Surgical Endoscopy

, Volume 18, Issue 12, pp 1742–1746

Robot-assisted vs laparoscopic adrenalectomy: a prospective randomized controlled trial


    • Department of Surgery, Minimally Invasive Surgery CenterUniversity of Turin
  • G. Benincà
    • Department of Surgery, Minimally Invasive Surgery CenterUniversity of Turin
  • G. Giraudo
    • Department of Surgery, Minimally Invasive Surgery CenterUniversity of Turin
  • G. M. Del Genio
    • Department of Surgery, Minimally Invasive Surgery CenterUniversity of Turin
  • F. Rebecchi
    • Department of Surgery, Minimally Invasive Surgery CenterUniversity of Turin
  • C. Garrone
    • Department of Surgery, Minimally Invasive Surgery CenterUniversity of Turin
Original article

DOI: 10.1007/s00464-004-9046-z

Cite this article as:
Morino, M., Benincà, G., Giraudo, G. et al. Surg Endosc (2004) 18: 1742. doi:10.1007/s00464-004-9046-z



The aim of this study was to assess the benefits and disadvantages of robot-assisted laparoscopic surgery for disorders of the adrenal gland in terms of feasibility, safety, and length of hospitalization.


Twenty consecutive patients with benign lesions of adrenal gland were randomized into two groups: Patients in the laparoscopic group underwent traditional laparoscopic adrenalectomy (LAP), whereas those in the robotic group underwent robot-assisted adrenalectomy (ROBOT) using the da Vinci robotic system.


There was no significant difference between the groups in terms of age, sex, body mass index, and size or locations of lesions. Operative times were significant longer in the ROBOT group (total operative time, 169.2 min [range, 136–215] vs 115.3 min (range, 95–155) p < 0.001. Skin-to-skin time was 107 m (range, 77–154) vs 82.1 min (range, 55–120) (p < 0.001). There were no conversions to open surgery. However, conversion to standard laparoscopic surgery was necessary in four of 10 ROBOT patients (40%; left, one right). Perioperative morbidity was higher in the ROBOT group (20% vs 0%). There was no difference in length of hospital stay. In the following ROBOT group, hospital stay was 5.7 days (range, 4–9) vs 5.4 days (range, 4–8) in the LAP group (p = NS). The total cost of the ROBOT procedure ($3,467) was significantly higher than that for LAP ($2,737) (p < 0.01).


Laparoscopic adrenalectomy is superior to robot-assisted adrenalectomy in terms of feasibility, morbidity, and cost.


Laparoscopic adrenalectomyRobot-assisted surgeryLaparoscopyAdrenal gland

Over the last decade laparoscopic surgical techniques have become the gold standard for the management of a variety of abdominal diseases, whether benign or malignant [6, 18]. In particular, adrenalectomy is now routinely performed via the minimally invasive approach, and laparoscopic adrenalectomy is accepted as the treatment of choice for patients with benign adrenal lesions [12]. Despite the improvement in outcomes and quality of life, the application of laparoscopy is not without its technical drawbacks, including the use of nonarticulated instruments with an ergonomically inadequate handle design, and inefficient handle-to-tip force transmission, an awkward, uncomfortable position for the surgeon at the operating table, and a flat, two-dimensional image [5, 8].

Recently, robotic technology has been introduced into laparoscopic clinical practice with the aim of increasing surgeon dexterity by eliminating tremors and fatigue and reducing the scale of movements [7]. The earliest implementation of any type of computer-aided surgery occurred during the 1980 s in neurosurgery with stereotactic surgery, which provided accurate positioning to locate the lesion and perform a precise biopsy of the brain [21]. In 1991, the development of robotic engineering led to the application in urology of the Probot (Imperial College of London, London, England) for transurethral resection of the prostate and in orthopedic surgery of the RoboDoc (Integrated Surgical Supplies, Sacramento, CA, USA), which is programmed to create a precise defect within the femoral shaft for the placement of a prosthesis [19].

The initial goal of combining robotic technology with laparoscopic surgery was to enable a solo surgeon approach. The AESOP program (Automated Endoscopic System for Optimal Positioning; Computer Motion, Goleta, CA, USA) was designed in the United States with the purpose of enabling the camera to be moved under direct control of the surgeon. Its development began in 1995 with the introduction of a foot pedal to handle the laparoscope by 1997, it incorporated a voice-controlled interface (AESOP 2000) and a three-dimensional visualization system (AESOP 3000) [21]. In 1997, Cadière performed the first laparoscopic procedure through a robot [7] — a laparoscopic fundoplication using the da Vinci Robotic Surgical System (Intuitive Surgical, Mountain View, CA, USA). This device adds to the advantages of laparoscopy the benefits of three-dimensional vision and the ability to make intraabdominal movements in three dimensions [22].

Different clinical applications of the da Vinci Robotic Surgical System have been described in recent years, including the entire spectrum of laparoscopic procedures [14, 15] as well as a limited number of carefully selected case reports and small series of adrenalectomies [3, 10, 13, 23]

At the Minimally Invasive Surgery Center of the University of Turin, we have recently adopted the da Vinci Surgical System. After a short period of clinical training, we designed a randomized, controlled clinical trial to evaluate robot-assisted adrenalectomy.

The aim of this prospective randomized controlled trial was to compare the feasibility and clinical results of robot-assisted adrenalectomy to standard laparoscopic adrenalectomy in a consecutive series of patients. To our knowledge, this is the first randomized study comparing a robotic procedure with another surgical approach.

Materials and methods

Patients with benign lesions of the adrenal gland were randomized to two different groups; one group underwent lateral flank laparoscopic adrenalectomy, (LAP) whereas the other group underwent robot-assisted lateral flank laparoscopic adrenalectomy (ROBOT). Patients exclusion criteria were the presence of a lesion >10 cm, bilateral lesions, and clinical or morphological suspicion of malignancy. A specific informed-consent form approved by our institution’s ethics committee, and signed by the candidates was required before inclusion in the trial. Patients were randomly assigned to groups at the time of surgical indication.

Patients with secreting diseases received the same preoperative pharmacological treatment in both groups (e.g., prior to operation all cases of pheochromocytomas received 15 days of α = lithics and β = blockers). The protocol for anesthesia was uniform in all patients. Analgesia was carried out with ketorolac and tramadol during the first 12 h postoperatively and thereafter at the request of the patient.

Variables collected included patient age, sex, body mass index (BMI), side (left or right), dimension and type of adrenal lesion, history of abdominal surgery, comorbidity, perioperative data, postoperative complications, length of hospital stay, and costs. We calculated total operative time (including anesthesia) and operative skin-to-skin time.

Cost evaluation included the use of operating room (e.g., nurse, technical staff, surgical devices, and maintenance), the cost of surgical tools (e.g., disposable instruments, trocars, and wires), and the charge for the hospitalization. At our hospital, the total cost of the operating room is ($304 per hour and the cost for each day of hospitalization is $249. The initial cost of the robot was not factored into the cost analysis.

Statistical analysis was carried out using the Student’ t-test. A p value <0.01 was considered significant.


The operating room for the LAP and ROBOT procedures was staffed by a team of nurses and technicians who were experienced in laparoscopic procedures. The LAP approach was transperitoneal, with the patient in lateral flank position rotated on the opposite side of the adrenal lesion, as described previously [20].

For robot-assisted procedures, we used the da Vinci system, which consists of a robotic manipulator with three arms (the central arm holds the camera and the two other arms hold the surgical instruments) and a remote console, where the first operator is seated during the procedure. A total of three laparoscopic-proficient surgeons (M.M., C.G., F.R.) were involved in the procedures. The operating surgeon controlled the robot at the console, whereas the first and second assistants, once the pneumoperitoneum was created, placed the trocars, connected the arms, and handled the laparoscopic instruments inserted in the accessory trocars. The system includes two controllers shaped as joysticks, from which the signal is transmitted to a computer, elaborated, and sent to robotic arms through a 10 m–long cable. The robot minimizes any tremors of the surgeon’s hands and relays smooth motions to the surgical instruments. The instruments were semi-disposable; the robot tracks the number of times they are used and will not operate an instrument after the 10th use. Most dissection was done at a 2:1 or 3:1 scale. In all cases, a 30° angled scope was used.

Left robotic adrenalectomy

The patient was placed in right lateral flank position, slightly rotated with the face upward. The robot’s optimal position was behind the patient’s left shoulder, 45° left from table’s the axis. After a 12-mmHg pneumoperitoneum had been created with a Veress needle, trocars were placed along the subcostal margin at the side of adrenal lesion. Three robotic trocars (one 12-mm disposable for the camera, two 7-mm nondisposable) were inserted for the da Vinci tools, and one additional trocar (12-mm disposable) was inserted for the auxiliary laparoscopic instruments. The operative steps reproduced those of the laparoscopic technique [20]. Once the abdominal cavity had been explored, the splenic flexure of the colon was refracted downward, while the spleen and pancreas rotated medially were enbloc, through the division of the spleno-colic and freno-lienal ligaments via robotic hook diathermy. Dissection continued into the periadrenal fat to identify the following landmarks: inferiorly, the left renal vein; laterally, the superior pole of the kidney; medially, the tail of pancreas and the splenic vessels; and posteriorly, the psoas muscle. Alternation of the robotic monopolar hook and the laparoscopic bipolar forceps, introduced from the auxiliary trocar, provided hemostasis. Next, the left adrenal vein was sectioned between clips applied from the auxiliary trocar at the origin with the renal vein, while the remaining dissection of the adrenal space was completed from the console by the first surgeon. The specimen was captured in a plastic bag and delivered through the auxiliary port site.

Right robotic adrenalectomy

The patient was placed in left lateral flank position, with a slight upside rotation. The robot’s optimal position was at the patient’s right shoulder, 45° right from the table’s axis. After a 12-mmHg pneumoperitoneum had been created, trocars were placed along the subcostal margin at the side of adrenal lesion. There were three robotic trocars (one 12-mm disposable for the camera, two 7-mm nondisposable for the da Vinci tools), one additional trocar (12-mm disposable for the auxiliary laparoscopic instruments), and (if required) an additional 5-mm nondisposable trocar. The operative steps reproduced those used in the laparoscopic technique [20]. A robotic monopolar hook was used to divide the right triangular ligament of the liver to better elevate the right hepatic lobe with a laparoscopic retractor. The surgeon at the console moved the robotic hook to achieve a precise dissection of the inferior vena cava along its lateral edge between the liver and the right renal vein. Others landmarks identified were laterally the superior pole of right kidney and posteriorly the psoas muscle. The right adrenal vein was isolated and sectioned between clips applied by the on-site assistant at the origin of the inferior cava vein. The dissection of the periadrenal space was completed using the robotic hook and bipolar forceps introduced from the auxiliary trocar. The specimen was delivered in a plastic bag.


Between March and December 2002, 22 patients with benign adrenal lesions underwent laparoscopic adrenalectomy at our center. Two patients were excluded from the study, one had a pheochromocytoma with a 12-cm diameter, and the other had with Cushing’s syndrome with two bilateral secreting lesions. Thus, 20 consecutive patients with benign lesions of adrenal gland were enrolled in the study (Table 1): Ten patients (five women, five men) underwent standard laparoscopic adrenalectomy, (seven left and three right), and 10 patients (six women, four men) underwent robotic adrenalectomy seven left and three right).
Table 1

Patient data




p value

No. of patients








Age (yr)

40.3 (range, 23–72)

38.7 (range, 19–68)



25.8 (range, 20.3–28.9)

22.2 (range, 18.6–31.2)


Mean diameter of lesion (cm)

3.1 (range, 1.5–6)

3.3 (range, 1.4–6.5)


Left/right side




Previous surgery









3 Conn’s adenomas

3 Cushing’s adenomas

3 pheochromocytomas

1 incidentaloma

3 Conn’s adenomas

4 pheochromocytomas

3 incidentalomas


LAP, standard laparoscopic adrenalectomy; ROBOT, robot-assisted adrenalectomy; BMI, body mass index

The indications in the LAP group were three adenomas in Conn’s syndrome, three adenomas in Cushing’s syndrome, and three benign pheochromocytomas, one incidentaloma with a diameter >4 cm (diagnosed at pathology as a nonsecreting adenoma). The indications in the ROBOT group were three adenomas in Conn’s syndrome, four benign pheochromocytomas, and three incidentalomas with diameter >4 cm (diagnosed at pathology as two nonsecreting adenomas and one myelolipoma).

The LAP group patients had a mean age of 40.3 years (range, 23–72) and, a mean BMI of 25.8 (range, 20.3–28.9). One patient had previously abdominal surgery, five patients had comorbidities (one hypertension, two diabetes, one myocardial ischemia and one chronic obstructive pulmonary disease). The mean diameter of the lesions was 3.1 cm (range, 1.5–6).

The ROBOT group patients had a mean age of 38.7 years (range, 19–68) and a mean BMI of 22.3 (range, 20.6–31.2). Two patients had previously had abdominal surgery, six patients had comorbidities (two hypertension, one diabetes mellitus, one coronary artery disease, and two chronic obstructive pulmonary disease). The mean diameter of the lesions was 3.3 (range, 1.4–6.5).

The mean number of trocars required was 4 and 3.2 (range, 3–4) for right and left laparoscopic adrenalectomy, respectively, and 4.1 (range, 4–5) and 4 for right and left robotic adrenalectomy, respectively. There were no intraoperative complications in the LAP group, but two patients (20%) in the ROBOT group with pheochromocytomas developed an episode of severe intraoperative hypertension. All LAP operations were completed laparoscopically. Conversion to standard laparoscopic surgery was required in four patients (40%) in the ROBOT group (Table 2). The reasons for the conversions were malposition of robotic trocars in two cases, difficulty in obtaining hemostasis in one case and a prolonged operative time in the last case. Therefore, in all cases the conversion was deemed necessary by the surgeon because of difficulty in pursuing the dissection via robotic techniques, not because of the onset of intraoperative complications. The rate of conversion was 33% (one of the three patients) in right adrenalectomies and 42% (three of seven patients) in left adrenalectomies (p = NS). The mean diameter of the lesion in the converted cases was 3.5 cm. The indications for adrenalectomy in the converted cases were two incidentalomas, one Conn’s syndrome, and one pheochromocytoma. The conversions rate was significantly higher in the first five cases (three of five, or 60%) than in the others (one of five, or 20%), (p < 0.01).
Table 2

Perioperative data







Intraoperative complications



Postoperative complications






Length of hospital stay (d)



LAP, standard laparoscopic adrenalectomy; ROBOT, robot-assisted adrenalectomy

a LAP group, conversion to open surgery; ROBOT group, conversion to standard laparoscopy

Total operative time was 115.3 min (range 95–155) vs 169 min (range, 136–215), and, skin-to-skin time was 82.1 min (range 55–120) vs 107 min (range, 77–154) in the LAP and ROBOT groups, respectively. Both time differences were statistically significant (p < 0.01).

There were no postoperative complications with either approach. Mean length of hospital stay was 5.4 days (range, 4–8) in the LAP group and 5.7 days (range, 4–9) in the ROBOT group (p = NS). However, it is noteworthy that the length of stay of the ROBOT patients successfully completed by the robotic technique was 5.4 days (range, 4–8), whereas the length of stay of the ROBOT patients converted to stardard laparoscopy was 6.25 days (range, 4–9) days (p = NS).

Costs per patient were significantly higher in the ROBOT group LAP group than in the Lap group: $3,466 vs $2,737 (p < 0.01) (Table 3). The following disposable or semi-disposable instruments were used one 12-mm trocar, one clip applier, one plastic bag, and the Ultrasonic Scalpel handle (Ethicon Endo-Surgery, Cincinnati, OH, USA) in the LAP group and two 12-mm trocars, one clip applier, one plastic bag, one robotic dissector, and one robotic hook in the ROBOT group.
Table 3

Costs in US dollars




Total operative time



(60 min = 305)

(115 min)

(169 min)

Disposable instruments



Hospital stay



(1 d = 249)

(5.4 d)

(5.7 d)

Total costs



LAP, standard laparoscopic adrenalectomy; ROBOT, robot-assisted adrenalectomy


Although favorable outcomes have recently been reported in small series of selected cases of robotic adrenalectomy [10, 13, 23], there is only a single nonrandomized prospective study comparing robot-assisted to the standard laparoscopic approach [6].

The present prospective randomized controlled trial, although it concerns only a limited number of patients, enables us to evaluate the clinical benefits and drawbacks of robot-assisted surgery when applied to adrenalectomy. The study was conducted in a surgical unit specially devoted to laparoscopic surgery; the two operating surgeons controlling the robot on the console (M.M., C.G.) had extensive experience in advanced laparoscopic surgery and specifically had performed >140 laparoscopic adrenalectomies. To reduce the effect of the learning curve, we began the study after a training period of 10 robotic antireflux fundoplications. This number of procedures was considered adequate to standardize the robot setup time and the surgical maneuvers [2].

Operative time is a reasonable parameter to evaluate when determining the efficacy of a new technology designed to improve a preexisting operation [17]. Total operative time and skin-to-skin time were both significantly longer in the ROBOT group. We believe this was mainly due to the lack of robotic instruments specifically designed for laparoscopic surgery. The time required for robot setup, which will probably shorten with improved experience, increased the turnover time for back-to-back cases.

The use of robotic technology led to a 40% rate of conversion to standard laparoscopy. Given that this was a series of consecutive nonselected cases, this is an extremely significant datum. The incidence of conversion was similar for both the right (one of three) and left (three of seven) sides and was not related to the specific pathology or the diameter of the lesions. We attributed this problem to interference between the robot arms, the lack of a bipolar energy source for the hemostasis, and the extra time needed to switch instruments. As a consequence, in some cases dissection and hemostasis became hazardous or difficult, prompting the surgeon to convert to traditional laparoscopy. As the technology advances and new multi-use instruments are developed we can expect these difficulties to be ameliorated. Nevertheless, it should be noted that with increasing experience the conversion rate was reduced from 60% in the first five cases to 20% in the following five. Therefore, the learning curve played a major role in determining the conversion rate. There were no conversions to open surgery.

From a clinical outcomes standpoint, robot-assisted surgery was comparable to standard laparoscopy in terms of postoperative morbidity and mortality (Table 3). Nevertheless, two patients in the ROBOT suffered a severe intraoperative crisis due to hypertension during the dissection of pheochromocytomas. This figure is not with in the range respected for recent series of patients undergoing traditional laparoscopic procedures [4, 11, 20] and should be verified in the context of larger number of patients before it can be ascribed to robotic manipulation.

Another drawback was the higher cost of the robotic procedure (ROBOT $3,466 vs LAP $2,737). This difference did not include the initial cost to buy the da Vinci system. The increased expense was mainly due to the use of semi-disposable robotic instruments and the longer operative time. Moreover, the ROBOT group did not benefit from the use of fewer trocars or a shorter postoperative hospitalization. Although no economic benefit could be shown for the robot-assisted approach, a multidisciplinary use of the system (e.g., in urology and cardiac, laparoscopic, and thoracic surgery) would result in a faster realization of return on the initial investment.

From a technical point of view, the da Vinci system has certain disadvantages that are most likely related to the fact that the technology is still in its infancy. First, the lack of tensile feedback is a major limitation of the device. Currently, the best way to address this limitation is by applying the “intuitive method,” based on training of the eye–brain pathway to improve the translation and elaboration of hand movements [22]. Second, the range of robotic instruments now available is somewhat limited, because this system was designed specifically for use in cardiac surgery ratter than laparoscopic procedures. Because bipolar cautery is not available, an auxiliary trocar has to be placed and controlled by an on-side assistant. The Ultrasound Scalpel, newly available, is similar to those familiar from standard laparoscopy and does not have an articulated tip [1]. Moreover, switching instruments is laborious and may lead to difficulties in cases with active bleeding.

In conclusion, our study suggests that standard laparoscopic adrenalectomy is superior to robot-assisted adrenalectomy in terms of feasibility, duration, and costs. These data are similar to those obtained in nonrandomized series comparing robotic and laparoscopic adrenalectomies [6] or antireflux fundoplication [17] and contrast with the purported benefits reported in nonrandomized prospective series of robotic adrenalectomies [9, 21].

In agreement with other authors, we believe that it is unlikely at the present time that robotic systems will be of significant benefit when applied to routine surgical practice. We therefore suggest that they should be adopted only in the setting of approved clinical trials [8]. However, robotic technology is under rapid and constant development. Improvements such as new instruments, smaller robotic arms, tactile feedback, and a fourth arm would represent an undeniable advantage and prompt us to reevaluate the clinical indications for its use in adrenal surgery.

Copyright information

© Springer-Verlag 2004