Transoral robotic surgery: does

the ends justify the means?Gregory S. Weinstein, Bert W. O’Malley Jr, Shaun C. Desai and Harry Quon

Department of Otorhinolaryngology – Head and NeckSurgery, the University of Pennsylvania, Philadelphia,Pennsylvania, USA

Correspondence to Gregory Weinstein, MD, Professorand Vice Chairman, Department ofOtorhinolaryngology – Head and Neck Surgery, 3400Spruce St, 5 Ravdin, University of Pennsylvania HealthSystem, Philadelphia, PA 19104, USATel: +1 215 349 5390; fax: +1 215 349 5977;e-mail: gregory.weinstein@uphs.upenn.edu

Current Opinion in Otolaryngology & Head and

Neck Surgery 2009, 17:126–131

Purpose of review

Head and neck surgical science has developed dramatically during the past 20 years

with a major focus on organ preservation surgery. Among these organ preserving

surgeries are the selective neck dissections, supracricoid partial laryngectomies,

transoral laser surgeries, and now a newcomer, transoral robotic surgery utilizing the da

Vinci surgical system. Transoral robotic surgery is in its infancy, but, indeed, there have

been some questions raised about the role of these innovative robotic surgical

techniques.

Recent findings

This article will review, point by point, the questions that have been raised concerning

the feasibility; safety and efficacy; teachability; and cost effectiveness of transoral

robotic surgery.

Summary

Although the present literature reports early findings, without long-term oncologic

outcomes, the results are consistently encouraging. Training programs have already

yielded successes. Indeed, multiple institutions have shown that transoral robotic

surgery programs can be successfully established yielding excellent clinical outcomes.

In addition, early studies of swallowing function following transoral robotic surgery show

swallowing outcomes that are superior to some of the reported chemoradiation results

for equivalent lesions.

Keywords

da Vinci surgical system, head and neck cancer, larynx cancer, tonsil cancer, transoral

robotic surgery

Curr Opin Otolaryngol Head Neck Surg 17:126–131� 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins1068-9508

IntroductionIn recent years, robotic surgery has increased in popularity

in various surgical specialties, including cardiac surgery,

urology, general surgery, and gynecology [1–4]. None-

theless, recent media coverage has led some skeptical

physicians to question whether the benefits of robotic

surgery outweigh the potential negatives. The high initial

cost of the robotic system (over a million dollars) and cost of

instruments have also led to skepticism about robotic

surgery in general. Specific concerns that have been

expressed about transoral robotic surgery (TORS) include

the issues such as feasibility of TORS; safety and efficacy;

‘teachability’ of complex new techniques; and competition

with the recent trend of nonsurgical treatment for equiv-

alent cancers. In the following article, we will discuss the

various controversies and limitations described in the

literature by reviewing the current clinical experience with

TORS at various institutions.

BackgroundFor the reader who may not be familiar with the term

‘robotic surgery’, it is performed utilizing the da Vinci

1068-9508 � 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins

surgical system. The surgeon sits at the console and

controls micromanipulators, which in turn are connected

to a robotic cart at the patient’s bedside. In TORS, three

arms are routinely utilized. The central arm has a double-

video endoscope with high-quality video that gives the

surgeon a three-dimensional view of the operative field

via the console. The two other arms carry interchangeable

instruments (approximately 5 mm wide and 2 feet long)

with miniaturized tools on the end that mimic standard

surgical instruments (i.e. electrocautery, pickups, etc.).

The tips of the double-video endoscope and the instru-

ment arms are inserted transorally and the assistant sits

at the bedside to aid with suctioning and retraction

(Fig. 1a–c). The tips of these robotic surgical instruments

are also ‘wristed’, so when surgeons move their wrist and

hands at the console, the entire motion is scaled down to

the miniaturized ‘robotic’ instruments, with benefits such

as tremor filtration (Fig. 2a and b). Traditional nonrobotic

transoral surgery can at times be surgically awkward

secondary to (1) the instruments, which are long and of

limited functionality; (2) the microscopic optics, which

are outside the oral cavity; or (3) the laser, which is a line

of sight beam far from the lesion. In contradistinction, the

robotic optics are in the oral cavity and the miniaturized

DOI:10.1097/MOO.0b013e32832924f5

Transoral robotic surgery Weinstein et al. 127

Figure 1

(a) The robotic cart is at the patient’s bedside and the tips of the arms are placed transorally. The assistant is positioned at the head of the bed, in orderto perform tasks such as suctioning. (b) The FK retractor, which is a cross between a mouth gag and a laryngoscope, is in place in the oral cavity and thetips of the robotic arms are placed intraorally. (c) The Crow–Davis mouth gag is in place, with the robotic endoscope centrally in the mouth and the two-wristed robotic instruments coming in at an angle on either side.

surgical instruments move exactly as the surgeons’ hands,

making the experience more like an actual open

surgical experience.

The concept and development of TORS using the da

Vinci surgical robot (Intuitive Surgical, Sunnyvale,

California, USA) was first successfully studied at the

University of Pennsylvania through a series of technical

and feasibility trials. Initial studies included preclinical

experiments involving manikin and cadaveric models in

which it was proven that the optimal way to perform

TORS was to work through mouth gags rather than

through traditional laryngoscopes [5,6]. Additional

cadaver and canine studies [7,8] showed feasibility for

multiple pharyngolaryngeal sites. This work laid the

foundation for the first application of TORS in human

patients in which three base of tongue neoplasms and

supraglottis were successfully extirpated in an institu-

tional review board (IRB)-approved human clinical trial

[9,10�]. The present number of TORS cases that have

been performed at the University of Pennsylvania

exceeds 225 cases.

Although the group at the University of Pennsylvania had

the first IRB-approved study of TORS in the world,

which started accruing patients in May 2005, it was the

128 Head and neck oncology

Figure 2

(a) This is the view seen by the surgeon sitting at the console of thewristed tip of the bovie and the Maryland pickups on the right during atransoral robotic surgery (TORS) radical tonsillectomy for cancer. (b)The surgeon’s console, which is placed a few yards away from theoperative field. The surgeon controls the wristed instruments on thebedside cart by moving manipulators on the console. The view seen bythe surgeon is a high-quality, magnified, three-dimensional video of theoperative field.

contention of O’Malley and Weinstein that ‘teachability’

was a critical factor in success of the program. Working

together with Intuitive Surgical Inc, in October 2006, a

research training seminar was held for 12 surgeons from

around the United States to share concepts, research

approaches and proctor a day-long cadaver training ses-

sion. This model for training proved to be a great success,

indicating the ‘teachability’ of these techniques, and

since then surgical teams trained by Weinstein and

O’Malley have developed TORS programs with much

success, demonstrating its potential niche for TORS in

head and neck surgery [11–13].

FeasibilitySome critics of robotic surgery have cited access and

bulky robotic instruments as main limitations to TORS

[14]. However, from our experience at the University of

Pennsylvania, access to almost any oropharynx lesion and

many supraglottic and hypopharyngeal lesions is not an

issue if appropriate triage is utilized in the form of a

separate preoperative endoscopy performed at a time

prior to the day of TORS. In a series of 27 patients with

T1–T3 tonsil cancer, surgical access was achieved in all

patients using either the Crow–Davis or FK retractors

(Gyrus ENT, LLC, Bartlett, Tennessee, USA). All

robotic arms functioned optimally using 5 mm, and less

commonly, 8 mm instruments, with no interference

noted between the robotic arms [15��]. Other groups

have had similar experience using the Crow–Davis,

FK, and Dingman retractors (Gyrus ENT, LLC) [16].

In another pilot study of 31 patients aimed at assessing

the indications and limitations associated with TORS

[17], five (16%) were ultimately terminated and deemed

inaccessible via this approach. Reasons for restricted

access in this series of patients with benign and malignant

lesions of the upper aerodigestive tract included retrusive

mandible (two), the inability to effectively retract per-

ipheral soft tissue (one), and the inability to access the

lesion because of the bulky nature of the robotic arms

[17]. In a similar study aimed at assessing surgical

exposure during TORS [13], 17 of 23 (73.9%) patients

had successful tumor removal via a robotic approach. Of

six patients who were ultimately terminated via a TORS

approach, four were found to have inadequate exposure

and two were found to have tumors too large for robotic-

assisted resection [13]. However, it is unclear whether

these other feasibility studies evaluated patients preo-

peratively under general anesthesia to assess for transoral

exposure, which could have potentially reduced the

number of cases that were intraoperatively terminated.

It is clear, however, that, in the University of Pennsylva-

nia series, the judicious utilization of preoperative panen-

doscopy was a very accurate way to triage patients to

TORS or to other treatment options when access was

considered a problem as, in the first 150 cases, only three

were terminated because of inability to gain adequate

access to perform TORS [18].

Total operative time and operating room setup time have

also been cited as potentially limiting routine use of the

da Vinci robot [14]. In our experience, as well as that of

others, total operative time (including set-up) seems to

fall with increased experience [9,10�,15��]. This trend

seems to parallel those from other robotic surgical sub-

specialties [4]. In a series of 150 patients with lesions in

Transoral robotic surgery Weinstein et al. 129

the oral cavity or laryngopharynx, an average additional

setup time needed to achieve exposure and robotic

positioning for TORS was just an additional 4 min when

compared with exposure time for standard transoral

resection [18].

Safety and efficacyIn general, there has been limited controversy in the

literature over the encouraging operative outcomes for

TORS with respect to estimated blood loss (EBL),

hospital length of stay, and postoperative complication

rates. EBL has been found to be very low (<200) in the

majority of TORS patients with no series to date report-

ing requirement of blood transfusion [13,15��,16]. In

comparison, open approaches have a much higher risk

of requiring blood transfusions as well as an overall higher

average EBL.

Although hospital length of stay has largely been institu-

tion-dependent, it appears that the overall stay, regard-

less of surgeon preference, has been shorter for TORS

patients than for those who would have otherwise under-

gone an open approach. In one series by Moore et al. [12],

all 35 patients with oropharyngeal squamous cell carci-

noma were discharged from the hospital within 6 days. In

another study by Boudreaux et al. [13], mean hospital stay

was just 2.7 days. Our experience has been similar, with

most patients staying between 5 and 7 days [9].

Another issue is the potential for postoperative compli-

cations. In a series of 27 tonsil cancer patients with

surgery done by TORS, five of 27 (19%) developed minor

complications that mostly resolved without significant

sequelae [15��]. These included delirium tremens from

alcohol withdrawal, transoral bleeding that was fixed by

cautery for minor mucosal bleeding, exacerbation of sleep

apnea from postoperative swelling, development of mod-

erate trismus, and occurrence of temporary hypernasality

of voice in one patient. Similar studies of 23, 35, and 31

patients, respectively, revealed no immediate compli-

cations [12,13,17]. None of the studies reported acute

sequelae, which have been reported in the alternative

treatments, such as death, pneumonia, or fistula or all of

them [19,20]. Furthermore, the complication rate for

primary chemoradiation has been cited as high as 80%

and often times with synergistic toxicity [20].

TeachabilitySurgeon and ancillary staff trainability is an important

concern in creating or sustaining a TORS program at any

institution. Although no proper set of guidelines has been

set forth, it is encouraging to see that the majority of the

surgeons of the original group (12 surgeons) who origin-

ally attended a training session by Drs Weinstein and

O’Malley at the Intuitive Surgical Six Robot Training

Center in Sunnyvale, California, went on to start success-

ful TORS programs at their respective institutions.

Furthermore, robotic surgical fellowships and training

sessions have been developed in other surgical fields

and show a feasible learning curve for resident training,

which hints at its potential in head and neck surgery [21].

Of note, since 2005 the graduating Head and Neck

Surgery Fellows at the University of Pennsylvania have

been trained in TORS. Although there is a hold on

further training of teams from the United States while

Food and Drug Administration (FDA) approval is being

sought, TORS is approved for use in many European,

Asian, and South American countries. The TORS train-

ing program at the University of Pennsylvania includes

observation in the clinic and the operating room as well as

robotic training in a porcine laboratory. Attendees of

the University of Pennsylvania training program have

returned to their home countries and developed active

TORS clinical programs, most notably in Korea, Belgium,

France, Italy, and Brazil.

Functional outcomesFunctional outcomes have largely remained one of the

most important factors in determining therapy as this has

such a major impact on quality of life. Although long-term

studies are not yet available for TORS, preliminary out-

comes have been extremely encouraging.

It is the authors’ opinion that a reasonable surrogate for

swallowing function, which allows comparison between

studies of various modalities that report this outcome, is

gastrostomy-tube dependence. In our series, 26 of 27

patients (96%) were swallowing without the use of a

gastrostomy tube at last follow-up [15��]. In another series

by Boudreaux et al. [13], of 23 patients, three were

gastrostomy-tube-dependent (however, one patient was

already dependent preoperatively). Genden et al. [16] did

not use preoperative gastrostomy tubes in one series

and began feeding pureed diet on postoperative day 1,

provided that the patient was evaluated by a speech

therapy team preoperatively and postoperatively and

the patient showed no evidence of aspiration. Other

groups have found similar results, with all 35 patients

in one series tolerating an oral diet within 2 weeks of the

procedure [12]. The use of a gastrostomy tube to ensure

proper nutrition has largely remained surgeon and institu-

tion preference and long-term comparison studies

are warranted. Regardless, the encouraging swallowing

function afforded by TORS may ultimately show an

advantage when compared with the alternative of primary

chemoradiation for head and neck cancer, which has

shown a gastrostomy-tube dependence, in one literature

review, in 17–30% of patients followed up for 1 year [22].

Nguyen et al. [22] had then focused on an often

130 Head and neck oncology

overlooked problem in the literature – the significant risk

of long-term swallowing dysfunction after chemoradia-

tion for head and neck cancer. It is reports such as this

that provide an important rationale for utilizing TORS, in

particular for oropharyngeal carcinomas.

Oncologic outcomesDespite a lack of conclusive evidence for the benefits of

chemoradiation over open or endoscopic primary surgery

followed by radiation or chemoradiation, there appears to

be a national trend toward increasing use of chemoradia-

tion as a primary modality for oropharyngeal carcinoma

[23]. Although, at this time, it is not possible to assess

long-term oncologic outcome in patients who underwent

TORS, predictions and observations can be made on the

basis of current trends and data.

In an important paper by Machtay et al. [20] at the

University of Pennsylvania, it was found that negative

surgical margins always resulted in local control.

Although this group could achieve negative margins in

all cases of tongue base carcinomas via an open surgical

approach, it was at the cost of significant functional

deficits. However, in our TORS radical tonsil series,

we could achieve negative margins in all 27 patients with

acceptable morbidity. In that series, there was no early

local or regional recurrence and only one distant oncolo-

gic failure [15��]. Other groups including Moore et al. [12]

Boudreaux et al. [13], and Genden et al. [16] all achieved

negative surgical margins in their respective experiences

as well.

Of interest, the results from a study at the University of

Pennsylvania comparing outcomes following TORS on

patients with and without human papilloma virus posi-

tivity are presently pending. The oncologic and func-

tional results of each group will then be compared with

equivalent lesions that have undergone chemoradiation,

previously reported in the literature. Indeed, if it is found

that the oncologic outcomes are equivalent or superior to

those of similar group of chemoradiated patients, then the

most important factor for triaging patients to TORS or

chemoradiation will be the swallowing outcomes. As this

review shows, the early data favor TORS over chemo-

radiation in terms of swallowing outcomes.

Cost considerationsSeveral skeptics cite cost as a primary limitation to wide-

spread use of the da Vinci surgical robot. The initial cost

of approximately 1.5 million dollars, coupled with an

approximately $100 000 yearly maintenance fee and

$200 per case per disposable instrument, is a significant

financial investment [10�]. However, as seen in the

robotic surgical literature, the main factor responsible

for the higher cost per case occurs when the initial cost

of purchasing the robot is factored in [24]. According to

their 17 October 2008 filings with the Securities and

Exchange Commission (http://investor.intuitivesurgical.

com/phoenix.zhtml?c=122359&p=irol-sec), Intuitive

Surgical Inc. has installed units in over 1032 academic

and community hospital sites worldwide (766 in North

America). Furthermore, according to one robotic cost

analysis, profitability increases with increased caseload.

Therefore, it is in the best interest of hospitals to utilize

their purchase to its maximum by encouraging use of the

robot by various surgical specialties [25]. It is our opinion

that the actual purchase price of the robotic system is

somewhat superfluous to the TORS discussion, as the

only surgeons who will be formally trained and poten-

tially create programs in TORS will be those at institu-

tions that already have a robot, which was likely pur-

chased for other more high volume uses (i.e. cardiac or

prostatectomy), and the addition of TORS cases, as noted

above, will only add to the value of the original purchase

by a given institution.

ConclusionIt does appear that TORS affords potential advantages

and benefits over current treatment modalities. Such

advantages include better visualization and access to

tumors via a minimally invasive, less morbid approach,

resulting in better overall functional outcome. The

expensive cost of the robot may clearly hinder wide-

spread use; however, the benefits of decreased operative

time and decreased length of hospital stay could counter-

act this cost. A direct cost analysis study comparing

different treatment modalities is necessary to draw

further conclusions. Although long-term oncologic out-

comes are not yet available, current evidence of local

regional control is extremely promising. Further studies,

including long-term studies, are warranted. However,

based on the current literature from numerous institu-

tions, it does appear likely that ultimately the ends will

justify the means for TORS.

AcknowledgementsG.S.W., B.W.O’M. Jr, and S.C.D. should be regarded as joint firstauthors.

References and recommended readingPapers of particular interest, published within the annual period of review, havebeen highlighted as:� of special interest�� of outstanding interest

Additional references related to this topic can also be found in the CurrentWorld Literature section in this issue (pp. 135–136).

1 Hollands CM, Dixey LN. Robotic-assisted esophagoesophagostomy. J PediatrSurg 2002; 37:983–985; discussion 5.

2 Patel VR. Essential elements to the establishment and design of a successfulrobotic surgery programme. Int J Med Robot 2006; 2:28–35.

Transoral robotic surgery Weinstein et al. 131

3 Advincula AP, Song A. The role of robotic surgery in gynecology. Curr OpinObstet Gynecol 2007; 19:331–336.

4 Nifong LW, Chitwood WR, Pappas PS, et al. Robotic mitral valve surgery: aUnited States multicenter trial. J Thorac Cardiovasc Surg 2005; 129:1395–1404.

5 Hockstein NG, Nolan JP, O’Malley BW Jr, Woo YJ. Robotic micro-laryngeal surgery: a technical feasibility study using the da Vincisurgical robot and an airway mannequin. Laryngoscope 2005; 115:780–785.

6 Hockstein NG, Nolan JP, O’Malley BW Jr, Woo YJ. Robot-assisted phar-yngeal and laryngeal microsurgery: results of robotic cadaver dissections.Laryngoscope 2005; 115:1003–1008.

7 Weinstein GS, O’Malley BW Jr, Hockstein NG. Transoral robotic surgery:supraglottic laryngectomy in a canine model. Laryngoscope 2005;115:1315–1319.

8 O’Malley BW Jr, Weinstein GS, Hockstein NG. Transoral roboticsurgery (TORS): glottic microsurgery in a canine model. J Voice 2006;20:263–268.

9 O’Malley BW Jr, Weinstein GS, Snyder W, Hockstein NG. Transoral roboticsurgery (TORS) for base of tongue neoplasms. Laryngoscope 2006;116:1465–1472.

10

�Weinstein GS, O’Malley BW Jr, Snyder W, Hockstein NG. Transoral roboticsurgery: supraglottic partial laryngectomy. Ann Otol Rhinol Laryngol 2007;116:19–23.

This paper presents the first patients to undergo TORS supraglottic laryngectomyand compares and contrasts the technique with transoral laser superaglottic partiallaryngectomy.

11 Desai SC, Sung CK, Jang DW, Genden EM. Transoral robotic surgery using acarbon dioxide flexible laser for tumors of the upper aerodigestive tract.Laryngoscope 2008; 118:2187–2189.

12 Moore EJ, Olsen KD, Kasperbauer JL, et al. Transoral robotic surgicalextirpation of oropharyngeal neoplasms. In: 7th International Conferenceon Head and Neck Cancer. San Francisco, CA; 2008.

13 Boudreaux BA, Rosenthal EL, Magnuson JS, et al. Feasibility andsafety of robotic surgery for upper aerodigestive tract neoplasms. In: 7thInternational Conference on Head and Neck Cancer. San Francisco, CA;2008.

14 Hillel AT, Kapoor A, Simaan N, et al. Applications of robotics for laryngealsurgery. Otolaryngol Clin North Am 2008; 41:781–791; vii.

Intuitive Surgical, Inc., www.intuitivesurg

15

��Weinstein GS, O’Malley BW Jr, Snyder W, et al. Transoral robotic surgery:radical tonsillectomy. Arch Otolaryngol Head Neck Surg 2007; 133:1220–1226.

This report presents the first 27 patients to undergo TORS radical tonsillectomy fortonsillar squamous cell carcinoma. This paper not only provides an excellentaccount of the expected early results following the procedure, but also providesan excellent overview and understanding of the management of TORS patientsfrom the preoperative, intraoperative, and postoperative perspective.

16 Genden EM, Desai S, Sung CK. Transoral robotic surgery for the manage-ment of head and neck cancer: a preliminary experience. Head Neck 2008;31:283–289.

17 Desai S, Sung C, Genden EM. Transoral robotic surgery: indications andlimitations for management of lesions of the upper aerodigestive tract. In: 7thInternational Conference on Head and Neck Cancer. San Francisco, CA;2008.

18 Weinstein GS, O’Malley BW. A pilot study assessing surgical exposure duringtransoral robotic surgery (TORS) using the da Vinci robotic system. In: 7thInternational Conference on Head and Neck Cancer. San Francisco, CA; 2008.

19 Holsinger FC, McWhorter AJ, Menard M, et al. Transoral lateral oropharyn-gectomy for squamous cell carcinoma of the tonsillar region. I: Technique,complications, and functional results. Arch Otolaryngol Head Neck Surg2005; 131:583–591.

20 Machtay M, Perch S, Markiewicz D, et al. Combined surgery and post-operative radiotherapy for carcinoma of the base of radiotherapy for carci-noma of the base of tongue: analysis of treatment outcome and prognosticvalue of margin status. Head Neck 1997; 19:494–499.

21 Blavier A, Gaudissart Q, Cadiere GB, Nyssen AS. Comparison of learningcurves and skill transfer between classical and robotic laparoscopy accordingto the viewing conditions: implications for training. Am J Surg 2007; 194:115–121.

22 Nguyen NP, Moltz CC, Frank C, et al. Dysphagia following chemoradiation forlocally advanced head and neck cancer. Ann Oncol 2004; 15:383–388.

23 Koch WM. Head and neck surgery in the era of organ preservation therapy.Semin Oncol 2000; 27:5–12.

24 Lotan Y, Cadeddu JA, Gettman MT. The new economics of radical prosta-tectomy: cost comparison of open, laparoscopic and robot assisted techni-ques. J Urol 2004; 172:1431–1435.

25 Scales CD Jr, Jones PJ, Eisenstein EL, et al. Local cost structures and theeconomics of robot assisted radical prostatectomy. J Urol 2005; 174:2323–2329.

ical.com, PN873156 Rev. A, 11/2009