Trendelenburg position during surgery are at risk of compartment syndrome, in particular obese

patients and those with peripheral vascular disease. These patients require monitoring of the peripheral

pulses, and position changes, to prevent compartment ischemia.

Principles of Total Mesorectal Excision

The principle of rectal cancer treatment is the eradication of the primary tumor and its lymphatic

drainage by en bloc removal of the rectum and the mesorectum, following well-defined anatomical

planes. This operation, known as total mesorectal excision, or TME, requires dissection under direct

vision along the areolar tissue plane situated between the MRF and the presacral fascia (Fig. 68-15). A

sharp dissection along this plane is associated with a higher probability of achieving a negative CRM,

lower risk of bleeding from inadvertent tearing of the presacral veins, and reduced risk of injuring the

hypogastric nerves.128

Adequate lymphadenectomy requires division of the lymphovascular pedicle at the origin of the

superior rectal vessels, caudal to the branching of the left colic artery from the inferior mesenteric

artery (usually defined as low tie). In patients with clinically suspicious nodes at the origin of the

inferior mesenteric artery, the lymphovascular control should be extended proximally by dividing the

inferior mesenteric artery close to the origin (high tie). In both cases the sigmoidal branches are

included in the surgical specimen; therefore, the proximal division of the bowel should ideally be

performed at the junction of the descending and the sigmoid colon, incorporating most of the sigmoid

colon in the surgical specimen.201 As distal tumor extension along the rectal wall is limited, a distal

margin of 2 cm of normal bowel wall is considered adequate for most tumors. However, distal spread in

the mesorectum may extend farther than intramural spread. Not infrequently, mesorectal deposits are

found 3 to 4 cm distal to the lower edge of the tumor. Therefore, for tumors of the upper rectum, the

mesorectal excision should be extended to 5 cm distal to the lower edge of the tumor. In these patients

the mesorectum should be divided perpendicular to the axis of the rectum. As some of the distal

mesorectum is left in the pelvis along with the distal rectal stump, this operation is known as tumorspecific TME (TSME), to distinguish it from the TME performed with dissection carried to the pelvic

floor, the rectum divided distal to the end of the mesorectum, and the entire mesorectum included in

the surgical specimen.123

Figure 68-15. Total mesorectal excision. A: APR specimen. B: ELAPE specimen. C: Planes of dissection for TME. D: Robotic TME.

Sphincter Saving Procedures

A TSME or a TME is compatible with sphincter preservation by anastomosing the divided end of the

descending or the sigmoid colon to the rectal stump. This operation is known as a low anterior resection

or LAR (Fig. 68-16). In this operation the rectum is divided distal to the tumor using a stapling device

that simultaneously staples and divides the rectum, leaving a short rectal stump. The anastomosis is

completed with a circular stapler that brings together the end of the colon to the rectal stump. The reestablishment of the large bowel’s continuity by suturing the colon to the anal canal, working through

the anus, is known as coloanal anastomosis or CAA (Fig. 68-17). The connection of the colon to the

rectum or anal canal can be performed by an end-to-end or a side-to-end connection, or with the

creation of a J-pouch reservoir (Fig. 68-18). A tension-free, well-vascularized low colorectal or coloanal

anastomosis requires a complete mobilization of the entire left colon with take-down of the splenic

1794

flexure, including division of the inferior mesenteric vein close to the ligament of Treitz. The blood

supply to the entire left colon is based on the left branch of the middle colic vessels through the

marginal artery.

Figure 68-16. A: Low anterior resection, section in the distal third of the rectum, candidate for colorectal anastomosis. B: Low

anterior resection, section above the sphincteric complex, candidate for coloanal anastomosis. C: Colorectal anastomosis.

Figure 68-17. Double stapling technique for low colorectal end-to-end anastomosis.

Many patients undergoing a sphincter-saving procedure report high bowel frequency, urgency,

soiling, and inability to defer defecation for 15 minutes.202 This constellation of symptoms is commonly

known as low anterior resection syndrome (LARS).203 The frequency and severity of these symptoms is

variable and depends, to some degree, on the location of the tumor and the anastomosis, the patient’s

prior bowel function, and the use of neoadjuvant or adjuvant therapy.204 Pelvic irradiation, before or

after surgery, increases significantly the risk of bowel dysfunction.205,206 The anatomical and

physiologic factors that contribute to LARS are not completely understood. Anal sphincter pressures are

not different in patients with or without LARS. What appears to be different is an exaggerated anal

sphincter relaxation to even small volumes of stool within the neorectum.207 Surgeons have tried to

overcome this low capacitance of the neorectum – which results from an end-to-end anastomosis – by

creating a colonic reservoir: either a colonic J-pouch, or a side-to-end anastomosis. A number of studies

have compared the functional outcomes after straight end-to-end anastomosis and a colonic J-pouch. A

systematic review of reconstructive techniques after LAR identified 9 trials that randomized a total of

473 rectal cancer patients to either colonic J-pouch or straight end-to-end anastomosis, reporting that

short-term bowel function was better for J-pouch compared to straight anastomosis.208 Frequency and

1795

urgency were less prevalent in patients with a colonic J-pouch, but the rate of fecal incontinence was

similar in both groups. The only two trials that followed patients for more than 118 months found

minimal or no long-term differences between groups. Although the methodology of these studies was

very heterogeneous, they do indicate that a colonic J-pouch is associated with a short-term

improvement in frequency and urgency, but no apparent long-term benefit, compared to end-to-end

anastomosis. The same systematic review found no differences in anastomotic leak rate, or other

complications, between groups. Of note, almost one-third of patients with colonic J-pouches often

complained of long-term difficulty emptying the rectum, requiring use of laxatives or

suppositories.209,210 These symptoms are more prevalent when larger pouches are created; therefore,

pouch size should be limited to 5 to 6 cm.211,212

Figure 68-18. Reconstructive options after LAR. A: End-to-end anastomosis. B: End-to-side anastomosis. C: J-pouch anastomosis.

The side-to-end anastomosis (created by inserting the anvil of a circular stapler 3 to 4 cm from the

end of the mobilized descending colon) can also potentially increase the reservoir capacity of the

anastomosis. It has similar functional results, but is easier to construct than the colonic J-pouch;

although it, too, is associated with defecatory problems.213 The creation of a transverse coloplasty

proximal to an end-to-end anastomosis is another option used to increase the capacity of the neorectum,

but is associated with a higher anastomotic leak rate.208

In summary, colonic reservoirs are associated with a temporary improvement in bowel frequency and

urgency, but have the risk of long-term defecatory problems. In general, the author prefers an end-toend anastomosis using the larger descending colon, rather than the sigmoid colon, for creation of the

neorectum.

The integrity of the anastomosis should be tested intraoperatively with a pneumatic test, or by

endoscopy, with distension of the bowel while submerging the anastomosis in water. Air leaks thus

detected may be treated with suture repair, re-do of the anastomosis, and/or the creation of a proximal

diverting stoma. Testing the integrity of the anastomosis during surgery has been associated with a

lower risk of clinical postoperative anastomotic leak.214

Low colorectal and coloanal anastomoses have a risk of leakage ranging from 3% to 34%, depending

on the patient population, the use of neoadjuvant radiation, the distance of the anastomosis from the

anal verge, and the surgical technique.215,216 An anastomotic leak can cause pelvic sepsis, which may

not only preclude sphincter preservation, but has also been associated with an increased risk of local

recurrence and poorer survival217 (although more recent data have questioned this association).218,219

To prevent these possible complications, a diverting stoma is recommended for patients having a TME,

and for patients receiving neoadjuvant radiation.220 Diverting stomas do not prevent complications, but

mitigate the consequences of a leak and reduce the need for urgent reoperation in patients developing a

leak.221,222 A loop ileostomy is preferred over a transverse loop colostomy because it is easier to create

and close.223 However, loop ileostomies are associated with postoperative dehydration and electrolyte

abnormalities requiring hospital readmission.224 Prevention of these complications requires adequate

patient education and training about stoma management.

Abdominoperineal Excision

1796

For many cancers located in the distal rectum, particularly those infiltrating the levator muscles or the

anal sphincter, an oncologically safe circumferential and/or distal resection margin is not compatible

with sphincter preservation. In these patients an R0 resection requires a TME with en bloc excision of

the levator muscles and the anal canal, and the creation of a permanent end colostomy. This operation is

known as abdominoperineal excision of the rectum or APE (Fig. 68-19). As the name indicates, the APE

has two parts: an abdominal part and a perineal part. During the abdominal part, the proximal

lymphovascular control, division of the colon, and dissection along the mesorectal plane are similar to

those in LAR, except that the mesorectal dissection should stop at the upper level of the levators, to

avoid disturbing the portion of the rectum resting on the levator muscles. During the perineal part, the

anal canal with the sphincter complex and the levator muscles are dissected off the ischiorectal fat, all

the way to the apex of the ischiorectal fossa. The levators are divided close to their insertion in the

white line near the obturator internus and coccygeus muscles. The levator muscles are left attached to

the rectum, and the resulting surgical specimen has a cylindrical appearance; therefore, this procedure is

called cylindrical APE or extralevator APE (ELAPE) (Fig. 68-20).225 However, some surgeons question

the need to entirely remove both levator muscles in every rectal cancer patient, and recommend

removing only the portion of the levators required to clear the tumor. This operation has been called

the standard APE or SAPE.226 The choice between an ELAPE and SAPE is controversial. The potential

oncologic benefit of larger tissue removal needs to be weighed against the increased morbidity

potentially associated with a larger perineal defect, particularly in patients treated with neoadjuvant

therapy.227 A recent meta-analysis with 949 patients from one randomized trial, one prospective casecontrol study, and six retrospective case series, indicated that ELAPE is associated with lower

intraoperative rectal perforation, lower positive CRM, and lower risk of local recurrence, but with

similar complication rates, compared to SAPE.228 At the present time, the hypothetical benefit of the

ELAPE over the SAPE has not been proven conclusively. It is possible that both techniques – the total or

the tumor-specific excision of the levator muscles – may be equivalent as long as an R0 resection, with

an intact rectal specimen, is achieved.

Figure 68-19. Abdominoperineal excision of the rectum (APE).

1797

Figure 68-20. Cylindrical APE.

Many surgeons will perform both parts of the APE operation with the patient in modified lithotomy,

also known as the Lloyd–Davis position. It provides relatively good access to the perineum and allows

two teams to work simultaneously – one in the abdomen, and one in the perineum. Others perform the

abdominal part in the supine position, and the perineal part with the patient prone over a hip roll and

padding of the bony prominences. The prone position, also known as jackknife position, provides better

exposure of the perineum, and facilitates assistance and teaching. It requires turning the patient, which

takes time, and does not allow a two-team approach. Results seem to be similar with both approaches,

and the selection is based on the surgeon’s preferences.229

Minimally Invasive Surgery for Rectal Cancer

5 Open surgery for rectal cancer requires long abdominal incisions. An APE can be performed though a

midline infraumbilical or low transverse incision. Most sphincter-preserving procedures (SSPs) require a

long midline incision, extending from the epigastrium to the symphysis of the pubis, in order to permit

access to the entire left side of the large bowel from the splenic flexure to the lower rectum. These

abdominal incisions are a source of patient discomfort, and short- and long-term morbidity. The goal of

minimally invasive surgery in rectal cancer is to reduce the size of the abdominal incision, thus

expediting recovery, without compromising the completeness of the mesorectal excision or the

oncologic outcomes. However, the adoption of minimally invasive surgery for rectal cancer has been

slow due to the difficulty of working in the relatively deep and narrow pelvic space using long, rigid,

nonarticulated laparoscopic instruments. According to a number of retrospective case series and small

randomized controlled trials, laparoscopic TME is associated with less postoperative pain, less surgical

morbidity, and shorter length of hospital stay, but equivalent completion of the mesorectal excision,

CRM positivity, and even LR, compared to open TME.230–232 While these studies have shown that

laparoscopic TME is feasible, less than 20% of all rectal cancer operations are performed

laparoscopically in the United States.233

Three large multi-institutional prospective, randomized trials have compared open and laparoscopic

TME for rectal cancer (Table 68-9).234–236 A fourth prospective study conducted in the United States, the

ACOSOG Z6051, has completed accrual, but the results are not available yet. The combined experience

of these trials indicates that laparoscopic TME results in longer operative time, less blood loss, faster

bowel recovery, and shorter hospital stay compared to open TME. Operative mortality and

intraoperative and postoperative complications were not different between groups. The proportions of

patients having an SSP, a complete mesorectal excision, or a positive CRM were not different between

groups, nor was the number of lymph nodes retrieved. However, conversion and positive CRM rates in

the laparoscopic arms varied widely between different studies. These differences could be explained by

trial design, inclusion criteria, randomization, and use of neoadjuvant therapy. Only the CLASICC and

COREAN trials have reported 3-year oncologic outcomes. LR, distant metastasis (DM) and survival were

not different between groups. However, these results should be interpreted with caution, because the

CLASICC trial was not specifically powered to detect differences between treatment groups in rectal

cancer patients, and the COREAN trial allowed a 15% difference as the noninferiority margin.

1798