The rectal ‘no man's land’ and sphincter preservation during rectal excision

Colorectal surgeons are familiar with the ‘holy plane’ as the key to total mesorectal excision (TME). Although this has revolutionized the surgical treatment of rectal cancer1, another term should now be introduced into the surgical lexicon—the rectal ‘no man's land’. This is the area of the distal rectum that lies within the pelvic floor musculature. It cannot usually be reached from the abdomen and has been relatively inviolate as far as surgical exploration is concerned, although its anterior compartment has been accessed by the coloproctologist and gynaecologist during rectocele repair and by the urologist during perineal prostatectomy. This region varies in length depending on age, sex and body habitus. It is this variability that determines the length of retained diseased rectal cuff (ranging from a few millimetres to several centimetres) during restorative proctocolectomy for ulcerative colitis. Retention of long lengths may mar the functional result and increases the risk of carcinoma development, particularly in the presence of dysplasia2. To avoid such a potential disaster, coloproctologists either resort to mucosectomy and transanal anastomosis or panproctocolectomy. Both have drawbacks; mucosectomy may be incomplete, risks damage to the underlying internal sphincter and may cause incontinence, whereas after panproctocolectomy the patient is left with a permanent stoma3,4. Similarly, in low rectal cancer, if tumour clearance does not leave sufficient rectal stump to facilitate a stapled anastomosis, the sphincter-saving resection is invariably abandoned in favour of an abdominoperineal excision. It is interesting to note that even in the modern era of stapling up to 52 per cent of patients undergoing surgery for rectal cancer may still be left with a permanent stoma5. Although a minority of surgeons in such circumstances may attempt a partial sphincter-conserving procedure, notably the intersphincteric approach, concerns about technical difficulty, tumour clearance and functional outcome have limited its use6–8. If access could be obtained easily to the crucial few centimetres of distal rectum lying between the superior border of the levator ani which forms the pelvic floor within the abdomen and the anorectal junction in the perineum, it should be feasible to preserve the sphincteric apparatus virtually intact. By doing so it might be possible to increase the proportion of sphincter-saving resections without jeopardizing cure. Such an approach has been developed and should significantly reduce the need for a permanent stoma in patients with benign and malignant rectal disease. So how can this ‘no man's land’ be explored? For many years urologists used the anterior compartment of this space for the perineal approach to prostatectomy. As a result of elegant anatomical studies, Walsh9 demonstrated the feasibility of excising the prostate without damaging the nervi erigentes that enter the prostate at its inferolateral poles. To access the prostate, the dissection by necessity is in the rectoprostatic plane and begins above the external sphincter. Gynaecologists also pointed the way by using the rectovaginal plane in the course of posterior colporrhaphy, albeit by entering this space via the posterior vaginal wall. Both groups take great pains to avoid damage to the anterior wall of the rectum during dissection and of course have no desire to mobilize it. Dissection in the rectovaginal/ prostatic plane has been used in the development of an operation for rectal intussusception (the Express procedure)10. During development of this procedure it was realized that the rectum and mesorectum could be mobilized completely within the funnel of the pelvic floor musculature, and that the ‘no man's land’ could be used for sphincter preservation. It was also demonstrated that the length of the rectum contained within this muscular funnel varied significantly. Although this variation was anatomical in that it was fixed according to body habitus, it was also obvious that some variation was operator dependent. Thus, in the course of an anterior resection for low rectal cancer, the extent of rectal mobilization down to the pelvic floor via the abdomen differs according to the surgeon's skill and perseverance, compounded by anatomical variation and the shape of the pelvis. This in turn influences the length of rectum and mesorectum that remains for construction of the colorectal anastomosis. Exploitation of the rectal ‘no man's land’ (which in our hands may vary from 4 to 13 cm) via the perineum, however, allows the operator to extend the rectal dissection as low as required while constructing an ultralow colo/ileoanal anastomosis under vision. If, as is believed, receptors for normal rectal sensation lie in the puborectalis and levator ani, as well as the rectal wall, restoration of gastrointestinal continuity by drawing the colon or ileum through the residual pelvic floor musculature, with retention of the internal and external sphincters and their somatic nerve supply, should result in an acceptable degree of continence11. Proof of principle has been demonstrated in a pilot study12. Although a heterogeneous group of patients with different underlying pathologies were treated, normal ‘rectal’ sensation together with normal sphincter function was achieved, thus ensuring acceptable continence. Only patients who would otherwise have required a permanent stoma were included, but the initial success suggests that this approach, termed the Anterior Perineal PlanE for ultralow Anterior Resection (APPEAR) technique deserves wider evaluation. Although minimal access surgery is being used more widely in colorectal surgery, a low anterior resection via the abdomen still remains technically demanding. Use of the APPEAR procedure together with laparoscopic abdominal mobilization of the rectum may simplify the operation, making sphincter preservation less of a challenge. Although the APPEAR technique was thought initially to be a novel approach, a similar procedure was described by Cuneo in 190813 and resurrected in 1963 by Lamy and Bricot14. Although these authors did not appreciate the significance of the ‘no man's land’ or TME, they were nevertheless able to construct low colorectal/anal anastomoses with excellent functional results, confirming its validity. Critics may raise concerns about oncological clearance in patients with cancer, and it remains to be seen whether there is a greater risk of local recurrence using the APPEAR technique compared with an abdominoperineal excision. Although careful audit over a 5–10-year period will be required, the omens seem favourable. The principles of proximal lymphatic excision and TME to deal with proximal and circumferential spread are adhered to. It might even be argued that excision of the mesorectum is more complete for some of these patients as access to its most distal extremity is achieved more easily. For the ultralow tumour the distal margin of clearance of 2 cm may be impinged upon, but previous data concerning close-shave anterior resection from above suggest that a margin of clearance of 1 cm or even less does not increase the risk of recurrence provided that TME has also been performed15. The pathological data obtained using the APPEAR technique in 16 patients with cancer showed that adequate clearance of the tumour was achieved, although it is acknowledged that in none of these patients was the pelvic floor musculature invaded. However, if this type of spread was present, there is no reason why resection of the rectum together with part of the pelvic floor musculature could not be performed without necessarily compromising continence. A complicating issue is the wider use of neoadjuvant radiotherapy with or without chemotherapy. The early toxic effects on wound healing may result in early postoperative complications. In addition, the damage now known to be inflicted on the sphincter mechanism in the longer term may be detrimental to patients who rely so heavily on a minimal residual continence mechanism. Although improvements in imaging should improve selection of patients for such additional therapies, there will need to be more thorough counselling and informed consent about options. Some patients may prefer an undamaged continence mechanism as opposed to a marginal reduction in the chances of local recurrence with such adjuvant therapy or even abdominoperineal excision. While it is true that some patients adapt easily to a stoma, many do not, as witnessed by the frequent demands by patients to avoid a stoma at all costs or for reversal from those that already possess one. Exploitation of the rectal ‘no man's land’ by the APPEAR technique deserves wider evaluation. If observational studies indicate that continence can be maintained with acceptable oncological outcomes, appropriate trials should ensue. N.S.W. is a Consultant to Frankenman International, which has a licensing agreement with Queen Mary, University of London, his employer, to produce instrumentation designed to facilitate the APPEAR procedure.