Abstract
Objective. To report a single institution’s experience with transoral robotic surgery (TORS) and its clinical outcomes.
Study Design. Preliminary clinical data from a prospective TORS study.
Setting. University tertiary care facility.
Subjects and Methods. Patients who underwent TORS at The Ohio State University Medical Center. Demographic, intraoperative, clinicopathological, and follow-up functional data were collected.
Results. Sixty-four patients underwent TORS with a median age of 56.9 years. A total of 113 TORS procedures were performed. Fifty-four patients with squamous cell cancer (SCCA) were included in the final analysis. Mean follow-up time was 11.8 months (range, 2-29). There was a trend toward longer TORS setup time, operative time, estimated blood loss, and hospital length of stay with advanced (T3) compared with early-stage tumors (T1-2). There were no major intraoperative complications, and none of the procedures were aborted because of inability to remove the tumor. Negative resection margins were achieved in 93% of cases of SCCA. No patients experienced immediate postoperative complications, and all of the patients tolerated an oral diet without any airway compromise on the day of surgery. Forty-nine patients (91%) underwent adjuvant radiation therapy (RT), with 11 patients requiring gastrostomy tube placement during RT. Addition of TORS to overall management of head and neck SCCA spared adjuvant RT or combined chemotherapy and RT (CRT) in 50% of stage I/II tumors and spared chemotherapy in 34% of stage III/IV tumors.
Conclusion. TORS is a safe procedure with minimal complications and favorable clinical and functional outcomes. It is a promising future alternative surgical treatment for laryngopharyngeal tumors.
Transoral robotic surgery (TORS) has been reported to be feasible at various sites of the upper aerodigestive tract, including the nasopharynx, oropharynx, and larynx, most frequently for the resection of squamous cell carcinoma. 1-6 The advantages of TORS have been attributed to better visualization and ability to manipulate the difficult-to-access oropharyngeal and laryngopharyngeal lesions with 2 hands 5 while minimizing damage to the surrounding musculature and nerves of the pharynx.
In 2006, O’Malley and colleagues 6 reported the feasibility of TORS in a canine model and in 3 human subjects for base of tongue (BOT) SCCA. In that series, TORS provided excellent 3-dimensional visualization and instrument access, allowing for complete resection to negative surgical margins with excellent hemostasis and no complications in the human patient surgeries. In 2007, Weinstein and colleagues 2 reported successful treatment of tonsillar SCCA with TORS, staged neck dissection, and adjuvant therapy. Final margins were found to be negative in 93%, with 96% of patients swallowing without the use of gastrostomy in that series. 2 In 2010, Weinstein and colleagues 7 reported a 100% control rate in 31 patients who underwent TORS and a selective neck dissection (ND) of nodal levels I to IV for the primary treatment of oropharyngeal SCCA, including palate, lateral wall, tonsil, base of tongue, and vallecula as the primary site for the tumor. Seventy-one percent of patients required adjuvant radiation therapy (RT) or a combination of chemotherapy and RT (CRT). None of the patients required a permanent gastrostomy tube, with all patients tolerating an oral diet on 2-year follow-up. 7
The gradual trend toward minimally invasive organ- and function-preserving therapy for oropharyngeal and laryngopharyngeal lesions, paralleled by the evolution of new technologies, has created more surgical options for the management of head and neck tumors, such as transoral CO2 laser microsurgery, TORS, and video-assisted and robotic surgery for the neck and thyroid. 8 The aim of the current study was to report our institution’s experience with TORS, including the feasibility, safety, and utility of the procedure as well as functional and long-term clinical outcomes. These are preliminary results of a larger prospective pilot study of patients undergoing TORS.
Methods
Institutional review board approval was obtained from The Ohio State University Office of Responsible Research Practices. Patients undergoing TORS were identified from The Ohio State University Medical Center transoral robotic surgery database.
TORS was performed under general anesthesia; a laser-safe endotracheal tube was used in all cases. The da Vinci Surgical System (Intuitive Surgical Inc, Sunnyvale, California) was positioned on the right side of the patient, and the bedside surgical assistant was seated at the patient’s head. A Crow-Davis retractor was used in the majority of the cases to expose the oropharynx or laryngopharynx. A Feyh-Kastenbauer (FK) retractor (Gyrus ACMI, Southborough, Massachusetts) was used if greater visualization of the BOT or laryngopharynx was necessary. The retractor was suspended to the bed with a Storz endoscope holder (Karl Storz, Tuttlingen, Germany). The position of the robotic arms included a central 0° (palatine tonsil, BOT, lingual tonsil) or 30° (BOT, laryngopharynx) video endoscope with the robotic surgical arms, with monopolar cautery and grasper trochars coming in on either side of the endoscope ( Figure 1 ).
Transoral robotic surgery exposure of left vallecular lesion with the Feyh-Kastenbauer retractor on the top and 2 robotic arms on the right and left sides.
For the oropharyngeal tumors, the procedure was commenced by making an incision on the lateral aspect of the palatine or lingual tonsil or BOT with a 5-mm spatula monopolar cautery. The specimen was retracted medially with a 5-mm Maryland forceps to provide visualization and tension for dissection. The dissection proceeded in a lateral to medial direction. Once the specimen was resected, it was removed from the oral cavity en bloc, oriented with a suture, and submitted to pathology services for histopathological analysis. The superior constrictor muscle was excised as a deep oncologic margin. It was not possible to preserve the buccopharyngeal fascia in most of the cases because every attempt was made to obtain negative margins during resection. Multiple circumferential and deep margins were submitted as an intraoperative frozen section specimen to determine completeness of tumor removal. The defect was allowed to heal by secondary intention.
For supraglottic tumors, a modification of the transoral robotic supraglottic laryngectomy described by Weinstein et al 4 was used. Exposure to the supraglottis was achieved with the FK retractor. A 30° camera was positioned centrally with 2 robotic arms placed laterally through the oral cavity, including monopolar cautery and Maryland dissector forceps. All the supraglottic lesions in the current series were located on the epiglottis. The epiglottis was retracted medially, and the epiglottic lesions were excised with a margin of normal surrounding tissues. Multiple mucosal margins were obtained and submitted for frozen section analysis. After complete tumor removal was ensured, the robotic arms and the retractors were removed from the oral cavity and the procedure was completed, or the patient was prepped and draped in a sterile fashion for a unilateral or bilateral concurrent neck dissection.
Clinicopathological data, including age at which TORS was performed, gender, type and site of lesion, intraoperative and postoperative complications, need for adjuvant chemotherapy and RT, need for gastrostomy tube (G-tube) and/or tracheostomy placement, and overall survival, were collected. Head and neck pathologists at our institution conducted histopathological review of the specimen to determine SCCA T stage, N stage, overall tumor stage, resection margins, human papillomavirus (HPV) status, protein p16INKa (p16) status, lymph node status, extracapsular lymph node spread, and lymphovascular invasion. Intraoperative details regarding TORS were collected, including diagnostic data, TORS setup, TORS operating room (OR) and total OR times, and intraoperative estimated blood loss (EBL). Diagnostic time was defined as the time during which endoscopies were performed, including all or a combination of direct laryngoscopy, esophagoscopy, and bronchoscopy. TORS setup time was defined as the time between the completion of the diagnostic procedure to the start of the TORS, which encompassed the time needed to place the appropriate retractor in the oral cavity to gain exposure of the lesion, place the endoscopic camera, and place robotic trochars in the oral cavity. TORS OR time was defined as the time between the first incision with the robotic system to the time of removal of the endoscope and robotic trochars from the oral cavity. Total OR time was defined as time between the first incision and completion of all procedures, including neck dissections. The indications for postoperative RT alone were presence of high-risk features, including T3 or T4 primary tumor, N2 or N3 nodal disease, and nodal disease in level IV or V. An option of concurrent CRT was discussed with patients who had these high-risk features. The main indication for postoperative neck radiotherapy alone was metastatic involvement of more than 1 node without the presence of extracapsular extension (ECE). Need for adjuvant RT and chemotherapy was based on high-risk features including (1) the presence of positive surgical margins or close (<2 mm) surgical margins and (2) the presence of extracapsular extension in the cervical lymph nodes.
Statistical analysis was performed with PASW 18.0 (Chicago, Illinois). The number of months of follow-up was defined as the time from the TORS to the time of last follow-up and was expressed as mean and as mean ± standard deviation. Standard descriptive statistics were used to summarize the data.
Results
Between April 2008 and August 2010, a total of 64 patients underwent TORS at our institution. The primary site of the lesion removed with TORS was palatine tonsils in 47 patients (73%), base of tongue in 6 (9%), larynx in 4 (6%), lingual tonsil in 5 (8%), retromolar trigone in 1 (2%), and parapharynx in 1 (2%). Fifty-four patients who underwent TORS as part of their treatment for head and neck SCCA were included in the final analysis. Two patients with an unknown primary lesion of the head and neck who underwent diagnostic bilateral tonsillectomies were excluded from the study because the primary site was not identified. The median age was 56.9 years (range, 37-81). There were 9 women and 45 men ( Table 1 ). All patients were alive at the time of the current study, with a mean follow-up time of 10.8 months (range, 2-29).
Patient Demographic and Clinical Characteristics
Abbreviations: HPV, human papillomavirus; p16, protein p16INKa; SCCA, squamous cell cancer; TORS, transoral robotic surgery.
The mean diagnostic time was 7.09 ± 3.26 minutes, TORS setup time was 23.0 ± 9.30 minutes (± standard deviation), and the mean TORS operating time was 20.4 ± 11.5 minutes, whereas the total OR time was 179 ± 78 minutes. EBL was 16.4 ± 14.8 mL, with an average hospital length of stay (LOS) of 3.00 ± 1.40 days. There was a trend toward longer TORS setup time, operative time, EBL, and LOS with advanced tumors (T3) compared with early-stage tumors (T1-2) ( Table 2 ). In addition, there was a trend toward decreased setup time, TORS tumor removal time, and total OR time after the first 10 cases ( Table 3 ). None of the procedures were aborted because of inability to remove the tumor. There was 1 intraoperative complication, a mucosal laceration of the posterior pharyngeal wall, which was repaired primarily without further sequela. There were 2 cases in which the pharyngectomy and neck dissection fields had communicated. The defects were repaired with local mucosal and muscle flaps. The 2 patients had nasogastric tube nutrition for 5 days, followed by oral intake on postoperative day 5 without further complications. The remaining patients tolerated an oral diet without any airway compromise on the day of surgery. There was 1 postoperative complication, consisting of oropharyngeal bleeding 5 days after TORS tonsillectomy, which required cauterization in the operating room.
TORS Outcomes by T Stage
Abbreviations: EBL, estimated blood loss; OR, operating room; TORS, transoral robotic surgery. Values are mean ± standard deviation.
Operating Room Times (Minutes) for TORS SCCA Cases
Abbreviations: OR, operating room; SCCA, squamous cell cancer; TORS, transoral robotic surgery. Times given as minutes, mean ± standard deviation.
Negative resection margins were achieved in 50 (93%) cases of SCCA. Among patients with SCCA, 46 tumors were stained for HPV. Among those, 37 (80%) were HPV-positive tumors and 84% were p16-positive. All 54 patients with SCCA underwent a unilateral or a bilateral neck dissection in addition to TORS. The mean number of positive lymph nodes on final histopathological diagnosis was 2.18 ± 4.98 (range, 0-37). Forty-nine of 54 patients with SCCA (91%) underwent adjuvant RT, and a total of 63% underwent both chemotherapy and RT, which also varied with the tumor stage, with 73%, 67%, 57%, and 68% of stage I, II, III, and IV tumors requiring adjuvant CRT, respectively ( Table 4 ). The rationale for adjuvant CRT in patients with stage I and II primary tumors was a positive surgical margin, or a close surgical margin (<2 mm), along with perineural and/or vascular invasion. The remaining 9% of patients who did not receive any adjuvant therapy were followed by close clinical surveillance. A total of 11 (19%) patients required G-tube placement due to dysphagia during RT, with a similar trend toward higher tumor stage necessitating need for gastrostomy tube (G-tube) placement compared with earlier stage. The range of time for G-tube placement following TORS was 2 to 5 months, with a mean of 2.7 ± 1.98 months. Five of the 34 (15%) patients with HPV-positive tumors required a gastrostomy tube placement for dysphagia during adjuvant RT or CRT, with all but 1 resuming an oral diet on last follow-up visit. None of the TORS patients required tracheostomy tube placement. There was 1 case of locoregional recurrence of a patient with a tonsil SCCA who failed to undergo the recommended adjuvant RT. There was 1 case of distant recurrence after completion of adjuvant CRT in a patient with a tonsillar SCCA.
TORS Clinical Outcomes by Overall Head and Neck Tumor Stage
Abbreviations: CRT, combined chemotherapy and radiation therapy; TORS, transoral robotic surgery. Values are n (%).
Discussion
The traditional surgical approach to oropharyngeal and laryngopharyngeal lesions is through the nonrobotic transoral approach either with a headlight and monopolar cautery or with laser microsurgery. 4 The challenges of transoral surgery are access, visualization, tissue manipulation, and protection of surrounding structures. Robotic optics allow for excellent visualization of the oropharyngeal and laryngopharyngeal mucosa. This visualization allows for a thorough inspection and transoral resection of oropharyngeal and laryngopharyngeal lesions. Although this technology is still in its developing stages, there is a considerable body of literature supporting the evidence base for robotic head and neck surgery. 2-7,9 This resulted in TORS being approved by the US Food and Drug Administration in 2009 for use in benign and selected malignant tumors of the head and neck. 10
The present series reports the feasibility of TORS in a variety of oropharyngeal and laryngopharyngeal SCCA lesions in 54 patients. The mean TORS setup time in the current series was 23.0 ± 9.30 minutes, and mean TORS operative time was 20.4 ± 11.5 minutes, which is concordant with the range reported in the literature, 9 to 68 minutes. 2,5,6 The TORS setup time varies with the surgeon’s experience with the robotic system and with patient factors such as retrognathia, narrow mandibular arch, and poor neck mobility. 5 Similar to the study by Moore et al, 5 who observed a decrease in the TORS setup time and total OR time with increased experience with TORS, the current series also demonstrated a decrease in TORS setup time (from 34.1 to 21.2 minutes) and total OR time (from 179 to 161 minutes) after the first 10 cases. In addition, in the present cohort there was a decrease in TORS OR time (from 23.0 to 17.4 minutes) after the first 10 cases, which was not seen in the Moore series. 5 This underlines the fact that as with any new technology, there is a certain learning curve in TORS, which in our series turned out to be approximately 10 cases that are set up and performed as primary surgeon.
Concordant with other series, 2,11 the present study demonstrated encouraging operative outcomes with respect to EBL (16.4 ± 14.8 mL) and postoperative LOS (3.00 ± 1.40 days). In a series of 45 patients with oropharyngeal SCCA, the average hospital stay was 3.8 days. 5 The O’Malley et al 6 series demonstrated a LOS of 5 to 7 days for patients who underwent TORS for BOT neoplasms. Five of the 64 patients were discharged home on the same day of the surgery. All of these patients underwent a diagnostic palatine tonsillectomy or a lingual tonsillectomy without a neck dissection.
Among the cases of SCCA, negative resection margins were achieved in 49 of 54 (91%) cases of SCCA in this series. Most cases with positive resection margins were those with extensive parapharyngeal extension. The rate of positive margins is comparable to the rate of 100% negative margin rate reported among 27 patients treated with TORS for tonsillar SCCA, with an excellent locoregional control and only 1 distant oncological failure. 2 Others groups show similar results with respect to a high rate of negative resection margins achieved with TORS. 11
Swallowing and breathing are most at risk after oropharyngeal and laryngeal surgery. A reasonable measure of swallowing function is the rate of G-tube dependence. None of our patients required G-tube placement in the immediate postoperative period. Eleven of 54 (20%) patients with SSCA eventually required G-tube placement. On the last follow-up visit, 4 (7.4%) patients were still using the G tube; the follow-up period for these 4 patients ranged from 2 to 5 months following TORS. Two of the 4 patients were still undergoing adjuvant CRT, and 2 have completed their adjuvant treatments. Similar results were reported by other groups. In one study, 96% of patients were tolerating an oral diet without the use of G tube on last follow-up visit. 2 In a series of 54 patients who underwent TORS for head and neck cancer, Iseli et al 12 reported a 69% rate of oral intake resumption prior to discharge from the hospital and 83% within 2 weeks. The TORS swallowing outcomes results are encouraging compared with the alternative of primary CRT for head and neck cancer–equivalent lesions, which showed a G-tube dependence of 17% to 30% in one review of patients followed for 1 year. 13 Thus, the use of TORS as part of the overall treatment of head and neck SCCA in our series has decreased the G-tube dependence from a rate of 17% to 30%, if primary CRT were used only, to a rate of 7.4% with the use of a combined modality of TORS followed by adjuvant RT or CRT.
Because HPV-positive tumors are common in the oropharynx and have been recognized as a separate entity in terms of risk factors, clinical features, sensitivity to treatment, and prognosis, compared with HPV-negative tumors, 14 we stratified the tumors based on the HPV status. HPV status was reported in 46 of the 54 patients with SCCA; 37 tumors were HPV-positive. Among the patients with HPV-positive tumors, 34 (92%) required postoperative RT, and 23 received both chemotherapy and RT (CRT). The reported oncologic outcomes of HPV-positive tumors treated with CRT are superior to those of HPV-negative tumors 15-17 ; however, the swallowing outcomes vary, with reported G-tube rates between 15% and 70%. 18-20 In a recent clinical trial of CRT for head and neck SCCA, among 47 patients, 33 (70%) required a G-tube placement, with 31 patients eventually resuming an oral diet within an average time of 3.6 months. 20 In the current series, 5 of the 34 (15%) HPV-positive patients required a G-tube placement for dysphagia during adjuvant RT or CRT, with only 1 patient requiring it long term. Our series shows that TORS may offer improved swallowing outcomes for the HPV-positive patients, with avoidance of G-tube placement, compared with the traditional CRT treatment modality offered to these patients. These early data favor TORS over CRT in terms of swallowing outcomes.
Overall, 5 of 10 patients with early stage I and II head and neck tumors received adjuvant treatment in the form of CRT. If a primary nonsurgical modality were used in these patients, all would undergo either RT or CRT; thus, the addition of TORS to the overall management of these patients spared RT or CRT in 50% of patients. Furthermore, among patients with advanced stage III and IV tumors, all have received adjuvant RT and 66% received CRT, sparing adjuvant chemotherapy in 34% of patients with the TORS multimodality approach.
Although the focus of this report is the feasibility and safety of TORS, we are encouraged by the results of this series showing that TORS not only is feasible and safe but also is efficient and has good functional outcomes.
Conclusion
TORS is a feasible and safe procedure with minimal complications and favorable outcomes. It is a promising future endoscopic alternative to surgical treatment of oropharyngeal and laryngopharyngeal lesions, resulting in low morbidity and excellent functional outcomes.
Author Contributions
Disclosures
Footnotes
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
This article was presented at the 2010 AAO–HNSF Annual Meeting & OTO EXPO; September 26-29, 2010; Boston, Massachusetts.