The frequency of benign and malignant thyroid tumors has experienced a significant rise in the last 20 years in the Mainland of China, partially due to an increase in public awareness, utilization of diagnostic tools and routine physical examinations. Conventional thyroidectomy (ConT) through a skin-crease incision in the anterior neck has been the proven, efficacious surgical method for treating various thyroid tumors for the past few decades. ConT requires a cervical collar incision, leaving an unsightly scar. However, numerous patients with thyroid tumors who require operation are young women and they are often concerned about the cosmetic appearance of the scar. Consequently, the demand for surgical procedures which avoid visible scars while maintaining optimal functional and ideal cosmetic results is increasing.

The goals of endoscopic thyroidectomy (ET) are to limit external scarring and improve cosmesis, to reduce postoperative pain, to enhance postoperative recovery and to achieve these ends without compromising treatment efficacy. ET was first described by Hüscher et al (1) in 1997. Most of the earlier endoscopic approaches to the thyroid gland employed small cervical incisions in the midline or laterally. To avoid a visible scar in the neck, non-cervical approaches to the thyroid gland have been employed. The most commonly used non-cervical approaches are the axillary (2), the breast (3), the lateral (4) and also certain hybrid approaches (5), such as the axillary-bilateral breast approach, the bilateral axillo-breast approach, the unilateral axillo-breast approach, and the postauricular and axillary approach. Among these, the breast approach has emerged as the clear favorite as evidenced by its exponential growth in number in the Mainland of China (6).

It may be due to the Chinese culture that avoidance of a cervical neck scar justifies the expense of technological investment combined with a lengthy procedure to approach the thyroid compartment remotely, which has promoted the widespread application of ET in the Mainland of China (6). The preponderance of females among patients requiring thyroid surgery has probably been another influence in the development and patient acceptance of ET. In our department, ET via breast approach was developed in 2007 and, since then, approximately 300 operations have been performed. The aim of this retrospective study was to compare the surgical results of patients who underwent ET and to those of patients who underwent ConT. Herein, we summarize the results of the ET and demonstrate its feasibility and safety via breast approach for the treatment of benign thyroid tumors.

Two hundred and eighty-five patients underwent ET via breast approach (ET group), and 352 patients underwent ConT (ConT group). The mean ages of the patients were 35.7±12.59 years (range 17–55) in the ET group and 40.0±11.14 years (range 22–68) in the ConT group (P<0.001). The male-to-female gender ratios were 10:275 (1:27.5) in the ET group and 42:310 (1:7.38) in the ConT group (P<0.001). A unilateral lobectomy was performed in 126 patients (44.2%) of the ET group and in 163 patients (46.3%) of the ConT group. A bilateral total thyroidectomy was performed in 159 patients (55.8%) of the ET group and in 189 patients (53.7%) of the ConT group. Patient characteristics are presented in Table I.

The total operative time for the ET group was significantly longer compared to the ConT group (79.9±20.10 vs. 45.4±11.90 min, P<0.001, for unilateral lobectomy; and 89.9±14.60 vs. 60.0±8.44 min, P<0.001, for bilateral total thyroidectomy) (Table I). The length of postoperative hospital stay of the ET group was significantly longer compared to that of the ConT group (5.5±0.50 vs. 5.3±0.75 days, P=0.002). There was no case of conversion to the conventional open surgery in the ET group. RLN and parathyroid glands were identified in all cases (Table I).

An endoscopic approach provides better, magnified views of the RLN and the parathyroid glands resulting in fewer complications. The postoperative complications of both groups are shown in Table III. Transient hypocalcemia was found in 1.4% (4/285) of the ET group and 4.3% (15/352) of the ConT group cases. No permanent hypocalcemia occurred in the ET group, but one occurred in the ConT group. Transient RLN palsy occurred in 3.2% (9/285) of the ET group and 3.7% (13/352) of the ConT group. No permanent RLN palsy occurred in each group. Postoperative pain in the ET and ConT groups occurred in 4.6% (13/285) and 4.3% (15/352), respectively, at 24 h after operation.

To evaluate the cosmetic results, a simple questionnaire was used (Table IV). Among the 282 responses in the ET group, 78.7% (222/282) were ‘Excellent’, 13.9% (39/282) were ‘Good’, 7.4% (21/282) were ‘Acceptable’ and there were no ‘Poor’ responses. Among the 352 responses in the ConT group, 7.4% (26/352) were ‘Excellent’, 51.4% (181/352) were ‘Good’, 37.5% (132/352) were ‘Acceptable’ and 3.7% (13/352) were ‘Poor’. Patients in the ET group had statistically significantly better cosmetic results compared to the ConT group (Table IV). No tracheal and esophageal injury occurred in each group.

Two hundred and twenty-one patients (77.5%) in the ET group and 256 patients (72.7%) in the ConT group had nodular goiter on histopathological examination (Table V). Forty-one patients (14.4%) in the ET group and 57 patients (16.2%) in the ConT group had follicular adenoma, and 20 patients (7.0%) in the ET group and 39 patients (11.1%) in the ConT group had Hurthle cell adenoma. Only 3 patients (1.1%) in the ET group had cystic lesions (Table V).

A new surgical technique when developed has to face up to its conventional standard and be comparable when it comes to results achieved, complications, safety, technical feasibility and cost-to-benefit ratio. A considerable recent surge has occurred in the development of endoscopic techniques in thyroid surgery. Since Gagner (7) and Hüscher et al (1) first reported successful endoscopic neck surgery in 1996 and 1997, several endoscopic thyroid surgical techniques have been developed (8–14). Minimizing surgical stress, cosmetic concerns and improving the postoperative quality of life are the main motivations for developing novel endoscopic thyroid surgical techniques. According to the insertion site of surgical instruments, endoscopic surgery can be classified as cervical (15,16), anterior chest wall (10), breast (17), axillary (18–20) and axillary breast approaches (21). These various techniques of endoscopic thyroidectomy have their own advantages and pitfalls. There were five published series involving 169 patients of ET performed via breast approach (17). We carried out this clinical observation after conducting 285 cases with ET via breast approach to provide an adequate match with our experience in conventional open surgery.

The mean operative time of ET in our series was 79.9±20.10 min for unilateral lobectomy and 89.9±14.60 min for bilateral total thyroidectomy (Table I). The acceptable operative time achieved was due to the short learning curve, the good surgical view, magnification of the surgical field, excellent instrumentation and the breast approach to the thyroid gland. Operative time is the only factor in our series that did not compare favorably to the open surgical procedure. However, the longer operative time compared to the open surgery group could be possibly reduced further through the accumulation of experience.

Although the conversion rates of ET via breast approach are reported to be 0–13% (22–24), we did not have any conversions. This directly relates to the prior endoscopic experience of the main surgeon in endoscopic thyroid surgery with formal training as well as his own experience with previous open thyroid lobectomies. As is the case with open surgery, exposure to a high volume of cases enables one to be comfortable with the surgical anatomy.

The common complications of endoscopic thyroidectomy were transient/permanent hypocalcemia and transient/permanent RLN palsy. With the lateral view of the thyroid gland and the paratracheal region, we easily identified and preserved the RLN and parathyroid glands while performing the ET. In the present study, the superior parathyroid gland was often easier to identify and dissect from the thyroid gland on a vascular pedicle than in open thyroidectomy. We experienced a lower incidence of transient hypocalcemia in the ET group than previous reports (22–24) (Table III). All patients with transient hypocalcemia recovered within 1 month. Several patients who underwent our endoscopic thyroid surgery reported mild hypesthesia and paresthesia in the neck and anterior chest wall. In addition, some of the patients reported discomfort while swallowing (Table III). Therefore, the invasiveness of significant dissection to achieve an adequate working space should be improved through the accumulation of experience.

Cosmetic results are difficult to evaluate because of the reliance on the subjective judgment of the patient. However, most of the patients in the ET group were satisfied with the operative scars (Table IV), and 222 patients (78.7%) were extremely satisfied with their cosmetic results.

ET via breast approach was not pioneered by Chinese endoscopic surgeons, but we did contribute towards improvement in technical details. Firstly, prior injection of 300 ml of diluted epinephrine saline solution into the subcutaneous space by a long needle (modified from a Veress needle) along the prospective operative tunnel up to the sternal notch greatly facilitates the creation of an adequate working space, which results in better pain and bleeding control. Secondly, low-pressure CO₂ (6 mmHg) was applied and was still able to obtain a satisfactory operative view. No subcutaneous emphysema occurred. Therefore, we believe that CO₂ insufflation for endoscopic thyroid surgery is safe if insufflation is maintained at a low pressure. Thirdly, the Harmonic Scalpel exclusively for hemostasis was used in ET without any ties or clips, and no postoperative hemorrhage or hematoma occurred in the ET group. The advantage of this device is that it does not smoke, causes minimal tissue injury, is safe for the surgeon and patient and facilitates dissection (22,25,26). Fourthly, the 30° endoscope was used, as it has a larger visual field, and thus makes the best use of operating space. Lastly, some major improvements and safer technologies, such as intraoperative neuromonitoring to prevent RLN paralysis and early measurement of quick intraoperative parathyroid hormone to avert symptomatic hypocalcemia, have been proposed and would be considered for application to thyroid surgery in subsequent years.

In conclusion, ET performed via breast approach is a technically feasible and safe procedure with excellent cosmetic results for patients with benign thyroid tumors. In addition to improved cosmesis, numerous patients experience decreased pain and faster recovery, and are at no increased risk for complications.