Balloon Dilation Eustachian Tuboplasty Combined with Tympanotomy Tube Insertion for Treatment of Refractory Otitis Media with Effusion: A Randomized, Prospective, Controlled Trial Comparing Local Anesthesia Combined Sedation Versus General Anesthesia

Background

Refractory otitis media with effusion (ROME) is a common disorder in ENT clinics. It is characterized by aural fullness and hearing loss, and it often leads to tympanic atelectasis and adhesive otitis media. In adults, it has a low prevalence (approximately 0.55–1%) [1]. Most ROME cases in adults are due to eustachian tube dysfunction (ETD) [2]. Traditional treatment options for ROME include medicines such as antibiotics, nasal steroids, and decongestants, physical training such as Valsalva maneuver, or surgical processes such as tympanic paracentesis (TP) or placement of a grommet to treat ETD [3]. However, these therapeutic approaches are not effective.

Balloon eustachian tuboplasty (BET) under general anesthesia is a therapeutic option with promising short-term results for ETD. It is a minimally invasive, simple, and safe approach [4–8]. Good treatment outcomes for ROME using BET and tympanostomy tube insertion (TBI) have been reported. BET+TBI was initially performed under general anesthesia [4]. However, there has been increased interest in performing the procedure in office settings under local anesthesia, and the feasibility of such an approach has been reported in recent years with varying local anesthesia protocols. However, there is a need to improve the technique [9,10]. In some areas, BET and TBI are still mainly performed under general anesthesia. Catalano et al first established the feasibility of BET under local anesthesia in 2012 [11]. Performing BET under local anesthesia with monitored anesthesia care (MAC) instead of general anesthesia confers the advantages of improved patient safety, faster patient recovery, and reduced treatment costs [12].

This study aimed to compare the clinical effects, safety, and cost-effectiveness of BET combined with TBI on ROME under local anesthesia with MAC vs general anesthesia.

Material and Methods

STUDY PARTICIPANTS:

This was a randomized, prospective, controlled trial. In total, 40 adults (age range=18–56 years) – 23 males (28 ears, including 5 binaural cases and 18 monaural cases) and 17 females (20 ears, including 3 binaural cases and 14 monaural cases) – were enrolled from December 2020 to November 2021 at the Affiliated Hospital of Jiangnan University. Patients with ROME diagnosed at least 6 months earlier were randomized based on computer software-assigned numbers allocated into 2 groups: the BET+TBI under local anesthesia group (Study group, 20 patients with 25 ears, including 5 binaural cases) and the BET+TBI under general anesthesia group (Control group, 20 patients with 23 ears, including 3 binaural cases). Eustachian tube dysfunction was diagnosed with tympanometry (TMM), the 7-item eustachian tube dysfunction questionnaire (ETDQ-7), and visual analog scale (VAS) with a 0–10 scale (0, no discomfort; 10, worst possible discomfort) [13]. Differences in gender, age, and disease duration since diagnosis between the groups were insignificant. This study was approved by the Ethics Review Board of the Affiliated Hospital of Jiangnan University (Ethics Approval Number: 2020023). All patients provided a signed informed consent form before enrollment.

INCLUSION AND EXCLUSION CRITERIA:

The inclusion criteria for this study were:

The exclusion criteria were:

OPERATION PROCEDURES:

For surgical procedures under local anesthesia with MAC, all patients were administered 15 mg/kg (1 g maximum) paracetamol according to their body weight before the procedure. The upper-limb venous access was opened and intravenously perfused with dexmetomidine at 1 μg/kg/h during the procedure for sedation. For patients experiencing unbearable pain, 5 ml of Dezocine was intravenously injected. For nasal nerve block, the nasal cavity was filled with cotton sheet supplemented with 3 ml of 0.1% adrenaline and 20 ml 1% tetracaine for about 15 to 20 min. The filling sites were the middle nasal meatus, olfactory sulcus, meatus nasi communis, nasopharynx, inferior nasal meatus, and nasal septum mucosa, among others. For local anesthesia and lubrication, 1 ml of lidocaine-prilocaine cream was applied to the eustachian tube lumen under endoscopy for 5 min. The endoscope was inserted into the ipsilateral nostril with the balloon dilatation catheter. If the nose was too narrow for both the endoscope and balloon dilatation catheter to pass, a 45° angulated endoscope was inserted in the contralateral nostril to visualize the eustachian tube orifice. Surgical techniques for BET+TBI were performed as previously described [14]. A balloon catheter (Carefree, Guangzhou Manxiang Medical Equipment Co., Ltd., China) was inserted into the ET with endoscopic assistance (0- or 30-degree view angle). The balloon was inflated with sterile water with a pressure of 10 bars for 2 min. After termination of inflation, the deflated balloon catheter was removed and the catheter tip was positioned correctly. After the procedures, the pharyngeal ostium of the eustachian tube was observed to be smooth and the openings enlarged without breakage (Figure 3). Then, the patients were instructed to turn the head to one side and, while keeping the affected ear up, perform the Valsalva maneuver. If entire tympanic atelectasis was restored when assessed by optic endoscopy, BET was proven effective and the patient would then undergo the TBI procedure (Figure 4). For general anesthesia, TBI was performed before BET and without the Valsalva maneuver.

QUESTIONNAIRE:

Once patients returned to the ward after the operation, they were asked to complete a questionnaire evaluating their operation experiences. The questionnaires were administered when patients felt comfortable. All patients were followed up at 1, 3, 6, and 12 months after the operation to observe the ventilation and tympanic membrane. ETDQ-7 results and VAS scores were recorded at each follow-up.

STATISTICAL ANALYSIS:

All analyses were performed using SPSS 22.0 software. The t test and LSD test were used for comparison of outcomes between the groups and for multiple comparisons, respectively. Changes in outcome between the preoperative period and postoperative period were assessed using the mixed effects regression model. P<0.05 were considered to be statistically significant.

Results

All patients successfully completed the BET+TBI procedure without any intraoperative anesthesia accidents, such as cardiopulmonary receptor reflex or marked fluctuations in blood oxygen levels, heart rate, or blood pressure. No difference in bleeding or abnormal patency of the eustachian tube or any other complications were observed between the groups. BET+TBI under local anesthesia with MAC was associated with higher discomfort compared with the general anesthesia group (mean VAS score=5.2±0.6 vs 0, P<0.0001) (Table 1), but had less postoperative discomfort when patients were awake (3.5±0.8 vs 2.3±0.8, P=0.049). About 90% and 95% of patients in MAC and general anesthesia groups, respectively, stated that anesthesia and pain relief were sufficient, with no significant difference between the groups. One patient reported discomfort under local anesthesia but no discomfort under MAC.

Three months after surgery, ETDQ-7 scores and R-values of TMM were significantly improved in both groups compared with preoperative scores (P<0.01) (Tables 2, 3). However, ETDQ-7 scores and TMM results between the 2 groups were not significantly different (P>0.05) (Figure 5, Tables 2, 3). Twelve months after the procedure, all tympanic membrane catheters were shed, and all tympanic membranes were integral under the otoscope (Figure 6).

The average total duration of the operation under local anesthesia was 49±17.26 min, whereas the total duration of anesthesia and operation in the general anesthesia group was 125.7±17.36 min, which significantly shortened the operation time (P<0.0001) (Table 4). In addition, the hospitalization costs in the local anesthesia group were lower than those in the general anesthesia group (P<0.0001) (Table 4).

Discussion

Since Catalano et al successfully performed BET under local anesthesia in 2012, this procedure has gradually become popular, but general anesthesia is still being used in many places [11]. In this study, we found that BET can be safely performed with MAC at a pressure of 10 to 12 atm and a dilatation time of 2 min. Our results are consistent with those reported by Veera et al, who reported that BET under MAC can be safely and effectively performed [15].

In this study, patients underwent surgery under local anesthesia while still awake, and reported mild nervousness and discomfort. However, since they were awake, they could communicate their discomfort for timely interventions during the operation. In this study, 19 patients in the local anesthesia with MAC group reported that the procedure provided adequate pain relief, whereas 1 patient reported the operation was uncomfortable during a retrospective review of the procedure. Further, all patients in the local anesthesia with MAC group reported complete relief from all discomfort within 1–2 h after the procedure. However, the experience of pain or discomfort during the operation in the local anesthesia group compared with the general anesthesia group suggests the need to improve our anesthetic techniques when performing eustachian tube and lumen procedures. Due to being awake during the operation, pain and discomfort could be generated from mucosal sensory, mechanical, or stretch receptors within the lumen of the cartilaginous eustachian tube (ET) or middle ear. This neuronal reflex arc also has the potential to trigger vasovagal responses [16]. Therefore, to reduce pain, Joonas et al suggested the rate of inflation not exceed 1 atm/s [17].

Compared with the general anesthesia group, postoperative VAS scores were slightly lower in the local anesthesia group. Under local anesthesia with MAC, the intraoperation procedures were tolerable, whereas the postoperative pain gradually increased after the analgesic effects subsided. Therefore, VAS scores after surgery were slightly but not significantly different between the 2 groups (P>0.05). Blood loss in both groups was less than 5 ml, and differences in postoperative ETDQ-7 scores between the groups were insignificant (P>0.05). Moreover, differences in short-term efficacy between the groups were insignificant, suggesting that anesthesia was not the main factor affecting the operation results. The procedure under general anesthesia was performed for about 120 min, which was longer than the 40 min under local anesthesia. Therefore, treatments under local anesthesia take less time and are associated with fewer postoperative complications such as nausea and vomiting [18].

Regarding the order of BET and TBI during the procedure, some operators believe that TBI should be performed before BET [19]. The reasoning behind this view is that removing the effusion or secretion in the tympanic cavity may balance the pressure between the external and middle ear. The middle ear could be in a negative pressure state caused by ETD, and if BET is performed first, the pressure in the middle ear cavity will change sharply when the balloon catheter is expanded in the eustachian tube lumen. However, whether performing TBI first followed by BET will affect the inner ear is still not clear. In this study, BET was performed first, followed by TBI, in the local anesthesia group. During the dilation process, the patient was clearly conscious and instructed to perform the Valsalva maneuver, which could indicate the patency of the eustachian tube.

Treatment costs are an important consideration in modern health practice. Surgery under local anesthesia with MAC, rather than general anesthesia, usually saves costs [20]. In this feasibility study, local anesthesia was performed in the operation room for monitoring. The anesthesiologist was on stand-by and constantly monitored the patients prepared for sedation. From the beginning of local anesthesia to the end of the operation, the average time consumed was about 49±17.26 min. Patients in the local anesthesia group experienced inapparent adverse reactions; therefore, anesthesia should be done in the preparation room before the patients enter the operating room. This will reduce the anesthesiologist’s time in the operating room by about 20 min, thereby reducing treatment costs [21]. Based on our findings, we believe that patients under the current local anesthetic scheme experienced some pain and discomfort, and some of them required intravenous drugs to relieve pain. Therefore, we do not recommend surgery without an intravenous approach.

The number of patients in this preliminary study was limited. While the local anesthesia group had no significant adverse reactions, we cannot rule out intraoperative pain and possible vagus reactions due to significant changes in heart rate or blood pressure. A patient was reported for the first time to experience bradycardia (heart rate of 30 beats per min) during BET under local anesthesia at the University of Helsinki hospital, which was relieved by intravenous atropine administration [12]. Therefore, we recommend proper intraoperative monitoring and intravenous access for BET under local anesthesia until more patients are safely treated and anesthetic procedures are improved. In addition, the high VAS score in the local anesthesia group indicated that some patients could still sense pain during the operation, suggesting that local anesthesia and operation approaches need further improvements.

Conclusions

BET+TBI under local anesthesia with MAC for treatment of refractory otitis media with effusion is feasible and safe. However, under the current program, patients still feel slight pain and discomfort, indicating the need to improve it. Clinically, BET is a potentially effective therapeutic approach, but further investigations are needed to improve pain control and vagal reactions.