Otolaryngology -- Head and Neck Surgery http://oto.sagepub.com/ Transoral Robotic Glossectomy for the Treatment of Obstructive Sleep Apnea-Hypopnea Syndrome Michael Friedman, Craig Hamilton, Christian G. Samuelson, Kanwar Kelley, David Taylor, Kristine Pearson-Chauhan, Alexander Maley, Renwick Taylor and T. K. Venkatesan Otolaryngology -- Head and Neck Surgery 2012 146: 854 originally published online 13 January 2012 DOI: 10.1177/0194599811434262 The online version of this article can be found at: http://oto.sagepub.com/content/146/5/854 Published by: http://www.sagepublications.com On behalf of: American Academy of Otolaryngology- Head and Neck Surgery Additional services and information for Otolaryngology -- Head and Neck Surgery can be found at: Email Alerts: http://oto.sagepub.com/cgi/alerts Subscriptions: http://oto.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsreprints.nav Permissions: http://www.sagepub.com/journalspermissions.nav >> Version of Record - May 1, 2012 OnlineFirst Version of Record - Jan 13, 2012 What is This?
434262OTOXXX10.1177/0194599811434262Fried man et alotolaryngology Head and Neck Surgery 2011 The Author(s) 2010 Reprints and permission: sagepub.com/journalspermissions.nav Original Research Sleep Medicine Transoral Robotic Glossectomy for the Treatment of Obstructive Sleep Apnea-Hypopnea Syndrome Otolaryngology Head and Neck Surgery 146(5) 854 862 American Academy of Otolaryngology Head and Neck Surgery Foundation 2012 Reprints and permission: sagepub.com/journalspermissions.nav DOI: 10.1177/0194599811434262 http://otojournal.org Michael Friedman, MD 1,2, Craig Hamilton, MBChB 2, Christian G. Samuelson, MD 2, Kanwar Kelley, MD, JD 2, David Taylor 2, Kristine Pearson-Chauhan 2, Alexander Maley 1,2, Renwick Taylor 2, and T. K. Venkatesan, MD 1,2 Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. Abstract Objective. In previous reports of transoral robotic surgery (TORS) for the treatment of obstructive sleep apnea hypopnea syndrome (OSAHS), patients underwent routine tracheotomy. We aim to assess the feasibility of performing robotically assisted partial glossectomy without tracheotomy and to assess efficacy by comparing OSAHS outcomes with those of established techniques. Study Design. Historical cohort study with planned data collection. Setting. Tertiary care center. Subjects and Methods. Forty consecutive patients underwent TORS for OSAHS between October 2010 and June 2011 and were followed up with regard to complications, morbidity, and subjective and objective outcomes. Data from 27 of these patients who underwent concomitant z-palatoplasty with 6-month follow-up were compared with those of 2 matched cohorts of patients, who underwent either radiofrequency (radiofrequency base-of-tongue reduction [RFBOT]) or coblation (submucosal minimally invasive lingual excision [SMILE]) reduction of the tongue base and z-palatoplasty. Results. No major bleeding or airway complications were observed. Postoperative pain and length of admission were similar between groups. All groups saw Epworth score and snore score improvement. Patients undergoing robot-assisted surgery took longer than their SMILE and RFBOT counterparts to tolerate normal diet and longer than RFBOT patients to resume normal activity. Apnea hypopnea index (AHI) reduction averaged 60.5% ± 24.9% for TORS versus 37.0% ± 51.6% (P =.042) and 32.0% ± 43.3% (P =.012) for SMILE and RFBOT, respectively. Only the robotic group achieved statistically significant improvement in minimum oxygen saturation. Surgical cure rate for TORS (66.7%) was significant compared with RFBOT (20.8%, P =.001) but not compared with SMILE (45.5%, P =.135). Conclusion. Robotically assisted partial glossectomy feasibly can be performed without the need for tracheotomy. This technique resulted in greater AHI reduction but increased morbidity compared with the other techniques studied. Keywords robotic surgery, transoral robotic surgery, glossectomy, obstructive sleep apnea hypopnea syndrome, sleep apnea, sleep apnea surgery Received September 4, 2011; revised November 30, 2011; accepted December 7, 2011. The base of the tongue is recognized as a significant site of obstruction in many if not most cases of obstructive sleep apnea hypopnea syndrome (OSAHS). 1 Multiple surgical measures aimed at improvement of tongue-base obstruction exist. However, this problem often remains inadequately treated owing to the unpredictable efficacy of the available procedures. These can be broadly classified as 1 of 2 approaches: tongue suspension or tongue reduction. Definitive reduction through traditional transoral midline glossectomy is infrequently performed. Insufficient visualization of 1 Rush University Medical Center, Chicago, Illinois, USA 2 Advanced Center for Specialty Care, Advocate Illinois Masonic Medical Center, Chicago, Illinois, USA Presented at the 2011 AAO-HNSF Annual Meeting & OTO EXPO; September 11 to 14, 2011; San Francisco, California. Corresponding Author: Michael Friedman, MD, Chicago ENT and Advanced Center for Specialty Care, 30 North Michigan Ave, Suite 1107, Chicago, IL 60602 Email: hednnek@aol.com
Friedman et al 855 crucial neurovascular structures restricts resection to the midline third of the tongue, ignoring the lateral tongue base. Open approaches through the neck can be performed to improve access, but these have significant associated morbidity. 1 In recent times, reasonable success has been achieved through the use of radiofrequency base-of-tongue reduction (RFBOT) and, latterly, submucosal minimally invasive lingual excision (SMILE) techniques. 1 Technological advances allowing improved access to the tongue base may theoretically allow for more aggressive tissue reduction. Robotically assisted surgery has gained popularity in a number of fields over the past decade. Transoral robotic surgery (TORS), pioneered by O Malley and Weinstein, 2 gained approval by the Food and Drug Administration in December 2009 for use in the treatment of benign and malignant conditions of the mouth, tongue, tonsils, pharynx, and larynx. Proponents of this technique cite improved 3-dimensional visualization and precise multiplanar tissue transection as its major advantages. 2-4 Potential disadvantages include a lack of tactile sensation, difficult hemostasis control in some cases, and high capital cost. To date, reports suggest TORS is well tolerated by patients, with complications related to the specific procedure performed rather than the use of the robot. 5 Vicini et al 3 presented a preliminary report on TORS for tongue-base reduction in OSAHS in 2010. Across their 10 patients, a mean apnea-hypopnea index (AHI) reduction from 38.3 to 20.6 was achieved. In 2011, a further report by the same group presented outcomes for 20 patients who underwent robotically assisted glossectomy, with AHI, minimum oxygen saturation, and Epworth Sleepiness Score (ESS) all significantly improved. 4 All patients in these series underwent routine tracheotomy as a precaution prior to TORS, although in the researchers retrospective analysis, they noted no serious problems with bleeding or airway compromise. The aim of our study is to assess the feasibility of performing robotically assisted midline glossectomy without tracheotomy for patients with moderate to severe OSAHS. In assessing the efficacy and associated morbidity of this procedure, we will directly compare outcomes from patients undergoing TORS with those of matched cohorts of patients undergoing radiofrequency treatment (RFBOT) or submucosal coblation (SMILE). We hypothesize that TORS will achieve improved outcomes for OSAHS as a result of the ability to better control the extent of tissue resection. Methods and Materials The standard in tongue-base surgery at our practice has evolved over the past 4 years as new evidence and new techniques have become available. Prior to 2008, RFBOT was the technique of choice for most of our patients. SMILE was then introduced in place of RFBOT from 2008 to 2010. Since October 2010, robotically assisted midline glossectomy has been adopted as the standard approach. This study was a retrospective analysis of data extracted from a prospective computerized database. The Advocate Health Care Institutional Review Board gave approval for review of data from all adult patients who presented for base-of-tongue reduction surgery between March 2007 and June 2011 at a single clinical site (Advanced Center for Specialty Care, Chicago, Illinois). Between October 2010 and June 2011, 40 patients underwent robotically assisted midline glossectomy using the davinci surgical robot (Intuitive Surgical, Inc, Sunnyvale, California). In all cases, tongue-base reduction was performed in tandem with a surgical or minimally invasive palatal procedure to address multilevel obstruction. For the purposes of this study, to control for the effect of the palatal procedure, only data from patients who underwent modified z-palatoplasty (ZPP) were included. Further inclusion criteria for review were (1) significant symptoms of snoring and/or daytime somnolence, (2) moderate to severe OSAHS confirmed by formal polysomnography (defined as an AHI 15), (3) preoperative and postoperative polysomnography and 6-month postoperative follow-up, (4) Friedman tongue position III or IV, 6 (5) the appearance of obstruction at the level of the soft palate, (6) documented failure/refusal of attempts at conservative treatment measures (including but not limited to continuous positive airway pressure), and (7) adults aged 18 years or older. Exclusion criteria were (1) failure to attend postoperative follow-up polysomnography within 6 months of surgery, (2) previous surgery to the base of the tongue or other surgical treatment of OSAHS, and (3) a history of malignancy or infection of the head and neck region, laryngeal trauma, or other previous oropharyngeal/laryngeal surgery. Data from 27 consecutive patients meeting the above criteria who underwent robotically assisted midline glossectomy + ZPP between October 2010 and January 2011 were reviewed. The other 13 patients who underwent robotic glossectomy were excluded on the basis of missing postoperative polysomnography data, not having undergone ZPP, or not having had 6 months elapse from their date of surgery. Two control groups of patients were selected in the following manner: (1) all patients treated consecutively with SMILE + ZPP between 2008 and 2010 and meeting the above inclusion criteria (22 patients) and (2) all patients treated consecutively prior to 2008 with RFBOT + ZPP and meeting the above inclusion criteria (24 patients). Baseline data obtained for all patients include age, sex, body mass index, and Friedman clinical stage for OSAHS (described in previous publications 6,7 ). All patients had either category II or III clinical obstruction. Objective OSAHS Outcomes Findings on formal 16-lead monitored polysomnography were compared preoperatively and postoperatively within and between groups. Outcomes of particular interest were the AHI and minimum oxygen saturation. Rates of surgical cure were also examined. Cure was defined as achievement of an AHI <20 as well as a reduction in AHI 50%. The weight of tissue removed (in grams) was recorded for all TORS patients. Subjective OSAHS Outcomes Preoperative and postoperative ESS (per patient) and visual analog scale (VAS; 0-10) snoring scores (per bed partner) were recorded.
856 Otolaryngology Head and Neck Surgery 146(5) Figure 1. Operating room setup. Anesthesia is set up at the patient s feet, and the surgeon sits at the operating console. The DaVinci robot is rotated 30 relative to the patient. Morbidity/Complications Any major complications related to airway compromise, bleeding, or requirement for revision surgery were recorded. Morbidity was assessed by number of nights spent in the hospital following operation, VAS (0-10) for day 1 postoperative pain, number of days taken to return to normal diet, and number of days taken to return to normal activity. Surgical Technique The objective of midline glossectomy is to enlarge the hypopharyngeal airspace and relieve obstruction. Glossectomy is performed at the midline to allow for aggressive tongue-base resection while minimizing the potential for compromise of the hypoglossal nerve and lingual artery. In all cases, ZPP was performed first. Doing so enlarges the oropharynx and thereby creates extra space for the camera of the robotic arms. ZPP, RFBOT, and SMILE techniques are described in detail in previous publications. 8,9 All patients received perioperative antibiotics (cephalosporin if nonallergic) and steroids. Robotically Assisted Midline Glossectomy The patient is placed in a supine position, with the patient s head placed at the foot of the bed, to allow for accommodation of the robotic cart. The anesthesiologist and anesthesia equipment are located at the patient s feet (Figure 1). The patient s eyes are protected using sterile drapes and eye covers, and the teeth are protected with a dental guard (Figure 2). Patients are placed under general anesthesia and undergo nasotracheal intubation. The oral cavity is then rinsed with
Friedman et al 857 Figure 2. The patient is intubated nasally, and a Jennings mouth gag is placed to allow for a transoral surgical approach. The patient s eyes and teeth are protected using eye covers and a mouth guard. 0.12% chlorhexidine gluconate. The course of the lingual arteries is identified bilaterally and marked using Doppler ultrasound. After the lingual artery is identified, 10 ml of saline with epinephrine (1:100,000) is injected into the tongue base. The lingual artery must be identified prior to injection of epinephrine since the epinephrine will cause vessel spasm. A small amount (1-2 ml) of bupivacaine may be injected for pain control. Anesthetic is not used with epinephrine for local primary injection because of the potential risk of bilateral hypoglossal paralysis. Although temporary, bilateral hypoglossal paralysis will obstruct the airway in the postoperative period. A Jennings mouth gag is placed to allow for a transoral surgical approach (Figure 2). Two retraction sutures are placed in the anterior aspect of the tongue to allow for forward retraction and better visualization and access to the posterior tongue (Figure 3). Tension on the anterior sutures should be applied laterally to help spread the tongue tissue. Using a Bovie, 2 straight parallel incisions are made at the circumvallate papillae and extended toward the valleculae. Two sutures are then placed at the lateral aspects of the incision site to allow for lateral retraction at the site of incision/glossectomy. The robotic patient side cart is brought into position at a 30 angle to the operating room table. The surgeon sits at a console away from the patient, operating the davinci surgical system (Figure 1). The Schertel and spatula robotic surgical instruments are attached to 5-mm robotic arms. For a righthanded surgeon, we attach the Schertel grasper hand piece to the left robotic arm and the spatula tip to the right arm. Figure 3. Two sutures are placed at the anterior aspect of the tongue to allow for forward retraction. Two straight parallel incisions are made at the midline area between the circumvallate papilla to the valleculae to allow for the raising of a mucosal flap, under which resection takes place. A small strip of mucosa is included with the specimen for grasping so that the submucosal and muscle resection can take place using the spatula tool (Figure 4). The spatula cutting tool resects mucosa and muscle inferiorly in a triangular area (Figure 5). At the end of the procedure, a Maryland tool is exchanged for the spatula on the right arm of the robot, which allows for bipolar cautery. Closure is performed with a 2-0 Vicryl suture. Patients are fully awake and follow commands before extubation is performed. Nasal trumpets should be readily available to alleviate potential hypopharyngeal obstruction. Statistical Analysis Continuous data are displayed as mean ± standard deviation (SD). Statistical significance is accepted when P <.05. The 2-tailed independent Student t test is used to compare differences in continuous variables between the robotic group and each of the other 2 groups in turn. The Student t test is used to compare preoperative and postoperative mean values within each group. χ 2 analysis is used to test the association
858 Otolaryngology Head and Neck Surgery 146(5) Figure 4. The Schertel tool is attached to the left robotic arm and is used to retract, while the spatula tool is attached to the right robotic arm and is used to resect tissue under the mucosal flap. Two sutures at the lateral margins of the incision provide lateral traction. Figure 5. Mucosa and muscle are resected inferiorly in a triangular area. between categorical variables. Note that no direct comparison is made between the SMILE and RFBOT groups. Results Twenty-seven consecutive patients underwent robotically assisted midline glossectomy + ZPP without tracheotomy. All procedures were completed successfully. A mean 2.28 ± 0.43 g of lingual tissue was excised. Intraoperative complications included mild bleeding (controlled with bipolar cautery) and edema. There were no incidences of significant bleeding or airway complications in the postoperative period. No complications related to tongue mobility, hypoglossal nerve injury, or speech were noted. No patients required revision surgery. Data from 22 consecutive patients who underwent SMILE + ZPP and 24 consecutive patients who underwent RFBOT + ZPP were independently compared with data from the robotic glossectomy group. The baseline characteristics of the 3 groups were homogeneous (Table 1). The average time from the date of surgery to postoperative polysomnography across all groups was 88 ± 78 days. Paired t-test analysis reveals a statistically significant reduction in AHI, snoring VAS, and ESS in all groups; however, only the robotic group achieved a significant improvement in the minimum oxygen saturation (Table 2). A significantly greater percentage improvement in AHI of 60.5% ± 24.9% is seen in the robotic group, compared with 37.0% ± 51.6% (P =.042) and 32.0% ± 43.3% (P =.012) in the SMILE and RFBOT groups, respectively. The proportion of patients who achieved surgical cure was higher in the robotic group (66.7%) compared with both SMILE (45.5%) and RFBOT (20.8%); however, this finding was statistically significant only compared with RFBOT. Within the robotic glossectomy group, no direct correlation was found between the weight of lingual tissue excised and the degree of improvement in AHI, minimum oxygen saturation, or ESS (Figure 6). Day 1 postoperative pain was not increased in the robotic group compared with the other groups. Length of hospital stay was increased in the robotic group compared with the RFBOT group (P =.034). Number of days to return to normal activity was increased compared with RFBOT (P =.015), and number of days to return to normal diet was increased compared with both SMILE (P =.012) and RFBOT (P =.006). Preoperative and postoperative findings within groups are summarized in Table 2. Findings are compared between the robotic glossectomy group and each of the other groups respectively in Table 3. An overall increase in operative time of approximately 1 hour per case was noted in the robotic group. This was largely attributed to time taken for setup of the robotic equipment between the ZPP and glossectomy phases. On average, an increase in turnover time between cases of approximately 30 minutes was also noted. Disposable costs were $730 per robotic procedure and $250 per SMILE or RFBOT procedure. Discussion Our results demonstrate that robotically assisted midline glossectomy can be feasibly performed without the need for
Friedman et al 859 Table 1. Comparison of Baseline Characteristics of the Robotic Group to Both SMILE and RFBOT Control Groups a Robot SMILE P Value RFBOT P Value No. 27 22 24 Age at surgery, y, mean ± SD 43.8 ± 9.2 41.7 ± 8.6.429 44 ± 9.2.932 Sex.816.739 Male 24 20 22 Female 3 2 2 BMI, kg/m 2, mean ± SD 32.3 ± 3.3 31.5 ± 4.6.473 31.6 ± 5.3.604 Clinical stage.649 II 7 7 NA III 20 15 NA OSAHS severity.476.810 Moderate 4 5 3 Severe 23 17 21 AHI, mean ± SD 54.6 ± 21.8 53.7 ± 29.3.899 54.7 ± 26.6.998 Min O 2 sat, mean ± SD 78.5 ± 7.4 79.7 ± 11.9.667 80.8 ± 7.8.291 Snoring VAS, mean ± SD 9.1 ± 1.0 7.4 ± 3.4.057 9.1 ± 0.8.914 ESS, mean ± SD 14.4 ± 4.5 15.2 ± 4.0.565 16.6 ± 2.8.054 Abbreviations: AHI, apnea-hypopnea index; BMI, body mass index; ESS, Epworth Sleepiness Score; Min O 2 sat, minimum oxygen saturation; NA, data not available; OSAHS, obstructive sleep apnea hypopnea syndrome; RFBOT, radiofrequency base-of-tongue reduction; SMILE, submucosal minimally invasive lingual excision; VAS, visual analog scale. a Statistical significance is accepted when P <.05. tracheotomy. To our knowledge, this represents the only study to date in which this approach has been examined. No significant intraoperative or postoperative complications requiring interventional airway management or revision surgery were observed in our series of 27 patients. However, having established that this approach can be safely performed, whether it should be undertaken in preference to other techniques is a different question entirely. Vicini et al 3,4 stated that TORS has a promising future as a treatment of OSAHS, having demonstrated its efficacy in their 2 published case series. In the current study, we find robotically assisted resection to be more effective than the other tongue-base reduction treatments examined with regard to improvement of objective OSAHS parameters, particularly percentage reduction in AHI. Of note, given the small sample sizes in our study, the robotic group was the only one that demonstrated a significant improvement in minimum oxygen saturation. Greater improvements in daytime somnolence were also noted, which were statistically significant compared with RFBOT. A higher rate of surgical cure was observed in the robotic group, and although this finding did not reach statistical significance, we feel it is deserving of further study. We hypothesized that any increase in efficacy observed when using the robotic technique compared with SMILE would be due to a greater degree of tissue resection achieved under direct visualization. This remains the most plausible explanation for the difference in outcomes noted between the otherwise homogeneous robotic and SMILE groups. However, because of the degree of tissue destruction with SMILE, specimens cannot be accurately weighed for comparison. Within the robotic group itself, there was little correlation observed between the mass of lingual tissue removed and the degree of improvement in either objective or subjective parameters. However, there was only minor variance in the tissue mass excised (SD = 0.43 g, maximum = 3.12 g). While we have information regarding the degree of obstruction at the tongue base (Friedman tongue position/clinical stage), when analyzing these trends, the initial tongue mass of individual patients cannot be adequately controlled for. As we gain experience with the robotic technique, a potential link between excised tissue mass and outcomes might yet be examined through further studies in which more aggressive tissue resection is attempted. Although visibility and dexterity within the surgical field are improved with the robotic technique, there remain a number of anatomic constraints that limit the degree of resection that may be attempted. While our triangular resection technique may pose greater risk to the neurovascular bundle than an inverted pyramid technique, it was performed to minimize mucosal and submucosal resection. The area of the vascular bundle was protected following identification with Doppler. The existing literature on TORS for tongue-base reduction in OSAHS as well as oncological purposes reports that the procedure is generally well tolerated. 3,4,5,10 As all patients underwent concomitant ZPP, the pain and postoperative morbidity associated with this procedure should be equivalent across groups, and any differences observed ought to be related to the respective tongue-base procedure. In our wider series of more than 40 patients, no significant complications relating to tongue mobility, hypoglossal nerve injury, or speech have been noted. Robotic patients in our series spent an average 0.5 nights longer in the hospital than RFBOT patients. Postoperative pain on day 1 was not increased in our robotic group compared with either minimally invasive control group. However, at follow-up appointments, a greater proportion of patients in the robotic group complained of
860 Otolaryngology Head and Neck Surgery 146(5) Figure 6. Relationship between excised tissue mass and outcomes in the robotic group (linear regression analysis). prolonged dysphagia. This is reflected in our data on return to normal diet, where robotic patients took significantly longer than both SMILE and RFBOT patients to achieve this end. These findings differ markedly from those reported by Vicini et al, 3,4 although this may be explained by differences in the way this metric was measured. We believe this increase in dysphagia may be due to both increased tissue removal and heat damage caused through the use of unipolar cautery. By contrast, tissue resection in SMILE occurs at much lower temperature. Return to normal activity was also slower in the robotic group than for RFBOT but equivalent with SMILE. Aside from the efficacy and associated morbidity of the procedure, an additional consideration should be the practicality of performing robotically assisted surgery, including time and cost. In a 2007 cost-analysis study, Barbash and Glied 11 estimated an additional variable cost across 20 robotically assisted procedures (not including head and neck procedures) of $1600. We found disposable costs for robotic glossectomy to be lower than this estimate, with an average per-procedure cost of $730. An increase in operative time and turnover time between cases is likely to drive costs significantly higher. However, with increasing operator experience and training for OR staff in particular, the effects of these delays may be minimized. While we currently report an increase in operative time of approximately 1 hour per case (largely attributable to setup time), in their experience with a series of more than 60 patients, Vicini et al 4 reported a mean setup time approaching 30 minutes.
Friedman et al 861 Table 2. Within-Group Comparison of Preoperative and Postoperative Objective and Subjective Treatment Outcomes a Robot SMILE RFBOT Preoperative AHI 54.6 ± 21.8 53.7 ± 29.3 54.7 ± 26.6 Postoperative AHI 18.6 ± 9.1 26.6 ± 23.9 34.6 ± 22.5 P value <.001 b.002 b.007 b Preoperative min O 2 sat 78.5 ± 7.4 79.7 ± 11.9 80.8 ± 7.8 Postoperative min O 2 sat 86.5 ± 6.3 84.8 ± 9.0 84.3 ± 7.5 P value <.001 b.113.116 Preoperative snoring VAS 9.1 ± 1.0 8.5 ± 2.7 9.1 ± 0.8 Postoperative snoring VAS 2.3 ± 2.9 2.3 ± 3.2 2.2 ± 2.6 P value <.001 b <.001 b <.001 b Preoperative ESS 14.4 ± 4.5 14.8 ± 4.0 16.6 ± 2.8 Postoperative ESS 5.4 ± 3.1 6.7 ± 4.7 10.8 ± 3.5 P value <.001 b <.001 b <.001 b Abbreviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Score; Min O 2 sat, minimum oxygen saturation; RFBOT, radiofrequency base-of-tongue reduction; SMILE, submucosal minimally invasive lingual excision; VAS, visual analog scale. a All values are expressed as mean ± standard deviation. b Significant (statistical significance is accepted when P <.05). Table 3. Comparison of Postoperative Subjective and Objective Treatment Outcomes and Morbidity between the Robotic Group and Both SMILE and RFBOT Control Groups Robot SMILE P Value RFBOT P Value AHI reduction, mean ± SD 36.1 ± 21.6 27.2 ± 32.2.254 20.0 ± 25.2.022 a % AHI reduction, mean ± SD 60.5 ± 24.9 37.0 ± 51.6.042 a 32.0 ± 43.3.012 a Improvement min O 2 sat, mean ± SD 8.0 ± 8.6 5.1 ± 9.1.257 3.5 ± 8.2.076 Surgical success, % 66.7 45.5.135 20.8.001 a Cure, n 18 10 5 No cure, n 9 12 19 Snoring VAS reduction, mean ± SD 5.9 ± 3.6 6.1 ± 4.3.844 6.8 ± 2.6.308 ESS reduction, mean ± SD 8.7 ± 4.3 8.0 ± 5.0.668 5.8 ± 3.5.024 a Day 1 pain VAS, mean ± SD 8.2 ± 2.5 8.5 ± 1.7.737 7.8 ± 1.8.650 Nights in hospital, mean ± SD 1.6 ± 0.7 1.5 ± 0.9.714 1.1 ± 0.9.034 a Days to normal activity, mean ± SD 12.9 ± 9.7 12.2 ± 8.9.833 7.3 ± 4.2.015 a Days to normal diet, mean ± SD 19.3 ± 8.4 13.9 ± 4.0.012 a 13.3 ± 4.7.006 a Abbreviations: AHI, apnea-hypopnea index; ESS, Epworth Sleepiness Score; Min O 2 sat, minimum oxygen saturation; VAS, visual analog scale. a Significant (statistical significance is accepted when P <.05). Conclusion Robotically assisted midline glossectomy may be safely performed without the need for preoperative tracheotomy. In our group of patients, robotic glossectomy combined with ZPP resulted in a greater reduction in AHI when compared with either SMILE/ZPP or RFBOT/ZPP. Morbidity was increased, with a longer recovery time needed for return to normal diet. Procedural costs and operating room time were increased with the robotic technique versus the other techniques studied. Acknowledgment We acknowledge Evie Baumann, Advanced Center for Specialty Care, Chicago, Illinois, for data collection. Author Contributions Michael Friedman, study design, surgical procedures, data collection/ interpretation, drafting/revision, presentation; Craig Hamilton, study design, data collection/analysis/interpretation, drafting/revision; Christian G. Samuelson, study design, data collection/analysis/interpretation, drafting/revision; Kanwar Kelley, study design, institutional review board (IRB) submission, data collection, drafting; David Taylor, study design, IRB submission, data collection, drafting; Kristine Pearson-Chauhan, study design, IRB submission, data collection, drafting; Alexander Maley, study design, IRB submission, data collection, drafting; Renwick Taylor, data collection; T. K. Venkatesan, study design, surgical procedures, data collection, revision. Disclosures Competing interests: Michael Friedman is the recipient of a grant for a study on nasal irrigation as treatment for sinonasal symptoms from TriCord Pharmaceuticals and is a member of the speaker s bureau for Glaxo-Smith-Kline. Sponsorships: None. Funding source: This study was funded in its entirety by the principal investigator (Michael Friedman, MD). The authors have no financial associations with Intuitive Surgical.
862 Otolaryngology Head and Neck Surgery 146(5) References 1. Friedman M. Sleep Apnea and Snoring: Surgical and Non-surgical Therapy. Philadelphia, PA: Saunders; 2009. 2. O Malley BW Jr, Weinstein GS, Snyder W, Hockstein NG. Transoral robotic surgery (TORS) for base of tongue neoplasms. Laryngoscope.2006;116:1465-1472. 3. Vicini C, Dallan I, Canzi P, et al. Transoral robotic tongue base resection in obstructive sleep apnoea-hypopnoea syndrome: a preliminary report. ORL J Otorhinolaryngol Relat Spec. 2010;72:22-27. 4. Vicini C, Dallan I, Canzi P, et al. Transoral robotic surgery of the tongue base in obstructive sleep apnea-hypopnea syndrome: anatomic considerations and clinical experience. Head Neck. 2012;34(1):15-22. 5. Hockstein NG, O Malley BW Jr. Transoral robotic surgery. Operative techniques in otolaryngology. 2008;19:67-71. 6. Friedman M, Ibrahim H, Bass L. Clinical staging for sleep-disordered breathing. Otolaryngol Head Neck Surg. 2002;127(1):13-21. 7. Friedman M, Ibrahim H, Joseph N. Staging of obstructive sleep apnea/hypopnea syndrome: a guide to appropriate treatment. Laryngoscope. 2004;114(3):454-459. 8. Friedman M, Ibrahim HZ, Vidyasagar R, et al. Z-palatoplasty (ZPP): a technique for patients without tonsils. Otolaryngol Head Neck Surg. 2004;131(1):89-100. 9. Friedman M, Soans R, Gurpinar B, et al. Evaluation of submucosal minimally invasive lingual excision technique for the treatment of obstructive sleep apnea/hypopnea syndrome. Otolaryngol Head Neck Surg. 2008;139:378-384. 10. Weinstein GS, O Malley BW Jr, Desai SC, et al. Transoral robotic surgery: does the ends justify the means? Curr Opin Otolaryngol Head Neck Surg. 2009;17:126-131. 11. Barbash GI, Glied SA. New technology and health care costs the case of robot-assisted surgery. N Engl J Med. 2010;363: 701-704.