Early Experience with Totally Robotic Roux-en-Y Gastric Bypass for Morbid Obesity

Abstract

Background:

Laparoscopic Roux-en-Y gastric bypass (RYGBP) for morbid obesity is a challenging operation. The application of robotic techniques has been shown to ease the technical difficulties and reduce perioperative morbidity, mainly because it facilitates the construction of the gastrojejunal anastomosis (GJ). Robotic laparoscopic RYGBP (LRYGBP) has been reported either as totally robotic with manual suturing of the GJ or as robotically assisted with the use of the robot only for the construction of the GJ. A totally robotic LRYGBP with a combined stapled and manual GJ has never been reported.

Patients and Methods:

Nine consecutive patients underwent totally robotic LRYGBP. The GJ was fashioned with a combination of the linear stapler and manual suturing.

Results:

Mean preoperative body mass index was 45.3 ± 4.7 kg/m². In 1 case, we had to undock the Da Vinci Surgical System at the time of the jejunojejunostomy due to unfavorable ergonomics. Mean time to dock the robot was 16.3 ± 3.3 minutes, whereas mean total operative time was 197.2 ± 12.3 minutes. Immediate postoperative morbidity and mortality equaled zero. One patient developed a stenosis of the GJ amenable to endoscopic dilatation. The mean excess weight loss rate 1-year postoperative was 79% ± 15%.

Conclusion:

Totally robotic LRYGBP can duplicate precisely any conventional technique without any compromise in operative time, short- or long-term results.

Introduction

Morbid obesity represents an endemic medical problem of the modern era with serious social and financial implications. It has been estimated that >5% of the U.S. population fulfill the criteria for surgical therapy of morbid obesity. Surgery represents currently the only viable option for a reasonably safe and long-term sustainable weight loss and it provides a frequent regression of the concomitant medical problems that obese persons face.¹ From the various types of bariatric surgery, Roux-en-Y gastric bypass (RYGBP) has been more extensively tested in the clinical practice during the last four decades since its introduction as an open procedure and mainly during the last decade after it has been popularized as a laparoscopic operation. It has proved to be a very reliable means of treatment of morbid obesity, with a long-term weight loss rate of 65%² and an impressively high rate of cure of the associated comorbidities.³ The total number of the yearly performed RYGBPs worldwide has grown from 95.257 operations in 2003 to 168.597 operations in 2008. This number represents approximately half of the total number of bariatric operations.⁴ Despite these data, it has been estimated that the number of the yearly performed bariatric operations does not exceed the 1% of the respective demand. The reason for this huge gap is, at least partly, that laparoscopic RYGBP (LRYGBP) is an extremely difficult procedure, with a notoriously steep learning curve that discourages a lot of surgeons from trying to get familiar with it. The skills that are required to master this procedure include the ability to operate in multiple abdominal quadrants, two-hand tissue manipulation, stapling techniques, suturing, and intracorporeal knot-tying. These surgical tasks become even more difficult due to the inherent limitations of laparoscopy, especially in obese patients. These limitations include the two-dimensional view, the counterintuitive movement of instruments when the surgical field changes between the various abdominal compartments, and the abdominal wall torque that causes fatigue to the surgeon and limits the range and precision of movements of the instruments.

The main motivation behind the development of advanced robotic equipment was to overcome all these problems, thus facilitating the performance of challenging operations such as LRYGBP and making them safer and more readily available. Indeed, the three-dimensional visualization, the increased freedom and precision of movements of the instruments, and the overall improved ergonomics have been shown to facilitate the performance of this technically demanding operation, thus modifying its steep learning curve.⁵ We started performing robotic LRYGBP in November 2008 for the first time in Greece. Our plan was to perform a totally robotic LRYGBP with a technique identical to the respective of our standard LRYGBP and we report herein our short experience.

Patients and Methods

Background data

Between December 2008 and May 2009, 9 consecutive patients underwent totally robotic LRYGBP. Our plan was to adapt the robotic technique so that the robotic operation will at the end of the day be identical with the respective of our conventional LRYGBP. Our standard technique involves the construction of the gastrojejunal anastomosis (GJ) with the use of the linear stapler and the manual suturing of the enterotomies.⁶ It starts with the mobilization of the angle of His and continues with the creation of the gastric pouch. Subsequently, the greater omentum is divided and the jejunum is transected 100 cm distally to the ligament of Treitz. The GJ is performed thereafter with the use of the 45 mm linear stapler (Ethicon ENDOGIA 45 mm) for the posterior wall, whereas the anterior wall, consisting of the gastrotomy and the enterotomy, is sutured manually in one layer with one or two continuous 3-0 Vicryl sutures. Once the upper anastomosis is checked for its water-tightness and the nasogastric tube lies in place, the jejunojejunal anastomosis (JJ) is fashioned 150 cm distally from the GJ in a side-to-side way with the use of the Echelon 60 mm linear stapler. The enterotomies are sutured once again manually in one layer with a running 3-0 Vicryl suture. Last but not the least, the mesenteric and Petersen's defects are manually sutured.

General arrangements and set-up for the robotic RYGBP

The surgical team consisted of a senior attending as the console surgeon, a certified surgical fellow as the scrubbed table surgeon, and one resident and was kept the same throughout the whole of this study. Both the console and the patient-side surgeon were trained and certified by Intuitive for the use of the Da Vinci Surgical System. Since it is extremely time consuming and creates a lot of upset in the operating room to undock and change the position of the robotic system, it is absolutely essential for a smooth operation to arrange the patient, the scrub-nurse table, the anesthesiologists, the robotic cart, the robotic arms, and the trocars on the abdominal wall ideally, so that the coordination will remain excellent throughout the whole procedure. To accommodate the numerous bulky devices in the restricted area of an operating room, especially around the head of the patient, we had to move the operating table toward one end of the room (caudally) and we changed its longitudinal axis 15° in relation to the respective axis of the operating theater. This manipulation permitted the accommodation of both the bulky patient cart of the robot above the left shoulder of the patient as close to the table as possible and the anesthesia cart over the right shoulder (Fig. 1). The anesthesiologists occupied the right side of the head of the patient. The table was tilted in a reverse 20° Trendelenburg position before the docking of the robot. The patient's right arm was extended to provide venous access for the anesthesiologists, whereas the left arm remained attached to the side of the trunk. Both the patient and the robot were draped by the table surgeon with the assistance of a scrub-nurse. The table-side surgeon was standing between the patient's legs for the first 2 cases, as during a conventional LRYGBP. We stopped abducting the patient's legs and the table surgeon was moved to the patient's side thereafter, as his main task was to dock and undock the robotic arms and insert the stapler, and this could be easily done from the patient's side. The scrub-nurse was situated beside the patient's right leg.

FIG. 1.

General set-up and position of trocars for the performance of Da Vinci RYGBP. The close to the umbilicus trocar corresponds to the camera. Left working trocar lies between the camera and the right subcostal arch, and right working trocar lies almost symmetrically on the left side.

Robotic port placement and cannulation technique

Pneumoperitoneum was established with the Verress technique in the midline, 5 cm above the umbilicus and, subsequently, a 12 mm port was inserted for the camera (C). The right 12 mm working port was positioned at the epigastrium, 5 cm toward the left of the midline of the patient, exactly as for the standard procedure, in a distance 8 cm or more from the camera port (right working trocar [RW]). The left 12 mm working port was located on the right side of the patient, a little bit lower than for the standard LRYGBP, so that it would also be at least 8 cm far away from the camera port and the contained angle between the three ports would be ∼120° (left working trocar [LW]). The 8 mm metallic robotic ports were inserted through the standard, disposable 12 mm trocars. This double-cannulation technique was used, because it greatly facilitates the retraction of the robotic arms with the robotic cannulas still attached, while the standard 12 mm trocars stay in place to serve the insertion of the staplers. The existence of the standard laparoscopic trocars on the abdominal wall carries an extra advantage, as they provide immediate access in the abdominal cavity, if an event occurs that necessitates the urgent conversion to a standard laparoscopic approach. Sufficient distance between the trocars and width of the contained angle are absolute prerequisites, to avoid collision of the robotic arms that would hamper the advancement of the procedure. Two further 12 mm trocars were inserted, one at the right anterior axillary line (RL), 2 cm below the subcostal arch for the liver retractor and one at the symmetrical site on the left side (LL) for the second assistant to retract the gastrosplenic ligament. For the first 2 patients in this series, we inserted a sixth trocar also at the umbilicus for the part of the operation below the transverse mesocolon, exactly as for a conventional LRYGBP; but soon, we abandoned this practice, as there was no real need to move the camera up and down between the two midline trocars to see clearly either the angle of His or the ligament of Treitz and the jejunal loops. All the trocars were inserted under the guidance of the standard 30° laparoscopic camera and the robotic system approached the patient and was installed (“docked”) afterward. The robotic cart was positioned over the patient's left shoulder so that the axis between the cart and the camera port would make an angle 10° with the patient's midline. Since our intention was to perform a totally robotic LRYGBP, the robot was docked from the very beginning of the operation and no dissection was carried out before the docking of the Da Vinci. The robotic cannulas and instruments were inserted in the abdominal cavity also with the guidance of the laparoscopic camera and the robotic camera was docked last, when everything else had already been set up.

Surgical technique

The operation started with the opening of a window in the lesser gastric curvature with the use of the modified for the Da Vinci harmonic scalpel. The console surgeon abducted the mesogastrium with a Cadiere forceps, whereas the patient-side surgeon abducted the stomach with a standard laparoscopic grasping forceps. We always try to dissect very close to the gastric wall so as to be sure that the trunk of the vagus nerve is preserved to avoid any disturbances of the gastric emptying. Once the window was opened and access to the lesser epiploic sac was gained, the left robotic arm was undocked from the (LW) for the insertion of the stapler and division of the stomach started. Usually, after the first firing of the stapler, the left robotic arm was docked once again and the rest of the firings were done through the RW. All staple lines were regularly reinforced with synthetic reinforcing material (SEAMGUARD; GORE Inc.). Once division of the stomach for the creation of the gastric pouch was completed, attention was diverted to the creation of the jejunal loops. Since we favor the construction of the GJ in an antecolic way, we always divide the greater omentum to shorten the distance between the upper and the lower part of the abdomen. The table surgeon grasps the free end of the major omentum left from its midline, and the console surgeon grasps with a Cadiere forceps exactly opposite and starts the division with the harmonic curved shears. Once the transverse colon is reached, transverse mesocolon is grasped by the table surgeon and the console surgeon identifies the first jejunal loop at the root of the mesentery and starts measuring 100 cm for the biliopancreatic limb. The Echelon 60 mm linear stapler is armed with a white cartridge and the jejunum is divided. The distal part of the jejunum is grasped and raised toward the gastric pouch and is anchored there with two Vicryl 3-0 stay-stitches. With the use of the harmonic curved shears, a gastrotomy and an enterotomy are created. The left robotic arm is undocked and the table surgeon inserts the 45 mm EndoGIA linear stapler for the construction of the posterior wall of the anastomosis. Subsequently, the oval gap of the attached gastrotomy and enterotomy is closed with one or two Vicryl 3-0 running sutures. Once the anastomosis is completed, it is checked for its water-tightness with instillation of diluted Povidone solution through the nasogastric tube, and attention is drawn to the construction of the Roux limb. Our preferred length for the alimentary loop is 150 cm, to enhance the malabsorptive part of the procedure. Once the appropriate jejunal loop is identified, it is attached to the proximal loop of the previously divided jejunum with two stay stitches, and the enterotomies for a side-to-side stapled anastomosis are opened. The right robotic arm is undocked once again and the Echelon 60 mm linear stapler armed with a white cartridge is inserted for the posterior wall of the anastomosis. Subsequently, the enterotomies are sutured with one or two Vicryl 3-0 running sutures. Finally, the mesenteric and the Petersen defects are sutured and the operation is completed.

Results

Patients' dataset comprised of 8 women and 1 man with a mean age of 40 ± 11.6 years. Mean preoperative body mass index was 45.3 ± 4.7 kg/m² (40–52). Preoperative comorbidities included 3 patients with diabetes mellitus, 1 with sleep apnea syndrome, 1 with hypertension, and 2 with significant musclesceletal problems.

Mean time to dock the robot was 16.3 ± 3.3 minutes (10–20), whereas mean total operative time was 197.2 ± 12.3 minutes (180–220) (Table 1). All but one operation ended robotically. In 1 case, we had to undock the Da Vinci Surgical System at the time of the JJ due to unfavorable ergonomics. The robot was undocked and, subsequently, the anastomosis was constructed and the ensuing mesenteric defects were sutured in the conventional laparoscopic way, after an additional 5 mm trocar was inserted in the right lateral abdominal wall. In 1 additional patient, the greater omentum was anchored to the anterior abdominal wall due to adhesions. We undocked the Da Vinci, cut the adhesions laparoscopically, and then re-docked the robot for the rest of the operation.

Table 1.

Time Parameters for Robotic LRYGBP ^a

Docking time (min)	16.3 ± 3.3 (10–20)
Net operative time (min)	180.9 ± 11.4 (165–200)
Total operative time (min)	197.2 ± 12.3 (180–220)
Total operative time/BMI (min/kg/m²)	4.4 ± 0.5 (3.6–5.3)
Net operative time/BMI (min/kg/m²)	4.0 ± 0.5 (3.2–4.9)

Mean ± standard deviation (minimum–maximum).

BMI, body mass index.

Immediate postoperative morbidity and mortality equaled zero. All patients were discharged on the fourth postoperative day according to our protocol. There were no readmissions in the immediate postoperative period. One patient developed a stenosis of the GJ and was submitted to an endoscopic dilatation twice.

Concerning the long-term results, rates of excess weight loss can be seen in Table 2. The mean excess weight loss rate for these 9 patients 1-year post-operative (post-op) was 79% ± 15%.

Table 2.

Percent Excess Body Weight Loss After Robotic LRYGBP (n = 9)

1 month	27.5 ± 1.8 (26.3–29.6)
3 months	44.4 ± 10.7 (30–57.5)
6 months	66 ± 16.8 (37.5–79.6)
1 year	79 ± 15 (57.8–100)

Discussion

Obesity represents a serious public health problem. Its incidence keeps on rising and although the number of the yearly performed bariatric operations is also rising, it is still far from meeting the demand.⁷ Roux-en-Y gastric bypass was introduced in 1967 by Mason and Ito⁸ and has been successfully tested all these years concerning its efficacy as well as its short- and long-term safety as a definitive surgical treatment for morbid obesity. Especially after its fashioning as a minimally invasive operation by Wittgrove et al.,⁹ the operation became better tolerated by the patients as a result of the less post-op pain and the far less frequent post-op pulmonary and wound problems. Despite this, there has always been a great obstacle that hampered the exponential rise in the availability of this operation: LRYGBP is technically a very demanding operation. It requires advanced laparoscopic skills such as difficult dissection in various abdominal compartments, both above and below the transverse mesocolon, suturing gastrointestinal anastomoses, and intracorporeal knot-tying. The price for the better tolerance on the patient's side was high on the surgeon's side, who has to fight against the torque of the very thick abdominal wall, in a less than ideal ergonomic position, for long hours, with less than ideally ergonomic tools to accomplish a sequence of very difficult surgical tasks.

With the introduction of the Da Vinci robotic surgical system, it seemed that the majority of these problems could be overcome. Indeed, the robot allows the surgeon to manipulate the instruments sitting at the console in a very comfortable position, provides an excellent three-dimensional view; its arms do not face problems of fatigue because of the torque of the abdominal wall, they have more degrees of freedom mimicking the movements of the human wrist, and the computer-controlled system filters the normal human hand tremor. Suddenly, the surgeon who is familiar with the Da Vinci becomes a never tired, ambidextrous, very skilled laparoscopic surgeon. It has been estimated that the steep learning curve of the LRYGBP is shortened from 100¹⁰ to 35⁵ cases as a result of all these.

We introduced robotic LRYGBP into our practice in November 2008. Our intention was to perform the operation with the Da Vinci exactly similar to our conventional LRYGBP to take the maximum advantage from the use of the robot without any compromise in our technique. Our small experience proved that a laparoscopic surgeon familiar with the standard LRYGBP can without big difficulty duplicate his standard technique with the Da Vinci. All parts of the procedure can be performed with equal or superior precision and safety. Ergonomic problems that necessitated the change of plans were encountered only once. In this case, the position of the two working robotic arms was not favorable for the construction of the JJ in the left lateral abdominal compartment. In a standard laparoscopic case for the construction of the JJ, the surgeon moves to the right side of the patient. Then he has a lot of options concerning the trocars he will use. Usually, he can use the right lateral trocar, previously used for the liver retraction, or insert a new 5 mm trocar in the right lateral abdominal wall to use it as his LW and the until-then LW as his RW for the suturing of the JJ or he can move the camera to the RW and vice versa. All these with the robot in place are extremely time consuming and technically difficult to accomplish. Thankfully, under normal circumstances, they are also needless. In the single case that the power and the freedom of movements of the robotic arms as well as the excellent three-dimensional view of the robotic camera were not enough to overcome the unfavorable position of the trocars, instead of undock and reinstall the robot in a better position, we decided to undock, insert another 5 mm conventional trocar, and complete the operation in the standard way.

Potential disadvantages of the use of the robot include the absence of haptic sense. This absence of the feeling of how much pressure is enforced on the tissues has been reported to increase the risk for intestinal tears during intestinal manipulation.¹¹ We did not encounter such a problem but any user of the Da Vinci must bear this in mind. Another issue that is commonly raised, mainly by surgeons not familiar with the Da Vinci, is a potential increase in the operative time due to the learning curve of the new technique and the time for the set up of the system. After having used the Da Vinci for 19 robotic sleeve gastrectomies¹² and the 9 LRYGBPs reported herein, we realized that time is not an issue. With accumulated experience, the docking time pretty soon fell well below 15 minutes. Other authors have reported docking times even shorter than 6 minutes.¹³ We also noticed that the operation per se was not more time consuming or was even shorter than our standard LRYGBP without any compromise concerning the long-term results (data not shown). The same observation for LRYGBP has been described by Mohr and colleagues⁵ after 100 robotic LRYGBPs, when they reported median operative times significantly lower for the robotic procedures as well as lower ratios of procedure time to body mass index.

The main use of the robot has always been considered the simplification of the GJ and, thus, the reduction of the morbidity associated with it. The majority of the authors who have reported their experience with robotic LRYGBP dock the robot only for the construction of this anastomosis and perform the rest of the operation in the standard way.¹⁴ All of them focus on the impressive decrease of the post-op leakage rate, almost to the point of nullification.¹⁵ The precision of the three-dimensional view, the accuracy and freedom of movements the robotic tools provide, and the equal dexterity for both arms transform such a big undertake as the GJ in a morbidly obese person to a straight forward task. It is this simplification that has contributed the most to the moderation of the steep learning curve of LRYGBP from 100 to 35 cases when it is performed robotically.⁵ We decided to use the robot not only for the construction of the GJ but we rather opted to perform a totally robotic LRYGBP similar to that performed by Mohr and colleagues.⁵ It turned out to be that in a very short period of time we were able to duplicate our standard technique for LRYGBP with the same good results in an acceptable or even shorter time frame.

The main drawback of this report is, of course, the small number of its patients. If we try to summarize any issues of special interest from this small experience, first of all we have to mention that this is the first report of this particular variation of surgical technique for the totally robotic LRYGBP. Various authors have reported either robotically assisted LRYGBP, where the robot is used only for the manual suturing of the GJ and the rest of the procedure is done in the standard way, or totally robotic LRYGBP with a different sequence of events and a different type of GJ. The immediately following conclusion is that experienced laparoscopic surgeons can precisely duplicate their conventional laparoscopic techniques without the need for any changes or compromises to what they consider as optimum. In addition to this, the adaptation from the conventional to the robotic technique did not involve a long, steep learning curve. Docking time did not prolong the total operative time significantly and had the tendency to become shorter and shorter with accumulated experience. The net and total operative times for robotic LRYGBP were equal or even shorter than the respective of conventional LRYGBP. The postoperative morbidity and long-term efficacy were also comparable to our historical data of LRYGBP. The reason for these observations might be the ease with which the console surgeon can accomplish such difficult tasks as manual suturing and knot-tying during the various parts of this procedure, due to the better three-dimensional view and the more ergonomic tools. The possible shortening of the precious operative time and the reported minimization of the leakage rate from the GJ and its associated morbidity are so far the claimed advantages of the robotic LRYGBP. In future studies, these have to be weighed against the increased cost for the initial purchase and the maintenance cost of the machinery, to justify its widespread use.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Steinbrook

. Surgery for severe obesity. N Engl J Med, 2004; 350:1075–1079.

Schauer

. Gastric bypass for severe obesity: Approaches and outcomes. Surg Obes Relat Dis, 2005; 1:297–300.

Pories

, Swanson

, MacDonald

, Long

, Morris

, Brown

, Barakat

, deRamon

, Israel

, Dolezal

et al. Who would have thought it? An operation proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg, 1995; 222:339–350, discussion 350–352.

Buchwald

, Oien

. Metabolic/bariatric surgery worldwide 2008. Obes Surg, 2009; 19:1605–1611.

Catherine

, Mohr

MSME

, Geoffrey

, Nadzam

, Myriam

, Curet

. Totally robotic Roux-en-Y gastric bypass. Arch Surg, 2005; 140:779–786.

Tichansky

, DeMaria

. Laparoscopic Roux-en-Y gastric bypass: Linear stapled technique. Inabnet

, Demaria

, Ikkramuddin

. Laparoscopic Bariatric Surgery. Philadelphia: Lippincott, Williams & Wilkins, 2005; 102–105.

Flegal

, Carroll

, Ogden

, Johnson

. Prevalence and trends in obesity among US adults, 1999–2000. JAMA, 2002; 288:1723–1727.

Mason

, Ito

. Gastric bypass in obesity. Surg Clin North Am, 1967; 47:1345–1351.

Wittgrove

, Clark

, Schubert

. Laparoscopic gastric bypass, Roux en-Y: Technique and results in 75 patients with 3–30 months follow-up. Obes Surg, 1996; 6:500–504.

10.

Schauer

, Ikramuddin

, Hamad

, Gourash

. The learning curve for laparoscopic Roux-en-Y gastric bypass is 100 cases. Surg Endosc, 2003; 17:212–215.

11.

Hubens

, Balliu

, Ruppert

, Gypen

, Van Tu

, Vaneerdeweg

. Roux-en-Y gastric bypass procedure performed with the da Vinci robot system: Is it worth it? Surg Endosc, 2008; 22:1690–1696.

12.

Diamantis

, Alexandrou

, Nikiteas

, Giannopoulos

, Papalambros

. Initial experience with robotic sleeve gastrectomy for morbid obesity. Obes Surg, Aug. 5, 2010. [Epub ahead of print]

13.

Jacobsen

, Berger

, Horgan

. The role robotic surgery in morbid obesity. J Laparoendosc Adv Surg Tech A, 2003; 13:229–283.

14.

Parini

, Fabozzi

, Brachet Contul

, Millo

, Loffredo

, Allieta

, Nardi

Jr. , Lale-Murix

. Laparoscopic gastric bypass performed with the Da Vinci intuitive robotic system: Preliminary experience. Surg Endosc, 2006; 20:1851–1857.

15.

Snyder

, Wilson

, Leong

, Klein

, Wilson

. Robotic assisted Roux-en-Y gastric bypass: Minimizing morbidity and mortality. Obes Surg, 2010; 20:265–270.