Review of 500 single incision laparoscopic colorectal surgery cases - Lessons learned

Abstract

Single incision laparoscopic surgery (SILS) is a minimally invasive platform with specific benefits over traditional multiport laparoscopic surgery. The safety and feasibility of SILS has been proven, and the applications continue to grow with experience. After 500 cases at a high-volume, single-institution, we were able to standardize instrumentation and operative steps, as well as develop adaptations in technique to help overcome technical and ergonomic challenges. These technical adaptations have allowed the successful application of SILS to technically difficult patient populations, such as pelvic cases, inflammatory bowel disease cases, and high body mass index patients. This review is a frame of reference for the application and wider integration of the single incision laparoscopic platform in colorectal surgery.

Key Words: Laparoscopic colectomy, Minimally invasive colorectal surgery, Single-incision laparoscopic surgery

Core tip: As the surgery paradigm progresses towards natural orifice transluminal endoscopic surgery, single incision laparoscopic surgery (SILS) is a valuable minimally invasive platform with specific clinical benefits. After experience from 500 SILS cases, we were able to standardize instrumentation and operative steps, and have developed technical adaptations for successful application in difficult patient populations. This review is a frame of reference for the application and wider integration of the single incision laparoscopic platform in colorectal surgery.

INTRODUCTION

Minimally invasive platforms continue to progress towards incisionless, natural orifice transluminal endoscopic surgery (NOTES) to reduce surgical trauma[1]. NOTES remains in experimental stages, and there is ongoing work is still to refine techniques[2]. In the meantime, single incision laparoscopic surgery (SILS) is the closest platform available for routine clinical practice. SILS, also referred to as Single Port Access (SPA) and laparoendoscopic single-site surgery (LESS), combines the cosmetic advantage and reduced incision principles of NOTES with the familiarity of performing surgical procedures with standard surgical instruments[3]. SILS was first reported in 1999 for cholecystectomy[4], then successfully applied to colorectal surgery in 2008[3,5]. Since that time, SILS has been proven safe and feasible for both benign and malignant colorectal disease[6-15]. In fact, SILS has been suggested as a method to improve the outcomes of traditional multiport laparoscopic surgery with reduced tissue trauma, reduced perioperative pain and narcotic use, improved cosmesis, lower rates of port-site related complications, and shorter lengths of stay[9,14,16-20].

SILS has been reported as the next major advance in the progression of minimally invasive surgical approaches feasible in generalized use[21,22]. Given the published support and our own outcomes, SILS has been the main minimally invasive platform used in our practice since 2008. Since that time, we have completed more than 700 single incision laparoscopic colorectal resection cases, continuously learning from our experiences, expanding the application of the SILS platform, and redefining our technique[23-31].

During the initial 500 cases at a high-volume, single-institution, we were able to standardize instrumentation and operative steps, as well as develop adaptations in technique to help overcome technical and ergonomic challenges. The goal is for this review is to be used as a frame of reference for the application and wider integration of the single incision laparoscopic platform in colorectal surgery.

DATABASE REVIEW

After Institutional Review Board approval, a prospective divisional database was reviewed for elective colorectal resection cases from July 2009 to the present. The initial 500 consecutive cases performed through a single incision laparoscopic approach were identified and included in the analysis. All cases were performed by 1 of 2 board-certified colorectal surgeons; both surgeons were trained in laparoscopy and moderate experienced (at least 150 colorectal resections and approximately 50 laparoscopic colorectal resection) at the start of the study period. Patients were excluded if under 18 years of age, had incomplete medical records, underwent a non-resection or stoma closure procedure, or case performed through a multiport, hand-assisted laparoscopy or open approach. Cases converted to multiport, hand-assisted laparoscopy or open surgeries were included for intention to treat analysis, and the reason for conversion was assessed.

Patient demographic, perioperative procedural, and short-term outcome data was evaluated. Data fields collected included age, gender, body mass index (BMI), ASA class, operative type (emergent/elective), operative procedure, operative time, blood loss, Intensive Care Unit stay required, length of stay, and postoperative complications, readmission, reoperation, and mortality rates. Complications were graded using the validated Clavien-Dindo Classification[32]. All patients were managed postoperatively with a standardized multimodal Enhanced Recovery Protocol (ERP). Alvimopan was used routinely in the ERP from 2011 onward. In 2013, local wound infiltration with long acting liposomal bupivacaine (EXPAREL^®, Pacira Pharmaceuticals, Inc., Parsippany, NJ) were performed at the port site at the completion of each case. In 2014, a post-induction pre-incision Transvs Abdominus Plane block with long acting liposomal bupivacaine was also routinely added to the protocol.

GENERAL TECHNICAL STEPS

For SILS procedures, there are several commercially produced access platforms available, with the most commonly used being the SILS™ Port (Covidien, Mansfield, Massachusetts, United States) and the GelPOINT^® platform (Applied Medical, Rancho Santa Margarita, California, United States).

Our operative steps for common SILS colorectal procedures have been previously described and standardized[25,27-29,31]. Port placement is critical, as proper placement can facilitate dissection, visualization, and the overall technical ease of a single-incision operation. Most procedures access the abdomen through a 2.5 cm skin incision at the umbilicus, allowing the incision to be “hidden” and giving full access to multiple quadrants. The umbilical stump is divided to expose the underlying fascia. The stump is completely detached from the fascia, allowing intraperitoneal access through the natural umbilical hernia defect, which is opened to 4 cm. The fascial incision can be opened further as needed without extending the overlying skin incision. For pelvic or cases requiring access to multiple quadrants, a Pfannenstiel incision may be favorable. In these cases, a 4 cm skin incision is made, and the underlying fascia is opened to 4 cm. In cases where a stoma is planned, the predetermined ostomy site may also be used for the SILS port, resulting in “scarless” surgery[7,33-35].

The sleeve of the access device (GelPOINT, Applied Medical, Rancho Santa Margarita, CA) is placed into the abdominal cavity, and a lap sponge is introduced to aid in small bowel retraction and cleaning the camera lens throughout the procedure. The instrumentation is also standardized. Three trocars are placed through the device - 1 for the camera, 1 for the atraumatic bowel grasper, and 1 for the energy device, the cap is secured, and pneumoperitoneum is created. As needed for retraction, an additional atraumatic grasper is placed in the 4^th trocar site. A 30° 5-mm camera with a right-angle light cord adaptor is used to allow rotation without interfering with the operator and to reduce external collisions. An experienced camera driver is helpful, as knowing tricks to decrease collisions and optimize visualization without forced movements can reduce the risk of inadvertent injury. The SILS set-up is seen is Figure 1.

Figure 1 Single incision laparoscopic surgery colorectal set-up.

A medial-to-lateral approach to bowel mobilization is performed with early identification and isolation of the vascular pedicle. We use a hand-over-fist triangulation technique for developing the retroperitoneal plane. With this technique, the nondominant hand hold an atraumatic grasper to lift the tissue, while the dominant hand holds an energy device to systematically advance the dissection through the retroperitoneal plane[28]. Once critical structures are identified, the vascular pedicle is divided and the retroperitoneal plane is further developed. When the full mobilization of the retroperitoneal plane has been achieved, a lateral-to-medial mobilization is performed to complete mobilization. In the lateral-to-medial approach, the non-dominant instrument retracts the colon medially while the dominant instrument takes down the lateral peritoneal attachments. When the intended resection segment is fully mobilized, the segment is extracted through the access device. For right-sided lesions and transverse colectomy, an extracorporeal anastomosis is performed, while for procedures requiring an anastomosis to the rectum, an intracorporeal circular stapled anastomosis is completed under laparoscopic visualization.

Overall experience

Between 2009 and 2014, we performed our initial 500 SILS colorectal resections. In our unselected case series, the mean patient age was 57.6 (SD 15.4) years, and the mean BMI was 26.5 (SD 4.7) kg/m². More than half of the patients (57.6%) had prior abdominal surgery. The main indications for operations were colon cancer (n = 156, 31.2%), diverticulitis (n = 133, 26.6%), and an unresectable polyp (n = 114, 22.8%). The main procedures performed were a right hemicolectomy/ileocolic resection (n = 197, 39.4%), an anterior rectosigmoidectomy (n = 134, 26.8%), and a Low Anterior Resection (n = 69, 13.8%) (Table 1). The mean operative time was 149.9 (SD 62.0) min. The average blood loss was 64.1 (SD 49.8) mL. Eleven patients had an intraoperative complication (enterotomy n = 4, bleeding n = 3, cystotomy n = 2, splenic injury n = 2, thermal injury n = 1). Conversion from SILS was required in 34 (6.8%) of these cases. The reasons for conversion were adhesions (n = 13), obesity/difficult reach (n = 5), poor visualization (n = 5), dense specimen (n = 4), enterotomy (n = 3), bleeding (n = 2), and anastomotic failure (n = 2).

Table 1 Diagnosis and procedure for 1^st 500 single incision laparoscopic surgery cases n (%).

Diagnosis
Colon cancer	156 (31.2)
Diverticulitis	133 (26.6)
Polyp	114 (22.8)
IBD	47 (9.4)
Constipation	19 (3.8)
Rectal cancer	18 (3.6)
Incontinence/prolapse	4 (0.8)
Endometriosis	2 (0.4)
Other	7 (1.4)
Procedure
Right hemicolectomy/ileocolic resection	197 (39.4)
Anterior rectosigmoidectomy	134 (26.8)
Low anterior resection	69 (13.8)
Total/subtotal abdominal colectomy	47 (9.4)
Sigmoid/left Hemicolectomy	31 (6.2)
Transverse colectomy	9 (1.8)
Ileal pouch anal anastomosis	9 (1.8)
Small bowel resection	4 (0.8)

IBD: Inflammatory bowel disease.

During the 500 case series, Alvimopan was used in 396 patients (79.2%), local wound infiltration with long acting liposomal bupivacaine was used in 104 patients (20.8%), and a post-induction pre-incision Trans vs Abdominus Plane block with local wound infiltration using long acting liposomal bupivacaine was used in 138 patients (27.6%).

Postoperatively, 49 patients had a complication (9.8%). Twelve were Clavien Class 1 (ileus, n = 7; dehydration, n = 5), 16 were Clavien Class 2 (superficial Site Infection, n = 6; bleeding, n = 5; and urinary tract infection, fever, pulmonary embolism, abdominal pain, and C. Difficile infection, all n = 1); 18 were class 3 (anastomotic leak, n = 9; intrabdominal abscess, n = 8, perforation, n = 1), 2 were class 4 (stroke and cardiac issue), and one Class 5 (pulmonary). Twenty-two patients were readmitted (4.4%), and 10 underwent an unplanned reoperation (2.0%) within 30 d of the index operation. The mean hospital length of hospital stay was 3.9 ± 3.1 (median 3, range: 2-31) d. There was one 30-d mortality in a patient with extensive pulmonary metastases who underwent a semi-urgent palliative resection. Postoperative details are seen in Table 2.

Table 2 Outcomes for the 1^st 500 single incision laparoscopic surgery cases n (%).

Variable	Value
Age (yr), mean ± SD	57.61 ± 15.54
BMI (kg/m2), mean ± SD	26.47 ± 4.70
ASA score, median (range)	2 (2-4)
Previous abdominal operation	288 (57.6)
Operative time (min), mean ± SD	149.89 ± 61.98
Blood loss (mL), mean ± SD	64.13 ± 49.78
Intraoperative complications	11 (2.2)
Stoma created	66 (13.2)
Intraoperative conversion	34 (6.8)
Postoperative ICU stay	8 (1.6)
Postoperative complications	49 (9.8)
Clavien class 1	12
Clavien class 2	16
Clavien class 3	18
Clavien class 4	2
Clavien class 5	1
Length of stay (d), mean ± SD	3.91 ± 3.06
Readmission	22 (4.4)
Unplanned re-operation	10 (2.0)
Mortality	1 (< 1.0)

Special populations

With experience, SILS has been expanded and applied to specific “difficult” patient populations. Using the knowledge gained from 500 cases, technical adaptations have been developed to successfully use the SILS platform in these cases.

PELVIC PROCEDURES

For pelvic procedures, the reach of the laparoscopic instruments and visualization are an issue. In 2011, after 140 SILS cases, a SILS + 1 approach was developed to address these issues of operating in the pelvis with a minimally invasive approach. In SILS + 1, a 4 cm Pfannenstiel skin incision is made for the access platform and a single port is placed through the umbilicus for the camera. This adaption allows for an expanded view of the pelvis and abdomen and addressed technical limitation of straight SILS, including better access to the splenic flexure and fewer external clashes between the surgeon and the camera holder. SILS + 1 has been described and validated in pelvic colorectal surgery, and demonstrates the versatility of the platform[27,28].

Outcomes with the SILS + 1 approach for pelvic colorectal cases have been described in a few reports in the existing literature. Kawamata et al[36] performed a comparison evaluation of SILS + 1 and multiport laparoscopic surgery for anterior rectosigmoidectomy, finding the platform was safe and effective. Case series describing SILS + 1 results in pelvic cases include a 16 patient series of anterior rectosigmoidectomy for rectal cancer patients[37], a 20 patient tumor specific mesorectal excision in colorectal cancer patients[38], and single cases of an APR with lateral pelvic lymph node dissection for treating rectal cancer[39], then for total mesorectal excision (TME) with lateral pelvic lymph node dissection[40]. In all reports, the authors found the technique was safe, technically feasible, and did not compromise oncologic safety.

In the present SILS 500 case series, there was a total of 138 SILS + 1 cases performed. The platform was applied for diverticulitis (n = 82), rectosigmoid junction cancer (n = 21), rectal cancer (n = 15), unresectable polyp (n = 12), and inflammatory bowel disease (n = 6). The procedures performed were an anterior rectosigmoidectomy (n = 117), segmental resection (n = 9), low anterior resection (n = 5), ileal pouch anal anastomosis (n = 4), and a total abdominal colectomy (n = 3). The mean operating time for these procedures was 181.65 (SD 58.69) min. With the technical adaption, only 4 cases required intraoperative conversion (< 1.0%); the reasons for conversion were bulky tumors (n = 2), a complex fistula (n = 1), and dense adhesions (n = 1). The mean final Pfannenstiel incision length was 4.2 (SD 0.8) cm. Postoperatively, the mean LOS was 3.44 (SD 2.02) d and there were 5 complications (ileus, n = 3; intrabdominal abscess, C. Difficile infection). Two complications required readmissions (intrabdominal abscess, C. Difficile infection), but there were no reoperations and no mortalities in the SILS + 1 sub-group.

In a direct, case-matched comparison of SILS and SILS + 1 in pelvic colorectal cases, we found a significantly shorter operative time (mean 166.6 min vs 178.0 min, P = 0.03) and significantly lower conversion rate to multiport or open surgery (n = 1, 1.1% vs n = 5, 11.4%, P = 0.02) using the SILS+1 vs the SILS approach (Table 3)[41]. Postoperatively, the length of stay. readmission or complication rates were comparable across the cohorts. Thus, the additional port improved visualization and outcomes without any impact on quality outcomes (Figure 2).

Figure 2 Single incision laparoscopic surgery + 1 approach for pelvic colorectal procedures.

Table 3 Single incision laparoscopic surgery vs single incision laparoscopic surgery + 1 outcome data n (%).

Value	SILS (n = 44)	SILS + 1 (n = 88)	P value
Diagnosis			0.77
Diverticulitis	40 (90.9)	77 (87.5)
Rectal cancer	4 (9.1)	11 (12.5)
Gender
Female	22 (50)	44 (50)	1.00
Male	22 (50)	44 (50)
Age (yr)	56.14 (12.83)	57.16 (10.98)	0.63
Body mass index (kg/m2), mean ± SD	26.57 ± 4.36	27.63 ± 4.50	0.42
ASA class, median	2 (range 2-3)	2 (range 2-4)	0.86
Procedure			1.00
Anterior rectosigmoidectomy	38 (86.4)	76 (86.4)
Low anterior resection	6 (13.6)	12 (13.6)
Operative time (min), mean ± SD	178.0 ± 70.0	166.6 ± 48.4	0.051
Conversion rate	5 (11.4)	1 (1.1)	0.021
Intraoperative complications	1 (2.3)	1 (1.1)	1.00
Length of stay (d), mean ± SD	3.45 ± 1.00	3.56 ± 1.49	0.45
Postoperative complications	2 (4.5)	-	0.11
Readmission (30 d)	1 (2.3)	-	0.33

¹Statistically significant values at an alpha level of 0.05.

ULCERATIVE COLITIS CASES

The SILS platform is valuable in Ulcerative Colitis patients undergoing a total abdominal colectomy or total proctocolectomy with a predetermined ileostomy, as the SILS port can be placed through the right iliac fossa stoma site, and the patient can have an “incisionless surgery”, where there are no incisions or extraction sites other than their stoma. In this patient population, that is predominately young and self-conscious, minimizing incisions is paramount. Using this approach, access is gained through a 2.5 cm circular incision at the planned ileostomy site. The colon is fully mobilized using an inferior to superior approach, and dissection of the colon and division of the mesentery takes place directly by the bowel wall to spare length, minimize manipulation of a thickened edematous mesentery, and protect the vasculature, especially in the ileocolic region, as the vessels need to be spared, as they are the blood supply for the future J pouch.

During our 1^st 500 SILS cases, we performed 30 “incisionless” total abdominal colectomies with an end ileostomy as described. The mean age was 44.3 (SD 15.7) years, and mean BMI was 24.5 (SD 4.5) kg/m². Forty percent of patients had a previous abdominal operation. The mean operative time was 216.1 (SD 62.5) min. One patient (3.3%) was converted intraoperatively for poor visualization. There were no intraoperative complications. The mean length of stay was 4.5 (SD 4.2) d. Postoperatively, 6 patients had complications (anastomotic leak, n = 3; ileus, n = 2; intrabdominal abscess, n = 1) - all patients with an anastomotic leak were on high-dose steroids pre-operatively. Four patients were readmitted within 30-d of the index operation, and 2 required unplanned reoperation for management of an anastomotic leak.

Several other reports have described this approach. Geisler et al[7] described the technique for an incisionless SILS total proctocolectomy with ileal anal J pouch anastomosis through the ileostomy site for a familial adenomatous polyposis patient. This patient had a scarless abdomen other than ileostomy, and the authors proved the safety and feasibility of this technique. Abarca et al[33] then reported on 17 patients that underwent a SILS ileal J-pouch anal anastomosis surgery with extraction of the specimen through the ileostomy site for resection and extracorporeal pouch construction). These patients had a shorter operative time, shorter length of stay, and lower complication rate than similar patients that had a Pfannenstiel extraction site. Thus, there may be more than cosmetic benefits to the modification of SILS in Ulcerative Colitis patients.

HIGH BMI

Laparoscopy has been proven beneficial in obese patients[14,42-50]; however, there is little data on single-incision laparoscopic outcomes in the obese. Most of the published literature with the SILS platform regulated it for experienced laparoscopic surgeons and non-obese patients[7,51-56]. In fact, in describing training colorectal surgeons in SILS, patients initially selected had an average BMI of less than 30 kg/m², with obese patients only incorporated after significant experience[53]. One study comparing outcomes for SILS and traditional multiport laparoscopic colorectal surgery in the obese found similar conversion rates, operative time, postoperative recovery and length of stay across the platforms[57].

In our institution, a case-matched cohort study of 80 obese patients obese (BMI ≥ 30 kg/m²) and 80 non-obese (BMI < 30 kg/m²) patients undergoing an elective single incision laparoscopic colectomy was performed, where we found similar intraoperative complication (2.5% non-obese vs 1.3% obese, P = 0.987), conversion (2.5% non-obese vs 5.0% obese, P = 0.682), postoperative complications (7.5% non-obese vs 12.5% obese, P = 0.430), length of stay (3.7 d non-obese vs 4.1 d obese, P = 0.332), readmission (3.8% in both groups, P = 1.000) and reoperation rates (2.5% non-obese vs 1.3% obese, P = 0.987) across groups[58]. The obese cohort did have significantly longer mean operative times (144.4 min non-obese vs 176.9 min obese, P < 0.001) and mean blood loss (51.6 mL non-obese vs 89.0 mL obese, P < 0.001); however, these measures did not impact outcome measures (Table 4). Thus, in the obese, where higher morbidity and conversion rates are common, by accepting longer operative times, SILS can be used to realize the minimally invasive benefits in these technically challenging cases.

Table 4 Single incision laparoscopic surgery for obese vs non-obese patients n (%).

Parameter	Non-obese (n = 80)	Obese (n = 80)	P value
Age (yr), mean ± SD	57.2 ± 12.8	57.0 ± 12.7	0.926
Gender			1.000
Male	41 (51.2)	41 (51.2)
Female	39 (48.8)	39 (48.8)
Body mass index (kg/m2)	22.8 (2.3)	33.4 (3.2)	< 0.001
ASA score, median (range)	II (I-III)	III (I-IV)	0.035*
Diagnosis			1.000
Diverticulitis	29 (36.3)	29 (36.3)
Colon cancer	23 (28.7)	23 (28.7)
Polyp	17 (21.3)	17 (21.3)
Inflammatory bowel disease	7 (8.7)	7 (8.7)
Constipation	4 (5.0)	4 (5.0)
Procedure			1.000
Anterior rectosigmoidectomy	39 (48.8)	39 (48.8)
Right hemicolectomy	29 (36.3)	29 (36.3)
Total abdominal colectomy	7 (8.8)	7 (8.8)
Left colectomy	3 (3.8)	3 (3.8)
Transverse colectomy	2 (2.5)	2 (2.5)
Operative time (min), mean ± SD	144.4 ± 47.2	176.9 ± 64.0	< 0.001
Blood loss (mL), mean ± SD	51.6 ± 38.0	89.0 ± 139.5	< 0.001
Intraoperative complications	2 (2.5)	1 (1.3)	0.987
Intraoperative conversion	2 (2.5)	4 (5.0)	0.682
Final incision length (cm), mean ± SD	3.5 ± 1.1	4.0 ± 1.5	0.080
Length of stay (d), mean ± SD	3.7 ± 2.1	4.1 ± 2.2	0.332
Postoperative complications	6 (7.5)	10 (12.5)	0.430
Reoperation	2 (2.5)	1 (1.3)	0.987
Readmissions	3 (3.8)	3 (3.8)	1.000

INTEGRATION WITH ROBOTIC ASSISTED LAPAROSCOPIC CASES FOR RECTAL CANCER

To facilitate the laparoscopic portion of a robotic low anterior resection for rectal cancer, we developed a novel reduced port set-up. The SILS port is placed through a predetermined right iliac fossa stoma site and a 12 mm port is placed at the umbilicus for the laparoscopic camera. Three 5 mm assistant ports are placed through the SILS port, as well as a 8 mm robotic assist port, reducing the port incisions. For the robotic portion of the case, two 8 mm accessory ports are placed along the left mid-axillary and left lateral sidewall. The patient is positioned, prepped, draped, and all ports are placed. However, the robot is not docked initially. The case begins with a laparoscopic splenic flexure dissection and ligation of the Inferior Mesenteric vessels. Then, the robot is docked for the pelvic portion. Arm 1 is docked to the patient’s right and used for dissection, while arms 2 and 3 are docked to the patient’s left and used for retraction. The mesorectum is divided beyond the tumor, a circumferential dissection is performed around the rectum, and the rectum is divided. The stapler is fired through the SILS port, eliminating the need for an additional 12 mm port. The specimen is can also be extracted through the SILS port, eliminating the additional Pfannenstiel extraction incision. An extracorporeal resection is performed, the colon is returned to the abdomen, and an end-to-end anastomosis is created. The robot is not undocked during this portion, and readily available to address any issues with the anastomosis. Per surgeon preference, a loop ileostomy can be pulled through the SILS port to protect the low rectal anastomosis, leaving the patient with 8 mm scars and a hidden umbilical scar.

During the study period for the 500 case series, 15 cases were performed using the reduced port robotic assisted low anterior resections technique for rectal cancer (Figure 3). Eleven patients had neoadjuvant chemoradiation. The mean patient age was 60.2 (SD 9.8) years, and 12 were male. The mean BMI was 30.2 (SD 7.55) kg/m². There were no intraoperative complications or conversions. The mean operative time was 306.6 (SD 95.2) min. The mean length of stay was 3.8 (SD 1.6) d. Postoperatively, there were 3 complications- dehydration (n = 2) and an abdominal wall abscess. All 3 patients were readmitted. There were no unplanned reoperations or mortality in the series.

Figure 3 Reduced port robotic assisted laparoscopic surgery.

CONCLUSION

SILS is a minimally invasive platform with specific benefits over traditional multiport laparoscopic surgery. The safety and feasibility of SILS has been proven, and the applications continue to grow with experience. After performing 500 cases, we developed best practices to approach common procedures, standardizing the instrumentation and operative steps. With this breadth of experience, we have also applied the SILS platform to special patient populations and developed adaptations to help overcome technical and ergonomic challenges faced with in minimally invasive colorectal surgery.