Key words: Bariatric surgery â GBP â Laparoscopic gastric bypass â LGBP â Robot-assisted gastric bypass â Robotic surgery. During the past 15 years, we ...
Surg Endosc (2006) 20: 1851–1857 DOI: 10.1007/s00464-004-9146-9 Ó Springer Science+Business Media, Inc. 2006
Laparoscopic gastric bypass performed with the Da Vinci Intuitive Robotic System: preliminary experience U. Parini, M. Fabozzi, R. Brachet Contul, P. Millo, A. Loffredo, R. Allieta, M. Nardi Jr., E. Lale-Murix Department of General Surgery, Valle dÕAosta Regional Hospital, Viale Ginevra 3, 11100 Aosta, Italy Received: 24 June 2004/Accepted: 25 November 2004/Online publication: 23 October 2006
Abstract Background: This study aimed to analyze retrospectively the authorsÕ preliminary experience using the Da Vinci Intuitive Robotic System for gastric bypass in managing morbid obesity, and to determine its efficacy and safety in relation to other standardized laparoscopic surgical techniques. Methods: From October 2000 to March 2004 the authors performed 146 laparoscopic gastric bypasses, 17 of which were robot assisted using the Da Vinci Intuitive Robotic System. The last patients were 7 men and 10 women with a mean age of 44 years. The mean weight was 139 kg, and the mean body mass index (BMI) was 49.8 kg/m2 at first postoperative recovery. The mean excess body weight (EBW) was 131%. Follow-up assessment, performed at months 1, 3, 6, and 12, then yearly thereafter, included evaluation of the variations in BMI and the percentage of excess body weight loss (EBWL%). All the patients were informed of the risks inherent with each surgical procedure as well as the potential benefits. Results: The mean operative time was 201 min (range, 90–300 min). No intraoperative complications and no conversion occurred in this series. The mean hospital stay was 9 days (range, 6–18 days). The patients in this series experienced a normal postoperative course without anastomotic complications. The mortality rate was zero. No robot-related complications were noted. The analysis of follow-up assessment at months 1, 3, 6, and 12 showed a progressive decrease in BMI and an increment of EBWL%. Conclusions: The authorsÕ early experience with robotic surgery suggests that it is safe and could be an effective alternative to conventional laparoscopic surgery. The authors believe that robotic surgery, with its ability to restore the hand–eye coordination and three-dimensional
Correspondence to: M. Fabozzi
view lost in laparoscopic surgery, could allow complex procedures to be performed with greater precision and better results. Key words: Bariatric surgery — GBP — Laparoscopic gastric bypass — LGBP — Robot-assisted gastric bypass — Robotic surgery
During the past 15 years, we have witnessed the development of video-assisted surgery, characterized by minimal invasiveness, small laparotomy, less pain, better cosmetic results, and faster postoperative recovery. These procedures have influenced the techniques in every specialty of surgery, leading to the replacement of conventional procedures and stimulating surgeons to find new approaches and solutions. In this scope, about 5 years ago [1], was born the era of robotic surgery, which uses more advanced technologies and represents a further step toward the development of minimally invasive surgery. Our institution is in the forefront of the laparoscopic surgery era that began in 1991. From that time, we have used laparoscopy to perform cholecystectomies, Nissen fundoplications, Toupet fundoplications, colectomies with total mesorectal excision, splenectomies, total and subtotal gastrectomies, duodenocephalopancreatectomies, esophagectomies, hepatectomies, nephrectomies, hysterectomies, adrenalectomies, lysis of adherences, and bariatric surgery. Specifically, we have performed 99 biliopancreatic diversions in open surgery, 59 nonadjustable laparoscopic gastric bandings according to Molina, 5 Swedish laparoscopic adjustable silicon gastric bandings, 171 laparoscopic vertical banded gastroplasties, and 146 laparoscopic gastric bypasses, 17 of which were robot assisted. The Da Vinci Intuitive Robotic System (Sunnyvale, California, USA) became available at our institution in
1852
Fig. 2. Operative unit of the robotic Da Vinci system.
Materials and methods
Fig. 1. SurgeonÕs site of the robotic Da Vinci system.
May 2003. According to the protocol, from September 2003, we performed GJ anastomosis using the robotic technique for all obese patients without randomization to verify its advantages over other techniques. Actually, we believe that the only limitation of this approach is the higher cost of robotic instruments, as compared with laparoscopic instruments. The Da Vinci Intuitive Robotic System has two main components: the surgeonÕs viewing and control console with its master manipulator (Fig. 1) and the surgical arm unit (Fig. 2). The surgeon sits comfortably at the control console, looks into a three-dimensional monitor in front of his face, and fixes his bilateral thumbs and index fingers into manipulators by which he can control robotic arms. The surgeonÕs screen presents a magnified three-dimensional operative field showing focalized robotic arms [15] that faithfully reproduce the surgeonÕs hand movements. Moreover, the surgeon can focalize the view and move the optic, which is the central third arm of the robot, by the help of a foot pedal. We use this system to perform advanced laparoscopic procedures, particularly bariatric surgery. This study was a retrospective analysis of our preliminary experience using the Da Vinci Intuitive Robotic System to perform gastric bypass procedures for the treatment of morbid obesity. The study aimed to show the efficacy and safety of this system in relation to other standardized laparoscopic surgical techniques including GagnerÕs technique, Cadie`reHimpenÕs technique, HigaÕs technique, the purse-string technique using Ko¨ckerling-StorzÕs forceps, LonrothDe MariaÕs technique, SchauerÕs technique, and WittgroveÕs technique.
From October 2000 to March 2004, we performed 146 consecutive laparoscopic Roux-en-Y gastric bypasses including 86 mechanical gastrojejunal anastomoses using GagnerÕs procedure, 43 purse-string anastomoses using StorzÕs Ko¨ckerling forceps, and 17 robotic anastomoses using the Da Vinci Intuitive Robotic System (beginning in September 2003). The robot-managed patients comprised 7 men and 10 women ages 26 to 59 years (mean age, 44 years). Their mean weight was 139 kg (range, 97–232 kg), and their mean body mass index (BMI) was 49.8 kg/m2 (range, 35.5–96 kg/m2). Their mean excess body weight rate (EBW%) was 122.3% ± 49.7% (range, 64.6–249.5%). The related pathologies included 7 cases of hypertension, 5 cases of sleep apnea, 4 cases of osteoarthritis, 2 cases of diabetes type II, endocrinopathy, and gastritis, and 1 case of chronic obstructive pulmonary disease, hypercholesterolemia, arrythmias, and varicose veins (Table 1). Presurgical evaluation was based on a multidisciplinary approach that included upper endoscopy, upper GI Rx series, abdomen ultrasound, and ultrasound Doppler of inferior members, as well as anesthesiologic, psycho-endocrine, dietologist, and pneumologic evaluation. For all patients, we performed eradication of Helicobacter pylori if gastric biopsy or fecal antigen results were positive. For two patients, we positioned a temporary inferior caval vein filter because their excessive size (weight, >200 kg) precluded preoperative diagnostic examinations. The table could not support their weight, and the TC o RMN machine could not accommodate their size. In one case, it was necessary to perform an emergency tracheotomy for acute respiratory failure before the operation. Follow-up evaluation was conducted after discharge, at the first, third, sixth and twelfth months, and then yearly. For all the patients, we evaluated the variations in BMI and the percentage of excess body weight loss (EBWL%) at 1, 3, 6, and 12 months. At 3 months (or earlier for patients with dysphagia), we also performed a routine gastrografin swallow. An upper gastrointestinal endoscopy (with eventual endoscopic dilation) was performed as a second choice. All patients were informed of the risks inherent in each surgical procedure as well as the potential benefits and therapeutic alternatives.
Surgical technique The patient is placed in a supine position with the legs abducted on straight leg boards. The legs are placed in pneumatic compression boots. A nasogastric tube and urinary catheter are placed. A video monitor for the assistant is positioned over the left shoulder. The operation begins with creation of the pneumoperitoneum, and six trocars are placed under direct vision of a 30° laparoscope for grasper, operating instruments, assistantÕs instruments, and retractor. For technical reasons related to the robotic system, we have been forced to change mini-invasive ports and the operative steps. Because of trocar positioning for robotic technique, all trocars are 5 to12 mm. They are placed, respectively, in the superior midline epigastrium 15 cm under the xiphoid (T1), in the left upper quadrant at
1853 Table 1. Results of a robot-assisted gastric bypass
Recovery (days)
Preop weight (kg)
BMI
EBW (%)
ASA
Op time (min)
Related pathologies Diabetes, hypertension Hypercholesterol, hypertension, osteoarthritis, diabetes. Diabetes, hypertension Hypertension — Hypertension Hypertension, osteoarthritis., sleep apnea Sleep apnea, respiratory failure, varicose veins — Sleep apnea, atrial fibrillation, osteoarthritis, severe chronic obstructive pulmonary disease, varicose veins, hypothyroid Hypertension, osteoarthritis Sleep apnea, hypertension Hypertension, osteoarthritis Hypertension, hypogonadism, gastritis Asthma, hypertension Diabetes, hypertension,varicose veins Gastritis HP+, hypertension, osteoarthritis,sleep apnea —
Patient
Sex
Age (yrs)
B.D. C.T.B.A.
F M
40 47
8 8
155 154
50.6 45.5
148 104
2 3
125 225
G.M. A.A. B.A. B.A. V.L.
M F F F M
34 35 41 49 48
7 6 6 7 7
127 97 111 126 136
49.6 39.4 45 49.2 40
120.9 81.3 107.5 129.1 80
3 2 2 2 3
300 180 210 210 170
G.M.
F
51
15
232
70.8
216.7
4
270
M.A. V.A.
F F
39 56
7 18
109 230
43 96
98.16 338.09
2 4
90 235
D.G.G. B.T. L.G. C.G.
F M M M
58 52 38 34
7 11 15 6
118 107 120 146
48 37 35.5 47.7
119 64.6 58.6 112.2
3 3 2 3
230 240 180 150
B.A. S.R.
F F
26 24
9 10
150 99
50 41.8
142 90.4
2 4
240 180
P.M.T.
F
58
9
131
51.1
138
2
180
Mean
7M/10F
42.9
9
138.5
50.3
129.9
—
201
Preop, preoperative; BMI, body mass index; EBW%, percentage of excess body weight; ASA, American Society of Anesthesiology score; Op, operative; HP+, Helyoobacter pylori
Fig. 3. Robotic gastro-jejunal anastomosis posterior plane.
the middle clavicular line level (T2), in the right upper quadrant at the middle clavicular line level (T3), in the left anterior axillary line for gastric retraction and insertion of the assistantÕs instruments during robotic time (T4), in the mesogastrium in the right para-umbilical space (T5), and in the right anterior axillary line for liver retraction (T6) (Fig. 3). The operation begins with laparoscopy, performed before the Roux limb with passage of the jejunal loop through the mesocolon, which is closed by simple suture to avoid internal hernias. Subsequently, the mechanical gastric resection is performed. Finally, we create the gastrojejunal (GJ) anastomosis using the Da Vinci Intuitive Robotic System. Our first three robotic gastric bypasses also included the gastric resection, but because no real benefit in terms of time was realized, we decided to create only the GJ anastomosis in subsequent operations.
We start the operation by opening the gastrocolic ligament with the harmonic scalpel. Care is taken to preserve the gastrohepiploic vessels and the root of transverse mesocolon below the inferior margin of pancreatic body. TreitzÕs ligament is found through the mesocolon. Creation of the Roux limb is accomplished from the first loop of the jejunum 50 cm to the ligament of Treitz through the mesentery. The bowel is divided with a vascular 60-mm endoscopic linear cutting stapler. A side-to-side enteroenterostomy is performed 100 to 150 cm from the end of the Roux limb (according to the BMI of the patient) using a vascular 60-mm endoscopic linear cutting stapler. The anastomotic incision is closed with a 3-0 absorbable running suture. The alimentary loop is carried through the mesocolon in the upper mesocolic field for the next GJ anastomosis, and the mesocolon orifice is closed with a 3-0 running suture to avoid internal hernias. A space adjacent to the lesser curvature of the stomach 6 cm down from the gastroesophageal junction is cleared through the lesser curvature mesentery to allow the passage of an endoscopic linear cutting stapler for gastric division. The harmonic scalpel (Autosonix (Tyco Healthcare Group, Boulder, CO, USA) or Ultracision (Ethicon Endosurgery, Cincinnati, USA)) can be useful for hemostatic dissection in the area. The linear endoscopic 45-mm blue-cartridge stapler is placed perpendicular to the lesser curvature and used to transect the stomach. A 36Fr gastric tube is placed by the anesthesiologist in the stomach through the mouth. Multiple applications of the linear cutting stapler (45–60 mm blue or green cartridge for patients with previous gastric intervention) are necessary to divide the stomach vertically toward the angle of His. Next, the tube is removed so the robotic GJ anastomosis can be performed.The Da Vinci Intuitive Robotic System is placed over the patient. The robotic three-dimensional optic is inserted in T1 while the operative arms are positioned, with the robotic trocars (8 mm) inserted inside the previous trocars (T2–T3). The robotic surgical instruments include the following: a Cadie`re forceps (Intuitive Surgical, Inc., Sunnyvale, California, USA), a large needleholder (Intuitive), and a permanent cautery hook (Intuitive). A Cadie`reÕs forceps is placed on the left and the needleholder on the
1854
Fig. 4. Robotic gastrojejunal anastomosis: anterior plane. right. The end of the Roux limb is fixed by running suture to the posterior layer of the gastric pouch under the transverse linear stapled line. Then a gastrotomy and a jejunotomy are performed using the robotic electrocautery hook. Finally, a double-layer gastrojejunal hand-sewn anastomosis is completed with three other running absorbable sutures (3-0 posteriorly and anteriorly, 4-0 in the interior) accomplished by a large needleholder (Fig. 4). During this procedure, the assistant alternatively uses a laparoscopic grasper to hold the right sutureÕs tension or a forceps to cut the thread (T4). The anastomosis is tested by inserting 60 ml of methylene blue solution through a nasogastric tube. A drain usually is placed next to the gastroenteric anastomosis through a left-sided port (e.g., T4).
Results At the beginning (for the first 3 cases), we also made the gastric pouches with the robotic system. However, the robotic system offered no real advantage over laparoscopy for this procedure because for this operative step, the harmonic scalpel (still not available with the robotic system) and a mechanical suturing device are necessary. We found that the hand-sewn robotic anastomosis results in greater safety, speed, and ease of performance. With the robotic system, the mean operative time was 201 min (range, 90–300 min). The mean operative robotic anastomosis time was 19.3 min (range, 16–35 min), excluding the setup time for the Da Vinci Intuitive Robotic System. No intraoperative complications and no conversion occurred in our series. The mean hospital stay was 9 days (range, 6–18 days). All the patients had a normal postoperative course. No anastomotic fistula, no intraabdominal bleeding, and no parietal infections were observed in our series. No complications related to the anastomotic techniques occurred. No robot-related complications were noted, and the mortality rate was zero (Table 2). All 17 patients had a minimum follow-up period of 6 months, and 9 of these patients also were evaluated 12 months after the operation. The follow-up analysis at 1, 3, 6, and 12 months showed a progressive reduction of BMI. Our data showed that the mean of BMI was 43.48 ± 9.99 at 1 month, 39.07 ± 9.21 at 3 months, 34.31 ± 7.98 at 6 months, and 31.84 ± 7.93 at 12 months. The mean EBWL% observed was 20.72% ± 6.6% at 1 month, 38.18% ± 9.28% at
3 months, 56.74% ± 13.6% at 6 months, and 65.20 ± 15.79% at 12 months (Table 3). Our findings showed good results in terms of weight loss. The patients lost at least 50% of their preoperative excess weight. This goal was achieved by 64.7% of 17 patients at 6 months, and by 77.7% of 9 patients at 12 months. According ReinoldÕs criteria, the results at postoperative month 12 were excellent (EBWL%, 75– 100%) for 33.3% of the patients and good (EBWL%, 50– 74%) for 44.4% of the patients (Table 4). No patient manifested malnutrition or incisional hernia during the follow-up period. All the patients showed significant improvement in terms of comorbidities. The Bariatric Analysis and Reporting Outcome System (BAROS) results were excellent, very good, or good for 100% of the patients who underwent surgery.
Discussion Laparoscopic gastric bypass surgery is a complex and challenging intervention requiring advanced laparoscopic skill that involves laparoscopic stapling, laparoscopic dissection of the gastroesophageal junction, and laparoscopic suturing. This surgery can be associated with significant morbidity including intraoperative splenic injury, anastomotic leaks, deep venous thrombosis, pulmonary embolism, wound infection, ventral hernia, and respiratory complications [2–4]. The GJ anastomosis can be performed in various ways: via circular stapler with the the anvil introduced through the mouth, according to GagnerÕs technique [5, 6]; via the introduction of the anvil through a gastrotomy, according to Cadie`re-HimpenÕs technique [7]; via laparoscopic hand-sewn, double-continued sutures, according to HigaÕs technique [2]; via a laparoscopic purse-string device using Ko¨ckerling-StorzÕs forceps, according to PariniÕs technique [12] (Fig. 5); via linear stapler, according to Lonroth-De MariaÕs technique [8, 9]; via mixed mechanical and manual anastomosis, according to SchauerÕs technique [3]; via WittgroveÕs technique [10], and via robotic hand-sewn anastomosis. Our laparoscopic gastric bypass experience started with GagnerÕs technique, but after the first 86 cases, we preferred to perform the purse-string device anastomosis to avoid damage to the esophagus resulting from introduction of the stapler anvil through the mouth, and to avoid port infection from introduction of the stapler (Table 5). Recently, with the introduction of the robotic Da Vinci Intuitive Robotic System into our surgical division, we have started to perform robotic GJ anastomosis after the laparoscopic preparation of the Roux-en-Y jejunal loop. For the first three cases, we used the robot to create the gastric pouch, but we saw no advantage in this step because the gastric transection is completely performed by Endogia (45–60) and the harmonic scalpel. Wittgrove et al. [10] already have reported the benefits of a decreased hospital stay with a better postoperative course for patients undergoing laparoscopic Roux-en-Y gastric bypass, and our results are consistent
1855 Table 2. Results for the different gastrojejunal anastomoses in the gastric bypass
Type of anastomosis
Patients (n)
Mean total operative time (min.)
Robotic hand sewn GagnerÕs technique (Parini) Purse-string device (Parini) Circular stapler (Gagner) Mixed mechanical and manual (Schauer) Linearstapler (Lonroth-De Maria) Lap hand sewn (Higa) Cadie`re-HimpenÕs technique WittgroveÕs technique
17 97 38 456 275 281 1,040 191 500
201 138 162 — 260 234 60 105 90
Anastomotic leak (%)
Wound infection (%)
Stenosis (%)
Bleeding (%)
Mortality (%)
0 1.03 0 1.32 1.81 5.1 1.15 1.57 1.8
0 3.09 3.22 — 4.72 1.1 0.09 1.04 0.8
0 2.06 2.63 3.29 4.72 6.6 4.9 2.09 1.6
0% 3.09 5.26 0.2 3.27 — 0.57 — 0.8
0% 0 0 0.5 0.36 0 0.09 0 0
Table 3. Data for follow-up assessment at 1, 3, 6, and 12 months Patient
BMI preop
BMI 1 mo postop.
BMI 3 mo postop
BMI 6 mo postop
BMI 12 mo postop
EBW% preop
EBWL% 1 mo Postop
EBWL% 3 mo postop
EBWL% 6 mo Postop
EBWL% 12 mo postop
B.D. C.T.B.A. G.M. A.A. B.A. B.A. V.L. G.M. M.A. V.A. D.G.G. B.T. L.G. C.G. B.A. S.R. P.M.T. Mean SD
50.61 45.49 49.61 39.35 45.03 49.22 40.17 70.82 42.72 76.89 49.12 37.02 38.58 47.67 49.54 41.74 51.17 48.52 10.62
44.73 41.35 42.97 36.11 40.57 44.53 35.74 64.71 38.65 71.77 43.29 32.18 33.02 42.45 45.91 36.68 44.53 43.48 9.99
39.84 36.03 37.70 32.46 36.51 40.23 33.08 60.44 34.81 64.08 38.29 30.45 28.70 37.55 41.29 32.05 40.63 39.07 9.21
38.53 34.85 29.30 28.40 32.05 35.94 26.29 55.40 30.06 51.69 33.51 27.68 25.15 32.33 35.01 29.52 37.50 34.31 7.98
36.57 32.79 26.56 27.59 30.43 31.25 25.11 50.52 25.71 — — — — — — — — 31.84 7.93
148.0 104.0 120.9 81.3 107.5 129.1 80.1 216.7 98.2 249.5 124.8 64.6 72.4 112.4 141.9 90.4 138.2 122.3 48.7
19.46 17.83 24.46 18.39 19.13 16.90 24.79 12.60 19.25 9.34 21.37 33.33 34.29 20.71 12.50 25.53 22.37 20.72 6.60
35.68 40.76 43.88 39.08 36.52 32.39 39.67 21.42 37.38 23.35 39.69 45.24 60.95 40.13 28.41 48.94 35.53 38.18 9.28
40.00 45.86 74.82 62.07 55.65 47.89 77.69 31.81 59.81 45.91 57.25 64.29 82.86 60.84 50.00 61.70 46.05 56.74 13.60
46.49 54.78 84.89 66.67 62.61 64.79 84.30 41.89 80.37 — — — — — — — — 65.20 15.79
BMI, body mass index; EBW%, percentage of excess body weight; SD, standard deviation
Table 4. Data for follow-up assessment at 1, 3, 6, and 12 months and results according ReinoldÕs criteriaa Follow-up (% of patients)
1 mo (n = 17)
3 mo (n = 17)
6 mo (n = 17)
12 mo (n = 9)
Excellent (EBWL%: 75–100) Good (EBWL%: 50–74) Fair (EBWL%: 25–49) Poor (EBWL%: 0–24) Failure (EBWL%:
