Transanal endoscopic surgery (TES)

INTRODUCTION — Surgical management of rectal lesions can be technically challenging. Transanal excision of benign or premalignant lesions using conventional retractors is limited by poor visualization of the mid- to proximal rectum and a lack of exposure for en bloc excision of larger lesions.

Although total mesorectal excision (TME) is the gold standard for treating locally advanced or node-positive rectal cancers [1], TME can be associated with technical difficulties due to anatomic constraints of the bony pelvis, postoperative morbidity due to injury to pelvic nerves or blood vessels, and poor postoperative bowel function [2]. Thus, for patients with early rectal cancer, defined as lesions limited to the bowel wall with no disease extension beyond the submucosa (T1) and no evidence of lymph node metastasis (N0) (table 1), local excision is a surgical option that exerts minimal impact on bowel function and negates colostomy in those who have low-lying diseases.

Transanal endoscopic surgery (TES) offers transanal access to resecting benign, premalignant, or early malignant lesions in the mid- to proximal rectum (figure 1). For benign or premalignant lesions, TES offers improved visualization, exposure, and access over transanal excision using conventional retractors. For early rectal cancers, TES may allow oncologic cure while maintaining postoperative bowel function. TES, however, does not address locoregional lymph nodes. Thus, only patients with low-risk tumors that are unlikely to harbor nodal metastasis should be treated with local excision alone.

The indications, techniques, platforms, and outcomes of TES are discussed in this topic. The diagnosis and staging of rectal cancer as well as treatment with chemoradiation therapy and surgery (eg, TME) are discussed separately. (See "Clinical presentation, diagnosis, and staging of colorectal cancer" and "Overview of the management of rectal adenocarcinoma" and "Radical resection of rectal cancer" and "Surgical treatment of rectal cancer" and "Neoadjuvant therapy for rectal adenocarcinoma" and "Adjuvant therapy for resected rectal adenocarcinoma in patients not receiving neoadjuvant therapy".)

The techniques and applications of endoscopic submucosal dissection (ESD) for rectal lesions are discussed elsewhere [3]. (See "Overview of endoscopic resection of gastrointestinal tumors", section on 'Endoscopic resection techniques'.)

INDICATIONS

Benign or premalignant lesions — TES can be used to treat a number of benign or premalignant rectal lesions or conditions, including [4-8]:

●Rectal polyps (adenomas)

●Rectal stenosis or stricture

●Solitary rectal ulcer

●Rectal prolapse

●Rectal endometriosis

●Tailgut cyst

●Rectourethral fistula repair

●Rectovaginal fistula repair

●Pelvic abscess drainage

Malignant lesions — In the United States [9] and Europe [10,11], local excision is an acceptable treatment for T1N0 rectal cancers (table 1) with low-risk pathologic features, including (see 'T1N0 rectal cancer' below):

●Moderately to well-differentiated tumor

●Absence of lymphovascular invasion

●Absence of perineural invasion

In a study of patients with clinically stage I rectal cancer from the National Cancer Institute (NCI) database, 12.2 percent with T1 tumors had positive lymph nodes; positive lymph nodes were present in 9.3 percent with neither poor differentiation nor lymphovascular invasion (LVI), 17.3 percent with poor differentiation alone, 34.7 percent with LVI alone, and 45.0 percent with both poor differentiation and LVI [12].

cT1 rectal cancers are subclassified into three categories: sm1, sm2, and sm3, based on the depth of submucosal invasion. Data primarily from Japan showed that when the depth of submucosal invasion was <1000 micrometers (ie, sm1), the rate of nodal involvement was 0 to 1.8 percent, whereas the rate of nodal involvement was 12.8 to 13.8 percent when the depth of submucosal invasion was ≥1000 micrometers (ie, sm2 or sm3) [13]. Consequently, the American Society of Colon and Rectal Surgeons (ASCRS) and European Society for Medical Oncology (ESMO) recommend local excision for cT1 rectal cancers that are sm1 but proctectomy with total mesorectal excision for those that are sm2 or sm3 [14,15]. Similarly, the United States Multi-society Task Force on Colorectal Cancer recommends local excision for cT1 colorectal cancer with <1 mm submucosal invasion but radical surgery for those that are ≥1 mm in submucosal invasion [16]. This criterion is more applicable to endoscopic resection, which is not always full thickness and is therefore more difficult to determine the sm status for. (See "Surgical treatment of rectal cancer", section on 'Criteria for local excision'.)

In the same study from the NCI database cited above, 18 percent of patients with T2 tumors had positive lymph nodes, among whom positive lymph nodes were present in 11.7 percent with neither poor differentiation nor LVI, 25.3 percent with poor differentiation alone, 47.3 percent with LVI alone, and 41.5 percent with both poor differentiation and LVI [12].

Although trials are being conducted to evaluate local excision of T1N0 tumors with high-risk pathologic features or T2N0 tumors following neoadjuvant chemoradiation therapy, such practice is not yet considered standard for all patients. (See 'T2N0 rectal cancer' below and 'T1-3N0 rectal cancer following neoadjuvant therapy' below.)

CONTRAINDICATIONS — TES is limited by technical factors such as the location and size of the tumor and the physical footprint of the platform. TES is ideal for local excision of lesions in the mid- to proximal rectum up to 15 cm from the anal verge. Lesions in the very distal rectum (<5 cm from anal verge) are best excised with conventional transanal excision (TAE) because they may be obscured by the TES platform.

Traditionally, contraindications to TAE, which included tumors involving >30 percent of the luminal circumference, >3 cm in size, or >8 cm from the anal verge, were regarded as contraindications for all local excision techniques, including TES [9]. With increasing surgeon experience, these formerly absolute contraindications have become relative. Some authors have advocated expanding the criteria for local excision (with TAE or TES) to include any locally resectable T1N0 (table 1) rectal cancer with low-risk pathologic features. (See 'Malignant lesions' above.)

A single-institution study of TES considered technically challenging lesions to be ≥5 cm in diameter, involving ≥50 percent of the lumen, or ≥10 cm from the anal verge [17]. TES performed for lesions possessing more than one of these attributes was associated with worse outcomes, including increased local recurrence, peritoneal perforation, tumor fragmentation, rectal stenosis, and incomplete resection.

SURGICAL TECHNIQUES — TES is performed with one of three platforms that enable the surgeon to visualize and manipulate rectal lesions beyond the reach of conventional transanal excision (TAE) (see 'TES platforms' below). Transanal endoscopic microsurgery (TEM) and transanal endoscopic operation (TEO) are rigid platforms, whereas transanal minimally invasive surgery (TAMIS) is a flexible platform (see 'TES platforms' below). Regardless of the platform used, TES is performed in the same sequence of steps.

Preparation — Patients are given a full bowel preparation the day before surgery, intravenous prophylactic antibiotics within one hour of incision, and prophylaxis against venous thromboembolism at the surgeon's discretion.

Anesthesia — TES can be performed with general or spinal anesthesia.

Patient position — Patient positioning is of critical importance in TES. Prior to TES, the surgeon must perform a rigid proctoscopy to determine both the level of the lesion within the rectum and its anatomic orientation (ie, anterior, posterior, or lateral). Patients are then placed in an appropriate position according to the orientation of the lesion and the platform used as detailed below. (See 'Patient positioning for rigid platforms' below and 'Patient positioning for TAMIS' below.)

Resection — For malignant lesions, the goal of TES is to achieve a full-thickness excision of the tumor down to the mesorectal fat with at least 1 cm of margin circumferentially. For benign lesions, dissection is ideally carried out in the submucosal plane. The specimen is usually oriented by the surgeon for pathologic assessment of the margins.

Closure — The defect in the rectal wall is typically closed with absorbable sutures transversely to avoid narrowing the lumen.

Peritoneal entry — TES performed for malignant lesions in the proximal rectum is at risk of entering into the peritoneal cavity. The risk of perforation is higher for anterior lesions, where the peritoneal reflection "dips" lower than it does posterolaterally. Full-thickness perforation of the rectum into the peritoneal cavity can usually be closed transanally via the TES platform, but larger perforations may require transabdominal repair, either open or laparoscopically. Patients with peritoneal entries are observed in the hospital for a longer period of time.

An example of TEM excision of a T1 rectal cancer can be found in the following link to the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) video library.

TES PLATFORMS — TES is performed with one of three platforms, which provide a constant pneumorectum; an angled, magnified view of the operative field; and the ability to insert and exchange instruments while maintaining pneumorectum. Two platforms utilize rigid operating proctoscopes; one utilizes a flexible multichannel laparoscopic port.

A multicenter matched cohort study comparing transanal endoscopic microsurgery (TEM) and transanal minimally invasive surgery (TAMIS) showed that high-quality local excision for benign and rectal neoplasms can be equally achieved using both techniques [18]. The choice of operating platform for local excisions of rectal neoplasms should therefore be based on surgeon preference, availability, and cost.

Rigid platforms — TEM and transanal endoscopic operation (TEO) use rigid operating proctoscopes manufactured by different companies (eg, Richard Wolf, Karl Storz).

●TEM – TEM was the first TES platform, introduced by Gerhard Buess in Germany in 1983 [19]. The TEM proctoscope has a diameter of 40 mm, a length of either 12 or 20 cm, and an end that is either flat or beveled. The proctoscope is inserted through the anus using a blunt obturator. Once the obturator is removed, the proctoscope is sealed with a faceplate containing ports for insertion of an angled stereoscopic optics system and surgical instruments. An insufflation system then generates and maintains a constant pneumorectum, while a suction and irrigation system driven by a roller pump evacuates smoke from the operative field. The stereoscopic (binocular) optics system provides the surgeon with a three-dimensional view of the operative field with depth perception. The entire TEM platform is attached to the operating table using a support arm (figure 2).

●TEO – TEO is a rigid platform that is similar to TEM (picture 1). Both platforms include a 40 mm proctoscope, a faceplate, a support arm, and angled instruments, but TEO offers proctoscope lengths of 7.5, 15, and 20 cm and an option to use a conventional 5 mm laparoscope instead of the binocular optics system [20].

Advantages and disadvantages of rigid platforms — The main advantage of the rigid platforms is that all components, including the insufflator, light source, suction, and instrumentation, come from the same vendor and therefore are integrated in their functionality. The closed-system insufflator permits maintenance of rectal distention and continuous operative exposure by regulating the instillation of carbon dioxide when suction or irrigation is used. Additionally, a support arm fixes the unit to the operating table, which eliminates the need for an assistant to hold the camera.

Disadvantages of the rigid platforms include more stringent requirements for patient positioning and higher cost compared with the flexible platform (TAMIS). (See 'Patient positioning for rigid platforms' below and 'Cost of rigid platforms' below.)

Patient positioning for rigid platforms — The rigid platforms are designed to provide optimal exposure of any lesion that is "down" relative to the optic scope; the angled optic scope is fixed to the proctoscope beveling downward. Thus, the patient must be positioned so that the lesion is "down" relative to the scope. As examples, the patient is placed in a prone position for an anterior lesion, a supine or lithotomy position for a posterior lesion, or a right (left)-side down for a right (left) lateral lesion. During the dissection, the surgeon may need to reposition the scope to center the lesion in the field of view.

Cost of rigid platforms — Because all instruments are supplied and serviced by the same vendor, upfront costs tend to be higher with rigid platforms than with TAMIS. The TEM equipment, for example, costs over USD $80,000. However, even after accounting for the initial equipment purchase, the overall cost of TEM resection of a distal rectal tumor is still much lower than that of open surgery (eg, low anterior resection) [21,22]. Compared with abdominal surgery, TES has a shorter length of stay, fewer complications, and no requirement for a diverting stoma that would require subsequent closure.

Learning curves of rigid platforms — In general, the rigid platforms (TEM or TEO) are more difficult to master than TAMIS. The number of procedures required before surgeons become proficient with TEM or TEO, however, varies between studies:

●In one study, a single surgeon performed 23 TEM procedures [23]. The average rate of excision (ARE) decreased and leveled off after four cases (13.8 minutes/cm² for the first four cases versus 7.9 minutes/cm² for the subsequent 19 cases). Experienced surgeons can perform TES more expediently or with a lower ARE.

●In another study, four surgeons performed 693 rectal resections using the TEM platform [24]. The conversion rate, procedure time, and complication rate were all influenced by surgeon experience as a continuum, with even the difference of a single case affecting outcomes. The recurrence rate, however, was not affected by the learning curve.

●In another study, two surgeons performed 95 TEM procedures [25]. The ARE decreased with each additional case until the 16^th case and then plateaued.

●In another study, 46 patients underwent TEO [20]. By using cumulative sum (CUSUM) analysis, the authors concluded that the learning curve for TEO was 17 cases, after which the operative time and length of stay decreased significantly.

●In a small ex vivo study, surgeons who were unfamiliar with laparoscopic suturing were able to perform suturing tasks more quickly and easily with TEM than with TAMIS and subjectively preferred TEM [26].

Transanal minimally invasive surgery — In 2010, single-incision laparoscopic surgery ports became commercially available. Two such devices, the SILS Port and GelPOINT Path, were adapted for use in TES [27]. TES performed using a flexible multichannel laparoscopic port is called TAMIS (figure 3). Most conventional laparoscopes and laparoscopic instruments can be used for TAMIS as well as angled laparoscopes and instruments specifically designed for single-incision laparoscopic surgery.

Advantages and disadvantages of TAMIS — Compared with rigid platforms, TAMIS has the advantages of lower costs and ease of patient positioning (see 'Patient positioning for TAMIS' below and 'Cost of TAMIS' below). Of note, the flexible platform is used most often for transanal TME (taTME), which is an emerging technology for minimally invasive TME and is discussed elsewhere. (See "Radical resection of rectal cancer", section on 'Transanal TME'.)

One disadvantage of TAMIS is the increased difficulty in accessing lesions in the distal rectum as the access port obscures the distal several centimeters of rectal mucosa. Since rigid platforms can also obscure distal rectal lesions, such lesions are best excised by standard transanal excision rather than by TES. (See 'Contraindications' above.)

Additionally, exposure of the operative field during TAMIS depends entirely on adequate insufflation, whereas the rigid platforms have a large rigid proctoscope to provide stable exposure. In a case report, an insufflation system has been specifically designed to maintain pneumorectum during TAMIS [28].

Patient positioning for TAMIS — TAMIS can be performed with less setup and patient positioning time than TEM or TEO. Because the laparoscopic port used in TAMIS is not oriented and the angled laparoscope allows 360° visualization, there is no absolute requirement for the patient to be positioned with the lesion "down." Consequently, some surgeons place all patients in the lithotomy position regardless of the orientation and location of the lesion [29]. Others place patients in the lithotomy position for all distal and midrectal lesions but the prone position for proximal anterior lesions [30]. The standardization of patient positioning dramatically reduces setup time and complexity.

Cost of TAMIS — Compared with the rigid platforms, TAMIS offers substantial savings in initial cost, which improves surgeon access to TES (see 'Cost of rigid platforms' above). The per-case cost of TAMIS is also comparable to those of the rigid platforms. As an example, a TAMIS port (USD $335 to $650) costs less than the disposable tubing and seals required for each case of TEM [27,29,31].

Learning curve for TAMIS — The learning curve for TAMIS has not been reported in the literature. However, because TAMIS uses the same platform and instruments as single-incision laparoscopic surgery, the learning curve is assumed to be shorter for surgeons who are already proficient in laparoscopy, especially laparoscopic suturing. Further efforts are underway to decrease the learning curve of TAMIS by combining it with robotic surgery, although the increased cost associated with robotic surgery may degrade the benefit of such efforts [32].

In a single-center study of 200 patients undergoing TAMIS for benign or malignant conditions, 7 percent had a positive margin and 5 percent had fragmentation of the specimen [33]. Postoperative complications occurred in 11 percent of patients, most commonly hemorrhage, urinary retention, and scrotal or subcutaneous emphysema. At a mean follow-up of 14 months, 6 percent recurred locally while 2 percent developed distant disease.

Robotic TAMIS — The surgical robot has been used for transanal excision via the single port (SP) robotic platform, which functions similarly to the gel port used for TAMIS procedures. Robotic TAMIS offers a three-dimensional view that is superior to that of the original TEM platform, access to wristed instruments, and freedom from the need for "lesion posterior" positioning. The short-term efficacy and feasibility of this approach has been demonstrated [34].

ONCOLOGIC OUTCOMES — Compared with transabdominal resection with total mesorectal excision (TME), local excision offers reduced perioperative morbidity and improved functional outcomes but yields no nodal information. Thus, only patients with minimal risk of harboring occult metastasis should be offered local excision. For malignant lesions, such risk largely depends upon the tumor grade (table 1).

Existing data support the use of TES for local excision of rectal carcinoid tumors <2 cm, rectal adenomatous polyps, and low-risk T1N0 rectal cancers. Local excision of T2N0 rectal cancers should only be performed in patients who are poor candidates for transabdominal surgery or as part of a clinical trial.

Patients should be informed that local excisional procedures, including TES, offer no nodal information and could potentially leave micrometastasis in the mesorectum or residual disease in the excision bed. To avoid local recurrence or distant metastasis, patients may require additional surgery (typically salvage TME) after local excision when the surgical specimen shows high-risk pathologic features, such as positive margins, tumor grade ≥T2, perineural or lymphovascular invasion, and poor differentiation.

Because the transanal endoscopic microsurgery (TEM) platform was introduced into clinical practice much earlier than the other platforms, most studies have been done with TEM. The indications for and oncologic results of TES performed with other platforms (eg, transanal minimally invasive surgery [TAMIS]) are similar to those of TEM [29].

Carcinoid tumors — The risk of metastasis from small carcinoid tumors is low. Therefore, rectal carcinoid lesions <2 cm and without radiologic evidence of metastasis can be locally excised with TES. In three separate studies, TEM excision of rectal carcinoid tumors with an average size of approximately 1 cm showed no recurrence after a mean follow-up of up to 70 months [35-37]. TEM has also been used to resect endoscopic polypectomy sites that contain residual carcinoid. In two studies, no recurrence was reported after patients underwent TEM after incomplete endoscopic polypectomy of rectal carcinoid [35,37].

Adenomas — For large, sessile, recurrent, or endoscopically unresectable rectal adenomas, TES has supplanted standard transanal excision (TAE) as the preferred method of local excision. In several studies, TEM was superior to TAE in achieving negative margins, avoiding specimen fragmentation, and reducing recurrence rates [38,39]. As an example, in a comparative single-institution review of 259 patients (43 TAE, 216 TEM), TEM achieved a higher negative resection margin rate (88 versus 50 percent), a lower specimen fragmentation rate (1.4 versus 23.8 percent), and a lower cumulative local recurrence rate at five years (6.1 versus 28.7 percent) compared with TAE [40]. In multiple large series, which included hundreds of patients, the recurrence rates after TES for benign polyps ranged from 4 to 7.6 percent [40-46].

Early rectal cancer — TES can be performed alone for T1N0 rectal cancers with low-risk pathologic features. TES can be performed for T2N0 rectal cancers after neoadjuvant chemotherapy and/or radiation therapy in patients who are unfit for transabdominal surgery or in clinical trials. Patients whose TES specimens show high-risk pathologic features should undergo prompt salvage TME or else risk local recurrences or distant metastasis. (See 'Malignant lesions' above.)

T1N0 rectal cancer — For patients with a clinically staged T1N0 rectal cancer without high-risk features, TES is an acceptable treatment option that is associated with better functional outcomes compared with transabdominal radical resection (eg, low anterior resection or abdominal perineal resection with TME). Although patients developed more local recurrences after TES than after TME, their survival was not negatively impacted, presumably because most local recurrences were salvageable with abdominal surgery. Patients should be appropriately counselled that TES is an alternative to standard oncologic resection (TME) but does not yield any nodal information.

In the only randomized trial reported, TEM was compared with transabdominal resection in 50 patients who had endorectal ultrasound-staged T1N0 rectal cancers. The five-year local recurrence (4 percent) and overall survival rates (96 percent) were identical between the two groups. Other outcome measures, such as hospital length of stay, blood loss, operative time, and overall early morbidity, were in favor of TEM [47].

Single-institution reviews of TEM for T1N0 cancers reported recurrence rates ranging from 0 to 24 percent with mean follow-ups of 21 to 97 months (table 2). Systematic reviews and meta-analyses have compared TEM with transabdominal radical resection:

●In a meta-analysis of 13 (mostly observational) studies, 2855 patients with T1N0 rectal cancers underwent either local excision (TAE or TEM) or radical resection [48]. Compared with radical resection, local excision was associated with decreased five-year overall survival (72 more deaths per 1000 patients, 95% CI 30-120) and increased local recurrences (relative risk [RR] 2.36, 95% CI 1.64-3.39) but decreased perioperative mortality, postoperative complications, and permanent ostomy. When the TEM subgroup was separately evaluated, the difference in overall survival was not significant.

●Another meta-analysis comparing only TEM with radical resection for T1N0 rectal cancers reached a similar conclusion [49]. Compared with radical resection, TEM was associated with a higher risk of local recurrences (odds ratio [OR] 4.62, 95% CI 2.03-10.53) but similar rates of overall survival and distant metastasis.

These results, however, are difficult to interpret. As most studies were nonrandomized, a selection bias existed against local excision as surgeons tend to offer local excision to patients who are sicker or have low rectal lesions (low rectal cancers have worse prognosis than more proximal lesions). Additionally, some studies combined the results of TAE with those of TES, despite the fact that the two techniques yielded different oncologic outcomes. Many patients in the local excision group subsequently required radical resection for local recurrences, which yielded similar overall oncologic outcomes.

T2N0 rectal cancer — Because the recurrence rate is higher and survival lower after local as compared with transabdominal excision of T2N0 rectal cancers, transabdominal surgery (TME) is the standard of care for these cancers. Local excision should only be performed in such patients if they are not candidates for abdominal surgery because of medical reasons or if the procedure is being performed as a part of a clinical trial.

In early small series, which often enrolled patients who were unfit for or declined radical resection and employed nonstandardized chemo- or radiation therapy protocols, the recurrence rates of T2N0 rectal cancer after local excision were much higher than those of T1 lesions [46,50,51].

Contemporary studies of locally excised T2N0 rectal cancers after neoadjuvant chemoradiation therapy reported recurrence rates ranging widely from 4.7 to 50 percent (table 3). As examples:

●In the Urbino trial, 70 patients with T2N0 rectal cancer were randomly assigned to TEM or laparoscopic resection (low anterior resection or abdominal perineal resection) following neoadjuvant chemoradiation. After a median follow-up of 84 months, the local recurrence rate was 5.7 percent after TEM and 2.8 percent after laparoscopic surgery. The disease-free survival was not different between the two groups (94 percent for both); a single patient in each group developed distant metastases. Although the overall complication rates were not different, TEM was associated with a shorter hospital stay (three versus seven days) [52]. Other studies by the same group of authors showed similar results [53-55].

●In the American College of Surgeons Oncology Group (ACOSOG) Z6041 trial, 90 patients with T2N0 rectal cancer underwent local excision six weeks after neoadjuvant chemoradiation. In the 77 patients who completed treatment per protocol, pathologic complete response was noted in 44 percent, and only a single patient had positive margins. Forty-two percent of patients developed major (grade ≥3) complications attributable to chemoradiation therapy; 58 percent developed surgery-related complications [56].

In that study, of the 72 patients who were followed for a median period of 56 months, the estimated three-year disease-free survival was 88.2 percent (95% CI 81.3-95.8) by intention-to-treat analysis and 86.9 percent (95% CI 79.3-95.3) by per-protocol analysis. Because the three-year disease-free survival was lower than anticipated, the authors of this study recommended that neoadjuvant therapy followed by local excision be reserved for patients with T2N0 rectal cancers who are not candidates for or refuse standard transabdominal radical resection [57].

T1-3N0 rectal cancer following neoadjuvant therapy — There has been increasing interest in the use of TES as a mechanism for evaluating pathologic response to neoadjuvant chemoradiation in patients with early rectal cancer. In the following studies, patients with T1-3N0 rectal cancer underwent neoadjuvant chemoradiation therapy followed by local excision (most often TEM). Further treatment was dictated by the pathologic findings on the local excision specimen.

Patients who had favorable pathologic features (negative margins, ypT0-1, absence of lymphovascular or perineural invasion, and moderate-to-well differentiation) on TES specimen were observed without further surgery:

●In the CARTS study, 55 patients with T1-3N0 rectal cancer underwent long-course neoadjuvant chemoradiation. After clinical reevaluation at six to eight weeks, those with significant clinical response (tumor downsizing) underwent TEM [58]. Thirty of 55 patients (55 percent) were ypT0-1 on TEM specimen and required no further surgery. After a median follow-up of 17 months, only one patient (of 30) developed a local recurrence and underwent salvage abdominal perineal resection.

At five years, the rectum was preserved in 64 percent of these 55 patients. The actuarial five-year local recurrence rate was 7.7 percent (one patient with stage ypT1 cancer after TEM and in three patients who initially refused completion TME for ypT2 stage cancer), and the five-year disease-free and overall survival rates were 81.6 and 82.8 percent, respectively [59]. Despite the favorable oncologic outcomes, 50 percent of patients with a preserved rectum experienced major symptoms similar to low anterior resection syndrome (LARS) (see "Low anterior resection syndrome (LARS)"), and one-third of patients could have been spared neoadjuvant radiation given that they went on to undergo TME. Two patients died from side effects from the neoadjuvant treatment.

●In another study, 89 patients with sessile T1-3N0 rectal cancers were treated with neoadjuvant chemoradiation or short-course radiation alone followed by local excision with TEM or TAE. Despite unreliable adherence to radiation protocols, the study reported a 10 percent local recurrence rate at two years in those who had ypT0-1 tumors in the local excision specimen [60].

●In the GRECCAR2 trial conducted in 15 centers in France, 186 patients with T2-3, N0-1 low rectal cancers (≤8 cm from anal verge) that were ≤4 cm received neoadjuvant chemoradiation therapy. One hundred forty-eight patients who had a good clinical response (residual tumor ≤2 cm on magnetic resonance imaging [MRI] six to eight weeks after neoadjuvant therapy) were randomly assigned to local excision or TME [61]. Local excision was carried out by transanal excision and TEM for 72 and 28 percent of patients, respectively. Patients who had an unfavorable pathology (ypT2-3 or R1) after local excision were offered completion TME; 26 patients accepted. At two years, one or more events of death, recurrence, serious morbidity, and long-term side effects (colostomy, incontinence, sexual dysfunction) occurred in 56 percent of the local excision group and 48 percent of the TME group. Although local excision was oncologically as safe as TME, it was not better in terms of morbidity and long-term preservation of organ (rectal) function.

Patients who had unfavorable pathologic features, including positive margins, ≥ypT2, neural or lymphovascular invasion, and poor differentiation, were offered radical resection; those who declined further surgery suffered high local recurrence rates and poor survival:

●In one study, 89 patients with sessile T1-3N0 rectal cancers underwent neoadjuvant therapy followed by local excision with TEM or TAE [60]. Eighteen patients were offered completion radical surgery due to high-risk pathologic findings on local excision specimen but declined. At two years, 8 of the 18 patients (44 percent) developed local recurrences; one additional patient developed a distant metastasis without local recurrence.

●In another study, 53 patients with early rectal cancer underwent neoadjuvant therapy followed by TEM [62]. Although 36 patients had high-risk pathologic features on TEM specimen, none underwent further surgery. At two years, 12 patients (33 percent) developed local recurrences. Of these 12 patients, only eight were candidates for salvage surgery, and negative circumferential resection margin was only achieved in one of eight patients at salvage surgery. At three years, four patients developed local re-recurrences. The two-year local recurrence-free survival was 77 percent, and the re-recurrence-free survival was 60 percent.

In a systematic review and meta-analysis of 20 studies (14 cohort, 5 comparative cohort, 1 randomized trial), over 1000 patients with early rectal cancer (23, 46, and 31 percent with T1-3 tumors, respectively) were treated with neoadjuvant therapy followed by local excision (TES or TAE) [63]. The pooled complete clinical response (cCR) rate was 46 percent (95% CI 31 to 61 percent); the pooled complete pathologic response (cPR) rate was 44 percent (95% CI 36 to 52 percent).

After a median follow-up of 54 months (range 12 to 81 months), ypT0, ypT1, ypT2, and ypT3 tumors had pooled local recurrence rates of 4 (95% CI 1.9 to 6.9 percent), 12 (95% CI 6.3 to 19.4 percent), 24 (95% CI 13 to 36 percent), and 60 percent (95% CI 32 to 84 percent), respectively. Based on the data, the authors of this meta-analysis concluded that local excision after neoadjuvant therapy should only be considered curative if cPR (ie, ypT0) was achieved. Radical surgery should be offered to any patient with an incomplete response to neoadjuvant therapy (ie, ≥ypT1) or else risk recurrences. However, because this study did not address the correlation between the pretreatment stage and cPR, it remains unclear which patients with early rectal cancer are likely to benefit from such a strategy of neoadjuvant therapy plus local excision.

Local excision as the sole surgical treatment for ≥T2N0 rectal cancer is still investigational out of concerns for micrometastasis in the mesorectum and residual disease in the excision bed. In addition, neoadjuvant therapy for clinical stage T1N0 or T2N0 patients is not standard practice and may not be necessary if such patients are treated with upfront curative TME. In at least some patients, pelvic radiotherapy and/or chemotherapy can cause morbidity or functional impairment that is comparable to that associated with radical surgery. (See "Overview of the management of rectal adenocarcinoma", section on 'Clinical T2N0 and cT1N0 not amenable to local excision' and "Neoadjuvant therapy for rectal adenocarcinoma", section on 'Complications'.)

Despite the evidence presented above, it remains inconclusive whether neoadjuvant therapy followed by TES is equivalent to radical surgery (TME) in treating cT1-3N0 rectal cancer. More trials are ongoing to compare the now three potential approaches to treating this cohort of patients: standard TME without neoadjuvant therapy, neoadjuvant therapy followed by TES, and neoadjuvant therapy followed by watchful waiting (for those who achieve a complete response to neoadjuvant therapy, which may further reduce morbidity).

For now, patients with ≥T2N0 rectal cancer should continue to be counseled to undergo abdominal surgery with TME for optimal oncologic outcomes. Local excision should only be used in patients who are frail or otherwise unfit for abdominal surgery or in the settings of clinical trials.

FUNCTIONAL OUTCOMES — In general, patients who undergo local excision have a better quality of life and bowel function than those who undergo transabdominal mesorectal excision [64].

The functional outcomes after TES, especially with regard to fecal incontinence, have been studied for transanal endoscopic microsurgery (TEM) and transanal minimally invasive surgery (TAMIS).

TEM — In early studies, TEM was associated with fecal incontinence attributed to the use of a large-diameter (4 cm) proctoscope. As an example, 18 of 58 patients (37 percent) who underwent TEM in one study suffered varying degrees of incontinence of liquid stool that persisted over time [65]. However, other more contemporary studies found TEM to be associated with no worse (or even, at times, better) clinical fecal incontinence scores compared with the baseline [66-71].

TAMIS — Several single-institution studies have reported preserved or improved fecal continence and overall quality of life after TAMIS [72-74].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Colorectal cancer" and "Society guideline links: Colorectal surgery for cancer".)

SUMMARY AND RECOMMENDATIONS

●Transanal endoscopic surgery (TES) offers transanal access to resecting benign, premalignant, or early malignant lesions in the mid- to proximal rectum. (See 'Introduction' above and 'Indications' above.)

●TES is ideal for local excision of lesions in the mid- to proximal rectum up to 15 cm from the anal verge. Lesions in the very distal rectum (<5 cm from anal verge) are best excised with conventional transanal excision because they may be obscured by the TES platform. (See 'Contraindications' above.)

●TES is performed with one of three platforms that provide a constant pneumorectum; an angled, magnified view of the operative field; and the ability to insert and exchange instruments while maintaining pneumorectum. Two platforms utilize rigid operating proctoscopes; one utilizes a flexible laparoscopic multichannel port (see 'TES platforms' above):

•Transanal endoscopic microsurgery (TEM) and transanal endoscopic operation (TEO) use rigid operating proctoscopes, which have the advantages of being integrated and not requiring a camera holder but the disadvantages of being costly, time consuming because of complex patient positioning, and difficult to learn compared with transanal minimally invasive surgery (TAMIS). (See 'Rigid platforms' above.)

•TAMIS uses commercially available laparoscopic ports, scopes, and instruments. Compared with TEM or TEO, TAMIS has the advantages of being less expensive, being quicker to set up because of standard patient positioning, and being easier to learn for surgeons with prior laparoscopic experience but has the disadvantage of a less stable pneumorectum. (See 'Transanal minimally invasive surgery' above.)

●TES can be performed for carcinoid rectal tumors of <2 cm and for rectal adenomatous polyps. (See 'Carcinoid tumors' above and 'Adenomas' above.)

●For patients with a clinically staged T1N0 rectal cancer without high-risk features, TES is an acceptable treatment option that is associated with better functional outcomes compared with a transabdominal radical resection (eg, low anterior resection or abdominal perineal resection with TME). Although patients developed more local recurrences after TES than after TME, their survival was not negatively impacted, presumably because most local recurrences were salvageable with abdominal surgery. Patients should be appropriately counselled that TES is an alternative to standard oncologic resection (TME) but does not yield any nodal information. (See 'T1N0 rectal cancer' above.)

●Because the recurrence rate is higher and survival lower after local excision of T2N0 rectal cancers compared with after transabdominal resection with TME, transabdominal surgery remains as the standard of care for T2N0 rectal cancer; local excision should only be performed for patients who are unfit for transabdominal surgery or as part of a clinical trial. (See 'T2N0 rectal cancer' above.)

●Patients whose TES specimen shows high-risk pathologic features, such as positive margins, tumor grade ≥T2, perineural or lymphovascular invasion, or poor differentiation, should promptly undergo salvage abdominal surgery with TME or else risk local recurrences or distant metastasis. (See 'T1-3N0 rectal cancer following neoadjuvant therapy' above.)

●Although TEM was associated with fecal incontinence in early reports, contemporary studies showed preserved or improved clinical continence scores after TEM or TAMIS. (See 'Functional outcomes' above.)