Discussion
This meta-analysis included four prospective and four retrospective studies, with a total of 330 patients, comparing the retroauricular hairline and conventional approaches for submandibular gland resection. The retroauricular hairline approach was found to be safe, with better cosmetic results and similar risk of postoperative nerve damage compared to the conventional approach. There was no significant difference in the intraoperative bleeding volume, postoperative drainage, and length of hospital stay between the two approaches.
Minimally-invasive surgical techniques emerged in the 1980s, and are now used in various surgical specialties, including abdominal, thoracic, and gynecological surgery. In recent years, head and neck surgery has made great strides in minimally-invasive surgery, aiming for improved esthetics, reduced blood loss, shorter hospital stays, and fewer complications23, 24. Technological developments and the demand for improved esthetics have changed the choice of surgical modalities for submandibular gland removal. The patients also expect perfect cosmetic results after submandibular gland surgery. Submandibular gland removal through the transcervical approach may result in unsightly keloids or hyperplastic scarring. This has led to the development of various techniques, including intraoral and retroauricular approaches, to reduce scar formation. Traditional submandibular gland resection is performed through a transcervical approach, which is a relatively safe procedure, but can lead to postoperative neurological complications. In young individuals, the scar is visible on the exposed part of the neck, which is cosmetically undesirable, particularly in cases of keloid formation. This places a tremendous burden on the patients and can have negative psychological and social effects. Based on the cosmetic requirements, minimally-invasive approaches for submandibular gland removal, including the transoral endoscopic approach25, smaller incisions combined with endoscopic assistance, and retroauricular hairline approach, have been investigated in the recent years. Intraoral submandibular gland surgery was described several years ago, but had limited applicability. Although intraoral surgery results in perfect cosmetic outcomes, it requires a higher level of surgical expertise, and the narrow surgical field can make it difficult to ligate the facial artery. This has limited the wider application of this approach, but it has been reintroduced in recent years with the development of robot-assisted surgery [26]. Despite the reduced length of neck incision, an important disadvantage of small neck incisions combined with endoscopy is the high level of surgical expertise required to handle the endoscope. Surgeons must be accustomed to endoscopic surgical views, and be able to handle a wide variety of instruments, including endoscopes, grasping forceps, harmonic scalpels, etc. In terms of cosmetic results, the posterior auricular approach offers significant advantages compared to conventional submandibular gland resection.
This meta-analysis demonstrates the feasibility of retroauricular hairline approach for submandibular gland resections. Although the retroauricular incision line is approximately 4–6 cm longer than the traditional transcervical incision, a “scarless” effect is achieved because the scar is hidden behind the ear and in the hairline, especially when covered by long hair. This meta-analysis compared the operative times, wound lengths, intraoperative bleeding volumes, postoperative drainage, durations of hospital stay, and postoperative nerve injuries between the postauricular hairline and conventional approaches for submandibular gland resection. The operative time and wound length were longer in the postauricular group compared to the conventional group. However, there were no statistically significant differences in the intraoperative bleeding volumes, postoperative drainage, and lengths of hospital stay between the approaches. The retroauricular hairline group had a higher risk of earlobe numbness, but this resolved during the follow-up period. There were no differences in the risk of damage to other nerves between the two approaches. The present meta-analysis had several limitations. First, half of the included studies were cohort studies; further prospective randomized controlled studies are needed to increase the level of evidence. Second, the differences in the study populations, surgical skill levels, and other factors could have contributed to heterogeneity among the studies. In the present study, the heterogeneity for continuous variables was greater than that for dichotomous variables, so we performed subgroup and sensitivity analyses. The subgroup analysis for the postauricular hairline group was performed by dividing the procedures into open, endoscopic, and robotic surgeries. There was significant heterogeneity in the surgical duration among the open, endoscopic, and robot-assisted retroauricular hairline submandibular gland resection subgroups. The heterogeneity was greater when the three groups were combined (I2 = 99%; Fig 2), indicating that grouping factors may not be the source of heterogeneity. In terms of wound lengths (I2 = 99%), there was no heterogeneity in the subgroup analysis for the endoscopic retroauricular hairline approach (I2 = 0%; Fig 3). This indicates that grouping factors may be the source of heterogeneity. Comparing the lengths of hospital stay between the two approaches (I2 = 89%; Fig 6), subgroup analysis revealed no heterogeneity for the robot-assisted retroauricular hairline submandibular gland resection (I2 = 0%). However, the heterogeneity was greater in the endoscopic group, indicating that subgroup factors may be the source of heterogeneity. Meanwhile, intraoperative bleeding volumes and postoperative drainage were more heterogeneous, but could not be analyzed in subgroups because of insufficient data. This was another important limitation of this meta-analysis.
This study showed that submandibular gland excision by the retroauricular hairline approach had significantly higher postoperative cosmetic outcome scores than the traditional approach; the retroauricular hairline group patients under the age of 30 years had four times higher postoperative cosmetic outcome scores than the traditional group21. This difference may be significant for patients with a cicatricial constitution. The retroauricular approach was also safer than the traditional cervical incision because no large neck vessels become exposed if sutures are lost, leading to necrosis or rupture, which prevents the risk of catastrophic bleeding27. However, surgeons may be uncomfortable with the top-to-bottom anatomical orientation for the retroauricular approach, and the limited access and visibility for anterior lesions. Although endoscopic surgery provides a magnified of the narrow surgical space, the assistant must keep the endoscope in place and move it in concert with the surgeon’s movements. This requires a close and skilled coordination between the assistant and surgeon. Robotic submandibular gland resection solves this problem as the surgeon controls the two robotic arms remotely from a console to perform the procedure. However, robotic and endoscopic submandibular gland resections were found to have comparable surgical and cosmetic outcomes28. With the advent of artificial intelligence, robotic surgery is emerging as a viable alternative to traditional open surgery, with potential safety and esthetic advantages. De Virgilio proposed that the first step in robotic submandibular gland resection through the retroauricular incision was to identify and ligate the facial artery and other major arterial and venous branches. This minimizes the risk of bleeding and improves visibility, further demonstrating that a large incision does not necessarily lead to poorer postoperative outcomes10. The robotic retroauricular hairline approach reduces the incidence of surgical complications associated with the postauricular anatomy because it does not require the creation of a surgical tunnel and postauricular flap elevation. It also reduces the risk of postoperative complications, including seromas and hematomas, caused by flap elevation. Because the robotic system is equipped with a magnifying camera and dexterous instruments, the gland boundaries are clearly identifiable, which facilitates the submandibular gland stripping and excision29. However, the use of robotic systems is technically demanding. The advantages of the retroauricular hairline approach are based on a smaller incision along the hairline. The large robotic arms of the currently available systems make the docking process critical, which makes the surgical technique and placement of robotic instruments more difficult. Without proper arm positioning, frequent collisions of surgical instruments may occur, making the procedure more time consuming and difficult30. The robotic systems are expensive and not widely available, particularly in developing countries. Although cost comparison was not done in the present study, it may be a deciding factor for the clinical application of robotic surgeries. Considering the economic realities of most countries, the endoscopic technique may be a more feasible alternative to conventional surgery for the optimization of esthetic and functional outcomes in selected head and neck tumor patients.