INTRODUCTION
Laparoscopic adrenalectomy has become the standard treatment for removal of adrenal tumors [
1]. This procedure has several advantages compared with open adrenalectomy, such as minimal invasiveness and shorter postoperative recovery and length of hospital stay [
2]. The recent introduction of robotic technology into surgical practice has led to robot-assisted laparoscopic prostatectomy, nephrectomy, and many other urologic procedures [
3-
5]. The first robot-assisted laparoscopic adrenalectomy was described in 1999 [
6].
Robotic surgery has several advantages over pure laparoscopic surgery, including its 3-dimensional magnified vision, eliminating surgeon tremors and reducing surgeon fatigue during the operation [
7]. These advantages are particularly important in optimizing surgery on patients with large-sized tumors and/or obesity [
8,
9].
Studies comparing robotic and laparoscopic adrenalectomy have shown similar perioperative outcomes, including operative time, estimated blood loss (EBL), complications and length of hospital stay, although robotic surgery has some advantages in difficult-to-treat patients [
9,
10]. For example, mean operation times in obese patients (body mass index [BMI]>30 kg/m
2) and patients with large-sized tumors (>55 mm) were shorter when performed robotically than by conventional laparoscopy [
10]. To extend these findings, this study compared perioperative outcomes in patients undergoing robotic and laparoscopic adrenalectomy.
MATERIALS AND METHODS
1. Study Population
With Institutional Review Board approval, the medical records of all patients who underwent laparoscopic or robot-assisted laparoscopic adrenalectomy from March 2005 to April 2016 at Seoul National University Bundang Hospital were retrospectively reviewed (IRB No. B-1607/353-101). A total of 63 patients underwent adrenalectomy during this period. Total 5 surgeons performed all adrenalectomies. Their clinicopathological characteristics, including age, sex, BMI, underlying disease, symptoms at diagnosis, laboratory findings, tumor size, pathologic results, operative time, EBL, complications, and length of hospital stay, were recorded. Tumor size was defined as the largest diameter of each lesion on preoperative computed tomography or magnetic resonance imaging. Pathologic tumor volume was calculated from the 3 diameters of each specimen. Operative time was defined as the time from skin incision to skin closure. EBL was estimated from drainage volume in the operating room. Postoperative complications were classified according to the modified Clavien-Dindo classification system [
11].
2. Surgical Procedure
Robot-assisted laparoscopic adrenalectomy was performed using the da Vinci S or Si robotic system (Intuitive Surgical, Sunnyvale, CA, USA). All laparoscopic adrenalectomies were performed via a transperitoneal approach. The patient was placed in the lateral decubitus position on a vacuum mat, and the table was flexed at the level of the kidneys. Three ports were used for left laparoscopic adrenalectomy, four ports for right laparoscopic adrenalectomy, and 6 ports for robot-assisted laparoscopic adrenalectomy. A 12-mm camera port was placed on the rectus margin at the level of the umbilicus. After pneumoperitoneum was established, two 8-mm ports were placed at the anterior axillary line and the rectus margin, with the latter 7 cm from the camera port, making an angle of 120°. For right adrenalectomy, a 5-mm port was placed at the right costal margin for liver traction. For robot-assisted laparoscopic adrenalectomy, an additional 8-mm port was placed just below the xiphoid process, and two 12-mm ports were placed at the midline in the epigastrium and hypogastrium. These 2 ports were used by the assistant for suction and clip application.
After Toldt line was incised, the colon was separated from the retroperitoneal space, exposing Gerota fascia. Meticulous dissection was performed, beginning at the medial side of the adrenal gland, with a hem-o-lok used to control the adrenal vein. The adrenal gland was removed with a wrap bag. After hemostasis, a Jackson-Pratt drain was inserted and the pneumoperitoneum was desufflated.
3. Statistical Analysis
Categorical variables were compared by chi-square tests and continuous variables by independent t-tests. Subgroup analysis was performed by Mann-Whitney test. All statistical analyses were performed using IBM SPSS Statistics ver. 19.0 (IBM Co., Armonk, NY, USA), with p-values <0.05 considered statistically significant.
RESULTS
The clinical characteristics of the 63 patients who underwent laparoscopic adrenalectomy during the study period are shown in
Table 1. Of these 63 patients, 29 underwent robot-assisted and 34 underwent conventional laparoscopic adrenalectomy. There were no significant differences in age, BMI, diabetes mellitus, and hypertension between the 2 groups. Seventeen patients (59%) in the robotic group and 11 (32%) in the laparoscopic group were male (p=0.037). Forty-seven patients (75%) were incidentally diagnosed, and 2 patients had uncontrolled hypertension. Mean tumor size (3.0±1.5 cm vs. 3.7±2.7 cm, p=0.134) and the percentage of patients with left adrenal tumors (44.8% vs. 61.8%, p=0.179) were similar in the robotic and laparoscopic groups, but the percentage of patients who had undergone prior abdominal surgery was significantly higher in the robotic than in the laparoscopic group (38% vs 12%, p=0.015).
Table 1.
Patient demographic and clinical characteristics
Variable |
Robotic (n=29) |
Laparoscopic (n=34) |
p-value |
Age (y) |
53.1±12.2 |
51.4±15.1 |
0.631 |
Sex |
|
|
0.037 |
Male |
17 (58.6) |
11 (32.4) |
|
Female |
12 (41.4) |
23 (67.6) |
|
Body mass index (kg/m2) |
25.9±3.8 |
25.2±3.5 |
0.461 |
Diabetes mellitus |
6 (20.7) |
10 (29.4) |
0.428 |
Hypertension |
14 (48.3) |
17 (50.0) |
0.891 |
Smoking |
|
|
0.564 |
Nonsmoker |
19 (65.5) |
20 (58.8) |
|
Ex-smoker |
4 (13.8) |
6 (17.6) |
|
Current smoker |
6 (20.7) |
6 (17.6) |
|
Symptom at diagnosis |
|
|
0.354 |
Incidentaloma |
22 (75.9) |
25 (73.5) |
|
Hypertension |
2 (6.9) |
- |
|
Flank pain |
2 (6.9) |
2 (5.9) |
|
Abdominal pain |
2 (6.9) |
- |
|
Palpitation |
- |
3 (8.8) |
|
Amenorrhea |
- |
1 (2.9) |
|
Etc |
1 (3.4) |
3 (8.8) |
|
Hemoglobin (g/dL) |
13.7±1.1 |
13.5±1.4 |
0.408 |
Creatine (mg/dL) |
0.86±0.25 |
0.77±0.21 |
0.100 |
eGFR (mL/min/1.73 m2) |
82.5±29.0 |
94.2±35.8 |
0.159 |
Tumor size (cm) |
3.0±1.5 |
3.7±2.7 |
0.134 |
Laterality |
|
|
0.179 |
Left |
13 (44.8) |
21 (61.8) |
|
Right |
16 (55.2) |
13 (38.2) |
|
Prior abdominal surgery |
11 (37.9) |
4 (11.8) |
0.015 |
Histopathologic diagnosis showed that the percentages of patients with pheochromocytoma (21% vs. 3%) and metastatic tumor (24% vs. 6%) were significantly higher in the robotic than in the laparoscopic group (p=0.019) (
Table 2). One patient who underwent robotic surgery for pheochromocytoma had a positive surgical margin (p=0.296). Mean pathologic tumor size (3.3±1.9 cm vs. 3.6±2.3 cm, p=0.568) and volume (18.4±36.6 cm
3 vs. 24.5±55.7 cm
3, p=0.606) were similar in the robotic and laparoscopic groups. Mean operation time (120 minutes vs. 125 minutes, p=0.656), EBL (100 mL vs. 128 mL, p=0.309), postoperative hemoglobin change (−0.8 mg/dL vs. −0.8 mg/dL, p=0.914), and length of hospital stay (6.5 days vs. 63 days, p=0.674) were also similar in the 2 groups.
Table 2.
Perioperative characteristics of patients undergoing adrenalectomy
Variable |
Robotic (n=29) |
Laparoscopic (n=34) |
p-value |
Pathology |
|
|
0.019 |
Adrenal cortical adenoma |
10 (34.5) |
17 (50.0) |
|
Adrenal cortical carcinoma |
- |
2 (5.9) |
|
Pheochromocytoma |
6 (20.7) |
1 (2.9) |
|
Myelolipoma |
1 (3.4) |
6 (17.6) |
|
Metastasis |
7 (24.1) |
2 (5.9) |
|
Etc |
5 (17.2) |
6 (17.6) |
|
Margin status |
|
|
0.296 |
Negative |
16 (55.2) |
18 (52.9) |
|
Positive |
1 (3.4) |
- |
|
Tumor size (cm) |
3.3±1.9 |
3.6±2.3 |
0.568 |
Tumor volume (cm3) |
18.4±36.6 |
24.5±55.7 |
0.606 |
Operation time (min) |
120±47 |
125±40 |
0.656 |
Console time (min) |
67± 15 |
- |
|
Estimated blood loss (mL) |
100±97 |
128±114 |
0.309 |
Transfusion |
|
|
|
Postoperative Hb (g/dL) |
12.9±1.2 |
12.6±1.2 |
0.358 |
Postoperative Hb drop (g/dL) |
0.8+0.9 |
0.8±1.0 |
0.914 |
Hospital stay (d) |
6.5±2.0 |
6.3±2.2 |
0.674 |
Three patients in the robotic group experienced intraoperative complications (p=0.158) (
Table 3). One patient experienced a pleural injury, which was treated by chest tube insertion; and 2 patients experienced a liver injury and were treated by electrocautery. Twelve patients experienced postoperative complication, 5 in the robotic group and 7 in the laparoscopic group (p=0.545). Most of these complications were wound problems such as skin bulla. No patient experienced a Clavien-Dindo grade III or IV complication.
Table 3.
Intraoperative and postoperative complications in patients undergoing adrenalectomy
Variable |
Robotic (n=29) |
Laparoscopic (n=34) |
p-value |
Intraoperative complication |
3 (10.3) |
|
0.158 |
Liver |
2 (6.9) |
- |
|
Pleura |
1 (3.4) |
- |
|
Postoperative complication |
5 (17.2) |
7 (20.6) |
0.545 |
Wound |
4 (13.8) |
4 (11.8) |
|
Emphysema |
1 (3.4) |
|
|
Voiding |
- |
1 (2.9) |
|
Weakness |
- |
1 (2.9) |
|
Urticaria |
- |
1 (2.9) |
|
Postoperative complication, grade |
|
|
0.408 |
1 |
5 (17.2) |
7 (20.6) |
|
2 |
4 (13.8) |
4 (11.8) |
|
3-5 |
1 (3.4) |
3 (8.8) |
|
To evaluate the advantage of the robot-assisted procedure, subgroup analysis was performed relative to tumor laterality, pathologic tumor volume, and BMI (
Table 4). Operative time, EBL, postoperative hemoglobin change, and length of hospital stay did not differ significantly in patients with left- or right-sided tumors undergoing robotic or laparoscopic adrenalectomy. When subdivided by the median pathologic tumor volume of 6.8 cm
3, mean operation time in patients with tumor volume ≤6.8 cm
3 was significantly shorter in those who underwent robotic than conventional laparoscopic adrenalectomy (96 minutes vs. 125 minutes, p=0.045). When subdivided by the median BMI of 29 kg/m
2, mean operation time in patients with a BMI >29 kg/m
2 was shorter in those who underwent robotic adrenalectomy (109 minutes vs. 166 minutes, p=0.048).
Table 4.
Comparison of operation time, estimated blood loss, postoperative hemoglobin drop, and length of hospital stay in patients who underwent robotic and laparoscopic adrenalectomy subgrouped by laterality, median pathologic tumor size, and median BMI
Variable |
Right
|
Left
|
Pathologic tumor volume ≤ 6.8 m3
|
BMI > 29 kg/m2
|
Robotic |
Laparoscopic |
p-value |
Robotic |
Laparoscopic |
p-value |
Robotic |
Laparoscopic |
p-value |
Robotic |
Laparoscopic |
p-value |
No. of patients |
16 (55.2) |
13 (38.2) |
|
13 (44.8) |
21 (61.8) |
|
16 (55.2) |
15 (44.1) |
|
7 (24.1) |
5 (14.7) |
|
Operation time (min) |
131±52 |
137±42 |
0.812 |
106±38 |
117±38 |
0.344 |
96±34 |
125±37 |
0.045 |
109±45 |
166±43 |
0.048 |
Estimated blood loss (mL) |
100±46 |
137±100 |
0.374 |
101±139 |
122±124 |
0.441 |
76±63 |
108±99 |
0.495 |
107±67 |
194±78 |
0.073 |
Postoperative Hb drop (g/dL) |
0.8±0.8 |
0.9±0.8 |
0.589 |
0.9±1.0 |
0.8±1.1 |
0.780 |
0.7±0.9 |
0.7±1.0 |
0.770 |
0.2±0.9 |
0.8±0.7 |
0.268 |
Hospital stay (d) |
6.4±2.3 |
6.3±1.9 |
0.983 |
6.6±1.8 |
6.3±2.3 |
0.484 |
6.3±2.0 |
7.1±2.5 |
0.358 |
7.3±2.2 |
6.8±2.5 |
0.876 |
DISCUSSION
This current study describes the experience of our center with robotic and conventional laparoscopic adrenalectomy. Outcomes of the 2 procedures were similar, with operation time, EBL, postoperative hemoglobin change, and length of stay not differing significantly in the two groups of patients.
Since its introduction, laparoscopic adrenalectomy has become the treatment of choice for resecting adrenal tumors [
1]. This approach, however, has several disadvantages, including the limited maneuverability of the instruments and a 2-dimensional screen, making laparoscopic adrenalectomy hard to learn and requiring a long learning curve [
12]. Robotic technology has overcome some of the limitations of laparoscopic adrenalectomy [
7]. Since the first robot-assisted laparoscopic adrenalectomy was performed in 2001 with the da Vinci system [
13], several studies comparing the 2 procedures have demonstrated similar perioperative outcomes [
9,
10]. The findings presented here are in agreement with previous studies, showing that robotic adrenalectomy is equivalent to laparoscopic adrenalectomy.
The mean operation times in the robotic and laparoscopic groups were similar (120 minutes vs. 125 minutes, p=0.656), in agreement with previous studies. Mean operating times were 120 minutes each in 30 patients who underwent robotic adrenalectomy [
14] and in 100 patients who underwent laparoscopic adrenalectomy [
15]. Moreover, a comparison of 50 patients who underwent robotic and 59 patients who underwent laparoscopic adrenalectomy reported mean operation times of 104 and 87 minutes, respectively [
10]. Thus, despite the advantages of the robotic procedure, including three-dimensional magnified vision and instruments with wrist joints, the mean operation times of the 2 procedures were similar. The robot system requires a longer setup time, including for docking of the robot and insertion of more ports than the laparoscopic procedure. The mean console time for robotic adrenalectomy was 67 minutes, indicating that half of the operating time was used to set up the robotic system.
Previous studies reported similar perioperative complication rates in patients who underwent robotic and laparoscopic adrenalectomy. For example, one study found that the postoperative complication rates were 7% and 11%, respectively, and the conversion rates were 0% and 9%, respectively [
14], and a second study found that the difference in postoperative complication rate was not statistically significant [
16]. Another study reported no intraoperative complications in both groups and postoperative complication rates of 13% for robotic and 25% for laparoscopic adrenalectomy [
9]. In the current study, 3 patients in the robotic group experienced intraoperative complications, with one having a pleural injury and 2 having liver injuries. These complications occurred during the early learning curve period. The complete loss of tactile sensation associated with robotic surgery may result in compromised power control, leading to organ injury [
17]. However, the overall postoperative complication rates were similar in the robotic and laparoscopic groups (17% vs. 20%) and no patient experienced a Clavien-Dindo grade III or IV complication.
The advantages of the robotic system suggest that robot-assisted laparoscopic adrenalectomy may be more appropriate in complex cases. Robotic adrenalectomy was associated with shorter operation times in obese patients (BMI>30 kg/m
2) and those with large tumors (>55 mm), with tumor size being an independent predictor of longer operation time [
10,
18]. Similarly, a comparison of the 2 methods in 63 patients with large adrenal tumors (≥5 cm) showed that the mean operation time was shorter in the robotic group [
19]. In contrast to previous studies, the present study showed that mean operation time in patients with smaller tumors was shorter in the robotic than in the laparoscopic group. Although this result may not be clinically relevant, the robotic procedure may be better able to identify and visualize tumors, shortening operation times in patients with smaller tumors. Although BMI and operation times did not differ significantly in total cohort, operation time of robotic group was shorter in obese patients (BMI>29 kg/m
2). The limited mobility of laparoscopic instruments makes laparoscopic surgery more difficult in obese patients; the robotic system may help surgeons overcome these difficulties [
20,
21].
This study had several limitations. Owing to its retrospective design, patient information was limited and there was a potential for selection bias. The number of patients in each group was small, and oncologic outcomes were not assessed. Operations were performed by 5 surgeons, which may have affected the data. Although patients were subdivided by a median pathologic tumor volume of 6.8 cm3 and a median BMI 29 kg/m2, these cutoffs may be arbitrary and not clinically relevant. Considering the small cohort and the number of surgeons, subgroup analysis of operation time may not mean much. Because we did not consider the tumor size and BMI at the time of decision of surgical methods, these parameters did not differ significantly in total cohort. Moreover, information on pathologic tumor volume was not available when the surgical procedure was chosen. However, when identifying and visualizing tumors, pathologic tumor volume is more appropriate than clinical tumor size on imaging modalities.
In conclusion, this study showed that robot-assisted laparoscopic adrenalectomy is technically feasible, with outcomes comparable to those of laparoscopic adrenalectomy. The robotic system has several advantages, including 3-dimensional magnified vision, elimination of surgeon tremor, and extended mobility of instruments with a wrist joint, making it more suitable than the laparoscopic procedure in difficult cases. However, the indications for robotic adrenalectomy must be more clearly defined.