International Multi-institutional Experience with the Transoral Endoscopic Thyroidectomy Vestibular Approach

Introduction

Surgical approaches to thyroidectomies have undergone a rapid evolution over the past three decades. Either as a result of spontaneous ingenuity or born from cultural demands to avoid cosmetically displeasing scars, less invasive approaches to thyroidectomies have been developed. ¹ The traditional open, trans-cervical thyroidectomy has been proven to be safe and efficacious for approaching a myriad of thyroid and cervical pathologies. ² Attempts to limit incision size for better cosmetic outcomes have ultimately been futile—given the fact that skin violation, however small, ultimately leads to scarring regardless of the length of the incision. In certain populations that scar can be more disfiguring with the development of keloids (due to genetic predispositions), or else can cause emotional distress.^1,3

The surgical techniques developed in an attempt to avoid neck scarring have included trans-axillary technique; bilateral axillo-breast approach; and retro-auricular (facelift) approach, among others. The ultimate goal of all of these approaches was to replace the trans-cervical incision with small incisions in remote, inconspicuous locations. Early on, these procedures were coined minimally invasive approaches. This was found to be a misnomer because these approaches require significant tissue dissection, can have long distances from the port sites to the thyroid gland, and because the unconventional incision sites can end up being cumulatively longer than the 3–5 cm trans-cervical incision. They have since been more aptly termed “remote access techniques” or just “extra-cervical approaches.”^1,4,5

The transoral approach is also categorized as natural orifice transluminal endoscopic surgery.^4,6,7 The thyroid gland is approached via lower lip incisions in the oral cavity, with the rest of the procedure occurring in the subplatysmal space of the neck. The first foray into the transoral technique was via the sublingual approach.^8,9 This was quickly abandoned due to technical limitation and high complication rates. The latest and most widely applied technique is the transoral endoscopic thyroidectomy vestibular approach (TOETVA).^1,10 TOETVA applies the benefits and avoids the drawbacks of all other remote access procedures. There is less tissue dissection and distance to reach the target gland, and most importantly, there is complete avoidance of a skin incision (with incisions hidden in vestibular mucosa).

The transoral approach was first described in 2013 and has since then been done at select international institutions, and multiple high volume centers in the United States. To this point there have been over 1500 TOETVA procedures performed worldwide.^4,10,11 The purpose of this article is to present the results of TOETVA from five different international institutions: United Hospital-Allina Health (St. Paul, MN), Bellvitge University Hospital (Catalonia, Spain), Swiss Medical Network (Clinique de Genolier and Clinique Valére, Switzerland), Keelung Chang Gung Memorial Hospital (Keelung, Taiwan), and Mount Sinai Hospital (New York, NY).

Materials and Methods

This study presents 152 TOETVA cases performed from 2016 to 2019 on 149 patients from five different international institutions. The discordancy between cases and patient number represent three cases of completion thyroidectomy. Outcomes were analyzed from prospectively maintained databases. All research was conducted after approval was obtained from the institutional review boards. All TOETVA cases were performed with the same technique as previously described.^2,7

The inclusion criteria were uniform among all five institutions and included the following: patient-motivated desire to avoid a cervical scar, symptomatic benign nodules ≤6 cm, cytologically indeterminate nodules (Bethesda III or IV lesions) <6 cm, estimated thyroid diameter ≤10 cm on ultrasound, estimated gland volume ≤45 mL on ultrasound, symptomatic Hashimoto's thyroiditis, Grave's disease, and differentiated thyroid cancer <3 cm without extra-thyroidal extension or lymph node metastasis appreciated on preoperative ultrasonography. Patients were excluded if they were unfit for surgery, unable to tolerate anesthesia, had sub-sternal goiters, or had previous neck surgery and/or radiation. In contrast to the inclusion criteria there was no uniformity in the perioperative antibiotics used, or in the length of time that they were administered for. Table 1 lists the varying antibiotics used, and the length of time.

Of the 149 patients who underwent a TOETVA, 21 (14.1%) patients were preoperatively found to have thyroid malignancy, 16 (10.7%) patients had symptomatic benign disease, and 112 (75.2%) patients had singular or multinodular disease with varying Bethesda classifications. For the purposes of presenting our results we designated a group name for the different institutions: United Hospital-Allina Health (UHAH)—Group A, Bellvitge University Hospital/Swiss Medical Network (BUH/SMN)—Group B, Keelung Chang Gung Memorial Hospital (KCGMH)—Group C, and Mount Sinai Hospital (MSH)—Group D.

We assessed for significant differences in mean outcomes measured between the two groups by using paired t-tests to generate a two-tailed P-value. One-way analysis of variance calculations were also utilized. Statistical significance was defined as a P-value less than .05. Calculations were performed with LaTeX.

Results

The patient population included 12 (8%) men and 137 (92%) women with a mean age and range of 41.5 ± 10.3 (27–69) and 46.8 ± 1.8 (17–78), respectively. The mean body mass index (BMI; kg/m²) was 27.7 ± 4.8. Across all institutions the mean thyroid nodule size was 29.1 (0.1–88 mm). Table 2 illustrates the variability in patient demographics encountered by the five international institutions.

One detail that we have highlighted is the differences in mean BMI. In groups A, B, C, and D the mean BMIs were 33.2 ± 11.2, 20 ± 2, 24 ± 3.7, and 26 ± 5.7, respectively. The differences in the mean BMI were found to be statistically significant between all five groups (P < .05). The entire sample size was then stratified according to BMI greater than or less than 30—there were 20 (13.4%) patients with a BMI ≥30 and 129 (86.6%) patients with a BMI <30. When the mean operative time (for lobectomies and total thyroidectomies) was compared between these two groups, there was no statically significant difference found (P < .05). With respect to postoperative complications the obese group (BMI ≥30) had 0 skin injuries, 17 (85%) transient nerve injuries, 0 persistent deficits secondary to nerve injury, and 2 (10%) cases of transient hypoparathyroidism. The group with a BMI <30 had 3 (2%) skin injuries, 82 (53.9%) transient nerve injuries, 1 (0.7%) persistent deficit secondary to nerve injury, and 5 (3.3%) cases of transient hypoparathyroidism.

A lobectomy (or hemi-thyroidectomy) was performed in 111 (73.0%) cases, a total thyroidectomy in 38 (25.0%) cases, and a completion thyroidectomy in 3 (2.0%) cases. The mean operative time was 161.8 ± 42.4 (83–304) minutes for the lobectomy (hemi-thyroidectomy), 213.4 ± 71.7 (120–430) for the total thyroidectomy, and 136.7 ± 109.8 (64–263) minutes for the completion thyroidectomies. The mean operative times (in minutes), for lobectomies (or hemi-thyroidectomies), of each institution were compared. There was a statistically significant difference between all institutions, with respect to mean operative time, except when comparing groups B and D (P < .05). The mean operative times (in minutes) for total thyroidectomies of each institution were also compared. There was a statistically significant difference between all institutions, with respect to mean operative time, except when comparing groups A and B (P < .05). The mean estimated blood loss was 18.6 ± 50.8 mL and there were 3 (2.0%) cases that required intraoperative conversion to an open procedure. The operative details of the individual international institutions are detailed in Table 3.

The average length of stay (LOS) in groups A, B, C, and D were 0.15 ± 0.37, 2.5 ± 0.82, 2.6 ± 0.77, and 0.2 ± 0.56, respectively. When comparing the mean LOS between institutions, there was a statistically significant difference except for when comparing Group B to Group C and Group A to Group D (P < .05). The duration of antibiotic therapy, peri- and postoperatively, varied between institutions. The mean duration (in days) of antibiotic therapy was 8.9 ± 1.5, 7 ± 0, 1 ± 0, and 1.1 ± 2.3 in Groups C, B, A, and D, respectively. The differences between the means were found to be statistically significant between all groups except for when comparing Group A and Group D. The mean length of time to follow-up (in months) was 5.5 (0.25–24).

There were 7 (4.7%) patients who developed postoperative transient hypoparathyroidism. Five (3.3%) patients developed transient recurrent laryngeal nerve (RLN) injury while 3 (2%) had persistent deficits secondary to injury to the RLN. Lower lip numbness was present in 51 (33.6%) patients lasting on average 9.2 weeks, with only 1 (0.7%) developing persistent lower lip numbness, beyond 6 months. Fifty-seven (38.5%) patients reported chin numbness with an average of 8.1 weeks to recovery. Most, if not all, patients who underwent the TOETVA procedure exhibited different degrees of postoperative ecchymosis/bruising in the overlying skin, but every case resolved within one to two weeks after the procedure. There were a total of 9 patients who suffered skin injuries during the procedure, ranging from skin ulceration to full-thickness perforations. There were 2 (1.4%) instances of significant bleeding, 1 of which required conversion for adequate control. One (0.7%) patient developed a postoperative infection. We also report 2 (1.4%) cases of tracheal perforation. The details of the intraoperative events and complications are outlined in Table 4.

Discussion

The TOETVA is a promising new technique for the removal of benign and malignant thyroid disease. There are several factors limiting widespread applicability of this novel technique. The most important has been the need to confirm the safety and efficacy, relative to the traditional trans-cervical approach. Thus far, multiple studies have shown that TOETVA can be performed safely and effectively—in select patients, with specific features.^3,10–12 Another factor that has limited expansion of its applicability has been the steep learning curve. This technique is technically demanding and requires expertise in conventional thyroid surgery, laparoscopic and endoscopic procedures—as prerequisites for safe introduction.^3,13,14 With that being said, TOETVA is currently being performed internationally, at multiple institutions in Asia, Europe, South American, and the United States.^1,3

A wide array of prophylactic antibiotics has been used for TOETVA. Just among the different institutions included in this study, no one utilized the same single and/or combination of antibiotic(s). The likely reason for the lack of uniformity is simply that there has been no guidelines set forth to determine which antibiotic(s) provides adequate prophylactic coverage. The mucosa of the oral cavity is colonized with a very diverse bacterial population (e.g., gram-positive aerobes/anaerobes and gram-negative aerobes/anaerobes). In the literature, there had been no reported cases of surgical site infection, with the oral vestibular approach, until this past year.^1,2,7,13 Because of the violation of the oral mucosa, TOETVA wounds are classified as clean contaminated.^15,16 The common thread between the antibiotics used, between all five institutions, was their ability to cover a diverse population found in the oral cavity. Length of time on antibiotics was another highly variable factor, with significant differences being found between the institutions, except when comparing the two in the United States (i.e., UHAH and MSH). Reasons for variation could be differences in cultural practices versus differences in hospital policies for coverage of regionally based bacterial resistance. There was a single postoperative infection in our cohort, an infected hematoma that required incision and drainage and treatment with intravenous antibiotic coverage—one month postoperatively. She received one preoperative dose of clindamycin and perioperative steroids for her coexisting comorbidities. We suspect that she developed a postoperative hematoma that became infected due to her transient immunocompromised state (secondary to perioperative steroid regimen and/or persistent smoking during the postoperative period).

Of the three conversions to an open procedure, one was to control bleeding. The bleeding continued to obscure the surgical field, making visualization of vital structures difficult, therefore the decision was made to convert. The other case with significant bleeding was secondary to an injury to the anterior jugular vein (with upward of 500 mL blood loss), but bleeding was adequately controlled without needing to convert. The other two cases requiring conversion were due to the size of the nodules (multinodular goiter) and overall gland size (Grave's thyroiditis). Due to size it was difficult to obtain adequate working space via the transoral route.

There were similar and overlapping intraoperative events and complications from all five international institutions. Complications that are new to surgeons performing thyroidectomies (particularly to those performing this new approach) have been the skin injuries and manifestations from mental nerve injury. The most commonly encountered skin lesion, in the immediate postoperative period, is ecchymosis/bruising of the chin and the anterior neck.^6,17 Although this is not so much a complication as it is a frequently noticed transient postoperative finding. The other skin lesions (i.e., skin abrasions and full thickness skin perforations) occurred during the creation of the subplatysmal working space. Lower lip and chin numbness, secondary to mental nerve injury, is a known complication of patients undergoing TOETVA. It was hypothesized that the course of the mental nerve could be avoided by repositioning the two vertical incisions for the 5 mm ports to the vestibular mucosa lateral to the level of the canines and just in the inner aspect of the inferior lip. Evaluation to determine whether incidence of lower lip and chin numbness/paresthesia has significantly decreased, as a result of this adjustment, still needs to be determined. Of the 152 cases in our series, 8 (5.3%) patients developed either transient or persistent (deficits lasting greater than 6 months) RLN injury. Other remote-access approaches have been associated with higher incidence of RLN injury and greater complications. ¹⁰

To the patients the greatest aspect of the TOETVA procedure is that it will result in no skin scarring, but to the surgeon it can be the improved visualization of the cervical anatomy. First, this approach allows excellent visualization of bilateral RLNs, as they insert into the larynx. Second, the bird's-eye view the camera provides can also present a more favorable angle of dissection along the plane of the nerve. Last, the limited soft tissue dissection needed for this particular remote access technique could prove useful in patients with obesity. ¹⁰

There were clinically and statistically significant mean BMI differences found among all five institutions. At UHAH (Group A) the mean BMI was greater than 30 (where obesity is defined as a BMI greater 30) while the rest of the institutions were below 27. The presence of obesity (BMI >30) had no significant impact on outcomes, with respect to efficacy or safety (i.e., operative time or complication rate). Therefore, in patients with obesity and significant soft tissue in the neck, along with obesity related to conditions that could affect wound healing, could benefit from the alternative route TOETVA provides.

Conclusion

TOETVA is currently the latest remote access option for the treatment of benign and malignant thyroid disease. It is a viable alternative to the traditional thyroidectomy, for highly motivated patients seeking cosmetically pleasing results. To date, the literature and the outcomes from these five international institutions have determined that TOETVA can be safe and efficacious for the treatment of specific thyroid pathologies. Uniform guidelines for which prophylactic antibiotic is to be used and length of time on it still needs to be determined. Prospective randomized control trials are necessary to determine whether adjustments in lateral port placement have ultimately had any effect on the incidence of mental nerve injury.

Longer operative times and the steep learning curve of this procedure have been the most common critiques. Mainstream applicability is only limited by a lack of understanding of the approach and lack of patient awareness that a cosmetically appealing alternative to the traditional open technique exists. As demand for this “scar-less” technique increases, so will volume, leading to greater experience with resultant decreased operative times and learning curves over time.