- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Patients with a history of prior abdominal surgery are at an increased risk of intra-abdominal adhesions [1,2] which may complicate successful completion of transperitoneal laparoscopic or robotic procedures. Previous groups have evaluated the effect of prior abdominal surgery on transperitoneal laparoscopic upper tract procedures [3–7]. The impact of prior abdominal surgery on robotic prostatectomy has also been evaluated . However, to our knowledge, the impact of previous abdominal surgery on robotic upper urinary tract procedures, such as robotic partial nephrectomy (RPN), has not been examined.
Partial nephrectomy is an established treatment for small renal masses less than 4 cm in size [9,10]. Laparoscopic partial nephrectomy (LPN) has demonstrated comparable cancer control and functional outcomes to open PN while potentially improving convalescence and perioperative outcomes [11–14]. However, LPN is technically challenging, requiring advanced laparoscopic skills for tumor excision and intracorporeal sutured reconstruction under the time constraints of warm ischemia. RPN has emerged as an alternative to LPN that may help with some of these technical challenges [15–21].
We evaluated the effect of previous abdominal surgery on perioperative outcomes in patients undergoing RPN via a transperitoneal approach. Using a large, prospectively collected database of RPN patients, we sought to answer the following questions:
What is the epidemiology of previous abdominal surgery at an experienced robotic center performing routine RPN?
What are the bowel and vascular complications associated with previous abdominal surgery during RPN?
What steps can be taken to minimize the morbidity of RPN in patients with previous abdominal surgery?
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Using a prospectively maintained, IRB approved database, we identified 197 consecutive patients who underwent robotic renal surgery from January 2004 to September 2009, of which 100 patients underwent RPN. Five patients underwent a retroperitoneal approach for RPN, leaving a total of 95 patients who underwent a transperitoneal RPN. Patients from four surgeons at our institution were included in the analysis. The da Vinci S surgical system (Intuitive Surgical Inc., SunnyVale, CA) was used in all operations.
Access to the peritoneal cavity for RPN was obtained via a Veress needle or Hasson technique . In those patients with previous abdominal surgery (Fig. 1), the Veress needle was placed in the ipsilateral abdominal quadrant farthest from the previous incision. After placing the Veress needle into the peritoneal cavity, pneumoperitoneum was established. Certain safety steps were used to confirm entry into the peritoneal cavity, including absent gas or blood at aspiration of a syringe through the Veress needle, injection of 5 cc saline that could not be aspirated, low initial intraperitoneal pressure and no rapid increase in intraperitoneal pressure at the commencement of insufflation. Direct-vision placement of the initial tocar was performed in our 10 most recent cases using an 8 mm robotic camera placed through the obturator of a 12 mm nonbladed optical trocar (Fig. 2). In cases in which peritoneal access using a Veress needle was unsuccessful, open trocar placement was performed utilizing the Hassan technique. Initial adhesiolysis was performed with laparoscopic scissors as needed to allow for placement of additional ports. A rigid nephroscope with a working channel can also be used for adhesiolysis through initial tocar prior to placement of other trocars. The remaining ports were placed as shown in (Fig. 3) and RPN was performed. Our technique of RPN has been previously described .
Figure 2. a, Robotic 8 mm camera placed through obturator of 12 mm clear-tipped trocar for direct-vision placement of the initial optical trocar. b, External view of direct-vision technique of gaining initial trocar access with 8 mm robotic camera through obturator of 12 mm optical trocar. The trocar is manually twisted to gain peritoneal entry under direct vision. c, Camera view of direct-vision placement of the initial trocar utilizing 8 mm camera.
Download figure to PowerPoint
Patients undergoing RPN were grouped according to whether or not they had undergone previous abdominal surgery. Patients with prior abdominal surgey were subcategorized as having an upper midline or ipsilateral upper quadrant scar or as having a lower abdominal, contralateral, or minimally invasive scar. Patient demographics and perioperative outcomes were compared between the surgery and no surgery groups. Statistical comparisons of continuous variables were performed using the Median test. Categorical variables were compared using Fisher’s exact test. A 5% significance level was used for all tests and all analyses were performed using commercially available software.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
During the study period a total of 197 robotic renal procedures were performed, of which 95 patients underwent transperitoneal RPN. A total of 41 (43%) patients undergoing RPN had a history of prior abdominal surgery and six had upper midline or ipsilateral upper quadrant scars. The types of abdominal surgeries patients had previously undergone are listed in Table 1. A total of six patients underwent seven types of previous upper midline or ipsilateral upper quadrant surgeries, including open cholecystectomy-2 patients (33%), open partial gastrectomy-2 patients (33%) and exploratory laparotomy-1 patient (17%). A total of 35 patients underwent 16 different types of other prior abdominal surgeries, including abdominal hysterectomy-10 patients (29%), umbilical/inguinal hernia repair-9 patients (26%), and appendectomy-7 patients (20%). The time between the initial abdominal surgery and RPN was more than one year in all patients.
Table 1. Types of previous abdominal surgeries in 41 patients who underwent transperitoneal RPN with a prior history of abdominal surgery
|Previous abdominal operation||No. of procedures (%)|
|Lower abdominal, contralat, min-invasive* (No. of Pts. = 35)|| |
| Abdominal total/partial Hysteretomy||10 (29%)|
| Umbilical/Inguinal Hernia Repair||9 (26%)|
| Appendectomy||7 (20%)|
| Cesarean Section||2 (6%)|
| Open Cholecystectomy, contralateral||3 (9%)|
| Laparoscopic Cholecystectomy||5 (14%)|
| Colectomy/Partial Colectomy||3 (8%)|
| Robotic Adrenal/Renal surgery||3 (9%)|
| Exploratory Laparotomy||1 (3%)|
| Robotic prostatectomy||3 (9%)|
| Diagnostic Laparoscopy||1 (3%)|
| Abdominal paraganglioma||5 (14%)|
| Open Partial/Radical Nephrectomy||3 (9%)|
| Enterocutaneous Fistula Surgery||1 (3%)|
| Tubal ligation||3 (9%)|
| PCNL||1 (3%)|
|Upper midline or ipsilat upper quadrant** (No. of Pts. = 6)|| |
| Open Partial Gastrectomy||2 (33%)|
| Open Cholecystectomy||2 (33%)|
| Exploratory Laparotomy||1 (17%)|
| Open Nephrectomy (Chevron)||1 (17%)|
| Abdominal Aortic Aneurysm Repair||1 (17%)|
| Splenic Artery Repair||1 (17%)|
| Drainage of Pancreatic Cyst||1 (17%)|
Table 2 lists baseline patient characteristics and perioperative outcomes for robotic partial nephrectomy in patients with prior surgery compared to patients with no surgery. There were no statistically significant differences between groups in age, sex, or BMI, median tumor size, or operative time. Patients with previous abdominal surgery were more likely to require adhesiolysis (41% vs 15%, P = 0.005). Adhesiolysis in patients with previous upper midline or ipsilateral upper quadrant surgery took a mean time of 32 min. Patients with prior abdominal surgery had a trend toward longer median warm ischemia time (21 vs 16 min) and median EBL (150 vs100 ml), although this did not reach statistical significance. There were no statistically significant differences in perioperative variables with stratification of patients with prior surgery into the two subcategories based on scar location (data not shown).
Table 2. Patient characteristics and perioperative outcomes for transperitoneal RPN based on prior surgery status
|Variables||All Surgeries (n = 41)||No Surgery (n = 54)||P value|
|Median Age, years (IQR)||61 (52–70)||59 (50–65)||0.406|
|Male Gender, n (%)||24 (58.5)||38 (70.3)||0.279|
|Median BMI, kg/m2 (IQR)||30.4 (25.2–36.3)||29.4 (26–33.6)||0.412|
|Median Tumor size, cm (IQR)||2.5 (1.5–3.3)||2.3 (1.7–3.0)||0.540|
|Median Operative Time, min (IQR)||246 (203–307)||250 (214–299)||1.000|
|Median Warm Ischemia time, min (IQR)||21 (11–25)||16 (0–25)||1.000|
|Median LOS, days (IQR)||2 (1–3)||2 (1–3)||0.513|
|Median EBL, cc (IQR)||150 (69–250)||100 (50–200)||0.144|
|Adhesiolysis, n (%)||17 (41)||8 (15)||0.005|
|Pathologic stage|| || || |
| pT1a||23||30|| |
| pT1b||2||8|| |
| pT2||0||0|| |
| pT3a||4||1|| |
Intra-operative and postoperative complications are listed in Table 3. Access related complications occurred in two patients with previous upper midline or ipsilateral upper quadrant surgery: an enterotomy during lysis of adhesions that was repaired robotically without sequelae and a mesenteric hematoma during Veress needle placement. There was no conversion from transperitoneal RPN to open partial or robotic radical nephrectomy. Complications in patients with a previous lower abdominal, contralateral, or minimally-invasive scar consisted of a postoperative urine leak – two patients (6%), bleeding – three patients (9%), and pulmonary embolism – one patient (3%). However, there was no statistically significant difference in complications between groups. There was no statistically significant difference in the proportion of medical vs surgical complications between groups. There were no access related injuries in the 10 cases in which the robotic 8 mm camera was used for initial trocar placement.
Table 3. Complications of patients with prior abdominal surgeries who underwent transperitoneal RPN
|Complications||All Surgeries (n = 41)||No Surgery (n = 54)||Other scar location (n = 35)||Upper midline or ipsilat. upper quadrant scar (n = 6)||P value|
|Intraoperative|| || || || || |
| • Enterotomy*||1 (2.4%)||0||0||1 (16.7%)||0.432|
| • Mesenteric hematoma||1 (2.4%)||0||0||1 (16.7%)|
|Postoperative|| || || || || |
| • Urine leak†||2 (5%)||0||2 (6%)||0||0.184|
| • Bleeding||3 (7%)||3 (6%)¶||3 (9%)§||0||1.000|
| • Pulmonary embolism||1 (2%)||0||1 (3%)||0||0.432|
| • Urinary retention||0||3 (6%)||0||0||0.256|
A subset analysis was performed of the five patients who underwent retroperitoneal RPN, which demonstrated a median tumor size of 2.2 cm, operative time of 259 min, warm ischemia time of 23 min, EBL of 75 ml, and length of stay of 2 days. Three of five patients (60%) had undergone prior abdominal surgery, including one patient with an upper midline and ipsilateral lower abdominal scar. There were no postoperative complications with the retroperitoneal approach.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Previous abdominal surgery is associated with formation of intra-abdominal adhesions [1,2,4], which may complicate successful completion of transperitoneal laparoscopic or robotic procedures. Previous abdominal surgery has traditionally been considered a relative contraindication to laparoscopy, and has been associated with increased operative times and complication rates [5,6]. Postmortem studies have shown that the most common cause of peritoneal adhesion formation is prior surgery . As complex urological procedures are increasingly being performed using robotic and laparoscopic techniques, the number of patients with a prior history of abdominal surgeries undergoing such procedures is likely to increase . Previous groups have evaluated the effect of prior abdominal surgery on transperitoneal laparoscopic upper tract procedures [3–7]. Robotic prostatectomy has also been evaluated in the setting of prior abdominal surgery . However, to the best of our knowledge, our study is the first to examine the impact of previous abdominal surgery on transperitoneal RPN.
The impact of previous abdominal surgery on non-urological laparoscopic procedures has been evaluated by several gropus. Earlier abdominal surgery did not have an impact on the postoperative stay or complication rates in patients who underwent laparoscopic cholecystectomy [25,26]. Approximately 5% of procedures were converted to an open procedure, with dense adhesions preventing trocar placement being the most common reason . Women undergoing laparoscopy for gynecologic procedures who had undergone prior laparotomy were also at a greater risk for access related complications . In patients undergoing laparoscopic tubal ligation, complication rates were almost twice as high in patients with previous abdominal surgery .
The impact of previous abdominal surgery on transperitoneal laparoscopic upper urinary tract procedures has been evaluated by several groups [3–7]. Some studies suggested higher complication rates in patients with prior abdominal surgery , while others showed no increase in complications in patients with a history of open abdominal or prior renal surgery [4,7]. In a study by Seifman et al. of 190 patients undergoing laparoscopic upper tract surgery, 76(40%) had undergone prior abdominal surgery . Patients with prior abdominal surgery had an increased risk of complications (16% vs 4%), which may have contributed to a longer mean hospital stay (3.8 vs 2.6 days). An upper midline scar/ipsilateral upper quadrant scar was associated with a greater access complication rate (12% vs 0%) but not a higher operative complication rate (21% vs 13%). When patients with previous surgery had a complication, the complication was more often operative than medical. In contrast, when patients without previous surgery had a complication, it was more likely to be medical than operative.
There are few reports on the effects of previous surgery on robotic surgery outcomes. Nazemi et al. published a study on 49 patients who underwent robotic surgery (25 had robotic radical prostatectomy) and found no difference in operative time, EBL, and complications when comparing those who had previous abdominal surgery vs no surgery . They found a higher incidence of peritoneal adhesions in patients with previous surgery (54% vs 10%). Almost a third (30%) of all patients had a history of previous surgery with the majority of the cases being hernia repairs or appendectomies.
In a study by Siddiqui et al. of 3950 patients who underwent transperitoneal robotic radical prostatectomy, 1049 (27%) patients had a history of abdominal or inguinal surgery . As in our study, there were no differences in EBL, total operative time, body mass index between groups. Adhesiolysis was more likely to be required in patients with prior abdominal surgery (23% vs 8%, P < 0.001). There were 4 access-related vascular injuries from Veress needle insertion, all of which were managed conservatively, with safe completion of the robotic prostatectomy within 2 weeks. Bowel injury occurred in 5 patients, 3 of which had a history of previous abdominal surgery requiring extensive adhesiolysis.
Our study is the first to analyze the impact of prior abdominal surgeries on perioperative outcomes after RPN. The incidence of peritoneal adhesions requiring adhesiolysis in our prior surgery group was (42% vs 15%) in patients without prior abdominal surgery, with achieved statistical significance (P = 0.005). Our rate of previous surgery of 43% is higher than the 30% and 27% rates reported by Nazemi et al. and Siddiqui et al., respectively, and is likely higher than what is seen in the community setting. This may be in part due to the fact that some patients are referred to us specifically because of a complex past surgical history. There was one enterotomy that occurred during lysis of adhesions in a patient with a prior major abdominal surgery and extensive adhesions, which was repaired robotically without sequelae. Some postoperative complications such as urine leak and pulmonary embolism occurred in patients with prior abdominal surgery, while bleeding occurred in both groups. However, our data do not show a statistically significant increase in complications in patients who have undergone previous abdominal surgery and there was not a statistical correlation in the location of the abdominal scar with operative complications. Also, our study did not show any increased proportion of medical complications in the no surgery group, contrary to what had been found by Seifman et al.  It is somewhat surprising that previous abdominal surgery did not seem to be associated with a statistically significant increase in operative times. Once the initial port was successfully placed, placement of other ports were relatively straightforward, even in most cases in which limited adhesiolysis was required. The direct vision port placement technique with the 8 mm camera further reduced potential delay in placing ports in the setting of prior abdominal surgery. Another possible explanation is that there could be a selection bias for more complex partial nephrectomy cases being done with an open extraperitoneal approach if dense adhesions were anticipated from prior abdominal surgery as opposed to attempting robotically for easier tumors, which might nullify extra time spent during adhesiolysis.
We used standard techniques for obtaining intraperitoneal access. Although many devices and techniques have been developed to decrease the morbidity of access injuries, no device is perfectly safe and there is no consensus regarding the optimal choice. Radially expanding trocars may be associated with fewer trocar injuries compared with bladed trocars [31,32]. Even the Hassan cannula can cause a major vascular injury at laparoscopy . Lecuru et al. described the feasibility and safety of initial blind intraperitoneal access . Patients with prior abdominal surgery had a higher rate of complications compared to patients without prior surgery. Audebert et al. determined the rate of umbilical adhesions at laparoscopy to be significantly higher in women with previous laparotomy . They recommended preliminary inspection with a microlaparoscope through the left upper quadrant and insertion of the umbilical trocar under direct vision. Vilos et al. recommended left upper quadrant access in those with suspected periumbilical adhesions .
The use of an optical access trocar technique for initial access was described by Thomas et al. in 1283 urological laparoscopic surgeries . Trocar injuries occurred in 4 patients (0.31%), but were recognized in 3 patients. They concluded that optical access trocars may result in fewer access-related complications or at least improved recognition if they do occur. We sought to recapitulate this technique using a robotic platform. The standard robotic 12 mm camera prevents the traditional application of this technique, as the camera is too large to fit through the obturator of an optical trocar. Therefore, setup of a separate conventional laparoscopic camera would be required in order to utilize this technique. We describe a modification of this technique utilizing the 8 mm robotic camera for direct-vision trocar placement that allows the vision trocar technique to be replicated using a robotic platform without the need for a separate laparoscopic setup. There were no access related injuries in the 10 cases in which the robotic 8 mm camera was used for initial trocar placement.
Limitations of our study include a single institution analysis with relatively low number of patients and the retrospective nature of the study. Other potential causes for adhesions, such as radiation therapy and inflammatory disorders, were not considered in this analysis. This study is based on a large cohort of patients at a major tertiary referral center. We also included multiple surgeons with varying level of experience and learning curve could potentially influence the results. In our study, operative experience did not decrease the complication rate, possibly due in part to liberalizing our criteria for patient selection as our experience increased. Consequently, the results may not be applicable to the urologic community at large. Our study focused on a transperitoneal approach for RPN, although we did include a subset analysis of patients who underwent a retroperitoneal RPN and we have previously described our experience with a retroperitoneal approach for robotic kidney surgery . Patients who underwent a retroperitoneal approach had similar perioperative outcomes. A retroperitoneal approach for RPN is feasible and may be another potential strategy for patients with prior abdominal surgery, particularly for posterior tumors. Our 8 mm robotic camera had a slightly darker view than the standard 12 mm robotic camera, so we only used it to obtain initial access and switched to the 12 mm camera for the rest of the case to optimize visualization. However, with upcoming improvements in the 8 mm robotic camera quality over time, it may be more practical to continue to use the 8 mm camera for the rest of the case. Our 8 mm robotic camera had been purchased for pediatric cases. For institutions that do not already have the 8 mm camera, it is debatable as to whether the ability to perform direct trocar placement warrants the purchase of a separate 8 mm robotic camera.