities and any carotid canal dehiscence. Patients were given 4 weeks of daily nasal saline irrigation unless used before enrollment. Then, BDET was performed, followed by 4 weeks of 3–4 daily nasal saline irrigation. Patients were followed up at 6 weeks and 6 months post BDET for all these measures. Primary outcome measures were defined as mean change in overall ETDQ-7 from baseline to 6 months and any BDET-related complication. Secondary outcome measures were changes in tympanometry, degree of tympanic membrane retraction/atelectasis and their response to the Valsalva maneuver, as well as a change in the pure tone average of four frequencies 500, 100, 2,000, and 4,000 Hz (PTA4). TM retraction was classified according to Sade classification (43), in which stage 1 is mildly retracted TM, stage 2 is retracted TM as medially as the incudo-stapedial joint, stage 3 is retracted TM over the promontory, and stage 4 is adhesive atelectatic TM. For objective tympanometry, improvement was defined per ear as a change from type B to type A or C, or from type C to type A. Subjective improvement was defined as an average ETDQ-7 score <3 points at the 6 months follow-up.

The description of technical procedure of BDET has been described extensively in the literature, while the technique to use EVB has been previously published (34, 35). Patients were instructed to fast for 2 h before the procedure. Patients were positioned at 45° angle. Two sprays of oxymetazoline were given to each nostril, followed by cottonoids soaked with tetracaine 2% left for 15 min in each nasal cavity. No systemic steroids, anxiolytics or vestibular suppressants were given. The ETs or tympanic membranes were not anesthetized. Before the procedure, each endovascular balloon (EVB: Advance 35LP, COOK Medical, Bloomington, IN) was tested for leaks after all air was removed from its system. Under 30° angled nasal endoscope, a deflated endovascular balloon of 20 mm in length and 6 mm in diameter was carefully introduced through a maxillary sinus suction to the ET lumen, until the distal end (silver marker) of the deflated balloon was identified in the ET meatus. The operator verified full EVB insertion by direct endoscopic visualization of the proximal silver marker being held at the ET meatus (torus tobarius) (Figure 1D). With the silver marker at torus tobarius, the balloon length would start from there and go up by 2 cm inside cartilaginous portion of ET. The EVB was never advanced against pressure or resistance, and tended to slightly slide out upon inflation. Out-fracture of the inferior turbinate was performed if necessary. When the ventilation port was connected to suction to remove topical anesthetic solution or mucous from the nasal cavity, it was disconnected before inserting the endovascular balloon into the ET to allow ventilation and avoid positive middle ear pressure during balloon insertion into ET. The endovascular balloon was then inflated by saline to 12 ATM for 2 min. The inflation rate was constant at approximately 1 ATM/s to avoid rapid middle ear pressure change and resultant otalgia and possible vertigo. Upon ET dilation for 2 min, the endovascular balloon was gently deflated and removed from the ET. The inferior turbinate was repositioned if necessary. The endovascular balloon was examined for kinks ex vivo. Patients were instructed to avoid the consumption of hot food or liquids for a day, to reduce the risk of epistaxis from minor mucosal laceration, as well as to avoid sneezing against closed nostrils to avoid submucosal and subcutaneous emphysema. Intraoperative endoscopic photos are found in Figure 1.

Categorical variables were summarized as frequencies and percentages. Continuous variables were evaluated for normal distribution by the use of histograms. The Chi-square and Fisher tests were used to compare categorical variables, and the Mann–Whitney test was used to compare continuous variables. All statistical tests were two-sided, and P < .05 was considered significant. SPSS software was used for all statistical analyses (IBM SPSS Statistics for Windows version 25, IBM Corporation, Armonk, NY, USA, 2017).

This systematic literature review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (44).

This systematic literature review of collective published data did not require approval from the institutional review board or the ethical committee according to local law because it does not use individualized patient data.

A comprehensive review of the scientific literature was conducted by means of an a-priori research protocol. We searched for articles in “Medline” via “PubMed”, “EMBASE”, and “Web of Science” without limitation of publication date up to March 1st, 2023. The exact search algorithm is shown in Supplementary I. Shortly, the Boolean operators used were either term [Eustachian tube], [Eustachian tuboplasty], [Balloon dilation tuboplasty], [Eustachius] or [Balloon tuboplasty], combined with the term [In office], [Local anesthesia] or [Awake], that appeared as a text or keyword anywhere in the paper. After all, publications had been identified, two investigators (OJU and MDB) independently excluded duplicate titles and then screened publications for suitability by consensus.

The included articles described BDET under local anesthesia, with or without anxiolytics Benzodiazepine (midazolam) medication. In papers describing BDET under local and general anesthesia, only the local anesthesia arm was extracted for data synthesis. The target population was restricted to adults (≥18 years). Case series and clinical trials were enrolled if a minimal number of 5 subjects were included. Primary outcomes for systematic review were mean change in overall ETDQ-7 (or non-English validated form) and complication rate. The need to induce general anesthesia to complete the procedure or abortion was defined as a complication. Publication, cohort and data collection times were not restricted, nor were the clinical settings (outpatient clinic, hospital department). Studies that were not in English and those that involved nonhuman studies or were review articles, abstracts, or letters were excluded. Also excluded were studies that lacked descriptions of treatment protocols or treatment outcomes. (Supplementary II). The abstracts and articles of the identified papers were reviewed to determine which investigations met the above selection criteria for inclusion in this study. Two investigators independently extracted the original data (OJU and MDB).

After the enrollment process was completed, a manual review of all the references cited in the enrolled studies was performed in search of additional papers for inclusion. This process was performed to allow studies which had not been identified according to our search algorithm to be identified and assessed for suitability for inclusion and was proved to be an effective method for the identification of additional papers for systematic literature reviews in otology (45–48). A quantitative data sheet was constructed, and each relevant publication was analyzed regarding study design, demographic data, selection criteria, local anesthesia protocol, balloon used and outcomes. All publications were assessed independently by two coauthors for the risk of bias using ‘‘The Cochrane Collaboration's tool for assessing the risk of bias in randomized trials” (49) and “Newcastle–Ottawa Quality Assessment Scale criteria” (50), as shown in Supplementary III. Reconciliation of disagreements was performed by a third coauthor, who was blind to the other authors’ assessments.

The literature review yielded 101 publications that met our search protocol. After 39 duplications were removed, 62 publications remained for screening. An additional 8 publications were excluded due to language other than English. Two more papers were excluded because of non-human subjects and the absence of indicating a BDET under local anesthesia (1 each). Additional 45 papers were excluded because BDET was performed under general anesthesia or because it was impossible to extract the clinical data of the local anesthesia BDET group from the total cohort. The resultant 7 publications were accessed for full text. One paper was excluded because of the absence of original data, leaving 6 papers for inclusion in the systematic literature review (Figure 2).

Five of the included papers were case series and one randomized controlled trial. Altogether, 193 patients (262 ETs) were included. Overall gender distribution could not be assessed due to missing reports in 3 papers (15, 31, 32), but 47/101 (47%) were males from the papers enrolled. Averaged (±SD) age ranged from 42 (±13) to 57 (±14). Indications for BDET widely varied: 2 papers did not report BDET indications (32, 51); out of the other 4, combinations of symptoms and tympanogram, ETDQ-7 and tympanogram, respectively, ETDQ-7 and symptoms and ETDQ-7 alone were used. ETDQ-7 mean item score for ETD diagnosis was not uniform, ranging from 2.1 to 3.0. Demographics and BDET criteria of published case series and our newly published case series are shown in Table 1 for the ease of comparison.

Eligibility for local anesthesia was missing in all but one paper (52). In this paper, only patients who tolerated the pre-procedural endoscopic exam well, had straightforward nasal anatomy, appropriate body habitus and absence of potentially exacerbated medical condition were enrolled. 6 dilatory balloon systems were used, making a comparison between them impossible. Anesthesia protocol varied as well. Five studies used systemic analgesia/anxiolysis and topical anesthesia. Topical anesthesia alone was used in one study (53). The nasal cavities were most commonly anesthetized by cottonoids soaked with a combination of tetracaine or lidocaine + decongestant like adrenaline or pseudoephedrine. The ET orifice was anesthetized in 5 papers with lidocaine cream ± tetracaine or pilocarpine. The tympanic membrane was anesthetized in 2 papers with 7% Tetracaine/Lidocaine. The anesthetic protocol and dilatory system used are shown in Table 2.

The most common outcome measured used was a tympanogram (3 papers), followed by ETDQ-7 (2 papers). Non-otologic outcome measures alone (pain, vitals and willingness to choose BDET under local anesthesia again) were present in 2 papers. Some papers compared the outcome measures between local vs. general anesthesia, while others compared baseline to follow-up performance. As a result, pooling outcome is impossible (Table 3).

The results of a bias risk assessment for the non-randomized trials (5 trials, at least 112 ears) and randomized controlled trials (1 trial, at least 38 ears) show low bias risk in all included papers. Supplementary III displays the stratification of each bias category.

In our case series, no major complications were recorded. BDET is a safe procedure. In their systematic literature review, Randrup and Ovesen reported minor self-limited complications following BDET in 2% of 1,800 BDETs, the most common of which was mild epistaxis (58). They also identified one report of a patient who suffered post-procedural C6–7 radiculopathy, which was related to neck extension, and probably is not related to the procedure itself. The bony part of the ET and the internal carotid artery (IAC) share a common bony wall, making the posteromedial aspect of the ET bony-cartilaginous junction a landmark to the IAC first genu (59, 60). This common bony wall is so thin that it can be easily medialized during balloon inflation, compressing the carotid canal with associated anoxic brain damage (61). This situation is further complicated by the fact that carotid canal dehiscence (CCD) is not uncommon (>7%) and that near-dehiscence can be interpreted as CCD in high-resolution CT (62). There are several FDA approved dedicated DBET systems in the market. TubaVent and TubaVent short (Spiggle & Theis, Germany) is 20 mm long. XprESS devices (Srtyker Corporation, Michigan, US) have several dedicated balloons, 8 or 20 mm long. Acclarent Aera Balloon (Acclarent, Irvine, CA) manufactures 16 mm length balloon devices. CCD is of clinical significance in the petrous portion, more than 20 mm way from the ET meatus. This observation highlights the need to avoid over insertion of the device up the ET. Because the OETD resides in the cartilaginous part of the ETD (63), a selective cartilaginous BDET reduces the risk of carotid canal compression. This is the rationale behind the importance of direct visualization of the dilation device in the ET orifice and avoiding advancing the suction guide to the ET lumen, blocking the vision of the dilation device. A recent study, including 510 BDETs, with a CCD incidence of 6.3%, reported no CCD-associated BDET complications, thereby concluding that “fear of injury to the internal carotid artery during balloon dilatation might be disproportionate”. However, since BDET is performed as an elective procedure, for quality of life- anoxic brain damage can be devastating sequala. In this term, in-office BDET seems to be safer because the conscious candidate can report real-time focal motor and/or sensory dysfunction.

Recently, a case report of iatrogenic sudden sensorineural hearing loss occurred during BDET (64). Likely pathophysiological mechanism suggested an inner ear concussion, or inner ear membrane injury from barotrauma. Other report of extensive cervico-facial emphysema and/or pneumo-mediastinum (65–67) also emphasizes that balloon insertion must be performed under utmost care, while the depth and speed of inflation must be rigorously monitored to avoid a sudden pressure change and tearing of mucosa.

The present study is an extension of a prior cadaver study demonstrating a proof of concept using the EVB (35). In our series of 20 patients using the EVB [ref#36 and this study], the risk profile and outcomes are similar to a systematic literature review, in which 262 BDETs were performed under local anesthesia (52–57). We believe there is clinical equipoise between an EVB and other commercially available BDET products,

Several dedicated dilation devices for BDET are Class II FDA-approved. These single- use devices are priced at over USD$1,000. The economy of scale in the pricing of the EVB has driven the cost to approximately USD $ 170 per device. This combined with diminished operating room resource in performing the procedure under local anesthesia within the ambulatory care setting, there would be considerable savings regardless of the type of healthcare delivery system in the context of cost-effectiveness. This is important when considering repeating the procedure for recurrent symptoms.

The effectiveness of the EVB appears comparable to other commercially available, dedicated balloons. In our study, one patient (#5) did not improve objectively (tympanometry and TM movement under Valsalva). Patient #3 (left ear) did not recover subjectively and objectively (ETDQ-7 score, TM movement under Valsalva and tympanometry). This is congruent with the reported 70%–90% success rate of BDET performed under local anesthesia (56, 68). Under general anesthesia, 63%–80% report improvement of tympanogram, with inferior success rate when chronic serous or adhesive otitis media coexist (30). While in-office procedures can potentially reduce the success rate due to limited intra-operative patient cooperation, proper patient selection can mitigate this type of selection bias. However, the EVB was proved in our study and a previous one (35) to be as effective as the dedicated ET dilation systems under general and local anesthesia.

Only 2 studies evaluated patients' satisfaction with BDET under local anesthesia. Chen et al. (53) reported 96% willingness to choose local anesthesia again, including patients who underwent bilateral BDET. This observation is probably in line with Luukkainen et al. (55), who reported a low visual analog scale (VAS) for pain and discomfort (∼1 ± 0.4 and ∼1.3 ± 0.7 on a 0–10 scale). Our subjects reported a 100% (8/8) willingness to repeat the procedure under local anesthesia if needed. Overall, patient satisfaction is closely related to the local anesthetic protocol. There was a significant difference in the reported local anesthesia methods, with non-uniform tympanic membrane and nasal cavity anesthesia. Tympanic membrane anesthesia controls baroreceptor activation in the middle ear, mainly in the tympanic membrane. During BDET, middle ear pressure can increase to as high as 206 deka Pascal (16 mm of Mercury) during insertion and inflation and as low as 253 deka Pascal (19 mm of Mercury) during deflation (69), while an abrupt 50 deka Pascal (4 mm of Mercury) change is associated with significant discomfort in 14% of the population (70). Another issue is the inter-aural middle ear pressure gradient. While this gradient exceeds 500 deka Pascal (38 mm of Mercury), a common scenario in bilateral ETD, where the operated ear reaches a peak pressure during balloon inflation and the contralateral middle ear pressure is still in the negatives, alterno-baric vertigo can result (71, 72). While various vestibular suppressants can control this adverse effect (55, 73), we found that a slow inflation rate of 1 ATM/s is enough to control pain from abrupt pressure change and alterno-baric vertigo. Hereby making ototopical anesthesia and systemic vestibular suppressants- unnecessary. Slowly balloon inflation has the advantage of keeping the vaso-vagal reflex, resulting from rapid tympanic cavity inflation, un-activated, as reported in the previous series (57).