Continent Ileostomy

David W. Dietz

INTRODUCTION

Complete removal of the colon and rectum is sometimes the only treatment option for patients with complicated or medically refractory inflammatory bowel disease (IBD) or certain hereditary colorectal cancer syndromes. Fortunately, patients undergoing total proctocolectomy (TPC) today are almost always able to maintain intestinal continuity with preservation of an anatomic route for defecation through creation of an ileal pouch-anal anastomosis (IPAA). Before the description of the restorative proctocolectomy (IPAA) by Parks and Nicholls in 1978, however, this was not the case. Patients in that era had an end ileostomy as the only option and the inconvenience and stigma attached to the ileostomy often made patients reluctant to consent to needed surgery. Appreciating the difficulties that such patients often encountered, Professor Nils Kock of the University of Gothenburg in Sweden first described a “continent ileostomy” (CI) in 1969. The advantages of the CI over an end ileostomy stemmed from an intussuscepted valve that allowed for creation of a flush stoma that did not require the patient to wear an external appliance to collect intestinal waste. A number of colorectal surgery centers around the world adopted the CI and began to gain experience with the procedure. However, the advent of the IPAA 10 years later, along with the difficulties and complications encountered during the early experience with CI surgery, led to a marked and rapid decrease in its popularity. As time passed, fewer and fewer centers offered CI as an option for patients wishing to avoid a conventional end ileostomy and today only a handful of surgeons familiar with the procedure can be found in practice. Despite this, however, the procedure still plays an important role in the surgical management of certain highly selected patients.

INDICATIONS/CONTRAINDICATIONS

Today, IPAA is the gold standard procedure for the surgical treatment of patients with ulcerative colitis (UC), but CI still remains an option in certain specific situations in patients with UC and in very highly selected patients with Crohn’s colitis. In an unpublished series of 423 patients undergoing CI surgery at the Cleveland Clinic, the final diagnosis was UC in 74%, indeterminate colitis in 5%, and Crohn’s colitis in 10%. Familial adenomatous polyposis (FAP) accounted for the remainder.

There are four common indications for CI surgery, mostly in patients with IBD. These indications are discussed in detail here.

Patients who are referred for redo-IPAA after failure of their index IPAA and are found not to be candidates (approximately 15% of patients).
Patients who are taken to the operating room for IPAA in whom pouch-anal anastomosis cannot be achieved because of inadequate mesenteric length (approximately 5%).
Patients who require TPC but have contraindications (e.g., fecal incontinence, very low rectal cancer, anal canal disease) to pouch-anal anastomosis (approximately 10%).
Patients with an existing conventional ileostomy who find it unacceptable because of severe pouching problems or interference with sex life, athletics, or occupation (approximately 60%).

Failed Ileal Pouch-Anal Anastomosis

Failure of an IPAA is defined as excision of the ileal pouch with closure of the anal canal, permanent fecal diversion with a proximal loop ileostomy, or having an unreversed diverting ileostomy. The long-term risk of IPAA failure varies between 3% and 28% and is influenced by characteristics of
the patient population. Pelvic sepsis complicating surgery and a diagnosis of Crohn’s disease (CD) are strongly associated with pouch failure.

Patients with a failed IPAA are often candidates for redo-IPAA and good results can be obtained. In 502 patients undergoing an IPAA reconstruction, the overall success rate was reported as 80%. Despite this high percentage of success, some patients who are referred for redo-IPAA are not candidates for that procedure. Factors such as sphincter dysfunction and extensive pelvic fibrosis due to peripouch sepsis are contraindications to redo-IPAA, because bowel function and quality of life can be anticipated to be extremely poor. Some of these patients, however, can be offered a CI if it is felt that they will not do well with a conventional end ileostomy. In 64 patients with a failed IPAA, 25% had the existing J-pouch converted into the CI pouch, whereas 75% required excision of the existing pelvic J-pouch with creation of a de novo CI pouch. The length of the remaining bowel is often an issue in these patients, because the previously constructed pelvic pouch may comprise 15% of their existing bowel length. Because a CI requires approximately 55-60 cm of bowel for construction, a patient who undergoes excision of a failed pelvic pouch with creation of a new CI may lose up to one-third of the functional bowel length, because the absorptive capacity of the CI pouch is largely unknown. It is critical to consider and discuss with the patient the “worst case scenario” of subsequent failure of the CI that requires pouch excision. This would potentially leave the patient with only 200 cm of small bowel proximal to an end ileostomy, a bowel length that most would consider on the borderline of short-bowel syndrome.

Technical or Patient-Related Factors that Preclude Ileal Pouch-Anal Anastomosis After Total Proctocolectomy

Occasionally, patients are taken to the operating room for restorative proctocolectomy but the surgeon finds that it is not possible to create an IPAA because of inadequate small bowel mesenteric length. Patients requiring TPC may also be poor candidates for IPAA because of preexisting problems such as poor sphincter function, low rectal cancer complicating chronic colitis, or anal canal/perineal CD. In these situations and when preoperative discussion has been held and consent obtained, a CI can be constructed at the same operation. However, in the former case, especially if not anticipated and discussed before surgery, it is best to construct a conventional end ileostomy. Consent can then be obtained from the patient after surgery and the conventional ileostomy can be converted to a CI 6-12 months later.

Existing Conventional Ileostomy Unacceptable

The largest groups of patients undergoing CI surgery nowadays are those who present with an existing conventional end ileostomy. These patients seek a CI for a variety of reasons, but the common denominator is that they feel their existing ileostomy adversely affects their quality of life. Common concerns relate to the effect of the ileostomy on sex life and limitations on physical activity, especially in young athletes. A small group of patients have skin conditions such as psoriasis or allergic dermatitis related to the stoma appliance. These patients have great difficulty in maintaining a seal and suffer from frequent and embarrassing leakage. In addition, the need for frequent pouch changes, sometimes several daily, can be financially prohibitive. Patients seeking CI for these reasons should be counseled extensively. The risks of CI surgery, both short- and long-term, must be balanced against the patient’s expectations and the likely benefits.

Contraindications to Continent Ileostomy Creation

Existing short-bowel syndrome, or risk of short-bowel syndrome if the CI were to fail and pouch excision was required
Obesity
Unfavorable anatomy
Small bowel CD
Intra-abdominal desmoid disease in patients with FAP

There are several situations where CI creation is ill-advised. Patients with existing short-bowel syndrome, or those who would be at risk of short-bowel syndrome should the CI fail and require excision, are not candidates for the procedure. As discussed, the latter determination requires considerable judgment by the surgeon who must calculate the risk of CI failure in the individual patient,
estimate or measure the length of existing small bowel, and predict the outcome if the patient lost an additional 55-60 cm of small bowel vis-a-vis pouch excision. Obesity, both truncal and visceral, may be considered as contraindications to CI creation. Obese patients have been found to be at increased risk of valve slippage, and a large pannus as well as a thick small bowel mesentery create technical difficulties with construction of both the intussuscepted nipple valve and its exit conduit. Patients with small bowel CD have been found to be at higher risk for CI failure, with failure rates of 20%, 35%, and 50% at 5, 10, and 20 years, respectively (unpublished data). Intra-abdominal desmoid disease, which complicates the course of approximately 10-15% of patients with FAP, can also preclude creation of a CI.

PREOPERATIVE PLANNING

The steps that should be taken to prepare a patient for CI surgery may vary, depending on the indication for surgery. Common to each is the preoperative marking of the patient by either the surgeon or enterostomal therapy nurse. Continent ileostomies are usually sited lower in the abdomen than a conventional end ileostomy, typically at the level of the anterior superior iliac spine and just above the pubic hairline, but still within the confines of the rectus abdominus muscle (Fig. 47-1). The reasons for this relate to the fact that the CI pouch usually resides in the lowest aspect of the right lower quadrant or in the pelvis, the abdominal wall may be less thick at this site, and that cosmesis is improved. In patients being considered for conversion of a failed IPAA to a CI, the health, size, and suitability of the existing ileal J-pouch should be assessed. Significant mucosal disease (pouchitis or CD) precludes use of the existing pouch and is likely a contraindication to formation of even a de novo CI. Extremely small or large ileal pouches may need to be either augmented or reduced in size during conversion to a CI and CI surgeons must possess sufficient experience and creativity to deal with these issues when they arise.

SURGERY

Evolution of the Continent Ileostomy

After its first description by Koch in 1969, the technique for CI creation has undergone a number of modifications. Kock’s initial description of the CI did not include an intussuscepted nipple valve. The primitive design was a U-shaped pouch constructed from the distal small bowel with a long efferent
limb pulled through an opening in the abdominal wall within the confines of the rectus abdominis muscle and terminating in a skin-level stoma. The rectus abdominis muscle was intended to act as a sphincter-type mechanism around the efferent limb to provide continence. Unfortunately, this design provided continence only in a small minority of patients. Kock’s initial, albeit unsuccessful, modification to overcome the problem of incontinence was to create an antiperistaltic efferent limb. It was not until 1973 that he described intussusception of the efferent limb to create the characteristic nipple valve that is the hallmark of most modern continent ileostomies. This modification was successful in providing continence to a majority of patients, but it also set the stage for the most common complication of the modern CI which is valve slippage.

FIGURE 47-1 A. Continent ileostomy patient. The stoma is typically located in the lowermost aspect of the right abdomen. B. Patient performing catheter intubation of the continent ileostomy.

During the past 30 years, a number of new methods have been developed in an attempt to reduce the rate of valve slippage, which has been reported in as many as 30% of CI patients. Kock attempted to address this problem through modifications of his technique that involved splitting and de-fatting of the valve mesentery, suture fixation and serosal scarring, partial rotation of the valve, and finally stapled fixation with the use of a fascial strip or Marlex mesh (Bard, Warwick, RI). Kock’s largest published series of 314 patients showed a steady reduction in valve complications and slippage with the evolution of his technique. Others have made similar modifications to the procedure, mostly aimed at fixation of the valve by chemical or physical means. Fibrosis between the two intussuscepted layers has been promoted by traumatizing the serosa of the efferent limb using an orthopedic rasp, deep diathermy scarring of the serosa, interposing synthetic mesh between the valve layers, and by chemical means with substances such as formalin, silver nitrate, talc, and even asbestos. Staple fixation of the valve, both to itself and the pouch sidewall, was first described by Fazio and Tjandra in 1992, and this method remains the author’s primary means of valve stabilization today. Although these modifications have undoubtedly reduced the incidence of valve slippage, they have also increased the risk of other complications such as valve or pouch fistulas.

More radical attempts at altering the basic design of the CI have also been undertaken in an attempt to decrease the risk of valve slippage. The most common of these is the Barnett continent ileostomy reservoir, or BCIR. The initial form of the BCIR was described by Spencer and Barnett in 1979 and relied on an isoperistaltic intussuscepted valve for continence. However, valve slippage continued to occur and the “living intestinal collar” that is the distinguishing feature of the modern BCIR was added in 1986 in an attempt to buttress the mesenteric side of the valve where slippage is felt to first develop.

Regardless of the technique of CI construction, valve slippage remains the “Achilles’ heel” of the operation. In an attempt to eliminate the problem of valve slippage altogether, Kaiser and Beart have developed a “valveless” CI. This design, known as the “T-pouch”, was initially described by Stein for urinary diversion after cystectomy. Although valve slippage is inherently avoided, analysis of the designers’ first 10 years of experience with the technique still found a reoperation rate of greater than 50%.

Current Technique

The CI operation can be divided into four stages: construction of the S-pouch, creation of the intussuscepted nipple valve, valve fixation, and siting of the pouch with stoma maturation. In the case of de novo CI creation, the entire pouch is constructed from the distal-most 60 cm of small bowel. In the first stage, an S-pouch is fashioned from three 12- to 15-cm limbs of ileum (Fig. 47-2). These limbs are first approximated with a serosal layer of interrupted or continuous 3-0 absorbable suture leaving an efferent limb of approximately 20 cm that will ultimately become the intussuscepted nipple valve. The exact length of the efferent limb is derived from doubling the desired length of the valve (6-7 cm) and adding the thickness of the abdominal wall through which the exit conduit will pass. An antimesenteric enterotomy is then created along the three limbs of the S-pouch and the back wall of the pouch is constructed with a running full-thickness 3-0 absorbable suture (Fig. 47-3). A 6- to 7-cm valve is then fashioned by intussuscepting the efferent limb into the pouch and fixing it to itself with two firings of a non-cutting 55-mm linear stapler placed along either edge of the valve mesentery (Fig. 47-4). Care must be taken to not include the mesentery in the staple lines because a hematoma or valve ischemia could result. If the efferent limb that will be used for valve construction has a bulky mesentery, it should be stripped of peritoneum and fat, taking care not to damage the underlying blood vessels. This “de-fatting” maneuver will make it easier to intussuscept the efferent limb to create the valve and promote fibrosis between the valve layers to inhibit slippage. The anterior wall of the S-pouch is then closed with either a running or interrupted 3-0 absorbable seromuscular suture(s). Suture closure of the anterior pouch wall is initiated at the apex of the pouch
and each stitch includes the mucosa and submucosa of the antimesenteric aspect of the valve. This is important for fixation of the valve to the pouch wall, a maneuver that helps to minimize the risk of valve slippage. When the suture line reaches the tip of the valve, a third firing of the non-cutting 55-mm linear stapler is applied along this suture line to further reinforce the suture fixation of the valve to the pouch wall (Fig. 47-5). Stitches are then transitioned to only include the pouch wall and the suture line is completed. “Fundoplication” stitches of a 3-0 nonabsorbable suture are then placed between the apex of the pouch and the exit conduit to further stabilize the valve. Pouch suture line integrity and continence of the valve are tested by intubating the pouch with a 30 French catheter,
filling the pouch to capacity with saline and air, and then withdrawing the catheter (Fig. 47-6). The pouch is then situated within the pelvis or lower abdominal cavity and the exit conduit is brought through the abdominal wall after creating a stoma aperture at the previously marked site. The apex of the pouch is then fixed to the underside of the abdominal wall with several 3-0 nonabsorbable sutures (Fig. 47-7), and the stoma is matured flush with the skin (Fig. 47-8). The stoma is again intubated with the catheter; the location of its tip is confirmed within the pouch (Fig. 47-9); and the catheter is secured to the skin to prevent it from becoming dislodged in the postoperative period (Fig. 47-10). The pelvis is drained, and the abdomen is then closed.

Only gold members can continue reading. Log In or Register to continue