Four anastomotic techniques following transanal total mesorectal excision (TaTME)

The formation of a colorectal or coloanal anastomosis is one of the critical steps post-TaTME that has been performed using both hand-sewn and stapling techniques. During a TaTME procedure, the distal rectal wall is divided at the start of the transanal dissection. This leaves an open distal rectal stump, which can easily be retracted and held in position for a hand-sewn anastomosis. The hand-sewn approach appears to be more suitable for very low coloanal anastomoses, as a pursestring closure is unlikely to be possible due to insufficient stump length. The level of the pursestring is dependent on the height of the tumour. If oncologically safe with an adequate margin, a rectal cuff just above the internal sphincter is preferred in order to have better functional outcome compared to the coloanal anastomosis. Conversely, a longer rectal stump may owe itself more readily to a stapling technique, as the visual exposure may be inadequate, and the distance from the anus too far for a hand-sewn anastomosis.

The EEA™ Haemorrhoid Stapler (Covidien) has been frequently used post-TaTME. The advantage of this stapler is the longer central rod on the anvil (13.5 cm) that allows connection to the stapler device before closure of the pursestring. However, there are two potential disadvantages associated with its use. The first is that the stapler’s large diameter of 33 mm could risk incorporating sphincter muscle or even the vagina into the stapler when forming a low coloanal anastomosis. This may lead to a worse functional outcome. Secondly, it is not always possible to fit the large-sized anvil into the new colonic conduit, even in a side-to-end orientation.

More recently, a stapling technique using the CEEA™ stapler has been described previously including a video and outlined above [5]. The addition of the 10Fr redivac drain acts as a guide and safety mechanism for the insertion of the spindle of the AutoSuture CEEA™ circular stapler through the pursestring. The diameter of the CEEA™ stapler is also smaller, 28 or 31 mm, compared to the 33-mm EEA™ stapler, posing less of a risk of incorporating sphincter muscle into the stapler. We have reported on a series of 12 cases using the AutoSuture CEEA stapler in which there were no anastomotic leaks, and to date, all patients have had a good functional outcome [5]. A potential drawback of this technique is that it demands good visualisation of the pelvic floor and the rectal stump from the abdominal side before completing the anastomosis since the anvil is placed onto the stapling gun using conventional laparoscopic methods. In the difficult narrow pelvis with a short rectal stump, this exposure is sometimes limited. To overcome problems with abdominal exposure, whilst still avoiding the disadvantages of the wide 33-mm stapling device, a standard 28-mm stapler can be utilised using the pull-through method which relies on a good transanal view rather than abdominal. Further, it creates the possibility of a transanal anastomosis with excellent control of the distal pursestring. A potential disadvantage of this technique is the relative short anvil, which has to be clamped inside the anal canal in order to attach the stapler. Therefore, its use is not recommended in higher anastomoses above 4–5 cm. The author, Tuynman, who pioneered this technique has performed 36 cases so far and experienced two clinical leaks, both managed by transgluteal drain positioning.

The potential advantages and disadvantages of each anastomotic technique are outlined in Table 1. However, the true benefits and optimal approach are yet to be tested and confirm in comparative studies (Table 1).

Since each patient and each tumour has their own characteristics, it may be reasonable for a surgeon to be able to perform a number of anastomotic techniques in order to tailor the approach to the patient’s anatomy. This has been suggested in Knol et al.’s recent publication on technical aspects of TaTME, a more individualised approach may be better depending on the distance of the tumour from the anorectal junction (ARJ) [4]. This will determine whether a platform is used at the start of the transanal TME dissection and what the most favourable anastomotic technique will be. For example, see Table 2.

Regardless of the technique used, care should always be taken to ensure well-vascularised anastomotic ends, optimal visualisation, and awareness of the potential risk to nearby structures such as the anorectal sphincters and vagina, especially when adherent to the rectal wall.

Recently, Tuech et al. [8] published the first functional outcome results in 56 consecutive patients who underwent endoscopic transanal proctectomy (ETAP) and hand-sewn coloanal anastomosis for low rectal cancer. The overall morbidity after surgery was 26 % with three patients developing a clinical anastomotic leakage (none required reoperation) and a local recurrence rate of only 1.7 % (median follow-up: 29 months, range 18–52). It is reassuring to find that the median Wexner score after stoma reversal was 5 (range 3–18), and only three patients (5.7 %) required a colostomy due to severe faecal incontinence. Given the more distal tumours included in this study, all of which had hand-sewn coloanal anastomoses, functional results are likely to be even better following more proximal stapled anastomoses.

Two further groups have published their initial experience with TaTME including the Dutch group, Veltcamp Helbach et al. [9], and Dr Lacy [10] from Barcelona. Eighty patients underwent TaTME in the Dutch group [9]; stapled anastomosis using the EEA™ haemorrhoidal stapler was used in cases in which gastrointestinal continuity was restored. Post-operative complications were seen in 39 % of patients, nine of whom required reoperation. One patient returned to theatre due to anastomotic leak.

Lacy et al. [10] have published the largest case series of 140 patients to date. Hand-sewn coloanal anastomosis was performed for patients with the most distal rectal tumours, whilst for mid- and proximal tumours, an EEA 33-mm circular stapler was used. Major complications were seen in 10 % of cases, with anastomotic leaks detected in 12 patients (8.6 %), three treated successfully conservatively, whilst one required percutaneous drainage and two had rectal tube transanal and intravenous antibiotics. The remaining nine patients returned to theatre with one of these patients requiring a stoma. Anastomotic bleeding occurred in three patients of whom one underwent a reoperation for transanal reinforcing stitches to control the bleeding.

Studies specifically comparing hand-sewn versus stapled coloanal/colorectal anastomosis following TaTME have yet to be published. Similis et al. [11] conducted a systematic review including 37 studies with a total of 628 participants who underwent TaTME resection. The review found that 66 % of anastomoses were hand-sewn coloanal and only 34 % were stapled. Anastomotic leak occurred in 25 cases, anastomotic stenosis in 11, and fistula formation in one case. Due to the heterogeneity of the studies included, with a low number of stapled anastomoses and cases likely to have been performed at an early stage in the surgeon’s learning curve for TaTME, firm conclusions as to the optimal anastomotic method cannot be made. Anastomotic techniques have been compared following traditional laparoscopic and open rectal resections, with conflicting results. Cong et al. [12] found significantly lower rates of anastomotic leakage and stricture formation following stapled coloanal anastomosis compared to manual anastomosis following laparoscopic intersphincteric resections. The complication rates were similar for fistula formation, bleeding, and neorectal mucosal prolapse between the two groups. An earlier randomised study comparing hand-sewn versus stapled techniques in colonic J-Pouch-Anal anastomosis for rectal cancer found that anastomotic stricture rates were lower in the stapled group but did not reach statistical significance [13]. Post-operative morbidity and functional problems were similar between the two groups, but intra-operatively, the time taken to perform a stapled anastomosis was significantly faster. In 2012, a Cochrane review found insufficient evidence to demonstrate superiority of stapled over hand-sewn techniques in colorectal anastomosis surgery, regardless of the level of anastomosis [14]. The only statistically different results were that stricture formation was more frequent with stapling (P < 0.05), and the time taken to perform the anastomosis was longer with hand-sewn techniques.

As with all emerging techniques, small modifications and technical optimisation are often required to further enhance the feasibility and safety profile. Three anastomotic colorectal techniques post-TaTME are in practice, and this description allows tailoring of the technique to length of the anal canal and height of anastomosis. However, studies comparing these techniques with functional outcome have yet to be published. Ideally, large randomised studies are required to compare post-operative outcomes between hand-sewn and stapling groups. However, as stated by Professor Wexner, ‘the rapid adoption by inadequately trained low-volume surgeons may sadly jeopardize the ultimate achievement’ of TaTME. Therefore, structured training, skills acquisition, mentorship, and credentialing with a standardised surgical approach are essential requisites in order to elicit and achieve the true potential benefits of TaTME.