John of Arderne is credited with first recognizing the signs and symptoms of rectal cancer in 1376 [1]. Although there appeared to be some rudimentary understanding of its natural history, no form of excisional surgery was performed for nearly another 400 years.

The earliest procedures were mostly palliative. Giovanni Morgani first proposed resection of the rectum in the eighteenth century [2]. Treating rectal cancers by some form of extirpative procedure had not been considered until then. In 1739, Jean Faget of France made history by first attempting a rectal resection [3]. He believed to be draining an ischio-rectal abscess but instead a perforated rectal cancer was encountered. Faget resected the rectum, leaving the patient with a sacral anus and a disastrous functional outcome.

The use of colostomy as a diverting procedure has been reported since ancient times and it played an early role in the management of rectal cancer. In 1776, Henry Pillmore of Rouen, France, performed the first colostomy in an adult for an obstructing “annular scirrhous” carcinoma though the patient eventually did not survive [4]. Colostomy achieved an important role when a French surgeon by the name of Amussat urged that it be the routine procedure for obstructing rectal cancer [5].

Jacques Lisfranc is credited for performing the first successful excision of a rectal tumor in 1826 [1]. Within 7 years, he performed nine additional perineal or posterior resections, of which five were considered successful [2]. These were performed without anesthesia or hemostasis. The patients were asked to bear down, the rectum was everted and a limited rectal amputation then performed. This would result in an incontinent perineal anus. Most patients would not leave the hospital and succumbed to hemorrhage and sepsis. The pain was unbearable, local recurrence was common and functional outcome dismal.

Anesthesia and antisepsis advances spurred a significant development of new techniques in the following decades. In 1873, Aristide Verneuil modified Lisfranc’s perineal resection and removed the coccyx to allow for better exposure and a more radical excision [6]. The conventional perineal approach had resulted in poor exposure of the upper rectum up to that point. In 1876, Theodore Kocher pioneered the transsacral resection with coccygectomy to excise the rectum and anastomose the colon to the anus [3, 7]. Around the same time, Paul Kraske had developed his own technique to remove the rectum, which he presented in 1885 at the Congress of the German Society of Surgery [1, 2]. He removed the coccyx and part of the left wing of the sacrum and preserved the anus and sphincters to allow for a potential anastomosis. Restoring intestinal continuity via the sacral approach was often problematic due to tension on the upper segment and inadequate blood supply. In general, the perineal and sacral approaches provided limited exposure, precluding radical resection of the tumors.

Others experimented with transvaginal resection of rectal tumors. These techniques are, at present, of historical value. Norton reported in 1889 the excision of a tumor of the anterior rectal wall not involving the vagina. The sphincter muscles were resected along with the rectum. In 1890, MacArthur was unable to mobilize the bowel enough to bring it to the skin while operating on a patient with recurrent rectal cancer. He, therefore, sutured it to the upper vagina. Byford reported in 1896 a singular method in which the vagina was used to replace the excised portion of the rectum. The proximal and distal portions were sutured to different portions of the vagina and the vaginal opening was closed [8].

Nearly 100 years after Lisfranc initial perineal resection, Lockhart-Mummery from St Mark’s Hospital in London revised the technique so it would allow for a relatively safer operation [9]. He would first perform a permanent loop colostomy and determine if the tumor was resectable. A week to 10 days later the perineal stage would take place. Removal of the coccyx with the patient in semi-prone position would allow for rectal and anal mobilization; the peritoneum was then opened and as much bowel as possible was pulled down and resected. In 1926, he reported a series of 200 patients in which an 8.5 % mortality was noted, much lower than that of the abdominoperineal resection at the time. A 50 %, 5-year survival without recurrence was observed, though it is said that he rejected about 50 % of his cases that were deemed unresectable [8, 9]. This posterior excision, as it was called, remained popular until the 1940s. The main drawback was that it left the superior lymphatics unresected; therefore, it was not an adequate cancer operation nor was it applicable for upper rectal tumors.

A small variant of the sacral resection, the York-Mason modification of the Kraske procedure, has been used to resect small distal rectal tumors through a presacral approach [10]. This technique of dividing and subsequently restoring the anal sphincter is rarely used anymore and has been replaced by either transanal procedures or ultralow resections with coloanal anastomosis.

Early attempts at abdominal resection of tumors were experimental and performed with little attention to oncological principles. Carl Gussen​bauer, an assistant to Billroth, performed the first abdominal resection of a rectal tumor with intraperitoneal closure of the distal rectum [11]. The first reported case combining abdominal and perineal approaches was performed by Vincenz Czerny in Germany [9]. In 1884, he was unable to remove a rectal cancer using a posterior perineal approach alone and decided to complete the extirpation through the abdomen by turning his patient supine. In 1904, Charles Mayo [8] first presented his technique of abdominoperineal resection (APR) at a meeting in Portland, Oregon, stressing the importance of resecting the lymphatics above the rectum, as high as the sacral promontory. The sigmoid colon was divided at that level and the inferior mesenteric artery transected as high as possible.

The problem of local recurrence was evident among surgeons at the time, including Sir William Ernest Miles. He had been a pupil of Harrison Cripps, who was well known for his work on rectal cancer and the introduction of the perineal approach in England [2, 7]. Miles had witnessed local recurrences within the pelvis in 54 of 57 of his patients excised by this mean [12]. He analyzed postmortem dissections and realized a more radical excision was needed, based on a better and new understanding of the perirectal lymphatic spread.

In 1908, Miles described a modification of Czerny’s operation and emphasized the downward, upward, and lateral spreads of the cancer, with the upward being the most important in his opinion [13, 14]. He considered even the most talented surgeons were unable to completely excise the mesorectal lymph nodes proximal to the tumor via the perineal approach. His operation started by creating a loop colostomy and dividing the bowel 2 in. below it. The distal bowel was mobilized until it could be pushed down into the pelvis and the peritoneum could be closed over it. The patient was then positioned in the right semi-prone position, the coccyx resected and the excision completed from the perineal approach. The procedure was based on five principles including resection of the rectosigmoid and its blood supply, resection of the mesorectum, removal of lymph nodes over the bifurcation of the common iliac artery, wide perineal resection including removal of the levator ani muscle and creation of an abdominal colostomy. Although his original series of 12 patients found 42 % mortality [14], seven survivors were tumor free in 1 year. In subsequent years, he was able to further reduce the mortality associated with the procedure as well as the overall recurrence rate, making the APR the standard of care for rectal tumors. Miles not only revolutionized the principles for a correct oncological resection of rectal cancers, but his approach was a landmark operation in the history of large bowel surgery.

The English pathologist Cuthbert Dukes published in 1930 that there was no significant difference between perineal and abdominoperineal operations for Stages A and B rectal cancer (negative lymph nodes, invasion into or through the bowel wall respectively); but the Miles operation was superior for Stage C (lymph node positives), because the perineal approach would leave the superior lymphatics unresected. This finding validated Miles pathologic premises [12, 15].

Several modifications of the abdominoperineal procedure popularized by Miles emerged in the following years. In 1915, Daniel Fiske Jones proposed a two-stage procedure consisting of an initial abdominal portion followed by a perineal stage 5–7 days later under spinal anesthesia [1]. Jones considered this would decrease sepsis and he reported a mortality of 18 % in 16 patients. Gabriel, a disciple of Lockhart-Mummery, proposed in 1934 a further modification of the APR designated as a perineoabdominal excision [16]. He performed a one-stage procedure starting with a perineal excision, then turning the patient supine and mobilizing the colon through an abdominal incision. Gabriel demonstrated a significant improvement in 5-year survival figures, 30 % vs 17.9 %, for those patients found to have positive lymph nodes via a perineoabdominal excision versus the perineal approach favored by his mentor.

As others emphasized the safety of a two-stage procedure, it was not until 1938 that the one-stage procedure originally described by Miles became commonplace. There was not longer the need to reposition the patient after Sir Hugh Devine introduced the adjustable leg rests in 1937, so the operation could be performed in the lithotomy-Trendelenburg position [2]. Oswald Lloyd Davies was the first to perform a synchronous combined radical abdominoperineal resection in the lithotomy-Trendelenburg position with two teams working simultaneously [1, 2]. The speed and efficiency of the procedure vastly improved with the two-team approach. By the 1960s, the Lloyd Davies technique was the most commonly performed excisional procedure for rectal cancer with a marked reduction in mortality.

With Miles’ operation and principles of resection well established, the focus of interest shifted towards new procedures centered on the restoration of intestinal continuity. The abdominoperineal resection was not only considered too radical by some surgeons but it submitted patients to a permanent colostomy and frequent genitourinary dysfunction.

Some of these techniques had originated in the late nineteenth century. The first documented attempt at restoration of intestinal continuity for rectal cancer is attributed to Reybard of Lyon when he performed a partial sigmoid resection for a colonic growth with immediate anastomosis of the ends [17]. In 1888, the “Durchzug” procedure (pull-through technique) was described by Hochenegg, in which the anorectal stump was everted, stripped of its mucosa and returned to its natural position followed by the distal colon drawn through the denuded rectum and sutured to the anal verge [7]. Despite achieving bowel continuity, this technique was not widely accepted due to the high mortality resulting from anastomotic leaks.

In 1892, Widenham Maunsell of New Zealand described a method for anastomosing the sigmoid colon to the anus [1, 17]. After dividing the anal sphincters in the posterior midline, the rectosigmoid was mobilized through the abdomen and invaginated out through the expanded anus. The tumor was resected and the two ends of the bowel anastomosed. Robert Weir, from Columbia University, New York, later modified this technique in 1901 [8]. Weir mobilized the rectosigmoid through the abdomen in similar fashion; but in contrast to Maunsell, he transected it 3 in. from the anus and pulled out the lower rectum from the perineum using an assistant. The upper bowel was dragged down through the lumen of the exteriorized everted rectum and anastomosed to it.

Babcock and Bacon offered a new procedure in 1939 and 1945 respectively, the delayed union and amputation technique, that basically involved removing the lining of the anal canal and bringing down the mobilized colon through it, leaving about 50 cm outside the body [9]. The previously divided anal sphincters were then sutured to the protruding colon and the excess intestine was removed after 2–3 weeks. With the temporary perineal colostomy, a proximal diversion was unnecessary. Bacon reported lower incidence of male impotence and fecal incontinence than with the APR, and yet similar cancer specific survival rates [9, 15]. In 1961, Turnbull and Cuthbertson from the Cleveland Clinic described their technique, a two-stage abdominoanal pull-through procedure [15, 17]. The rectum was resected, the colon pulled out through the everted rectal stump and the rectum sutured to the seromuscular layer of the protruding colon. Ten days later, and to the patient’s relief, the bowel was finally excised above the dentate line and the end-to-end anastomoses performed.

During the second half of the twentieth century, restoration of intestinal continuity by means of primary anastomosis evolved through the abdominosacral resection championed by Localio [18]. He placed the patient in the right lateral position with the hips flexed, thus avoiding the need for repositioning between the abdominal and sacral portions of the procedure. The abdominal incision was made above the left inguinal ligament, the resection was completed from the sacral approach and a primary anastomosis performed with a 4–5 cm distal stump [8, 18].

Many surgeons were in disagreement with Miles’ oncologic principle regarding downward lymphatic spread as an important pathway for rectal cancer propagation. By preserving the sphincters a radical downward resection could be avoided and therefore better functional outcomes would be achieved.

In 1910, the American surgeon Donald Balfour described a technique of anterior resection through an abdominal approach with the construction of an end-to-end anastomosis between the rectum and the sigmoid colon [2, 8]. In this setting he utilized a “tube support” for the anastomosis after accidentally injuring the sigmoid colon during a procedure. He later suggested his operation could have a role in cancer resections. This technique never gained widespread acceptance due to the high mortality rate related to anastomotic leaks.

The French surgeon Henri Hartmann offered an alternative operation for the treatment of cancer of the middle to upper rectum. In 1921 he described an anterior resection without end anastomosis for high rectal lesions [3]. After resecting the involved segment and its mesentery, the rectum was inverted and left in place. This procedure succeeded in removing the tumor with establishment of a colostomy and avoided the perineal dissection. It was associated with less blood loss and lower mortality than the abdominoperineal resection. The main disadvantage was the necessity of a permanent colostomy. The Hartmann’s resection is frequently applied today in the initial management of complicated sigmoid diverticulitis.