Historical Overview: From Miles' Abdominoperineal Resection to Today's Sphincter-Preserving
Techniques
Rectal Resections before Miles
Giovanni Morgagni in the early 18th century was the first to propose rectal resection
as a treatment for cancer.[1] However, it took over a century before the first successful rectal resection was
performed. Over the years, there have been constant changes in the approach used by
surgeons to perform a rectal resection.
Rectal resections were first performed using a perineal approach as described in 1826
by Jacques LisFranc.[1]
[2] He operated in the era before anesthesia and proper antiseptic techniques. Since
inadvertent entry into the peritoneal cavity could prove lethal, he chose the perineal
approach to evert the rectum and perform a limited resection below the peritoneal
reflection. A limited amount of rectum was resected with this approach and no attempt
was made to purposely include the mesorectum and draining lymph nodes.[2] Success at that time was based upon whether the patient survived to leave the hospital,
and so the risk of local recurrence had very little influence. With the advent of
anesthesia and aseptic techniques, operations could now be performed that would provide
for a more radical resection. Paul Kraske developed a technique similar to Kocher's
technique, where he would incise and detach the left side of the coccyx and sacrum
to provide more exposure. This exposure allowed him to resect the rectal cancer with
½ inch margins on either side of the rectum. The proximal bowel would then be pulled
down and sutured to the anal sphincter complex. However, this “sacral anus” was difficult
for the patient to manage.[1]
[3]
Carl Guessenbauer performed the first transabdominal rectal resection and with closure
and colostomy in 1879. This procedure would be later popularized by Henri Hartmann
for the purpose of treating diverticulitis.[4] Vincent Czerny is credited with performing the first combined abdominal and perineal
approach. This combined approach was never intended but was performed only after an
unsuccessful attempt at resection through the perineal approach.[1] In Vogel's review of 1,500 cases performed by 12 of the most prominent 19th century
surgeons including Billroth, Kocher, Kraske, and Czerny there was a 21% operative
mortality rate with a high (80%) recurrence rate.[3] Surgery certainly had progressed beyond LisFranc's perineal resections; however,
the results were merely palliative in the vast majority of cases.
Sir William Ernest Miles and the Battle to Prevent Local Recurrence
The improvement in mortality for the surgical treatment of rectal cancer from 100
to 4% over the past 250 years is also due to a great many advancements in medicine.[1] None has been more important than the growing knowledge of rectal anatomy and sophisticated
pathology evaluating tumor spread and lymph node involvement. Sir William Ernest Miles
can be credited with the first to emphasize the importance of these factors in preventing
local recurrence and reducing mortality. Unfortunately, the knowledge he gained in
these areas came at the expense of his surgical failures. Like all of his predecessors
and surgeons of his era, Miles would witness recurrence of cancer following perineal
resection. Frustrated by these failures, Miles exhaustively studied the results of
his operations from 1899 to 1906. He observed the clinical natural history in inoperable
patients and performed extensive post mortem dissections on these and his own patients.
Through all of this, he hoped to gain a better understanding of the usual routes by
which rectal cancer spread.[1]
[5] He first observed recurrences in the ischiorectal fat, leading him to make wider
perineal excisions. After this modification, all 14 of his patients had recurrence,
but now these recurrences followed a more lateral spread. In the 11 operations to
follow, he decided to excise the rectal fascia propria and mesorectum, along with
the levator ani muscles and opening up the peritoneum to excise another 3 cm of proximal
bowel. All 11 had recurrence, with the majority involving the mesenteric tissues more
proximal and intra-abdominal to his dissection. Further attempts to excise the entire
mesorectum proved impossible with the majority of recurrences occurring in the pelvic
mesocolon. Through all of these modifications, Miles doubled the time before recurrence,
but a 100% recurrence rate mandated a new approach.[1]
[2]
[5]
By now Miles had distinguished, based upon his observations, three zones of spread.
With his own modifications in technique, he was able to successfully eliminate both
the downward and lateral spread of rectal cancer. However, he knew from his own operations
that the zone of upward spread could not be eradicated through the perineal approach.
This would lead him to first performing an abdominal midline approach followed by
a perineal resection. He was not the first to combine these approaches and he himself
acknowledged such contributors in his landmark article. He would, however, further
clarify that these prior attempts had failed to eradicate the zones of upward spread.[5] Like Czerny and surgeons after him, an abdominal approach was mainly used to provide
more mobility to the rectum to facilitate the rectal resection through a perineal
approach. Miles used this abdominal approach to gain better access to and eradicate
his perceived zone of upward spread. Specifically, the areas involved included the
pelvic mesocolon, lymph nodes over the left common iliac artery, and the peritoneum.
From his extensive observations in the spread of rectal cancer, he then formulated
“certain essentials in the technique of the operation.”[5]
-
The abdominal anus is a necessity.
-
The whole of the pelvic colon must be removed because the zone of upward spread involves
its blood supply.
-
The whole of the pelvic mesocolon below the point where it crosses the common iliac
artery, together with a strip of peritoneum at least an inch wide, on either side
must be cleared.
-
The lymph nodes around the bifurcation of the common iliac artery are to be removed
in all instances.
-
The perineal portion of the operation must be performed as widely as possible to fully
eradicated both lateral and downward zones of spread.
In his landmark article, Miles revealed the results of his first 12 operations for
which he called an abdominoperineal excision. He acknowledged that the 41.6% mortality
rate was indeed too high and that with further experience he could decrease this rate.
Overtime, Miles would be able to decrease the rate of recurrence to 29.5%, which was
a monumental achievement when compared with the usual rates of recurrence at that
time.[1] Miles' abdominoperineal excision would eventually become the standard procedure
in the treatment of rectal cancer, but only after further refinements in technique,
in addition to advances in anesthesia and blood transfusions.
Surgeons now began to recognize the importance of the upward spread of rectal cancer.
Some, like Lockhart-Mummery, recognized this pattern of spread but believed that actual
cancer involvement of these nodes indicated incurability.[6] Both Miles and Lord Moynihan emphasized this pattern of spread and each advocated
for proximal ligation of the inferior mesenteric artery. Unlike Miles, Moynihan, however,
believed the best point of ligation was even higher and should be done proximal the
left colic branch. This debate continues to this day.[7]
Birth of Sphincter-Sparing Techniques
Miles' Major Error
The abdominoperineal resection (APR) as described by Miles was recommended for all
rectal cancers including those above the peritoneal reflection, which are currently
treated with sphincter-sparing low anterior resections (LARs) of the rectosigmoid.
This belief was predicated on Miles' erroneous assumption that these more anterior
rectal tumors could spread in a downward direction. Since the APR was the gold standard
of the day, this theory would have to be disproven before colostomies could be prevented
and bowel continuity restored.
While some surgeons questioned the morbidity involved in performing the Miles' APR,
others began to question its need for all rectal cancers.[1] Cuthbert Dukes, a pathologist at St. Marks Hospital in London, well known for his
classification system of rectal cancer, saw a disparity between his clinical observations
and that of Miles. He often saw lymph node spread parallel and proximal to the tumor
and felt that downward spread in rectal cancer had been overemphasized. Other pathologic
studies during the 1930s likewise contradicted Miles. There now became a shift toward
anterior resection and anastomosis for mid and proximal rectal cancers.[7]
Unfortunately, the mortality rate was high for anterior resections, as first described
by Balfour, secondary to leaks from the anastomosis.[1]
[7] Claude Dixon, in 1948, was the first to prove that anterior resections could be
safely done. He presented to the American Surgical Association his results from some
426 anterior resections and showed that a very low 2.6% mortality rate could be achieved.[1] Further, he proved with his results that such resection was oncologically sound.[2] In 1970, Sir Alan Parks, at Saint Marks Hospital, showed that rectal cancers even
closer to the dentate line cutoff could be safely resected and a coloanal anastomosis
performed. He achieved comparable results for cancers treated with APR. All of Dixon's
resections involved cancers 6 cm or more from the dentate line.[1] Sphincter-sparing operations became the goal once the operation was proved to have
comparable oncological outcomes as the gold standard of APRs. Progress continues toward
this goal.
Challenging the Distal Margin
One of the major concerns in sphincter preservation is the ideal distal margin. The
initial 5 cm “safe margin” as set forth by Goligher, Dukes, and Bussey was a severe
limitation for the surgeon treating a patient with a low rectal tumor.[7] Even during its initial adoption, the 5 cm rule was constantly being challenged.
Not until the 1990s was this changed to the 2 cm rule. Currently, in the setting of
neoadjuvant chemoradiation, even this 2 cm margin has been challenged to require only
a negative margin on the pathologic specimen. Those tumors that are in the distal
3 cm of rectum require close margins to spare the sphincter and transition zone of
the anal canal.[7] Surgeons are attempting to resect even lower tumors with subsequent reconstruction.
There remains a need to improve and create new techniques to meet these new technical
demands.
Technical Improvements
The major challenges of rectal cancer surgery are due to its location within the bony
confines of the pelvis. These challenges increase significantly in morbidly obese
patients, male patients with a narrow pelvis, in some patients with prior chemoradiation,
and in large, locally advanced cancers. Sphincter preservation is less likely in the
larger but completely resectable tumors. Current data support the belief that surgeon
volume, and presumably skill, is one of the two most important factors in sphincter
preservation.[8]
[9]
While technical ability has always varied among surgeons, technology has usually been
able to help equalize this variation. For rectal cancers, the surgical stapler was
the device which helped surgeons work more efficiently in this technically demanding
location, facilitating safe anastomoses and even lower resections. Staplers were first
used by Humer Hultl in 1908 for a gastrectomy.[1] Fifty years later, Mark Ravitch brought back the design of earliest Russian staplers
and, following a few modifications, began using them in the United States.[1] The most important of all staplers for the colorectal surgeon, the end-to-end anastomosis
(EEA) stapler, was first successfully used in 1977.[10] Until 1977, the low anastomosis was performed much like a cardiac valve procedure
by “parachuting” a hand sewn end-to-end suture line into the deep pelvis. The EEA
stapler allowed for more efficient and technically sound anastomosis in the low pelvis.
The widespread use of the EEA stapler in the late 1970s and 1980s significantly reduced
the need for APR.[10]
Laparoscopy has provided to colorectal surgery its well-known benefits, such as decrease
length of stay and better pain control. However, the recovery of bowel function and
anastomotic healing is more influential on length of stay than reduction in abdominal
incision length as seen in other operations. The equivalency of oncologic outcomes
between laparoscopic and open colon resections has been proven in multiple randomized
control trials. The equivalencies of laparoscopic and open treatment of rectal cancers
have been an ongoing debate, and currently randomized trials are in progress. The
one overwhelming advantage when using laparoscopy is the improved visualization during
dissection. R. J. Heald, the major influence in the adoption of the total mesorectal
excision (TME), has likewise acknowledged the superior visualization when using the
laparoscope. TME requires dissection in an areolar tissue plane in the pelvis outside
the mesorectal fascia. However, the laparoscopic technique within the pelvis is limiting
due to the angles of approach and linear effector positions.
Robotic surgery has the possibility to extend this further, not only by improving
the visualization during surgery but also by potentially overcoming some of the technical
limitations that exist when performing laparoscopy in the pelvis. Improved retraction
and the advantage of wristed or articulating instruments facilitate dissection within
the narrow pelvis. Indeed, the literature has shown evidence of its merits for rectal
resections in the male pelvis as well as mid- and low-rectal tumors.[11]
[12]
[13] Whether any of these benefits will outweigh the cost of robotic assisted surgery
will be addressed in the RObotic versus LAparoscopic Resection for Rectal cancer (ROLARR)
study.
Neoadjuvant Therapy
In the era before the introduction of TME, surgical resections which did not focus
on maintaining the mesorectal envelope resulted in a 15 to 45% rate of local recurrence.
Even though surgical techniques had advanced substantially, these rates of recurrence
proved there was a need for additional recurrence-lowering therapies. Radiation to
the pelvis to treat locally advanced rectal cancers began in 1914.[1] Based upon trials in the 1980s, 5FU-based postoperative chemoradiation became the
standard of care up to the 1990s.[14] Postoperative chemoradiotherapy was seen to significantly improve the low recurrence
rates achieved with TME. Despite the improvement in recurrence rates, postoperative
chemotherapy was associated with high toxicity rates and poor functional outcome after
sphincter-sparing surgery. In an effort to improve the tolerability of chemoradiation,
multiple clinical trials were performed testing preoperative with postoperative administration.
Trials like the German Rectal Cancer Study Group showed that improved local control
and reduced toxicity could be achieved when chemoradiation was given preoperatively.[1] Additional studies showed that the addition of neoadjuvant chemoradiation to TME
could reduce the rate of local recurrence from 8.2 to 2.4%.[14]
Currently, neoadjuvant chemoradiation is believed to improve the rate of sphincter
preservation in two ways. First, as the tumors are downsized by a response to chemoradiation,
the tumor becomes smaller and more manageable with less collateral tissue resection,
which may convert the patient to an LAR instead of an APR. Second, for patients whose
tumors have had a complete pathologic response, the option for local excisions or
observation exists, but is currently under investigation. Unfortunately, most large
studies have been unable to show any significant increase in the rate of sphincter
preservation with the use of neoadjuvant therapy despite clearly showing its oncological
advantages.[15]
[16] These results are still not clear when one looks at whether short or long course
therapies were given or how long of an interval to wait.[16] The use of local excision offers the potential for sphincter preservation, but it
is still being explored.
The skilled surgeon is ultimately much more likely to have increased rates of sphincter
preservation. This fact is exemplified by Heald's 89.6% rate of sphincter preservation,
far surpassing even specialized centers. In this highly selected group of people,
only 9% of this group received preoperative radiotherapy.[17] In Europe and Scandinavian countries, it was not until after education-based workshops
that the rates of TME and sphincter preservation significantly improve.[18] At present, chemoradiation should mainly be considered for its improvement in the
rate of local recurrence and tumor downstaging, but in the future better knowledge
of tumor biology may improve the oncological success rates for techniques of local
excision.
Sphincter-Sparing Techniques for Distal Tumors
Reported rates of sphincter-sparing resections among colon and rectal specialist have
been reported as high as 70 to 90%.[9] Nationally, only about one out of every two patients will be given a sphincter-sparing
operation.[19] Even with the ability for neoadjuvant therapies to provide tumor downsizing, tumors
located in the distal one-third of the rectum continue to challenge our ability to
provide both an optimal oncological resection and a sphincter-sparing operation. These
challenges are mainly due to funnel shape of the pelvis anatomy. It becomes increasingly
more difficult to adequately resect the cylindrical piece of tissue consisting of
the rectum and mesorectum from the progressively narrowing funnel-shaped pelvis. This
is why obtaining adequate lateral margins is often much more difficult than to obtain
clear distal margins. The challenges of the pelvic anatomy are even worse in patients
who have a large amount of intra-abdominal and pelvic fat resulting in positive margins
and noncurative resections.
The introduction of the EEA stapler in 1979 and the universal adoption of the double
stapling technique, first described by Knight and Griffin in 1980, overcame some of
the technical challenges in performing LARs and certainly facilitated the progress
toward performing more sphincter-sparing operations. The conventional double stapling
technique is much less helpful, however, in attaining an adequate distal resection
margin for tumors located in the distal one-third of the rectum. The narrowness of
the low pelvis makes placement of a linear stapler difficult even in an open operation.
Laparoscopic endostaplers are more likely to produce tangential resection line and
require multiple staple lines. This may result in positive distal margins or the risk
of anastomotic leak, respectively.[20]
[21]
Technical challenges that are faced when resecting these very low rectal cancers has
stimulated surgeons to invent an array of new techniques which overcome the inadequacies
of the existing conventional methods to overcome these challenges. These new techniques,
however, have only been able to be utilized by the more highly trained and specialized
surgeons, and are still an emerging field of Colon and Rectal surgery. In the United
States, most rectal cancers are resected by surgeons experienced only in the conventional
open or laparoscopic methods that are more suited for more proximal cancers.
The first methods created to combat low rectal tumors were pull through procedures
first described by Maunsell in 1892 and later described by Cutait and Turnbull in
the 1960s and 1970s. These operations removed the entire dentate line and pulled the
colon through the anus to allow the fusion of the bowel to the anal canal. The redundant
colon, which became gangrenous, was then amputated at the anal verge 7 days later.
Sir Alan Park later refined the pull-through into performing just a coloanal anastomosis
in 1972.[22] Later, surgeons would begin to exploit the natural plain between the internal and
external sphincter muscles to begin the dissection before excising the rectum along
with some or all of the internal sphincter. This procedure would be further refined
by Dr. Gerald Marks in 1982 and given the name transabdominal transanal proctosigmoidectomy
or TATA.[23]
Sphincter-preserving techniques such as the intersphincteric resection were created
to better delineate the distal margin and allow for an easier resection.[24] These bottom-up dissections continue to be extended cephalad further and further
owing to the ease with which the mesorectum can be visualized and resected in the
narrow male pelvis. This in-line visualization of the pelvic structures aids in a
better quality mesorectal resection overall, particularly in the distal most pelvis.[24] A different array of platforms such as the transanal endoscopic microsurgery (TEM)
and transanal minimally invasive surgery (TAMIS) have allowed for improved visualization
through endoscopy and use of longer instruments passed through the anal canal opening.
Local Therapy
Due to the historically high rates of local recurrence in rectal cancer, transanal
excision without radical resection was only rarely considered. Recently, three factors
have resulted in an increase in the overall interest in not only sphincter preservation
but also organ-preserving treatments. First, despite improvements in technique and
postoperative care, rectal resections with TME are associated with significant postoperative
morbidity. Second, the rates of tumor downstaging and even complete pathological response
following neoadjuvant therapy have brought to question the utility of performing a
resection of the mesorectum when there are no viable cells found in the tumor. Third,
new technologies such as TEM and TAMIS have made local resections technically more
feasible.
Currently, the National Comprehensive Cancer Center Network Guidelines state that
candidates for full thickness local resection include Tis and T1 tumors up to 3 cm
in size which are well to moderately differentiated, less than one-third the circumference
of the rectal lumen and within 8 cm from the anal verge. Any local resection that
results in a final margin less than 1 mm, or cancers revealing lymphovascular invasion,
poor differentiation, or occupy the lower one-third of the submucosa requires a more
radical resection.
One of the main issues with local resection is that even small lesions, up to T2,
can have a 19% lymph node positivity.[25] Determination of which tumors (based upon depth, differentiation, or imaging characteristics)
are likely to have lymph node involvement has yet to be truly defined. Most recently,
excellent staging has been achieved through imaging of the pelvis with a specific
type of MR that uses body surface coil phased array scanning and processing of images
to provide cross-section images of the rectum. Prior studies have shown that local
excision for T1 lesions has a much higher recurrence rate (13.2 vs. 2.7%) than LAR
with TME.[26] A recent study using the Surveillance, Epidemiology, and End Results (SEER) database
has shown equivalent oncologic outcomes in patients with T1 cancers as a whole. This
same study showed that local excision is safe for downstaged tumors following neoadjuvant
chemoradiation, including T2 tumors.[27] These results are quite improved over the prior studies which showed up to a 15%
recurrence rate.[28] Certainly, there will need to be improvements in our ability to characterize tumor
response to neoadjuvant therapy radiologically before local excision can be routinely
recommended.
Endocavitary contact radiation (ECR) or radiotherapy is another technique for treatment
of early-stage rectal cancers in selected patients. ECR was first used in 1946 to
treat rectal cancer by Lamarque and Gros.[29] ECR offers an additional form of local treatment of T1 and possibly T2 rectal cancers
with similarly reported failure rates when compared with local excision.[30] ECR does not disrupt any anatomical planes and therefore, if it fails, does not
inhibit salvage surgery. Disadvantages include lack of a surgical specimen for a full
histological examination to better predict the likelihood of failure as is used with
local excision. Overall, this form of therapy is not likely to gain any more support
in the 21st century than it did in the 20th century.
Rectal Cancer Today: From Personalizing Our Approach to Standardizing Our Care
The surgeon has always been at the forefront in the fight to constantly find better
methods to improve the outcomes and the rate of sphincter preservation when treating
rectal cancer. However, some of the improvements in the treatment of rectal cancer
are due to efforts from physicians of other disciplines.
Early in the history of treating rectal cancer, the pathologist began to play an important
role, not just in staging the resected tumor, but also in guiding the surgeon on what
tissues outside the rectum to include in the resection. Through the observations of
the pathologist Cuthbert Dukes, surgeons began to transition from the more radical
APRs to the sphincter-sparing LAR.[1] In 1986, the pathologist Phil Quirke demonstrated the importance of lateral tumor
spread in local recurrence. He showed that inadequate resections which lead to positive
radial or circumferential resection margins increase the rate of local recurrence.[7] These results would add further emphasis to the importance of a TME previously established
in 1982 by British surgeon Bill Heald.[7] Through their efforts, the TME would become the gold standard technique for low
rectal cancers and decrease local recurrence rates from a high of 20% down to 4%.[7]
Neoadjuvant chemoradiation is now the standard of care and can lower local recurrence
rates when combined with TME compared with TME alone.[1] Improvements in imaging have dramatically increased our ability to clinically stage
a tumor prior to surgery or treatment with chemoradiation. An accurate assessment
of the pretreatment stage can help individualize treatment and prevent the morbidity
associated with chemoradiation in patients with early rectal cancers. With this advanced
knowledge and individualized approach, the treatment of rectal cancer has now become
a complex decision-making process that requires specialized knowledge from multiple
disciplines.
The Multidisciplinary Approach
Once a disease whose treatment involved only the technical skills of a surgeon, the
treatment of rectal cancer now requires the accurate input and assessment from an
array of specialties. Unfortunately, many patients have been treated based upon the
old U.S. generalist-centered model resulting in variability of care as seen in local
recurrence rates, mortality rates, and permanent stoma rates.[31] To address these discrepancies and improve patient care, many countries, including
the United States, have created centers of excellence. Paramount to these centers
of excellence is the multidisciplinary team (MDT) which administers the treatment
of rectal cancer to each individual patient using standard care pathways. These standard
care pathways are based upon five evidenced-based principles of rectal cancer treatment
as outlined by the OSTRiCh (Optimizing the Surgical Treatment of Rectal Cancer) group.[31]
-
TME
-
Measuring the quality of surgery or TME through specific pathology assessment techniques
-
Specialist imaging techniques identifying patients at high risk of local recurrence
-
Administering newer and more effective neoadjuvant and adjuvant therapies
-
Using an MDT approach to identify, coordinate, deliver, and monitor the ideal treatment
on an individual patient basis.
The ability of the TME to reduce local recurrence has been discussed. This method
is easy to teach and, with its widespread use, has been shown to decrease permanent
stoma rates, decrease local recurrence, and even improve 5-year survival rates in
certain population-based studies.[31] Pathologic assessment of these specimens provides important prognostic factors related
to circumferential margin status as well as a quality indicator of the type of resection
performed. Pathologic assessment of the quality of resection provides surgeons with
direct feedback and allows surgeons to continue to improve and refine their techniques.
MRI is currently the standard for pretreatment imaging of rectal cancer in Europe
and will soon become the standard in the United States. The use of MRI within the
MDT format has been shown to reduce the incidence of positive circumferential tumor
margins.[32]
[33] MR volumetric analysis can reliably predict a tumor's clinical response following
neoadjuvant therapy and identify patients with low rectal tumors that are amenable
to a sphincter-sparing resection.[32] The use of MRI in the pretreatment assessment of tumors could also allow for a more
tailored approach when using neoadjuvant therapy, thereby reducing some of the morbidity
associated with rectal cancer treatment.[31]
The MDT consists of surgeons, radiation and medical oncologists, pathologists, and
radiologists. These MDTs have been shown to improve clinical decision making and clinical
outcomes in rectal cancers. The American College of Surgeons Commission on Cancer
has listed the rectal cancer MDT as a key requirement in the treatment of rectal cancer.
