Key points

Introduction

The earliest surgical attempts to treat rectal cancer were via the perineum and the techniques used were exclusively extraperitoneal with extremely poor results. The perioperative mortality was high, functional results appalling, and local control very bad, with local recurrence rates up to 90%. Sir Ernest Miles, a surgeon at St Mark’s Hospital in London, took an important step in the development of surgery for rectal cancer when he published an article, “A Method of Performing Abdomino-Perineal Excision for Carcinoma of the Rectum and of the Terminal Portion of the Pelvic Colon,” on December 19, 1908, in The Lancet . This was a thorough description of an APE of the rectum and has since been called the Miles operation . In his original description of the procedure, the rectum was bluntly mobilized down to the sacrococcygeal articulation, to the prostate, and to “the upper surface of the levatores ani” laterally, thus leaving the mesorectum attached to the pelvic floor. After this mobilization of the rectum, a colostomy was created and the abdominal wall was closed. The patient was turned over and placed in the right lateral and semiprone position. Miles emphasized that the levator muscles should be divided “as far outwards as their origin from the white line so as to include the lateral zone of spread”; therefore, the perineal part of the operation included a wide excision of skin, fat, and pelvic floor (levator muscles).

The Lancet article had an enormous impact on the surgical community and for many decades the Miles operation was the gold standard procedure for all rectal carcinomas. The concept of removing the entire rectum, the anus, and the perineum in all patients with rectal cancer, however, was gradually abandoned. Increasing experience with bowel reconstruction, including developments of stapling instruments, led to a new concept of anterior resection (AR) and low AR (LAR), which became the standard procedures for tumors of the upper and middle rectum.

For tumors of the lower rectum, most surgeons continued to perform APE, although the extensive perineal approach described by Miles was more or less neglected and the synchronous combined APE was introduced as a feasible procedure that became popular and gained widespread use in the treatment of low rectal cancer. During the synchronous combined operation, the perineal part is carried out simultaneously with the pelvic part of the abdominal procedure, with the patient in the supine lithotomy, or Lloyd-Davies position; the rectum with its mesorectum is first mobilized down to the pelvic floor and the perineal surgeon then enters the pelvic cavity just in front of the coccyx, the levator muscles are divided on both sides, and, finally, the rectum is dissected off the prostate or the vagina and the specimen is delivered through the perineum.

Although there were gradual improvements in the treatment of rectal cancer during the twentieth century, local control remained a major problem after surgery, with local recurrence rates of up to 40% after potentially curative resections. Therefore, irradiation to the rectum and to the pelvis, both preoperatively and postoperatively, was tried in order to improve local control. Preoperative radiotherapy has been evaluated in several large randomized trials and was shown to reduce local recurrence rates by 50% and to improve cancer-specific survival.

With the development of total mesorectal excision (TME), as described by Heald and colleagues, treatment results improved dramatically, both concerning local control and survival. Heald and colleagues reported a local recurrence rate of approximately 5% and a cancer-specific survival of approximately 70% at 5 years, without radiotherapy. Initially, these results were mistrusted by many surgeons but, due to extensive educational efforts, the technique was gradually accepted. During the recent 15 to 20 years, the TME technique for rectal cancer resection has been introduced in many countries and, subsequently, the results with regard to local control and cancer survival have improved significantly. Local recurrence rates are now reported to be less than 10% in population-based studies. The acknowledgment of TME as the standard surgical technique in the treatment of rectal cancer has resulted not only in improved local control and survival but also in increasing rates of sphincter-saving procedures and improved results concerning urogenital function.

Consequently, in the past 15 to 20 years, teaching rectal cancer surgery mainly focused on the operative technique of TME and AR. Although the technique used for the abdominal part of an APE was modified along the lines of TME, little attention was given to the perineal part of this procedure. Thus, most surgeons adopted the technique of sharp dissection under direct vision outside the mesorectal fascia down to the pelvic floor, with the aim of saving autonomic nerves and creating perfect specimen with an intact mesorectal fascia. The perineal part, however, was often completed in the conventional way, with dissection close to the external sphincter and with the division of the levator muscles close to the rectal wall. With a patient in the supine lithotomy position, it is difficult to achieve an optimal view, especially anteriorly, and, therefore, parts of the perineal dissection are often done with blunt dissection when this approach is used.

Introduction

The earliest surgical attempts to treat rectal cancer were via the perineum and the techniques used were exclusively extraperitoneal with extremely poor results. The perioperative mortality was high, functional results appalling, and local control very bad, with local recurrence rates up to 90%. Sir Ernest Miles, a surgeon at St Mark’s Hospital in London, took an important step in the development of surgery for rectal cancer when he published an article, “A Method of Performing Abdomino-Perineal Excision for Carcinoma of the Rectum and of the Terminal Portion of the Pelvic Colon,” on December 19, 1908, in The Lancet . This was a thorough description of an APE of the rectum and has since been called the Miles operation . In his original description of the procedure, the rectum was bluntly mobilized down to the sacrococcygeal articulation, to the prostate, and to “the upper surface of the levatores ani” laterally, thus leaving the mesorectum attached to the pelvic floor. After this mobilization of the rectum, a colostomy was created and the abdominal wall was closed. The patient was turned over and placed in the right lateral and semiprone position. Miles emphasized that the levator muscles should be divided “as far outwards as their origin from the white line so as to include the lateral zone of spread”; therefore, the perineal part of the operation included a wide excision of skin, fat, and pelvic floor (levator muscles).

The Lancet article had an enormous impact on the surgical community and for many decades the Miles operation was the gold standard procedure for all rectal carcinomas. The concept of removing the entire rectum, the anus, and the perineum in all patients with rectal cancer, however, was gradually abandoned. Increasing experience with bowel reconstruction, including developments of stapling instruments, led to a new concept of anterior resection (AR) and low AR (LAR), which became the standard procedures for tumors of the upper and middle rectum.

For tumors of the lower rectum, most surgeons continued to perform APE, although the extensive perineal approach described by Miles was more or less neglected and the synchronous combined APE was introduced as a feasible procedure that became popular and gained widespread use in the treatment of low rectal cancer. During the synchronous combined operation, the perineal part is carried out simultaneously with the pelvic part of the abdominal procedure, with the patient in the supine lithotomy, or Lloyd-Davies position; the rectum with its mesorectum is first mobilized down to the pelvic floor and the perineal surgeon then enters the pelvic cavity just in front of the coccyx, the levator muscles are divided on both sides, and, finally, the rectum is dissected off the prostate or the vagina and the specimen is delivered through the perineum.

Although there were gradual improvements in the treatment of rectal cancer during the twentieth century, local control remained a major problem after surgery, with local recurrence rates of up to 40% after potentially curative resections. Therefore, irradiation to the rectum and to the pelvis, both preoperatively and postoperatively, was tried in order to improve local control. Preoperative radiotherapy has been evaluated in several large randomized trials and was shown to reduce local recurrence rates by 50% and to improve cancer-specific survival.

With the development of total mesorectal excision (TME), as described by Heald and colleagues, treatment results improved dramatically, both concerning local control and survival. Heald and colleagues reported a local recurrence rate of approximately 5% and a cancer-specific survival of approximately 70% at 5 years, without radiotherapy. Initially, these results were mistrusted by many surgeons but, due to extensive educational efforts, the technique was gradually accepted. During the recent 15 to 20 years, the TME technique for rectal cancer resection has been introduced in many countries and, subsequently, the results with regard to local control and cancer survival have improved significantly. Local recurrence rates are now reported to be less than 10% in population-based studies. The acknowledgment of TME as the standard surgical technique in the treatment of rectal cancer has resulted not only in improved local control and survival but also in increasing rates of sphincter-saving procedures and improved results concerning urogenital function.

Consequently, in the past 15 to 20 years, teaching rectal cancer surgery mainly focused on the operative technique of TME and AR. Although the technique used for the abdominal part of an APE was modified along the lines of TME, little attention was given to the perineal part of this procedure. Thus, most surgeons adopted the technique of sharp dissection under direct vision outside the mesorectal fascia down to the pelvic floor, with the aim of saving autonomic nerves and creating perfect specimen with an intact mesorectal fascia. The perineal part, however, was often completed in the conventional way, with dissection close to the external sphincter and with the division of the levator muscles close to the rectal wall. With a patient in the supine lithotomy position, it is difficult to achieve an optimal view, especially anteriorly, and, therefore, parts of the perineal dissection are often done with blunt dissection when this approach is used.

Problems related to the conventional synchronous combined APE

In recent years, several investigators have acknowledged that local control and survival after APE have not improved to the same degree as those seen after AR. In one study based on 561 patients from Leeds, United Kingdom, it was reported that patients undergoing APE had a higher local failure rate (22.3% vs 13.5%) and a poorer survival (52.3% vs 65.8%) compared with patients who had an AR during the same time period.

Another article based on data from five different European trials reported that the APE procedure was associated with an increased risk of circumferential resection margin (CRM) involvement, an increased local recurrence rate, and a decreased cancer-specific survival. A large cohort study from Norway also reported a higher local recurrence rate (15% vs 10%) and a poorer 5-year survival (55% vs 68%) after APE than after AR.

These differences in oncological outcomes between the two procedures may be explained by several factors, including anatomic difficulties and the surgical technique associated with standard APE surgery. In the lower rectum, the surrounding mesorectum is reduced in size and disappears at the top of the sphincters. Below this level, the sphincter muscle forms the CRM. As discussed previously, the abdominal dissection during a conventional synchronous combined APE is often carried out along the mesorectum, all the way down to the pelvic floor and the top of the puborectalis muscle, with the mesorectum mobilized off the levator muscles. The perineal dissection then follows the external sphincter to meet the pelvic dissection at the top of the anal canal ( Fig. 1 A). With this technique the retrieved specimen often has a typical waist at 3 to 5 cm from the distal end, corresponding to the top of the external sphincter at the level of the puborectalis muscle and the lowest part of the mesorectum (see Fig. 1 B).

( A ) The pelvic dissection in a conventional synchronous combined APE is carried along outside the mesorectal fascia down to the top of the anal canal ( blue line ) and the perineal dissection is carried along the external sphincter ( red line ). The two dissection planes meet at the level of the puborectal muscle, which creates a waist on the specimen. ( B ) Photograph of a fresh specimen after a conventional APE, with the typical waist at the level of the puborectal muscle.

This inward coning at the pelvic floor carries the dissection close to the rectal wall and several studies have reported higher rates of bowel perforation and tumor involvement of CRM after APE compared with AR. Nagtegaal and colleagues assessed 846 AR specimens and 373 APE specimens from the Dutch TME trial and found that the plane of resection was within the sphincter muscle, the submucosa, or lumen in more than one-third of the APE cases, and in the remainder was on the sphincter muscles. This resulted in a positive CRM rate of 30.4% after APE versus 10.7% after AR and a perforation rate of 13.7% after APE versus 2.5% after AR. Similarly, population-based reports from Sweden, Norway, and Holland have shown a 3-fold increase in perforation rates after APE compared with AR (14%–15% vs 3%–4%) and that perforation is a significant risk factor for adverse outcomes regarding local control and survival. In addition, a publication based on the Dutch TME trial reported that tumor involvement of the CRM was an independent risk factor, both for local recurrence and survival, in patients undergoing APE. Thus, the differences in oncological outcomes between the conventional type of APE and AR may to a substantial part be explained by the increased risk of tumor-involved margins and inadvertent bowel perforations, because both these factors are significantly related to local control and survival.

With the development of TME leading to substantially improved results after AR, many surgeons have advocated low or ultralow AR, even for tumors of the lower rectum. It has also been shown that these procedures are feasible and oncologically safe, provided that the tumor can be removed with a clear distal and circumferential margin. In dedicated and highly specialized centers, adopting intersphincteric AR for appropriate cases, the overall APE rate may be below 10%.

The functional results after an ultralow AR may be poor, however, especially if patients have received preoperative radiochemotherapy. In patients with a preoperative history of gas or fecal incontinence, careful counseling is, therefore, mandatory and information should be given about the risk of a poor functional outcome after AR. In such patients, a permanent stoma may be preferable.

If a tumor in the lower rectum is more advanced, growing close to or into the distal mesorectal fascia, the levator muscle, or the external sphincter and, thereby, threatens the potential CRM, it may not be possible to perform a safe AR and in these cases an APE is necessary. The decision of when to recommend an APE is, therefore, related to both the patient and the tumor characteristics. Because such variables are interpreted differently between different surgeons, the rate of APE varies greatly between individual surgeons and between different institutions. Morris and colleagues reported that the rate of APE varied from 8.5% to 52.6% between different English hospitals. In Sweden, the rate of APE for low rectal cancer, as defined by tumors within 6 cm from the anal verge, has varied between 80% and 92% during the past 15 years ( Fig. 2 ). Thus, APE is still a common operation for low rectal cancer and because the results have been suboptimal, it is important to change the concept of APE in order to reduce the rate of inadvertent bowel perforations and tumor-involved margins and, thereby, obtain improved oncological outcomes.

Graph from the Swedish Rectal Cancer Registry showing the proportion of patients with rectal cancer below 6 cm, operated on with APE, LAR, or Hartmann procedure annually since 1995.

( Adapted from Påhlman L, Bohe M, Cedermark B, et al. The Swedish rectal cancer registry. Br J Surg 2007;94(10):1285–92; with permission.)

The new concept of APE

One problem associated with the conventional type of synchronous combined APE is the lack of standardization and a clear definition of the details of the perineal part of this procedure. Although the abdominal part of the operation follows the standard TME principles, there has been no agreement on the surgical details of the perineal part of the operation. This probably explains the significant variability in the observed rates of tumor-involved margins, bowel perforations, local recurrence, and survival. Due to this variability and the suboptimal results after APE, there has been a call for a different concept and a more standardized approach to APE. In recent years, a new concept of APE has, therefore, evolved, which takes into account the specific anatomic structures of the perineum and the pelvic floor and which aims to adopt and standardize the procedure according to the characteristics of the patient and the tumor. Basically, three types of APE can be described in relation to the perineal approach and the extent of dissection—the intersphincteric APE, the extralavator APE (ELAPE), and the ischioanal APE—and the indications are different for these three procedures (shown in Box 1 ).