Return to specialty index |
The development and refinement of pelvic pouch surgery now allows the excision of a diseased colon while maintaining transanal faecal continence. The success of restorative proctocolectomy is largely dependent on careful patient selection combined with meticulous surgical technique. The authors discuss the main indications for restorative proctocolectomy and describe the surgical procedure.
Keywords: Familial polyposis coli, restorative proctocolectomy, pouch, ulcerative colitis, technique
Restorative proctocolectomy (RP) or ‘pouch’ procedure is arguably the gold standard operation in the surgical management of ulcerative colitis (UC). It is also indicated in selected cases of familial polyposis coli (FPC) and constipation. Since the initial description by Parks and Nicholls in 1978 the operation has undergone many modifications. (1) Surgeons have adopted a variety of techniques based on their own experience, their understanding of the pathology and individual surgical abilities. This difficult operation calls for not just one technique, but requires versatility to manage problems as they arise. The authors present their surgical preferences and the rationale behind choosing certain procedures.
INDICATIONS FOR SURGERY
The main indications and relative frequency for which the pouch procedure is performed are summarized below.
Ulcerative Colitis
This is the main indication for this procedure. The majority of patients will have a chronic form of the illness that has become resistant to medical management while a substantial number will present with a fulminant form of the disease. There is little disagreement regarding the choice of surgery for those patients presenting acutely. In such a situation, total colectomy and ileostomy is the safest and most expedient operation. Most patients recover within a few months and may wish to consider pouch surgery at a later date if histological examination confirms ulcerative colitis. Restorative proctocolectomy in such urgent situations is potentially dangerous as the patient is often underweight, anaemic and hypoalbuminaemic and is invariably on high-dose steroids. This combination increases the risk of complications and morbidity. However, there are reports of pouches being performed on patients falling between these two extremes. Such patients are unwell but are on low doses of steroids, have lost only a moderate amount of weight and have mild anaemia and hypoalbuminaemia. The decision to perform pouch surgery on these patients requires fine judgement and considerable experience if minimal morbidity is to be achieved. In patients with chronic disease who are generally well, a restorative procedure can be offered without prior colectomy. The decision to perform a RP in patients with indeterminate colitis has been debated.(2) The patient who is keen to have a pouch in these circumstances must be warned of the increased risk of failure.
Familial Polyposis Coli
This is a well-recognised indication for RP. The rationale is similar to that for UC in that if the colonic and rectal mucosa is removed then the colonic manifestations of the disease can be eradicated. It is particularly useful when there are multiple large polyps or dysplasia in the rectum when a colectomy with ileo-rectal anastomosis is contraindicated.
Crohn's Disease
Most surgeons believe Crohn's disease to be a contraindication for restorative proctocolectomy. The disease often involves the terminal ileum and the perianal area, which precludes RP due to poor pouch and sphincter function. Recently, RP has been used in highly selected patients with obvious absence of small bowel and anal disease with reasonable success. (3)
Constipation
Severe constipation is a complex clinical problem and is due to a number of causes. In a small proportion of patients RP has been used, but with variable results. (4)
Age
Is age a bar to this operation? Several experienced surgeons have argued that if the patent is well motivated, medically fit, understands the complexity of the operation and the possibility of complications, and if anal sphincter function is sound, then pouch surgery can be performed on elderly patients. Comparable results with those in the young have been obtained. (5) Others argue that such a difficult operation should only be offered to patients up to 50 years-of-age. However, all surgeons involved in the pouch procedure would agree that the ideal patient is one who is young, well informed and highly motivated.
Pre-operative Discussion
It is important to spend adequate time with the patient and family to discuss the aims of the operation, the functional results and the potential complications. It has to be emphasised that the aim of the operation is to eradicate the disease, whenever possible, without the need for a permanent stoma. Patient expectations have to be discussed in this context. In the patient with severe ulcerative colitis, the acute diarrhoeal illness and associated faecal urgency can be eradicated by the operation but bowel function will never be normal, even if the operation goes without mishap. The positive feature is that body image will be maintained. Many patients will already have had an urgent colectomy and possess a stoma. Some patients will have adapted to life with a stoma and are completely well. They may not wish to undergo another complicated operation and this decision should be respected. The low risk of post-operative sexual problems and infertility should be specifically discussed.
Although the benefits of pouch surgery are more obvious in the colitic patient, those patients with FPC face a dilemma. The operation is performed to prevent the inevitable development of colorectal malignancy in a young patient with normal bowel function and good quality of life. These patients, therefore, may perceive good pouch function in these circumstances as far from satisfactory.
The message is that in the preoperative phase, several consultations should take place to allow full and frank discussion. It is important to have a team approach involving the surgeon, stoma therapist and, when possible, a patient who has undergone the operation. An information leaflet describing the operation in simple terms, which the patient and family can read at their leisure, is useful. This ideal approach is not practical when the patient requires a pouch in the semi-urgent situation but in all cases the patient must be aware that the operation is a complex procedure, that the end result is unpredictable and that there is a risk of certain complications.
General Considerations and Procedures
Basic steps:
The decision to use bowel preparation should not be a problem. In the elective case, mechanical bowel preparation is preferable but not essential since the bowel is to be removed and with careful dissection there should be little contamination. In those patients presenting semi-urgently bowel preparation is unnecessary. The rectum can be washed out intra-operatively.
Careful positioning of the patient is important to allow good access to all aspects of the abdomen and to the anus. A soft pad placed under the sacrum will lift the anus anteriorly, which is especially useful in the heavy male patient (not all patients are malnourished - especially those who have had a prior colectomy). The patient will lie in the Lloyd-Davis position for several hours and is at risk of developing a neuropathy (lateral popliteal nerve palsy) or compartment syndrome if the legs are not properly placed and protected. ‘Wrapping the patient in cotton wool’ is an important protective effort.
SPECIFIC SURGICAL PROCEDURES
The main steps of the operation are:
Removal of the Colon
After a thorough laparotomy, the colon is mobilised from the ileo-caecal area to the recto-sigmoid junction. With the surgeon standing on the patient's left and an assistant retracting the lateral abdominal wall from the right, the right colon is raised towards the midline. Careful diathermy technique should produce a dry field. The duodenum is recognised and protected and the plane anterior to this is developed. At this point, the thin-walled mesenteric veins are apt to tear if the assistant does not support the mobilised mesentery. Failure to do so can result in troublesome venous bleeding. As the transverse colon is approached, a decision has to be made as to whether the omentum should be preserved. Some surgeons hold the view that if the omentum is left behind then it may lead to the development of band adhesions with a subsequent tendency to intestinal obstruction. It can often be densely adherent to the colonic wall in inflammatory bowel disease and if separation is difficult then removal en bloc with the colon may be easier. In the majority of cases, however, the omentum can be separated from the colon allowing entry into the lesser sac as the left transverse colon is approached. Dissection of the splenic flexure is aided if the surgeon moves between the patient's legs. Retraction by an assistant on each side produces a clear view of the upper abdomen. Care is taken with retraction in order not to damage the spleen. The transverse colon is mobilised up to, but not including, the splenic flexure. Firstly the descending colon is mobilised towards the midline following the white line of Tolt. This should also be an avascular approach. Finally all that remains is to dissect away the remaining peritoneal connections at the splenic flexure. In long-standing UC, the splenic flexure lies some distance from the spleen making mobilisation of the flexure relatively easy and safe.
The surgeon returns to the patient's left side and an assistant stands between the legs to retract the bladder, uterus and anterior abdominal wall forwards and caudally. A second assistant retracts the small bowel, which has been placed in a suitable pack towards the right. This allows the surgeon to mobilise the remaining descending and sigmoid colon. The left ureter is recognised and protected. Even at this point it is important to protect the sympathetic nerves as they course over the aortic bifurcation to avoid inadvertent injury. Once the colon is fully mobilised, the blood supply is divided close to the colonic wall. The ileo-colic vessel is preserved giving the surgeon a choice of which vessels to retain at the time of pouch formation in order to allow a tension-free, well vascularised anastomosis. This process is continued around the colon. The area of the inferior mesenteric artery origin is left undisturbed in order to protect the sympathetic nerves. The mobilised bowel, still attached to the rectum, is placed in a pack and the rectal dissection begins.
Rectal Dissection
The patient remains in the supine position or in a slight Trendelenburg position. Too much tilt can be counterproductive, may produce vascular complications in the lower limbs and tends to hinder the view of the posterior rectal dissection. The assistants play a crucial role in this part of the procedure. It is important that anterior pelvic structures are retracted forwards and caudally (Figure 1), and that the peritoneum overlying the upper rectum is retracted laterally. This allows the surgeon to perform a ‘nerve-sparing’ excision of the rectum using sharp dissection. Again this should produce a relatively bloodless field. Posteriorly, the dissection is initially taken down to the recto-sacral ligament, keeping very close to the fascia of the mesentery.
 |
|
The lateral dissection is continued in this plane before dividing the peritoneum that lies anteriorly across the recto-vesical septum in the male and the pouch of Douglas in the female. It is at this point that the surgeon may stray too far anteriorly resulting in troublesome bleeding. A two-inch Dever retractor can be used in most patients to lift the vagina or the seminal vesicles anteriorly and allow the correct plane to be entered. Denonvilliers fascia is divided by sharp dissection and the anterior rectal wall is identified (Figure 2).
 |
|
The nerves to the bladder and sexual organs should lie anteriorly and be protected. The surgeon returns to the posterior dissection and the recto-sacral ligament is divided and the plane developed down to the pelvic floor (Figure 3). Having achieved this, the lateral and most difficult area of dissection is approached. Again good retraction is critical. If the Dever retractor does not produce good anterior exposure, a Lloyd-Davis retractor replaces it. This allows the anterior structures to be lifted forward. Laterally, the right lateral ‘stalk’ is stretched (a ‘swab on a stick’ is useful here).
 |
|
The surgeon firmly retracts the rectum away from the stalk, which is divided by using a mixture of scissors and diathermy dissection. Occasionally, the middle rectal artery will be found although its presence and size can vary considerably. The nervi erigentes can be identified as the parasympathetic nerves course from the lateral aspect of the sacrum. Retraction is adjusted frequently to allow good visibility. Following a similar approach to the left side, the rectum is freed. At this point the ano-rectal junction is still approximately 2 cm away. The rectum is gently retracted towards the patient's head and a Lloyd-Davis retractor is placed between the rectum and vagina or prostate. Having divided the fascia of Denonvilliers earlier, the retractor can be used to develop this plane by gently separating the areolar tissue aided by scissor dissection down the upper aspect of the anal canal. All that remains is to free any loose connections to the lateral aspects of the rectum. It is useful at this point to define the level of dissection by inserting the gloved left index finger into the anal canal while the point of the assumed ano-rectum is ‘pinched’ from above with the right hand. It is not unusual to be misled as to how far the dissection has reached. In an obese male, it is easy to overestimate the degree of progress and the lower aspect of the dissection may still be several centimetres above the ano-rectum. Conversely, in a thin female it is easy to stray into the anal canal resulting in injury to the internal sphincter. This risk is compounded by the fact that the freed ano-rectum is now very mobile and can be pulled upwards leading to a "shortening" effect of the anal canal. Before dividing and removing the rectum, the rectum is washed out with chlorhexidine acetate BP (0.02%) using a rectal tube passed up through a proctoscope (Figure 4).
 |
|
The ano-rectum is then divided using a TD 30 mm linear stapler. It is important that the assistants provide a good view at this stage to allow the stapler to be placed safely and accurately, avoiding damage to structures such as the vagina. The stapler can be angulated to allow relatively more of the anterior wall of the ano-rectum to be preserved. This leaves a safer distance between the ano-rectum and the vagina to help reduce injury to this structure when the anal-pouch anastomosis is performed. Before finally firing the stapler, the level of division is again checked by either using the finger, proctoscope or both. When the surgeon is completely satisfied, the gun is fired and the rectum removed (Figure 5). The pelvis is washed out with antibiotic solution (cefotaxime 1g in 1 litre normal saline) and careful haemostasis is completed. A dry swab is left in the pelvis.
 |
|
Mucosectomy
Mucosectomy may be necessary in patients with FPC, where there is severe ano-rectal disease in UC or where dysplasia has been detected on preoperative biopsies. It may also be necessary in cases where there have been technical difficulties intra-operatively preventing the use of a stapler. The operation proceeds as described above, but the exact level of division of the rectum is less critical. At a convenient point, for example three centimetres above the dentate line, the rectum is clamped with a right-angled clamp and divided using diathermy. The surgeon sits between the legs, using a headlamp to improve visibility of the anal canal. A well-lubricated Eissenhammer retractor is gently inserted into the anal canal. It is important to open the retractor gently in order to prevent injury to the sphincters. As the assistant holds the retractor, the surgeon injects a few millimeters of a 1:200,000 adrenaline solution into the submucous plane to ‘balloon’ the mucosa. Occasionally, there is very little mucosa present in severe UC and this must be recognised if the internal sphincter is not to be injured. The mucosa is then excised, usually in strips, starting just above the dentate line and protecting the underlying smooth muscle. Fistula scissors are useful for this stage. The retractor will have to be gently repositioned repeatedly to allow sufficient exposure. Complete haemostasis is critical.
Mobilisation of Small Intestine and Pouch Formation
If a prior colectomy has been performed then the ileostomy will need to be taken down. Otherwise, the stapled ileum is retracted to allow the free edge of the peritoneum to be defined and dissected up to the pancreas. All connections to the lateral and posterior aspects are divided and the small intestine totally freed. Any adhesions between loops of bowel are divided allowing it to be freed from the duodenal-jejunal flexure to the divided ileum. If a J-pouch is to be used then the point chosen for the anastomosis is held with a Babcock clamp and a trial descent is performed. The point on the ileum that reaches down to the anus without undue tension is variable and may be a considerable distance from the divided ileum. It may be necessary to excise excess distal ileum otherwise the pouch created may be too large. Ensuring that the ileum can be brought down to the anus without tension can be frustrating, but all attempts must be made to get as much length as possible. The following techniques all help to add length:
Checks are made to ensure that all adhesions have been divided, that the mesentery has been fully freed up to the pancreas and that the rotation is correct. The peritoneum overlying the mesenteric vessels can be divided transversely which results in straightening of these vessels, in turn producing extra length. If further length is needed then the risky step of dividing mesenteric vessels may be necessary. This has to be planned carefully if ischaemia of the pouch is not to occur. A trial application of vascular clamps to those vessels chosen for division is important to ensure that the remaining blood supply is adequate to maintain pouch viability. The area for anastomosis should lie without undue tension over the pubic symphysis. If after all attempts the ileum will not reach, without tension, then the pouch operation should be abandoned and a completion proctectomy performed. Neither the patient nor the surgeon will benefit from pelvic sepsis as a result of a failed anastomosis.
Once the surgeon is satisfied that the ileum will reach the anus then the pouch can be fashioned. A J-pouch with 15-20 cm limbs will usually achieve satisfactory pouch function. The ileum is folded back on itself and stay sutures are placed at the apex. The enterotomy will be made between these sutures. Further stay sutures are placed on the mesenteric border of both limbs at 7 and 15 cm from the apex. These additional stays, when approximated, allow the stapler to be fired without incorporating the mesentery. Inserting the index and middle fingers behind the limbs of the ileum can further prevent this complication while the stapler is fired. This manoeuvre allows the mesentery to be pushed laterally and away from the staple line.
A small enterotomy is made in the apex of the ileum. If this is made too large it becomes difficult to achieve a tight closure of the purse string suture. The arms of a multi-fire 80 mm GIA are lubricated and passed separately up into the limbs of the ileum. They are then locked together and with the mesentery carefully excluded the gun is fired (Figure 6).
 |
|
This manoeuvre is repeated on two further occasions advancing up into the pouch. The gun is cleaned of any excess staples before re-insertion. Once completed, a suture is placed distal to the staple line to hold the ileal limbs together and, thus, prevent tension on the staple line. Infusing warm saline with a bladder syringe placed via the enterotomy checks integrity of the pouch (Figure 7).
 |
|
A suitable sized (e.g. 29 mm) intra-luminal circular stapler is selected. The head is then inserted into the pouch and the enterotomy closed using 2/0 Prolene purse string suture. It is easiest to place the head into the pouch prior to placing the purse string. The purse string sutures are placed 5-7 mm apart and include the full thickness of the ileal wall (Figure 8).
 |
|
Pouch-anal Anastomosis
Although the authors routinely use the totally stapled technique a sutured anastomosis may be necessary after a mucosectomy and so both techniques will be described here. (6)
Stapled Method
The pelvic swab is removed and haemostasis is checked. An assistant positioned between the legs is instructed to pass the gun into the anus. This can be more difficult than it appears as the anus is short and it can be awkward to fully insert the gun. To help with the insertion, the gun and anus are lubricated and the gun ‘shoe-horned’ in. The abdominal surgeon places his right hand down into the pelvis and gently presses onto the anal staple line as the gun is pushed from below. This protects against staple line disruption. Good retraction is maintained as the gun is passed upwards and gently opened. The spike is identified as it emerges through the centre of the staple line, or preferably just behind it. The abdominal surgeon may find it useful to direct this action by holding the shaft of the gun with his left hand. This ensures correct positioning of the spike. As the spike begins to point through the tissue a scalpel is used to stroke the tissue, allowing it to penetrate without suture line disruption.
The abdominal surgeon turns his attention to the pouch, which is orientated correctly, before connecting the two parts of the stapling gun together. The assistant then closes the gun while the abdominal surgeon controls the two ends. The vagina and any extraneous tissue are retracted away from the gun as it closes. The gun is angled to the correct position and is fired by the assistant. The gun is partially opened and gently removed. The doughnuts are inspected for completeness and the anal pouch anastomosis is checked for air tightness.
Sutured Anastomosis
The abdominal surgeon places a 3/0 waxed vicryl suture from within the lumen outwards at the 3 and 9 o'clock positions at the opening of the pouch. The needles are left attached and the sutures are passed down to the perineal operator. This operator has passed a Robert forceps up into the pelvis to grasp the sutures and direct them down and out of the anus, ensuring correct orientation. As the pouch is guided down, each suture is passed through the mucosa of the anal canal picking up a part of the internal sphincter at the level of the mucosectomy. This is aided by use of the Eissenhammer retractor. The retractor is removed as each suture is tied anchoring the pouch at 3 and 9 o'clock. Inserting the Eissenhammer retractor back into the anal canal and the pouch lumen allows sutures to be placed to complete the anastomosis. These sutures are placed individually by ‘backhanding’ the needle, picking up the mucosa and internal sphincter and then the pouch from out to in. The retractor is removed each time the suture is tied. Approximately 12-16 sutures are needed to complete the anastomosis. Integrity is checked by inserting a gloved index finger and by inspection using the Eissenhammer retractor.
Formation of a Loop Ileostomy
The use of a protective loop ileostomy is slightly controversial. It would be fair, however, to say that it should only be omitted when an experienced pouch surgeon feels that the operation has been straightforward and a tension-free anastomosis has been created in a fit patient. The segment of the ileum chosen for the stoma must be able to be brought out without tension. Therefore, it may lie at a varying point from the pouch but it is preferable not to have too long an afferent loop from the pouch. The bowel is brought out through the previous stoma site if a prior colectomy has been performed. A Babcock forceps is placed via the opening to grasp the ileum, which is then delivered. Placing one diathermy mark distally and two proximally ensures correct orientation. In an obese patient it may be impossible to bring out a tension-free stoma. In this situation one may choose to omit a stoma, but if required then it maybe be necessary to divide the ileum to allow the proximal end to be brought out without tension. The closed distal end may be tacked to the underside of the abdominal wall immediately adjacent to the ileostomy for ease of closure.
If there has been no prior colectomy then a trephine has to be made at the pre-marked spot. The skin at the site chosen is picked up with a Lane's tissue forceps and a disc of skin removed. The subcutaneous tissues are spread with Langenbeck retractors and the aponeurosis split in a cruciate manner with diathermy. The muscle is then split with Mayo scissors and the posterior fascia and peritoneum divided to allow access to the abdomen. Two fingers should pass quite easily into the opening. The operation progresses as described above.
The abdomen is washed out with antibiotic solution and a suction drain passed down to the area of the anastomosis. If the omentum remains, it is passed down to the pelvis and over the small intestine to prevent adherence to the wound. The wound is closed using a mass closure technique and a subcuticular suture to skin. The wound is covered and a Brooke's eversion ileostomy fashioned with catgut sutures. A suitable ileostomy bag is fitted.
The mortality rate for such a complicated procedure is extremely low (0.5%). Conversely, the morbidity rate is high, as summarised in Table 1.
The most worrying immediate postoperative complication is pelvic sepsis or, if a covering ileostomy is not used, generalised peritonitis. This complication is caused by anastomotic dehiscence, which is often partial. This is usually secondary to tension at the anastomosis. It is important to recognise this complication early and instigate appropriate management if the anastomosis and subsequent pouch function are to be preserved. Staple line haemorrhage is unfortunate and difficult to manage. Staples are not haemostatic and with a 15-20 cm staple line it is not surprising that this problem occasionally occurs. Intestinal obstruction is the most common complication reported in all published series. It can occur early or late and must be recognised and rectified as necessary since there is a potential loss of important small bowel. Because of the serious nature of these complications, it is important that the surgical team remains vigilant and that the patient is carefully monitored in the immediate postoperative period.
Table 1: Complications after restorative proctocolectomy
| Timing |
Complications
|
Percentage |
|---|---|---|
| Early post- operative period |
Staple line
haemorrhage
|
<2
|
|
Anastomotic
leakage and peritonitis
|
<2
|
|
|
Pelvic abscess |
5-10
|
|
|
Pelvic
haematoma
|
5
|
|
|
Intestinal
obstruction
|
13
|
|
|
Ileostomy
dysfunction
|
5
|
|
|
Urinary
disturbance (temporary)
|
5
|
|
| Late-after hospital discharge |
Pelvic
sepsis
|
5
|
|
Intestinal
obstruction
|
10
|
|
|
Ileostomy
dysfunction
|
<5
|
|
|
Anastomotic
dehiscence
|
<5
|
|
|
Sexual and
bladder problems
|
<4
|
|
| After ileostomy closure |
Peritonitis
- related to stoma closure
|
<2
|
|
Peritonitis
- delayed anastomotic failure
|
<3
|
|
|
Intestinal
obstruction
|
9
|
|
|
Pouch
malfunction
|
5-12
|
ACKNOWLEDGEMENTS
We are grateful to the staff of the Medical Illustration Department at the University of Aberdeen for their skill and patience in the production of the operative photographs.
|
Return to specialty index |