Local excision

This treatment option has often been used in elderly patients not fitting for major surgery. In those patients the treatment rationale is more a matter of local control to diminish symptoms from the local tumor burden, than to have a curative operation. However, with the new surgical rectoscope, transanal endoscopic microsurgery (TEM), it is possible to perform a locally curative operation (2). In patients with a tumor in stage T1 and T2 there are proposals that the treatment should be done with local excision. In many cases with huge villous adenoma it has often turned out to be an early rectal cancer when the specimen has been thoroughly examined by the pathologist. The problem is to find those patients preoperatively. Staging with ultrasound is an important new tool and with this technique it is possible to find tumors in stage T1 and T2 (3). By doing so an interesting alternative is of course a local treatment instead of major surgery like an anterior resection or an abdominal perineal excision. In cases with T1-tumors data do support that a local excision is as good as a bowel resection in terms of long term survival. One German randomized trial has shown that the outcome is equal in patients with a T1 lesion (4). The question is whether or not it is possible to safely treat patients with a T2 or T3 tumor in the same way. With local excision and pre- or postoperative radio-chemotherapy it looks like the outcome is as good as compared with similar tumors in a historical control series (5). If a local excision done with an optimized technique, i.e. TEM, is equivalent to a more radical procedure (anterior resection or abdominoperineal excision) in terms of longterm survival will be addressed in a new randomized trial, which is going to be launched.

Abdominal procedures

The surgical technique has been discussed extensively the last 15 years. The objectives are to have a good circumferential margin, an appropriate lymph node clearance and finally, if possible, a safe and well functioning reconstruction of bowel continuity. Although no randomized trial has shown the necessity of meticulous dissection taking into account the embryological fascias, there is some evidence that this is the correct way to do the procedure. As proposed by Mr. Heald, the circumferential dissection is important in order not to damage the embryological plane (6). The easiest way to do the cancer operation is to start posteriorly after having divided the vascular trunk. After that division, the hypogastric plexus is identified and by following the rectal fascia posteriorly in the avascular plane, it is easy to find the proper cleavage. The hypogastric nerves can be held laterally to the dissection making it possible to avoid nerve damage. By following the rectal fascia laterally and in a circumferential dissection anteriorly the plane of dissection will be outside the rectal fascia as well as the Denonvilliers fascia. It is essential to gradually following this plane with sharp incision avoiding tear into the mesorectal fat. Finally the whole rectum including the tumor and mesorectum can be excised at a level of the choice of the surgeon. Only if there are macroscopic signs that the tumor has broken through this fascia envelope, it is justified to go even further laterally or anteriorly.

One important question is whether or not an abdominoperineal excision or an anterior resection is to be done. No randomized trials have been done on this topic, mainly due to an unethical concept to randomize between those to options. In older literature, when the anastomosis have been rather high, no difference in local recurrence rates have been found (7, 8).

Margins

The proximal margin is still a controversy. According to the oncological outcome there are data, although not randomized, indicating that there is no need for a high ligation flash to aorta, i.e. ligation of the inferior mesenteric artery. If technically feasible a ligation of the origin of the superior hemorroidal artery is sufficient (9–11). However, if the splenic flexure is planned to be taken down and a colonic pouch is going to be created on the descendent colon, it is essential to have a ligation of the inferior mesenteric artery as well as the inferior mesenteric vein as high as possible, which means ligation flush to aorta regarding the artery and ligation of the vein under the pancreas.

The lateral margin, or more correct the circumferential margin, has been proven to be a very important prognostic factor. If the tumor is growing out to this margin, it indicates a high risk of local recurrence but also a bad prognosis (12–13). If the operation has been done according to the TME concept the risk of having a positive lateral margin is diminished. As mentioned above, only if there is a gross tumor overgrowth through the perirectal fascia, there is an indication to do a wider excision in the pelvis.

The distal margin is also o topic for debate. A tumor rarely grows distally more than 5 mm indicating that 1 cm should be enough. However, due to lymph node deposits in mesorectum (see below) a distal margin of at least 5 cm in the mesorectum should be achieved in upper and middle third rectal cancers. The most important part to discuss regarding the distant margin is of course very low rectal cancers. When there is very little mesorectum down to the pelvic floor and if the sphincters are going to be preserved, a distant margin of just 1 or 1.5 cm seems to be appropriate in very low rectal cancers (14). In tumors situating in the middle and lower third of the rectum, it is often technically easier to do a total mesorectal excision and divide the rectum just below the tip of the mesorectum indicating a division of the rectum just at the top of the anal canal.

Lymph nodes

As mentioned above the oncological rational for doing a high ligation of the interior mesenteric artery, is not proven to be essential. Patients with lymph node metastases in this area, theoretically cured by a high ligation instead of a ligation of the superior hemorroidal artery, do have a disseminated disease and therefore there is no rational other than mentioned above to do such a high ligation.

The mesorectal lymph nodes are excised in a proper way if a mesorectal excision is done according to the guidelines given above. The question of the necessity in doing a total mesorectal excision in all cases is not clear. There are reports of lymph node deposits distal to the tumor at least at the level of 4 cm indicating that a minimum margin of mesorectum should be 5 cm (17), but in very low rectal cancer where there is no mesorectum left the 5 cm role can be questioned (se above).

The Japanese surgeons have addressed the question of clearance of lymph nodes along the pelvic sidewall. After meticulous mapping they have found metastases in those lymph nodes to a certain number and have claimed that it is essential to clear also the lateral compartment (16). However, again no strong evidence based data exist since most of the trials are single centers series compared to historical controls. Again, much evidence does support that if the lateral lymph nodes outside the rectum fascia are involved, this is an indication of disseminated disease.

Nerve function

If care is not taken during the dissection in the lesser pelvis, the risk of damaging the hypogastric nerves as well as the nerves coming laterally to the pelvic organ, is high. The Japanese surgeons have claimed that it is possible to do a nerve sparing dissection in the lateral area by avoiding dividing the nerves. However, in most Japanese material the incidence of impotence and bladder function is high (17). The crucial area to avoid retrograde ejaculation is to identify the hypogastric nerve plexus and keep them laterally during the first phase of the dissection in the lesser pelvis. The other important area of challenge is the so-called lateral ligament, which involve the nerves to the bladder and also the nervii errigente. In those ligaments there are sometimes a mid rectal artery which are not of that size that has to be clamped. Most often it can be secured by cautery. Therefore, the predominant standpoint today to avoid nerve damage in this area is to divide the structure what has been called the lateral ligaments under vision so the nerves will be spared, avoiding dysfunction in the bladder or in the male patients' sexual dysfunction.

Reconstruction

The question of a good function after sphincter saving surgery can be divided in two main options: the quality of the anus sphincter and the level of the tumor. It is therefore essential to evaluate the sphincters preoperatively. If the sphincters are weak and there is a history of bad bowel function or a slight or severe incontinence, it is more or less unethical to perform a low anterior resection due to the risk of faecal incontinence. In such cases it is much better to give the patient a stoma. The other reason when the stoma is necessary is of course if the tumor is growing very low. If the sphincters are involved, it is impossible to create a good function with a safe procedure and therefore an abdominal perineal excision is to be recommended.

In all other cases a reconstruction will improve the quality of life. After a total mesorectal excision, which means a division of the rectum in the top of the anal canal, there are several randomized trials, which have shown that a small J-pouch is to be preferred (18–20). Again, based upon randomized trials the length in the pouch should not be longer than 6 cm (21–22). It might be that an end to side anastomosis is as good as a pouch anal anastomosis. Although trials are ongoing no real evidence based data support this.

In the sphincter saving procedure it has been recommended to wash out the rectal stump before dividing the rectum. This recommendation is based upon the findings of viable cells from the tumor during surgery in the distant part of rectum (23). Although no randomized trials have been performed it is a general opinion that the rectal stump should be irrigated. Some surgeons use cytotoxic agents, others use alcohol or just plain water. Again, no evidence-based data do support what solution should be used for the irrigation.

Another question is the use of a covering stoma after a low anterior resection. We know from several trials that the lower the anastomosis is created the higher is the risk of an anastomatic breakdown. Therefore, a diverting colostomy or ileostomy has been recommended. The rational for such a diversion is not to decrease the leak rate but merely to reduce the consequence of a leak. However, this statement is again not based on trials but more on hypothetical assumptions.