Robotic Surgery 2015
Robotic Surgery 2015
Robotic Surgery
Current Applications and New Trends
In collaboration with
Alessandra Marano
Giampaolo Formisano
Updates in Surgery
Giuseppe Spinoglio
Editor
Robotic Surgery
Current Applications and New Trends
In collaboration with
Alessandra Marano and Giampaolo Formisano
Forewords by
Giorgio De Toma
Francesco Corcione
~ Springer
Editor
Giuseppe Spinoglio
Department of General and Oncologie Surgery
··ss. Antonio e Biagio·· Hospital
Alessandria, ltaly
In co/laboration with
Alessandra Marano and Giampaolo Fonnisano
The publication and the distribution of this volume have been supported by the
Italian Society of Surgery
ISSN 2280-9848
ISBN 978-88-470-5713-5 ISBN 978-88-470-5714-2 (eBook)
DOI 10.1007/978-88-470-5714-2
1bis work îs subject to copyright. Ali rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recita-
tion, broadcasting, reproduction on microfllms or in any other physical way, and transmission or infor-
mation storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed. Exempted from this legal reservation are brief
excerpts in connection with reviews or scholarly analysis or material supplied specifically for the pur-
pose of beîng entered and executed on a computer system, for exclusive use by the purchaser of the
work. Duplication of this publication or parts thereof is permitted only under the provisions of the
Copyright Law of the Publisher's location, in its current version, and permission for use must always be
obtained from Springer. Permissions for use may be obtained through Rightslink at the Copyright
Clearance Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive narnes, registered narnes, trademarks, service marks, etc. in this publica-
tion does not imply, even in the absence of a specific statement, that such narnes are exempt from the
relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed tobe true and accurate at the date of publi-
cation, neither the authors nor the editors nor the publisher can accept any legal responsîbility for any
errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect
to the material contained herein.
The symbol of technical evolution, in the last few years , in general surgery is
undoubtedly robotic-assisted surgery. The introduction of this technique goes back
a decade and as usually happens with advanced technologies , enormous steps for-
ward have been made in the development of devices and equipment that is increas-
ingly sophisticated and of reduced dinlensions. A very significant impetus has been
given to this kind of surgery by Prof. Giuseppe Spmoglio, who recently had the
honor of being elected to the position of President to the Clinical Robotic Surgery
Associatiol/ in Chicago, a surgical society that is a key player of this specific
branch, and so is fOf\vard looking and of relevance.
The hospital in which Prof. Spinoglio is Head of Department has become an
intemationally renowned point of reference for robotics in surgery, where all kinds
of general surgery interventions are practiced, that can be treated through a mini-
mally invasive approach. To Prof. Spinoglio's further credit is the energy devoted
to create a European Training Center to infonu and train younger surgeons in tIus
particular type of surgery. This present text has been carefully edited by him and
will certainly have an undeluable worth for Italian surgeons and will be of great
didactic value for the youngest amongst them.
vii
Foreword
Throughout the ages , there have always been men gifted with visions that in
time proved to almost be foresight . By way of example , at the end of the 15th
century, Da Vinci had already imagined the helicopter and the parachute among
other inventions, and Jules Vente conceived the submarine decades before the
first model was built.
Keeping the above in mind, what happened much more swiftly in the world
of surgery under our own eyes, is without comparison. If someone had told me
25 years ago (not 250) that I would have been perfonlling surgical interventions,
even challenging ones , by introducing a thin tube into an abdominal cavity filled
with C02 while watching operations on a screen, I would have been flabbergast-
ed. And if someone had predicted I would even be able to one day operate on a
patient remotely, and follow each step of the intervention in full 3D HD on a flat
screen, while maneuvering sophisticated instmments that are as articulate as a
human hand, I 'would have taken the speaker as mad. Nevertheless all this
proved to be tme , and in a short time span of 25 years! It may thus easily be
understood that the surgeon relies increasingly on technology, to the point of
being dependent of it. Robotic-assisted surgery is the latest and most demanding
technological development in the operating room and , leaving aside all consid-
erations arising from correct surgical indications and the still critical cost-bene-
fit analysis, it surely represents a possibility of inlprovement that surgeons can-
not disregard.
Even robotics evolves at a stunning pace , almost on a daily basis: after the
older robots that ""ere made available 15 years ago , nowadays devices are
incredibly more practical and efficient. Further mention could be made of the
monotrocar, surgical sealants and dissection devices , robotic suturing, imaging
systems for lymphatic and vascular diagnosis and intervention, and many others.
All this should be a cause for reflection from the clinical and scientific point
of view, before Isaac Asimov 's dream can come tme in the surgical sphere as
well: "A robot may not injure a human being, or, through inaction, allow one to
come to harm .. ."
ix
x Foreword
I am thus very pleased to introduce this work that has very rightly been
assigned to a friend and colleague Giuseppe Spinoglio, one of the Italian sur-
geons most devoted and committed in this specific field of surgery in recent
years, acquiring an undisputed and intemationally renowned competence.
To be aware of the latest technological innovations in robotics, to know the
correct related clinical applications, how the new devices work and to be
informed of all associated issues is an important function of professional growth
that should not be missed, and reading this monograph will certainly enrich our
knowledge and supply a most welcome update on one of the most revolutionary
aspects of modem surgery.
Robotics entered the lives of humans more than fifty years ago , with the technol-
ogy being applied to engine construction .
In medicine, the use of robotics was established in the early 2000s " 'ith the
dissemination of the outcomes of the nerve-sparing prostatectomy for cancer. In
urology, and later in gynecology, robotics has achieved an immediate and
favorable response and subsequent widespread usage. On the other hand, in
general surgery its application was limited initially to procedures with a high
degree of difficulty that could especially benefit from the advantages of the
robotic suturing . Only at the end of the first decade of the 2000s , was robotics
continuously applied to colorectal and digestive surgery, especially by Korean
and Italian surgeons.
The da Vinci# System of Intuitive Surgical Inc . is the only robotic platfonn
existing today and has undergone several evolutions from 1998 to present: standard
three and four-ann , STM HD, Si™ HD (including Si™ -e) and the newest Xint .
The complex nattue of the settlp of the first models and the fearnre of work-
ing in fixed and narrow surgical fields has hanlpered its routine use in general
surgery. Indeed, while its users consider the robotic system as a revolutionary
innovation, the opponents emphasize costs, tinle-consuming procedures and the
lack of clinical evidence when compared to laparoscopy. Many of the issues relat-
ed to the robotic settlp and its time-spending applications have been overcome by
the technical characteristics of the new models and by the standardization of sur-
gical procedures, sintilar to what happened to surgery after the pioneering phase.
Regarding robotic benefits , three different types of problems can be identified:
1. Robotics has been applied to general surgery for little more than five years
and for less time with a standardized technique: randontized clinical trials
with adequate follow-up providing sound data are expected to be available in
approximately ten years , as occurred with the CLASSIC and COLOR studies
for oncological outcomes.
2. About 3,000 da Vinci® systems of different generations are currently installed
worldwide and less than 2 ,000 are applied in general surgery for different pro-
cedures: in such conditions, it is difficult to collect a large enough sample to
be compared with single laparoscopic procedures . Comparative studies on
xi
xii Preface
short-term outcomes have been published to date but they have been per-
fOfllled on small series.
3. In robotic surgery, as occurred with other approaches, there is a learning
curve . The number of robotic-experienced surgeons is a small percentage of
those who have decided to implement it: this specific issue makes it hard to
recmit homogeneous centers involved in studies (the same surgeon may be
very experienced in laparoscopic teclmiques but not in robotic ones).
Nevertheless, the following main issues need to be underlined:
• To date , robotics technology has advantages over other present techniques
that are absolutely clear. Its superiority over an action perfonned manual-
ly results mainly in the 3DHD vision , up to lOx magnification, the
EndoWrist® instruments with seven degrees of freedom and the intuitive
motion.
• In every part of the world, surgeons vvho had the patience and persever-
ance to overcome the learning curve have not abandoned robotic surgery.
• Surgeons who are not convinced of the advantages of robotic surgery are
sticking to their opinion , as for the reasons mentioned above, these bene-
fits cannot be demonstrated as yet.
The purpose of this book is to spread the robotic surgical technique for stan-
dardized procedures with the most recent updates. Since Italian surgeons were
among the pioneers of this surgery, I had the pleasure of hosting in this volume
both experiences of the finest ones and those of surgeons who have spread robot-
ics abroad (i.e. Profs. Enrico Benedetti, Pier Cristoforo Giulianotti and Alessio
Pigazzi). I sincerely thank President Giorgio De Toma and all SIC's Board for
entmsting me with the lecture at the 116th SIC Congress and the subsequent writ-
ing of this book, with a focus on innovation, even \\'hen it is still questioned.
Finally, I would like to thank my assistants for their sound collaboration, mainly
Alessandra Marano and Gianlpaolo FOfllusano: without them, this book would
never have been completed.
4 Thoracic Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 29
Giulia Veronesi
xiii
xiv Contents
9 Hepatic Resections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 83
Alberto Patriti, Graziano Ceccarelli, and Luciano Casciola
14 Left Colectomy and Segmental Resections for Cancer .... . .... . .. 133
Paolo Pietro Bianchi. Igor MonseUato, and Wanda Petz
16 Full robotic Technique for Rectal Cancer . . ... ..... .. ........... 159
Giuseppe Spinoglio, Gianlpaolo Fonllisano, Luca Matteo Lenti.
Fabio Melandro, and Alessandra Marano
Part VI Miscellany
The volume editor and the publisher gratefully acknowledge the educational
contribution offered by ab medica S .p.A. , and wish to thank Intuitive Surgical@ ,
Inc., for granting permission for the use of operating room setup and trocar lay-
out images .
xvi
Contributors
xvii
xviii Contributors
1.1 Introduction
G. Spinoglio ( ~ )
Department of General and Oncologic Surgery, "Ss . Antonio e Biagio" Hospital,
Alessandria, Italy
e-mail: [email protected]
gery and led to laparoscopic techniques for minimally invasive surgery. This
culmination of technological advancements led to the first laparoscopic appen-
dectomy by Gemlan gynaecologist Dr. Kurt Semm [1] and to the first report-
ed laparoscopic cholecystectomy in 1987 by French surgeon Dr. Philippe
Mouret [2] ; actually, the first laparoscopic cholecystectomy ,vas performed in
1985 by German surgeon Erich Mohe [3].
Soon after these landmark operations, laparoscopic technology and tech-
niques continued to gain popularity into the 1990s. However, the tools being
used for manual laparoscopy only worked well for relatively simple surgical
procedures that involved the removal of tissue and basic tissue closure .
Sophisticated mechanisms , such as staplers and other tissue closure devices
were developed but still did not allow laparoscopic techniques to gain traction
in more complex surgical procedures.
Though, in the strict sense of the word, the robotic systems developed for sur-
gical applications are not actually "robots" but remote "perfomlers" that work
using the master-slave style, robotic telepresence was beguUling to flourish at the
same time laparoscopy was experiencing lunitations. Many ulstitutions recognized
a potential opportunity to blend minimally invasive surgery (MIS) with robotics to
overcome the limitations of laparoscopic surgery. The goal was to make MIS an
option for even complex procedures with help from computers and mechanics .
The first robot-assisted surgical procedure was performed in 1983 with the
use of "Arthrobot" , which was designed to assist in orthopedic procedures. Later
on, a transurethral resection of the prostate and standard prostatectomy were car-
ried out in London vvith the PUMA (Programmable Universal Machine for
Assembly) and the SARP (Surgeon-Assistant Robot for Prostatectomy) systems ,
respectively. Further development on SARP led to the creation of PROBOT,
VRobot and SPUD, \vhich are abbreviation for "Prostate Robot", "Urology
Robot" and "Surgeon Programmable Urological Device" , respectively [4-6].
The first application of robotics in abdomulal surgery dates back to 1993 ,
when Yulin Wang succeeded in developing the first FDA-approved robotic
device for use in general surgery. The system, Automated Endoscopic System
for Optimal Positioning (AESOP) [7], consisted of a table-mounted articulatulg
arm that was used to control the movements of the camera during laparoscopic
surgery. Originally the AESOP was manipulated by hand or foot controls, but the
later version was capable of utilizing voice commands and ulcorporated voice
control of the endoscope and OR room lights .
The introduction in 2001 of the ZEUS system (Computer Motion Inc.) rep-
resented the real step towards the modem concept of robotically-assisted laparo-
scopic surgery. This platfonn allowed the surgeon to control a robotic slave
device that was docked to the patient remotely from a console.
The ZEUS robotic system had a camera arm that was voice controlled
(AESOP System), along with two other operating anus that provided four
degrees of freedom and were able to hold a variety of instruments that were tele-
manipulated with joysticks from the surgical console . The software that inter-
1 History of Robotic Surgery 3
faced the surgeon console with the robotic arms allovved tremor filtration and
motion scaling by a factor of 2-10 . The surgical field was visualized through a
regular two-dimensional screen or through polarized glasses that allowed for
three-dimensional images. This system was used for the first time in a full
laparoscopic procedure for fallopian tube anastomosis at the Cleveland Clinic in
1998 [8] and for coronary bypass by Reichenspumer in 1999 [9].
In 2001, Jacques Marescaux utilized the ZEUS system to perfoml a robot-
assisted cholecystectomy on a patient in Strasbourg , France , who was 4000 km
away from the surgeon in New York [10] . This procedure , nicknamed "Operation
Lindbergh" , gave an impressive demonstration of telepresence in surgery.
hI almost a parallel path, another group of researchers in Califomia set about
to develop a surgical robotic system for civilian use and founded Intuitive
Surgical Intemational in 1995: this group was eventually able to develop the
first FDA-approved fully robotic system for application in laparoscopic surgery
with licensed technologies from MIT, IBM and SRI Intemational.
1.2.1 Introduction
The design of the da Vinci® is the result of a long developmental process which
integrated many ideas and technologies to produce an intuitive and functional
surgical platfonn. The original project of the US Department of Defense was
started ,,,ith the aim of providing injured soldiers with a frontline surgical suite
controlled by surgeons operating from a safe remote location. Although imprac-
tical at the time, the ultimate goal of the surgical platfoml was to produce a reli-
able system which would deliver the benefits of minimally invasive surgery to
patients while preserving the benefits of the open approach to surgeons . The
technology specifically aimed to address port-access limitations in dexterity,
intuitiveness , visualization and ergonomics through advances in telepresence,
telemanipulation and stereoscopic capnlre.
Animal trials started in 1996 and clearly demonstrated the promise of seven-
degrees-of-freedom manipulators as well as the need for a mobile patient-side
cart . The next major i1l1lovation was called "Mona" and featured exchangeable
sterile components, which allowed human trial to proceed in 1997. In December
1998, the first commercial version of the da Vinci® system was delivered to the
Leipzig University Heart Center in Germany.
Refinement of the original da Vinci® design continued with the addition of a
fourth arm and with the expansion of instmment families.
The da Vinci® STM and da Vinci® Si™ (released in 2009 , Fig. 1.l) represent
an evolution of the fIrst generation system and are currently available on the mar-
ket . Novvadays , moreover, access to the latest da Vinci@ technology and future
innovation is available in an upgradable three-ann configuration: the da Vinci@
4 G. Spinoglio et al.
Si-e™ System has been designed to deliver da Vinci® core functionality, provid-
ing a flexible and economical solution for many robotic-assisted procedures .
The latest product interaction is the fourth generation of the da Vinci® sys-
tem (XiTM , Fig. 1.2 , 1.3), which features improvements to the vision , control
system and functionality.
1 History of Robotic Surgery 5
The da Vinci® system is designed to represent the natural extension of the sur-
geon 's eyes and hands and its motion capabilities are intended to mimic those of
its human operator, as in open surgery.
The shared core technology of all systems offers the following distinguished
features:
• physical separation of the surgeon from the patient by operating at an
ergonomic console rather than at the patient's side, with less fatigue ;
• EndoWrist® technology: robotic instruments provide seven degrees of free-
dom with intuitive control (compared with five degrees of freedom for stan-
dard laparoscopic instruments), which replicates the experience of open sur-
gery by preserving natural eye-hand-instrument alignment (in contrast with
standard laparoscopy);
• three-dimensional stereoscopic HD images with magnification (up to lOx);
• complete tremor filtering and motion scaling.
All systems consist of three main components: the surgical console , the
patient cart and the vision cart .
The surgical console is the workplace of the robotic surgeon and contains
the following core elements: master controllers , stereo viewer, touchpad for sys-
tem configuration, left-side pod for ergonomic controls, right side pod for power
and emergency stop, and a foot-switch for operative mode selection and energy
devices activation.
The masters are built essentially like a human arm, with a wrist portion and
the elbow/shoulder joints for positioning: the wrist portion moves the instrument
tip in the surgical field , while the more proximal joints orient the instrument to
the desired location. The master controllers also possess finger clutches, which
decouple the joystick from the control of the corresponding instnunent to allow
for ergonomic repositioning during surgery.
The stereoviewer provides the 3D image of the operative field to the surgeon,
including extended instruments and system infomlation. The touchpad is the
main control interface for system functions: it provides system status, instru-
ment arm selection and control selections.
The patient cart has the prinlary function to support the operative amlS and
the camera aml. It contains five main components: the setup joints, instrument
amlS, camera arm, EndoWrist® instruments and an endoscope.
The setup joints enable movements of the instrument and camera ann to
position them for sterile draping by the scrub nurse and docking of the system
to the patient by the bedside assistant after trocar placement and correct table
positioning according to the procedure. Clutch buttons are used by the assistant
to free the setup joints, both for cart docking and ann repositioning, if needed,
during the procedure. To ensure patient safety, any actions of the patient-side
assistant should always preclude the telepresence of the robotic surgeon.
EndoWrist® instruments are installed onto the instrument amlS after the
6 G. Spinoglio et al.
patient cart is docked. Most instnuuents are capable of seven degrees of freedom
and about 90° of articulation in the wrist (distal) joint. Many different instm-
ments and energy devices (monopolar, bipolar, ultrasonic , radiofrequency) are
available with a diameter of 8 mm. A selection of 5 l111U instmments is also avail-
able for use with smaller access ports.
The instmments are made by the following elements:
• a tip that represents the effector of a specific surgical task;
• an EndoWrist® articulating joint (with the exception of some instmments , as
required by the underlying technology, i .e ., Harmonic ACE®);
• a shaft that represents the rotating amI of the instmment and through move-
ment are transferred from the robotic amI to the wristed tips;
• release levers for the instmment's removal;
• instmment housing that engages with the sterile adapter of the robotic arm.
The da Vinci@ endoscope provides a three-dimensional high-definition imag-
ing system. It is available with either a straight (0°) or angled (30°) tip and with
a shaft diameter of 8.5 mm or 12ll1111. In keeping with the anthropomorphic con-
cept , the endoscope contains two separate optical channels and focusing ele-
ments , and the camera head contains two separate cameras . When displayed on
two monitors to the left and right eye of the surgeon, a tme and natural three-
dimensional image is achieved.
The vision cart includes the image processing and vision equipment, a 24
inches touch screen monitor (touch screen not available on the Si-e™ system)
for the bedside assistant and system settings , and adjustable shelves for other
surgical devices such as insuffiators and electro surgical generators. The vision
cart is the system's central connection point ,,·here all auxiliary equipments and
audiovisual connections are routed.
A comparison between the different da Vinci~' Surgical Systems (STM /Si-
e™ /SiTM) currently available on the market is shown in Table 1.1.
have been also FDA cleared for the da Vinci® Xi nt system, ,,'hile CE mark is
expected to be obtained in the near future. The Single-Site nt technology will be
designed and integrated into a completely innovative platfoml.
Three out of these five lumens are straight and are designed for the 8.5 nm}
3D HD endoscope, the insufflation adaptor and the 5 nml assistant port, respec-
tively. The two more lateral lumens are curved and cross in the midline of the
Single-Sitent port (remote center) with the outlet holes on the opposite side of
entry: the semi-rigid instrument that enters the abdomen from the left reaches
the operative field on the right and vice versa . The curved 5 mm robotic camlll-
lae are inserted into these channels so that the instrument arms outside the body
wall are separated, maximizing the range of motion and minimizing potential
intemal and extemal crowding (Fig. 1.4).
Indeed, thanks to the abovementioned Single-Site™ configuration, the
instruments and the camera, crossed within the port, use remote center technol-
ogy to avoid cannula collisions , am} interferences and port-site movement.
Subsequently the da Vinci® system software automatically detects and re-associ-
ates the user' s hands with the instrument tips to create intuitive movement
through crossed cannulae. At the end , the triangulation is restored and the cor-
rect hand/instrument correlation is obtained with the da Vinci# software .
The placement of the remote center at the level of the abdominal wall and the
curvature of the cannulae guarantees an optimal focal distance of work allowing
the instruments to converge correctly on the anatomical target . If the target is clos-
er or further away to the optinlllm focus it will be necessary to advance or retract
the cannulae. These modifications can flrstly cause an incorrect positioning of the
remote center and secondly produce excessive stress on the port and on the
abdominal wall resulting in improper working of the instruments and loss of C02 .
History of Robotic Surgery 9
Additionally, if the instruments come out too far from the call1lulae to reach
a distant target , during the lateral traction they could overly flex with a poten-
tially bullwhip effect. These issues can be mitigated by using one of the two sets
of robotic curved cannulae of different lengths (250 mm or 300 mm). However,
nowadays , the main drawback of this innovative platfonu is represented by the
absence of the EndoWrist® technology, as in traditional robotic instruments.
To date , the da Vinci® Single-Site™ piatfoffil is made up of the follovving
components:
• Single-Site™ accessories: Single-Site™ Port (with insufflation tubing and
stopcock) , a 8.5 mm endoscope cannula for the introduction of Intuitive
Surgical fluorescence or Basic three-dimensional high-definition endoscope
(30° or 0°), two fixed-shape 5 mm curved cannulae (250 nml or 300 mm
length) with flexible blunt obturator, a 5 nl1l1 and a 10 nml straight accesso-
ry call1lula for illsertion of mauuallaparoscopic illstruments with correspon-
ding flexible obturator, an 8 mm semi-rigid blunt obturator (250 mm and
300 nun length), a dock assist tool;
• Single-Site™ semi-rigid instruments: Maryland dissector, crocodile grasper,
fundus grasper, cadiere forceps , curved scissors , monopolar cautery hook,
Hem-o-Lok~ clip applier, Hem-o-lok® ML clips (Weck®), suctiOll irrigator,
curved needle driver and bipolar Marylaud;
• EndoWristed Single-Site™ lleedle driver is expected to be put on the market
in the near future.
In 2011, a new optical system has been developed and integrated into the da
Vinciil Surgical System . It is capable of emitting laser light that is close to
infrared light with the ability to switch with dedicated conunands at the console
between white light and near-infrared (NIR) light view in real tinIe , thus offer-
ing the opportunities to perfonn fluorescence-guided surgery thanks to the prop-
erties of the Indocyanine Green (leG) vital dye [13] (Fig . 1.5).
leG has been widely used for the study of blood flow and microcirculation
for more thau 40 years . It binds to plasma proteins when injected in the blood
stream and reaches all the organs and body regions. Its routine use has widely
spread through different specialties (cardiac surgery, neurosurgery, ophthalmol-
ogy, hepatology, etc.) thanks to its excellent tolerability, few side effects,
extremely low toxicity and few allergic reactions; the optimal dose rauge is
between 0.1 aud 0 .5 mglkg and should not exceed 2 mglkg.
The leG has the ability to absorb light in the NIR wavelengths between 600
and 900 1l1l1. If its molecules are excited with infrared laser light, they emit a
very intense fluorescent signal. At the frequency of 780 nm, it is possible to sup-
press the exciting laser light through dedicated filters in order to detect only the
fluorescent signal.
10 G. Spinoglio et al.
Fig. 1.5 Indocyanine Green Near-Infrared fluorescence for the da Vinci@ Sint System. (© 2014
Intuitive Surgical, Inc.)
• The EndoWrist# One T )'! vessel sealer is a single-use, sterile instrument with
independent sealing and cutting functions and a dual-hinged jaw. It bears the
CE mark and is cleared in the US for bipolar coagulation and mechanical
transection of vessels up to 7 mm in diameter and tissue bundles that fit in
the jaws of the instrument. It is cleared for commercial distribution in the US
for use with the da Vinci® Si™ and Xi T ).! Surgical System and the ERBE VIO
300 D electrosurgical generator (identified by a label on the front bezel).
Like other da Vinci® instruments , it features fully-wristed architecture and
can he activated from the surgeon console. The single-use EndoWrist®
One™ vessel sealer comes with an integrated cable and requires a propri-
etary and dedicated upgrade mounted to the vision cart of the da Vinci~' Si™
System (ERBE VIO 300 D) that controls and activates the instrument and the
1 History of Robotic Surgery 11
/C5OIJ
Motor pack
1m9sJ
Instrument
C~~)
Reloads
/
r./lMp· r2".,tIOl/
1.4 Conclusions
Robotic surgery was born to extend the frontiers of minimally invasive surgery.
New robotic platforms are emerging for use in different surgical specialties and
distinct ne,,' features will enable more procedures to be perfornled with the help
of a computer-enhanced system. Additional technologies will be developed into
currently existing or new platforms to hold the promise of becoming the central
workstation of surgical care. However, although robotic surgery is growing , con-
cerns regarding the high costs involved still exist and the market is yet to be
fully matured.
References
In the last ten years, minimally invasive approaches have increased their appli-
cations in neck surgery.
The minimally invasive surgery can be distinguished in two groups:
• direct cervical approach: mininlally invasive video-assisted thyroidectomy
(MIVAT) [I] or endoscopic technique with an anterior approach carried out
by Gagner and Cougard or with a lateral approach described by Henry and
Inabnet [2-5];
• indirect or extracervical approach [6-8]: endoscopic techniques outside
the neck region through a chest , axillary, or combined axillary-breast access,
described for the first time by two Japanese surgeons, Ikeda and Takami in
1999 [2,9].
Some of these techniques are totally gasless, such as the robotic transaxillary
access; others are performed with gas insufflation. However, endoscopic proce-
dures are limited by video camera platfonll instability, straight endoscopic instru-
ments, two-dinlensional inlaging and a difficult manipulation of the anatomical
structures. In the last years , some Asiatic surgeons have attempted to incorporate
surgical robots in thyroid surgery, reducing the limitations of conventional
endoscopy: improving freedom of motion through the use of multi-articulated
instruments; providing the surgeon with an ergonomically perfect position at the
robotic console; providing a three-dinlensional, stable, magnified imaging:
enabling the surgeon to perfonll minute, precise movements and allowing the
dampening of physiologic tremors [10). Prof. W.Y. Chung, from South Korea , in
M. Piccoli ( ~ )
General Surgery Unit, "Sant' Agostino-Estense" New Hospital,
Baggiovara (MO), Italy
e-mail: m [email protected]
Fig. 2. 1 Patient
positioning
2 Transaxillary Thyroidectomy and Parathyroidectomy 17
a
Fig. 2.2 a Left axillary access _b Right axillary access _(:© 2014 Intuitive Surgical. Inc _)
The da Vinci® Sin! robot is docked contralateraUy to the axillary access (Fig.
2.2a , b). Three arnlS are generally sufficient for lobectomy and parathyroidecto-
my; four amlS are needed, however, for total thyroidectomy.
2.4 Procedure
A 4-5 cm skin incision is made in the axilla , following the lateral edge of the
major pectoralis muscle; the subplatysmal skin flap from the axilla to the ante-
rior neck area is dissected over the anterior surface of the pectoralis major mus-
cle using laparoscopic instn1ments (such as the Johann grasper and monopolar
hook), under endoscopic vision with a 30° camera. The endoscopic vision
instead of direct vision (used by Korean surgeons) , allows all the surgical team
to follow the creation of the flap , reducing the learning curve of this step. An
external retractor - the so called "Modena Retractor" (CEATEC@
Medizintechnik) - is used from the beginning to create the flap . A suction tube
is directly connected to the retractor 's blade in order to avoid field fogging. The
myocutaneous flap is raised until the sternal and clavicular heads of the stern-
ocleidomastoideum muscle (SCM) are visualized; then the dissection continues
through the two SCM branches. During this time, care must be taken not to
cause thermal bums on the overlying skin . The "Modena retractor" is reposi-
18 M. Piccoli et al.
tioned beneath both the sternal branch of the SCM and the strap muscles and the
thyroid is discovered.
The omohyoid muscle is the superior landmark; behind it , there is the upper
thyroid pole . Care must be taken not to damage the internal jugular vein that is
the first main venous vessel that appears in the operative field. The contralater-
al strap muscles are identified and raised if a total thyroidectomy must be per-
fornled.
The robot is positioned in the operative field and three robotic anns are intro-
duced all through the axillary incision: 30° optic , Hannonic curved shears
(5 mm) and ProGrasp forceps (8 nml). In the case of the three robotic arms tech-
nique , only t\\'0 anns are introduced in the incision: endoscope and Harmonic
shears (the operation is conducted without the help of the ProGrasp). The fourth
(or the third in case of three robotic arms technique) robotic ann is inserted
through an independent incision at the inferior part of the axillary incision for
the Maryland forceps . (Fig . 2.3a , b; Fig. 2.4a , b) . During the operation, the
Maryland forceps and Hannonic shears are interchangeable .
Fig. 2.3 a Right axillary access with four arms. b Left axillary access with four anns
Fig. 2.4 a Right axillary access with three anns . b Left axillary access with three anns
of the RLN. At this time also the contralateral thyroid lobe is resected and extract-
ed through the axillary skin incision. In the case of parathyroidectomy, the thy-
roid gland is turned medially and , with cautious dissection, the parathyroid ade-
noma is identified, than circumferentially dissected and excised. The venous
bleeding is checked ,""ith the Valsalva maneuver. A closed suction drain is insert-
ed through the separate incision under the axillary skin incision. The robotic anns
are removed and the wound is cosmetically closed. The small incision scar in the
axilla is completely covered when the arm is in its natural position .
20 M. Piccoli et al.
From September 2010 to May 2014, 191 patients underwent robot-assisted thy-
roidectomies (only four endoscopic, due to robotic technical problems) and five
target parathyroidectomies using a unilateral trans axillary approach. The indica-
tions were: benign tumor, not larger than 5 cm; follicular neoplasm from FNAB;
Graves' disease , well-differentiated thyroid cancer and parathyroid adenoma.
The final pathologies were: 107 nodular hyperplasia , 25 follicular adenoma, II
Hurtle adenoma, 2 diffuse hyperplasia , 38 differentiated thyroid carcinoma , 13
thyroid totalization (12 free thyroid tumor tissue and I contralateralmicrocarci-
noma), 5 parathyroid adenoma. The median size of the dominant thyroid nodule
was 28 .6 mm (range: 5-60) . Mean total operative time (from the end of induc-
tion of anesthesia to the completion of skin closure) was 160.2 min for total thy-
roidectomy (IT) and 115.1 min for lobectomy (LT) . The mean time required to
create the working space was 63 .8 min (30-180 min) , to dock the robot was
14.9 min (4-60 min). The mean console time was 53.7 min (10-135 min): mean
LT, 39.0 min; mean TT, 77 .1 min.
These intraoperative complications have been observed: two bums skin
lesions , one external jugular vein lesion, and one internal jugular vein lesion, all
resolved without conversion. As regards postoperative complications: 7 tempo-
rary RLN injury (3.5%), 27 transient hypocalcemia (13.7%), 6 transient ipsilat-
eral arm paralysis (I brachial plexus injury), 4 seromas (only one treated by per-
cutaneuos aspiration), I subcutaneous tunnel infection, 4 postoperative
hematoma: 2 conservatively treated and 2 reoperated. Median hospital stay is
2.0 days (1-15).
2.8 Conclusions
The robotic gasless trans axillary thyroidectomy is feasible and safe if perfornled
by a surgeon skilled in endocrine and robotic surgery. Robotic thyroid and
parathyroid surgery is often associated with longer operation time than conven-
tional open/endoscopic surgery but with same outcomes in surgical complete-
ness and safety. Compared to the conventional cervical approach , robotic neck
surgery is associated not only with excellent cosmetic results, but also with
reduced postoperative neck discomfort (pain, decreased voice and swallowing).
Furtheonore , another advantage is that the robotic surgery improves ergonomics
and has a shorter learning curve than open or endoscopic surgery [13-15] .
Prospective, controlled randomized studies with long-teon follow-up are
needed.
22 M. Piccoli et al.
References
3.1 Introduction
M. Piccoli ( )
General Surgery Unit, "Sant' Agostino-Estense" New Hospital,
Baggiovara (MO), Italy
e-mail: m [email protected]
Fig. 3.1 a Patient positioning and trocar layout . I, Hannonic curved shears; 2, ProGraspnl for-
ceps; 3, Maryland dissector. b OR setup (© 2014 Intuitive Surgical, Inc)
3 Zenker Diverticulum Treatment 2S
duced all through the axillary incision: dual channel endoscope (central, down) ,
Harmonic curved shears (right) and ProGraspT~ forceps (up) . The fourth robot-
ic amI is inserted through an independent incision in the inferior part of the axil-
lary one , for Maryland dissector (left) (Fig. 3.la). All vessel dissections are per-
fomled using the HamlOnic curved shears . Under robotic guidance, the thyroid
is drawn medially by the ProGrasp forceps in order to identify and spare the
inferior thyroid artery and the inferior laryngeal nerve . It is necessary to cut the
middle thyroid vein , and sometimes also the omohyoid muscle. The prevertebral
fascia is identified , and the diverticulum can be isolated from its adhesion with
the hypopharynx. The loose connective tissue surrounding the pouch is dissect-
ed to identify its neck on the posterior pharyngeal wall. This procedure must be
carried out under esophagoscopic control. The neck of the diverticulum is fully
exposed by tractiouing the diverticulum to the left with the Maryland dissector.
A complete myotomy must be performed, including the cricopharyngealmuscle
and the first 5 cm of the circular layer of the cervical esophagus . The myotomy
could be perfomled with a robotic monopolar hook allowing dissection and
resection of the fibers, proceeding downward, starting from the upper
esophageal sphincter. Then a surgical linear stapler (Endopath RTS-FLEX
Endoscopic Articulating Linear Cutter 35 mm; Ethicon Endo-surgery, LLC) with
a blue cartridge is inserted through the axilla and it is applied to the neck of the
diverticulum, which is cut and sutured and then removed (Fig . 3.3).
The complete diverticultml removal and its staple line could be checked endoscop-
ically. A suction-type drain is left in place lUltil the X-ray examination is perfonlled.
Intravenous broad-spectrum antibiotics are administered only perioperatively.
Treatment options for a Zenker diverticulum include open surgery, flexible endo-
scopic and rigid endoscopic therapy. Based on current evidence, traditional open
surgery is suitable for all kinds of diverticula, providing satisfactory long-tenll out-
comes, although with a significant incidence of complications including medias-
tinitis , recurrent laryngeal nerve injury, esophageal stricture , fistula, esophageal
perforation, hematoma, wound infection, pneumonia and even death, with a 11 %
median incidence of major morbidity [I] . It also needs general anesthesia.
Rigid endoscopic treatment can be done under general anesthesia and hyper-
extension of the neck. It might be technically difficult when the diverticular sep-
tum cannot be well exposed. Flexible endoscopic therapy can be conducted
without general anesthesia or neck hyperextension; however it is only suitable
for selected patients.
The poor quality of current evidence renders it difficult to establish a sound
conclusion for the optimal treatment of ZD [3, 7, 8]. Actually there is not strong
evidence for a gold standard technique .
transaxillary robotic approach. The median age was 70.3 years , seven patients
were male and one female . The preoperative work-up included a barium swal-
lovv, which allowed detemunation of the diverticulum size and location and
upper gastrointestinal endoscopy, and was mandatory to rule out malignancy.
An X-ray check with an oral soluble contrast swallow study was carried out
4 days after the operation . In the absence of leakage, oral feeding was adnunis-
tered to the patient and the patient could be discharged. Patients were scheduled
for a clinic visit 5 days after discharge , and after 1 month a barium swallow
study was perfomled .
The mean diameter of the ZD, on preoperative barium swallow, was 4.25 cm
(range: 3-9 cm). The mean time for setting up the working space was 66 nlln-
utes (range 40-90 nlln .), the mean docking time was 12.8 nun . (range 6-23 nun.)
and the mean robotic time 85.1 nunutes (range 60-115 min) . The mean hospital
stay was 7 days. There vvere no conversions to open cervicotomy, no transient or
definitive left recurrent nerve palsies . There vvas only one postoperative leak,
which was conservatively treated and fully recovered. There was no evidence of
persistence or recurrence at the postoperative , although short , follow-up .
3.6 Conclusions
The results of this technique show that the left trans axillary robotic approach is
a safe and effective surgical procedure in the treatment of any kind of ZD . This
procedure camlOt be considered a cosmetic treatment, as ZD mostly occurs in
men between the seventh and eighth decade of life , but the advantage offered by
the da Vinci® Si™ surgical robot system is a three-dimensional field of view and
a more accurate sense of perspective. Moreover, the magnification of target
structures, made possible by the system, facilitates the preservation of the recur-
rent laryngeal nerve , thus preventing both transient and definitive left recurrent
nerve palsy, and spares the esophageal mucosa during the myotomy, reducing
the incidence of postoperative leakage.
It is also worth considering that the robot system incorporates features for
hand-tremor filtration , fine motion scaling, negative motion reversal (allovving
minute and precise tissue manipulation); in conjunction with the ergononucally
designed console, they help decrease the surgeon's fatigue .
The robotic assisted trans axillary Zenker's diverticulectomy is a procedure
that requires experience in perfomung the flap and, due to the rarity of this
pathology, the only way for the surgeon to complete the learning curve is to per-
foml robotic transaxillary thyroid surgery; so tIus procedure can be attempted
only by surgeons well versed in neck robotic surgery, and skilled in esophageal
diseases surgery.
Open diverticulectomy has low but significant adverse events, including
mediastinitis , recurrent laryngeal nerve injury, esophageal stricture, fistula ,
esophageal perforation, hematoma , wound infection, pneumonia and even death.
28 G. Melotti et al.
Various techniques have been described for rigid and flexible endoscopic treat-
ment of ZD . Despite overall good results, follow-up has been often inadequate [2 ,
3] and technical refinements are still in progress to avoid adverse events [7, 8].
The robot-assisted trans axillary Zenker's diverticulectomy is a technically
demanding procedure. According to our preliminary data , this procedure appears
safe and effective.
However, our results need to be confirmed in larger cohorts of patients and
further randomized controlled studies comparing this technique with rigid and
flexible endoscopic diverticulotomy.
References
1. Dzeletovic I, Ekbom DC, Baron TH (2012) Flexible endoscopic and surgical management of
Zenker's diverticulum. Expert Rev Gastroenterol Hepato16:449-466
2. Law R, Katzka DA, Baron TH (2014) Zenker's diverticulum. Clin Gastroenterol Hepatol.
doi:lO.1016/j .cgh.2013 .09 .016
3. Yuan Y, Zhao Y-F, Hu Y, Chen L-Q (2013) Surgical treatment of Zenker's diverticulum . Dig
Surg 30:214-225
4. Kang SW, Jeong JJ, Yun JS et al (2009) Robot-assisted endoscopic surgery for thyroid can-
cer: experience with the first 100 patients . Surg Endosc 23:2399-2406
5. Kang SW, Lee SC, Lee SH. Lee KY, Jeong JJ, Lee YS, Nam KH, Chang HS, Chung WY, Park
CS (2009) Robotic thyroid surgery using a gasless, transaxillary approach and the da Vmci S
system: the operative outcomes of 338 consecutive patients. Surgery 146: 1048-1055
6. Lee S, Ryu HR, Park JH (2011) Excellence in robotic thyroid surgery: a comparative study
of robot-assisted versus conventional endoscopic thyroidectomy in papillary thyroid micro-
carcinoma patients . Ann Surg 253: 1060-1066
7. Huberty V, EI Bacha S, Blero D et al (2013) Endoscopic treatment for Zenker's diverticulum:
long-term results (with video). Gastrointest Endosc 77:701- 707
8. Manno M, Manta R, Caruso A et al (2014) Altemative endoscopic treatment of Zenker's di-
verticulum: a case series (with video). Gastrointest Endosc 79:168-170
Thoracic Surgery
4
Giulia Veronesi
G . Veronesi ( ~ )
Division of Thoracic Surgery, European Institute of Oncology,
Milan, Italy
e-mail: [email protected]
The patient is positioned in a lateral decubitus position with the arnIS in front of
the face as in Figs. 4.lb and 4.lc . The operating table should ideally fold down
at the level of the fifth intercostal space , to lower the level of the hips; alterna-
tively a pillow can be placed under the chest. The patient is intubated with a dou-
ble lumen endotracheal tube. The operation starts with the introduction of the
high-definition endoscope through the port at the level of the seventh intercostal
space in the anterior axillary line (Fig. 4.la). Under endoscopic control, the util-
ity incision is made through the fourth or fifth intercostal space. In 10ng-linIbed
patients, use of the fifth intercostal space may make it easier to reach the
diaphragm and the lower ligament.
Entry port positioning is standard for all lobectomies and segmentectomies
(Fig . 4.la), although on the left the endoscope port may moved 2 cm posterior-
ly compared to the right side inorder to avoid the heart obscuring hilar stmc-
tures. Furthenllore, if CO~ insufflation is required (obese patients and small cav-
ities), the utility incision for specimen removal is made at the end of the proce-
dure , and in a lower position thannonllal- just above the diaphragm . The robot
is positioned behind the patient's head, 20-30° off the midline (Figs . 4.lb and
4.lc). Docking requires 5-10 minutes, starting " 'ith the endoscope amI , fol-
lowed by the two posterior arms, and then the anterior arm through the utility
mC1Slon.
For mediastinal lesions the patient is positioned in a semilateral decubitus
position with the preferred side (usually left) above. The arms are positioned
along the side of the body. The room setup is described in Figs. 4.2b and 4.2c
for the left and right approach.
Three ports are used (Fig . 4.2a): fifth intercostal space in the middle axillary
line (submammary sulcus), fifth intercostal paramediastinic site , and third inter-
costal space at the anterior axillary line . A 30° endoscope is used, introduced via
the central port. Lifting of the chest wall with the endoscope helps enlarge the
chest cavity; C02 insufflation (8-10 mmHg) is also used. In some cases a right
approach is used depending on the position of the lesion or surgeon preference .
4.3.1 Lobectomy
An anterior approach lobectomy using four robotic arms is described. The resec-
tion begins by isolating hilar elements using a hook (or Maryland bipolar for-
ceps) and two forceps. The hook is manipulated by the right robotic arm intro-
duced through the utility port for right lung lobectomies, or the posterior eighth
4 Thoracic Surgery 31
c
Fig. 4.1 a Positions of entry ports for right lobectomy. b, c OR setup for right (b) and left (c) lung
resection. « (~ 2014 Intuitive Surgical. Inc.)
32 G.Veronesi
'~~
..,...
(j),
...•
a
c
Fig. 4.2 a Port placcmcnt for resection of anterior mediastinal lesions. b OR setup for right
approach to anterior mediastinal lesions. c OR setup for left approach to anterior mediastinal
lesions. (<<-j 2014 Intuitive Surgical, Inc.)
4 Thoracic Surgery 33
intercostal port for left-side lobectomies . One of the forceps (fourth robotic ann)
is used to retract the lung and expose the stmctures . The other forceps are
manipulated by the left robotic arm and used to grip stmctures during dissection:
it is introduced through the utility thoracotomy for left-side lobectomies or the
posterior eighth intercostal space for right-side lobectomies. When a hilar ves-
sel or bronchus is ready to be surrounded with a vessel loop and stapled, a third
pair of forceps is introduced (substituting the hook). The vessels and bronchus
are sectioned with a stapler introduced by the assistant surgeon through the util-
ity incision for upper lobes and through the posterior port after removing the
robotic arm for lower lobes . The pulmonary vein is usually the first stmcture to
be exposed and divided. If the lesion is in the right upper lobe, vein division is
followed by exposure of the pulmonary artery branches and sectioning, and then
bronchus exposure and sectioning. Anterior and posterior parenchyma division
is completed with an EndoStapler (fissureless technique). If the lesion is in the
right lower lobe, the sequence is usually: vein, artery, fissures and bronchus . If
the anterior fissure is incomplete it can be completed after bronchus stapling.
When performing a middle lobectomy, the most favorable sequence is vein,
bronchus, artery.
For the left upper lobe , after pulmonary vein sectioning, the first mediastinal
artery is exposed and transected to allow for the introduction of the stapler for
bronchus sectioning. The posterior and lingular arteries are then isolated and sta-
pled or cut between Hem-o-lok@ clips, and the lobe is removed.
Left lower lobectomy starts with vein exposure and transection, followed by
the artery and bronchus. Incomplete fissures are usually completed with an
EndoGIA introduced by the assistant through one of the ports. The lobe 1S
extracted through the anterior utility thoracotomy using an EndoCatch.
4.3.2 Segmentectomy
For suspected primary lung cancer but no preoperative diagnosis, a VATS wedge
resection with frozen section examination may be performed to confirm malig-
nancy before proceeding to segmentectomy. However, if possible this phase
should be avoided as intersegmental planes can be dismpted by the atypical
resections . For centrally located cancers , the operation proceeds directly to
anatomical segmentectomy. Pulmonary artery and vein branches are divided
with a Hem-o-lok@' or a stapler; bronchi and parenchyma are divided with the
stapler, which is introduced by the assistant. Dissection is usually perfonned
with Cadiere forceps in one hand and hook in the other. A spatula with monopo-
lar cautery or a bipolar tool (Maryland) in the right hand can be used as an alter-
native to hook cautery. Fibrin glue is used to control parenchyma air leakage
when required.
For right upper lobe dorsal segmentectomy, the first step is dissection of the
interlobar portion of the pulmonary artery so as to identify the posterior ascend-
34 G.Veronesi
ing segmental artery, to the dorsal upper lobe, and the 10vI'er lobe apical segmen-
tal artery. The ascending branch is then cut between Hem-o-Iok® clips. The
upper lobe is pulled anteriorly with the forceps on the fourth arm and the
bronchial tree is isolated to expose the upper lobe and intennediate bronchus.
The dissection continues by identifying the branch to the dorsal segment, which
is encircled and divided by the stapler. When visible, the segmental vein is
divided between Hem-o-Iok® clips. The segmentectomy is completed by divid-
ing the parenchyma betVl'een the dorsal and apical segments with a stapler. The
resected specimen is extracted in an Endobag through the anterior utility inci-
SIon.
For right upper lobe anterior segmentectomy, the mediastinal pleura is com-
pletely incised anteriorly from the middle lobe vein to the truncus arteriosus .
The anterior segmental vein is isolated and divided; the anterior segmental
artery coming from the lower truncus arteriosus is isolated and divided. The
anterior portion of the horizontal fissure is completed with the endovascular sta-
pler, and the anterior segmental bronchus is isolated and divided. The remaining
segmental fissures are divided with staplers, and the segment is removed .
For right lower lobe upper segmentectomy, if performed from posterior to
anterior, the sequence of hilar structure transection is vein, bronchus , artery;
when the approach is from the fissure , the sequence is artery, bronchus, vein.
The posterior to anterior technique is described: with the lung retracted anteri-
orly, the pulmonary ligament is divided, and the lower pulmonary vein dissect-
ed bluntly to expose the lower lobe apical segmental vein. After division of the
vein, the bronchus is exposed with the lung still retracted anteriorly. The lower
lobe apical segmental artery is then identified in the fissure and stapled (or
clipped with Hem-o-Iok® clips). Lastly, the parenchyma is transected using a
stapler introduced through the anterior utility incision.
For right lower lobe basilar segmentectomy, the sequence is vein, artery,
bronchus. The lung is first retracted posteriorly, the lung liganlent is transected,
and the vein branch is isolated and divided. The next step is to identify the seg-
mental artery within the fissure , which is then cut by a stapler. The bronchus is
then isolated and divided by a stapler followed by parenchyma division with an
articulated stapler.
Left lower lobe trisegmentectomy with lingual sparing proceeds in the
sequence vein, upper artery, bronchus, ventral artery, dorsal artery. The proce-
dure starts with blunt dissection of the upper pulmonary vein. The main branch
is isolated from lingular branch and divided by the stapler. This is followed by
gentle dissection of the artery to the upper segment (one or two branches) . The
culmen bronchus is than isolated and divided by the stapler. One or more ven-
tral and dorsal branches are then isolated and cut after Hem-o-Iokilt clipping.
Fissure division is the last step.
The sequence for lingual segmentectomy is artery, vein, bronchus. The artery
in the fissure is first exposed so as to identify the lingular artery, which is then
divided between Hem-o-Iok® clips or with stapler. The segmental vein is isolat-
4 Thoracic Surgery 3S
ed and divided betvveen Hem-o-Iok® clips while the bronchus is divided by a sta-
pler. Fissure division is the last step.
The techniques for superior segmentectomy and basilar segmentectomy of
the lo\\'er left lobe are similar to those for right-side segmentectomies.
Radical lymph node dissection can be perfomled before or after lobectomy. but
suspicious lymph nodes are usually removed before lobectomy. The technique is
similar to that used in open surgery. The paratracheallymph nodes on the right
side are removed first, usually avoiding division of the azygos vein. The medi-
astinal pleura, between the superior vena cava and the azygos vein, is then
incised and the lymph nodes, together with the fatty soft tissue of the paratra-
cheal region are removed en bloc using the hook and forceps . Sometimes a PK
system, UltraCision, is used in patients with large quantities of mediastinal fat
or enlarged lymph nodes .
The nodes of the subcarinal station are removed after resection of the pul-
monary ligament and retraction of the lung toward the anterior mediastinum to
expose the posterior mediastinum. The bronchial arteries can usually be avoid-
ed since visibility is generally good , if not they are coagulated and a clip is not
usually required. Fibrin sealant (Tissucol, Baxter) may be used at lymphadenec-
tomy sites to reduce lymphorrhea, but is rarely required; sealant may also be
applied to the bronchial stump and fissure surface, to reduce air leakage.
Absorbable hemostatic agent is applied , and a single 28Ch pleural drain is
emplaced at the end of the operation .
The stereoendoscope is introduced through the central port of the fifth inter-
costal space. The forceps for the right hand , and hook for the left hand, are intro-
duced under endoscopic visual control. Radical thymectomy begins by incising
the mediastinal pleura above the phrenic nerve . The lower part of the thymus
and mediastinal fat are dissected moving toward the apical region. The anony-
mous vein is best isolated from the right side where the subclavian vein is clear-
ly visible . Intercostal mammary vessels are spared and mark the boundary of the
anterior chest wall incision . Radical extended thymectomy is usually perfonned
en bloc with mediastinal fat tissue . The fatty tissue of the superior poles, peri-
cardium, and brachiocephalic truncus is removed. Hook cautery or ultracision is
used for the dissection and a Cadiere forceps for retraction. The boundaries of
the resection are the diaphragm below, the thyroid above, and the phrenic nerves
laterally. The small veins are controlled with monopolar cautery or ultracision.
Large vessels are clipped. The controlateral pleura can be left intact if the lesion
36 G.Veronesi
is small; for larger tumors the contralateral pleura can be opened to remove all
the mediastinal tissue without injury to the phrenic nerve, which is identified
and spared. The specimen is removed through the parasternal incision, which
may sometinles require enlargement. A single 28Ch chest tube is emplaced via
the same port.
Intensive care is not nonnally necessary. Patients are typically awakened in the
operating room soon after surgery has been completed and are then brought to
the ward. Chest X ray and blood tests are done in the inmlediate postsurgical
period. Patients are mobilized and start pulmonary rehabilitation (for those who
have undergone lung surgery) on the first postoperative day; the vesical catheter
is removed on the same day if diuresis is adequate (>30 cclh) and there are no
other contraindications. The drain is removed when less than 350-400 cc has
accumulated over the preceding 24 hours, and air leaks are absent. Discharge is
possible on the same day as the drain removal (third postoperative day at the ear-
liest). In the event of prolonged air leakage, a Heinllich valve is attached to the
chest tube , and discharge planned for the fifth postoperative day in the absence
of further contraindications .
Robotic thoracic surgery is still relatively new, but sufficient experience has
accumulated to justify considering it as the future of minimally invasive surgery.
Nevertheless, the claimed advantages of robotic surgery - high-definition
stereoscopic view, improved dexterity due to more degrees of movement of the
instuments, lack of the instrument fulcnlln effect that occurs with VATS , tremor
filtration, and greater surgeon comfort - have not been shown to produce supe-
rior clinical outcomes, so data supporting these perceived benefits are urgently
required. The high capital and running costs [7], together with the requirement
for the entire operating teanl to learn a new set of skills, have slowed down the
adoption of robotic systems . Although it may be easier for the surgeon to learn
robotic thoracic surgery than VATS, team learning may be longer and more dif-
ficult. Other disadvantages of robotic systems are the lack of tactile feedback
with an inability to feel small lung nodules. However robotic technology is
evolving quickly and the use of robotic thoracic surgery increased more than in
any other surgical discipline in 2012-2013. Our own experience with the da
Vinci® Surgical System (Intuitive Surgical) suggests that about 20 operations are
required, for a surgeon experienced in open thoracic surgery (but not VATS), to
achieve competence in robotic lobectomy [9]. Available data on complications,
number of lymph nodes removed, and patient survival are encouraging, [3-6,9,
4 Thoracic Surgery 37
10] supporting the safety and oncological radicality of the robotic approach to
lobectomy in lung cancer patients.
There are a number of variations in surgical technique. We use a four-ann
approach - three robot arms and the utility incision [1]. Some surgeons use only
three anns [5]; others make the utility incision at the end of the procedure
because they insufflate the chest cavity with C02 [5, 6]. The position of the util-
ity incision varies with surgeon preference. Other teams use a hybrid robot-
VATS approach [11].
hl these cases , the robot is used for vascular, hilar, and mediastinal dissec-
tion, followed by VATS lobectomy.
The radicality of minimally invasive approaches to mediastinal lymph node
dissection has long been debated [2] . Extent of pathologic nodal upstaging can
be a surrogate for the completeness of node dissection and the quality of surgery.
Wilson et al. [12] detemuned the rate of hilar (pNl) nodal upstaging in 303
patients with stage I non-small-celliung cancer, who underwent robotic surgery
for their disease , and compared them historic lular upstaging rates for VATS, and
thoracotomy. They found that nodal upstaging follovving robotic resection
appeared superior to VATS and comparable to thoracotomy, suggesting that the
completeness of node dissection by robotic surgery is sinular to that obtained by
open surgery. In a study by our group [3], the median number of lymph nodes
removed from patients undergoing lobectomy did not differ between robotic and
open lobectomy, again suggesting that node dissection is reassuringly complete
with the robotic approach.
Literature data on robotic mediastinal surgery are linuted. Rea et al. [13]
reported on 33 thymectomies performed on myasthenia gravis patients . Total
thymomectomy is often necessary to obtain long-lasting or pennanent renussion
of muscle weakness in patients with myasthenia gravis and thymoma. More con-
troversially it is also used on myasthenic patients without thymoma. The study
concluded that robotic thymectomy was safe and effective, with complete remis-
sion or much improved symptomatology in 92% of the series. The advantages of
the left-sided approach were emphasized, but it was considered that long-tenn
follow-up was essential to COnfiml benefit. Ruckert et al. [14] compared tradi-
tional monolateral VATS with a robotic approach for thymectomy myasthenia
gravis patients. Duration of surgery, rate of conversion and morbidity vvere sim-
ilar in the tvvo groups, however patients receiving robotic surgery had a higher
rate of complete renussion that was attributed to more complete removal of thy-
mus tissue with the robotic approach.
The largest published experience to date on robotic thymectomy is a study
that assessed early and late results in patients with early-stage (Masaoka I or II)
thymoma, treated at four European centers betvveen2002 and 2011 [7]. Seventy-
nine patients of median age 57 years received left-sided (82.4%), right-sided
(12.6%), or bilateral (5%) robotic surgery. Forty-five (57%) had myasthenia
gravis. The mean operating tinle was 155 minutes and median hospital stay was
3 days. The median resected tumor diameter was 3 cm (range: 1-12 cm). After
38 G.Veronesi
Table 4.1 Summary of results of major pUblications (>30 cases) on robotic lung resections (lobec-
tomies and segmentectomies)
a median follow-up of 40 months , 74 patients were alive and five had died (four
from non-thymoma causes and one of disseminated intrathoracic recurrence).
Five-year survival vvas 90% comparing favorably vvith historical results.
It is important to remove all thymic tissue when perfomung thymectomy in
myasthenic patients in order to maxinuze the chance of long-tenll renussion .
When using a unilateral approach, identification of the contralateral phrenic
nerve is a key landmark for ensuring maxinlUm removals of thynuc tissue.
Waegner et al. [15] used fluorescence imaging with indocyanine green dur-
ing right robotic thymectomy to assist with identification of the contralateral
phrenic nerve. The technique consists of bolus iv injection of indocyanine green
solution. The robot visual system is switched to fluorescent mode and a fluores-
cence response in mediastinal blood vessels is observed about to seconds after
the injection. The contralateral phrenic nerve was thus identified as it nIns par-
allel to the pericardiophrenic neurovascular bundle, which shows up as fluores-
cent. It was found that the contralateral pericardiophrenic neurovascular bundle
was visualized in 80% of patients from a left pleural view, rarely from a medi-
astinal vie ..", and never distal to the aortopuhnonary window. The authors con-
cluded that this technology had the potential to maximize thynuc tissue resec-
tion using the lllulateral approach, while reducing operating time and nerve
injury.
4 Thoracic Surgery 39
References
Before incision, the right lung is desufflated after double lumen tube intubation.
A lO-mm camera port is placed at the sixth intercostal space, posterior to the
posterior axillary line. Under vision, two 8-nlll robot-ports are placed just ante-
rior to the scapular rim in the fourth intercostal space and more posterior in the
ninth intercostal space. These ports are used for the monopolar thermal hook at
arm 1 and Cadiere forceps at amI 2. Two thoracoscopic ports are used in the fifth
and seventh intercostal spaces just posterior to the posterior axillary line . These
ports are used for conventional thoracoscopic assistance such as suction, trac-
tion , and clipping (Fig . 5 .2). C02 insufflation of the thoracic cavity with
6 nuuHg penu.its excellent vision , without the need for retracting the lung from
the operative field. In the case of a noncompliant lung, a retractor can be used.
After division of any pulmonary adhesions and when a proper overview of the
operating field is achieved, the right pulmonary ligament is divided. The parietal
pleura is dissected at the anterior side of the esophagus from the diaphragm up
to the azygos arch. The azygos arch is carefully ligated ..vith robotic Hem-o-lok®
ligation clips in ann 1. (Fig. 5.3) . Then dissection of the parietal pleura is con-
tinued above the aortic arch for a bilateral paratracheal lymph node dissection
(Fig . 5.4). The right vagal nerve is dissected just above the level of the carina .
Right para tracheal lymph nodes (2R and 4R) are dissected over the superior
5 Esophagectomy for Cancer 45
<D o
• 0 Asststant 4
Camera
o
<D . Assistant 6"
.,.
.,.r
a
FIg. 5.2 a Port position. Robotic arms 1 (reI/ow), 2 (greell) and camera (blue) . Two assisting ports
(white) . b Port position. Robotic arms 1 (veIlOlv) , 2 (greell) and camera (blue). Two assisting ports
(white) . (t) 2014 Intuitive Surgical , Inc .)
caval vein up to the level of the right subclavian artery and vein (Fig . 5.5) . Left
paratracheallymph nodes (2L and 4L) are dissected along the left pleura and left
recurrent nerve , that runs just at the tracheal rim. Often the left carotid artery is
visualized. Subsequently, the parietal pleura is dissected at the posterior side of
the esophagus, cranially to caudally, along the azygos vein toward the
diaphragm, including the thoracic duct. At the level of the diaphragm, the tho-
racic duct is clipped vvith a IO-nml endoscopic clipping device (EndoclipTM II;
Covidien, Mansfield, Massachusetts, USA) to prevent postoperative chylous
leakage.
At the level of the diaphragm , a Penrose drain is placed around the esopha-
gus to provide traction by the assistant , facilitating esophageal lifting off the
pericardium and pulmonary veins. The esophagus is then resected en bloc with
the periesophageal (station 8) mediastinal lymph nodes and the thoracic duct
from the diaphragm up to the thoracic inlet. The aorta is completely exposed and
aortoesophageal vessels are identified and clipped with Endoclip by the assist-
ing surgeon . Finally the aortopulmonary window (station 5), subcarinalnodes
(station 7) are dissected. The subcarinalnodes are very fragile and bleed easily,
therefore first the nodes are dissected along the carinal cartilage and then the
surrounding fascia is dissected, preventing grasping of the nodes . A 24-Fr chest
tube is placed, and the lung is insufflated under direct vision.
After completion of the robot-assisted thoracoscopic esophageal mobiliza-
tion , the patient is put in the supine position. An 11-n1111 camera port is intro-
duced left paraumbilically, and an 11-m111 working port is placed at the left mid-
clavicular line at the umbilical level. A 5-nml working port is placed more cra-
46 R. van Hillegersberg
nially at the right midclavicular line . A 5-n1111 assisting port is placed in the left
subcostal area, and a 12-nml port is placed pararectally right for the liver retrac-
tor. The abdomen is insufflated to a carbon dioxide pressure level of 15 n1111Hg.
The hepatogastric ligament is opened. The greater and lesser curvatures are
dissected with ultrasonic hannonic scalpel (Hannonic Ace®, Ethicon Endosurgery,
Johnson&Johnson , New Bnmswick, New Jersey, USA). The hiatus is opened, and
the distal esophagus is dissected from the right and left crus. The carbon dioxide
pressure level is reduced to 6 nmilig to avoid excessive intrathoracic pressure and
a chest tube is placed in the left pleural sinus. Dissection and lymphadenectomy
5 Esophagectomy for Cancer 47
then continues around the celiac tmnk . The left gastric artery and vein are then
transected at their origin with Hem-o-Iok® ligation clips (Teleflex Medical, NC,
US) . Abdominal lymphadenectomy includes lymph nodes surrounding the left
gastric artery and the lesser omental lymph nodes.
The cervical esophagus is mobilized through a left-side longitudinal neck inci-
sion along the stemocleidoid muscle. No fomlal cervical lymph node dissection is
carried out, but cervical lymph nodes are dissected if lymph node metastases are
suspected macroscopically during the cervical phase of esophagectomy. The esoph-
agus is dissected and a cord is attached to the proxinlal part of the specimen to
enable pull-up of the gastric conduit along the anatomical tract of the esophagus.
The esophagus and surrounding lymph nodes are pulled into the abdomen under
laparoscopic vision. A 7-cm transverse incision is made at the level of the left paralUll-
bilical port for extraction of the specimen and stomach using a wOlUld protector.
Outside the abdomen, a 5-cm-wide gastric tube is constmcted with staplers (GIA
TM 80, 3.8 nUll; Covidien, Dublin, Ireland), and the stapled line is oversewn with
3-0 polydioxanone. Routine extracorporal oversewing \vas reintroduced as two seri-
ous complications occurred when the staple line was not oversewn [3] .The specinlen
consisting of the esophagus and cardia of the stomach is sent for pathological exam-
ination. After the gastric tube has been pulled to the neck, a hand-sewn end-to-side
esophagogastrostomy is perfonned in the neck using 3-0 polydioxanone single-layer
mmung sutures. Excess gastric tubing is removed using a GIA stapler.
A feeding jejunostomy (Freka® FCJ-Set, Fresenius Kabi AG, Bad Homburg
vd H., Gennany) is placed at the level of the transverse incision.
48 R. van Hillegersberg
and co-morbidity. The minimally invasive approach may offer a greater percent-
age of patients a potentially curative surgical resection. Patients with stage I-III
disease, i.e ., TI-T4a tumors, and no evidence of distant metastases are eligible
to RAMIE [8].
With our growing experience , especially patients with tumors in the upper
mediastinum and ..vith para tracheal lymph node metastases were resected suc-
cessfully with the thoracoscopic robotic approach (Fig. 5.5). These tumors are
often in close contact with the upper mediastinal blood vessels (superior caval
vein , carotid artery and subclavian artery and vein) . The thoracic inlet is very
difficult to reach with an open or conventional thoracoscopic approach . With the
robot, this area can be reached without any limitations . This extends the opera-
tive and potentially curative options in these specific groups of patients substan-
tially. A recent report of a Korean group shows that a very extensive complete
paratracheallymph node dissection can be performed with the robot [9] .
Following our initial report of RAMIE in 2009 , we analyzed the following consec-
utive series of 108 patients until 2011 . Esophageal dissection vvas completed thora-
coscopically in 88 patients. Conversion to thoracotomy was necessary in 12
patients due to: bulky adhesive tumor in the mediastinum (4): bleeding that could
not be controlled thoracoscopic ally (4); insufficient collapse of the right lung (2);
or inadequate thoracoscopic trocar position (2). Conversion to a transhiatal proce-
dure was necessary in 9 patients due to: insufficient collapse of the right lung (6);
inadequate thoracoscopic port position (1); pleural adhesions (1); or enlarged right
cardiac atrium (unusual anatomy) (1). Conversion of the laparoscopic abdominal
phase was required in 3 patients due to bleeding that could not be controlled laparo-
scopic ally (1) , locally advanced tumor requiring total gastrectomy with colonic
interposition (1) or very low position of the greater curvature (1) . There was a sig-
nificant decrease in the percentage of conversions between the first 54 and second
54 patients (13 (24%) vs. 7 (13%) respectively; p <0·001).
The median set up time for the robot was 17 minutes (range 5-91) . The medi-
an duration of the total procedure was 381 minutes (range 264-550) . The thora-
coscopic phase (88 patients) had a median duration of 175 minutes (range
108-241). There vvas a significant decrease in thoracoscopic operative time
betv.'een the first 44 and second 44 patients who completed the thoracic phase
thoracoscopic ally (199 min versus 166 min respectively; p < 0·001) .
We found a high percentage (95%) of radical resections despite the high rate
of T3 tumors (78%) and only 64% neoadjuvant therapy. A median of 26 dissect-
ed lymph nodes were retrieved . Follow-up was at least 25 months with a medi-
an follow-up of 34 months. Median disease-free survival was 21 months and
median overall survival was 29 months , with a 5-year overall survival of 40%.
The percentage of in-hospital puhuonary infections after RAMIE in our series
50 R. van Hillegersberg
References
1. Luketich JD, Pennathur A. Awais 0 et al (2012) Outcomes after minimally invasive esophagec-
tomy: review of over 1000 patients. Ann Surg 256:95-103
2. van Hillegersberg R, Boone J, Draaisma WA et al (2006) First experience with robot-assist-
ed thoracoscopic esophagolymphadenectomy for esophageal cancer. Surg Endosc 20:1435-1439
3. Boone I , Rinkes ill, van Hillegersberg R (2006) Gastric conduit staple line after esophagec-
tomy: to oversew or not? I Thorac Cardiovasc Surg 132: 1491-1492
4. Biere SS, van Berge Henegouwen MI, Maas KW et al (2012) Minimally invasive versus open
oesophagectomy for patients with oesophageal cancer: a multicentre, open-label, randomised
controlled trial. Lancet 379: 1887-1892
5. Verhage RJ, Hazebroek EJ, Boone J, van Hillegersberg R (2009) Minimally invasive surgery
compared to open procedures in esophagectomy for cancer: a systematic review of the liter-
ature . Minerva Chir 64: 135-146
6. Ruurda JP, Draaisma WA. van Hillegersberg R et al (2005) Robot-assisted endoscopic sur-
gery: a four-year single-center experience. Dig Surg 22:313-320
7. Kerustine KH (2004) Robotics in thoracic surgery. Am I Surg 188: 89S-97S
8. Boone J, Schipper ME, Moojen WA et al (2009) Robot-assisted thoracoscopic oesophagec-
tomy for cancer. Br I Surg 96:878-886
9. Kim DI , Park SY, Lee S et al (2014) Feasibility of a robot-assisted thoracoscopic lym-
phadenectomy along the recurrent laryngeal nerves in radical esophagectomy for esophageal
squamous carcinoma. Surg Endosc 28: 1866-1873
10. Clark I , Sodergren MH, Purkayastha S et al (2011) The role of robotic assisted laparoscopy
for oesophagogastric oncological resection; an appraisal of the literature. Dis Esophagus
24:240-250
11 . van der Sluis PC, Ruurda IP, van der Horst Set al (2012) Robot-assisted minimally invasive
thoraco-Iaparoscopic esophagectomy versus open transthoracic esophagectomy for resectable
esophageal cancer, a randomized controlled trial (ROBOT trial). Trials 13:230
Anti-reflux Procedures
and Cardioesophagomyotomy
6
Gianluigi Melotti, Vincenzo Trapani, Marzio Frazzoni,
Michele Varoli, and Micaela Piccoli
6.1.1 Introduction
M. Piccoli ( ~ )
General Surgery Unit, "Sant' Agostino-Estense" New Hospital,
Baggiovara (MO), Italy
e-mail: m [email protected]
In 1991 , Dallemagne [8] described the laparoscopic technique for the Nissen fun-
doplication. The laparoscopic approach had a lot of benefits over the traditional
open approach, including improved cosmesis, reduced morbidity, decreased hos-
pital stay, decreased respiratory complications , and faster recovery.
The Nissen fundoplication , consisting of a total (360°) vvrap , is the most
commonly perfonlled anti-reflux operation. Despite general reflux alleviation,
dysphagia and gas bloating are the primary causes of dissatisfaction after Nissen
fundoplication. In order to reduce the postoperative dysphagia and gas bloating,
a variety of procedures in which the fundus is only partially v,'Tapped have been
proposed, including the Toupet fundoplication, consisting of a posterior (270°)
wrap, and the Dor fundoplication , consisting of an anterior (180°) wrap. Sinular
efficacy on heartbUrn/regurgitation and reflux parameters but less dysphagia and
gas bloat have been reported when the Toupet fundoplication has been compared
with the Nissen fundoplication [9 , 10]: it should be noted, however, that with
both techniques the fundoplication was perfornled by fixing the wrap to the
anterior wall of the esophagus . The Nissen procedure can also be carried out
without anchoring the fundoplication: no wrap slipping, negligible gas bloating,
and cumulative incidence of postoperative dysphagia quite sinular to that report-
ed with the Toupet procedure, in conjunction vvith nomlal reflux parameters and
sustained symptom remission in the vast majority of cases have been found with
tlUs technique [11, 12] . To prevent postoperative dysphagia , other key technical
issues must be considered, including division of the short gastric vessels v.'hen-
ever deemed necessary in order to adequately mobilize the esophagus and to
make the fundoplication tension-free , and insertion of a 52-Fr bougie into the
esophagus during construction of the wrap .
Robot-assisted surgery was developed to overcome the linutations of laparo-
scopic technology and to expand the benefits of minimal invasive surgery. The
first report of robot-assisted fundoplication was published by Meininger in 2001
[13] and its use has been increasing since then.
6 Anti-reflux Procedures and (ardioesophagomyotomy 53
The patient-side cart is positioned at the head of the patient and it includes four
robotic arms; the vision system is at the left side of the bed, and the surgeon con-
sole is distant from the bed, but in the same operating room (Fig. 6.1).
/~
gical , Inc.)
• <D . camera • 0 0
Assistant 0
to retract the liver. A S-ll1ll111 assistant port is placed about 2 cm under the umbil-
icalline , between the midline and the mid-clavicular line, in the left side . The senlp
of the robot is usually perfomled by the assistant at the bedside.
(LNF) was carried out by three expert surgeons: nonsignificant differences were
found in the postoperative clinical outcome and in pH-metric parameters between
the two groups of patients. In the study by Draaisma and co-workers [17], 25
patients treated with conventional LNF were compared to 25 patients treated with
robot-assisted LNF: similar postoperative pH-metric results were found in the
two groups of patients but the study detected a statistically significant difference
of only 35% and the two surgical procedures were carried out by seven surgeons.
In a recent study comparing 44 patients who underwent robot-assisted
laparoscopic fundoplication with 44 patients who underwent conventional
laparoscopic fundoplication. at a 3-month postoperative follow-up no difference
in temlS of heartbum remission was found but normalization of acid reflux was
detected significantly more frequently in patients treated with robotic surgery
(100% vs. 86%. P =0.026) [18] . It should be noted that in this study all the inter-
ventions were carried out by the same surgeon. The potential advantages of
robotic surgery include greater precision due to the anti-tremble filter. better
vision as a result of the three-dimensional imaging of the surgical field. and
reduced bleeding . Taken together. these advantages can explain these slightly
but significantly better results after robot-assisted laparoscopic fundoplication in
temlS of postoperative acid reflux control. Currently. anti-reflux surgery is main-
ly advised for PPI-refractory GERD patients and altematives of proven efficacy
are not available [5]. Even a 14% gain in nonllalization of postoperative acid
reflux parameters provided by the robot-assisted technique can be clinically rel-
evant for PPI-refractory patients. It can be concluded that in centers vvhere
robot-assisted laparoscopic fundoplication is available, it should be preferred to
the conventional technique in patients with PPI-refractory GERD.
6.2 Cardioesophagomyotomy
6.2.1 Introduction
Treatment options include medical therapy, such as nitrates and calcium channel
blockers; endoscopic therapy, such as pneumatic dilation or botulinum toxin
injection; and operative cardioesophagomyotomy [19].
Botulinum toxin injection is safe but less effective than balloon dilation; it
requires retreatment and leads to scar fornlation in the submucosal plane, which
can result in a more difficult myotomy and greater rates of mucosal perforation
(up to 30%) during dissection. It should be reserved for patients who are poor
candidates for other more effective treatment options.
Currently, except for high-risk or elderly patients, pneumatic dilation or
laparoscopic cardioesophagomyotomy with anterior fundoplication (Heller-Dor)
represent the standard management for achalasia. Using a graded approach vvith
increasing diameters of the polyethylene balloon dilator from 3.0 to 4 .0 cm, up
to a 93% response rate has been achieved during a follow-up period of up to 4
years vvith a relatively low perforation risk (3%). In a recent multicenter ran-
domized trial, endoscopic dilation (up to three series of pneumatic dilations) and
laparoscopic Heller-Dor had comparable success at two years: 92% for dilation
and 87% for myotomy [20].
The robot-assisted cardioesophagomyotomy has been described for the first
time by Horgan et al. in 2005 [21].
Recently, peroral endoscopic myotomy (POEM) has been introduced as a
promising alternative to the current treatments [22]. Several open label studies
have been published showing excellent short tenll results with only a few com-
plications. However, there is currently insufficient evidence to perform POEM
as routine achalasia treatment. Until the long-ternl efficacy data from random-
ized controlled trials are available, POEM should be restricted to centers partic-
ipating in clinical trials.
References
1. Katz P, Gerson L, Vela M (2013) Guidelines for the diagnosis and management of gastroe-
sophageal reflux disease . Am J GastroenterolI08:308-328
Fitzgerald RC, Di Pietro M, Ragunath K et al (2014) British Society of Gastroenterology guide-
lines on the diagnosis and management of Barrett's oesophagus . Gut 63:7-42
3. Frazzoni M, Conigliaro R. Mirante VG et al (2012) The added value of quantitative analysis
of on-therapy impedance-pH parameters in distinguishing refractory non-erosive reflux dis-
ease from functional heartburn. Neurogastroenterol MotiI24:141--e87
4. Frazzoni M , Manta R, Mirante VG et al (2013) Esophageal chemical clearance is impaired
in gastro-esophageal reflux disease - A 24h impedance-pH monitoring assessment. Neuro-
gastroenterol MotiI25 :399--e295
5. Katz P, Gerson L, Vela M (2013) Guidelines for the diagnosis and management of gastroe-
sophageal reflux disease. Am J GastroenteroI108:308-328
6. Dallemagne B , Perretta S (2011) Twenty years oflaparoscopic fundoplication for GERD. World
J Surg 35:1428-1435
7. Engstrom C, Cai W, Irvine T et al (2012) Twenty years of experience with laparoscopic an-
tireflux surgery. Br J Surg 99: 1415-1442
8. Dallemagne B , Weerts 1M , Jehaes C et al (1991) Laparoscopic Nissen fundoplication : pre-
liminary report. Surg Laparosc Endosc 3: 138-143
9. Broeders JA, Bredenoord AJ, Hazebroek EJ et al (2012) Reflux and belching after 270 de-
gree versus 360 degree laparoscopic posterior fundoplication . Ann Surg 255 :59-65
10. Koch 00 , Kaindlstorfer A , Antoniou SA et al (2013) Comparison of results from a random-
ized trial 1 year after laparoscopic Nissen and Toupet fundophcations . Surg Endosc
27 :2383-2390
11. Frazzoni M, Conigliaro R, Melotti G (2011) Reflux parameters as modified by laparoscopic
fundoplication in 40 patients with heartburn/regurgitation persisting despite PPI therapy. A
study using impedance-pH monitoring. Dig Dis Sci 56: 1099-1106
12 . Frazzoni M, Piccoli M , Conigliaro R et al (2013) Refractory gastroesophageal reflux disease
as diagnosed by impedance-pH monitoring can be cured by laparoscopic fundoplication.
Surg Endosc 2013; 27:2940-2946
58 G. Melotti et al.
13. Meininger DD, Byhahn C , Heller K et al (200 1) Totally endoseopic Nissen fundoplication with
a robotic system in a child. Surg Endose II : 1360
14. Dallemagne B (1999) Treatment of the gastroesophageal reflux syndrome. In Meinero M.
Mouret Ph. Melotti G, (eds.): Laparoseopic surgery; the nineties. Masson, Milan pp. 251-260
15 . Cadiere GB (1995) Trattamento video-Iaparoseopico del reflusso gastro-esofageo-E.M .C . Ro-
ma Tecniche Chirurgiche-Addominale, 40 189 , p 10
16. Morino M , Pellegrino L , Giaccone C et al (2006) Randomized clinical trial of robot-assisted
versus laparoseopic Nissen fundoplication . Br J Surg 93 :553-558
17. Draaisma WA, Ruurda IP, Scheffer RCH et al (2006) Randomized clinical trial of standard
laparoseopic versus robot-assisted laparoscopic Nissen fundoplication for gastro-oesophageal
reflux disease . Br J Surg 93:1351-1359
18 . Frazzoni M, Conigliaro R, Colli G et al (2012) Conventional versus robot-assisted laparoseop-
ic Nissen fundoplication. A comparison of postoperative acid reflux parameters . Surg Endose
26:1675-1681
19. Stefanidis D, Richardson W, Farrell TM et al (2012) SAGES guidelines for the surgical treat-
ment of esophageal achalasia. Surg Endose 26:296-311
20 . Boeckxstaens GE, Annese V, Bruley des Varannes S et al (2011) Pneumatic dilation versus
laparoscopic Heller's myotomy for idiopathic achalasia . New Engl J Med 364:1807-1816
21 . Horgan S, Galvani C, Gorodner MV et al (2005) Robotic-assisted Heller myotomy versus la-
paroseopic Heller myotomy for the treatment of esophageal achalasia: multicenter study. J Gas-
trointest Surg 9 :1020-1030
22 . Bredenoord AI , Rosch T, Fockens P (2014) Peroral endoscopic myotomy for achalasia . Neu-
rogastroenterol Motil26: 3-16
23. Ancona E , Zaninotto G, Costantini M (2003) Miotomie Esofagee. In: Encycl Med Chir (Edi-
tions Scientifiques et Medicales Elsevier SAS, Paris) Tecniche Chirurgiche-Addominale, pp.
40-183
24. Huffmanm LC, Pandallai PK, Boulton I et al (2007) Robotic Heller myotomy: A safe oper-
ation with higher postoperative quality-of-Iife indices. Surgery 142:613-20
25 . ShaligramA, Unnirevi J, Simorov Aet al (2012) How does the robot affect outcomes? A ret-
rospective review of open, laparoscopic, and robotic Heller myotomy for achalasia. Surg En-
dosc 26:1047-1050
Gastrectomy for Cancer
7
Andrea Coratti, Mario Annecchiarico,
and Stefano Amore Bonapasta
M. Annecchiarico ( )
Department of Oncology, Division of Oncological and Robotic General Surgery,
Careggi University Hospital, Florence, Italy
e-mail: [email protected]
According to the articles present in the up-to-date literature, RAG with limi-
ted lymphadenectomy is indicated for Stage la, while RAG with D2 lymphade-
nectomy is indicated for Stage Ib-2a. Mucosal and submucosal tumors, if eligi-
ble, are considered for endoscopic resection first [3, 8]. RAG for more advanced
stages is the object of investigations and at present should be performed only in
highly specialized centers.
We describe the technique of distal subtotal and total gastrectomy with D2
lymphadenectomy.
The robotic cart is docked from the patient's head (Fig. 7.1a). Correct positio-
ning of the robotic cart is fundamental because its axis must coincide with the
working axis , coming from the opposite site.
Because the operating table call11ot be changed once the robot has been
docked, the height and slope of the operating table must be reconfirmed before
the robot is docked.
a b
(i)
/ (i)
• G>• 0
•
C~1IM!n
Anlstllll
9l\. S(.,l
./
~
~ I ~
Fig 7.1 OR setup (a) and trocar placement (b) . (© 2014 Intuitive Surgical , Inc.)
bipolar forceps . R3 is armed with a grasper and used for retraction and expo-
sure;
• The assistant surgeon aids the console surgeon using the accessory port for
aspiration/irrigation, for clip or stapler application and other additional
maneuvers .
7.5.1.1 Exploration
The abdominal cavity is explored in order to assess a definitive staging of dis-
ease . Diagnostic peritoneal lavage is perfonlled routinely. Intraoperative ultra-
sound is performed only if hepatic metastases are suspected. Intraoperative
endoscopy can be conducted to establish the location of the tumor and to mark
the proximal resection margin.
After laparoscopic exploration, the robotic cart is docked as previously
described.
Fig. 7 .2 Lymph node dissection. a Station 6 . b Section of right gastroepiploic artery. c Retraction
of hepatic artery by vessel loop. d Station 7
ded and the nodes of station 7 are removed. The left gastric vein is divided bet-
ween clips or ties .
The gastropancreatic connective tissue is dissected up to esophagogastric
junction along the lesser curvature and the lymphatic nodes of station 1-3 (along
the right cardia and lesser curvature) are removed.
7.5.1.8 Reconstruction
The digestive continuity is restored by intra-corporeal gastrojejunal anastomosis
(GJA) on a Roux-en-Y jejunal loop. In elderly patients , a Billroth II GJA is usu-
ally preferred. In both cases, the jejunal loop is transposed with antecolic recon-
stmction .
A mechanical GJA is carried out with a side-to-side fashion , on the posterior
64 A.(oratti et al.
The surgical steps are the same as for distal gastrectomy, except for the follo-
wing: division of the short gastric vessels and dissection along the gastro-sple-
nic ligament, extent of lymphadenectomy, restoring of the digestive continuity.
7.5.2.2 lymphadenectomy
In order to achieve a correct D2 lymphadenectomy, the dissection is extended
removing the nodes on the distal splenic artery (station lId) and the splenic
hilum (station 10); the spleen is preserved, unless there is massive lymphatic
metastasis of the hilum or direct infiltration of the organ.
The lymphatic dissection is enlarged to remove the nodes on the left side of
the cardia (station 2) .
7.5.2.4 Reconstruction
A long Roux-en-Y is made extracorporeally through the left-side mini-laparoto-
my. The esophagojejunal anastomosis (EJA) can be fashioned both with a circu-
lar stapler or robot-assisted hand-sewing technique.
Up to a year ago, we mainly perfomled mechanical EJA. To fashion this anasto-
mosis, a purse string is carried out on the esophageal stump by robotic hand-
sewn technique; the anvil of a 25 nilll or 29 mm circular stapler is inserted into
the esophageal stump, and the purse string is closed (Fig. 7.3c, d). The circular
stapler is then inserted into the jejunal loop and introduced inside the abdomen
7 Gastrectomy for Cancer 6S
7.6.2 D2-Lymphadenectomy
The robotic system makes easier the dissection of difficult lymphatic areas,
namely stations 7-12a and 14v, \"hich represent the "hot point" in conventional
laparoscopy.
The removal of station 14v is controversial. Originally part of the D2-lym-
phadenectomy for tumors of the lower third, this station has since been exclud-
ed in the latest edition of Japanese Classification. However, the removal of sta-
tion 14v can still be beneficial in cases of macroscopic metastasis of lymph
nodes in station 6. Furthermore, if dissection of the infrapyloric area is started
from station 14v, a good exposure of the pancreas can be obtained before
approaching station 6 . This technique can reduce the amount of mistakes of dis-
section plane, particularly in obese patients. Another difficult area to dissect is
the splenic hilum. By using the robotic system, the surgeon can perform a cor-
rect spleen-preserving dissection thus reducing the likelihood of mistakes of
plane and the risk of bleeding [6, 7,9].
Additionally, the enhanced three-dimensional vision and the high precision
of the robotic movements allo\\' an optimal identification of vascular anomalies,
such as an accessory left hepatic artery coming from the left gastric artery, and
pemuts the dissection of the lymphatic tissue surrounding the main tnlllk.
Given the short follow-up times, the numbers of harvested lymph nodes and the
resection margin are used as indicators of oncological adequacy. Most authors
report a mean number of nodes superior to 30 that are in-line with the recom-
mended standard for conventional open D2-lymphadenectomy. In their recent
meta-analysis including a total of 7200 patients (663 , RAG; 1236, LG; 5301,
open gastrectomy (OG», Hyun et al. [10] reported that the number of retrieved
lymph nodes with RAG was sinular to that for LG and OG, even with a subgroup
analysis matched for the extent of lymphadenectomy and type of gastrectomy.
Likewise, in the meta-analysis from Xiong et al. [3] , there were no differences
observed in the number of nodes retrieved between the RAG and LG procedures.
The majority of studies reported free resection margins at pathological exam-
ination in 100% of cases following RAG [7, 9]. This likely reflects an accurate
preoperative selection of cases, with prevalence for early versus advanced
stages. In our experience, intraoperative endoscopy can be useful to confiml the
location of the gastric lesion and to evaluate the adjacent mucosa.
Currently, there is no significant data available in terms of long-tenn onco-
logical results and survival in robotic gastric surgery. The longest follow-up to
date was reported by Pugliese et al. [11], with a mean observation of 53 months.
According to their data, there are not significant differences in 5-year survival
between LG and RAG.
MIS plays an important role in the therapeutic strategy for advanced disease
offering a less aggressive procedure that can help to reduce the time between
68 A.(oratti et al.
In several studies , it has been reported that digestive restoration was performed
extracorporeally through the same mini-laparotomy used for specimen removal
[9, 10] . This hybrid-open technique ..vas used both in gastrojejunostomy and gas-
troduodenostomy following distal gastrectomy, as vveU as in esophagojejunosto-
my following total gastrectomy. This approach is possible for patients with a
very low BMI, indeed the technique of extracorporeal anastomosis is primarily
used by Eastern surgeons. In cases with high BMI, it is very difficult to perfornl
an extracorporeal anastomosis, unless an incision larger than that for a mini-
laparotomy is used . Other disadvantages of extracorporeal anastomosis are the
lack of appropriate vision and the excessive traction put on the viscera, which
make the application of stapling devices potentially difficult and dangerous.
This is why other authors described an intracorporeal digestive restoration using
a linear stapler for GJA and a circular stapler for EJA [1, 7]. Extracorporeal
anastomosis may negate the potential mininlally invasive surgical approach ,
especially for obese patients, and the intracorporeal techniques appear to be the
preferred solution. The robotic system, moreover, pemlits full hand-sewn
C"robot-sewn") technique of anastomosis . All reconstructions, including Roux-
en-Y jejunal limb, EJA, GJA, or even gastroduodenal anastomosis, can be fully
carried out by the intracorporeal robot-sewn method.
In our present clinical practice we always perfonll intracorporeal anastomo-
sis : mechanical side-to-side GJA after distal gastrectomy, and hand-sewn end-to-
side EJA after total gastrectomy. The hand-sewn anastomosis is achievable
thanks to the ability of the robotic system to provide the surgeon with the nec-
essary tools to perform precise sutures even in deep and narrow spaces .
RAG compared with LG: no differences were found in time to start mobilization
and time to resume diet [6, 10] . The hospital stay is shorter in patients undergo-
ing RAG than in those having LAG, but such a difference is not statistically sig-
nificant [3 , 8]. Hospital stay for RAG is significantly shorter than OG in all
series [10].
7.6.5.4 Complications
In their series of 5839 patients (4542 OG, 861 LAG and 436 RAG) , Kim and
colleagues [12] found that overall rates of complications , reoperation and mor-
70 A.(oratti et al.
tality were similar between the three groups. Postoperative ileus and intestinal
obstmction, as well as intra-abdominal fluid collections and abscesses , occurred
more frequently after open surgery, while anastomotic leakage was significant-
ly more conmlon after minimally invasive approach (LG, 2.1 %; RAG, 2.3% ;
OG , 1.1 %; P = 0·017). The authors hypothesized that the higher rate of leaks in
LG and RAG may be associated with the limited tactile feedback or differences
in staple-line reinforcement. Indeed, almost all anastomoses were made using
staplers and the staple lines were always reinforced by over-sewing sutures dur-
ing OG , vvhereas reinforcement was not perfonned in laparoscopic and robotic
procedures .
In the meta-analyses from Hyun [10] and Xiong [3], the complication rate
and mortality rate does not differ significantly between RAG , LG and OG.
Conversion to open surgery is critical because converted patients have high-
er complication rates and worse oncological outcomes. The conversion rate does
not differ significantly between LG and RAG [3] .
7.7 Conclusions
The most convincing indications for robotic surgery are procedures that involve
a small, deep, fixed operating field or where MIS requires extreme accuracy,
fine dissection and endoscopic suturing [1]. Therefore , the major technical
advantages of the robot-assistance in gastric surgery may be appreciated during
lymph node dissection , bleeding control, intracorporeal reconstmction, enlarged
resections and complex reconstmctions . Finally, the learning curve and repro-
ducibility of RAG seem to be shorter and more feasible than with conventional
laparoscopy.
For these reasons , robotics has the potential to contribute to a standardization
and major diffusion of MIS for the treatment of gastric cancer, making it a rOll-
tine approach even in advanced stages . Hence , RAG could better integrate min-
imally invasive resection with neoadjuvant and adjuvant multimodal therapies .
Longer operation tinIe , higher costs and oncologic equivalency to its coun-
terparts are still unresolved issues , which need further development and investi-
gation.
7 Gastrectomy for Cancer 71
References
I. Giulianotti PC. Coratti A.Angelini M et al (2003) Robotics in general surgery: personal ex-
perience in a large community hospital . Arch Surg 138:777-784
Hashizurne M. Sugimachi K (2003) Robot-assisted gastric surgery. Surg Coo North Am
83: 1429-1444
3. Xiong B. Ma L. Zhang C (2012) Robotic versus !aparoseopic gastrectomy for gastric cancer:
a meta-analysis of short outcomes. Surg Onco! 21 :274-280
4. Hyun MH. Lee CH. Kwon YI et al (2013) Robot versus laparoscopic gastrectomy for cancer
by an experienced surgeon: comparisons of surgery. complications. and surgical stress . Ann
Surg OncoI20:1258-1265
5. Kim MC. Heo GU. lung GI (2010) Robotic gastrectomy for gastric cancer: surgical techniques
and clinical merits. Surg Endose 24:610-615
6. Coratti A.Annecchiarico M . Di Marino M et al (2013) Robot-assisted gastrectomy for gas-
tric cancer: current status and technical considerations . World J Surg 37:2771-2781
7. D ' Annibale A . Pende V, Pemazza G et al (2011) Full robotic gastrectomy with extended (D2)
lymphadenectomy for gastric cancer: surgical technique and preliminary results . J Surg Res
166:e113-120
8. Marano A, Hyung WJ (2012) Robotic gastrectomy: the current state of the art . J Gastric Can-
cer 12:63-72
9. Song J, Oh SJ, Kang WH et al (2009) Robot-assisted gastrectomy with lymph node dissec-
tion for gastric cancer: lessons learned from an initial 100 consecutive procedures. Ann Surg
249:927-932
10. Hyun MH, Lee CH . Kim HJ et al (2013) Systematic review and meta-analysis of robotic sur-
gery compared with conventionallaparoseopic and open resections for gastric carcinoma. Br
J Surg 100: 1566-1578
11 . Pugliese R , Maggioni D. Sansonna F et al (2010) Subtotal gastrectomy with D2 dissection
by minimally invasive surgery for distal adenocarcinoma of the stomach: results and 5-year
survival. Surg Endose 24:2594-2602
12. Kim KM,An IY, Kim ill et al (2012) Major early complications following open. laparoseop-
ic and robotic gastrectomy. Br J Surg 99:1681-1687
Robotic Subtotal Gastrectomy:
a Modified Korean Technique
8
Giuseppe Spinoglio, Giampaolo Formisano, Ferruccio Ravazzoni,
Francesca Pagliardi, and Alessandra Marano
The robotic approach for the treatment of gastric cancer (GC) has been initial-
ly adopted mainly in Asia , where tIus malignant disease is more conlllon than
in Westen! countries and it is diagnosed at earlier stages thanks to a screening
program. Among Asian countries, South Korea started to embrace the robotic
technique in 2005 and now it has become one of the leading countries in robot-
ic gastric cancer surgery. In this chapter vl'e present our experience of robotic
subtotal gastrectomy (RSTG) with D2 lymph node (LN) dissection for GC
where the step-by-step procedure is based mostly on the technique of Dr Woo
Jin Hyung [1].
The patient is secured in the supine position with both anns alongside the body
and the legs are closed; a nasogastric tube is inserted during the anesthesiolo-
gist's preparation. The table is then placed in a 15° reverse-Trendelenburg
position. The surgical cart is placed at the head side of the patient, the 1ll00utor
to the right side of the patient and the assistant stands to the left of the table
(Fig . 8.Ia) .
G_ Spinoglio ( i8l)
Department of General and Oncologic Surgery,
" Ss _Antonio e Biagio" Hospital,
Alessandria , Italy
e-mail: giuseppe _spinoglio@gmaiLcom
. <D
/ 00
<D . 0
•
Camen
Assistant
<,\)\.. SU{
/
!Q.
~
~ ~ ~
~
Fig. 8.1 Subtotal gastrectomy. a Overhead view of the OR setup. b Trocar layout . (02014 Intui-
tive Surgical, Inc .)
to achieve the correct angle with the ultrasonic shears to perfonn a complete LN
dissection of station No . Up.
After the laparoscopic exploration of the abdominal cavity to exclude the pres-
ence of metastatic disease, the left lobe of the liver is retracted toward the
abdominal wall with a liver- suspension method [2].
The teclmique requires: two X-ray-detectable 5 x 5 cm gauze pads, one
90 cm 2-0 nylon monofilament with a 60 oml double straight taper needle and
two plastic surgical clips (Hem-o-lokQt, Weck; Teleflex Medical Europe Ltd) .
The gauze pads are folded in half, threaded VI'ith the 2-0 nylon suture, then both
straight needles, with the attached suture-threaded gauze pads, are introduced
into the abdominal cavity through the assistant's port .
One of the straight needles is brought out through the anterior abdominal wall,
directly anterior to the middle of the left lobe of the liver.
The nylon suture is then secured to the pars condensa of the hepatogastric liga-
ment with two Hem-o-Iok® clips, after having divided the pars flaccida up to the
right of the esophageal cms. The other straight needle is passed extemally through
the abdominal wall immediately to the right of the falcifornl ligament. While the
assistant slowly pulls upward on the two sutures, the two gauze pads are brought
together and stretched to cover the undersurface of the retracted liver. Finally, the
two external sutures are tied together to rest on the external abdominal wall.
Originally, in the case of small or undetectable GC, portable plain radiogra-
phy was perfornled to identify endoscopic metallic clips, applied just proximal
to the tumor, during the preoperative upper endoscopy [3] . Actually, we use an
ultrasound evaluation with a curved linear probe (ProARTTM Robotic
Transducer), which is introduced into the abdominal cavity through the assistant
port. Once the tumor location is identified, the patient-side cart is moved next to
the patient and the robot is docked, as previously described .
An omentectomy is perfonned using ultrasonic shears (R2) toward the lower pole
of the spleen where the left gastroepiploic vessels are sectioned at their roots.
Once entered into the lesser sac, coloepiploic detachment is continued toward the
pyloms and the right gastroepiploic vessels are identified and ligated with clips
(LN station No. 4sb, No. 4d). The dissection proceeds along the gastroduodenal
artery to where it springs from the common hepatic artery (CHA).
76 G. Spinoglio et al.
FIg. 8.2 Complete dissection oflymph node No. 12a around proper hepatic artery (PHA). The me-
dial side of portal vein (P V) is pictured
The lesser omentum is dissected and the right gastric artery is identified and
divided betvveen clips at its origin. The supra- and infrapyloric nodes (LN sta-
tion No. 5, No.6) are removed: in the case of a suspected metastatic LN at sta-
tion No. 6 , the dissection of LN station No. 14v might be of benefit to the
patient. The duodenum, after being circumferentially cleared, is transected 2 cm
distal to the pyloms using a 45 mm Endo-linear stapler inserted through the
assistant port . The removal of LN station No. 12a is then carried out (Fig. 8.2) .
The dissection continues by ablation with ultrasonic shears of LN station No. 8a (Fig.
8.3) and No. 9; during this step the left gastric vein is exposed and divided. Then, the
left gastric artery is carefully transected at its root with clips and the soft tissues
around it are retrieved for LN station No.7 (Fig. 8.4). The splenic vessels are [wally
skeletonized with the removal of LN No. IIp.
The retroperitoneal detachment of the stomach is completed along the lesser cur-
vature up to the right esophageal cms together with the retrieval of LN station
No . 3 and No. 1. Once the mobilization of the stomach is completed , the resec-
8 Robotic Subtotal Gastrectomy:a Modified Korean Technique 77
Fig. 8.3 Complete dissection of lymph node No. 8a around common hepatic artery (CHA). GOA,
gastroduodenal artery; RGA, right gastric artery
Fig, 8.4 Clipping of the left gastric artery (LGA) with removal of lymph nodes Nos . 7 and 9 . CHA,
common hepatic artery
8.4.6 Reconstruction
After the transection, a gastrotomy on the posterior wall of the stomach and then an
enterotomy 15-20 cm away from the Treitz ligament are perfomled by means of
ultrasonic dissection. A gastrojejunostomy is perfomled intracorporeally, using a
60 nUll Endolinear stapler; the COnltllOn entry hole for the anastomosis is closed
with interrupted 3/0 PDS sutures. The robot is undocked; a suction drain is posi-
78 G. Spinoglio et al.
tioned close to the duodenal stump and finally the specimen is removed through a
Pfannenstiel incision. The nasogastric tube is left in place and removed on POD I .
8.5 Advantages
Table 8.1 Experience of our Institution of Robotic Subtotal Gastrectomy for Cancer
Variable G group (0=26)
Age (Y). meoll ± D 67 .7 ± 10.6
Gender ( :J/F) n (%) 1719
BMI k glm 2), meall ± D 24 ± 3.8
o-morbidit ,II (%)
HypenensioniCOPD/CAOD/ 7 (_6.9)/3 (11.5)/2 (7.6)/
CVDlHistory of previous rumorJOther* 2 (7 .6)/2 (7 .6)/3 (115)
A score, II (%)
IIIIIIII 6 (23.1)/14 (53 .8)/6 (23.1)
Extent of lymph-node Iii tion , II (%)
DIID2 2 (7.6)/24 (92.3)
onsole tim e (min), meall ± D 190.3 ± 48.9
Intraoperati ve complications, II (%) 0(0)
iquid diet Illrt day), meall ± SD 2.9 ± l.6
Ho pital stay (days, meoll ± D 92 ±2.9
omplicalions 11(%)
onelPresent 19 (73.1)n (26 .9)
Hyperamylasemia/Prolonged ileus! Pu.eumonia 1 (3.8)12 (7 .6)/4 (15.3)
Thmor location , II (%)
middle third/lower third 7 (26.9)/19 (73 .1)
Hi tology t pe, n (%)
intestinal/diffuse/mixed type 11 (42.3)n (26.9)/8 (30.7)
Grading, TI (%)O/1I213 2 (7 .6)/3 (11.5)110 (38 .4)111 (423)
AJ C tage, II (%)
0/ W m/ ITA! lIB/ mAl IIIB 2 (7 .6)/4 (15.4)/4 (15 .4)/1 (3 .8)/5
(19 2 )13 (1l5)n (26.9)
egatiYe proximal margin, II (%) 26 (100)
umber of r trieved L , mean ± D 295 ± 12.6
GC, gastric cancer; 8MI, body mass index; COPD, chronic obstructive pulmonary disease;
CAOD, coronary artery obsbUctive disease; CVD, cerebrovascular disease; LN, lymph nodes.
*Other, co-morbidity includes inflammatory bowel disease , diabetes and lithiasis of gallbladder
80 G. Spinoglio et al.
References
I. Hyung W, Woo Y, Noh S (2011) Robotic surgery for gastric cancer: a technical review. Jour-
nal of Robotic Surgery 5:241-249
2. Woo y , Obama K, Son TI et al (20 II) Minimizing hepatic trauma with a novel liver retrac-
tion method during robotic and laparoscopic gastric cancer surgery, a simple liver-suspension
with gauze-suture technique . Int J Med Robotics ComputAssist Surg 7:56
3. Kim HI , Hyung WJ , Lee CR et al (2011) Intraoperative portable abdominal radiograph for
tumor localization: a simple and accurate method for laparoscopic gastrectomy. Surg Endosc
25:958-963
4. Kim MC, Heo GU, Jung GJ (2010) Robotic gastrectomy for gastric cancer surgical techniques
and clinical merits . Surg Endosc 24:610-615
5. Japanese Gastric Cancer Association (1998) Japanese classification of gastric carcinoma - 2nd
English edition. Gastric Cancer 1: 10-24
6. Japanese Gastric Cancer Association (2011) Japanese classification of gastric carcinoma: 3rd
English edition. Gastric Cancer 14:101-112
7. Japanese Gastric Cancer Association (2011) Japanese gastric cancer treatment guidelines
2010 (ver. 3). Gastric Cancer 14: 113-123
8. Edge S8, Compton CC (2010) The American Joint Committee on Cancer: the 7th edition of
theAJCC cancer staging manual and the future ofTNM.Ann Surg OncoI17:1471-l474
Part III
Hepatob iI iopa nc reatic Su rgery
Hepatic Resections
9
Alberto Patriti, Graziano Ceccarelli, and Luciano Casciola
A. Patriti (183)
Department of Surgery Division of General, Minimally Invasive, and Robotic Surgery,
"San Matteo degli Infermi" Hospital , Spoleto (PG)
e-mail: albertopatriti@gmail .com
the possibility given by the robot to preserve liver parenchyma even in the case
of tumor location in the P-S segments or close to a major liver vessel [5] . In the
present chapter, technical details of robot-assisted liver resection will be provid-
ed along vvith a discussion on the current clinical applications of robotics in
hepatobiliary surgery.
The Pringle maneuver is the easier and safer way to control the inflow but its use
in laparoscopic surgery has been neglected for years. An ideal system for laparo-
scopic Pringle maneuver should be cost-effective , as user-friendly as to pemnt
intermittent inflow occlusion during parenchymal transection and guarantee a
fast vascular control in case of bleeding . The device for inflov.·· occlusion set in
our institution is composed of a 20 Fr chest tube, an umbilical tape and a plug
used for occlusion of the Foley catheter [6] .
The chest tube is inserted in the right upper abdominal quadrant and the
umbilical tape is passed around the hepatoduodenal ligament with the use of
an endowristed Cadiere forceps (Intuitive Surgical Inc ., Sunnyvale , CA) or an
articulated laparoscopic device (Endoflex) . The umbilical tape is then exteri-
orized through the chest tube with the use of a 5-n1111 laparoscopic forceps. The
chest tube is finally closed with the plug in order to avoid air loss . When
inflow occlusion is needed the on-table surgeon removes the plug and pulls the
umbilical tape . When the desired tape tension is achieved , the chest tube is
closed with the plug.
For major hepatectomies individual dissection and control of portal and arte-
rial pedicles is possible and greatly facilitated by endowristed instruments.
Dissection is carried out as in open surgery with scissors or hook and a bipolar
forceps (Fig . 9.1) .
86 A. Patriti et al.
Fig. 9.1 Hilar dissection for right hepatectomy. The right portal vein is encircled and exposed
T\\'o robotic devices are available for parenchymal transection. The Kelly
clamp-cmshing technique associated with an intemuttent Pringle maneuver is
considered the safer and more accurate method of parenchymal transection in
open liver surgery. The major advantages of the Kelly clamp-cmshing tech-
nique are the low costs and the ability to fragment the parenchyma, preserving
the vascular stmctures that can be ligated and divided or preserving according
to the resection plan . Using the endowristed PreCise™ bipolar forceps (Intuitive
Surgical Systems, Sunnyvale , CA , USA) the parenchyma can be easily frag-
mented exposing the inner vessels as in open surgery. The on-table surgeon
uses the forceps to perfonn the intermittent inflow occlusion, thereby allowing
the console surgeon to focus luslher attention only on the transection line . The
robotic clamp-cmshing technique allows parenchymal preservation even for
deeply located lesions, widening the indications for a minimally invasive
approach to lesions in the P-S segments and those located close to major liver
vessels. Hemostasis of small vessels is obtained with monopolar or bipolar
cautery. To secure larger vessels on the transection line, we use Hem-o-Iock®
clips or ligatures with Vicryl® and Prolene® . The hepatic veins (HVs) are usu-
ally divided with the laparoscopic linear stapler or sutured with Prolene® .
Biliostasis is assessed by observation and the bile leaks controlled with sutures
as in open surgery.
The hamlonic scalpel is the other device used to transect the liver. It can be
used to cut and coagulate the liver or, by taking advantage of ultrasonic dissec-
tion , using it to fragment the liver parenchyma exposing the underlying vessels.
The only drawback of this device is related to its four degrees of freedom mak-
ing it ideal for straight-line resections (left and right hepatectomies) but less
effective for parenchyma-preserving resections.
9 Hepatic Resections 87
For resections on anterior segments, left lateral sectionectomy included, some stay
sutures can be useful to retract the specimen using the fourth robotic ann or the
assistant grasper. A vessel-loop held by the fourth amI around the left lobe can be
used to stretch the parenchyma along the transection plane thus facilitating the
effect of the Kelly clamp-cmshing technique. For deeply located lesions or when
the tumor is close to a major vessel the "corkscrew technique" can be reproduced.
After identification of the lesion by inspection and intraoperative ultrasound,
Glisson's capsule is marked with electrocautery 1-2 cm away from the tumor mar-
gin. According to the location of the tumor the marked area is anchored by stitch-
es, with caution, in order to prevent the needle from entering the tumor. The suture
is held together by clips and upward traction is perfonned, facilitating the transec-
tion of the parenchyma and correct identification of vascular and biliary stmctures.
Parenchymal transection is perfonned with the monopolar shears for the first liver
layer (1 cm from the Glisson capsule) and then "vith the Kelly clamp-cmshing tech-
nique . For steady exposure of the parenchymal transection plane in major hepatec-
tomies, the "mbber band technique" can be used. Two sterile mbber bands are
introduced into the abdominal cavity. One end of each mbber band is anchored with
stay sutures at the right and left resection margins . The other end is pulled outside
using an EndoClose needle and fixed with appropriate tension.
The patient is placed in mild reverse Trendelenburg position, with semi-left lat-
eral decubihls and spread legs . The pneumoperitoneum is made using a Veress
needle in the left upper quadrant, maintaining a pressure of 12 mmHg . Four tro-
cars are placed for the robot, one in the right pararectal for the camera, two
8 mm trocars in the right flank and left pararectal, and another 8-nlll trocar for
the fourth arm in the left flank . Two additional lO-nl1ll trocars are inserted to be
used by the assistant (Fig. 9.2a). The OR setup in shown in Fig. 9.2b.
Cholecystectomy is completed and the hepatic hilum is dissected using the
monopolar hook in the right antI of the robot, and the bipolar forceps in the left.
The right hepatic artery is the first element to be sectioned between ligahues or
Hem-o-lock~' clips. The right portal vein is exposed , looped and sectioned
between Hem-o-lock® clips or 5-0 polypropylene suhlre .
The dissection and evaluation of the biliary tree requires special attention
because sectioning will be made either at this time or later during the procedure,
depending upon the location of the biliary bifurcation. Recent visualization tech-
nology with ICG (Indocyanine Green) fluorescence enables the evaluation of the
A. Patriti et al.
FIg. 9.2 a Trocar disposition for right hepatectomy. b OR setup for right and left hepatectomy.
For left hepatectomy arm 3 is positioned on the opposite side . (© 2014 Intuitive Surgical, Inc .)
extrahepatic biliary tree without requiring any invasive procedure; this is due to
the physical characteristics of ICG , which allows for unique hepatic metabolism
and biliary excretion (Fig. 9.3) . The right liver is retracted to the left using the
fourth arm. The first step is to section the right triangular ligament, using the
hook. The liver is gently retracted to the left while dissection proceeds to the
bare area of the liver. At this point, the space behveen the vena cava and the liver
is dissected in a caudal-to-cephalad direction, sectioning between Hem-o-lock®
clips the accessory veins to segments 6 and 7. For a better hemostasis,
polypropylene 5-0 sutures can be applied on the side of the vena cava. A hepa-
tocaval dissection is then made up to the drainage of the RHV.
The ischemic delineation of the transection along Cantlie's line is marked
following the section of the vascular vessels of the right liver. The monopolar
hook can be used to mark this line and open the capsule of Glisson, although it
is not mandatory. Prior to starting the transection , two stay sutures at the tran-
section margins can be used for a better retraction. The transection is performed
using the hannonic scalpel , closing gradually as ultrasonic energy is applied.
The transection is performed in a caudal-to-cephalad direction. Once the vascu-
lar branches are crossing between right and left liver lobes, they are ligated or
secured with Hem-o-lock® clips. The surgical specimen is then extracted in an
endoscopic plastic bag , through a Pfannenstiel incision. The surgical hemostasis
and the trocars ' orifices are again evaluated and two closed Jackson-Pratt drains
are left in place.
9 Hepatic Resections 89
Fig. 9.3 leG fluorescence highlighting the extrahepatic biliary tree during a right hepatectomy
./
my and left lateral sectionectomy. (© 2014
Intuitive Surgical. Inc.)
00
• •
o Camera Assi~ant
AssIstant
The patient is placed in mild reverse Trendelenburg position, vvith semi-right lateral
decubitus and with spread legs. Patient's anus are either tucked along the body to
avoid collision with robotic anus. The assistant stands in between the patient's legs.
The robot is docked directly over the head of the patient with two operating
amlS on the patient's right side.
A total of 5-6 ports are placed, replacing the specular shape of the trocar dis-
position of the right hepatectomy. Pneumoperitoneum is maintained at 12 mmHg
(Fig . 9 .4) .
Surgery starts with visual exploration of the abdomen, followed by contact
ultrasonography of the liver. As soon as resectability is confirmed, the left lobe
90 A. Patriti et al.
The patient is placed in mild reverse Trendelenburg position, with right lateral
decubitus and with spread legs . Patient"s arms are either tucked along the body
to avoid collision \vith robotic anus. The assistant stands in between the
patient's legs. The robot is docked directly over the head of the patient. Trocar
disposition is the same as that used for left hepatectomy (Fig. 9.4) . We general-
ly encircle the liver pedicle for the Pringle maneuver in case of bleeding . A 2-0
stay suture is passed at the inferior border of segment 3 to retract the left later-
al section to the left. The Rex fossa is therefore exposed. The fa1cifonnligament
is transected sparing the round ligament. After opening of the bare area of the
hepatocaval confluence, the anterior and posterior layer of the left triangular lig-
anlent are sectioned with monopolar hook or scissors . Retracting the left lateral
section to the left allows the pedicle of segment 3 (P3) to be identified.
Sometimes a small bridge of parenchyma has to be separated before reaching the
pedicle (the so called hepatic bridge) . Liver transection is carried out with bipo-
lar forceps in the left hand and a mono polar scissor in the right hand.
Transection starts 5 mm to the left of the falcifonu ligament. The first portion of
liver parenchyma is transected to allow a complete exposition of P3. The small
arteries to S3 are identified and ligated with titanium clips or coagulated with
bipolar forceps. P3 is finally encircled using the bipolar forceps and secured
vvith Hem-o-lock® clips. After its transection parenchymal division is continued
9 Hepatic Resections 91
till reaching the portal pedicle of segment 2 (P2), which is clipped and divided
in the same way. The LHV is finally secured and divided "'ith a single fire of an
endovascular stapler. The specimen can be extracted through the umbilicus or a
Pfannenstiel incision. Two closed suction Jackson-Pratt drains are placed near
the transection surface of the liver.
(j)
I <D.
The patient is tilted on the left flank and the
right robotic trocar in inserted between the
10th and 11th rib. (© 2014 Inruitive Surgical .
• (j) O Inc .)
I •
Camera (
0
0
AssIstant AssI<tant
,,>S-- SlI{
is;
/
~ 90
P ~. ~
r-
I
92 A. Patriti et al.
To date, only a few reports have focused on robotics in liver surgery, the num-
ber of patients included is rather small and the majority of revie,,"s are carried
out on a retrospective basis [3, 5, 7]. Therefore there is still not good evidence
9 Hepatic Resections 93
References
l. AJdrighetti L, Cipriani F, Ratti F et al (1013) The Italian experience in minimally invasive
surgery of the liver: A national survey. In: Calise F, Casciola L (eds) - Minimally invasive
surgery of the liver. Springer, Milan
Buell IF, Cherqui D, Geller DA et al (1009) The international position on laparoscopic liver
surgery: The Louisville Statement, 1008. Annals of Surgery 150:813-830
3. Casciola L, Patriti A , Ceccarelli G et al (10 II) Robot-assisted parenchymal-sparing liver sur-
gery including lesions located in the posterosuperior segments . Surg Endosc 25:3815-3814
94 A. Patriti et al.
4. Giulianotti PC, Coratti A , Sbrana P et al (2010) Robotic liver surgery: Results for 70 resec-
tions. Surgery 149:29-39
5. Troisi RI, Patriti A, Montalti R , Casciola L (2013) Robot assistance in liver surgery: a real
advantage over a fully laparoscopic approach? Results of a comparative bi-institutional analy-
sis. Int J Med Robot 9:160-166
6. Patriti A, Ceccarelli G , Bartoli A , Casciola L (2011) Extracorporeal Pringle maneuver in r0-
bot-assisted liver surgery. Surg Laparosc Endosc Percutan Tech 21 :e242-244
7. Tsung A , Geller DA , Sukato DC et al (2014) Robotic versus laparoscopic hepatectomy. An-
nals of Surgery 259 :549-555
8. Giulianotti PC , Sbrana P, Bianco PM, Addeo P (2010) Robot-assisted laparoscopic extended
right hepatectomy with biliary reconstruction. J Laparoendosc Adv Surg Tech A 20: 159-163
Biliary Tract Tumors
(Resection and Reconstruction)
10
Pier Cristoforo Giulianotti, Vivek Bindal, and Despoina Daskalaki
10.1 Classification
Bile duct tumors can be broadly classified into benign, premalignant and malig-
nant tumors as per the histologic findings, and into intrahepatic and extrahepat-
ic tumors based on their location [1]. Most of these tumors are malignant. Bile
duct adenoma is the most conmlon benign neoplasm of the bile ducts, though it
is rare. Premalignant bile duct tumors include biliary intraepithelial neoplasia
(BilIN) and intraductal papillary neoplasm of the bile duct (lPN-B).
Cholangiocarcinoma (CC) is the commonest primary malignancy of the bile
ducts [2]. It is predominantly adenocarcinoma (95% of cases), although other
histologic types including squamous cell carcinoma , small cell carcinoma and
sarcomas have been described.
CC can be classified as intrahepatic or peripheral, periliilar and distal
depending on their location. The distal CC, if resectable, is treated by pancre-
aticoduodenectomy (described in next chapter). Periliilar CC, also called
Klatskin tumor, accounts for 60% cases of extrahepatic CC [3]. It is a challeng-
ing situation, as it may require a formal hepatectomy with lymphadenectomy
and reconstruction. We shall elaborate on these details in the sections below.
RO resection of the tumor offers the only possibility for long term survival and
has been regarded as the gold standard for the treatment of resectable disease.
P. C. Giulianoni ( :81)
Division of General, Minimally Invasive and Robotic Surgery,
University of lllinois, Chicago, IL, USA
e-mail: [email protected]
For Klatskin tumor, which forms the majority of cases, this may include extend-
ed hepatectomy, combined with complete extrahepatic bile duct resection and
radical lymphadenectomy [4]. This aggressive surgical strategy has increased
the rate of curative resection and long term survival for the patients harboring
this disease [5]. There are many challenging technical complexities in this
approach like inflow control, inclusion of caudate lobe and reconstruction with
separate small caliber ducts. The feasibility and safety of laparoscopic tech-
niques in liver resections have been reported, especially for anterior and lateral
segments. However, the adoption of laparoscopy for biliary tumors has been
restricted due to technical limitations and oncologic concerns [6].
Robotic surgery may help overcome certain limitations of laparoscopy and
provide the minimally invasive advantage to these patients, who otherwise usu-
ally undergo extensive open surgery. The main challenges faced during these
procedures are right lobe mobilization, hepatic hilum dissection, control of
bleeding during parenchymal transection, and complex biliary reconstruction.
The robotic platform provides many advantages which have been already well
described. This helps the surgeon increase the precision of dissection while
facilitating suturing in difficult situations. It also provides him with the
ergonomic comfort in these long and challenging procedures so as to allow him
to work to his full potential.
In all these advanced procedures, it is important to understand that with the
use of a minimally invasive method, indications and patient selection for a pro-
cedure does not change. The fundamental principles of a safe R0 surgical resec-
tion and adequate lymphadenectomy should not be compromised. A low thresh-
old for conversion to an open procedure should be kept if the goals of the oper-
ation cannot be accomplished safely by using minimally invasive technique.
One has to assess these patients for their performance status and fitness for
major surgery that may include a partial hepatectomy. Chronic liver disease or
portal hypertension generally makes these patients bad candidates for surgery. If
there is any evidence of cholangitis, it should be treated with adequate drainage
and antibiotics before surgery [2].
The main goal of the surgery is R0 resection with free proximal and distal
margins, resection of tumor bed (including caudate lobe and vascular elements)
and adequate lymphadenectomy. A complete resection with histologically nega-
tive resection margins is a very important criterion for long term survival in
cholangiocarcinoma.
The criteria for unresectable disease are [7]:
• major comorbidities precluding safe surgery;
• metastatic disease;
10 Biliary Tract Tumors (Resection and Reconstruction) 97
Under general anaesthesia, the patient is placed in the supine position with part-
ed legs with approximately a 20° reverse-Trendelenburg tilt. The abdomen is
cleaned and draped and an orogastric tube and urinary catheter are inserted. The
assistant stands in between the legs. Pneumoperitoneum is achieved to 15mmHg
using a Veress needle at Palmer’s point. A 10/12mm trocar is placed in supraum-
bilical position (which is used as an assistant port in the operation). One optical
and three da Vinci® trocars are placed as follows (Fig. 10.1a):
• optical (12mm): in the right midclavicular line, approximately 10cm from
the assistant trocar, above the level of the umbilicus;
• R1: in the left midclavicular line approximately 10cm away from the assis-
tant trocar, above the level of umbilicus;
• R2: in right anterior axillary line, at least at 10cm from the optical trocar;
• R3: in left anterior axillary line, used for retraction purpose.
The port placement needs to be adjusted based on the body habitus of the
patient so as to prevent external arm collision and provide optimal exposure.
A diagnostic laparoscopy is done to look for any metastatic deposits or free
fluid. If a suspicious deposit is seen, it is biopsied and sent for frozen examina-
tion to rule out metastatic disease. Any free fluid is sent for cytology to look for
malignant cells. Intraoperative ultrasonography is performed to rule out any
undetected metastatic deposits in the liver.
The da Vinci® patient cart is brought from the head of the patient, and the
arms are docked to the placed ports. The third arm of the robot comes from the
left side of the patient. To start the procedure, a monopolar hook is taken in R1,
bipolar forceps in R2 and grasping forceps in R3. The assistant surgeon stands
in between the legs for complementary maneuvers (i.e., suction, stapling, retrac-
tion, and laparoscopic ultrasonography). A sample OR setup for right extended
hepatectomy for CC is depicted in Fig. 10.1b.
98 P. C. Giulianotti et al.
a b
Fig. 10.1 Robotic-assisted right extended hepatectomy. a Port position. SUL, spino-umbilical li-
ne; MCL, midclavicular line; 1, 2, 3 represent positions for robotic arm 1, 2 and 3 respectively. b
OR setup. (© 2014 Intuitive Surgical, Inc.)
Hepatic flexure of the colon is mobilized medially and caudally and a partial
Kocher maneuver is performed. Dissection is commenced at the hepatic hilum
while it is retracted using R3. If the gallbladder is in place, it is taken down
keeping the cystic duct attached to the common bile duct (CBD). Indocyanine
green fluorescence aids in the detection of CBD and any aberrant biliary anato-
my. The CBD is dissected and transected distally at the superior border of pan-
creas, and the distal stump is oversewn. The distal margin is sent for frozen sec-
tion, which if positive, calls for a pancreaticoduodenectomy with or without
liver resection, as per disease location, spread and performance status of the
patient. Now, the proximal bile duct is lifted up and dissection is continued to
separate the hepatic artery and portal vein from the bile duct. Starting at the
superior border of the pancreatic head, a lymphadenectomy along the common
hepatic artery is performed. The origin of gastroduodenal artery is exposed.
Generally, lymph node dissection in hepatoduodenal ligament is adequate. The
fat and lymph nodes are resected en bloc with the bile duct. At the hilum, right
and left hepatic ducts are dissected and encircled with vessel loops to aid in trac-
tion. If one can get proximal to the tumor, bile ducts are divided and margins are
sent for frozen section to confirm R0 resection.
In case of Bismuth type III tumors, an extended right or left hepatectomy along
with bile duct resection may be required for R0 resection. The detailed technique
of these procedures is described below in separate sections.
Vascular resection and reconstruction may be required in case of portal
venous or hepatic arterial involvement. This may be done only by experienced
10 Biliary Tract Tumors (Resection and Reconstruction) 99
surgeons and institutions with good perioperative results in such high risk pro-
cedures.
Fig. 10.2 Interrupted stitches to perform the anterior layer of hepaticojejunostomy. The interrup-
ted sutures are held in position by clips. Use of robotic platform provides the distinct advantage of
ability to perform microsurgical interrupted stitches in a minimally invasive environment. HD, he-
patic duct
100 P. C. Giulianotti et al.
two stitches of Prolene 3-0 (Ethicon, Somerville, NJ, USA) are placed to fix the
jejunal loop to the hilum.
Use of fibrin glue may be considered at the end of the anastomosis. At the
end of the procedure, a drain may be left near the biliary anastomosis.
Fig. 10.3 Left branch of portal vein (PV) being ligated. The left hepatic artery (LHA) has been di-
vided and lifted up. One can see the gastroduodenal artery (GDA) and right hepatic artery (RHA)
10 Biliary Tract Tumors (Resection and Reconstruction) 101
terior branches of right hepatic duct are dissected, transected and margins are
sent for frozen section. The caudate lobe is dissected from the inferior vena
cava. The parenchymal transection is completed and hemostasis confirmed.
The reconstruction is done in a Roux-en-Y fashion. The Roux limb is created
by stapled jejunojejunostomy (as described in Sect. 10.3.3) and it is brought
cranially in a transmesocolic (sometimes retrogastric) fashion. Using 5-0 PDS,
the two right ducts are connected to each other. An opening made in the Roux
limb and a posterior layer of the anastomosis is performed using continuous
suturing. An interrupted anterior layer is thrown and a few stitches are placed
between liver and Roux limb so as to avoid undue tension on the anastomosis.
Fibrin glue is used over the anastomosis and the raw surface of transected liver
parenchyma. The specimen is retrieved in an Endobag usually through a
Pfannensteil incision.
Initially, the hepatic flexure of the colon is mobilized, and a partial Kocher
maneuver is performed. Starting at the superior border of the pancreatic head,
a lymphadenectomy of the common hepatic artery is performed, using a
monopolar hook and bipolar forceps, to expose the origin of the gastroduode-
nal artery. The inferior aspect of segment IV is retracted upward by using the
third robotic arm, and the CBD is dissected and sectioned at the superior bor-
der of the pancreatic head. The distal stump of the CBD is sutured, and a
frozen section is sent to rule out neoplastic invasion. The right hepatic artery
is dissected and divided at its origin from the proper hepatic artery. The left
hepatic duct is transected at the left umbilical fissure and a frozen section is
sent at this level as well. Following this, the right portal vein is dissected and
divided between ligatures. The right liver lobe is mobilized from its peritoneal
attachments. This is done by sectioning the falciform ligament and the anteri-
or half of the coronary ligament, until the anterior side of the inferior vena
cava (IVC) and the right hepatic vein is reached. The hepatorenal ligament and
the right triangular ligament are divided by using a monopolar hook. The third
arm is used to retract the inferior aspect of the right liver lobe upward. In this
way, the right side of the IVC is exposed. The accessory hepatic vein is suture
ligated and the dissection proceeds until the inferior aspect of the right hepat-
ic vein is reached. After sectioning the bridge of parenchyma between seg-
ments IV and III, the parenchymal transection is carried out along the right
aspect of the falciform ligament, by harmonic scalpel, starting at the anterior
border of the liver. The recurrent vessels from the umbilical fissure to segment
IV, middle hepatic and right hepatic veins are divided using staplers. The
reconstruction is done as per the already described method. The specimen is
retrieved in an Endobag.
102 P. C. Giulianotti et al.
lack of perineural invasion [13]. In general, the best outcomes are in patients
who undergo R0 resection and this is the best predictor for five-year survival.
References
1. Joo I, Lee JM (2013) Imaging bile duct tumors: pathologic concepts, classification, and ear-
ly tumor detection. Abdominal Imaging 38:1334–1350
2. Jarnagin W, Winston C (2005) Hilar cholangiocarcinoma: diagnosis and staging. HPB
7:244–251
3. Lazaridis KN, Gores GJ (2005) Cholangiocarcinoma. Gastroenterology 128:1655
4. Giulianotti PC, Sbrana F, Fransesco BM, Addeo P (2010) Robot-assisted laparoscopic extend-
ed right hepatectomy with biliary reconstruction. J Laparoendosc Adv Surg Tech 20:159–163
5. Ito F, Agni R, Rettammell RJ et al (2008) Resection of hilar cholangiocarcinoma: Concomi-
tant liver resection decreases hepatic recurrence. Ann Surg 248:273–279
6. Simillis C, Constantinides VA, Tekkis PP et al (2007) Laparoscopic versus open hepatic re-
sections for benign and malignant neoplasms—a meta analysis. Surgery 141:203–211
7. Whang EE, Duxbury M, Rocha FG; Zinner MJ (2013). Cancer of the gall bladder and bile
ducts. In: Maingot R, Zinner M, Ashley SW, (eds.) Maingot’s abdominal operations, 12th edn.
McGraw-Hill Medical, New York
8. Date RS, Siriwardena AK (2005) Current status of laparoscopic biliary bypass in the man-
agement of non-resectable peri-ampullary cancer. Pancreatology 5:325–329
9. Smith AC, Dowsett JF, Russell RC et al (1994) Randomised trial of endoscopic stenting ver-
sus surgical bypass in malignant low bile duct obstruction. Lancet 344:1655–1660
10. Lesurtel M, Dehni N, Tiret E et al (2006) Palliative surgery for unresectable pancreatic and
periampullary cancer: a reappraisal. J Gastrointest Surg 10:286–291
11. Buchs NC, Addeo P, Bianco FM et al (2011) Robotic palliation for unresectable pancreatic
cancer and distal cholangiocarcinoma. Int J Med Robotics Comput Assist Surg 7:60–765
12. Giulianotti PC, Tzvetanov I, Jeon H et al (2009) Robot assisted right lobe donor hepatecto-
my. Transpl Int 25:e5–e9
13. Santibanes ED, Ardiles V (2012) High malignant biliary tract obstruction. In: Fischer JE, (ed.)
Fischer’s Mastery of Surgery. 6th ed. Lippincott Williams & Wilkins
Pancreatic Surgery for Cancer
11
Pier Cristoforo Giulianotti, Despoina Daskalaki,
and Francesco Mario Bianco
11.1 Introduction
Pancreatic cancer is the fourth leading cause of cancer-related death in the
United States, with more than 45,000 estimated new cases per year [1]. The only
potentially curative treatment is surgery, but unfortunately only 10–20% of
patients have a resectable disease at the time of diagnosis [2]. The most impor-
tant factors that can improve the oncologic outcome after surgery are an ade-
quate lymphadenectomy and a negative resection margin [3]. Even though the
overall morbidity and mortality rates after pancreatic surgery have been pro-
gressively decreasing over the past twenty years, pancreas-related complica-
tions are still a major concern.
The last major innovation in pancreatic surgery has been the introduction of
minimally invasive (MI) techniques [4–6]. Even though the rates of pancreatic
fistula and delayed gastric emptying have not been proven to be lower with the
MI approach, several other postoperative outcomes are in favor of these tech-
niques: less blood loss and transfusion, less wound infections, shorter hospital
stay, faster recovery and faster start of adjuvant treatment are all indisputable
advantages of MI surgery [7, 8].
The laparoscopic approach to the pancreas has proved to be feasible and safe
[9, 10]. However, while laparoscopic distal pancreatectomy has been widely
adopted even for malignant disease, laparoscopic pancreaticoduodenectomy
still remains relatively uncommon [11]. This is most likely due to the technical
complexity of this procedure and the steep learning curve.
The introduction of robotic technology more than a decade ago, has slowly
P. C. Giulianotti ()
Division of General, Minimally Invasive and Robotic Surgery,
University of Illinois, Chicago, IL, USA
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 105
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_11, © Springer-Verlag Italia 2015
106 P.C. Giulianotti et al.
The pancreas is now exposed and the pancreatic neck is identified. The dis-
section continues along the inferior border of the pancreas in order to expose the
SMV. The retropancreatic tunnel can be created at this point, paying attention
not to disrupt any vascular branches.
Usually at this point we perform an intraoperative ultrasound in order to
localize the mass and assess its relationship to the vessels.
The duodenum (or the stomach, depending on the type of pancreaticoduo-
denectomy) is now transected, using an Endostapler. The jejunum is also divid-
ed using a stapler device, distal to the Treitz ligament. At this point, the jejunum
is retracted to the right, rotating the specimen. This maneuver facilitates expo-
sure of the SMV and artery.
The pancreas is now completely dissected. Prolene® sutures are placed at the
superior and inferior border of the pancreas as a way of anchoring and retract-
ing it, prior to division of the gland. The parenchyma is now divided using the
robotic ultrasonic shears (Intuitive Surgical, Sunnyvale, CA) being careful not
to occlude the duct. A plastic stent is placed in the main pancreatic duct and
secured with a stitch. The proximal margin of the pancreatic remnant is sent to
pathology for frozen-section analysis.
The next step is the uncinate process dissection. The pancreas is lifted
upward and rotated, facilitating the dissection from the SMV; the fourth robotic
arm facilitates the retraction. The dissection begins distally, following the SMV
upward (Fig. 11.3). The small jejunal branches are transected either with the
robotic ultrasonic shears or between sutures. The value of robotic technology is
particularly evident during this step of the procedure. The stability of the plat-
form and the magnified vision, together with the EndoWrist® instruments, allow
a precise dissection with minimal blood loss. Moreover, even in the case of
bleeding, suturing can be performed like in open surgery, with less technical dif-
ficulty compared to laparoscopy.
After confirmation of the negative resection margins, the reconstructive
phase of the operation can begin. In the case of a pylorus-preserving procedure,
we prefer performing a dunking transgastric pancreaticogastrostomy (PG) (Fig.
11.4). The PG is performed by making an anterior gastrotomy and then pulling
the pancreatic stump into the gastric cavity through a posterior gastrotomy. Once
the pancreas is well mobilized (for at least 2 inches) inside the stomach, the
sutures are placed and the anterior gastrotomy is closed with a running PDS
suture. Attention must be made to the accurate hemostasis of the gastric mucosa,
in order to avoid postoperative bleeding. The hepaticojejunostomy (HJ) and
duodenojejunostomy are performed with a single loop of jejunum, with running
PDS suture, or interrupted stitches in the case of a HJ with a small/normal CBD.
In the case of the classical Whipple, without preservation of the pylorus, the
jejunal loop is used for the pancreatic reconstruction. We usually prefer placing
two drains, one near the PG and one near the HJ.
11 Pancreatic Surgery for Cancer 109
Fig. 11.3 Uncinate process dissection. The dissection follows the superior mesenteric vein cau-
dal to cephalad. SMV, superior mesenteric vein; SMA, superior mesenteric artery; RPL, retropor-
tal lamina; PD, pancreatic duct
Fig. 11.4 Dunking transgastric pancreaticogastrostomy during the reconstructive phase of pancre-
aticoduodenectomy. An anterior gastrotomy is performed and the pancreatic stump is then pulled
into the gastric cavity through a posterior gastrotomy. PG, posterior gastrotomy; AG, anterior gas-
trotomy; PS, pancreatic stump
110 P.C. Giulianotti et al.
11.3 Innovations
A robotic EndoWrist® stapler is available in the market; this allows the surgeon
sitting at the console to control the device directly.
Integrated imaging is one of the most important innovations that will soon be
available. The robotic console could allow integration of preoperative imaging
that could be superimposed onto the intraoperative view and may even allow vir-
tual-reality options. This would be an important step toward better training of
robotic surgeons and patient safety.
encouraging and even though larger studies are needed, they open the way for
wider acceptance of robotic technology in the pancreatic field. It is important to
note that robotic pancreatic surgery should be performed by experienced robot-
ic, hepatopancreatobiliary surgeons. The initial learning curve of a robotic sur-
geon, should include simpler procedures and then be followed by more complex
operations, such as pancreatic resections.
Our experience with robotic pancreatic surgery started in 2000. Since then
115 cases of PD and 77 cases of DP, for different indications, have been per-
formed by a single surgeon, in two different centers (Grosseto, Italy and
Chicago, USA). The results of the series were presented in an oral communica-
tion at the 2013 Clinical Robotic Surgery Association Meeting (Washington
DC). Out of 114 PDs, 66 were standard Whipple procedures and 48 were
pylorus-preserving; the mean operative time was 443.5min (including the dock-
ing time), with a mean estimated blood loss of 343mL; the conversion-to-open
rate was 13% and the postoperative morbidity (Clavien III/IV) rate was 24.7%;
the mortality rate was 2.8%; the surgical indication was malignant disease in
70% of cases; a negative resection margin (R0) was achieved in more than 90%
of cases, with a mean of 19.4 lymph nodes retrieved.
Of the 77 DPs, 41 included splenectomy and 36 were spleen-preserving
(these last ones performed for benign disease); the intention-to-treat spleen-
preservation rate was 97%; the mean operative time was 236min (including the
docking time), with a median estimated blood loss of 100mL; the rate of clini-
cally significant postoperative pancreatic fistula, grade B according to the
ISGPF definition [13], was 12%; no grade C fistulas occurred in our series; the
conversion-to-open rate was 5% and the mortality was nil; surgery was per-
formed for malignant disease in 36% of the patients; the mean number of lymph
nodes harvested was 10.3, with a 90% negative resection margin.
We should mention at this point that opting for a robotic approach in pancre-
atic surgery also has some limitations, especially for the less experienced sur-
geons that have yet to learn how to troubleshoot the potential problems that
could arise by using the robotic system. The lack of tactile feedback represents
a drawback, especially at the beginning of the learning curve. With experience
though, the surgeon learns how to compensate by taking advantage of the excel-
lent vision that the robotic system provides. Another limitation is the difficulty
of performing multiquadrant surgery and the inability of changing the patient’s
position, once the robotic system is docked. This can be overcome by careful
preoperative planning of the robot setting and a correct placement of the robot-
ic trocars, in a way that collisions will be avoided. Also, all the advanced robot-
ic procedures require a trained team of assistant surgeon and scrub nurse. As the
main surgeon is separated from the patient while performing robotic surgery, the
assistant surgeon has to be trained enough to help in performing difficult tasks
and also to take care of any emergency situation that could arise during the pro-
cedure. The role of a trained scrub nurse and operating room technician is also
very important in streamlining the conduct of the procedure and preventing any
11 Pancreatic Surgery for Cancer 113
wastage of time and resources. Thus, team work is the key to success in
advanced robotic surgery.
References
1. Siegel R, Naishadham D, Jemal A (2013) Cancer statistics, 2013. CA: Cancer J Clin 63:11–30
2. Bilimoria KY, Bentrem DJ, Ko CY et al (2007) Validation of the 6th edition AJCC Pancreat-
ic Cancer Staging System: report from the National Cancer Database. Cancer 110:738–744
3. Konstantinidis IT, Warshaw AL, Allen JN et al (2013) Pancreatic ductal adenocarcinoma: is
there a survival difference for R1 resections versus locally advanced unresectable tumors? What
is a “true” R0 resection? Annals of Surgery 257:731–736
4. Venkat R, Edil BH, Schulick RD et al (2012) Laparoscopic distal pancreatectomy is associ-
ated with significantly less overall morbidity compared to the open technique: a systematic
review and meta-analysis. Annals of Surgery 255:1048–1059
5. Giulianotti PC, Sbrana F, Bianco FM et al (2010) Robot-assisted laparoscopic pancreatic sur-
gery: single-surgeon experience. Surgical Endoscopy 24:1646–1657
6. Milone L, Daskalaki D, Wang X, Giulianotti PC (2013) State of the art of robotic pancreatic
surgery. World J Surg 37:2761–2770
7. Kendrick ML (2012) Laparoscopic and robotic resection for pancreatic cancer. Cancer J
18:571–576
8. Kooby DA (2006) Laparoscopic surgery for cancer: historical, theoretical, and technical con-
siderations. Oncology (Williston Park) 20:917–927
9. Jayaraman S, Gonen M, Brennan MF et al (2010) Laparoscopic distal pancreatectomy: evo-
lution of a technique at a single institution. J Am Coll Surg 211:503–509
10. Kim SC, Song KB, Jung YS et al (2013) Short-term clinical outcomes for 100 consecutive
cases of laparoscopic pylorus-preserving pancreatoduodenectomy: improvement with surgi-
cal experience. Surgical Endoscopy 27:95–103
11. Subar D, Gobardhan PD, Gayet B (2014) Laparoscopic pancreatic surgery: An overview of
the literature and experiences of a single center. Best Pract Res Clin Gastroenterol 28:123–132
12. Zhang J, Wu WM, You L, Zhao YP (2013) Robotic versus open pancreatectomy: a system-
atic review and meta-analysis. Ann Surg Oncol 20:1774–1780
13. Bassi C, Dervenis C, Butturini G et al (2005) Postoperative pancreatic fistula: an internation-
al study group (ISGPF) definition. Surgery 138:8–13
Part IV
Lower Gastrointestinal Surgery
Right Colectomy for Cancer:
Three-arm Technique 12
Domenico Garcea, Francesca Bazzocchi, and Andrea Avanzolini
F. Bazzocchi ()
General, Gastrointestinal, and Minimally Invasive Surgery Unit,
“G.B. Morgagni - L. Pierantoni” Hospital,
Forlì, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 117
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_12, © Springer-Verlag Italia 2015
118 D. Garcea et al.
axis of the superior mesenteric vein: it represents the target and horizon of the
operative field.
When the surgeon is sure about the feasibility of the procedure the robotic
cart is docked. A medial to lateral approach is used. Sometimes in obese patients
it is necessary to use a lateral to medial approach; an inferior to superior
approach has also been described [3]. In order to avoid intraoperative complica-
tions, proper tractions and countertractions should be exerted by robotic and
laparoscopic forceps to create an adequate exposure: the surgeon can better visu-
alize the vascular anatomy and its variations, can avoid ureteral injuries and can
execute a safe dissection along the correct planes without unnecessary bowel
manipulation. The technique is described step-by-step.
The gastrocolic ligament is put in tension, divided and then dissected up to the
right colic flexure so that the lesser sac and the duodenum are exposed. For neo-
plasms of the right colic flexure or proximal transverse colon, it is necessary to
extend the dissection of the gastrocolic ligament toward the distal transverse
colon, in order to facilitate the execution of the intracorporeal ileocolic anasto-
mosis in the right side of the abdomen without changing the robotic docking.
Moreover, in these neoplasms, it is mandatory to carry out the dissection of the
gastrocolic ligament along the greater curve of the stomach rather than along the
margin of the transverse colon, performing in this way the lymphadenectomy of
the right gastroepiploic vessels.
The transverse mesocolon is divided from its root to the colonic wall for final
transection. The ileal mesentery is divided approximately 10–20 cm from the
ileocecal valve. Marginal arteries will need to be controlled with clips or
cautery. A linear articulated stapler with tri-staple technology is used by the
assistant at the operating table for the transection of the transverse colon and
ileum.
the mechanical suture and the bowel defect is closed with a two-layer running
suture with an absorbable self-locking thread. The mesenteric defect is then
closed with an absorbable suture.
The specimen extraction is carried out through a paraumbilical incision or a
sovrapubic minilaparotomy with plastic wound protection.
system in colectomies, we believe that the da Vinci® System may improve accu-
racy during lymphadenectomy and intracorporeal anastomosis. The high costs of
robotic surgery represent the main limitation to the widespread diffusion of this
technology. In order to reduce costs, we have chosen to use only three robotic
arms and no more than three robotic instruments (monopolar hook, Maryland
bipolar forceps, needle driver). Therefore, the approximate additional cost for a
right colectomy is 2250 USD.
In conclusion, the robotic system improves the performance of the surgeon
and allows to reproduce open procedures faithfully. Further studies comparing
clinical and cost-effectiveness outcomes with laparoscopy are required to better
assess the role of this technology in colorectal surgery.
References
1. Mirnezami AH, Mirnezami R, Venkatasubramanian AK et al (2010) Robotic colorectal sur-
gery: hype or new hope? A systematic review of robotics in colorectal surgery. Colorectal Dis
12:1084–1093
2. Antoniou SA, Antoniou GA, Koch OO et al (2012) Robot-assisted laparoscopic surgery of
the colon and rectum. Surg Endosc 26:1–11
3. Park SY, Choi GS, Park JS et al (2012) Robot-assisted right colectomy with lymphadenecto-
my and intracorporeal anastomosis for colon cancer: technical considerations. Surg Laparosc
Endosc Percutan Tech 22:271–276
4. de Souza AL, Prasad LM, Park JJ et al (2010) Robotic assistance in right hemicolectomy: is
there a role? Dis Colon Rectum 53:1000–1006
5. D’Annibale A, Pernazza G, Morpurgo E et al (2010) Robotic right colon resection: evalua-
tion of first 50 consecutive cases for malignant disease. Ann Surg Oncol 17:2856–2862
6. Rawlings AL, Woodland JH, Vegunta RK et al (2007) Robotic versus laparoscopic colecto-
my. Surg Endosc 21:1701–1708
7. Fung AKY, Aly EH (2013) Robotic colonic surgery: is it advisable to commence a new learn-
ing curve? Dis Colon Rectum 56:786–796
8. Spinoglio G, Summa M, Priora F et al (2008) Robotic colorectal surgery: first 50 cases ex-
perience. Dis Colon Rectum 51:1627–1632
9. Zimmern A, Prasad L, Desouza A et al (2010) Robotic colon and rectal surgery: a series of
131 cases. World J Surg 34:1954–1958
Right Colectomy with Complete Mesocolic
Excision: Four-arm Technique 13
Giuseppe Spinoglio, Alessandra Marano, Fabio Priora,
Ferruccio Ravazzoni, and Giampaolo Formisano
13.1 Introduction
Minimally invasive surgery is gaining worldwide acceptance in the treatment of
colonic cancer and the advantages over the traditional open approach are well
known [1–3]. Unfortunately, during recent decades, the outcomes of patients
after colon cancer resection have not improved to the same degree as for rectal
cancer, whose treatment with total mesorectal excision (TME) is universally
accepted as the standard of care. The complete mesocolic excision (CME), first
reported by Hohenberger and colleagues in 2008 [4], seems to produce better
long-term outcomes when compared to standard lymphadenectomy by follow-
ing the same embryological-based principles introduced by Heald for rectal can-
cer more than 20 years ago [5]. However, well-conducted randomized studies
are needed to confirm its efficacy.
Additionally, laparoscopic right colectomy with intracorporeal anastomosis
is still considered by some surgeons as one of the most difficult procedures to
perform [6] and radical lymphadenectomy with CME is technically challenging
when performed with conventional laparoscopic instruments.
To date and to the best of our knowledge, few reports analyzing the safety and
feasibility of laparoscopic CME for right-sided colonic malignancies exist in the
literature [7–14]. No studies have focused selectively on the robotic approach.
This chapter describes the technical approach to robotic CME in right colec-
tomy, including technical tips culminating from the author’s nearly 10-year
experience in robotic colonic surgery.
G. Spinoglio ()
Department of General and Oncologic Surgery,
“Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 125
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_13, © Springer-Verlag Italia 2015
126 G. Spinoglio et al.
a b
Fig. 13.1 Robotic right colectomy. a Trocar layout. b Overhead view of OR setup. (© 2014 Intui-
tive Surgical, Inc.)
The first important step is to achieve exposure of the right and transverse meso-
colon. R2 and R3 trocars are used to keep the superior mesenteric axis in trac-
tion and the monopolar cautery hook/scissors in R1 are used for dissection.
128 G. Spinoglio et al.
Fig. 13.2 Final view of the operative field after complete mesocolic excision. MCA, middle colic
artery; SMV, superior mesenteric vein; MCV, middle colic vein; RGEV, right gastroepiploic vein;
ARCV, accessory right colic vein; AIPDV, anterior inferior pancreaticoduodenal vein
13 Right Colectomy with Complete Mesocolic Excision: Four-arm Technique 129
The transverse colon and the ileum are approximated to choose the correct
enterotomy sites, but no stay sutures are placed since they are not useful, in our
opinion. Monopolar curved scissors in R1 are used to create enterotomies on the
antimesenteric border of the ileum and the free taenia of the transverse colon.
The monopolar device in R1 is then replaced with a needle driver. A 60-mm lin-
ear stapler is introduced through the assistant port to perform an isoperistaltic
anastomosis and the enterotomies are subsequently closed with a robotically
hand-sewn double-layer running suture (using absorbable monofilament barbed
knotless sutures; V-Loc™, Covidien). The mesenteric defect is also closed with
a continuous suture to prevent internal hernias.
The specimen is then extracted into a plastic bag through a mini-Pfannenstiel
incision performed at the suprapubic port site, avoiding any squeezing. The
advantages of intracorporeal anastomosis are minimal colonic mobilization, lim-
ited chance for bowel and anastomotic twisting, and also the possibility to
choose the specimen extraction site (according to the patient’s history of prior
abdominal surgery).
Once the specimen is removed, the mini-laparotomy incision is closed and the
pneumoperitoneum re-established for a final check of the operative field. No
drain is routinely left in place. The trocars are removed under direct vision and
all the sites greater than 8mm in diameter are closed with absorbable sutures at
the fascial level.
130 G. Spinoglio et al.
Table 13.1 Perioperative outcomes and pathological characteristics of robotic vs. laparoscopic
right colectomy for cancer with complete mesocolic excision
RRC LRC p value
(101 pts) (101 pts)
Operating room time (min)
Mean (SD) 279 (80) 236 (68) p<0.001†
Range 135–540 95–465
Conversion rate
n (%) 0 (0) 7 (6.9) p=0.014§
Length of stay
Mean (SD) 7.9 (5.2) 7.9 (3.5) p=0.948†
Range 4–37 4–19
Time to return of bowel function (days)
Mean (SD) 1.9 (1) 1.8 (0.8) p=0.563†
Range 1–7 1–4
Oral re-intake (days)
Mean (SD) 1.3 (1.1) 1.1 (0.5) p=0.150†
Range 1–9 1–4
Harvested lymph nodes (n)
Mean (SD) 28.2 (10.6) 30.4 (13.1) p=0.188†
Range 13–66 12–74
Specimen length (cm)
Mean (SD) 35.2 (9.9) 36.2 (10.8) p=0.489†
Range 21–70 20–88
AJCC staging (n) p=0.088‡
I 21 26
II 38 28
III 37 33
IV 5 14
Values are expressed as mean (SD=standard deviation) or n (%). RRC, robotic right colectomy;
LRC, laparoscopic right colectomy; AJCC, American Joint Committee on Cancer; †Student’s t
test; ‡Pearson’s chi-squared test; §Fisher’s exact test
13 Right Colectomy with Complete Mesocolic Excision: Four-arm Technique 131
To date, however, no robotic series directly addresses the issue of CME for
right-sided colonic malignancies.
We recently conducted a retrospective analysis on prospectively collected
data of 101 patients who underwent robotic right colectomy with CME at our
institution between October 2005 and November 2013. The results of a consec-
utive contemporary series of 101 standard laparoscopic resections carried out
with the same operative technique have been retrospectively analyzed. The
groups were comparable in terms of baseline characteristics and we found only
two statistically significant differences: the robotic series showed a lower con-
version rate (0% vs. 6.9%; p=0.014) but longer operative times (279 min. vs.
236 min.; p <0.001) than the laparoscopic one (Table 13.1), thus confirming
results from previously published studies. Anastomotic leak rate (1%) and reop-
eration rate (2%) were equal in both groups, as well as minor complication rates.
The 3-year disease-free, cancer related and overall survival were 91.1%, 93.7%
and 91.2% in the robotic group, respectively, and 87.1%, 90.5% and 86% in the
laparoscopic series, respectively. Though encouraging and superior to the 3-year
results from previous randomized studies (from 67.8% to 84.2%) [1–3], at least
a 5-year follow-up is required (Table 13.2).
Robotic right colectomy with CME is feasible and safe. The inherent prop-
erties of the robotic system might eventually lessen the technical difficulties of
vascular control, extended lymphadenectomy and intracorporeal anastomosis
during right colonic resections, but longer operating room times are required.
Nevertheless, scientific evidence about the clinical advantages of both the
robotic approach and CME, if compared to laparoscopy and standard lym-
phadenectomy, is still insufficient at present to recommend their adoption in
routine practice.
132 G. Spinoglio et al.
References
1. Jayne DG, Thorpe HC, Copeland J et al (2010) Five-year follow-up of the Medical Research
Council CLASICC trial of laparoscopically assisted versus open surgery for colorectal can-
cer. Br J Surg 97:1638–1645
2. Fleshman J, Sargent DJ, Green E et al (2007) Laparoscopic colectomy for cancer is not infe-
rior to open surgery based on 5-year data from the COST Study Group trial. Ann Surg
246:655–662
3. Buunen M, Veldkamp R, Hop WC et al (2009) Survival after laparoscopic surgery versus open
surgery for colon cancer: long-term outcome of a randomised clinical trial. Lancet Oncol
10:44–52
4. Hohenberger W, Weber K, Matzel K et al (2009) Standardized surgery for colonic cancer: com-
plete mesocolic excision and central ligation—technical notes and outcome. Colorectal Dis
11:354–364
5. Heald RJ (1988) The ‘Holy Plane’ of rectal surgery. J R Soc Med 81:503–508
6. Jamali FR, Soweid AM, Dimassi H et al (2008) Evaluating the degree of difficulty of laparo-
scopic colorectal surgery. Archives Surg 143:762–767
7. Bae SU, Saklani AP, Lim DR et al (2014) Laparoscopic-assisted versus open complete meso-
colic excision and central vascular ligation for right-sided colon cancer. Ann Surg Oncol
21:2288–2294
8. Adamina M, Manwaring ML, Park KJ, Delaney CP (2012) Laparoscopic complete mesocol-
ic excision for right colon cancer. Surg Endosc 26:2976–2980
9. Feng B, Ling TL, Lu AG et al (2013) Completely medial versus hybrid medial approach for
laparoscopic complete mesocolic excision in right hemicolon cancer. Surg Endosc 28:477–483
10. Feng B, Sun J, Ling TL et al (2012) Laparoscopic complete mesocolic excision (CME) with
medial access for right-hemi colon cancer: feasibility and technical strategies. Surg Endosc
26:3669–3675
11. Gouvas N, Pechlivanides G, Zervakis N et al (2012) Complete mesocolic excision in colon
cancer surgery: a comparison between open and laparoscopic approach. Colorectal Dis
14:1357–1364
12. Shin JW, Amar AH, Kim SH et al (2014) Complete mesocolic excision with D3 lymph node
dissection in laparoscopic colectomy for stages II and III colon cancer: long-term oncologic
outcomes in 168 patients. Tech Coloproctol [Epub ahead of print]
13. Kang J, Kim IK, Kang SI et al (2014) Laparoscopic right hemicolectomy with complete meso-
colic excision. Surg Endosc 28:2747-2451
14. Storli KE, Sondenaa K, Furnes B, Eide GE (2013) Outcome after introduction of complete
mesocolic excision for colon cancer is similar for open and laparoscopic surgical treatments.
Digestive Surg 30:317–327
15. Park JS, Choi GS, Park SY et al (2012) Randomized clinical trial of robot-assisted versus stan-
dard laparoscopic right colectomy. Brit J Surg 99:1219–1226
16. Zimmern A, Prasad L, Desouza A (2010) Robotic colon and rectal surgery: a series of 131
cases. World J Surg 34:1954–1958
17. Trastulli S, Desiderio J, Farinacci F et al (2013) Robotic right colectomy for cancer with in-
tracorporeal anastomosis: short-term outcomes from a single institution. Int journal J Colorect
Dis 28:807–814
18. D’Annibale A, Pernazza G, Morpurgo E et al (2010) Robotic right colon resection: evalua-
tion of first 50 consecutive cases for malignant disease. Ann Surg Oncol 17:2856–2862
19. Shin JY (2012) Comparison of short-term surgical outcomes between a robotic colectomy and
a laparoscopic colectomy during early experience. J Korean Soc Coloproctol 28:19–26
20. de Souza AL, Prasad LM, Park JJ et al (2010) Robotic assistance in right hemicolectomy: is
there a role? Dis Colon Rectum 53:1000–1006
21. Casillas MA Jr, Leichtle SW, Wahl WL et al (2014) Improved perioperative and short-term
outcomes of robotic versus conventional laparoscopic colorectal operations. Am J Surg
208:33–40
Left Colectomy and Segmental
Resections for Cancer 14
Paolo Pietro Bianchi, Igor Monsellato, and Wanda Petz
14.1 Introduction
One of the goals of surgical oncology is to attempt to reduce the invasiveness
of surgery while maintaining or further improving the outcomes of traditional
open surgery. Therefore minimally invasive techniques have been applied to
oncologic surgery since the end of 90s, on the basis of the good clinical results
obtained by laparoscopic surgery in some benign diseases. The diffusion of
laparoscopic colonic surgery has increased in recent years thanks to specific
educational and training programs, but mainly in academic and high-volume
hospitals [1]. The most frequently-performed procedures are the less difficult,
such as sigmoidectomy for benign diseases and right colectomies with extra-
corporeal anastomosis [2]. Despite the improvement of technology some dis-
advantages of standard laparoscopy are still present, such as poor ergonomics,
the difficulty to perform intracorporeal sutures and the problem of the assistant
being able to hold the camera stable. Robotic surgery is an emerging technique
that seems to overcome some difficulties of the standard laparoscopic
approach, and its use in colorectal cancer is increasing quickly [3]. In this
chapter, the technical details of robotic left colectomy and segmental colonic
resections are reported.
P. P. Bianchi ()
Minimally Invasive Surgery Unit, European Institute of Oncology,
Milan, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 133
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_14, © Springer-Verlag Italia 2015
134 P. P. Bianchi et al.
The patient is placed in a supine position with arms alongside the trunk and legs
abducted. A slight Trendelenburg position and a right tilt are maintained in order
to expose the operative field from the ileal loops. The procedure starts with the
insertion of the Veress needle in the left hypochondrium through which a 12-
mmHg pneumoperitoneum is induced. A 12-mm standard laparoscopic trocar is
placed 2 cm right of the midline along an ideal line passing through the left
anterosuperior iliac spine and right hypochondrium. The procedure is performed
with a five-trocar technique: three 8-mm robotic trocars and two 12-mm stan-
dard laparoscopic trocars (Fig. 14.1a). The robotic cart approaches the operative
table from the left side of the patient and the robotic arms are connected to the
trocars, starting from the camera. Arm 1 is connected to the trocar in the right
iliac fossa, arm 3 is connected to the trocar in the right hypochondrium, arm 2 is
connected to the trocar in left iliac fossa (Fig. 14.1b). The initial instrument
positions are configured as follows: the robotic monopolar hook/scissors are
mounted on arm 1, the robotic grasper on arm 2 and the robotic bipolar grasper
on arm 3.
a b
Fig. 14.1 Robotic left colectomy/splenic flefure resection. a Trocar positioning. 1, monopolar
robotic cautery device (hook or scissors); 2, robotic grasper (Cadiere forceps); 3, robotic bipolar
grasper. b OR setup. (© 2014 Intuitive Surgical, Inc.)
the robotic monopolar hook on arm 1; the bipolar grasper on arm 3 helps this
surgical step by granting a stable tension on the root. This maneuver allows the
opening of the lesser sac, anterior to the pancreas (Fig. 14.2a). The IMV is freed
from the peritoneal envelope by the robotic monopolar hook on arm 1 and sec-
tioned between clips positioned by the assistant or by a robotic clip applier on
arm 1. Medial dissection is carried out to the end of the pancreatic tail; a gauze
may be positioned under the left mesocolon as a marker of the plane of dissec-
tion. Thereafter, the descending colon is retracted medially by the robotic
grasper on arm 1 and by the assistant with a grasper. Lateral mobilization of the
descending colon is achieved by the dissection of the parietocolic ligament per-
formed by the robotic monopolar hook on arm 2. During this step, robotic arm
3 is not utilized. The left colon is then freed laterally in the caudocranial path-
way up to the splenic flexure. The phrenocolic and splenocolic ligament, as well
as the sustentaculum lienis, are sectioned by the robotic monopolar hook on arm
1 and bipolar coagulation. The robotic grasper on arm 3 gently retracts the
spleen laterally with a gauze. Complete mobilization of the splenic flexure is
obtained by dissection of the coloepiploic ligament. Then the transverse colon is
retracted inferiorly by the grasper on arm 1 and by the assistant; the bipolar
grasper on arm 3 lifts the coloepiploic ligament cranially, which is dissected by
the robotic scissors on arm 2.
a b
grasper on arm 3, enhancing the shape of the IMA (Fig. 14.2b). The angle
between the IMA and the aorta is opened. This maneuver helps the surgeon to
identify the paraaortic nerves, which lie over the preaortic plane, the surgical
field is clear because of the stable retraction of the IMA by the locked robotic
grasper. A careful dissection of the IMA and a locoregional lymphadenectomy is
carried out preserving the paraortic nerves and the superior hypogastric plexus.
This step is performed with the robotic monopolar hook on arm 1 and bipolar
grasper on arm 3, which work synergically. The articulated tip of the robotic
monopolar hook facilitates dissection of the IMA, which is freed by the sur-
rounding lymphatic tissue, providing an optimal locoregional lymphadenectomy
(Fig. 14.2c). As for the IMV, the IMA is isolated between clips positioned by the
assistant or by robotic clip applier and then sectioned by the robotic scissors on
arm 1 or by the assistant.
or mesorectal fat. The proximal rectum is sectioned with a linear stapler con-
trolled by the assistant. This step can also be performed by a robotic stapler on
arm 1, if available. The anastomosis is fashioned according to the Knight &
Griffen technique. A minilaparotomy is created in the left iliac fossa or supra-
pubically. The descending colon is extracted through the protected incision and
transected proximally. The anvil of a circular stapler is inserted into the colon
stump and fixed by a manual purse-string suture. The colon is then reintroduced
into the abdomen and the minilaparotomy is closed. A laparoscopy is carried
out to perform the transanal end-to-end mechanical colorectal anastomosis.
During this step the robot is undocked, but not deactivated, as it could be use-
ful to perform additional sutures in the case of weak anastomosis.
The patient is placed in the supine position with arms alongside the trunk. A
slight anti-Trendelenburg position with a 10° right tilt are maintained in order to
138 P. P. Bianchi et al.
expose the operative field from the ileal loops. The positioning of trocars, OR
setup, and instruments are the same as for coloctomy (Fig. 14.1a, b).
14.3.3.2 Dissection of the Left Branch of the Middle Colic Vessels and
Locoregional Lymphadenectomy
During this phase robotic arm 2 is turned upward and the robotic grasper main-
tains the distal transverse colon retracted, exposing the transverse mesocolon
and facilitating vascular dissection. The assistant pulls the transverse colon
upward providing tension on the root of the transverse colon. The identification
of the middle colic vessels (MCV) starts at the level of the origin from the supe-
rior mesenteric axis near the inferior margin of the pancreas. The bipolar grasper
on arm 3 provides a stable tension of the middle colic vessels and a precise iso-
lation of the left branch of the MCV is performed by the robotic electrocautery
hook on arm 1. The vessels are clipped by the assistant or by the robotic clip
applier and sectioned by the robotic scissors on arm 1.
scopic linear stapler. The use of both laparoscopic and robotic instruments per-
mits a wider angle of motion and allows an easy transection of both the descend-
ing and transverse colon. After transection a stitch is placed to join together the
colonic stumps; the wide articulation of the tip of the robotic needle-holder
facilitates the intraoperative knotting, reproducing precise movements as in
open surgery (Fig. 14.3a). The robotic bipolar gasper on arm 3 holds the
descending colon stump while a colotomy is performed at the level of the tenia,
with the robotic monopolar hook on arm 1. Then the assistant holds the trans-
verse colon stump and a second incision is perfomed at the level of the tenia of
the transverse colon stump, again with the robotic monopolar hook on arm 1.
The laparoscopic linear stapler is introduced by the assistant through the trocar
in the right flank and the surgeon at the console helps the introduction of the two
branches of the stapler inside the colonic stumps with the robotic grasper on arm
1 and 3. A colo-colic side-to-side mechanical anastomosis is then performed
(Fig. 14.3b). The vascular perfusion of the site of the colic transection can be
evaluated by the use of indocyanine fluorescence and near-infrared image, if
available, in order to reduce the risk of an ischemic anastomotic damage [6]
(Fig. 14.3c). Thereafter, the entry hole of the stapler is closed by two running
sutures starting from the opposite angles. The robotic grasper on arm 2 pulls the
tail of the upper suture up favoring a stable position of the anastomosis, thus
facilitating the closure of the defect. The first running suture is performed from
the inferior angle upward. The tail of the upper suture and the inferior suture are
tied together. After completing the first layer, the second suture is performed
from the upper angle downward. As for the first layer, the upper suture is tied
with the tail of the inferior suture; the wide articulation of the robotic needle-
holder facilitates knot-tying (Fig. 14.3d). Then the robot is disconnected from
the patient and a suprapubic minilaparotomy is performed through which the
specimen is extracted inside a bag and with the placement of a wound protector.
The patient is placed in anti-Trendelenburg position with the arms along the
trunk and the legs abducted with a slight tilt to the right, which allows the small
bowel to roll off the operative field (Fig. 14.4a).
The procedure starts with the insertion of the Veress needle in the left
hypochondrium through which a 12-mmHg pneumoperitoneum is induced. A
12-mm standard laparoscopic trocar is placed 2cm right of the midline along an
ideal line passing through the left anterosuperior iliac spine and the right
hypochondrium. The procedure is performed with a five-trocar technique: three
8-mm robotic trocars and two 12-mm laparoscopic trocars. Three 8-mm trocars
140 P. P. Bianchi et al.
a b
d
Fig. 14.3 Splenic flexure resection. a Approach of the transverse and descending colonic stumps
by a stitch. b Fashioning of a side-to-side colo-colonic anastomosis by linear stapler. c Evaluation
of colo-colonic anastomosis perfusion by fluorescence and near-infrared imaging. d Robotic su-
ture of the service hole, after stapler colo-colonic side-to-side anastomosis
are placed in the left and right hypochondrium, and in the right flank, respective-
ly. A 12-mm trocar is placed in the left flank for the assistant. The robotic cart
approaches the operative table from the patient’s head and the robotic arms are
14 Left Colectomy and Segmental Resections for Cancer 141
a b
Fig. 14.4 Robotic transverse colonic resection. a OR setup. b Trocar positions. 1, monopolar
robotic cautery device (hook or scissors); 2, robotic grasper (Cadiere forceps); 3, robotic bipolar
grasper. (© 2014 Intuitive Surgical, Inc.)
connected to the trocars, starting from the camera. Arm 1 is connected to the tro-
car in the left hypochondrium, arm 2 is connected to the trocar in the right
hypochondrium, while arm 3 is connected to the trocar in the right flank (Fig.
14.4b). The initial instrument positions are configured as follows: the monopo-
lar hook/scissors are mounted on arm 1, the robotic grasper on arm 2 and the
robotic bipolar grasper on arm 3.
ic bipolar grasper on arm 3 pulls up the mesocolon while the robotic monopolar
hook or scissors on arm 1 perform a dissection of the transverse mesocolon from
the posterior peritoneal layer toward the right colic flexure. Dissection of the trans-
verse mesocolon is now achieved, resembling a V-shape with the base on the trans-
verse colon, which is now pulled down.
14.4.1.2 Mobilization of the Left and Right Colic Flexures and of the
Transverse Colon
The assistant provides a medial retraction of the proximal part of the descend-
ing colon and of the splenic flexure, as well as the bipolar grasper on arm 2,
while the parietocolic ligament is sectioned by the robotic monopolar hook up
to the left colic flexure. The phrenocolic and splenocolic ligament, as well as the
sustentaculum lienis, are sectioned by a robotic monopolar hook on arm 1. The
assistant now pulls the transverse colon down with a grasper, while the robotic
bipolar grasper on arm 2 lifts up the proximal side of the gastrocolic ligament.
The ligament is then dissected below the gastroepiploic vessels by the robotic
monopolar hook on arm 1 up to the right flexure. The right colon is then retract-
ed medially by the assistant by a laparoscopic grasper and by the robotic grasper
on arm 3, allowing the dissection of the right parietocolic ligament, performed
by the robotic monopolar hook on arm 2. Total mobilization of both the right and
left colic flexures and of the transverse colon is completed without changing the
docking of the robot.
The first robot-assisted colectomy was reported by Weber et al. in 2002 [6].
Since then several reports on robotic colorectal surgery have been published, but
originating mainly from single-center experiences. In Table 14.1, the results are
reported of the studies with more than 10 cases of robotic colonic resections. In
all of these studies the main significant data was that robotic surgery resulted in
a lower percentage of conversion to open surgery, compared to the laparoscopic
14 Left Colectomy and Segmental Resections for Cancer 143
Fig. 14.5 Robotic transverse colonic resection. Robotic sewing end-to-end colo-colonic anasto-
mosis
Year Study Type of study Total No. Mean Postoperative Conversion Mortality Type of
[Reference] of patients operative complications n (%) n (%) procedure
time (min) n (%)
2004 D’Annibale [7] Comparative 53 240 4 0 0 17 LC, 11 SC
2006 DeNoto [8] Case Series 11 197 2 1 0 11 SC
2006 Rawlings [9] Case Series 30 225 6 2 (LPT) 0 13 SC
2008 Soravia [10] Case Series 40 162 3 5 (3 LPS, 2 LPT) 0 28 SC
2008 Spinoglio [11] Comparative 50 383 7 2 (1 LPS, 1 LPT) 0 10 LC
2009 Luca [12] Case Series 55 290 12 0 0 27 LC
2011 Huettner [13] Case Series 102 229.7 19 0 0 43 SC
2011 Patel [14] Comparative 30 247 4 0 0 23 SC
2013 Helvind [15] Comparative 101 243 22 5 1 12 LC, 34 SC
2013 Bianchi Case Series 31 270 2 0 0 11 LC, 6 SF, 1
(unpublished TCR, 13 SC
data)
Total 503 248 81 (16) 15 (2.9) 1 (0.2) 66 LC, 150 SC
LC, left colectomies; SC, sigmoidectomies; TCR, transverse colon resection; LPT, laparotomy; LPS, laparoscopy; SF, splenic flexure resection
P. P. Bianchi et al.
14 Left Colectomy and Segmental Resections for Cancer 145
by Keller et al. [17] in the low volume users when compared to middle- and
high-volume centers and surgeons.
Although the results available on robotic surgery are still few, robotic assis-
tance seems to reduce the percentage of conversions to open surgery among
expert surgeons and is promising as a method to attenuate the learning curve of
more difficult procedures, such as segmental and splenic flexure resections with
intracorporeal anastomosis. At the moment, the robotic system has higher costs
than laparoscopy and its use should be planned within the remit of a clearly-
defined educational program, preferably in a hospital conducting middle/high
volumes of minimally invasive surgery and colorectal procedures, in order to
avoid an increase in complication rates.
References
1. Kang CY, Halabi WJ, Luo R et al (2012) Laparoscopic colorectal surgery. A better look to
the latest trends. Arch Surg 147:724–731
2. The Surgical Care and Outcomes Assessment Program (SCOAP) Collaborative, Kwon S,
Billingham R, Farrokhi E, Florence M, Herzig D, Horvath K, Rogers T, Steele S, Symons,
Thirlby S, Whiteford M, Flum D (2012) Adoption of laparoscopy for elective colorectal re-
section: a report from surgical care and outcomes assessment program. J Am Coll Surg
214:909–918
3. Bianchi PP, Pigazzi A, Choi GS (2014) Clinical Robotic Surgery Association Fifth Worldwide
Congress, Washington DC, 3–5 October 2013: Robotic Colorectal Surgery. Ecancermed-
icalscience 8:385
4. Schlachta CM, Mamazza J, Poulin EC (2007) Are transverse colon cancers suitable for la-
paroscopic resection? Surg Endosc 21:396–399
5. Kim HJ, Lee IK, Lee YS et al (2009) A comparative study on the short-term clinicopatholog-
ic outcomes of laparoscopic surgery versus conventional open surgery for transverse colon
cancer. Surg Endosc 23:1812–1817
6. Weber PA, Merola S, Wasielevski A, Ballantyne GH (2002) Telerobotic-assisted laparoscop-
ic right and sigmoid colectomies for benign disease. Dis Colon Rectum 45:1689–1694
7. D’Annibale A, Morpurgo E, Fiscon V et al (2004) Robotic and laparoscopic surgery for
treatment of colorectal diseases. Dis Colon Rectum 47:2162–2168
8. DeNoto G, Rubach E, Ravikumar TS (2006) A standardized technique for robotically performed
sigmoid colectomy. J Laparoendosc Adv Surg Tech A. 16:551–556
9. Rawlings AL, Woodland JH, Crawford DL (2006) Telerobotic surgery for right and sigmoid
colectomies: 30 consecutive cases. Surg Endosc 20:1713–1718
10. Soravia C, Schwieger I, Witzig JA et al (2008) Laparoscopic robotic-assisted gastrointestinal
surgery: the Geneva experience. J Robotic Surg 1:291–295
11. Spinoglio G, Summa M, Priora F et al (2008) Robotic colorectal surgery: first 50 cases ex-
perience. Dis Colon Rectum 51:1627–1632
12. Luca F, Cenciarelli S, Valvo M et al (2009) Full robotic left colon and rectal cancer resection:
technique and early outcome. Ann Surg Oncol 16:1274–1278
13. Huettner F, Pacheco PE, Doubet JL et al (2011) One hundred and two consecutive robotic-
assisted minimally invasive colectomies—an outcome and technical update. J Gastrointest Surg
15:1195–1204
14. Patel CB, Ragupathi M, Ramos-Valadez DI, Haas EM (2011) A three-arm (laparoscopic, hand-
assisted, and robotic) matched-case analysis of intraoperative and postoperative outcomes in
minimally invasive colorectal surgery. Dis Colon Rectum 54:144–150
146 P. P. Bianchi et al.
15. Helvind NM, Eriksen JR, Mogensen A et al (2013) No differences in short-term morbidity
and mortality after robot-assisted laparoscopic versus laparoscopic resection for colonic can-
cer: a case-control study of 263 patients. Surg Endosc 27:2575–2580
16. Baek SJ, Kim SH, Cho JS et al (2012) Robotic versus conventional laparoscopic surgery for
rectal cancer: a cost analysis from a single institute in Korea. World J Surg 36:2722–2729
17. Keller DS, Hashemi L, Lu M, Delaney CP (2013) Short-term outcomes for robotic colorec-
tal surgery by provider volume. J Am Coll Surg 217:1063–1069
Hybrid Robotic Technique for
Rectal Cancer: Low Anterior Resection 15
and Perineal Resection
15.1 Introduction
Colorectal cancer is the third most commonly diagnosed cancer with over 1.4
million new cases each year [1]. As surgical technology has evolved so has the
treatment for this disease. Rectal cancer resection is complicated by the anatom-
ic configuration of the pelvis and the proximity of these tumors to the anus.
Evolving optics have allowed for the increased use of laparoscopy to allow for
better visualization during pelvic surgery however its use was not implemented
initially. Both the technical challenge of laparoscopic rectal surgery and the
concern over oncological outcome have made its widespread adoption limited.
As more surgeons gain comfort with advanced laparoscopic techniques the only
concern is of the oncologic benefit [2–4]. The Conventional vs. Laparoscopic-
Assisted Surgery in Colorectal Cancer (CLASICC) trial examined oncologic
outcomes between laparoscopic and open rectal resections. Laparoscopic resec-
tion was associated with a higher rate of positive circumferential margin; how-
ever this did not translate into an increase in local recurrence when compared to
the open procedures [5]. Long-term follow-up from the CLASICC trial has con-
tinued to provide support for the safe use of laparoscopy in colon and rectal can-
cer. The overall survival at 5 years after a low anterior resection was 56.7% in
the open group and 62.8% in the laparoscopic one; abdominal perineal resection
showed similar results with an overall survival of 41.8% in open cases and
53.2% in laparoscopic cases [6]. The Colorectal cancer Laparoscopic or Open
Resection (COLOR II) trial has also advanced the use of laparoscopy and
helped to show similarity in the completeness of mesorectal resection with a
A. Pigazzi ()
Department of Surgery, University of California Irvine,
Irvine, CA, USA
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 147
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_15, © Springer-Verlag Italia 2015
148 T. F. Feldmann et al.
patient’s factors (i.e., incontinence) and tumor location, low anterior or abdom-
inal perineal resection can be completed after total mesorectal excision has been
performed. The robot is then undocked and the perineal or reconstructive por-
tion of the operation may be accomplished. We outline the basic steps in our
approach below.
Fig. 15.3 Dissection plane beneath the inferior mesenteric vein. The pancreas is visible at the left
edge of the picture. The inferior mesenteric artery will be found as dissection is carried toward the
right edge of the picture
mesenteric vein (IMV) or the inferior mesenteric artery (IMA) can be isolated
first and patient positioning should be changed as needed to allow for visualiza-
tion. We find that identification of the IMV near the ligament of Treitz allows for
development of the appropriate plane between the mesocolon and the retroperi-
toneum. The vein is identified lateral to the ligament of Treitz and the peri-
toneum below it is incised. Blunt dissection is carried laterally and inferiorly
until the IMA is encountered. This dissection plane will be bordered anteriorly
by the IMV and colonic mesentery, posteriorly by the retroperitoneum, superior-
ly by the pancreas, and inferiorly by the takeoff of the IMA (Fig. 15.3).
Dissection is carried out within this space. Care should be taken if the dissection
plane is continued superiorly as the natural dissection plane will continue under-
neath the pancreas. If this course is chosen then the surgeon will need to “step
up” over the pancreas to avoid injury to the splenic vessels. Dissection over the
pancreas can be accomplished later and therefore we recommend that blunt dis-
section be continued laterally and inferiorly until the IMA is reached.
Mobilization of the splenic flexure will be completed once the lateral attach-
ments have been divided. The IMV can be ligated at any point during the dissec-
tion through a variety of means (clips, staplers, bipolar cautery devices). Traction
must be avoided on the vein and earlier division can avoid an avulsion injury.
Once the superior aspect of the IMA is identified, the sigmoid colon is grasped
and retracted toward the anterior abdominal wall. The superior rectal artery
should be seen coursing through the mesenteric plane. Opening the peritoneum
underlying this pedicle will allow for blunt dissection of the mesocolon once
15 Hybrid Robotic Technique for Rectal Cancer: Low Anterior Resection and Perineal Resection 153
Fig. 15.4 Dissection plane beneath the superior hemorrhoidal artery. To the left of the picture is
the pedicle of the inferior mesenteric artery. The ureter is coursing directly through the middle of
the retroperitoneal plane
again. This dissection is again taken laterally and identification of the ureter and
gonadal vessels is paramount. These structures should be kept on the retroperi-
toneal side of the dissection. Visualization of the psoas muscle tendon often
implies a “deep” dissection and the ureter may have been swept into the tissue of
the mesocolon. Once this plane has been fully developed it should expose the
origin of the IMA and its branch point into the left colic and superior rectal
artery. This dissection plane will now be bordered superiorly by the IMA pedi-
cle, inferiorly by the mesorectum, anteriorly by the superior rectal artery and
mesocolon, and posteriorly by the retroperitoneum (Fig. 15.4). The ureter and
gonadal vessels travel in a crania-caudal direction within this area of the
retroperitoneum. In order to allow optimal mobilization of the left colon, as well
as to provide appropriate oncologic resection, the IMA should be taken at its ori-
gin. A classic view in the shape of a “T” should be seen at this point (Fig. 15.5).
The IMA will be the body of the T with the left colic artery as the left branch and
the superior rectal artery as the right branch. These structures should be free of
the retroperitoneum so that the ureter is not taken with the artery. Just as with the
IMV, the IMA can be taken using the surgeon’s preferred method.
Once the mesocolon has been dissected from the medial side, the lateral
colonic attachments must be taken down. The white line of Toldt is incised and
carried superiorly toward the splenic flexure. The omental attachments to the
descending colon should be taken as well. Dissection can continue from this
direction or it may also be approached from the transverse colon. This allows
mobilization of the omentum off of the colon and then entrance into the lesser
154 T. F. Feldmann et al.
Fig. 15.5 Classic view of “T”. The inferior mesenteric artery branches into the left colic (left
side) and the superior hemorrhoidal artery (right side). Both previous dissection planes can be
visualized
sac. The colon is retracted toward the midline and the splenocolic ligament is
taken down to allow for full mobilization of the flexure. This step will ensure
adequate colonic length to construct a tension free anastomosis once the resec-
tion is completed. After the mobilization is completed the patient is prepared for
docking of the robot.
• R1 – monopolar scissors
• R2 – bipolar fenestrated grasper
• R3 – ProGraspTM grasper
• L1 – suction irrigation device
• L2 – locking grasper (i.e., Davis & Geck)
The laparoscopic instruments will be used by the assistant, who will stay at
the patient’s right side. The operating surgeon should now proceed to the robot-
ic console. The assistant should grasp the upper rectum with the locking grasper
to allow for retraction throughout this part of the procedure.
Dissection of the mesorectum is then undertaken. Care should be taken to
stay in the avascular plane that exists between the endopelvic visceral fascia and
endopelvic parietal fascia. This will help to avoid the hypogastric nerve plexus
and sacral venous plexus located deep in the parietal layer within the presacral
space. R2 is used during this portion to provide upward traction on the mesorec-
tum while the assistant places countertraction with the suction device when able.
This allows for dissection to be carried out in the avascular plane posteriorly.
Frequent repositioning of R2 to maintain adequate countertraction will allow the
dissection plane to be continued to the level of the pelvic floor and for the later-
al stalks to be identified. R2 should be used with the wrist joint in an L shape to
allow for a larger surface area of retraction. The lateral stalks are then subse-
quently taken and dissection is carried onto the anterior surface of the rectum.
Here the vagina or the seminal vesicles and prostate are dissected free and pro-
tected. R3 should be used as a static retractor to move these structures anterior-
ly using a similar wrist configuration as used previously with R2. As the middle
rectal vessels are encountered they can be taken with bipolar cautery. The auto-
nomic nerves lie laterally on the pelvic sidewalls and should be kept free of the
dissection plane. The mesorectal excision continues downward in a cylindrical
fashion until past the level of the lesion. If no margin is feasible then dissection
should be carried as far as possible to allow for an easier perineal dissection. As
the plane below the mesorectum is reached (bare rectum area) further dissection
will lead to the intersphincteric plane. If a margin is obtainable, the rectum is
divided using a 45-mm stapler in either an anterior-posterior or right-to-left
fashion. Ideally, one or two loads are required. Once the rectum is free it is
grasped with the locking grasper at the site of the staple line and the robot can
be removed.
the specimen side and an end colostomy can be constructed without tension.
Once the abdominal portion of the procedure is completed, and the focus shifts
to the perineal resection, two options are available from a patient positioning
standpoint. The choice can be made based on surgeon preference; however loca-
tion of the tumor may assist in decision making. Tumors located in the posteri-
or portion of the rectum may be more easily approached from the lithotomy
position while anterior or lateral lesions may be more amenable to prone posi-
tioning.
• If a decision is made to continue the operation in the lithotomy position. The
legs are raised and a perineal incision outside the external sphincters is per-
formed circumferentially. Dissection is carried superiorly and the levator ani
divided. Starting the dissection posteriorly, toward the coccyx, allows for
discovery of the superior plane of dissection. The rectum is then cleared cir-
cumferentially. Once free, the specimen can be extracted through the perineal
incision. Closure of the perineal wound can then be performed, ideally in
multiple layers. If preferred, tissue flaps, such as a vertical rectus abdominus
myocutaneous (VRAM) flap, gluteal flap, or a gracilis flap can be transposed
for reinforcement. The defect is closed and an end colostomy is brought out
laparoscopically. The R2 port can be upsized into a permanent colostomy site
with maturation in standard fashion after the laparoscopic sites are closed.
• The second option involves repositioning into the prone position for the per-
ineal resection. Some studies would suggest this provides a better circumfer-
ential resection margin as opposed to lithotomy [13, 14]. In this case, and as
mentioned above, laparoscopic or robotic stapling is used to transect the
colon at the desired location and is then brought out through the R2 site as
an end colostomy prior to changing to a prone position. Ports are then
removed and the incisions closed and dressed prior to maturation of the
colostomy. After the abdominal portion is complete the patient is then repo-
sitioned into a prone jackknife position. Again a circumferential incision is
made outside the level of the external sphincter and dissection is carried
down in an extralevator fashion until the dissection planes are joined. The
specimen can then be extracted and the wound closed as per the surgeon’s
preference.
15.10 Conclusions
Robotic platforms have created new opportunities for rectal surgery. Through
our experience we have found a hybrid laparoscopic and robotic approach that
is safe, feasible, and efficacious. This technique allows us to tailor our approach
for each patient and the location of the tumor. As technology moves forward and
further research emerges, robotic rectal cancer surgery will allow for excellent
cosmetic, oncologic, and overall outcomes.
158 T. F. Feldmann et al.
References
1. Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, Rebelo M, Parkin DM,
Forman D, Bray F. GLOBOCAN 2012, Cancer Incidence and Mortality Worldwide: IARC
CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer;
2013. Available from: http://globocan.iarc.fr, accessed on 3/20/14
2. Clinical Outcomes of Surgical Therapy Study Group (2004) A comparison of laparoscopical-
ly assisted and open colectomy for colon cancer. N Engl J Med 350:2050–2059
3. Colon Cancer Laparoscopic or Open Resection Study Group, Buunen M, Veldkamp R, Hop
WC, Kuhry E, Jeekel J et al (2009) Survival after laparoscopic surgery versus open surgery
for colon cancer: long-term outcome of a randomised clinical trial. Lancet Oncol 10:44–52
4. Veldkamp R, Kuhry E, Hop WC et al (2005) Laparoscopic surgery versus open surgery for
colon cancer: short-term outcomes of a randomised trial. Lancet Oncol 6:477–484
5. Jayne DG, Guillou PJ, Thorpe H et al (2007) Randomized trial of laparoscopic-assisted re-
section of colorectal carcinoma: 3-year results of the UK MRC CLASICC Trial Group. J Clin
Oncol 25:306–308
6. Jayne DG, Thorpe HC, Copeland J et al (2010) Five-year follow-up of the Medical Research
Council CLASICC trial of laparoscopically assisted versus open surgery for colorectal can-
cer. Br J Surg 97:1638–1645
7. van der Pas MH, Haglind E, Cuesta MA et al (2013) Laparoscopic versus open surgery for
rectal cancer (COLOR II): short-term outcomes of a randomized, phase 3 trial. Lancet Oncol
14:210–218
8. Baik SH, Ko YT, Kang CM et al (2008) Robotic tumor-specific mesorectal excision of rec-
tal cancer: short-term outcome of a pilot randomized trial. Surg Endosc 22:1601–1608
9. Pigazzi A, Ellenhorn JD, Ballantyne GH, Paz IB (2006) Robotic-assisted laparoscopic low
anterior resection with total mesorectal excision for rectal cancer. Surg Endosc 20:1521–1525
10. Dalton RS, Smart NJ, Edwards TJ et al (2012) Short-term outcomes of the prone perineal ap-
proach for extra-levator abdomino-perineal excision (elAPE). Surgeon 10:342–346
11. Bianchi PP, Luca F, Petz W et al (2013) The role of the robotic technique in minimally inva-
sive surgery in rectal cancer. Ecancermedicalscience 7:357
12. Obias V, Sanchez C, Nam A et al (2011) Totally robotic single-position ‘flip’ arm technique
for splenic flexure mobilizations and low anterior resections. Int J Med Robot 7:123–126
13. Hellan M, Stein H, Pigazzi A (2009) Totally robotic low anterior resection with total mesorec-
tal excision and splenic flexure mobilization. Surg Endosc 23:447–451
14. Stelzner S, Hellmich G, Schubert C et al (2011) Short-term outcome of extra-levator ab-
dominoperineal excision for rectal cancer. Int J Colorectal Dis 26:919–925
Full-robotic Technique
for Rectal Cancer 16
Giuseppe Spinoglio, Giampaolo Formisano, Luca Matteo Lenti,
Fabio Melandro, and Alessandra Marano
16.1 Introduction
Since the first robotic total mesorectal excision (TME) was reported in 2006 [1],
two main methods for robotic rectal surgery, hybrid versus totally robotic tech-
nique, have been described.
In the hybrid technique [2, 3], the surgeon performs a splenic flexure mobi-
lization laparoscopically and in almost all cases the vascular handling, while the
robotic approach is reserved for pelvic dissection. Among the totally robotic
procedures, there are the single and dual docking strategies with several sub-
types, according to redocking of robotic arms and reorientation of the robotic
cart (or patient table) during surgery [4–8].
Since the introduction of the da Vinci® Si™ system at our institution in 2010,
we have developed a full-robotic single-docking technique for rectal anterior
resection with TME. In this chapter we outline each step of our surgical strate-
gy; the indications have already been discussed in Chapter 15.
G. Spinoglio ()
Department of General and Oncologic Surgery,
“Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 159
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_16, © Springer-Verlag Italia 2015
160 G. Spinoglio et al.
16.4 OR Setup
First assistant is on the patient’s right side. The scrub nurse is at the lower right
side of the table and the assistant’s monitor is located at the patient’s left shoul-
der. The cart is placed at the patient’s left side and is docked from the left lower
quadrant over the left hip, at approximately 45° to the perpendicular to the
patient (Fig. 16.1).
• one 8-mm robotic port (robotic arm 2, R2) in the left lower quadrant, 2 cm
lateral to the MCL and 16–18 cm to the pelvis;
• one 12-mm laparoscopic port, in the right flank area, 2 cm lateral to the
MCL.
column, the camera arm, the left anterior superior iliac spine and the camera
trocar
4. When the camera arm joins the camera trocar, the robotic cart is in the cor-
rect position and does not require further adjustments.
This method of positioning and docking the robot is the best one for the TME
step that is the core for the whole surgery.
Next, the patient is placed in a 15° reverse-Trendelenburg position and tilted
completely to the right side. Robotic arms are now docked to the trocar.
1. From top-to-bottom
The assistant grasps the omentum and pulls it down while the robotic instrument
in R2 lifts the stomach up. The gastrocolic ligament is incised in its more
translucent portion (the so-called Bouchet’s area): a section from the right to the
left side of the patient is progressively carried out considering the inferior
splenic pole as a landmark. Once the gastrocolic ligament is sectioned, the pos-
terior peritoneal layer of the lesser sac is incised 1 cm caudal to the inferior bor-
der of the pancreas in order to expose the Gerota’s fascia.
The tranverse mesocolon is separated from the inferior border of the pan-
creas and the dissection proceeds in order to fully mobilize the splenic flexure
and a portion of the descending colon, separating the Gerota’s fascia from the
Toldt’s one or the two layers of the Toldt’s fascia (Fig. 16.3).
2. From bottom-to-top (after primary vascular control step)
After having pulled up the greater omentum, the transverse mesocolon is opened
from its inferior aspect just above the body of the pancreas to enter the lesser
sac. Dissection of the transverse mesocolon continues toward the distal trans-
verse colon and the base of the descending colon. The omentum, attached to the
transverse colon, is then dissected in the avascular plane, beginning from the
middle third of the transverse colon: the renocolic and splenocolic ligaments are
divided and the splenic flexure is fully mobilized.
Our standardized technique involves the splenic flexure takedown from the
top to the bottom for the following reasons:
• This strategy allows easy access to the avascular plane between the Toldt and
Gerota’s fascia (or the two layers of the Toldt’s fascia) since the largest space
between these two layers is at the level of the inferior border of the pancreas.
164 G. Spinoglio et al.
We believe that a dissection starting from this area is easier than the other
one;
• In the case of the splenic flexure being in a high position, its takedown is eas-
ier to perform (also when an adhesiolysis is needed) and is safer since the
spleen is always under visual control;
• The anatomical plane created after the splenic flexure mobilization from the
top to the bottom makes the dissection easy to be implemented after the divi-
sion of the inferior mesenteric vein (IMV).
• The patient is placed in a steep Trendelenburg position with the right side
down.
• The arm setup is shown in Fig. 16.4:
- R1: Hot Shears™ (monopolar curved scissors) or cautery hook, Hem-o-
lok® clip applier
- R2: ProGrasp™ forceps
- R3: Bipolar forceps
• The assistant port is used for tissue retraction, suction/irrigation or clipping
of vessels.
a b
a b
Fig. 16.7 White light (a) and NIR-fluorescent view (b) of rectal transection
16 Full-robotic Technique for Rectal Cancer 167
Our personal experience includes 114 RAR with TME performed from March
2010 to May 2014. The patient demographic and perioperative data are shown in
Table 16.1. The procedure has been successfully performed in 109 out of 114
patients (conversion rate: 4.4%) with a mean operative time of 341 ± 98.6 min-
utes (range: 175–723 minutes, including loop ileostomy fashioning).
No intraoperative complications were observed. Bowel continuity was
restored with transanal stapled end-to-end or end-to-side anastomosis in 91
patients and with manual coloanal anastomosis for ultralow anterior resections
in 23 patients. A diverting loop ileostomy was matured in 90 patients (78.9%).
No positive distal resection margin was reported and the number of harvest-
ed lymph nodes was 20.1 ± 8. Four specimens (3.5%) showed circumferential
16 Full-robotic Technique for Rectal Cancer 169
References
1. Pigazzi A, Ellenhorn JD, Ballantyne GH et al (2006) Robotic-assisted laparoscopic low an-
terior resection with total mesorectal excision for rectal cancer. Surg Endosc 20:1521–1525
2. Hellan M, Anderson C, Ellenhorn JD et al (2007) Short-term outcomes after robotic-assisted
total mesorectal excision for rectal cancer. Ann Surg Oncol 14:3168–3173
3. Baik SH, Lee WJ, Rha KH et al (2008) Robotic total mesorectal excision for rectal cancer
using four robotic arms. Surg Endosc 22:792–797
4. Hellan M, Stein H, Pigazzi A (2009) Totally robotic low anterior resection with total mesorec-
tal excision and splenic flexure mobilization. Surg Endosc 23:447–451
5. Choi DJ, Kim SH, Lee PJ et al (2009) Single-stage totally robotic dissection for rectal can-
cer surgery: technique and short-term outcome in 50 consecutive patients. Dis Colon Rectum
52:1824–1830
6. Park YA, Kim JM, Kim SA et al (2010) Totally robotic surgery for rectal cancer: from splenic
flexure to pelvic floor in one setup. Surg Endosc 24:715–720
7. Obias V, Sanchez C, Nam A et al (2011) Totally robotic single-position ‘flip’ arm technique
for splenic flexure mobilizations and low anterior resections. Int J Med Robot 7:123–126
8. Bianchi PP, Ceriani C, Locatelli A et al (2010) Robotic versus laparoscopic total mesorectal
excision for rectal cancer: a comparative analysis of oncological safety and short-term out-
comes. Surg Endosc 24:2888–2894
9. Heald RJ (1988) The ‘Holy Plane’ of rectal surgery. J Royal Soc Med 81:503–508
10. Spinoglio G, Summa M, Priora F et al (2008) Robotic colorectal surgery: first 50 cases ex-
perience. Dis Colon Rectum 51:1627–1632
11. Kim SH, Kwak JM (2013) Robotic total mesorectal excision: operative technique and review
of the literature. Tech Coloproctol 17 Suppl 1:S47–53
12. D’annibale A, Pernazza G, Monsellato I et al (2013) Total mesorectal excision: a comparison
of oncological and functional outcomes between robotic and laparoscopic surgery for rectal
cancer. Surg Endosc 27:1887–1895
13. Alasari S, Min BS (2012) Robotic colorectal surgery: a systematic review. ISRN Surg
2012:293894
Robotic Surgery for Complicated
Diverticulitis 17
Giuseppe Spinoglio, Giampaolo Formisano, Francesca Pagliardi,
Ferruccio Ravazzoni, and Alessandra Marano
17.1 Introduction
The severity of complicated diverticulitis includes a broad spectrum of diseases
and is classified according to the Hinchey classification system [1]:
• Stage I: pericolic abscess, confined to the mesentery of the colon, usually
responsive to conservative management, with a radiological drainage in the
case of an abscess larger than 5 cm;
• Stage II: distant abscess amenable to percutaneous drainage (Stage IIa) or
complex and multiple abscesses with or without a digestive fistula (Stage IIb);
• Stage III: diffuse purulent peritonitis;
• Stage IV: diffuse fecal peritonitis;
Hinchey Stage I and II are amenable to laparoscopic lavage-drainage, when
percutaneous drainage has failed or in the case of clinical deterioration, which
may spare the patient from major emergency colorectal surgery or postpone an
elective delayed resection. Hinchey III diverticulitis represents a controversial
stage: a broad spectrum of surgical options is considered, which range from a
lavage-drainage to a major resection. An ongoing randomized trial (LADIES,
NTR2037) is expected to provide further recommendations [2, 3]. In the case of
Hinchey Stage IV, sigmoid resection with a primary anastomosis with or with-
out a diverting loop ileostomy can be performed [3]; however this stage repre-
sents the major indication for Hartmann’s procedure [2].
The laparoscopic approach for colorectal resection has been shown to be
feasible and safe in an emergency setting if performed by experienced surgeons
G. Spinoglio ()
Department of General and Oncologic Surgery,
“Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 171
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_17, © Springer-Verlag Italia 2015
172 G. Spinoglio et al.
[4, 5] but to date, due to the weak evidence, no recommendation can be made.
In elective settings, this minimally invasive technique provides a reduction in
postoperative pain, systemic analgesia requirements, hospital stay, overall post-
operative morbidity, total hospital cost and improved quality of life [6, 7].
However, the laparoscopic management is associated with a conversion rate of
up to 65% due to the presence of bulky mesenteric tissue and ill-defined planes
of dissection [8].
The robotic technique has been recently applied to the treatment of compli-
cated diverticular diseases in elective settings. Thanks to its technology, the da
Vinci® application seems to be promising and might overcome some limitations
of the laparoscopic approach. This chapter deals with our experience of robotic
delayed sigmoidectomy for complicated diverticulitis.
The details for this procedure are the same as those for our personal technique
for rectal anterior resection (See Chapter 16).
Table 17.1 Personal experience of robotic delayed sigmoidectomy for complicated diverticulitis
Variable Robotic sigmoidectomy group, n=25 pts
Sex (F/M) 11/14
Mean age, yrs (range) 60 (32–85)
Mean BMI, kg/m2 (range) 27.3 (18.7–33)
Hinchey I, n(%) 10 (40%)
Hinchey II, n(%) 14 (56%)
Stenosis, n(%) 1 (4%)
Conversion rate %, (n) 0% (0/25)
Central/peripheral vascular ligation (n) 16/9
Mean OT, min (range) 327 (200–515)
Mean LOS, days (range) 6.6 (3–13)
Intraoperative complications rate % 0%
Postoperative complications rate %, (n) 12% (3/25); 1 perianastomotic collection;
1 pulmonary embolism; 1 proctorrhage
OT, operative time; LOS, length of stay
176 G. Spinoglio et al.
complicated diverticulitis; this group was comparable to the robotic one in terms
of demographic characteristics and Hinchey disease stratification. A statistically
significant reduction in conversion rate (0% vs. 17.4%, p =0.027) was observed
in favor of the robotic group, with comparable operating room times (282 min
vs. 255 min; p =0.167). No statistically significant differences were observed
regarding hospital stay or 30-day postoperative morbidity between the compared
groups.
Our results support the robotic approach as a safe and effective tool in max-
imizing the advantages of minimally invasive surgery in the management of
complicated diverticular diseases. The 3DHD vision and the Endowristed tech-
nology make the robotic approach particularly useful during surgery performed
in these settings. The dissection of hard and severely inflamed tissue in the
abdomen (and also in a narrow pelvis) and the identification of the proper plane
are likely to be the areas that benefit from this technology mainly.
References
1. Hinchey EJ, Schaal PG, Richards GK (1978) Treatment of perforated diverticular disease of
the colon. Adv Surg 12:85–109
2. Mutter D, Marescaux J (2013) Appendicitis/diverticulitis: minimally invasive surgery. Dig Dis
31:76–82
3. Agresta F, Ansaloni L, Baiocchi GL et al (2012) Laparoscopic approach to acute abdomen
from the Consensus Development Conference of the Societa Italiana di Chirurgia Endoscop-
ica e nuove tecnologie (SICE), Associazione Chirurghi Ospedalieri Italiani (ACOI), Societa
Italiana di Chirurgia (SIC), Societa Italiana di Chirurgia d’Urgenza e del Trauma (SICUT),
Societa Italiana di Chirurgia nell’Ospedalita Privata (SICOP), and the European Association
for Endoscopic Surgery (EAES). Surg Endosc 26:2134–2164
4. Zdichavsky M, Granderath FA, Blumenstock G et al (2010) Acute laparoscopic intervention
for diverticular disease (AIDD): a feasible approach. Langenbecks Arch Surg 395:41–48
5. Cirocchi R, Cochetti G, Randolph J et al (2014) Laparoscopic treatment of colovesical fistu-
las due to complicated colonic diverticular disease: a systematic review. Tech Coloproctol
doi:10.1007/s10151-014-1157-5
6. Klarenbeek BR, Veenhof AA, Bergamaschi R et al (2009) Laparoscopic sigmoid resection for
diverticulitis decreases major morbidity rates: a randomized control trial: short-term results
of the Sigma Trial. Ann Surg 249:39–44
7. Alves A, Panis Y, Slim K et al (2005) French multicentre prospective observational study of
laparoscopic versus open colectomy for sigmoid diverticular disease. Br J Surg 92:1520–1525
8. Ragupathi M, Ramos-Valadez DI, Patel CB et al (2011) Robotic-assisted laparoscopic sur-
gery for recurrent diverticulitis: experience in consecutive cases and a review of the litera-
ture. Surg Endosc 25:199–206
9. Purkayastha S, Constantinides VA, Tekkis PP et al (2006) Laparoscopic vs. open surgery for
diverticular disease: a meta-analysis of nonrandomized studies. Dis Colon Rectum 49:446–463
10. Maciel V, Lujan HJ, Plasencia G et al (2014) Diverticular disease complicated with colovesi-
cal fistula: laparoscopic versus robotic management. Int Surg 99:203–210
11. Guillou PJ, Quirke P, Thorpe H et al (2005) Short-term endpoints of conventional versus la-
paroscopic-assisted surgery in patients with colorectal cancer (MRC CLASICC trial): multi-
centre, randomised controlled trial. Lancet 365:1718–1726
Part V
New Technologies in Robotic Platform
Single-Site™ Surgery
18
Giuseppe Spinoglio, Giampaolo Formisano,
Luca Matteo Lenti, Fabio Priora, and Alessandra Marano
18.1 Introduction
The Single-Site™ platform was primarily designed to work in a narrow opera-
tive field and with a specific anatomical target. To date, the most consistent pub-
lished experiences are regarding the use of this technology to perform cholecys-
tectomy [1–9] but, recently, it has been applied in other fields of general surgery
[10–12]. This chapter will focus on its current application in performing chole-
cystectomy and right colectomy.
Under general anesthesia, the patient is secured in a supine position with both
arms tucked at sides. The robotic cart should approach the patient at 45° (from
perpendicular) over the right shoulder ensuring that the target anatomy is in-line
with the center column, umbilicus and arrow on the port. The assistant and the
scrub nurse are positioned at the patient’s left side and at the patient’s feet,
respectively; the main assistant monitor is located at the right of the patient in
the line of vision of the assistant (Fig. 18.1a).
G. Spinoglio ()
Department of General and Oncologic Surgery,
“Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 179
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_18, © Springer-Verlag Italia 2015
180 G. Spinoglio et al.
a b
Fig. 18.1 a Overhead views of OR setup for Single-Site™ robotic cholecystectomy and Single-
Site TM robotic right colectomy. b Single-Site™ port placement for cholecystectomy. (© 2014
Intuitive Surgical, Inc.)
a 45° angle over the right shoulder ensuring that the target anatomy is in-line
with the center column, umbilicus and arrow on the port.
For SSRC, only robotic arm 1 (R1) and arm 2 (R2) are used in addition to the
camera arm:
• the camera arm has to be in line with the center of the column and bent at an
angle of 45° (sweet spot);
• R1 is placed to the left of the patient and its instruments reach the surgical
field from the right: Cadiere forceps (alternatively crocodile or Maryland
grasper);
• R2 is placed to the right of the patient and its instruments reach the surgical
field from the left: monopolar cautery with hook tip, Hem-o-lok® clip appli-
er, curved scissors, suction/irrigator;
• accessory port: laparoscopic grasper.
It is worth pointing out that one of the main innovations is that the da Vinci®
software restores the correct hand/instrument correlation: the surgeon uses the
right hand to drive the instrument visualized on the right side of the operative
screen, even if the instrument comes from the left and vice versa.
The endoscope is reinserted vertically; under direct vision the lubricated
curved cannulae are inserted by sight to avoid visceral injury, in order that they
cross inside the port keeping the remote center at the level of the abdominal
wall. Then, the robot is docked.
At the patient’s bedside, the assistant grasps the fundus of the gallbladder and
lifts up and toward the patient’s right shoulder to expose the hepatoduodenal lig-
ament. At the console, the da Vinci® software has automatically associated the
surgeon’s hand to the ipsilateral instrument tip: the surgeon retracts the scope,
moves it under the grasper, and pushes it toward the gallbladder while the assis-
tant lifts the grasper (and therefore the fundus of the gallbladder) upward.
Finally, the surgeon can easily retract the gallbladder infundibulum laterally to
open the Calot’s triangle, as in multitrocar laparoscopy.
Dissection begins and follows the traditional steps until the critical view of
safety is achieved. The peritoneum, close to the gallbladder neck, is incised with
the monopolar hook by gentle dissection of the peritoneal elements. The cystic
duct and artery are identified and skeletonized (Fig. 18.2). The ligation is per-
formed with Hem-o-lok® ML clips (clip applier in R2) and the division is per-
formed with curved scissors (R2).
The dissection of the gallbladder from the liver bed is carried out using the
Cadiere forceps (R1) and monopolar cautery hook (R2). During this last step,
the scope is repositioned above the grasper to lift the liver and expose the surgi-
cal field. Before completion of the dissection of the gallbladder from the liver,
it is important to inspect the gallbladder bed and ducts for evidence of bleeding
182 G. Spinoglio et al.
a b
Fig. 18.2 Final view of cystic duct and cystic artery after Calot’s dissection in white light (a) and
near-infrared fluorescent (b) view. CD, cystic duct; CA, cystic artery; CHD, common hepatic
duct; CBD, common bile duct
18.3 Discussion
18.3.1 Advantages
• the correct triangulation of the target anatomy is restored by the rigid can-
nulae, crossing at the monoport and the robotic arm collisions are mini-
mized;
• the bedside-assistant port enables dynamic retraction to perform a safe dis-
section;
• any internal clashing with other instruments are avoided because of the
placement of the robotic camera in the middle of the curved cannulae zone
(instead of being parallel as in SILS);
• the same-sided hand-eye control of the instruments is maintained by the da
Vinci® intuitive software: “what is right is right and vice versa”;
• the added benefits of the stable 3DHD view, the precision, the better
ergonomic comfort and the intraoperative dynamic fluorescent cholangiogra-
phy increase the safety and make the surgery easy to perform.
18.3.2 Limitations
The most consistent experiences of SSRC are reported in Table 18.1. Nine arti-
cles evaluating over 600 SSRC with Single-Site™ kit have been published
between 2011 and January 2014 [1–9]. Based on available data, the approach is
safe and feasible and has been easily adopted by surgeons mainly of American
and European origin.
SSRC was successfully carried out not only in the case of symptomatic
cholelithiasis but also for cholecystitis with a conversion rate of about 3.8%. The
mean length of hospital stay, derived from seven articles, was 30.2 hours and
few early or late major complications have been reported. It is worth saying that
no bile duct injuries occurred in any of the above mentioned papers.
Our single institutional experience consists of 151 patients whom underwent
SSRC for symptomatic cholelithiasis (n = 139, 92%), cholecystitis (n = 8, 5.3%)
and gallbladder polyposis (n = 4, 2.7%) from July 2011 to May 2014. The mean
total operative time was 70.7 minutes; no conversions occurred and no addition-
al port was required. With a mean hospital stay of 1.2 days, no major intraoper-
184 G. Spinoglio et al.
The patient is placed in a partial Trendelenburg supine position with a slight roll
to the left and the cart is on the patient’s right shoulder. The assistant is at the
patient’s left side and the scrub nurse at the patient’s feet. The main assistant
monitor is located at the patient’s right side (Fig. 18.1).
The first surgical step is represented by the lateral retraction of the last ileocol-
ic loop with the Cadiere grasper tenting up the ileocolic vessels; then, in order
to create a window under these, the peritoneum is opened with the cautery hook
to visualize the duodenum. The ileocolic vessels are clipped, sectioned (Fig.
18.5) and following the superior mesenteric vein (SMV) left anterior surface
upward, the right branches of the middle colic vessels are clipped and sectioned
too, as a result of the cephalad traction of the transverse mesocolon with the
assistant’s laparoscopic grasper. A medial-to-lateral colonic mobilization is per-
formed in the avascular plane between Gerota’s and Toldt’s fasciae keeping the
right ureter and the gonadic vessels down.
For malignant diseases, a complete mesocolic excision (CME), according to
the principles of Hohenberger et al. [14], is carried out as we usually perform in
Table 18.1 Single-Site™ robotic cholecystectomy (SSRC): major clinical series of at least 25 cases
Study Study No. of Preoperative Age, years; Docking; console; Conversion; Incision, LOS, Major early;
[Reference] design patients diagnosis BMI, kg/m2; total time, min. adding port; cm hours late complica-
(F/M) ASA score, (mean, SD) n (%) (mean) (mean) tions, n (%)
(mean, SD)
Konstantinidis Case series 45 36 choleli- 47 ± 12; 28.8 ± 4; 5.8 ± 1.56; 43 ± 21.98; 0(0); 3(6.7) 2 23.8 1(2.2); 0; 1
[1] (2012, (22/23) thiasis; 1.48 ± 1 84.5 ± 25.5 hemorrhage
18 Single-Site™ Surgery
Greece) 5 cholecystitis;
4 other1
Pietrabissa Prospective 100 100 53.4 ± 12.9; 13 ± 6; 32 ± 13; 2 to OC(2); 0 2 - 0; –
[2] (2012, observational (71/29) cholelithiasis 24.4 ± 3.7; – 71 ± 19
Italy)
Spinoglio [3] Retrospective 25 23 cholelithiasis; 54.2 (19–78)*; 5.2 ± 1.1; 22.3 ± 10.9; 0; 0 2.5 26.4 0;0
(2012, Italy) comparativea (20/5) 2 other1 23.7(16.5–32.4)*; – 62.7 ± 16.6 p
Angus [4] Case series 55 37 cholelithiasis; 46.01± 4.25; 11.34±3.74; 0; 0 2 - 0;0
(2013, USA) (34/21) 18 other3 26.57± 4.25; – 28.74±11.04;
61.84±14.66
Gonzalez [5] Retrospective 166 127 cholelithiasis; 51.6 ± 15.9; – ; – ; 63.0 ± 25.2 p 0; 3(1.8) 2.5 28.8 2(1.2); 0; 2
(2013, USA) comparativeb (131/35) 20 cholecystitis; 29.4 ± 6.2; intrabdominal
19 other1,2,3 1.84 ± 0.73 abscesses
Spinoglio [6] Case series 45 45 cholelithiasis 48(23–76)*; – ; 24.1 (7.3–59.3)*; 0; 0 - 26.4 0; –
(2013, Italy) (33/12) 24.7(19–43)*; – 67(35–110)*
Uras [7] Case series 36 36 cholelithiasis 40.1(21–64)*; 9.8 (4–30)*; 0; 1(2.7) 2.5 24 0; 1(2.7); 1
(2013, (31/5) –;– 24.9 (7–60)*;
Turkey) 61.8 (34–110)* incisional hernia
(cont.)
185
Table 18.1 (continued)
186
Study Study No. of Preoperative Age, years; Docking; console; Conversion; Incision, LOS, Major early;
[Reference] design patients diagnosis BMI, kg/m2; total time, min. adding port; cm hours late complica-
(F/M) ASA score, (mean, SD) n (%) (mean) (mean) tions, n (%)
(mean, SD)
Vidovszky [8] Prospective 95 80 cholelithiasis; 45.2 ± 16.1; 4.9 ± 3.0; 7 (6 to LC, 1 2.5 24 4(4.2); 0; 3
(2013, USA) cohort study (68/27) 8 cholecystitis; 30.1 ± 7.1; – 39.1 ± 15.3; to OC, 8.4)/ 0 (84%pts) residual CBD
7 other1,3,4 88.63 ± 32.0 lithiasis,
1 biloma
Morel [9] Case series 82 82 cholelithiasis 48.74 ± 13.23; 6.60 ± 7.58; 50.90 ± 1 to OC(1.2); 2.93 58.1 1(1.2); 1(1.2);
(2014, (72/28) 26.33 ± 4.18; 23.25; 91.05 ±29.92 2(2.4) 1 duodenal
Switzerland) 1.73 ± 0.54 lesion/1
incisional
hernia
aSSRC vs. Single-Incision Laparoscopic Cholecystectomy (SILC); bSSRC vs. SILC vs. SPIDER approach; 1 gallbladder polyps; 2 biliary pancreatitis; 3 biliary dyskinesia;
4 Recurrent pyogenic cholangitis; *median(range); p Value with statistical significance between compared group; CBD, common bile duct; OC, open cholecystectomy; LC,
laparoscopic cholecystectomy; LOS, length of stay
G. Spinoglio et al.
18 Single-Site™ Surgery 187
Table 18.2 Demographics and perioperative outcomes of our series of Single-Site™ robotic
cholecystectomy
Variable Total No. of patients, n=151
Sex, n (F/M) 109/38
Age, years (mean ± SD) 49.4 ± 14.0
BMI, kg/m2
mean ± SD 24.5 ± 4.2
<25, n (%) 89 (58.9)
25–29.9, n (%) 46 (30.5)
>30, n (%) 16 (10.6)
Previous major abdominal surgery, n (%) 54 (37.2)
Preoperative diagnosis
Symptomatic cholelithiasis, n (%) 139 (92)
Cholecystitits, n (%) 8 (5.3)
Gallbladder polyps, n (%) 4 (2.7)
Total operative time, min (mean ± SD) 70.7 ± 24.8
Conversion/adding ports, n (%) 0 (0)
IOC, n (%) 1 (0.7)
Major intraoperative complication, n (%) 0 (0)
Major early postoperative complication, n (%) 0 (0)
LOS (mean ± SD) 1.17 ± 0.5
Mean follow-up, months 26.3
IOC, intraoperative cholangiography; LOS, length of stay
Fig. 18.5 Skeletonization and division of ileocolic vessels at their root. ICA, ileocolic artery; ICV,
ileocolic vein; SMA, superior mesenteric artery; SMV, superior mesenteric vein
Fig. 18.6 The ileum and the transverse colon are mechanically transected with two stapler appli-
cations, including the remaining enterotomies, after having performed the side-to-side anisoperi-
staltic anastomosis
18.5 Discussion
18.5.1 Advantages
teric axis with the Pfannenstiel minilaparotomy. The extraction of the specimen
through this incision is carried out in order to obtain an improved cosmesis, a
decreased pain and a lower rate of incisional hernia compared to median supra
and infraumbilical laparotomy.
Fluorescence imaging with indocyanine green (ICG), integrated into the da
Vinci® system, can also be used to verify bowel stump perfusion before bowel
transection.
18.5.2 Limitations
Up to date, one case of SSRRC has been published [11]. It was in a patient
affected by a malignant polyp of the caecum and both resection and anastomo-
sis have been carried out extracorporeally. The authors reported about the safe-
ty and feasibility of this procedure underlying the main technical benefits of this
system.
Our experience consists of three full SSRRC with the robotic monoport
inserted through a left sovrapubic transverse incision [12]. Overall SSRRC oper-
ative time was 218.3 ± 75.9 min. A side-to-side anisoperistaltic anastomosis was
fashioned intracorporeally in two out of three cases. All patients were discharged
within 5 days from surgery; oncological principles have been satisfied in both
two candidates affected by colon cancer.
References
1. Konstantinidis KM, Hirides P, Hirides S et al (2012) Cholecystectomy using a novel Single-
Site® robotic platform: early experience from 45 consecutive cases. Surg Endosc 26:2687–2694
2. Pietrabissa A, Sbrana F, Morelli L et al (2012) Overcoming the challenges of single-incision
cholecystectomy with robotic single-site technology. Arch Surg (Chicago, Ill: 1960)
147:709–714
18 Single-Site™ Surgery 191
19.1 Introduction
The imaging technique based on indocyanine green (ICG) fluorescence has been
widely used for more than forty years, especially to study blood flow and micro-
circulation. This method was first applied in general surgery to perform sentinel
lymph node (SLN) biopsies in patients affected by breast and colorectal cancer.
In 2010, a near-infrared (NIR) laser light system was integrated with the da
Vinci® Si™ HD robotic system (Intuitive Surgical Inc., Sunnyvale, CA, USA).
This imaging system is able to provide both white light and near-infrared light
images through dedicated endoscopic illuminators and filters by simply press-
ing a pedal on the surgical console, thus allowing real-time fluorescence-guid-
ed surgery. There are many fields of application of ICG fluorescence in robotic
general surgery, some experimental and still evolving, that include:
• intraoperative (IO) fluorescent cholangiography to assess biliary anatomy;
• evaluation of bowel stump perfusion;
• lymph node (LN) mapping and the SLN biopsy in colorectal cancer surgery;
• colorectal tattoing.
In this chapter, we describe our nearly 3-year experience with fluorescence-
guided robotic general surgery.
G. Spinoglio ()
Department of General and Oncologic Surgery,
“Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 193
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_19, © Springer-Verlag Italia 2015
194 G. Spinoglio et al.
Bile duct injury (BDI) is a rare but serious and feared complication of cholecystec-
tomy and in the current era of laparoscopy the incidence of these lesions increased
to 0.2–0.5% [1] when compared to open surgery (0.1–0.2%) [2]. Moreover, single-
incision laparoscopic cholecystectomies may be associated with an even higher rate
of BDI (0.7–0.8%) [3] due to the loss of the linchpins of conventional laparoscopy,
such as triangulation and respect of optimal working angles. A “misperception” of
the biliary anatomy is considered the primary cause of BDI rather than the lack of
skill or knowledge and the use of a routinely intraoperative cholangiography (IOC)
has been emphasized recently [4]. While recognizing the importance of IOC, it has
several disadvantages, such as the interruption of the regular workflow with
increased operative time, requirement for a multidisciplinary team and equipment,
patient and staff exposure to radiation and the necessity to cannulate the cystic duct.
Moreover, it provides static images that are often difficult to interpret. As already
mentioned by some authors, there is a need for a “simpler method of locating the
course of the ductal system during the operation, something simpler than cholan-
giography or ultrasonography” [5].
Single-Site™ robotic cholecystectomy (SSRC) certainly allows easier and safer
surgical procedures than single-incision laparoscopy to be performed by overcom-
ing its inherent technical limitations; however, the difficulty in visualizing the bil-
iary structures can still remain. The use of the da Vinci® NIR-fluorescence imaging
system could be a valid solution to allow real-time view of the anatomy of the
extrahepatic biliary tract and a time-efficient dissection of Calot’s triangle, thus fur-
ther increasing safety during single-site surgery.
19.2.2 Technique
a b
Fig. 19.1 White light (a) and near-infrared view (b) of biliary anatomy prior to Calot’s dissection.
CD, cystic duct; CBD, common bile duct; CHD, common hepatic duct; RHD, right hepatic duct
a b
Fig. 19.2 White light (a) and near-infrared view (b) of biliary anatomy after Calot’s dissection.
CD, cystic duct; CA, cystic artery; CBD, common bile duct; CHD, common hepatic duct; RHD,
right hepatic duct; LHD, left hepatic duct
iary confluence. If there are concerns with the vascular anatomy, it is possible to
proceed with a further injection of 2.5mg of ICG: after 10–20seconds, hepatic and
cystic arteries can be visualized to avoid any damage to both main and anomalous
branches, especially the branch to the sixth hepatic segment. During the detachment
of the gallbladder from the liver bed, the use of fluorescence is useful to define the
limit between the gallbladder and liver bed (especially in cases of a thin or intra-
hepatic gallbladder) and to visualize any aberrant Luschka ducts. At the end of the
procedure, a final fluorescent view of the operative field may be prudent.
19.2.3.1 Advantages
The advantages of FC over standard IOC are numerous:
• no interruption of the regular workflow: the surgeon operates both in white and
fluorescent light;
196 G. Spinoglio et al.
• real-time interpretation of the images, in contrast to the traditional IOC that pro-
vides static information;
• control of the cystic duct section with a clear distinction of its confluence with
the common bile duct;
• real-time detection of any aberrant Luschka ducts and bile leaks;
• identification of the vascular anatomy of the hepatic and cystic arteries, if nec-
essary.
19.2.3.2 Limitations
The main limitations of FC are obesity and inflammation since fluorescent NIR light
has a limited tissue penetration (5–10mm). The ability of NIR light to reach deep-
er, inflamed and edematous tissues must be further investigated. Moreover, the capa-
bility of this method to recognize common bile duct stones or other obstructions is
low and, of course, IOC remains the best method to assess choledocholithiasis.
Table 19.1 Visualization of fluorescent biliary ducts prior to and after Calot’s dissection
No. of visualization/total procedures %
Before Calot’s dissection
Cystic duct 114/120 95
CHD 92/120 76.6
CBD 95/120 79.2
At least one duct 116/120 96.6
After Calot’s dissection
Cystic duct 119/120 99.2
CHD 111/120 92.5
CBD 116/120 96.6
At least one duct 120/120 100
CHD, common hepatic duct; CBD, common bile duct
We can conclude that FC is a simple, fast, safe and effective procedure. It allows
clear real-time identification of extrahepatic biliary anatomy in almost all patients,
thus implementing the well-known advantages of SSRC over the traditional single-
incision laparoscopic approach.
The anastomotic leakage is one of the most feared complications in colorectal sur-
gery occurring in 1.2–19.2% of all operations, with rates reaching 39% for rectal
cancer cases requiring low or ultralow anastomosis.
The underlying pathogenic mechanism has not yet been fully clarified and is
thought to be a multifactorial problem: adequate exposure and access, correct sur-
gical technique and gentle handling of the bowel, adequate hemostasis, absence of
tension at anastomosis, avoidance of fecal contamination and approximation of
well-vascularized bowel are all cornerstones to safely performing an intestinal
anastomosis and have historically been considered as “good practice” by the surgi-
cal community.
Besides technical aspects, adequate bowel perfusion is certainly the main factor
in ensuring the integrity of an anastomosis: it has been traditionally based on sub-
jective parameters such as the presence of active bleeding from the section line,
pulsatility of the mesenteric vessels, as well as peristalsis and the lack of discol-
oration of bowel segments. The clinical judgment alone, however, could represent
an inaccurate method to evaluate the risk of anastomotic leakage, with limited pre-
dictive accuracy [11].
Though many different solutions have been proposed [12–14], the da Vinci®
NIR-fluorescence imaging system represents the latest innovation in the field: it
198 G. Spinoglio et al.
a b
Fig. 19.3 White light (a) and near-infrared view (b) for evaluation of bowel stump perfusion
can be used to assess in real time both macroscopic vascular anatomy and perfu-
sion of intestinal stumps thus supporting surgeons in choosing the correct section
point during colorectal resections. Moreover, it can be extremely useful in nonstan-
dardized colonic procedures (i.e., transverse colon, splenic flexure), since vascular
abnormalities can strongly impair blood supply.
19.3.2 Technique
Surgery is performed with the da Vinci® Si™ HD surgical system according to the
usual technique of vascular control and mesenteric division to prepare the intestinal
segments. Afterward, the optimal transection line is evaluated by the surgeon with
white light and a dose of 0.1 mg/kg of ICG is administered intravenously: after
approximately 15–30seconds, the surgeon switches to fluorescent vision (Fig. 19.3).
If the site chosen for the section does not appear to be sufficiently perfused, the sec-
tion line may be revised and the stapler can be moved to a more proximal/distal
location according to the best fluorescence point reflected. In case of doubt, the test
can be repeated after waiting for a few minutes to allow the dye to washout.
However, depending on the tissue, the green fluorescent intensity appears different
and at different times.
Regarding the colonic stumps, the vessels of the epiploic appendices and
mesentery turn green first, then the green spreads across the intestinal wall. The
antimesenteric border of the descending and transverse colon is always a little
paler, because the vascularization of the tenia is less intense due to the thickness of
muscle tissue. The perfused segments gradually become green until they assume a
bright green color, in contrast with the gray segments that are not well vascularized.
Further checks may be carried out before and after performing the anastomosis.
With regard to the rectal stump, the pelvic wall turns green first (as does the
uterus in women), because it is highly vascularized. After a few seconds, the rectal
stump colors up allowing the assessment of perfusion at the selected point. The
ends of the section lines of the rectum in a Knight-Griffen anastomosis or of the
colon in a laterolateral anastomosis are often referred to as critical points of leak-
19 ICG Fluorescence: Current and Future Applications 199
age for their potential minor irroration: particular attention should be paid to their
perfusion.
19.3.3.1 Advantages
Visual inspection and caution are often sufficient to choose a well-vascularized
intestinal stump for anastomosis, especially for experienced surgeons. However
there are difficult cases, both for patient conditions (such as obesity, diabetes,
inflammatory disease, etc) and for the type of anastomosis (i.e., ultralow anterior
rectal resection, splenic flexure resection, right colectomy), in which the evaluation
of perfusion is important, even if only as confirmation, especially in the presence of
a thick mesocolon and short mesentery. Moreover, this application might clearly be
useful during minimally invasive surgery where there is a loss of tactile feedback.
19.3.3.2 Limitations
A limitation of this method can be represented by the fluorescence of the peri-
toneum due to an accidental spillage in the course of injection for tattoo or LN map-
ping. In this case, it may be difficult to appreciate the fluorescence of the stump tied
to the perfusion, as distinct from the one linked to impregnation of the tissue.
Unfortunately, at present it remains a linear-graded result that requires subjec-
tive interpretation as to the cutoff point between sufficient and insufficient perfu-
sion. However, in our opinion, there is a very clear visible cutoff in fluorescence
mode between not perfused and perfused bowel during the first seconds after ICG
administration. Quantitative real-time analysis of the fluorescence image would be
desirable but is currently not available on the robotic system.
Promising outcomes have also been reported by Bae et al. [16] in three case
reports. The authors applied the ICG fluorescence during robotic LAR for cancer
to better demark the ischemic area in the distal rectum so that the surgeon was
helped to define the distal resection margin.
A prospective multicenter study [17] has recently evaluated the impact of fluo-
rescence imaging on visualization of perfusion and subsequent change of transection
point during left-sided robotic colorectal surgery. Fluorescence imaging was applied
on 40 patients and resulted in a change of the proximal transection location in 40%
(16/40) of the candidates: two patients (5%) with a change in transection line devel-
oped an anastomotic leak at postoperative days 15 and 40. The authors concluded
that fluorescence imaging during colorectal procedures provides important addition-
al information about bowel perfusion at the transection site; this may eventually help
to decrease leaks caused by hypoperfusion.
From September 2011 to May 2014, we have performed bowel stump evalua-
tion with fluorescence imaging system in 128 full-robotic multiport colorectal
resections for benign and malignant diseases (unpublished data), which included:
51 right colectomies, 39 low and ultralow anterior rectal resections, 32 left colec-
tomies and 6 splenic flexure resections. We routinely choose the transection line,
after mesenteric section and bowel preparation, according to intestinal perfusion as
shown by fluorescence imaging system. There were no intraoperative or anesthetic
complications associated with the injection of ICG dye and the fluorescence bowel
assessment was easily accomplished in real-time in 100% of cases with no addi-
tional operative time.
We registered one case of anastomotic leak (0.8%) in a 78-year-old obese male
patient affected by sigmoid cancer. He presented a Grade C anastomotic leakage on
postoperative day 4 and therefore he underwent emergency laparoscopic surgery
with toilette and endoscopic clip placement.
To conclude, considering both promising published outcomes and our large
experience, ICG fluorescence imaging is an effective and not time-consuming
method to evaluate bowel stump perfusion. Whether this will eventually translate
into decreased leakage rates, if compared to standard subjective parameters,
remains to be definitively proven by large and well-conducted randomized trials.
Integrated software to quantify the fluorescent signal could be useful.
The concept of the SLN biopsy is based on the premise that drainage from a solid
organ tumor occurs in an orderly and near linear manner via lymphatics to region-
al LNs. Identification of SLNs during surgery through dedicated tracers would
identify the “proper nodes” for pathologic examination. The promise of an accurate
nodal staging with reduced surgical trauma and morbidity pushed the development
19 ICG Fluorescence: Current and Future Applications 201
Fig. 19.4 Intraoperative subserosal injection of indocyanine green for sentinel lymph node biop-
sy for caecal cancer
of the SLN biopsy for melanoma and breast cancer in 1992 and 1994, respectively.
When compared with breast cancer or melanoma, however, SLN biopsy in
colon cancer is used primarily for improving the accuracy of staging since typical-
ly it does not determine a more limited surgical resection, which remains clearly
investigational because of the still high rate of false-negative SLNs.
Therefore, although its clinical value in colon cancer is far from being fully
established, the aim is to upstage tumors (stage I and II) by providing the patholo-
gist with one to three lymph nodes for detailed immunohistochemical evaluation. A
reliable sentinel node harvesting technique may eventually alter the management in
colon cancer treatment (i.e., adjuvant chemotherapy) where nodal micrometastases
or isolated tumor cells would remain undetected by conventional pathological
examination.
19.4.2 Technique
The dye (1.5–2.5mg) is injected intraoperatively into the subserosa, by the inser-
tion, through the assistant trocar, of a butterfly infusion set (Fig. 19.4). About
10–20minutes after the injection, from one to three fluorescent LNs are displayed
(Fig. 19.5). SLNs are removed selectively with a “berry picking” technique or
marked with clips for ex vivo identification: each SLN will be stained with H&E
and if no metastases are identified, the pathologist will select from each SLN a
superficial, intermediate and deep section, respectively, that will be immunostained
for cytokeratin (AE-1/AE-3).
Removed tissues are labeled according to anatomical location and sent separate-
ly for pathological evaluation.
202 G. Spinoglio et al.
a b
Fig. 19.5 Near-infrared view of sentinel lymph nodes (a). Final white light view of clipped sen-
tinel lymph nodes (b)
19.4.3 Discussion
The prognosis and quality of life of patients affected by colorectal cancer depends
on the extent of the tumor (i.e., stage), its biological features (i.e., differentiation
grade) and characteristics of onset. Complete surgical removal with en-bloc region-
al lymphadenectomy is pivotal for patient oncological outcomes and correct stag-
ing. Nevertheless, concerns about lymphatic spread in rectal cancer still exist.
Examples of involved nodal basins, which are not usually removed through a
standard lymphadenectomy, are represented by lateral pelvic and periaortic LNs in
rectal cancer, that typically have a variable drainage pattern. Although more radical
surgical treatments (such as lateral pelvic lymphadenectomy in Japan) have been
advocated over the years, surgical oncology is evolving toward less aggressive
approaches: in this scenario, LN mapping with ICG might be considered a valid
tool in performing a “tailored” surgery, eventually changing the operative strategy
by performing an extended but selective lymphadenectomy according to the lym-
phatic migration of the fluorescent dye. The aim of LN mapping in rectal cancer is
to potentially improve the staging process without higher morbidity figures if com-
pared to standard resections.
19.5.2 Technique
From 1.5 to 2.5mg of ICG solution are injected endoscopically around the tumor
in the submucosa from 3 to 24 hours before surgery. Within 10–15 minutes, it
reaches the first LN, one or two hours later it reaches the regional LNs where it
remains for about 24–48hours. Operating in NIR light, the LNs, from where the
ICG has been drained, are highlighted during dissection. The LNs are removed en
bloc if present in typical sites, whilst they are removed with a “berry picking” tech-
nique when present in unusual locations after LN mapping (i.e., periaortic, pericav-
al and lateral pelvic nodes, Fig. 19.6).
19.5.3 Discussion
To date, no studies about ICG lymph node mapping in robotic rectal surgery have
been reported.
We have performed 19 LN mappings for rectal cancer, at our institution, to eval-
uate the feasibility of using the da Vinci® NIR-fluorescence imaging system. The
overall detection rate was 78.9%. LN mapping was not successful in 4 out of 19
patients: two patients underwent preoperative radiochemotheraphy, which could
compromise the detection rate by obliteration of lymphatic channels or by scleros-
ing of LNs draining the primary tumor. An overdose of ICG was administered in
204 G. Spinoglio et al.
a b
Fig. 19.6 White light (a) and near-infrared view (b) of interaortocaval lymph nodes following
submucosal indocyanine green injection for rectal cancer
the remaining two cases, making the surgical field appear completely green in the
fluorescent view.
Important caveats in LN mapping are the same as for the SLN biopsy, and
include gentle manipulation of fluorescent LNs to avoid inadvertent staining of the
surrounding tissues and ICG dose.
The ICG fluorescence imaging system during robotic colorectal surgery is a
novel and challenging procedure: it may represent an additional tool to perform LN
mapping and SLN biopsies. Our experience is still at a feasibility stage and we hope
that our on-going study on SLN biopsy could provide useful information.
References
1. Connor S, Garden OJ (2006) Bile duct injury in the era of laparoscopic cholecystectomy. Brit
J Surg 93:158–168
19 ICG Fluorescence: Current and Future Applications 205
2. Rowe CK, Franco FB, Barbosa JA et al (2012) A novel method of evaluating ureteropelvic
junction obstruction: dynamic near infrared fluorescence imaging compared to standard
modalities to assess urinary obstruction in a Swine model. J Urol 188:1978–1985
3. Joseph M, Phillips MR, Farrell TM, Rupp CC (2012) Single incision laparoscopic cholecys-
tectomy is associated with a higher bile duct injury rate: a review and a word of caution. Ann
Surg 256:1–6
4. Berci G, Hunter J, Morgenstern L et al (2013) Laparoscopic cholecystectomy: first, do no harm;
second, take care of bile duct stones. Surgical endoscopy 27:1051–1054
5. Way LW, Stewart L, Gantert W et al (2003) Causes and prevention of laparoscopic bile duct
injuries: analysis of 252 cases from a human factors and cognitive psychology perspective.
Ann Surg 237:460–469
6. Ishizawa T, Bandai Y, Ijichi M (2010) Fluorescent cholangiography illuminating the biliary
tree during laparoscopic cholecystectomy. Brit J Surg 97:1369–1377
7. Ishizawa T, Kaneko J, Inoue Y et al (2011) Application of fluorescent cholangiography to sin-
gle-incision laparoscopic cholecystectomy. Surg Endosc 25:2631–2636
8. Buchs NC, Hagen ME, Pugin F et al (2012) Intra-operative fluorescent cholangiography us-
ing indocyanin green during robotic single site cholecystectomy. Int J Med Robot Comput As-
sist Surg 8:436–440
9. Spinoglio G, Priora F, Bianchi PP et al (2013) Real-time near-infrared (NIR) fluorescent cholan-
giography in single-site robotic cholecystectomy (SSRC): a single-institutional prospective
study. Surg Endosc 27:2156–2162
10. Daskalaki D, Fernandes E, Wang X et al (2014) Indocyanine Green (ICG) fluorescent cholan-
giography during robotic cholecystectomy: results of 184 consecutive cases in a single insti-
tution. Surg Innov. doi: 10.1177/1553350614524839
11. Karliczek A, Harlaar NJ, Zeebregts CJ (2009) Surgeons lack predictive accuracy for anasto-
motic leakage in gastrointestinal surgery. Int J Colorectal Dis 24:569–576
12. Kudszus S, Roesel C, Schachtrupp A, Hoer JJ (2010) Intraoperative laser fluorescence angiog-
raphy in colorectal surgery: a noninvasive analysis to reduce the rate of anastomotic leakage.
Langenbeck’s Arch Surgery 395:1025–1030
13. Sherwinter DA, Gallagher J, Donkar T (2013) Intra-operative transanal near infrared imag-
ing of colorectal anastomotic perfusion: a feasibility study. Colorectal Dis 15:91–96
14. Ris F, Hompes R, Cunningham C et al(2014)Near-infrared (NIR) perfusion angiography in
minimally invasive colorectal surgery. Surg Endosc 28:2221-2226
15. Jafari MD, Lee KH, Halabi WJ et al (2013) The use of indocyanine green fluorescence to as-
sess anastomotic perfusion during robotic assisted laparoscopic rectal surgery. Surg Endosc
27:3003–3008
16. Bae SU, Baek SJ, Hur H et al (2013) Intraoperative near infrared fluorescence imaging in ro-
botic low anterior resection: three case reports. Yonsei Med J 54:1066–1069
17. Hellan M, Spinoglio G, Pigazzi A, Lagares-Garcia JA (2014) The influence of fluorescence
imaging on the location of bowel transection during robotic left-sided colorectal surgery. Surg
Endosc 28:1695–1702
18. Hirche C, Mohr Z, Kneif S et al (2012) Ultrastaging of colon cancer by sentinel node biop-
sy using fluorescence navigation with indocyanine green. Int J Colorectal Dis 27:319–224
19. Cahill RA, Anderson M, Wang LM et al (2012) Near-infrared (NIR) laparoscopy for intraop-
erative lymphatic road-mapping and sentinel node identification during definitive surgical re-
section of early-stage colorectal neoplasia. Surg Endosc 26:197–204
20. Kusano M, Tajima Y, Yamazaki K et al (2008) Sentinel node mapping guided by indocyanine
green fluorescence imaging: a new method for sentinel node navigation surgery in gastroin-
testinal cancer. Digestive surgery 25:103–108
21. Nagata K, Endo S, Hidaka E et al (2006) Laparoscopic sentinel node mapping for colorectal
cancer using infrared ray laparoscopy. Anticancer Res 26:2307–2311
22. Hutteman M, Choi HS, Mieog JS et al (2011) Clinical translation of ex vivo sentinel lymph
node mapping for colorectal cancer using invisible near-infrared fluorescence light. Ann Surg
Oncol 18:1006–1014
206 G. Spinoglio et al.
23. Jewell EL, Huang JJ, Abu-Rustum N et al (2014) Detection of sentinel lymph nodes in min-
imally invasive surgery using indocyanine green and near-infrared fluorescence imaging for
uterine and cervical malignancies. Gynecol Oncol 132:274–277
24. Saha S, Johnston G, Korant A et al (2013) Aberrant drainage of sentinel lymph nodes in colon
cancer and its impact on staging and extent of operation. Am J Surg 205:302–305
25. Yano H, Okada K, Monden T (2003) Adhesion ileus caused by tattoo-marking: unusual com-
plication after laparoscopic surgery for early colorectal cancer. Dis Colon Rectum 46:987
26. Gianom D, Hollinger A, Wirth HP (2003) Intestinal perforation after preoperative colonic tat-
tooing with India ink. Swiss Sur 9:307–310
27. Miyoshi N, Ohue M, Noura S et al (2009) Surgical usefulness of indocyanine green as an al-
ternative to India ink for endoscopic marking. Surg Endosc 23:347–351
Part VI
Miscellany
Splenectomy and Hemisplenectomy
20
Giuseppe Spinoglio, Alessandra Marano,
Luca Matteo Lenti, Francesca Pagliardi,
and Giampaolo Formisano
20.1 Splenectomy
G. Spinoglio ()
Department of General and Oncologic Surgery, “Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 209
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_20, © Springer-Verlag Italia 2015
210 G. Spinoglio et al.
legs and the thorax, with the umbilicus above the pivoting point of the table.
This position widens the space between the costal margin and anterior superior
iliac spine for introducing trocars. A mild reverse-Trendelenburg tilt is applied
to improve exposure of the operative field.
The robot is brought to the table from the left shoulder with a 45° angle and
docked (Fig. 20.1a).
a b
Fig. 20.1 Splenectomy and hemisplenectomy. a OR setup. b Trocar layout. (© 2014 Intuitive
Surgical, Inc.)
20 Splenectomy and Hemisplenectomy 211
20.1.4.1 Step 1 - Access to the Lesser Sac and Splenic Hilum Exposure
Robotic Splenectomy (RS) is performed with an anterior approach (i.e., vessel divi-
sion without a posterior mobilization of the spleen). The stomach is retracted medi-
ally by the assistant’s grasper and a gentle and constant lateral traction on the spleen
is applied with the fenestrated grasper in R3. Gastrocolic and gastrosplenic liga-
ments, within which short gastric vessels lie, are opened with the bipolar forceps in
R2 and the cautery hook/monopolar scissors in R1; clips may be applied if required.
There are two vascular patterns of the splenic pedicle: the magistral type, in
which the main splenic artery enters the splenic hilum as a compact structure
without branching; and the distributed type, which is the most common pattern,
in which multiple arterial branches arise from the main splenic artery 2–3 cm
from the hilum. Indeed, two or three lobar arteries can be found in the majority
of cases (86% and 12%, respectively) [7].
Fig. 20.3 After arterial control, the vein is fully dissected and divided
20.2 Hemisplenectomy
The value in performing partial splenectomy for selected benign diseases has been
well demonstrated in preserving the immune function, particularly in young patients.
20 Splenectomy and Hemisplenectomy 213
The robotic assisted approach may be the most indicated for hemisplenectomy
because it theoretically allows a more meticulous dissection of small arterial branch-
es if compared to laparoscopy. However, to date, different studies have shown
laparoscopic partial splenectomy to be feasible and safe with good clinical results
[8–16] and only few robotic series have directly addressed this issue [3, 5, 6].
Patient and robot positioning, trocar placement and splenic hilum exposure are
the same as previously described for total splenectomy, except for an extensive
division of the more cranial short gastric vessels in the case of lesions located at
the lower pole of the spleen.
The splenic artery is dissected first and passed under with a loop as distally
as possible for bleeding control, if necessary, during hilar and parenchymal dis-
section. The hilar dissection is then undertaken and all arterial branches are
identified and progressively dissected free, depending on the vascular pattern.
Any arterial or venous branches to be preserved are identified, while the other
vessels are clipped proximally and distally or ligated with absorbable sutures
(Fig. 20.4). The cautery hook in R1 guarantees for precise dissection of even
small branches of the splenic vessels.
The demarcation line is then identified in white light or through the integrat-
ed fluorescence imaging system after the intravenous administration of
0.1 mg/kg of indocyanine green (ICG); also arterial vascular branching can be
assessed in real-time during the procedure. The spleen is divided using the
Fig. 20.4 Dissection and ligation of the inferior lobar artery during hemisplenectomy. The demar-
cation line is clearly visible
214 G. Spinoglio et al.
Fig. 20.5 Final view after hemisplenectomy. An absorbable fibrin sealant patch is applied on the
transected splenic parenchyma
• fragile parenchyma and huge spleen volume, which is the main predictor for
conversion to open surgery in the case of laparoscopic splenectomy;
• previous radiotherapy with extensive fibrosis and adhesions;
• hemisplenectomy, for selective small vessel ligation, parenchymal transec-
tion and parenchymal sparing.
In the case of hemisplenectomy, different methods of parenchymal transec-
tion can be used [6]; however, we do not routinely oversew the remaining
splenic parenchyma with pledgets as described by other authors [3]. Moreover,
Vasilescu et al. [6] have demonstrated lower blood loss and vascular dissection
time during robotic subtotal splenectomy if compared to the laparoscopic
approach.
To date, we performed 15 total splenectomies and 5 hemisplenectomies.
There were 9 men (45%) and 11 (55%) women, with a mean age of 45.2 years.
Indications for surgery with total splenectomies included hematological disor-
ders (8 pts), lymphoproloferative diseases (5 pts) and voluminous cysts (2 pts);
the mean spleen size for total splenectomy was 13.2 cm. Hemisplenectomies
were performed for benign focal lesions. The mean operative times were
229 min and 260 min for total and partial splenectomy, respectively. Overall,
there were no conversions to open surgery and no intraoperative complications
or major postoperative complications were observed.
In conclusion, our series and previous studies report RS as safe and feasible;
it provides a valid alternative to laparoscopic splenectomy, especially in more
challenging cases and during hemisplenectomy.
References
1. Delaitre B, Maignien B (1991) Splenectomy by the laparoscopic approach. Report of a case.
Presse Med 20:2263
2. Gamme G, Birch DW, Karmali S (2013) Minimally invasive splenectomy: an update and re-
view. Can J Surg 56:280–285
3. Giulianotti PC, Buchs NC, Addeo P (2011) Robot-assisted partial and total splenectomy. Int
J Med Robot Comput Assist Surg 7:482–488
4. Gelmini R, Franzoni C, Spaziani A (2011) Laparoscopic splenectomy: conventional versus
robotic approach–a comparative study. J Laparoendosc Adv Surg Tech A 21:393–398
5. Vasilescu C, Tudor S, Popa M (2010) Robotic partial splenectomy for hydatid cyst of the spleen.
Langenbeck’s Arch Surg 395:1169–1174
6. Vasilescu C, Stanciulea O, Tudor S (2012) Laparoscopic versus robotic subtotal splenecto-
my in hereditary spherocytosis. Potential advantages and limits of an expensive approach. Surg
Endosc 26:2802–2809
7. Liu DL, Xia S, Xu W et al (1996) Anatomy of vasculature of 850 spleen specimens and its
application in partial splenectomy. Surgery 119:27–33
8. Seims AD, Breckler FD, Hardacker KD, Rescorla FJ (2013) Partial vs total splenectomy in
children with hereditary spherocytosis. Surgery 154:849–853
9. Zhang Y, Chen XM, Sun DL, Yang C (2014) Treatment of hemolymphangioma of the spleen
by laparoscopic partial splenectomy: a case report. W J Surg Oncol 12:60
10. Slater BJ, Chan FP, Davis K, Dutta S (2010) Institutional experience with laparoscopic par-
tial splenectomy for hereditary spherocytosis. J Ped Surg 45:1682–1686
216 G. Spinoglio et al.
11. Morinis J, Dutta S, Blanchette V et al (2008) Laparoscopic partial vs total splenectomy in chil-
dren with hereditary spherocytosis. J Ped Surg 43:1649–1652
12. Hery G, Becmeur F, Mefat L et al (2008) Laparoscopic partial splenectomy: indications and
results of a multicenter retrospective study. Surg Endosc 22:45–49
13. Breitenstein S, Scholz T, Schafer M (2007) Laparoscopic partial splenectomy. J Am Coll Surg
204:179–181
14. Uranues S, Grossman D, Ludwig L, Bergamaschi R (2007) Laparoscopic partial splenecto-
my. Surg Endosc 21:57–60
15. Rescorla FJ, West KW, Engum SA, Grosfeld JL (2007) Laparoscopic splenic procedures in
children: experience in 231 children. Ann Surg 246:683–687
16. Dutta S, Price VE, Blanchette V, Langer JC (2006) A laparoscopic approach to partial splenec-
tomy for children with hereditary spherocytosis. Surg Endosc 20:1719–1724
Transperitoneal Adrenalectomy
21
Giuseppe Spinoglio, Alessandra Marano,
Ferruccio Ravazzoni, Francesca Pagliardi,
and Giampaolo Formisano
G. Spinoglio ()
Department of General and Oncologic Surgery, “Ss. Antonio e Biagio” Hospital,
Alessandria, Italy
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 217
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_21, © Springer-Verlag Italia 2015
218 G. Spinoglio et al.
anatomic perspective. In 2010, the robotic posterior approach was also reported
by Berber et al. [16].
In this chapter, we will focus on the robotic lateral transperitoneal approach
for adrenal gland tumors.
• One 8-mm robotic trocar (R1) is placed in the left flank for Hot Shears™
(monopolar curved scissors), cautery hook or robotic clip applier. We do not
routinely use an ultrasonic scalpel;
• One 8-mm robotic trocar (R2) is placed in the epigastric area, 2–3 cm left
paramedian, for bipolar forceps;
• One 8-mm robotic trocar (R3) is placed cranially to R2, a few centimeters
below the xyphoid process, for the ProGrasp™ forceps or Cadiere forceps;
they are used for traction on the adrenal gland or on the splenic parenchyma
along with the pancreatic tail;
• One 12-mm assistant’s trocar is placed between R1 and the camera ports for
suction/irrigation, clip application (if necessary) or other maneuvers to
achieve optimal exposure of the adrenal gland.
Trocars are placed in a mirror image for right-sided adrenal masses, with R1
in the epigastric area and R2 in the right flank (Fig. 21.3).
We perform a totally robotic approach, though a hybrid technique with
laparoscopic mobilization of the liver or of the splenopancreatic block has been
described.
220 G. Spinoglio et al.
It is not necessary to divide the triangular and coronary ligaments with the robot-
ic approach, whilst this maneuver is required in laparoscopic surgery. The liver
is retracted upward with the fenestrated grasper in R3 and the inferior caval vein
(ICV) is identified.
The peritoneum and the Gerota’s fascia are opened along the right border of
the ICV; the dissection is continued upward along the inferior surface of the liver
and then counterclockwise to gain access to the right margin of the superior renal
pole. Once the peritoneum has been incised, the liver can be further mobilized
and pulled cranially with the grasp in R3 without any risk of capsular tears
because of the stable and constant traction exerted by the robotic instruments.
The inferior part of the right border of the caval vein and the superior aspect
of the right renal vein are fully exposed to gain access to the posterior muscular
plane. We separate the adipose capsule from the posterior abdominal wall open-
ing a wide dihedral angle that represents our first working space and allows the
control of the right middle adrenal vein more easily and safely, as a result of an
anterior and posterior access. If present, upper polar renal arteries must be iden-
tified and preserved. Attention should be paid to a small accessory hepatic vein
(Fig. 21.4).The middle adrenal vein is dissected and clipped with laparoscopic
or robotic nonabsorbable self-locking clips (Hem-o-lok®, Weck-Teleflex Europe
Ltd.) and divided (Fig. 21.5). This maneuver is a crucial step of the operation,
especially in the case of a large and short adrenal vein: the robotic clip applier
should be used in these situations because of the advantages offered by the
EndoWrist® technology.
Fig. 21.4 An accessory hepatic vein (AHV), crossing the upper pole of the adrenal gland is clipped
and divided. ICV, inferior caval vein
21 Transperitoneal Adrenalectomy 221
Fig. 21.5 Medial dissection during right adrenalectomy. MAV, middle adrenal vein; ICV, inferior
caval vein; AHV, accessory hepatic vein
Other small venous or arterial branches are usually controlled with bipolar
energy and are divided. The adrenal gland is dissected completely free along with
the surrounding epinephric fat, with complete respect for oncological principles.
The robot is undocked and the lesion is removed in a specimen retrieval bag
by slightly enlarging a trocar site. A suprapubic minilaparotomy can be per-
formed in the case of larger masses.
The Toldt’s white line is incised starting from the proximal descending colon in
a caudal-to-cephalad direction, thus gaining access to the avascular plane and
exposing the lateral aspect of the Gerota’s fascia. The peritoneum is incised up
to the greater curvature of the stomach behind the spleen by the division of the
splenophrenic ligament; the dissection is carried out progressively upward and
the splenopancreatic block, together with the splenic flexure, is fully mobilized
and can be retracted medially to expose the adrenal gland.
The Gerota’s fascia is then incised along the medial aspect of the adrenal
gland thus entering into the epinephric fat. The left renal vein is identified as a
landmark for medial dissection and isolation of the middle adrenal vein, paying
great attention not to damage the spermatic vessels, as well as the underlying
renal artery. If present, upper polar renal arteries must be identified and pre-
served. The middle adrenal vein is dissected free and clipped with laparoscopic
or robotic nonabsorbable self-locking clips (Hem-o-lok®, Weck-Teleflex Europe
Ltd.) and divided. A small superior adrenal artery, arising from the inferior
phrenic artery, may be found; it can be controlled with bipolar energy or clips.
Once the specimen is completely freed, the procedure continues as already
described.
222 G. Spinoglio et al.
The same group compared the robotic vs. standard laparoscopic technique in
the surgical treatment of large adrenal tumors >5 cm [3]. The operative time and
conversion rate were lower in the robotic group (159 min vs. 187 min, and 4%
vs. 11%, respectively). Additionally, the length of hospital stay was shorter in the
robotic vs. laparoscopic group (1.4 vs. 1.9 days). These results suggest that the
robotic platform could potentially facilitate the resection of large adrenal tumors.
An additional benefit of the robotic approach may be glimpsed in partial
adrenalectomy for patients requiring bilateral adrenalectomy and therefore life-
long steroid supplementation. Few cases, however, have been reported in the lit-
erature [19–21].
Recently, a metanalysis has been published by Brandao et al. [22] to critical-
ly analyze the available evidence of studies comparing laparoscopic and robotic
adrenalectomy. Studies with both the transperitoneal and posterior approach
have been included. Authors showed that the robot-assisted adrenalectomy can
be performed safely and effectively with operative times and complication rates
similar to laparoscopic adrenalectomy and it can provide the potential advantage
of a shorter hospital stay and less blood loss.
To date, we have performed 14 robotic adrenalectomy (unpublished data),
mainly for right-sided adrenal tumors (12 out of 14). The indications for surgery
were five nonfunctional adenomas, four pheochromocytomas, three metastasis
and two Cushing syndromes. Mean BMI was 25 kg/m2. The mean tumor size
was 5.2 cm and the mean operative time was 180 min. Neither intraoperative nor
early postoperative complications were recorded, and there were no conversions
to laparoscopic or open surgery.
Our experience, even though limited, confirms the feasibility and safety of
robotic adrenalectomy with the transperitoneal approach. The technological
properties of the da Vinci® system strongly improve the dissection during adre-
nal surgery, especially in the case of large right-sided tumors (>6 cm) posterior
to the caval vein. The stable platform and the EndoWrist™ technology also
enables the surgeon to safely control and divide large and short right adrenal
veins at the caval confluence.
Future studies involving larger case series and randomized trials with ade-
quate power will determine the exact role of robotics in adrenal surgery with
both the anterior and posterior approach. Moreover, the experience with mini-
mally invasive surgery is limited and therefore controversial in the case of
malignant lesions.
References
1. Gagner M, Lacroix A, Bolte E (1992) Laparoscopic adrenalectomy in Cushing’s syndrome
and pheochromocytoma. N Engl J Med 327:1033
2. Piazza L, Caragliano P, Scardilli M et al (1999) Laparoscopic robot-assisted right adrenalec-
tomy and left ovariectomy (case reports). Chirurgia Italiana 51:465–466
224 G. Spinoglio et al.
22.1 Introduction
R. Garcia-Roca ()
Department of Surgery, Division of Transplantation, University of Illinois Hospital and Health
Sciences System, Chicago, IL, USA
e-mail: [email protected]
G. Spinoglio (Ed), Robotic Surgery: Current Applications and New Trends, 225
Updates in Surgery
DOI: 10.1007/978-88-470-5714-2_22, © Springer-Verlag Italia 2015
226 R. Garcia-Roca et al.
inherent risk of a surgical intervention without a direct health benefit. The avail-
ability of a minimally invasive, laparoscopic surgical technique greatly
enhanced living donation rates by reducing postoperative pain, achieving faster
recovery and minimizing the surgical incisions. The da Vinci® Surgical System
has been used in living donor nephrectomy as a logical extension of the widely
adapted laparoscopic approach. To date there are no randomized trials compar-
ing robotic versus laparoscopic donor nephrectomy. Intuitively, with the expand-
ed range of movements, it simplifies complicated procedures in the presence of
multiple renal arteries or other vascular anomalies. The first worldwide transab-
dominal hand-assisted robotic donor nephrectomy was performed successfully
at the University of Illinois at Chicago Hospital in 2000 [5]. Since then, our
institution has performed over 800 robotic donor nephrectomies with excellent
outcomes.
Potential donors are evaluated by a multidisciplinary transplant team, which
includes a surgeon, pharmacist, social worker, nutritionist, financial counselor
and nurse coordinator. The medical screening involves assessment of immuno-
logic compatibility, the renal function and the current medical status of the
patient to exclude conditions that can be transmitted to the recipient or jeopard-
ize the donor health. The decision regarding which kidney to be harvested is
based on the function and anatomy of the kidneys determined by a 3D recon-
struction from an abdominal CT scan with arterial contrast. Usually the left kid-
ney is procured, due to its favorable anatomy (longer left renal vein) and the
lower complexity of the left nephrectomy, even in the presence of multiple arter-
ies on that side.
For a left nephrectomy, the patient is rolled into the right lateral decubitus posi-
tion with a cushioned beanbag and axillary roll, and the table is flexed. The
patient should be secured to the operative table, because any instability after
docking the robotic system could jeopardize the safety. The table should be in
the Trendelenburg position to avoid injury of the arm and shoulder from the left
robotic arm (Fig. 22.1a).
a b
Fig. 22.1 Living donor nephrectomy. a OR setup. b Trocar placement. (© 2014 Intuitive Surgical,
Inc.)
a b c
Fig. 22.2 a Anterior dissection of the artery. b Posterior exposure of the artery and transection
of lymphatic tissue. c The left robotic arm gently retracts the artery to facilitate introduction
of the stapling device. RV, renal vein; RA, renal artery
lok® clips. Along the upper border of the renal vein, the left adrenal vein is also
dissected and transected between robotic clips. In most of the cases at least one
lumbar vein will be joining the left renal vein and should be similarly transect-
ed. In these cases the articulating skills of the robotic system and the 3D vision
give significant advantage over conventional laparoscopic instruments.
In the dissection plane between the upper pole of the kidney and the adrenal
gland lies the adrenal artery, which originates from the left renal artery and
should be divided between clips whenever present. The upper pole of the kidney
is then fully mobilized close to the renal capsule, leaving behind Gerota’s fascia
and fat. With the kidney in the anatomical position, the artery can be found just
behind the renal vein, the lymphatic tissue surrounding the artery is transected
with the hook forceps.
The previously mobilized ureter is clipped with two robotic Hem-o-lok® clips
at the pelvic rim and sharply transected proximal to them. The posterior attach-
ments of the kidney are divided with the help of the assisting surgeon, exposing
the renal hilum posteriorly. At this point the orientation of the kidney should be
maintained anatomical avoiding twisting of the vessels. This could potentially
result in ischemia of the kidney, or worse, intimal dissection of the renal artery.
Additionally, the robotic surgeon may injure the renal artery if the anatomy is dis-
torted due to torsion. With the kidney in the medial position, the ganglionic and
lymphatic tissue surrounding the renal artery needs to be transected. The vessel
needs to be circumferentially dissected at the level of its origin from the aorta
(Fig. 22.2). If multiple arteries are present, every vessel has to be dissected free
as described. If a lower polar artery, originating from distal abdominal aorta, is
present, this vessel needs to be identified and exposed carefully, because its unin-
tentional injury would deprive the ureter from blood supply.
sected free. The renal artery is localized after posterior mobilization and medial
retraction of the kidney. The right renal artery may bifurcate behind the vena
cava; ligation of lumbar veins may become necessary to medially rotate the IVC
adequately exposing the renal artery. The rest of the procedure follows the same
steps as described for left nephrectomy.
Multiple vessels: The transplant team at our institution prefers to remove the
left kidney even in the presence of multiple arteries. In the hands of experienced
surgeons, reconstruction of multiple vessels has similar outcomes in terms of graft
function and risk of graft loss [6]. The surgeon needs to evaluate the 3D reconstruc-
tions of the computerized tomography to visualize the relationship between the
arteries and renal vein. Lower polar arteries should be preserved in every case, as
they are the sole blood supply of the ureter. Upper polar arteries may be sacrificed
if they are small and do not feed to a large portion of the renal parenchyma. This
can only be assessed once the kidney is removed and flushed in the back table
through the main renal artery, as it will demarcate the area dependent on the upper
polar artery better. Vascular reconstruction is necessary if the area is significant.
Arteries need to be fully dissected individually; this is better done from the
posterior aspect of the kidney. The arterial occlusion is performed in a similar
fashion as when there is a single artery, they are sequentially clipped and tran-
sected from the lower polar to the upper one.
Retroaortic and circumaortic renal vein: The reported incidence of cir-
cumaortic and retroaortic left renal vein is 9 to 14%, this common variant has
been considered a relative contraindication for left donor nephrectomy due to
the potential for inadvertent venous injury. Evaluation of the renal anatomy pre-
operatively via computerized tomography and 3D reconstructions provides
superior details of the renal vascular anatomy including the location, size and
spatial interrelationship of the renal, adrenal, gonadal, and lumbar veins.
Left donor nephrectomy was efficaciously performed in the presence of left
renal venous anomaly, with surgical outcomes comparable to patients with a nor-
mal left renal vein. Importantly, the warm ischemia time and harvested length of
the left renal artery and vein were similar [7]. The two components of the cir-
cumaortic renal vein are meticulously dissected. In most situations, the posteri-
or component of the circumaortic renal vein can be safely ligated, similar to a
larger lumbar vein.
In the presence of a retroaortic renal vein, the aorta limits the posterior dis-
section and the vein will be significantly shortened. This should not compromise
the transplant procedure and in our experience, it has never required backbench
repair or extension patch prior to implantation.
be wider than the operating table. After proper preparation and draping, a 7-cm
midline incision approximately 5cm below the xyphoid process is made for the
placement of the hand access device.
Port placement can be done under direct laparoscopic visualization through
a 5-mm port inserted through the epigastric incision. The position of the laparo-
scopic ports are as follows:
• A 12-mm long laparoscopic port for the 30° robotic scope is inserted just
above the umbilicus;
• Two 7-mm robotic ports are inserted triangulating to the target vessels in the
pelvis, one is placed in the right flank and the other in the left lower quad-
rant (or mirror image in the case of left implantation);
• Another 12-mm assistant laparoscopic long port is placed to the left of the
camera one (Fig. 22.3a).
The patient is placed in a 45° Trendelenburg position and the table is rotated
to the left (for implantation to the right external iliac vessels). This positioning
uses gravity to retract the bowels away from the surgical field. The robot system
is docked into position parallel to the patient’s right leg and slightly diagonal to
the body (Fig. 22.3b).
a b
Fig. 22.3 Kidney transplantation in obese patients. a Trocar placement. b OR setup.
(© 2014 Intuitive Surgical. Inc.)
22 Robotic-assisted Organ Transplantation 233
a b
Fig. 22.4 Dissection of the iliac vein: (a) using vessel loop for retraction; (b) application of
robotic vascular clamps on the iliac vein
a b c
Fig. 22.5 a Placing corner stitch, 5-0 Gore-Tex 12-cm double needle tied ends. b Posterior
wall running suture. c Completion of anastomosis
a b c
Fig. 22.6 a Corner stitch made with 6-0 Gore-Tex. b Posterior wall with running suture. c
Followed with anterior wall running suture
22.3.2 Discussion
In the last four years, we have used this technique and performed more than 80
robotic-assisted kidney transplants in obese recipients. We include any patient
with a BMI >30kg/m2, without an upper limit. The mean BMI of the group was
45kg/m2. We have not observed any SSI within the first 30 post-transplant days.
High immunologic risk or multiple previous surgeries were not considered con-
traindications for the procedure. The only exclusion criteria were severe athero-
sclerosis in the iliac vessels of the recipient or in the graft vessels of kidney
coming from a deceased donor. To evaluate our initial hypothesis of improving
wound complications with the robotic approach, we performed a case-control
study comparing our first 28 robotic-assisted kidney transplants to a matched
retrospective cohort of obese recipients who underwent kidney transplantation
22 Robotic-assisted Organ Transplantation 235
section of the splenic vein is performed at the confluence with the mesenteric
vein. The splenic artery was dissected from the celiac trunk.
The parenchymal transection was performed at the junction between the
body and the head of the pancreas with a robotic UltraCision device. After sys-
temic heparinization, the splenic artery was stapled at the take-off with a TA vas-
cular stapler and then sharply divided with a robotic scissor. Finally the splenic
vein was divided with a GIA vascular stapler and the pancreas immediately
retrieved through the midline incision.
During the recipient’s operation, the renal graft was anastomosed to the left
external iliac vessels and the pancreas graft to the right external iliac artery and
vein. The exocrine secretions were drained to the bladder via pancreaticocys-
tostomy. The allografts became functional immediately on transplantation into
the recipient.
The potential postoperative complications for the donor include pancreatitis,
pancreatic leak, pseudocyst formation and splenic infarction. But a more devas-
tating long-term complication is the development of diabetes in the donor. In
2005, Tan et al. [11] reported their first initial experience with five hand-assist-
ed laparoscopic donor pancreatectomies. The donor and recipient survival rate
was 100% at up to 3 years of follow-up. None of these donors have had compli-
cations or required antidiabetic medications. More recent publications acknowl-
edge the risk of diabetes in the donor but support the procedure in selected cases
of highly-sensitized recipients, owing to long waiting times for deceased pan-
creas donors [12].
In reference to the future of minimally invasive pancreas transplantation,
Boggi et al. reported the first three whole pancreas transplants performed with
the assistance of the da Vinci® Surgical System [3]. The mean warm and cold
ischemia times were 30 minutes and 7.3 hours, respectively. Hemorrhage that
required intervention occurred in the third pancreas graft; however, it was con-
trolled successfully using the robotic system. None of the recipients needed
blood transfusions. Their experience proves the feasibility of robotic-assisted
laparoscopic surgery in pancreas transplantation. However, further studies are
necessary and larger series for it to become an alternative approach to the con-
ventional open technique.
ral center for robotic liver procedures, such experience in combination with liv-
ing donation expertise made possible the first robotic right donor hepatectomy.
Our first case was conducted in a 53-year-old healthy man, donating the right
lobe to his brother who had hepatocellular carcinoma, which was complicating
his hepatitis C cirrhosis [4]. The entire procedure was performed using the da
Vinci® Robotic Surgical System. After the laparoscopic trocars were placed and
the robotic system was installed, a cholecystectomy was performed. The hepat-
ic artery and the right portal vein were dissected free, followed by isolation and
transection of the right hepatic duct. The right lobe was completely mobilized
exposing the retrohepatic cava upward from the caudate lobe. Consequently,
parenchyma transection was performed with the robotic Harmonic scalpel while
preserving vascularization of the lobe. The vascular transsection was performed
using Endo-GIATM vascular stapler, first to the right hepatic artery, then the right
portal vein and lastly the right hepatic vein. The graft was rapidly removed
through a lower abdominal incision. The operative time was approximately
8 hours, with an estimated blood loss of 350 mL and no transfusion require-
ments. His postoperative course was uneventful and he was discharged home on
the fifth postoperative day.
Besides facilitating the vascular and biliary dissection, the robotic technolo-
gy also offers the possibility of identifying biliary structures using Indocyanine
Green, minimizing dissection around the biliary tree, and thus reducing the
chances of ischemic injury of the biliary anastomosis [13]. The lower incision
provides better pulmonary care and minimizes analgesic requirements, thus
facilitating early mobilization. This incision decreases the pain and risk of pul-
monary complications associated with an upper midline incision. However, liv-
ing donor hepatectomy should only be undertaken by experienced surgical
teams, due to the significant morbidity and mortality associated to this proce-
dure for donor and recipient.
References
1. Giulianotti P, Gorodner V, Sbrana F et al (2010) Robotic transabdominal kidney transplanta-
tion in a morbidly obese patient. Am J Transplant 10:1478–1478
2. Boggi U, Vistoli F, Signori S et al (2011) Robotic renal transplantation: first European case.
Transpl Int 24:213–218
3. Boggi U, Signori S, Vistoli F et al (2012) Laparoscopic robot-assisted pancreas transplanta-
tion: first world experience. Transplantation 93:201–206
4. Giulianotti PC, Tzevetanov I, Jeon H et al (2012) Robot-assisted right lobe donor hepatecto-
my. Transpl Int 25:e5–9
5. Horgan S, Vanuno D, Sileri P et al (2002) Robotic-assisted laparoscopic donor nephrectomy
for kidney transplantation. Transplantation 73:1474–1479
6. Kay MD, Brook N, Kaushir M et al (2006) Comparison of right and left laparoscopic live donor
nephrectomy. BJU Int 98:843–844
7. Lin CHSteinberg AP, Ramani AP et al (2004) Laparoscopic live donor nephrectomy in the
presence of circumaortic or retroaortic left renal vein. J Urol 171:44–46
238 R. Garcia-Roca et al.
8. Aalten J, Christriaans MH, de Filter H et al (2006) The influence of obesity on short- and long-
term graft and patient survival after renal transplantation. Transpl Int 19:901–907
9. Lynch RJ, Ranney DN, Shijine C et al (2009) Obesity, surgical site infection, and outcome
following renal transplantation. Ann Surg 250:1014–1020
10. Oberholzer J, Giulianotti P, Danielson KK et al (2013) Minimally invasive robotic kidney trans-
plantation for obese patients previously denied access to transplantation. Am J Transplant
13:721–728
11. Tan M, Kandaswamy R, Sutherland DE, Gruessner RW (2005) Laparoscopic donor distal pan-
createctomy for living donor pancreas and pancreas-kidney transplantation. Am J Transplant
5:1966–1970
12. Sutherland DE, Radosevich D, Gruessner RW et al (2012) Pushing the envelope: living donor
pancreas transplantation. Curr Opin Organ Transplant 17:106–115
13. Rotellar F, Pardo F, Benito A (2013) Totally laparoscopic right-lobe hepatectomy for adult liv-
ing donor liver transplantation: useful strategies to enhance safety. Am J Transplant
13:3269–3273
14. Samstein B, Cherqui D, Rotellar F (2013) Totally laparoscopic full left hepatectomy for liv-
ing donor liver transplantation in adolescents and adults. Am J Transplant 13:2462–2456