British Journal of Anaesthesia, 119 (S1): i72–i84 (2017)

doi: 10.1093/bja/aex383
Clinical Practice

The evolution of robotic surgery: surgical and

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

anaesthetic aspects
H. Ashrafian1,2, O. Clancy3, V. Grover3 and A. Darzi1,2,*
1
Institute of Global Health Innovation, Imperial College London, UK, 2Department of Surgery and Cancer,
Imperial College London, UK and 3Department of Anaesthesia and Critical Care Medicine, Royal Marsden
Hospital, London, UK
*Corresponding author. E-mail: a.darzi@imperial.ac.uk

Abstract
Robotic surgery pushes the frontiers of innovation in healthcare technology towards improved clinical outcomes. We dis-
cuss the evolution to five generations of robotic surgical platforms including stereotactic, endoscopic, bioinspired, microbots
on the millimetre scale, and the future development of autonomous systems. We examine the challenges, obstacles and
limitations of robotic surgery and its future potential including integrated real-time anatomical and immune-histological
imaging and data assimilation with improved visualisation, haptic feedback and robot-surgeon interactivity. We consider
current evidence, cost-effectiveness and the learning curve in relation to the surgical and anaesthetic journey, and what is
required to continue to realise improvements in surgical operative care. The innovative impact of this technology holds the
potential to achieve transformative clinical improvements. However, despite over 30 yr of incremental advances it remains
formative in its innovative disruption.

Key words: anesthesiology; robotic surgical procedures; surgery, computer-assisted; video-assisted surgery

The birth of robotic surgery took place at a time where there was several others suggested that a digitally programmed tool linked
an increasing demand for greater surgical precision and safer to a surgical cutting device could offer higher levels of operative
operations, and in an era where surgeons were increasingly adopt- accuracy when compared with conventional surgical methods.
ing minimal invasive surgical (MIS) technologies to enhance their This in-turn led to a paradigm shift in surgical thinking,
outcomes. The benefits of these minimally invasive approaches whereby surgical robots could potentially offer more than “an
(such as laparoscopy and thoracoscopy) included: (i) reduced equivalent-to-open operation with smaller incisions”, to one
wound access trauma, (ii) shorter hospital stay, (iii) improved visu- where an operation with a robot would allow a higher level of
alisation, (iv) less postoperative wound complications (ranging tissue discrimination, dissection and repair.
from wound infections to incisional hernias), and (v) less disfigure- Over 30 yr since their introduction, surgical robots occupy an
ment. As such they were designed to offer an equivalence to open influential role in today’s surgical ecology. Their increasing
surgery with less tissue trauma and speedier discharge that in application is derived from the technical benefits of modern
turn was anticipated to offer (vi) increased cost-efficacy. robotic platforms, but also from the conceptual science-fiction
The clinical introduction of the Puma 560 in 1985 led to the effect of robotics on modern society where robots represent the
first surgical robot being applied to perform selective brain biop- pre-eminence of cutting-edge technology.
sies. It was designed to outperform hand biopsies in terms of Technical advantages for the surgeon include: the potential
accuracy and surgical precision. Work with this robot and for better visualisation (higher magnification) with stereoscopic

C The Author 2017. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
V
For Permissions, please email: journals.permissions@oup.com

i72
 The evolution of robotic surgery | i73

M
icr
o-
an
d
Na us
no
-s omo
ca ton
Bi lin Au
om g 5th generation
im autonomous
icr
y
M
IS 4th generation
pr
in

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

cip microbots
le
s
3rd generation
bioinspired
m Concentric
y ste tube-based
es 2nd generation
slav Catheter-
r- endoscopic
ste based
Ma
Tendon- Electromagnetic
1st generation
driven steering
stereotaxic
Prototypes Mechanical
steering

Fig 1 The evolution of five generations of surgical robotics.

views; elimination of hand tremor allowing greater precision; and included the steam-driven “Flying Pigeon of Archytas” circa 400
improved manoeuvring as a result of the “robotic wrist” which in BC and the “string-coded” three-wheeled cart of Heron of
some systems allows up to seven degrees of freedom (angles at 
Alexandria circa 40 AD. Capek first introduced the word robot in
which surgeons can use their instruments to operate on target 1920 as a Czech term relating to “serf labour”,5 although the term
organs). There are improved kinematics where large external became popularised by Isaac Asimov in the 1940s.6 Over the next
movements of the surgical hands can be scaled down and trans- 40 yr, the concept of robotics became increasingly recognised in
formed to limited internal movements of the “robotic hands”. technological frontiers, receiving a prominent uplift during the
This in turn improves ergonomics that extend the surgical ability Space Race of the 1960s and the Apollo Mission by NASA.7 Its
to perform complex technical tasks in a limited space. Here the application in healthcare was commenced through the design of
surgeon is able to work in an ergonomic environment with less tools specifically geared at achieving a sophisticated level of
stress, achieving higher levels of concentration. The computer- precision for brain biopsies, where subtle inaccuracies could
ized nature of the surgical robot allows integration of real-time potentially lead to devastating outcomes. Driller and Neumann8
and previously recorded data utilisation, so that it could accom- reported on such an electromagnetic device in 1967, although
modate complex intra-operative factors such as compensating the first commercially and clinically available robot designed for
for the beating movement of the heart, making it unnecessary to a similar task was not available for another two decades. In the
stop the heart during cardiothoracic surgery. There may also be interim two main innovative were was taken: first there was a
less need for assistance once surgery is under way. global adoption of mechanical surgical reconstructive devices
We provide an overview of the evolution of modern surgical (anastomotic and haemostatic staplers) that had been present
robots (based on our modification1 2 of Camarillo and col- for over a century but had become industrially applied by the
leagues3 and the Rebello4 classification) applying the SEBMA Russian military.9 10 Second, Minimally Invasive Surgery (MIS)
acronym (Sterotaxic, Endoscopic, Bioinspired, Microbots and which again had been in experimental existence for over a cen-
Autonomous robots of the future) (Fig. 1). We identify their tury had become increasingly propagated, as was exemplified by
innovative role in operative healthcare and highlight essentials the first laparoscopic cholecystectomy by Mühe in 1985.11
for the anaesthetist. Are surgical robots as ground-breaking for
surgeons as the laryngeal mask airway is for anaesthetists? And
what are the likely advancers of this technology for the future? First generation – stereotaxic robots
The first surgical robot in clinical practice was the PUMA 200,
first utilised in the same year as the first laparoscopic cholecys-
Prototype robots tectomy.12 This robot was utilised for stereotaxic brain biopsy
The first robots were mechanical machines designed (or pro- with the surgeon placing the arms of the robot in a position to
grammed) to perform specific human-selected tasks. These perform its task. This device acted as a forerunner to a modified
 i74 | Ashrafian et al.

brain tumour excision device known as the Neuromate.13 between Computer Motion and Intuitive Surgical Inc, and as a
Similarly in orthopaedics, robots were introduced to perform result Intuitive Surgical acquired Computer Motion in 2003,
procedures that had a clear-cut mathematical and mechanical whereupon the da Vinci robot became the only commercially
strategy where tissue tactility and tissue vulnerability had been available endoscopic robotic system.
limited; the predictable geometry of the end-result was of crit- Although other endoscopic robotic platforms have subse-
ical importance in the outcome. As a result, devices such as the quently entered the market, the da Vinci system (Fig. 2) has
SCARA robot, the ROBODOC and the AcroBot were introduced14 retained its market dominance. This has been achieved through
to perform these tasks and were initiated from both industrial both its unique market place for several years and its multiple
(IBM) and non-industrial medial sources. At the end of this first features that allow: (1) a comfortable environment and ergo-
era of surgical robotics, several paradigms had been clarified. nomic console from which the surgeon can operate remotely
First, the robotic technology at the time required human surgi- from the patient, (2) 3-D imaging that offers accurate depth per-
cal review at the end of every step of robotic surgery and robots ception with multiple degrees of magnification, (3) in-line

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

could not perform multiple sequential tasks on human subjects “intuitive” eye-hand control of the instruments at the console, (4)
unsupervised (as the safety concern for tissue damage and life physiological tremor negation, and (5) movement of the mechani-
could not be guaranteed). So a master-slave paradigm was cal “endo-wrists” with 7 degrees of freedom beyond that of the
introduced where a robot was operating as a direct extension human wrist during laparoscopy/thoracoscopy with 4 degrees of
(or “slave” without independent choice) of its surgeon supervi- freedom. This allows the completion of complex microsurgical
sor (this is level 1 autonomy or pure human control according to tasks (one of the da Vinci’s predecessors, MIT’s Black Falcon19
the US Department of Defence Scale). Second, even the first gen- offered 8-degrees-of-freedom in articulation to overcome the loss
eration of robotic surgeons had questioned the whole future of of wrist articulation in laparoscopic/thoracoscopic procedures) (6)
robotic surgery15 in view of its limitations to simplistic and less Access to “hard-to-reach” areas of the body (such as the pelvis)
variable and static tissue platforms (such as the brain and where traditional open or MIS methods result in excessive torque
bones). As a result there was a need for increased dexterity and forces or the requirement for large incisions. This system has also
flexibility in operating tissue (soft and more elastic) targets. gone through several iterations ranging from the original da Vinci
that eventually went from a 3-arm system to a 4-arm system, the
da Vinci S with improved vision technology (including 3 D HD)
and an easier set-up, the da Vinci Si with further visual enhance-
Second generation – endoscopic robots ments and upgradeable architecture, and the da Vinci Xi with
The introduction of the second generation of surgical robots has enhanced vision and laser targeting in addition to capability for
resulted in the greatest expansion of the concept of robotic surgery adding future technologies. There are currently just under 4000 da
to date. This has been as a result of the introduction of soft-tissue Vinci robot systems worldwide, 66% are in the USA, 17% in
surgical capability such as the PROBOT from Imperial College Europe, 12.5% in Asia and the remainder at other sites.20
London that can remove pre-defined prostate gland volumes. It is The da Vinci device has been used in every organ system to
also because of the market-need for highly accurate robotic sys- varying degrees. But whilst it had initially been spearheaded as
tems that can augment established MIS surgical technology by a platform particularly well suited to cardiothoracic surgery,
building on established stereo-endoscopic platforms such as lapa- the robot has not been adopted universally for coronary artery
roscopy or thoracoscopy. Here surgical robots could potentially surgery21 as had been anticipated although it has seen adoption
offer four core advantages over traditional MIS surgery by over- in other cardiothoracic pathologies. Rather, it seems particularly
coming: (i) difficulty in access to tissue places and organ systems favoured in surgery of the pelvis (in urology and gynaecology)
as a result of anatomical restraints such as the pelvis or thoracic where for example, between 2003 and 2010, the national robot-
cavity causing torque and needing sheer physicality to address, assisted radical prostatectomy (RARP) adoption rate in the USA
(ii) instruments that lack precision for tasks such as vascular anas- increased from 0.7% to 42%.22
tomosis that are possible by hand but rendered more complex Whilst the da Vinci remains a clear market leader in the robotic
when performed via the intermediary of a basic MIS instrument as surgery market, other second generation robotic platforms exist
they can require counter-intuitive hand-eye coordination (iii) diffi- and include the University of Washington (UW) Raven23 which is a
culties in visualisation, which have traditionally been limited to 6-degree-of-freedom, master-slave system, programmable, modu-
2-D in MIS; and (iv) lack of tactile or haptic feedback from some lar robot (that has been devised to offer a degree of autonomy in
tissues whilst operating. surgery) and the German Aerospace Center’s DLR MicroSurge.24
Two of the best-known endoscopic robotic systems were These platforms share console and utility similarities such that for
simultaneously developed and introduced just before the mil- example, one platform has been utilised to train surgeons for tele-
lennium. These were The Zeus robotic system (Computer operative experience with one another.25
Motion, Goleta, CA, USA) which first became commercially As these second generation platforms are the most wide-
available in 1998, closely followed by the da Vinci robotic spread and established platforms in current clinical use, they
system (Intuitive Surgical Inc, Mountain View, CA, USA) in 2000 are the platform on to which many novel technological innova-
(Fig. 2).16 Between 1998–1999 both the da Vinci system and sub- tions are currently being applied. These range from improved
sequently the Zeus were successfully used in coronary artery surgical visibility and visual information transfer to improved
surgery as a proof-of-concept operative principle. Furthermore, robot-survival interactions ranging from haptic tactile feedback
the Zeus system was applied in Canada to complete the first to ease of application in surgical environments.
beating-heart coronary operation, and in 2001 this system was
applied to complete the Lindbergh operation; the first trans-
Atlantic operation performed by utilising a tele-robotic system
where the robotic surgical device (and French surgeons) were in
Third generation – bioinspired robots
New York and the patient was in Strasbourg in France.17 18 Advances in biomimicry, bionics and autobionics have been
Market forces led to competition over intellectual property evolving in parallel with modern robotics since the 1950s.26
 The evolution of robotic surgery | i75

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

Fig 2 The da Vinci Xi Surgical System (A) robotic arms, (B) console, (C) in the operating environment and (D) performing a coronary anastomosis (courtesy of Dr Leanne Harling).

However, during the development and evolution of second producing and utilising MIS instruments with articulated tips to
generation robots, MIS and endoscopic methodologies began allow MIS surgery to be possible in hard-to-reach areas with
adopting bioinspired technologies. In endoscopy the NOTES minimal access.
(Natural Orifice Transluminal Endoscopic Surgery) platform The third generation of surgical robots has adopted these
developed from a predilection for scarless surgery, whilst MIS principles of biomimicry and multiple articulation technology
surgery wanted to innovate to even less surgical exposure for multiple surgical pathologies. Most of these robots have
impact, so that multiple MIS ports was transformed into one- been designed in the past decade with a few exceptions (such as
port SPL (Single-port laparoscopy). The NOTES technology Ikuda’s microminiature SMA-Shape Memory Alloy27 servoactua-
allowed for surgical tools to be applied at the end of highly tor robot system). Current systems can be classified into
articulated snake-like endoscopes, whilst SPL offered a platform (i) Tendon-driven flexible systems such as Imperial College’s
for MIS technology to perform surgery with standard laparo- i-SNAKE and the CardioArm, and (ii) Catheter-navigated sys-
scopic/thoracoscopic equipment through enhanced ergonomics tems that have been derived for cardiovascular percutaneous
that had not previously been available at one site (transabdomi- intervention technology. These include mechanical-steering
nal, transumbilical or transluminal). These combined innova- systems such as the Amigo from Catheter Robotics Inc. and the
tions increasingly progressed to the next logical step of Magellan from Hansen Medical Inc. Catheter systems also
 i76 | Ashrafian et al.

include those that have electromagnetic steering such as Niobe perform individual pre-programmed tasks, the concept of fully
from Stereotaxis Inc., the CGCI from Magnetecs Inc. and sys- autonomous, human-level consciousness robots remains pre-
tems that can offer Magnetic Resonance Imaging (MRI) scanner- dominantly conceptual. Autonomous robots will likely benefit
guided steering28 that would allow rapid adoption for the large from enhanced machine-learning capability that will require
number of hospitals that have this modality in-house. next generation Turing Test intelligence (comparable to
As catheter systems are lengthened they can suffer from lack human-level intelligence and consciousness).35 36 They will take
of force of action at distance., This has led to an alternative the form of the first four generation of robots with added auton-
subgroup of third generation robots utilising concentric tube omous decision-making capability. These may range from a
devices, where this distance effect can be decreased. Both cyborg humanoid-type platform to a swarm-type system with
catheter-based and concentric tube systems have been proposed comparable swarm intelligence.
as the next generation in-use robotic system particularly suited
to tubular organ groups such as the cardiovascular, neurovascu-

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

lar, urological and respiratory/airway systems in addition to sur- Obstacles to robotic surgical adoption
gical imaging at the microscopic level (imaging foetuses29 in Cost
pregnancy for example) with a miniature “needle-sized” surgical
footprint. Concentric robot systems have a particular advantage A contemporary da Vinci robotic platform (Fig. 2) costs approxi-
as they can be very small in size, offer high dexterity, and have mately £1.55 million, with a yearly service charge of £125 000 and
the capability to move in highly curved paths.30 They also have instrument cost of approximately £2000 per case. This remains
the potential to be reproduced with increased automation beyond the financing capability of the majority of UK hospitals.
through methods such as 3-D printing.31 Whilst it is clear that robotic surgery costs remain high compared
with open and MIS cases,37 38 there is increasing evidence to sug-
gest the long-term cost efficacy of robotic approaches compared
Fourth generation – microbots with traditional open operations (such as for radical prostatecto-
mies). This was demonstrated in terms of lower inpatient admis-
The concept of microscopic robots has been present in the pub-
sions, hospital bed-days and excess bed-days for robotic surgery
lic sphere for some time and even gained international promi-
that in turn suggest more cost benefit of robotic procedures in the
nence from science fiction films such as “Fantastic Voyage”
long term. However these effects were non-significant (at 360 days
from 1966. Robots at a microscopic level could enter the body
£779 vs £1242 and at 1080 days £2122 vs £2889).39 A similar trend
with minimal surgical footprint and work as a solitary robot or
has also been reported in European40 and US private health insur-
more likely as a group of robots to image and treat not only con-
ance and the Medicare reimbursement system where long-term
ventional surgical diseases but also non-surgical diseases such
(over three years) robot prostatectomy saved $1451 per case. This
as infective and immune processes that could be managed at a
is largely as a result of lower overall complications, lower inconti-
cellular level. Microbots occupy a millimetre scale ( a fraction of
nence and lower sexual dysfunction costs with a 38–99% probabil-
a millimetre to several millimetres but larger than the nano-
ity that robotic prostatectomy provides cost savings according to
metre scale).
R Monte-Carlo probabilistic sensitivity analysis.41 A formal Health
Currently capsule endoscopes (such as the PillCamV WCEs-
R Technological economic assessment of the cost-effectiveness of
Wireless capsule endoscopes ranging from PillCamV SB3,
R R R laparoscopic and robotic surgery revealed a 10-yr time horizon
PillCamV Colon2, PillCamV UGI, and PillCamV PATENCY) are in
incremental cost per QALY of<£30 000 for robotic prostatectomy
clinical use, although these function as predominantly imaging
(providing>150 procedures are performed each year). Superiority
modalities that are passively mobile capsules being transported
of robotic outcomes was predominantly because of differences in
by the peristaltic motility of the gastrointestinal system whilst
positive margin rate (which had some limitations on data capture).
taking images of the gastrointestinal tract. The next generation
This identified that with an NHS-type financing system, fixed cap-
of these microbots would include further advances in each com-
ital and maintenance charges for the robotic platforms remain
ponent of these robots ranging from vision, locomotion, local-
core barriers to adoption, although this could be negated to a
isation, telemetry, power, diagnosis and tissue manipulation.32
degree by commercial negotiation and achieving high volumes of
Just as in third generation robots, these microbots will benefit
cases in each centre (more than 100–150 annual cases).42
from advances in biomimicry where for example locomotion
There remain a handful of companies with one dominant
would be based on electromagnetic steering or even autono-
market leader offering these robotics platforms, predominantly
mous locomotion based on insect-like, fish-like, snake-like or
of the second generation. One method of overcoming these cost
bacteria or parasite-like (flagellate)33 technology. These systems
effects is to increase market completion by a concerted effort of
will have the capability of working with established imaging
clinicians, robot scientists and policy-makers supporting new
modalities but can also offer a higher resolution micrometre
entrants to the market. Additionally, the wider adoption of
real-time imaging of diseases and patient anatomy. This is an
these devices based on appropriate evidence may also offer
opportunity to take advantage of imaging systems that allow
improved cost schemes to allow their utilisation. This may be
both electromagnetic navigation and imaging at the same time;
coupled to enhanced economic strategies such as institutional
these microbots could be integrated with MRI systems34 (just as
sharing of devices and costs to ensure easier access and financ-
in third generation robots), where the scanner will offer external
ing for robotic surgery in a wider patient population.
imaging and electromagnetic navigation whilst the microbot(s)
will offer internal imaging and disease treatments.
Learning curve
One of the core advantages of robotic surgery has been its
Fifth generation – autonomous systems “promise” to offer a shorter learning curve when compared with
Whilst systems since the first generation of surgical robots have MIS platforms as a result of its “intuitive” technical adoption.
been designed to carry a degree of autonomous capacity to This has been demonstrated in some studies,43 though there is
 The evolution of robotic surgery | i77

no large scale randomized evidence to support this finding at Whilst in MIS many of these devices simply need to function
this time. Limitations for this evidence suggest that in the cur- within the constraint of available port sizes, the nature of tele-
rent era, most practising surgeons almost universally become robotic surgery not directly in proximity to the patient, limits
familiarized with MIS techniques before they go on to practice the number of devices available on the robotic platforms. Two
robotic surgery, so that a true comparison of their learning processes are underway to overcome this issue: (i) the robotic
curves could be biased in this setting. Additionally, one system- companies are designing and implementing their own devices
atic review of the literature has identified that the measurement to accommodate surgical need (for example the da Vinci Xi
of MIS learning curves remains multifaceted and ill-defined in EndoWrist Stapler 30 Instruments And Reloads), that can be
the majority of studies, with only a handful of analyses utilizing expensive for the robotic company who are not traditionally
the recognized CUSUM model of assessing trends in multiple designers of MIS equipment; and (ii) device companies can
surgical outcomes within a clinical setting.44 Future work will design instruments that can work on the robotic platforms.
need to offer increased robustness in data comparability However this also has barriers regarding the ownership of

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

between robotic and MIS learning curves if this factor is to be device intellectual property and its transfer between device
used as a source to decide on the platform utilized in a particu- companies and robotic companies, and carries cost considera-
lar clinical setting. tions that would affect healthcare institutions and their
Learning curves to achieve surgical proficiency with robotic patients. One possible solution lies within the academic sector
platforms differ widely between procedures, pathologies and where surgeons and engineers can attempt to design and solve
anatomical sites (just as in in open and MIS procedures). As a robotic surgical needs at an equitable price.
result, comparing learning curves and proficiency rates between
procedures and techniques can be problematic as utilizing
open, MIS and robotic procedures to achieve the same end
Overview of clinical outcomes and current evidence
result may not follow the same steps and therefore are difficult Robotic surgery has been described by some as the natural evo-
to compare clinically and statistically. It has been suggested lution to laparoscopic surgery along the minimally invasive con-
that the technical advantages of robotic surgery, which reduce tinuum.48 Essentially offering similar advantages in reducing
the cognitive and physical demands of minimally invasive sur- the systemic inflammatory and metabolic insult whilst provid-
gery, would ameliorate the challenging surgical learning curve. ing improved precision and accuracy in surgical technique
This would allow surgeons, including those without previous because of superior 3 D dexterity and it offers potential in future
laparoscopic experience, to provide the benefits of minimally developments including digitally enhanced analysis of tissues
invasive surgery to their patients. However, the evidence to with integrated immunofluorescence and improved outcomes
support these assertions is limited and there is a paucity of in benign and malignant disease.49
comparative data. Utilising an example of robot-assisted laparo- Current robotic surgical evidence points towards a convinc-
scopic radical prostatectomy (RALP), sources identify a learning ing reduction in postoperative surgical and non-surgical compli-
curve that ranges between 12 and 250 cases based on the defini- cations, reduced blood loss, improved recovery rates, improved
tion of “learning curve” utilised.45 It typically takes 150–250 cosmesis and reduced length of stay in comparison with open
cases to achieve the learning curve for operative time, though surgery.50–52 The comparison with MIS however is equivocal,
the learning curve for oncological and biochemical outcomes in although several studies do show some advantages in length of
this case lies at approximately 750 cases.46 stay, conversion rate and estimated blood loss.53–55
Concerns regarding robotic surgery predominantly focus on
increased length of operating time (and cost), although gains in
Operational and environmental limitations improved recovery times and benefits of robotic techniques
Most current robotic platforms carry multiple operational chal- in more complex surgery and with specialist groups may go
lenges for day-to-day application. These include (i) sufficient some way to counter this. One area of particular superiority
theatre space that can accommodate the large dimensions of in comparison to MIS is that of a reduced conversion to open
current devices, (ii) theatre staff (not only surgeons and anaes- surgical technique, which has particular benefit in obese and
thetists) that are familiar with the robotic platform set-up, elderly patient groups.56
(iii) managing the complex ergonomics of a busy theatre space There is evidence from a variety of robotic surgical specialities
with a robotic device in-situ and (iv) the ability to minimise that in comparison to the non-obese population there is no
robotic operating room turnover time. Attempts at managing increase in intraoperative or postoperative complications, conver-
the latter point derive from the application of “pit-stop” models sion to laparotomy or operative time in obese patients.57–59 In fact
originally practised in the motor racing industry for rapid but in some specialities they have demonstrated a shorter operative
exact changeovers within a surgical environment. This includes time in obese patients using robotic surgery in comparison with
the use of multiple anaesthetic teams and anaesthetic rooms.47 open surgery,60 and in comparison with laparoscopic surgery it
Addressing the former points however will increasingly rely on has been shown that return of bowel function and discharge
adoption of the next generation of robots with a smaller surgical home is faster by 24 h, with otherwise comparable operative time,
footprint to allow increased ease of use and accessibility to blood loss, conversion rates, resection margins and complica-
these devices. Smaller devices would also offer ease of transport tions.61 However, robotic procedures do not universally demon-
which in turn could increase the possibility of sharing devices strate speedier results, such that increased operative times
as part of a business model, or ease of transporting devices to and length of stay has been reported when using robotic vs
manufacturers for repairs and updates. laparoscopic techniques in bariatric surgery.61
Outcomes in the elderly have been shown to be better after
robotic surgery in comparison with open surgery, with reduced
Ancillary Equipment and intellectual property surgical and medical complications, improved length of stay
MIS operations remain highly dependent on ancillary equip- and quicker discharge home. This may be as a result of reduced
ment such as stapling guns, scissors and haemostatic devices. blood loss and transfusion rates, alongside reduced wound
 i78 | Ashrafian et al.

Table 1 Perioperative considerations for robotic surgery

Perioperative Considerations Rationale

Stage

Preoperative Environmental considerations and ergonomics – may ben- Need space and ergonomic layout for table, robot, surgical,
efit from visiting other departments if new to robotics anaesthetic and nursing teams for safe and efficient care49 69
Multi-disciplinary team training – consider simulation Robot set up and patient positioning takes time and experi-
ence but can be efficient. Critical incidents (cardiac arrest
for example) require special consideration49 69
Preoperative assessment and multi-disciplinary deci- More high risk patients are considered for minimally invasive
sion-making regarding benefits and risks of robotic surgery and robotic surgery has particular physiological

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

surgery and anaesthetic technique impact
Induction Induce on theatre table Avoids transfer and can increase efficiency
Consider second anaesthetic team to anaesthetise Minimize turnover time and improve robot utilisation
further patients efficiency
Consider opioid based neuroaxial block Evidence base for improved pain management70 71
Positioning of lines in most accessible position/away Improved ergonomics and access, reducing complications
from robot
Invasive lines only in high risk patients Routine cases may be managed without invasive monitoring
Trial of positioning in high risk patients Ensure patient can tolerate Trendelenburg positioning before
robot docking
Nasogastric tube not always required Regurgitation may be absent in short low-risk cases
Maintenance Close attention to positioning and padding – gamgee for Complications because of peripheral nerve injury, lower limb
arms, inflated gloves to protect hands. Lacrilube to compartment syndromes, oedema and ocular compro-
eyes, eyes further padded and taped with tegaderm, mise72 73
with regular checks
Shoulder supports and horizontal bar to protect face Protection of patient from moving and from robot arm
from robot arm movement
Consider TIVA approach Potential benefits for cancer management74
Maintain muscle paralysis (may not be required with Movement may cause significant patient injury once robot
remifentanil infusion) docked
Restrictive fluid management where possible Avoid complications of oedema
Use of mobile phone app spirit level to measure degree Achieve accurate angle of Trendelenburg
of Trendelenburg
Emergence Consider degree of oedema Particular issue in prolonged steep head down surgery
Cuff leak check, consider airway exchange catheter
Consider overnight intubation and dexamethasone in Aim to avoid difficult emergency re-intubation72
high risk cases
Postoperative Enhanced recovery principles Consolidate benefits of minimally invasive surgery and ensure
Post anaesthetic care unit/critical care for high-risk optimal outcomes particularly for high risk or frail patients
patients

and fascial complications despite longer operating times.62 Assessing all available data-sets, we identified 108 studies on
Interestingly several studies across a variety of surgical special- 14 448 patients. Those reporting on robotic vs open surgery (OS)
ities have found there to be no differences in outcomes between included 11 RCTs and 39 prospective studies, which together
younger and older patient groups having robotic surgery,63–65 demonstrated lower blood loss at 50.5%, lower transfusion rate
indicating age alone is not a risk factor. One study demonstrated at 27.2%, lower length of hospital stay at 69.5%, and reduction of
that older patients having robotic surgery vs younger patients 30-day overall complication rate at 63.7% in favour of robotic
having open surgery had significantly lower early complication surgery when compared with open surgery. For robotic vs MIS,
rates (17% vs 59%).52 there were 21 RCTs and 37 prospective studies, which demon-
In terms of specialist surgeries, the benefits of robotic tech- strated mildly reduced blood loss at 85.3% and transfusion rate
niques have enabled increasingly complex procedures such as at 62.1% in favour of robotic surgery but similar length of hospi-
retroperitoneal lymph node clearance for treating testicular tal stay (98.2%) and 30-day overall complication rate (98.8%)
cancer. Using an open approach would be extremely invasive, when robotic surgery was compared with MIS. In both compari-
but using a robotic approach there is potential for a return to sons, robotic surgery prolonged operative time (7.3% longer
full physical fitness within three weeks.66 These are important than open surgery and 13.5% longer than MIS). In our analysis,
considerations in terms of improving outcomes in cancer man- for the first 30 yr, there were relatively few RCTs, and those that
agement and patient satisfaction and quality of care; these are were present suffered from inadequate statistical power and a
outcome markers that can demonstrate the incremental gains high risk of bias. As a result, there has been a recent communal
offered by robotic surgery. effort to produce high quality randomized data on robotic out-
We performed a systematic review67 of all the papers in the comes. For example the recent68 Australian randomised con-
literature for the first 30 yr of robotic surgery (1985–2015). trolled phase 3 study comparing robot-assisted laparoscopic
 The evolution of robotic surgery | i79

prostatectomy and open radical retropubic prostatectomy, morbid obesity, this can be managed with protective pressure
revealed that at 12 weeks, there was no significant difference in control ventilatory strategies including positive end-expiratory
standard oncological or quality of life outcomes. Studies such as pressure and optimal fluid management, plus invasive lines
this (which was the first such RCT in prostatectomy surgery), and vasopressor support in the high-risk patient. Furthermore,
with much longer outcome data (years rather than weeks) will a potential technique in counselled high-risk morbidly obese
help clarify the decision-making in robotic surgery for the patients is to use steep Trendelenburg in the anaesthetic room
future. Lastly, the notable lack of evidence for robotic proce- after induction of anaesthesia and tracheal intubation to moni-
dures requires the development of validated robot-specific tor respiratory and cardiovascular effects. If significantly com-
methodological tools to assess and evaluate this evolving tech- promised in this position, surgery may be conducted as an open
nology. This includes methodologies to appraise well-defined procedure, or postponed for further discussion and decision-
clinical endpoints including specific quality of life (QoL) and making. Patients are fully consented for this before induction.
patient reported outcome measures (PROMS) combined with Awaited are the results of a current study looking at ventilatory

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

robust cost-efficacy and economic analyses. strategy and pulmonary outcomes in robotic surgery (AVATaR),
a collaboration between groups in Brazil, Italy, Holland and
Germany (clinicaltrials.gov number NCT02989415).
Anaesthetic perspectives Limited access to the patient once docked, the precarious
Whilst all care providers consider the patients’ surgical journey positioning of the patient, and the absolute need for no move-
through the treatment pathway by appraising whether the ment intraoperatively raises several challenges for the anaes-
advances made by robotic surgery demonstrate benefits in thetist. Care and attention to positioning and padding is one of
terms of outcomes and whether these outweigh associated the most crucial elements. There is some evidence that there is
risks, anaesthetists also consider several core anaesthetic- a slightly higher incidence of peripheral nerve injury in the
specific factors as a part of patient management (Table 1). upper and lower limbs for robotic surgery vs laparoscopic sur-
The anaesthetic risks associated with robotic surgery are gery, and whilst in many cases this resolves in under six weeks,
largely associated with length of operating time and position- it may persist for more than six months.72 Shoulder braces and
ing.72 The most frequent complications are peripheral neuropa- beanbags have been specifically implicated in brachial plexus
thies, corneal abrasions, vascular complications including injuries and so should be avoided.79 Some centres advocate
compartment syndrome, rhabdomyolysis and thromboembolic chest banding to stabilise position, but this may compromise
disease, and the effects of oedema (most significantly cerebral, lung compliance.49 Compartment syndrome and rhabdomyoly-
ocular and airway), which are elegantly described in a recent sis are rare but significant consequence of positioning, long pro-
review.72 Additionally, obesity is widely accepted as increasing cedures and tight leg braces.53 Attention needs to be paid
the risk of perioperative complications in robotic surgery. to ensure straps do not compromise blood supply, and addition-
Robotic prostatectomy is considered to be the index operation ally that systemic cardiovascular integrity is maintained.
and studies in obese patients undergoing robotic-assisted radi- Gluteal compartment syndrome is a specific risk, which
cal prostatectomy (RARP) show higher complication rates com- although rare80 81 has significant consequences and so gluteal
pared with non-obese individuals, such that in some cases the cushioning is recommended.82 Additional large bore access
robotic approach fails to decrease the risks of obesity-associated with a long venous line connection and muscle paralysis (in the
surgical complications.75 Together these complications form the form of neuromuscular blocking agents or a remifentanil
basis for the anaesthetic technique. infusion) would also be advised. High postoperative vigilance
As the learning curve for robotic surgery has progressed for complications and appropriate management protocols
from both a surgical and anaesthetic perspective, a reduction in optimise outcome and patient satisfaction.
surgical time and perceived requirement for invasive lines has After patient safety is considered, the aim for anaesthesia is
been observed. However as confidence grows, more complex to contribute to the incremental gains offered by robotic surgery
surgeries and patients with increasing comorbidities will be by providing optimal fluid management, analgesia, reducing
considered for surgery, highlighting the need for thorough postoperative nausea and vomiting (PONV) and cognitive dys-
preoperative assessment, multidisciplinary decision-making function, improving recovery and discharge times and overall
and assessment on a case-by-case basis. patient satisfaction. With a newly realised opportunity to con-
Robotic surgery has been utilised in a range of specialities tribute to overall cancer outcomes with evidence suggesting the
including urology, gynaecology, cardiac, thoracic, upper and superiority of total i.v. anaesthesia (TIVA) over volatile anaes-
lower gastrointestinal and endocrine surgery. Anaesthetic thetic techniques,74 it seems sensible to consider the use of
approaches for each procedure will be unique. However, similar TIVA in robotic procedures for oncological surgery. Despite the
considerations will be based around patient positioning, the advantages of TIVA in terms of PONV and recovery times, there
physiological impact of surgery, and potential patient safety currently remains limited evidence to make recommendations
issues including the limitations of restricted patient access. On for its use in all types of robotic surgery.83 Postoperative analge-
an organisation level teamwork and communication between sia may be improved with neuroaxial techniques such as intra-
anaesthetic staff, surgical staff and nursing teams is imperative thecal opioids as reduced systemic opiate use, reduced pain
in terms of robotic setup, and docking and undocking, particu- scores and increased patient and nursing staff satisfaction have
larly in the event of an emergency such as cardiac arrest.49 been demonstrated with this approach.70 71
Simulation has been recommended to improve efficiency in Oedema can become problematic, particularly in dependent
these circumstances.69 areas after long surgeries in the steep head down position.
The required position for many types of robotic procedures Laryngeal oedema may occur, and presents as respiratory dis-
is the steep Trendelenburg.76–78 The respiratory and cardiovas- tress and airway compromise in the immediate postoperative
cular implications of this extreme and exaggerated position period. The overall incidence of reintubation after robotic
have been well described.49 69 76–78 However, in the majority of surgery is around 0.7%, and delayed extubation 3.5%; but
patients, including those with chronic respiratory disease and the incidence of airway oedema may be up to 26%.72 Many
 i80 | Ashrafian et al.

anaesthetists perform direct laryngoscopy and use a leak test Visualization

before extubation;84 some centres also use airway catheters in
Dynamic View Expansion or Mosaicing have already been intro-
case of the need for re-intubation. Facial and periorbital oedema
duced in MIS and can offer robotic platforms a wider field of
can be indicative of laryngeal oedema, and be implicated in
view than standard MIS camera technology.89 90 Whilst multi-
other problems such as corneal abrasions, as oedema may
modal visualisation technology is already being applied in
cause eyelids to separate. Vigilance, careful lubrication, taping,
robotic procedures, such as augmented reality (overlaying of CT,
padding, and positioning of drapes is advised, and consideration
MRI, ultrasound or other imaging) to guide intraoperative deci-
to inserting a nasogastric tube to prevent gastric content con-
sions, these techniques continue to need enhancement by
tamination. The increase in intraocular pressure intraopera-
improved depth perception with inverse realism, and to offer
tively has also meant that glaucoma can be considered as a
see-through vision of an embedded virtual object while sustain-
relative contraindication to some forms of robotic surgery.
ing the vision of standard operative anatomical landmarks.91
Visual loss is rare, but has been described in association with
Real-time intraoperative ultrasound (USS) had added a tech-

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

posterior ischaemic optic neuropathy.85–87 Cerebral oedema can
nically simple yet diagnostically powerful imaging modality to
also be a significant complication causing confusion or reduced
robotic surgery and is used extensively for robotic partial neph-
levels of consciousness postoperatively. The pathogenesis is
rectomy. It has potential in prostatectomy and other procedures
likely because of increased venous pressure in the
where it can help differentiate tissues based on their
Trendelenburg position with pneumoperitoneum leading to
echogenicity.92 93 A predominance of 3-D systems with smaller
increased intracranial pressure and capillary leak. Preventative
sized cameras at 4 mm (such as the Visionsense VSiii) and
strategies include limiting operative time, minimising the angle
smaller will offer stereoscopy with increased use of other ports
of Trendelenburg, limiting insufflation pressure to 8 mm Hg
for surgery. Tissue imaging with photodynamic capture and
where possible, and fluid restriction.88 It is also advisable to
enhanced microscopy ranging from Narrow Band Imaging (NBI),
maintain a normal end-tidal carbon dioxide concentration.
Fluorescence Lifetime Imaging (FLIM), Optical coherence tomog-
High-risk patients can be kept intubated for a period of time
raphy and Flexible Confocal Microscopy (FCM)94–96 will offer
postoperatively. Despite a potential delay to extubation, the
increased real-time visual histological data that can identify
recovery and discharge from ITU and overall hospital stay can
tumour cells and margins. Robotic cameras can emit and quan-
remain lower than that after open surgery. It is clear that a fluid
tify tissue autofluorescence of reflectance spectra to highlight
restriction strategy may help to reduce complications in associ-
any microscopic surgical pathology for resection. This coupled
ation with oedema, but this obviously needs to be balanced
with enhanced diagnostic computation, neuromorphic visual
against compromise to the cardiovascular and renal systems.69
processing tools and machine-learning algorithms97 will allow
As the experience with robotic technology has expanded for
imaging at a number of tissue scales (including molecules, cells
surgeons, the anaesthetic considerations for avoiding potential
tissues and organs)98 99 through a new generation of real-time
hazards have become better understood and realised, as have
disease diagnostic capability well beyond that of traditional
the postoperative benefits of the robotic technique. However,
assessment tools.
there remain many unresolved questions. More multidiscipli-
nary considerations and evidence to support the benefits of dif-
ferent surgico-anaesthetic approaches for individual cases Somatosensory perception and beyond
should be investigated. Overall we need to aim towards conclu- Operating with a refined sense of touch to assess bodily tissues
sive evidence as to which anaesthetic and analgesic approach can be of critical importance in differentiating pathology and
offers the most postoperative benefits to patients. We should be making on-table surgical decisions. This has been largely lack-
able to justify the cost of interventions in terms of patient bene- ing or exists at a blunted level in current MIS systems, although
fits from a physiological impact, pathological outcome this has not necessarily been associated with poor results.100
and quality satisfaction perspective, which will also ensure Nevertheless, increased tactility will offer a new level of tissue
that multidisciplinary teams and patients can make fully perception for robotic surgeons that could be translated into
informed decisions and choices, particularly those in high-risk increased precision and safety. Increased understanding of the
categories. neurophysiology and mechano-transduction of tactile percep-
tion through vibrotactile cueing101 and traction loads are allow-
ing the next generation of wearable haptic systems for robotic
The future of robotic surgery
platforms offering tactile enhancement.102 103 This increased
The future of robotic surgery hinges on five core TECAT dimen- tactility will allow surgery through ever smaller operative uten-
sions: (i) Technology, continual application of advancing and next sils with advanced kinematics and higher degree-of-freedom
generation technologies to offer improved surgical precision in joint capacity.104
a wider range of cases with increased usability to achieve better Every element of the operative environment can now con-
clinical outcomes, (ii) Evidence, increased evidence to select the tribute to robot surgical decision-making. One prominent novel
best robotic platforms for the most appropriate population- example is Imperial College’s intelligent scalpel or i-Knife which
base, (iii) Cost, cost-efficacy for individuals, institutions and can utilise diathermy smoke to offer pathological diagnoses
nations to afford robotic surgical healthcare (iv) Awareness, (for example cancer vs non-cancerous tissue) based on the
increased societal and patient awareness and comfort in having metabolic profile of the tissues being diathermied.105
surgical procedures performed when appropriate, and finally
(v) Training, enhanced training of surgical, anaesthetic and asso-
ciated healthcare staff to have increased familiarity and
Robot-surgeon interactions
improved team outcomes when applying surgical robotics. Many novel robotic technologies focus on offering technology to
Whilst we have already highlighted some of these factors in the enhance surgical decisions, where the surgeon is the hub and
text, this section will focus on future technologies that can information can be given to the surgeon who then processes
enhance the next generation of robotic procedures. this to formulate a conscious plan which in turn is executed
 The evolution of robotic surgery | i81

manually into an operative manoeuvre. Increasingly an which can be enhanced within the simulation environment.
additional approach has been generated where the robot can These factors all require underpinning with the highest levels of
also pick up information from the surgeon to support surgical evidence to develop the optimum multi-disciplinary approaches
decisions in parallel as a “supportive partner”. to integrate surgical, anaesthetic and allied specialties to deliver
For example, neural integration is being developed to allow robotic surgery into its next stage of innovative evolution.
robots to derive information from their surgeon thorough EEGs,
magnetic EEGs and near infra-red spectroscopy (NIRS), so that it
may (i) utilise machine learning algorithms to help record the steps
Acknowledgements
of an operation and (ii) possibly offer a modification of the surgical We would like to express our thanks to Dr Leanne Harling
environment to optimise surgical precision, accuracy and safety. for images of the da Vinci Xi system.
Video-oculography (eye-tracking) is a non-invasive technology
that can be utilised to assess regional brain activity106 through

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

optical topography (OT) and gaze-behaviour, to better understand Authors’ contributions
surgeon behaviour and decision-making which in turn can be
Study design/planning: H.A. and A.D.
applied to enhance the next generation of surgical trainees.
Study conduct: all authors
Additionally, gaze-contingent information can also offer the
Data analysis: all authors
assessment of saccadic eye movements and ocular vergence to
enable an understanding of surgeon 3 D depth perception through Writing paper: all authors
wearable eye-trackers. This also allows a deeper appreciation of Revising paper: all authors
real-time surgical behaviour and decision-making that can help
augment training and improve surgical safety.107 However it also
carries a novel value of “feeding-back” modified visual informa-
Declaration of interest
tion through a tele-robotic module to allow a surgeon to overcome None declared.
operative environmental complexities. For example, gaze-
contingency information can be used to overcome the difficulties
of operating on a beating heart in off-pump cardiac surgery, by
References
generating a non-moving “phantom heart” image so as to enable 1. Athanasiou T, Ashrafian H, Rao C, Yang GZ, Darzi A. The
the surgeon to visualise a still heart for performing surgery.108 tipping point of robotic surgery in healthcare: from master-
slave to flexible access bio-inspired platforms. Surg Technol
Int 2011; 21: 28–34
2. Bergeles C, Yang GZ. From passive tool holders to microsur-
Conclusions geons: safer, smaller, smarter surgical robots. IEEE Trans
Robotic surgery applies actuators and computer control into all Biomed Eng 2014; 61: 1565–76
surgical specialties with an overarching aim to combine a mini- 3. Camarillo DB, Krummel TM, Salisbury JK Jr. Robotic tech-
mally invasive approach with improved surgical precision and nology in surgery: past, present, and future. Am J Surg 2004;
accuracy. The healthcare sector’s “learning curve” for robotic 188: 2S–15S
technologies has so far met with some challenge and resistance 4. Rebello KJ. Applications of MEMS in surgery. Proc IEEE 2004;
including pertinent concerns towards cost and lack of 92: 43–55
evidence.However the benefits in terms of postoperative recov- 5. Capek K. R.U.R. Mineola, New York: Dover Publications Inc.,
ery and advantages in particular patient groups are becoming 2001
increasingly realised in selective procedures and cases. Future 6. Asimov I. The Complete Robot (Robot Series). London: Voyager,
promises towards the integration of current robotic systems 1983
with advanced real-time anatomical and immunohistological 7. Berry CA, Hoffler GW, Jernigan CA, Kerwin JP, Mohler SR.
imaging technologies, alongside more discrete and manoeu- History of space medicine: the formative years at NASA.
vrable instruments with improved visualisation and tactile Aviat Space Environ Med 2009; 80: 345–52
feedback, offer exciting surgical opportunities. These opportuni- 8. Driller J, Neumann G. An electromagnetic biopsy device.
ties have the potential to translate into improved clinical out- IEEE Trans Biomed Eng 1967; BE-14: 52–3
comes in terms of cancer survival and overall quality of care for 9. Anan’Ev MG, Antoshina NV, IuIa G. [Tissue suture appara-
a wide range of complex and high-risk patients. These proce- tus with tantalum staples]. Eksp Khirurgiia 1957; 2: 28–35
dures have the potential to offer improvements in stronger 10. Kaintanov NN, Petrova NP, Iurasova NV. [A multiple sta-
measures of outcome evidence such as quality of life, cost- pling device for longitudinal suturing of vessels and a new
efficiency and patient reported outcome measures (PROMS). vessel suturing apparatus for end to end anastomosis].
In its current form, robotic surgery continues to have the Khirurgiia (Mosk) 1966; 42: 135–8
potential to become dramatically transformative in global 11. Blum CA, Adams DB. Who did the first laparoscopic chole-
healthcare, although it has not achieved this accolade yet. The cystectomy? J Minim Access Surg 2011; 7: 165–8
future of this field includes exposure to continual innovation in 12. Kwoh YS, Hou J, Jonckheere EA, Hayati S. A robot with
technology, but also costing strategies and healthcare value net- improved absolute positioning accuracy for CT guided ster-
works, to allow the next generation of robotic platforms to gain eotactic brain surgery. IEEE Trans Biomed Eng 1988; 35:
establishment in the modern healthcare market. This also 153–60
requires training and adoption of evidence-based robotic 13. Benabid AL, Cinquin P, Lavalle S, Le Bas JF, Demongeot J, de
approaches and gaining experience and confidence in the skills Rougemont J. Computer-driven robot for stereotactic sur-
necessary for managing the complexities and complications of gery connected to CT scan and magnetic resonance imag-
patients undergoing robotic surgery. These skills need to be ing. Technological design and preliminary results. Appl
included within the anaesthetic and surgical training curricula, Neurophysiol 1987; 50: 153–4
 i82 | Ashrafian et al.

14. Shenoy R, Nathwani D. Evidence for robots. SICOT J 2017; 3: 38 36. Ashrafian H. Intelligent robots must uphold human rights.
15. Benabid AL, Hoffmann D, Lavallee S, et al. Is there any Nature 2015; 519: 391
future for robots in neurosurgery? Adv Tech Stand Neurosurg 37. Bolenz C, Gupta A, Hotze T, et al. Cost comparison of robotic,
1991; 18: 3–45 laparoscopic, and open radical prostatectomy for prostate
16. Watanabe G. Are you ready to take off as a robo-surgeon? cancer. Eur Urol 2010; 57: 453–8
Surg Today 2010; 40: 491–3 38. Tandogdu Z, Vale L, Fraser C, Ramsay C. A systematic
17. Marescaux J. [Code name: “Lindbergh operation”]. Ann Chir review of economic evaluations of the use of robotic
2002; 127: 2–4 assisted laparoscopy in surgery compared with open or
18. Marescaux J, Leroy J, Rubino F, et al. Transcontinental robot- laparoscopic surgery. Appl Health Econ Health Policy 2015; 13:
assisted remote telesurgery: feasibility and potential appli- 457–67
cations. Ann Surg 2002; 235: 487–92 39. Hughes D, Camp C, O’Hara J, Adshead J. Health resource use
19. Madhani AJ, Niemeyer G, Salisbury JK. The Black Falcon: a after robot-assisted surgery vs open and conventional

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

teleoperated surgical instrument for minimally invasive laparoscopic techniques in oncology: analysis of English
surgery. Proc IEEE/RSJ Int Conf Intell Robot Syst 1998; 936–44 secondary care data for radical prostatectomy and partial
20. Intuitive_Surgical. Investor FAQ, 2017. http://phx.corpo nephrectomy. BJU Int 2016; 117: 940–7
rate-ir.net/phoenix.zhtml?c¼122359&p¼irol-faq (accessed 40. Niklas C, Saar M, Berg B, et al. da Vinci and Open Radical
10 August 2017) Prostatectomy: comparison of clinical outcomes and analy-
21. Desai PH, Tran R, Steinwagner T, Poston RS. Challenges of sis of insurance costs. Urol Int 2016; 96: 287–94
telerobotics in coronary bypass surgery. Expert Rev Med 41. Bijlani A, Hebert AE, Davitian M, et al. A multidimensional
Devices 2010; 7: 165–8 analysis of prostate surgery costs in the United States:
22. Chang SL, Kibel AS, Brooks JD, Chung BI. The impact of robotic-assisted versus retropubic radical prostatectomy.
robotic surgery on the surgical management of prostate Value Health 2016; 19: 391–403
cancer in the USA. BJU Int 2015; 115: 929–36 42. Ramsay C, Pickard R, Robertson C, et al. Systematic review
23. Hu D, Gong Y, Hannaford B, Seibel EJ. Semi-autonomous and economic modelling of the relative clinical benefit and
Simulated Brain Tumor Ablation with RavenII Surgical cost-effectiveness of laparoscopic surgery and robotic
Robot using Behavior Tree. IEEE Int Conf Robot Autom 2015; surgery for removal of the prostate in men with localised
2015: 3868–75 prostate cancer. Health Technol Assess 2012; 16: 1–313
24. Hagn U, Konietschke R, Tobergte A, et al. DLR MiroSurge: a 43. Huang KH, Lan YT, Fang WL, et al. Initial experience of
versatile system for research in endoscopic telesurgery. Int J robotic gastrectomy and comparison with open and laparo-
Comput Assist Radiol Surg 2010; 5: 183–93 scopic gastrectomy for gastric cancer. J Gastrointest Surg
25. Glassman D, White L, Lewis A, et al. Raven surgical robot 2012; 16: 1303–10
training in preparation for da vinci. Stud Health Technol 44. Barrie J, Jayne DG, Wright J, Murray CJ, Collinson FJ, Pavitt
Inform 2014; 196: 135–41 SH. Attaining surgical competency and its implications in
26. Ashrafian H, Darzi A, Athanasiou T. Autobionics: a new surgical clinical trial design: a systematic review of the
paradigm in regenerative medicine and surgery. Regen Med learning curve in laparoscopic and robot-assisted laparo-
2010; 5: 279–88 scopic colorectal cancer surgery. Ann Surg Oncol 2014; 21:
27. Ikuta K. Micro/miniature shape memory alloy actuator. IEEE Proc 829–40
Int Conf Robot Autom 1990, doi: 10.1109/ROBOT.1990.126323 45. Wolanski P, Chabert C, Jones L, Mullavey T, Walsh S,
28. Lalande V, Gosselin FP, Vonthron M, et al. In vivo Gianduzzo T. Preliminary results of robot-assisted laparo-
demonstration of magnetic guidewire steerability in a MRI sys- scopic radical prostatectomy (RALP) after fellowship train-
tem with additional gradient coils. Med Phys 2015; 42: 969–76 ing and experience in laparoscopic radical prostatectomy
29. Dwyer G, Chadebecq F, Tella Amo M, et al. A Continuum (LRP). BJU Int 2012; 110 (Suppl 4): 64–70
Robot and Control Interface for Surgical Assist in Fetoscopic 46. Vickers AJ, Savage CJ, Hruza M, et al. The surgical learning
Interventions. IEEE Robot Autom Lett 2017; 2: 1656–63 curve for laparoscopic radical prostatectomy: a retrospec-
30. Morimoto TK, Greer JD, Hsieh MH, Okamura AM. Surgeon tive cohort study. Lancet Oncol 2009; 10: 475–80
design interface for patient-specific concentric tube robots. 47. Souders CP, Catchpole KR, Wood LN, et al. Reducing operat-
Proc IEEE RAS EMBS Int Conf Biomed Robot Biomechatron 2016; ing room turnover time for robotic surgery using a motor
2016: 41–8 racing pit stop model. World J Surg 2017; 41: 1943–9
31. Morimoto TK, Okamura AM. Design of 3-D printed concen- 48. Lavoue V, Gotlieb W. Benefits of minimal access surgery
tric tube robots. IEEE Trans Robot 2016; 32: 1419–30 in elderly patients with pelvic cancer. Cancers (Basel) 2016;
32. Ciuti G, Valdastri P, Menciassi A, Dario P. Robotic magnetic 8: 12
steering and locomotion of capsule endoscope for diagnos- 49. Lee JR. Anesthetic considerations for robotic surgery. Korean
tic and surgical endoluminal procedures. Robotica 2010; 28: J Anesthesiol 2014; 66: 3–11
199–207 50. Herling SF, Moller AM, Palle C, Grynnerup A, Thomsen T.
33. Zhang L, Abbott JJ, Dong L, et al. Characterizing the swim- Robotic-assisted laparoscopic hysterectomy for women
ming properties of artificial bacterial flagella. Nano Lett 2009; with endometrial cancer. Dan Med J 2017; 64: pii: A5343
9: 3663–7 51. Parisi A, Reim D, Borghi F, et al. Minimally invasive surgery
34. Ullrich F, Bergeles C, Pokki J, et al. Mobility experiments for gastric cancer: A comparison between robotic, laparo-
with microrobots for minimally invasive intraocular sur- scopic and open surgery. World J Gastroenterol 2017; 23:
gery. Invest Ophthalmol Vis Sci 2013; 54: 2853–63 2376–84
35. Ashrafian H, Darzi A, Athanasiou T. A novel modification of 52. Knox ML, El-Galley R, Busby JE. Robotic versus open radical
the Turing test for artificial intelligence and robotics in cystectomy: identification of patients who benefit from the
healthcare. Int J Med Robot 2015; 11: 38–43 robotic approach. J Endourol 2013; 27: 40–4
 The evolution of robotic surgery | i83

53. Hu LD, Li XF, Wang XY, Guo TK. Robotic versus laparoscopic 72. Maerz DA, Beck LN, Sim AJ, Gainsburg DM. Complications of
gastrectomy for gastric carcinoma: a meta-analysis of effi- robotic-assisted laparoscopic surgery distant from the sur-
cacy and safety. Asian Pac J Cancer Prev 2016; 17: 4327–33 gical site. Br J Anaesth 2017; 118: 492–503
54. Cucinella G, Calagna G, Romano G, et al. Robotic versus lap- 73. Sukhu T, Krupski TL. Patient positioning and prevention of
aroscopic sacrocolpopexy for apical prolapse: a case- injuries in patients undergoing laparoscopic and robot-
control study. G Chir 2016; 37: 113–7 assisted urologic procedures. Curr Urol Rep 2014; 15: 398
55. Lairmore TC, Folek J, Govednik CM, Snyder SK. Improving 74. Wigmore TJ, Mohammed K, Jhanji S. Long-term survival for
minimally invasive adrenalectomy: selection of optimal patients undergoing volatile versus IV anesthesia for cancer
approach and comparison of outcomes. World J Surg 2016; surgery: a retrospective analysis. Anesthesiology 2016; 124: 69–79
40: 1625–31 75. Ellimoottil C, Roghmann F, Blackwell R, et al. Open versus
56. Health Quality Ontario. Robotic-assisted minimally inva- robotic radical prostatectomy in obese men. Curr Urol 2015;
sive surgery for gynecologic and urologic oncology: an 8: 156–61

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

evidence-based analysis. Ont Health Technol Assess Ser 2010; 76. Danic MJ, Chow M, Alexander G, Bhandari A, Menon M,
10: 1–118 Brown M. Anesthesia considerations for robotic-assisted
57. Harr JN, Luka S, Kankaria A, Juo YY, Agarwal S, Obias V. laparoscopic prostatectomy: a review of 1,500 cases. J Robot
Robotic-assisted colorectal surgery in obese patients: a Surg 2007; 1: 119–23
case-matched series. Surg Endosc 2017; 31: 2813–9 77. Kilic OF, Borgers A, Kohne W, Musch M, Kropfl D, Groeben
58. Kissane LM, Calixte R, Grigorescu B, Finamore P, Vintzileos H. Effects of steep Trendelenburg position for robotic-
A. Impact of obesity on robotic-assisted sacrocolpopexy. assisted prostatectomies on intra- and extrathoracic
J Minim Invasive Gynecol 2017; 24: 36–40 airways in patients with or without chronic obstructive
59. Baukloh JK, Reeh M, Spinoglio G, et al. Evaluation of the pulmonary disease. Br J Anaesth 2015; 114: 70–6
robotic approach concerning pitfalls in rectal surgery. Eur J 78. Ghomi A, Kramer C, Askari R, Chavan NR, Einarsson JI.
Surg Oncol 2017; 43: 1304–11 Trendelenburg position in gynecologic robotic-assisted sur-
60. Girgis MD, Zenati MS, Steve J, et al. Robotic approach miti- gery. J Minim Invasive Gynecol 2012; 19: 485–9
gates perioperative morbidity in obese patients following 79. Devarajan J, Byrd JB, Gong MC, et al. Upper and middle trunk
pancreaticoduodenectomy. HPB (Oxford) 2017; 19: 93–8 brachial plexopathy after robotic prostatectomy. Anesth
61. Gorgun E, Ozben V, Costedio M, Stocchi L, Kalady M, Remzi Analg 2012; 115: 867–70
F. Robotic versus conventional laparoscopic rectal cancer 80. Matta A, Kelley SR. Gluteal compartment syndrome after
surgery in obese patients. Colorectal Dis 2016; 18: 1063–71 robotic left adrenalectomy. Am Surg 2014; 80: E101–2
62. Backes FJ, ElNaggar AC, Farrell MR, et al. Perioperative out- 81. Keene R, Froelich JM, Milbrandt JC, Idusuyi OB. Bilateral glu-
comes for laparotomy compared to robotic surgical staging teal compartment syndrome following robotic-assisted
of endometrial cancer in the elderly: a retrospective cohort. prostatectomy. Orthopedics 2010; 33: 852
Int J Gynecol Cancer 2016; 26: 1717–21 82. Mattei A, Di Pierro GB, Rafeld V, Konrad C, Beutler J,
63. Coward RM, Smith A, Raynor M, Nielsen M, Wallen EM, Danuser H. Positioning injury, rhabdomyolysis, and serum
Pruthi RS. Feasibility and outcomes of robotic-assisted lap- creatine kinase-concentration course in patients under-
aroscopic radical cystectomy for bladder cancer in older going robot-assisted radical prostatectomy and extended
patients. Urology 2011; 77: 1111–4 pelvic lymph node dissection. J Endourol 2013; 27: 45–51
64. Nakamura LY, Nunez RN, Andrews PE, et al. Older age does 83. Herling SF, Dreijer B, Wrist Lam G, Thomsen T, Moller AM.
not impact perioperative complications after robot-assisted Total intravenous anaesthesia versus inhalational anaes-
radical prostatectomy. J Robot Surg 2011; 5: 201–8 thesia for adults undergoing transabdominal robotic
65. Buchs NC, Addeo P, Bianco FM, et al. Perioperative risk assisted laparoscopic surgery. Cochrane Database Syst Rev
assessment in robotic general surgery: lessons learned 2017; 4: CD011387
from 884 cases at a single institution. Arch Surg 2012; 147: 84. McLarney JT, Rose GL. Anesthetic implications of robotic
701–8 gynecologic surgery. J Gynecol Endosc Surg 2011; 2: 75–8
66. Pearce SM, Golan S, Gorin MA, et al. Safety and early onco- 85. Lee LA. Perioperative visual loss and anesthetic manage-
logic effectiveness of primary robotic retroperitoneal lymph ment. Curr Opin Anaesthesiol 2013; 26: 375–81
node dissection for nonseminomatous germ cell testicular 86. Weber ED, Colyer MH, Lesser RL, Subramanian PS. Posterior
cancer. Eur Urol 2017; 71: 476–82 ischemic optic neuropathy after minimally invasive prosta-
67. Tan A, Ashrafian H, Scott AJ, et al. Robotic surgery: disruptive tectomy. J Neuroophthalmol 2007; 27: 285–7
innovation or unfulfilled promise? A systematic review and 87. Gkegkes ID, Karydis A, Tyritzis SI, Iavazzo C. Ocular compli-
meta-analysis of the first 30 years. Surg Endosc 2016; 30: 4330–52 cations in robotic surgery. Int J Med Robot 2014
68. Yaxley JW, Coughlin GD, Chambers SK, et al. Robot-assisted 88. Barr C, Madhuri TK, Prabhu P, Butler-Manuel S, Tailor A.
laparoscopic prostatectomy versus open radical retropubic Cerebral oedema following robotic surgery: a rare complica-
prostatectomy: early outcomes from a randomised con- tion. Arch Gynecol Obstet 2014; 290: 1041–4
trolled phase 3 study. Lancet 2016; 388: 1057–66 89. Lerotic M, Chung AJ, Clark J, Valibeik S, Yang GZ. Dynamic
69. Hsu RL, Kaye AD, Urman RD. Anesthetic challenges in view expansion for enhanced navigation in Natural Orifice
robotic-assisted urologic surgery. Rev Urol 2013; 15: 178–84 Transluminal Endoscopic Surgery. Med Image Comput
70. Segal D, Awad N, Nasir H, Mustafa S, Lowenstein L. Comput Assist Interv 2008; 11: 467–75
Combined spinal and general anesthesia vs general anes- 90. Stoyanov D, Yang GZ. Stabilization of image motion for
thesia for robotic sacrocervicopexy: a randomized robotic assisted beating heart surgery. Med Image Comput
controlled trial. Int Urogynecol J 2014; 25: 369–74 Comput Assist Interv 2007; 10: 417–24
71. Bae J, Kim HC, Hong DM. Intrathecal morphine for postoper- 91. Lerotic M, Chung AJ, Mylonas G, Yang GZ. Pq-space based
ative pain control following robot-assisted prostatectomy: non-photorealistic rendering for augmented reality. Med
a prospective randomized trial. J Anesth 2017; 31: 565–71 Image Comput Comput Assist Interv 2007; 10: 102–9
 i84 | Ashrafian et al.

92. Mohareri O, Ischia J, Black PC, et al. Intraoperative registered study of 9/0 and 10/0 knot tying operated by 24 surgeons.
transrectal ultrasound guidance for robot-assisted laparo- Hand Surg 2011; 16: 1–3
scopic radical prostatectomy. J Urol 2015; 193: 302–12 101. Okamoto S, Wiertlewski M, Hayward V. Anticipatory vibro-
93. Hekman MCH, Rijpkema M, Langenhuijsen JF, Boerman OC, tactile cueing facilitates grip force adjustment during per-
Oosterwijk E, Mulders PFA. Intraoperative imaging techni- turbative loading. IEEE Trans Haptics 2016; 9: 233–42
ques to support complete tumor resection in partial neph- 102. Pacchierotti C, Sinclair S, Solazzi M, Frisoli A, Hayward V,
rectomy. Eur Urol Focus Advance Access published on May Prattichizzo D. Wearable Haptic Systems for the Fingertip
12, 2017, doi: 10.1016/j.euf.2017.04.008 and the Hand: Taxonomy, Review, and Perspectives. IEEE
94. Galletly NP, McGinty J, Dunsby C, et al. Fluorescence lifetime Trans Haptics. Advance Access published on May 9, 2017,
imaging distinguishes basal cell carcinoma from surround- doi: 10.1109/TOH.2017.2689006
ing uninvolved skin. Br J Dermatol 2008; 159: 152–61 103. Yao HY, Hayward V, Ellis RE. A tactile enhancement instru-
95. van den Broek FJ, Reitsma JB, Curvers WL, Fockens P, ment for minimally invasive surgery. Comput Aided Surg

Downloaded from https://academic.oup.com/bja/article/119/suppl_1/i72/4638479 by guest on 20 January 2025

Dekker E. Systematic review of narrow-band imaging for 2005; 10: 233–9
the detection and differentiation of neoplastic and nonneo- 104. Hammond FL, Howe RD, Wood RJ. Dexterous High-Precision
plastic lesions in the colon (with videos). Gastrointest Endosc Robotic Wrist for Micromanipulation. IEEE Int Conf Robot Autom
2009; 69: 124–35 2013, doi: 10.1109/ICAR.2013.6766578
96. von Delius S, Feussner H, Wilhelm D, et al. Transgastric 105. St John ER, Al-Khudairi R, Ashrafian H, et al. Diagnostic
in vivo histology in the peritoneal cavity using miniprobe- accuracy of intraoperative techniques for margin assess-
based confocal fluorescence microscopy in an acute porcine ment in breast cancer surgery: a meta-analysis. Ann Surg
model. Endoscopy 2007; 39: 407–11 2017; 265: 300–10
97. Esteva A, Kuprel B, Novoa RA, et al. Dermatologist-level 106. Leff DR, James DR, Orihuela-Espina F, et al. The impact of
classification of skin cancer with deep neural networks. expert visual guidance on trainee visual search strategy, vis-
Nature 2017; 542: 115–8 ual attention and motor skills. Front Hum Neurosci 2015; 9: 526
98. Mountney P, Giannarou S, Elson D, Yang GZ. Optical biopsy 107. Kogkas AA, Darzi A, Mylonas GP. Gaze-contingent percep-
mapping for minimally invasive cancer screening. Med tually enabled interactions in the operating theatre. Int J
Image Comput Comput Assist Interv 2009; 12: 483–90 Comput Assist Radiol Surg 2017; 12: 1131–40
99. Madabhushi A. Digital pathology image analysis: opportu- 108. Visentini-Scarzanella M, Mylonas GP, Stoyanov D, Yang GZ.
nities and challenges. Imaging Med 2009; 1: 7–10 i-BRUSH: a gaze-contingent virtual paintbrush for dense 3D
100. Panchulidze I, Berner S, Mantovani G, Liverneaux P. Is haptic reconstruction in robotic assisted surgery. Med Image
feedback necessary to microsurgical suturing? Comparative Comput Comput Assist Interv 2009; 12: 353–60