Long‐term outcome and quality of life after continent ileostomy for ulcerative colitis: A systematic review

The continent ileostomy allows evacuation of an ileal reservoir at a time convenient to the patient. It is a surgical option for patients with ulcerative colitis (UC) when a restorative option is not suitable or has not succeeded and the patient does not want a conventional end ileostomy. Continent ileostomy types include the Kock pouch, Barnett continent intestinal reservoir and T‐pouch. All of the published evidence on the long‐term outcome and quality of life after continent ileostomy for UC was systematically reviewed.


INTRODUC TI ON
The aim of a continent ileostomy is to provide the patient control over the evacuation of an ileal reservoir at a time convenient to them. The continent ileostomy was first described by Kock in 1969 [1]. It comprises a small bowel pouch which leads to an intussusceptive valve that prevents flow of ileal contents (see Figure 1). It was constructed for patients after colectomy for ulcerative colitis (UC) or familial adenomatous polyposis in the era before restorative surgery. When Sir Alan Parks and Professor John Nicholls described an ileoanal pouch that restored continuity of the bowel in 1978 [2] the use of the continent ileostomy declined. However, the continent ileostomy still plays a role in those patients for whom an ileal pouch anal anastomosis (IPAA) is not suitable, or where an IPAA may have failed. Additionally, patients with a conventional ileostomy who would prefer not to have an external appliance may also opt for conversion to a continent ileostomy.
Although Kock described his pouch first, other types of continent ileostomy have been developed such as the Barnett continent intestinal reservoir (BCIR) [3] and the T-pouch [4]. The BCIR utilises a "living intestinal collar" to support the intussusceptive valve (see Figure 2). The T-pouch does not have an intussusceptive valve but instead places an outflow segment of small bowel in a serosa lined tunnel that leads from an intestinal pouch (see Figure 3). Most recently, a medical device, the transcutaneous implant evacuation solution (TIES), has been invented to perform the same function as a continent ileostomy [5]. The TIES utilises a titanium mesh device into which the patient's ileal tissue will biointegrate. The stoma is then closed with a cap that connects to the titanium device.
This systematic review will investigate the long-term outcome and quality of life of a continent ileostomy in patients with ulcerative colitis. This review will consider all types of continent ileostomy as well as novel devices which perform the same function.

ME THOD
The systematic review was performed in accordance with the Preferred Reporting in Systematic Reviews and Meta-analyses (PRISMA) guidelines and the Cochrane Handbook for Systematic Reviews of Interventions.
F I G U R E 1 Construction of a Kock Pouch. A U-shaped pouch is created from the distal small bowel. The efferent limb (dashed white arrow) of the distal small bowel is intussuscepted to create a nipple valve (solid white arrow) that provides continence. Staple fixation stabilises the nipple valve configuration [33] Search strategy MEDLINE and EMBASE (searched via Ovid) and Web of Science were searched in a systematic manner. Reference lists in articles that were analysed for full text review were also searched. The last search was conducted on 6 January 2021.
The two main concepts of continent ileostomy and ulcerative colitis were searched for. Terms within each concept were combined with the Boolean operator "OR" and then the concepts themselves were combined with the Boolean operator "AND". The concepts were searched both by subject headings and as keywords depending on the database. Only keywords were used to search the Web of Science database. See Appendix 1 for the search strategy used. No search limit was put on year of publication or on language published; however, only articles in English were screened. Two reviewers (MD and GW) screened the abstracts and discrepancies were resolved by consensus after discussion with a third reviewer (KP). Covidence systematic review software was used to perform the search, screen abstracts, review full-text articles and construct the PRISMA diagram (available at www.covid ence.org).

Inclusion criteria
All articles describing studies of long-term outcome and quality of life after continent ileostomy in patients with ulcerative colitis were included. Long-term outcome was defined as a surgical complication, reoperation and/or pouch/device excision at any point during the study follow-up period.

Exclusion criteria
Review articles, editorials, guidelines, letters, published abstracts and audits were all excluded. Small case series with less than five patients were excluded (this was waived for innovative surgical technologies to ensure novel devices were included in the review).
Articles published before 1990 were excluded because of the change in surgical techniques that have occurred over 30 years. Where the outcome of a single surgical cohort was reported multiple times after 1990, the most recent article was selected for data extraction and the older articles excluded.

Data analysis
A descriptive synthesis was used to analyse the data collected due to clinical heterogeneity. Risk of bias was assessed (by MD, verified by KP, see Table S1) with the Joanna Briggs Institute Cohort Studies Critical Appraisal Tool (available from https://synth esism anual.jbi.global) [6].

F I G U R E 2 Construction of a Barnett Continent Intestinal
Reservoir (BCIR). A U-shaped pouch is constructed from the distal small bowel (solid arrows). As with a Kock pouch, the efferent limb is intussuscepted to create a nipple valve and stapled to in place to stabilise the configuration. A "living intestinal collar" is wrapped around the efferent limb to provide a further continence measure (dashed arrow). A side to side, small bowel to pouch, anastomosis is created connecting the ileum to the BCIR (not illustrated) [20] F I G U R E 3 Construction of a T-pouch. An isolated segment of distal small bowel is placed in a serosa lined tunnel formed from the base of adjacent ileal segments (blue arrow). The valve is opened and tapered along a 30 Fr tube (blue catheter pictured). The adjacent bowel segments are opened and the two large flaps of ileal tissue are brought over the isolated segment to create the valve. The pouch is then closed (white arrow) [34] F I G U R E 4 PRISMA diagram.

Characteristics of included studies
A total of 1655 abstracts were returned following the search and 245 duplicates were removed. A total of 1410 abstracts were screened, and 1259 studies were found to be irrelevant. One hundred and fifty-one full text articles were reviewed and 19 were included in the final review for data extraction and qualitative analysis (see PRISMA diagram, Figure 4). All studies were retrospective cohort studies involving 1602 patients with UC (see Table 1).

Outcomes and complications
Berndtsson et al. [11,25] and Borjesson et al. [12] only reported patient characteristics and outcomes after Kock pouch and not the incidence of postoperative complications and are therefore not included in Table 3.

Most common indication for reoperation
Average time to reoperation (years)

Reoperation
Reoperation rates were high for all types of continent ileostomy.
The most common indication for reoperation after Kock [20].
The T-pouch does not utilise a nipple valve but instead places an isolated segment of bowel in a serosa-lined tunnel of two apposed limbs of small bowel [24]. However, the main complication requiring reoperation was also related to the valve mechanism and the valve would together [22]. The incontinence rate after T-pouch was 8.3% [24].

Leaks
There were four studies that reported leak rates after Kock pouch.
These leaks were from suture lines, staple lines and anastomoses.
Aytac et al. reported the highest rate of 9% [17] and the lowest rate of 2.3% was reported by Leijonmarck et al. [7]. A leak rate of 1.96% was found after BCIR by Mullen et al. [20]. A leak rate of 12.5% was found after T-pouch by Kaiser et al. [24]. No leaks were found after TIES device which does not utilise stapler technology or anastomoses in its technique [5].

Stomal stenosis and necrosis
Stomal stenosis after continent ileostomy was reported by six studies that investigated outcomes after Kock pouch. Rates of stenosis were noted to range from 2% [19] to 29% [18]. These studies reported that the stenotic segments could be dilated under local anaesthetic.
The highest rate of stenosis was found in the series by Denoya et al.
which looked at patients with functioning Kock pouches for at least 10 years suggesting stomal stenosis remains a possible complication in the long term. Similar rates of stomal stenosis were found for BCIR, 2.4%-7.8% [20,21], and for studies that looked at both BCIR and Kock pouches, 8.3%-10% [22,23]. A quarter of all patients who underwent T-pouch developed stomal stenosis [24]. Stomal stenosis was not found in those patients who had the TIES device [5].
Stomal necrosis was reported in all types of continent ileostomy.
A patient who had the TIES procedure underwent necrosis of the distal 2 cm and the small bowel retracted exposing the titanium mesh in the device [5]. Stomal necrosis occurred in 3.4%-4% of pa-   [28]. The average CGQL score was 0.77 with an average score of 8 for current quality of health, current quality of life and current level of energy [28].

Conversion from failed IPAA to continent ileostomy
See Table 5 for a summary of the studies that investigated outcome after conversion from IPAA to continent ileostomy. There were four studies that looked at the specific clinical scenario of converting a failed IPAA to a continent ileostomy [12,17,21,28]. Aytac  Overall continent ileostomy pouch survival rates were high after conversion from IPAA ranging from 79% [17] to 95% [28]. Nipple valve slippage was also the main surgical complication in this set of patients. Again, reoperation rates were high after conversion from IPAA and ranged from 33%-61%.
Some of the highest rates of fistula formation were also reported in these studies with 9.5% developing fistulae in Behrens et al series [21]. Conversion from IPAA to the TIES has not been undertaken.

Outcome Summary Comparison for Different Types of Continent Ileostomy
The outcome summary comparing the different types of continent ileostomy is outlined in Table 6.

DISCUSS ION
This review has found that the different types of continent ileostomy all have low postoperative mortality. However, they are all associated with a high reoperation rate which is primarily due to valve mechanism failure. Other significant complications from surgery for continent ileostomy include fistulae, stomal stenosis and stomal necrosis. Despite this the rates of pouch removal are relatively low.
Patients might have their quality of life greatly improved by a continent ileostomy in a number of clinical scenarios -as a primary procedure, after a failed IPAA, or conversion from a conventional end ileostomy -and many patients report high levels of satisfaction afterwards.
Valve mechanism failures occur for different reasons in intussusceptive and non-intussusceptive types of continent ileostomies.
In intussusceptive valves the fault is due to nipple valve slippage.
Generally, the rates of slippage have decreased with improvements in surgical technique, but rates remain stubbornly high. Techniques to prevent nipple valve slippage include removing the fat from the valve mesentery and staple fixation of the valve [30]. Other mechanical and chemical methods have been used to promote scarring of the intussuscepted segment and prevent slippage.
It was nipple valve slippage that spurred Spencer and Barnett to invent the BCIR with its living intestinal collar to buttress the intussusceptive valve and prevent this complication. But Kaiser et al.
note that their non-intussusceptive valve also has problems -specifically the outflow segment being too long or too short [24]. If it is too long the patient will have difficulty intubating the pouch and if it is too short it will lead to incontinence and leakage. They describe a learning curve to be able to predict the length of outflow segment needed. The general trend with continent ileostomy studies is that the learning curve reduces the complications and reoperation rate as time goes on. Second, the risk of fistulation is also higher in those who have had conversion from IPAA as noted above.
Published data on outcome after TIES is limited. There is no larger published series than the study of four patients included in this review [5]. There are concerns regarding the lack of biointegration of the small bowel into the titanium implant. It is known that a number of these devices have been explanted [32]. Other innovative technologies to replicate the continent ileostomy may be developed in the future.
A limitation of this review is the retrospective nature of all the studies identified. No prospective studies on continent ileostomies and how they compare to conventional end ileostomies has ever been TA B L E 6 Overall outcome comparison. performed. Furthermore, there may well be a significant survivorship bias to the quality of life data and outcome data available to the centres that perform these procedures. Some of the outcome data is not reported by a majority of studies, such as average time to reoperation (which is particularly important for this procedure). Interpretation of the BCIR data requires caution: the results are dominated by a single study of 510 patients, although this was a collaborative study covering 12 surgeons' practice at five different institutions across the United States. Likewise, data on the T-pouch is from a single publication and this limits the interpretation of the observed results.

Kock pouch BCIR T-pouch TIES
Although IPAA remains the gold standard restorative option after proctocolectomy for ulcerative colitis, good pouch function is not guaranteed, and pouch failure represents a devastating outcome for such patients. We expect the future may also include novel devices to improve the options of a continent ileostomy for those with UC who do not want or are not suitable for IPAA. This review highlights that continent ileostomy is still an option for various patients. However, careful counselling of the risks and the possible difficult journey ahead with revision surgery must be made clear to the patient.

CO N FLI C T O F I NTE R E S T
None of the authors have any relevant disclosures.

E TH I C S S TATEM ENT
Ethics approval was not sought nor required for this work.