j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 4 ( 2 0 1 5 ) 3 6 1 e3 6 6
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Postoperative pancreatic fistula after robot distal gastrectomy Ho Seok Seo, MD, Jung Ho Shim, MD, PhD, Hae Myung Jeon, MD, PhD, Cho Hyun Park, MD, PhD, and Kyo Young Song, MD, PhD* Division of Gastrointestinal Surgery, Department of Surgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
article info
abstract
Article history:
Background: To compare the incidences of postoperative pancreatic fistula (POPF) between
Received 15 July 2014
robot-assisted distal gastrectomy (RADG) and laparoscopy-assisted distal gastrectomy (LADG).
Received in revised form
Materials and methods: A total of 40 patients with gastric cancer who underwent RADG were
9 September 2014
compared with 40 initial patients who underwent LADG by a single surgeon. We evaluated
Accepted 17 October 2014
and compared the clinicopathologic characteristics, surgical outcomes, and operative
Available online 22 October 2014
complications including POPF in two groups. Results: The POPF was observed more frequently in the LADG group than in the RADG group
Keywords:
(22.5% versus 10%, P < 0.001). Although the serum amylase levels in the 20 first-half cases
Robot-assisted distal gastrectomy
did not statistically differ between LADG and RADG (P ¼ 0.32), those in the 20 latter-half
Laparoscopic-assisted distal
cases were significantly lower in the RADG group (P < 0.05). Univariate and multivariate
gastrectomy
analyses identified laparoscopic surgery and visceral fat area as POPF-associated risk
Postoperative pancreatic fistula
factors.
Gastric cancer
Conclusions: RADG is feasible and safe for distal gastrectomy in terms of POPF. ª 2015 Elsevier Inc. All rights reserved.
1.
Introduction
Although the incidence of gastric cancer has declined in recent decades [1], it remains the second-leading cause of death worldwide. In fact, in Korea and Japan, the incidence of early gastric cancer has increased [2,3]. Early detection of gastric cancer enhances the feasibility and facilitates the development of minimally invasive surgical options. Laparoscopic gastrectomy with lymph node dissection, for instance, has been established as standard treatment for gastric cancer [4,5]. It can aid recovery of bowel functionality and minimize postoperative pain, whose benefits have profoundly positive effects on patients’ clinical recovery. Several retrospective and prospective studies have confirmed the operative safety of laparoscopy-assisted distal gastrectomy (LADG) and its compatibility with open gastrectomy [6e8].
However, LADG with radical lymph node dissection, specifically as performed in and around the suprapancreatic area, remains challenging even for experienced surgeons, because of factors such as the limited range of movement, amplification of hand tremors, two-dimensional imaging, and inconvenient surgical positioning. Major postoperative complications, such as bleeding or aggravated postoperative pancreatic fistula (POPF), are not uncommon results [9,10]. As a means of overcoming the limitations and drawbacks of LADG, a new-concept robotic surgery system entailing three-dimensional imaging, tremor filtering, and ergonomic instrumentation has been introduced. This system has inaugurated a new era of minimally invasive surgery; its feasibility and benefits have been demonstrated in a variety of surgical fields including urological and cardiac surgery, among others [11e13].
* Corresponding author. Division of Gastrointestinal Surgery, Department of Surgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Seocho-gu, Seoul 137-701, Korea. Tel.: þ82 2 2258 2876; fax: þ82 2 595 2822. E-mail addresses: [email protected], [email protected] (K.Y. Song). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.10.022
362
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 4 ( 2 0 1 5 ) 3 6 1 e3 6 6
Recently, robot-assisted distal gastrectomy (RADG) has been trialed for application to patients with gastric cancer. Thus far, studies comparing robotic with laparoscopic surgery in this regard have been few. The benefits of RADG over LADG remain controversial; it’s considerably greater expense is one of the most important factors determining the use or non-use of the robotic system in gastric cancer surgery [14,15]. In this study, to elucidate the benefits of RADG, we examined post-RADG clinical outcomes with surgical complications including POPF. Additionally, the RADG outcomes were compared with those of LADG in patients with gastric cancer.
2.
Material and methods
2.1.
Subjects and study design
Patient data were prospectively collected from The Gastric Cancer Patients Registry maintained by Seoul St. Mary’s Hospital in Seoul, South Korea. A total of 80 cases were selected: 40 consecutive cases of LADG and RADG, respectively. Both type of surgery was performed by a single surgeon (K.Y.S.) and has been begun on a different period; LADG June 2004, RADG March 2009. Because the surgeon had performed more than 500 conventional open gastrectomies and 350 LADG before undertaking RADG, the initial 40 cases of LADG and RADG were compared to minimize the effect of the learning curve. The indication of both types of surgery was basically same and RADG has been chosen only when patient selected it. At our institution, laparoscopic or robotic gastrectomy has been performed for patients with clinical T1N0, T1N1, or T2N0 at preoperative staging. Preoperative staging was performed by endoscopy, endoscopic ultrasonography, and computed tomography. The following parameters were recorded prospectively: age and gender, body mass index (BMI), visceral fat area (VFA), tumor characteristics, docking times for RADG, operative time, reconstruction time, estimated blood loss, postoperative hospital stay, postoperative complications (denoted according to the ClavieneDindo classification system) [16], and postoperative white blood cell- and amylase serum levels. The VFA were preoperatively measured (both in squared centimeters) using a cross-sectional computed tomography scan at the level of the umbilicus. The Institutional Review Board of Seoul St. Mary’s Hospital approved this study (KC13RISI0029).
2.2.
Definition of pancreatic fistula
The serum amylase and drain amylase level was checked on postoperative day 3 and rechecked later if suspicious for pancreatic fistula. POPF was defined according to the International Study Group on Pancreatic Fistula definition: output, via an operatively implanted drain, of any measurable volume of drain fluid on or after postoperative day 3, with an amylase content greater than three times the upper normal serum value [17]. The POPF grading system used was as follows: grade A, “transient fistula” with no significant impact on postoperative course; grade B, change in management or clinical pathway required; and grade C, significant change in management or clinical pathway required, even aggressive intervention considered. In this study, grade B or C was regarded as significant.
2.3.
Operative procedure
The procedure of LADG was performed as described previously [6]. In RADG, it is essential that the central axis of the robot be in line with the head-to-leg axis of the patient. And, when this docking process is properly completed, each robot arm must be parallel to that axis. Furthermore, because the operating table cannot be changed once the robot has been docked, the height and slope of the operating table must be reconfirmed beforehand. The two 8-mm ports on the right side of the patient were used for Cadiere forceps and an ultrasonic shear, and the one on the left side was used for bipolar Maryland forceps. The procedure is similar to the previously reported one [18]: partial omentectomy, resection margin >2 cm from the lesion, and at least D1 þ b (including 8a, 9) lymphadenectomy based on Japanese Gastric Cancer treatment guidelines, and D2 lymphadenectomy [6,19]. The robot-arm positioning and utilization procedure for suprapancreatic lymph node dissection is as follows. First, Cardiere forceps are used as a third arm to dissect lymph node-bearing tissues around the common hepatic artery and portal vein while applying constant traction toward the right side of the patient; with Maryland forceps, the operator holds the soft tissues of the portal vein medially, and a harmonic scalpel is used to dissect the lymph nodes around the vessels (Figure A). Second, Cardiere forceps are set as a third arm to lift up the lymph nodebearing tissues around the left gastric artery; the Maryland forceps and a harmonic scalpel are used to dissect the soft tissues (Figure B). Third, Cardiere forceps are applied for constant compression of the upper border of the pancreas in approaching the splenic artery and vein, where lymph node station number 11 is located (Figure C). Billroth I or II reconstruction was used to facilitate the confluence of the gastrointestinal tract. Roux-en Y reconstruction was used for the later periods of RADG, and intracorporeal reconstruction was facilitated by the laparoscopic procedure even for the RADG.
2.4.
Statistical analysis
Post-normality test and categorical-variable analysis were done by chi-square test, continuous variables were analyzed using the ManneWhitney test or independent sample t-test. Risk factors that might complicate POPF were investigated by univariate analysis, and a multivariate analysis was performed by logistic regression. All the statistical analyses were carried out using SPSS version 17 (SPSS, Chicago, IL), and a P value 0.05).
3.2.
Surgical performance
The mean operation time was slightly longer in the RADG group (243 versus 224 min, P ¼ 0.06), though the difference was not statistically significant. Estimated blood loss was significantly smaller in the RADG group (76 versus 227 mL, P < 0.05). However, the extent of lymph node dissection or total number of retrieved the lymph nodes were not different between two groups. When we looked at the retrieved node counts at each station, there was difference only at numbers 1e3 (lesser curvature) lymph nodes (14.6 versus 10.1 in RADG and LADG, respectively, P ¼ 0.045).
3.3.
3.4.
Postoperative pancreatic fistula
The POPF observed in 13 patients in both groups and in most of them (69.2%) were grade A or transient fistula, which has no clinical impact. The incidence of POPF was significantly different between two groups (22.5 versus 10% (P < 0.05) in LADG and RADG groups, respectively).
3.5.
Risk factors for POPF
According to the univariate analysis, BMI, VFA, and operation type were significantly associated with POPF (Table 3). Furthermore, the multivariate analysis identified operation type (P ¼ 0.006; odds ratio [OR], 2.347; 95% confidence interval [CI], 1.411e12.3) and VFA (P ¼ 0.034; OR, 2251; 95% CI, 1.321e7.391) as an independent risk factor associated with POPF (Table 4).
Early postoperative outcomes and complications
Table 2 shows the early postoperative outcomes and complications in the RADG and LADG groups. The hospital stays were shorter, and the diet was started earlier, in the RADG group (P < 0.05). Moreover, the postoperative white blood cell count and amylase serum levels were lower in the RADG group (P < 0.05).
4.
Discussion
In the previous comparative reports, the surgical outcomes between RADG and LADG have been equivocal or even, in some cases, have favored LADG over RADG. The merits of robot-assisted surgery, such as three-dimensional vision,
Table 1 e Patient clinicopathologic characteristics. Clinicopathologic characteristics Mean age (y SD) Gender (M/F) BMI (kg/m2SD) ECOG performance 1 2 Comorbidity Previous abdominal surgery Stage I II III Extent of LN dissection D1þb D2 Total number of retrieved LNs
LAG (N ¼ 40)
RAG (N ¼ 40)
P value
55.1 5.1 20/20 23.8 1.9
51.6 4.5 19/21 23.6 2.1
0.046 0.621 0.819 0.405
38 2 14 1
37 3 20 2
37 2 1
36 3 1
23 17 35.4 2.3
22 18 40.4 3.3
0.201 0.794 0.649
0.382
0.051
ECOG ¼ Eastern Cooperative Oncology Group; LDG ¼ laparoscopic distal gastrectomy; LN ¼ lymph node; RDG ¼ robotic distal gastrectomy; SD ¼ standard deviation.
364
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 4 ( 2 0 1 5 ) 3 6 1 e3 6 6
Table 2 e Early postoperative outcomes. Variable Hospital stays (d SD) Starting of sips of water (d SD) Starting of liquid diet (d SD) WBC count (uL) POD# 1 POD# 3 Serum amylase (U/L) POD# 1 POD# 3 Postoperative complication Duodenal stump leak Atelectasis Pneumonia Surgical site infection Gastric stasis Adhesive ileus POPF (grade) A B C
LADG (n ¼ 40)
RADG (n ¼ 40)
P value
7.37 1.5 3.71 0.5 4.30 0.7
6.75 1.7 2.63 0.6 3.52 0.5
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
Postoperative pancreatic fistula after robot distal gastrectomy Ho Seok Seo, MD, Jung Ho Shim, MD, PhD, Hae Myung Jeon, MD, PhD, Cho Hyun Park, MD, PhD, and Kyo Young Song, MD, PhD* Division of Gastrointestinal Surgery, Department of Surgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
article info
abstract
Article history:
Background: To compare the incidences of postoperative pancreatic fistula (POPF) between
Received 15 July 2014
robot-assisted distal gastrectomy (RADG) and laparoscopy-assisted distal gastrectomy (LADG).
Received in revised form
Materials and methods: A total of 40 patients with gastric cancer who underwent RADG were
9 September 2014
compared with 40 initial patients who underwent LADG by a single surgeon. We evaluated
Accepted 17 October 2014
and compared the clinicopathologic characteristics, surgical outcomes, and operative
Available online 22 October 2014
complications including POPF in two groups. Results: The POPF was observed more frequently in the LADG group than in the RADG group
Keywords:
(22.5% versus 10%, P < 0.001). Although the serum amylase levels in the 20 first-half cases
Robot-assisted distal gastrectomy
did not statistically differ between LADG and RADG (P ¼ 0.32), those in the 20 latter-half
Laparoscopic-assisted distal
cases were significantly lower in the RADG group (P < 0.05). Univariate and multivariate
gastrectomy
analyses identified laparoscopic surgery and visceral fat area as POPF-associated risk
Postoperative pancreatic fistula
factors.
Gastric cancer
Conclusions: RADG is feasible and safe for distal gastrectomy in terms of POPF. ª 2015 Elsevier Inc. All rights reserved.
1.
Introduction
Although the incidence of gastric cancer has declined in recent decades [1], it remains the second-leading cause of death worldwide. In fact, in Korea and Japan, the incidence of early gastric cancer has increased [2,3]. Early detection of gastric cancer enhances the feasibility and facilitates the development of minimally invasive surgical options. Laparoscopic gastrectomy with lymph node dissection, for instance, has been established as standard treatment for gastric cancer [4,5]. It can aid recovery of bowel functionality and minimize postoperative pain, whose benefits have profoundly positive effects on patients’ clinical recovery. Several retrospective and prospective studies have confirmed the operative safety of laparoscopy-assisted distal gastrectomy (LADG) and its compatibility with open gastrectomy [6e8].
However, LADG with radical lymph node dissection, specifically as performed in and around the suprapancreatic area, remains challenging even for experienced surgeons, because of factors such as the limited range of movement, amplification of hand tremors, two-dimensional imaging, and inconvenient surgical positioning. Major postoperative complications, such as bleeding or aggravated postoperative pancreatic fistula (POPF), are not uncommon results [9,10]. As a means of overcoming the limitations and drawbacks of LADG, a new-concept robotic surgery system entailing three-dimensional imaging, tremor filtering, and ergonomic instrumentation has been introduced. This system has inaugurated a new era of minimally invasive surgery; its feasibility and benefits have been demonstrated in a variety of surgical fields including urological and cardiac surgery, among others [11e13].
* Corresponding author. Division of Gastrointestinal Surgery, Department of Surgery, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 505 Banpo-dong, Seocho-gu, Seoul 137-701, Korea. Tel.: þ82 2 2258 2876; fax: þ82 2 595 2822. E-mail addresses: [email protected], [email protected] (K.Y. Song). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2014.10.022
362
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 4 ( 2 0 1 5 ) 3 6 1 e3 6 6
Recently, robot-assisted distal gastrectomy (RADG) has been trialed for application to patients with gastric cancer. Thus far, studies comparing robotic with laparoscopic surgery in this regard have been few. The benefits of RADG over LADG remain controversial; it’s considerably greater expense is one of the most important factors determining the use or non-use of the robotic system in gastric cancer surgery [14,15]. In this study, to elucidate the benefits of RADG, we examined post-RADG clinical outcomes with surgical complications including POPF. Additionally, the RADG outcomes were compared with those of LADG in patients with gastric cancer.
2.
Material and methods
2.1.
Subjects and study design
Patient data were prospectively collected from The Gastric Cancer Patients Registry maintained by Seoul St. Mary’s Hospital in Seoul, South Korea. A total of 80 cases were selected: 40 consecutive cases of LADG and RADG, respectively. Both type of surgery was performed by a single surgeon (K.Y.S.) and has been begun on a different period; LADG June 2004, RADG March 2009. Because the surgeon had performed more than 500 conventional open gastrectomies and 350 LADG before undertaking RADG, the initial 40 cases of LADG and RADG were compared to minimize the effect of the learning curve. The indication of both types of surgery was basically same and RADG has been chosen only when patient selected it. At our institution, laparoscopic or robotic gastrectomy has been performed for patients with clinical T1N0, T1N1, or T2N0 at preoperative staging. Preoperative staging was performed by endoscopy, endoscopic ultrasonography, and computed tomography. The following parameters were recorded prospectively: age and gender, body mass index (BMI), visceral fat area (VFA), tumor characteristics, docking times for RADG, operative time, reconstruction time, estimated blood loss, postoperative hospital stay, postoperative complications (denoted according to the ClavieneDindo classification system) [16], and postoperative white blood cell- and amylase serum levels. The VFA were preoperatively measured (both in squared centimeters) using a cross-sectional computed tomography scan at the level of the umbilicus. The Institutional Review Board of Seoul St. Mary’s Hospital approved this study (KC13RISI0029).
2.2.
Definition of pancreatic fistula
The serum amylase and drain amylase level was checked on postoperative day 3 and rechecked later if suspicious for pancreatic fistula. POPF was defined according to the International Study Group on Pancreatic Fistula definition: output, via an operatively implanted drain, of any measurable volume of drain fluid on or after postoperative day 3, with an amylase content greater than three times the upper normal serum value [17]. The POPF grading system used was as follows: grade A, “transient fistula” with no significant impact on postoperative course; grade B, change in management or clinical pathway required; and grade C, significant change in management or clinical pathway required, even aggressive intervention considered. In this study, grade B or C was regarded as significant.
2.3.
Operative procedure
The procedure of LADG was performed as described previously [6]. In RADG, it is essential that the central axis of the robot be in line with the head-to-leg axis of the patient. And, when this docking process is properly completed, each robot arm must be parallel to that axis. Furthermore, because the operating table cannot be changed once the robot has been docked, the height and slope of the operating table must be reconfirmed beforehand. The two 8-mm ports on the right side of the patient were used for Cadiere forceps and an ultrasonic shear, and the one on the left side was used for bipolar Maryland forceps. The procedure is similar to the previously reported one [18]: partial omentectomy, resection margin >2 cm from the lesion, and at least D1 þ b (including 8a, 9) lymphadenectomy based on Japanese Gastric Cancer treatment guidelines, and D2 lymphadenectomy [6,19]. The robot-arm positioning and utilization procedure for suprapancreatic lymph node dissection is as follows. First, Cardiere forceps are used as a third arm to dissect lymph node-bearing tissues around the common hepatic artery and portal vein while applying constant traction toward the right side of the patient; with Maryland forceps, the operator holds the soft tissues of the portal vein medially, and a harmonic scalpel is used to dissect the lymph nodes around the vessels (Figure A). Second, Cardiere forceps are set as a third arm to lift up the lymph nodebearing tissues around the left gastric artery; the Maryland forceps and a harmonic scalpel are used to dissect the soft tissues (Figure B). Third, Cardiere forceps are applied for constant compression of the upper border of the pancreas in approaching the splenic artery and vein, where lymph node station number 11 is located (Figure C). Billroth I or II reconstruction was used to facilitate the confluence of the gastrointestinal tract. Roux-en Y reconstruction was used for the later periods of RADG, and intracorporeal reconstruction was facilitated by the laparoscopic procedure even for the RADG.
2.4.
Statistical analysis
Post-normality test and categorical-variable analysis were done by chi-square test, continuous variables were analyzed using the ManneWhitney test or independent sample t-test. Risk factors that might complicate POPF were investigated by univariate analysis, and a multivariate analysis was performed by logistic regression. All the statistical analyses were carried out using SPSS version 17 (SPSS, Chicago, IL), and a P value 0.05).
3.2.
Surgical performance
The mean operation time was slightly longer in the RADG group (243 versus 224 min, P ¼ 0.06), though the difference was not statistically significant. Estimated blood loss was significantly smaller in the RADG group (76 versus 227 mL, P < 0.05). However, the extent of lymph node dissection or total number of retrieved the lymph nodes were not different between two groups. When we looked at the retrieved node counts at each station, there was difference only at numbers 1e3 (lesser curvature) lymph nodes (14.6 versus 10.1 in RADG and LADG, respectively, P ¼ 0.045).
3.3.
3.4.
Postoperative pancreatic fistula
The POPF observed in 13 patients in both groups and in most of them (69.2%) were grade A or transient fistula, which has no clinical impact. The incidence of POPF was significantly different between two groups (22.5 versus 10% (P < 0.05) in LADG and RADG groups, respectively).
3.5.
Risk factors for POPF
According to the univariate analysis, BMI, VFA, and operation type were significantly associated with POPF (Table 3). Furthermore, the multivariate analysis identified operation type (P ¼ 0.006; odds ratio [OR], 2.347; 95% confidence interval [CI], 1.411e12.3) and VFA (P ¼ 0.034; OR, 2251; 95% CI, 1.321e7.391) as an independent risk factor associated with POPF (Table 4).
Early postoperative outcomes and complications
Table 2 shows the early postoperative outcomes and complications in the RADG and LADG groups. The hospital stays were shorter, and the diet was started earlier, in the RADG group (P < 0.05). Moreover, the postoperative white blood cell count and amylase serum levels were lower in the RADG group (P < 0.05).
4.
Discussion
In the previous comparative reports, the surgical outcomes between RADG and LADG have been equivocal or even, in some cases, have favored LADG over RADG. The merits of robot-assisted surgery, such as three-dimensional vision,
Table 1 e Patient clinicopathologic characteristics. Clinicopathologic characteristics Mean age (y SD) Gender (M/F) BMI (kg/m2SD) ECOG performance 1 2 Comorbidity Previous abdominal surgery Stage I II III Extent of LN dissection D1þb D2 Total number of retrieved LNs
LAG (N ¼ 40)
RAG (N ¼ 40)
P value
55.1 5.1 20/20 23.8 1.9
51.6 4.5 19/21 23.6 2.1
0.046 0.621 0.819 0.405
38 2 14 1
37 3 20 2
37 2 1
36 3 1
23 17 35.4 2.3
22 18 40.4 3.3
0.201 0.794 0.649
0.382
0.051
ECOG ¼ Eastern Cooperative Oncology Group; LDG ¼ laparoscopic distal gastrectomy; LN ¼ lymph node; RDG ¼ robotic distal gastrectomy; SD ¼ standard deviation.
364
j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 4 ( 2 0 1 5 ) 3 6 1 e3 6 6
Table 2 e Early postoperative outcomes. Variable Hospital stays (d SD) Starting of sips of water (d SD) Starting of liquid diet (d SD) WBC count (uL) POD# 1 POD# 3 Serum amylase (U/L) POD# 1 POD# 3 Postoperative complication Duodenal stump leak Atelectasis Pneumonia Surgical site infection Gastric stasis Adhesive ileus POPF (grade) A B C
LADG (n ¼ 40)
RADG (n ¼ 40)
P value
7.37 1.5 3.71 0.5 4.30 0.7
6.75 1.7 2.63 0.6 3.52 0.5
