With the advent of minimally invasive techniques in thyroid surgery, conventional
open-access surgery for bilateral multinodular goiter was extended to encompass total
thyroidectomy. At the same time, the surgical approach to the thyroid gland was reduced
to a minimum. Totally endoscopic and video-assisted procedures through a minimal neck
incision were shown to be better tolerated by the patient, resulting in improved cosmesis
with no increase in surgical morbidity . In terms of perioperative and cosmetic
benefit, minimally invasive procedures are superior to conventional open surgery when
the surgical trauma inflicted to gain access to a fairly small target organ is substantial
(e.g. in laparoscopic cholecystectomy or adrenalectomy). Owing to the short distance
to the target organ, the thyroid and parathyroid glands stand least to gain from the
use of minimally invasive surgery. Keeping the neck incision as short as possible
was a first step towards achieving a better cosmetic result in the neck. Unfortunately,
only some 15% of thyroid patients were suitable candidates for this type of surgery.
Some other patients were dissatisfied because their expectations of having no visible
neck scar were not met. Nowadays, more and more people from all walks of life, whatever
their physical build or ethnic, geographic or cultural background, yearn for better
surgical cosmesis after thyroidectomy – why should they not opt straightaway for a
neck without a scar?
From a medical point of view, transaxillary thyroid surgery, the most popular approach
in order to avoid scarring, needs not only to be feasible but also as safe as conventional-access
thyroidectomy. From the patient perspective, the greater exposure of the thyroid facilitated
by the transaxillary route is not as important as the potential of this new surgical
approach for less damage in the short and long term. These considerations clinically
trump the issue of cost and reimbursement policies [2,3,4] because most endoscopic
procedures are less cost-effective than conventional open surgeries.
Is transaxillary, more specifically robot-assisted transaxillary surgery (RATS), as
it stands today [5,6,7], as safe as conventional open-access thyroid surgery (CATS)?
That is to say, is it safe enough for patients with uncomplicated, small, nonretrosternal
goiter, or low-risk localized thyroid cancer to serve as an alternative to Miccoli's
minimally invasive video-assisted technique [1,8]? At present, this is not at all
clear. Kang et al. , from the Yonsei University College of Medicine in Seoul (South
Korea), were the first to describe RATS in 2009. Within a 5-year period, a total of
10,000 RATS procedures (without being able to avoid double counting of procedures
reported by the same group in more than 1 publication) have been described in about
31 publications [4,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39].
Thirteen publications performed a head-to-head comparison of RATS and CATS [4,9,11,12,16,18,19,23,29,33,40,41,42].
All 31 RATS studies, 22 of which were from South Korea and 9 from the USA, were retrospective
by design (table 1). Systematically reviewed, they revealed three areas of concern
regarding the safety of RATS: (1) perioperative surgical quality control, (2) surgical
complications and (3) long-term oncological outcome in the setting of thyroid cancer.
Perioperative Surgical Quality Control
As a matter of fact, less than one third of the above-mentioned 31 RATS studies systematically
looked for complications typical of thyroidectomy, notably recurrent laryngeal nerve
and parathyroid function (table 1), as was outlined in the multicentric Scandinavian
 and German  quality-evaluating studies and in the ‘framework for new technology
assessment and safe implementation’ . Only some of the above-mentioned RATS studies
detailed the common, though often transient, complications of the transaxillary approach,
such as wound hematoma, postoperative neck and anterior chest pain and paresthesia.
There are two major complications that can result in serious long-term morbidity,
which are not encountered with CATS: injury to the aerodigestive tract (especially
the trachea) and the brachial plexus. Intraoperative monitoring of the brachial plexus
[46,47] was systematically performed in only 1 study , and postoperative evaluation
of brachial plexus function was exceptional. Not a single study utilized intraoperative
monitoring of the vagus nerve routinely, although this technology reliably permits
prediction of recurrent laryngeal nerve function after the operation , and helps
avoid bilateral vocal cord palsy by postponing completion of the other side in benign
goiter and low-risk differentiated thyroid cancer . Given the lack of outcome
data regarding the function of the recurrent laryngeal nerves, the brachial plexus
and the parathyroid glands, the published complication rates are likely underestimates
of the surgical risk inherent in RATS.
When total thyroidectomy without or with the addition of central node dissection was
performed (38% of all RATS patients reviewed), postoperative transient hypoparathyroidism
was the most frequent complication of RATS (36%; table 2). The unusually low rate
of permanent hypoparathyroidism, no more than 1/1,000 patients undergoing RATS, suggests
that the majority were not systematically screened for this specific condition. Among
the 29% of patients who underwent laryngoscopy after RATS on a routine basis, transient
and permanent vocal cord palsies were noted in 3.9 and 0.5%, respectively. The actual
rate of postoperative vocal cord dysfunction likewise seems to have been underdiagnosed,
because the absence of hoarseness cannot exclude asymptomatic vocal cord palsy.
Brachial plexus neuropathy during RATS is probably due to positioning of the arm to
gain sufficient access. Considering the absence of brachial plexus injury after CATS,
this is at least believed to be a contributory cause. The neuropathy resolved in 0.2%
of patients after RATS, but was permanent in 0.04%. As the transaxillary approach
usually requires elevation and flexion of the arm above the level of the head, brachial
plexus neuropathy is thought to be a complication inherent in RATS. Although most
of the time it is rare and transient, considerations of patient safety and possible
medicolegal consequences warrant a great deal of attention to prevent this type of
complication, more specifically, intraoperative electrophysiological monitoring and
postoperative clinical controls of the brachial plexus. Tracheal injury, reported
in 0.2% of patients undergoing RATS, seems to be another procedure-related surgical
complication that may be more experience-dependent than inherent in RATS as brachial
plexus neuropathy is considered to be. Although it can mostly be managed by closing
the tracheal defect endoscopically, tracheal injury remains a serious complication
of RATS that is not observed with CATS.
Even after 5 years of experience with RATS, it remains unclear whether the complication
rates of RATS and CATS are comparable or not. As a matter of fact, the reported incidence
of postoperative permanent hypoparathyroidism is, in all likelihood, a gross underestimate
of the actual rate. As many as one third of RATS patients qualify for total thyroidectomy,
so in the future, studies need to be designed such that postoperative hypoparathyroidism
after RATS is rigorously ascertained after total thyroidectomy has been performed.
Brachial plexus neuropathy and tracheal injury are grave procedure- and experience-related
complications, virtually unheard of in conventional thyroidectomy, which will hopefully
become rarer as surgeons progress on the learning curve.
Long-Term Oncological Risk
In stark contrast to the entire spectrum of thyroid diseases, 96% of RATS procedures
were carried out for papillary thyroid cancer (PTC). Strikingly, one third of these
patients revealed PTC with a T-category ≥T2, most of which fell into the T3 category
for minimal extrathyroidal extension. Even if one were to concede that minimal tumor
growth through the thyroid capsule does not entail a greater oncological risk [50,51],
it remains highly debatable whether a novel surgical technology should first be assessed
in patients with higher-risk cancer. Owing to the fairly short follow-up period of
less than 5 years, it is impossible to quantify the risk of neck recurrence after
the possible seeding of cells from a pT3 thyroid cancer. A few retrospective studies
used lymph node retrieval during node dissection and postoperative thyroglobulin levels
as a surrogate for clinical outcome [11,12,16,18,23,33,42]. Based on these criteria,
in some studies at least, RATS seemed to fare less well than CATS [12,16].
In the quest for superior surgical cosmesis, RATS represents the culmination in the
development of endoscopic thyroidectomy that is not minimally invasive in nature.
The Korean surgeons are to be congratulated on having introduced the robot into thyroid
surgery and having perfected its routine use. Robot thyroidectomy is now fit for clinical
use if one is prepared to foot the bill for the incremental direct (i.e. robotic system
including annual service fee and costs for single-use instruments) and indirect costs
(i.e. prolonged anesthesia and operative time).
The incremental expenditure associated with the use of robot surgery raises important
questions. Should we adopt a new technology solely because it affords a better cosmetic
result? Should the use of this technology be governed by economic principles, such
as a patient's or institution's willingness and ability to pay? Is it ethical to divert
significant financial resources from the healthcare system merely for improved cosmesis,
even though the novel technology has not yet been proven to be as safe as the former
gold standard? As a matter of fact, most series published on RATS do not measure up
to the usual requirements for surgical quality assessments. The surgical complication
rates of robot thyroidectomy have still to be rigorously ascertained beyond the early
postoperative period. Are we willing to accept serious complications over and above
the usual spectrum of surgical complications, just for the use of a ‘hip’ technology?
Are we willing to subject our patients to greater surgical morbidity until our learning
curve has leveled off? After the introduction of laparoscopic cholecystectomy in the
early 1990s, longer operative times were observed but also more severe complications
such as hilar and vascular damage than with the former standard of conventional open
cholecystectomy. In the light of such experience, the surgical community agreed that
learning curves involving longer operative times were acceptable, unlike learning
curves associated with higher and more serious complication rates. The same should
now apply to RATS. Should we really operate on higher-risk cancer patients in the
first place before the safety of a novel technology has been established in benign
Despite these reservations, RATS is a truly fascinating technology conferring superior
cosmesis compared to CATS. Rather than abandoning RATS , national registries should
be set up , flanked by rigorously conducted prospective, ideally randomized clinical
trials to determine the benefit-risk profiles of RATS and CATS in a head-to-head comparison.
The author declares not to have conflict of interest.