Subtotal colectomy in ulcerative colitis—long term considerations for the rectal stump

Abstract

BACKGROUND

The initial operation of choice in many patients presenting as an emergency with ulcerative colitis is a subtotal colectomy with end ileostomy. A percentage of patients do not proceed to completion proctectomy with ileal pouch anal anastomosis.

AIM

To review the existing literature in relation to the significant long-term complic-ations associated with the rectal stump, to provide an overview of options for the surgical management of remnant rectum and anal canal and to form a consolidated guideline on endoscopic screening recommendations in this cohort.

METHODS

A systematic review was carried out in accordance with PRISMA guidelines for papers containing recommendations for endoscopy surveillance in rectal remnants in ulcerative colitis. A secondary narrative review was carried out exploring the medical and surgical management options for the retained rectum.

RESULTS

For rectal stump surveillance guidelines, 20% recommended an interval of 6 mo to a year, 50% recommended yearly surveillance 10% recommended 2 yearly surveillance and the remaining 30% recommended risk stratification of patients and different screening intervals based on this. All studies agreed surveillance should be carried out via endoscopy and biopsy. Increased vigilance is needed in endoscopy in these patients. Literature review revealed a number of options for surgical management of the remnant rectum.

CONCLUSION

The retained rectal stump needs to be surveyed endoscopically according to risk stratification. Great care must be taken to avoid rectal perforation and pelvic sepsis at time of endoscopy. If completion proctectomy is indicated the authors favour removal of the anal canal using an intersphincteric dissection technique.

Key Words: Stump; Ulcerative colitis; Surveillance; Endoscopy; Rectal; Screening

Core Tip: Rectal stumps require long term surveillance due to well documented risk of malignancy, the authors provide a summary of current guidance and recommendations for this. Patients may require completion proctectomy due to dysplasia, malignancy or persistent symptoms, options for completion proctectomy are explored. Endoscopic surveillance of the rectal stump poses certain challenges, potential complications and their management are explored.

INTRODUCTION

Ulcerative colitis is an idiopathic inflammatory condition in which there is relapsing and remitting inflammation involving predominantly the colon and rectum, generally limited to the superficial mucosal layer[1]. There is a rising incidence worldwide, with prevalence highest in Europe and North America[2-4]. While the need for surgical intervention in ulcerative colitis is decreasing with the advent of biologic therapies, a significant proportion of patients still progress to operative management[5-7]. In recent analysis, 10-year rates of colectomy were still found to be as high as 17% in some patient cohorts[8]. It is estimated that 12%-25% of ulcerative colitis patients will require hospitalisation due to severe exacerbation, and the risk of colectomy during these admissions is as high as 20%-30%[9]. A multidisciplinary approach involving gastroenterology, colorectal surgery, stoma therapy, dietetics and other ancillary services is important early in this process.

Total proctocolectomy remains a very safe single stage procedure and is the gold standard for elective management of ulcerative colitis. In patients abhorrent to a permanent ileostomy, creation of an ileal pouch anal anastomosis (IPAA) most often through a staged approach has good long-term outcomes. In episodes of acute severe ulcerative colitis requiring urgent or emergent surgical input a large proportion of patients are initially managed via a subtotal colectomy (STC) and end ileostomy with associated retention of a rectal stump[10-12]. In the acute setting STC offers a safe procedure for patients in the setting of an active inflammatory state, while also preserving the option of future return of intestinal continuity via an IPAA or, less commonly, an ileorectal anastomosis (IRA)[2,11,13]. In addition to an emergency presentation the rectum may also be left in situ due to patient factors including concerns regarding the associated risk of pelvic nerve damage, infertility and sexual function with proctectomy, this is an important consideration in females of child bearing age[14,15].

Of interest, studies from Sweden and the United Kingdom have shown that in some cohorts, only one third to half of patients undergo reconstruction post colectomy[16,17]. This may be due to a number of factors ranging from satisfaction with quality of life post STC, to being unfit for further operative intervention, as well as lack of patient awareness regarding reconstructive options. Unfortunately the retained rectal stump is not without its complications. Patients can suffer from persistent symptoms including ongoing mucous or bloody discharge, low grade fevers and feelings of rectal discomfort or urgency due to ongoing ulcerative proctitis or diversion proctitis[18,19]. Rates of diversion proctitis may be as high as 90% in varying degrees of severity[20]. Symptoms may start as soon as a few months after faecal diversion and in many patients this ongoing discharge has a deleterious effect on their quality of life. In this cohort medical or surgical intervention may be required[21].

Long term, there is also a well-documented risk of malignancy in the retained rectum[22]. A recent meta-analysis by Derikx et al[22] has shown this risk to be as high as 3% in those with a retained rectal stump, vs 1% in those who undergo resection and reconstruction with an ileoanal pouch. For primary IPAA surgery the majority of colorectal surgeons have now moved to a stapled technique when anastomosing the ileal pouch to the anal canal. This is technically easier and associated with improved functional outcome[23]. However, this involves leaving 1-2 cm of rectum in-situ which can give rise to “cuffitis” and long-term potential for malignant changes[24,25]. Thus the frequency of endoscopic surveillance of the retained rectum or ileal pouch to observe for dysplastic changes or malignancy must be considered.

Thus, for the reasons outlined above a considerable number of patients with a retained rectal stump who do not undergo reconstruction with IPAA or IRA will progress to require completion proctectomy. Traditionally, this would have been carried out via an open abdominoperineal approach. This however has been associated with significant morbidity (up to 41%) and mortality (6%), as well as long term perineal wound healing issues[22,23]. Advances in surgical practice in the form of laparoscopic[24] trans anal[25], robotic[26] and endoscopic[27] techniques have reduced this morbidity. Whether the anal canal is left-in situ or removed at time of proctectomy is an area of debate.

In this article, we aim to conduct a review of the published literature concerning the long-term management and surveillance of rectal stumps. As part of this, we hope to provide consolidated recommendations on surveillance guidelines for patients with rectal remnants following colectomy in ulcerative colitis. We also seek to provide a brief overview of medical management options of diversion proctitis, as well as a compendium of surgical methods of completion proctectomy in patients in whom medical management ultimately fails. As far as the authors are aware, there is currently no concise collection of this data in relation to the rectal stump available in the literature.

MATERIALS AND METHODS

A systematic review was conducted according to the PRISMA guidelines[26]. Literature review of all papers providing recommendations for long term surveillance of patients with retained rectum following colectomy for ulcerative colitis with a retained rectum. The search was performed using multiple databases including MEDLINE (PubMed), Ovid and Cochrane databases including the timeframe between January 1980 and March 2020. Search protocol was cross checked in PROSPERO but no existing ongoing studies were found[27]. The following search criteria were used: Keywords (rectal) AND (stump) OR (remnant) OR (retained) AND (surveillance) AND (ulcerative colitis). Articles were included if they discussed recommendations on surveillance in patients with retained rectal tissue following colectomy for ulcerative colitis or indeterminate colitis. Papers were excluded if they dealt exclusively with Crohn’s colitis, paediatric populations and if recommendations made were specific to IPAA. Papers were also excluded if they made a general recommendation for surveillance, but gave no specific recommendation regarding timeframe or method of same.

A search and review of paper titles and abstracts was performed by the lead author, and papers were filtered to a list of those requiring full text evaluation. The references section of the papers undergoing evaluation were also examined for further relevant articles. The search was limited to English language studies. Data was extracted from each of the papers undergoing full text review including authors, year of publication, surveillance interval recommended and surveillance method. Descriptive statistics were used to analyse the data. Given the heterogeneity of the study populations and design, it was not possible to perform a meta-analysis. Results have been reported in line with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and AMSTAR (Assessing the methodological quality of systematic reviews) Guidelines.

Following this, a secondary search was carried out using the keywords (proctectomy), (rectal), (stump), (completion), (surgery), (ulcerative colitis) and (management). Again a search and review of paper titles and abstracts was performed by the lead author, and papers were filtered to a list of those requiring full text evaluation. The references section of the papers undergoing evaluation were also examined for further relevant articles. All articles reporting methods of completion rectal stump proctectomy were analysed.

RESULTS

For the systematic review, initial search yielded 117 papers. After removal of duplicates and screening of abstracts, 37 papers were selected for further screening along with 7 additional papers found through review of references in the papers retrieved. Following analysis of each paper, 10 papers were included in the study. The reasons for exclusion are as per the criteria above and are outlined in the PRISMA flow diagram in Figure 1. Data was collected as above and the relevant papers and findings are summarised in Table 1[28-37]. Of the 10 papers, 2 (20%) recommended an interval of 6 mo to a year, 5 recommended yearly surveillance (50%), 1 (10%) recommended 2 yearly surveillance and the remaining 3 (30%) recommended risk stratification of patients and different screening intervals based on this. All studies agreed that follow up should be with endoscopy and biopsy, but there was no distinction made in terms of number or placement of biopsies. It is also worth noting that 9 further studies did highlight the importance of surveillance in this patients cohort, but made no specific recommendations regarding timing and/or method.

Figure 1 PRISMA flow diagram indication exclusion for screened papers.

Table 1 Summary of papers.

Ref.	Year	Surveillance interval	Surveillance method
Pastore et al[28]	1997	Yearly	Endoscopy with biopsy
Khubchandani et al[29]	1994	6 monthly to yearly	Endoscopy with biopsy
Petersen et al[30]	2008	2 yearly	Endoscopy with biopsy
da Luz Moreira et al[31]	2010	Yearly	Endoscopy with biopsy
Lutgens et al[32]	2012	Rectal stump, PSC, and disease duration > 8 yr: 1 to 2 yearly	Endoscopy with biopsy
Munie et al[15]	2013	Yearly	Endoscopy with biopsy
Andersson et al[33]	2014	Yearly	Endoscopy with biopsy
Scoglio et al[34]	2014	6 mo to yearly	Endoscopy with biopsy
Myrelid et al[36]	2015	< 20 yr of age + < 10 yr duration: yearly; early onset of the disease and > 10 yr duration: twice yearly; all others: yearly	Endoscopy with biopsy
Abdalla et al[35]	2017	Yearly	Endoscopy with biopsy
Derikx et al[37]	2018	High risk: yearly; Hx of PSC: 2-3 yearly; low risk: 5 yearly	Endoscopy with biopsy

PSC: Primary sclerosing cholangitis.

For the narrative aspect of the review, 253 articles were screened. Articles describing operative techniques for completion proctectomy in ulcerative colitis were assessed in full. Overall, a number of techniques were described in the literature including abdominoperineal resection, either open, laparoscopic or laparoscopic assisted, transanal approach either open or endoscopic, intersphincteric dissection and robotic proctectomy. These techniques will be explored in further detail within the discussion.

DISCUSSION

The long term risk of colorectal cancer (CRC) in patients with rectal remnants following colectomy in inflammatory bowel disease is well described[14,22]. In addition to the general risk of CRC associated with ulcerative colitis, patients post colectomy have the added factor of an “out of circuit,” rectum. This may mean that patients are less likely to be aware of signs of malignancy including changes in bowel habit or bleeding[15]. A high proportion of these patients may also suffer from diversion proctitis, which can present with discharge and bleeding and may mask signs of a more sinister pathology[20]. As a result these patients represent a more vulnerable population than those with intestinal continuity and pose a particular challenge for surveillance. In contrast, despite restored continuity, recent systematic review and meta-analysis by Derikx et al[22] also found the risk to be higher in those patients with IRA vs patients undergoing IPAA (2%-2.5% vs 5%). This is likely due to the higher proportion of retained rectal tissue in IRA and stump patients.

However, while recommendations exist for long term CRC surveillance in patients with inflammatory bowel disease[38-41], there is no specific guidance for patients with a retained rectal stump or IRA. The only guidelines to provide any form of recommendation for post-colectomy patients are the British Society of Gastroen-terology, however they do not distinguish between different operative interventions and the presence or absence of rectal remnants[42]. These guidelines stratify patients into low and high risk dependent on factors such as previous CRC history, dysplasia or history of primary sclerosing cholangitis (PSC). High risk patients are recommended for yearly screening and lower risk 5 yearly.

In in our literature review, many of the studies recommend between 6 moly and 2 yearly screening for patients with significant amounts of retained rectal tissue. More recent studies however have shifted focus toward risk stratification of patients based on disease duration and activity, history of previous CRC or dysplasia and disease related factors such as PSC. In 2015 Myrelid et al[36] advised guidelines based on patient disease duration. Based on analysis of the BSG guidelines, as well as guidelines in place for patients with an intact colon, Derikx et al[37] propose their own guidelines based on risk stratification into low intermediate and high risk groups. High risk cohorts with a history of CRC or dysplasia were advised to undergo yearly screening, intermediate cohorts with a history of PSC were advised 2-3 yearly screening and low risk patients 5 yearly[37]. In light of increased current knowledge regarding CRC risk in this patient cohort, this risk stratification strategy seems to be an appropriate method for providing guidance on surveillance. One major advantage of stratifying patients based on risk, is that it may reduce the number of follow up endoscopies needed overall in this patient group. This is an important consideration given that rectal stump patients are at a theoretical risk of stump blowout which each endoscopic procedure undertaken[43]. There is also a well-documented issue of overall compliance with endoscopic screening in this patient cohort[15,43,44] in particular with yearly follow up, and increasing the interval between scopes may improve compliance.

Rectal stump perforation

As introduced above, when discussing endoscopic surveillance it is also pertinent to discuss the risk of stump blowout[10]. Perforation or dehiscence of the rectal stump may occur following the initial colectomy or secondary to endoscopic surveillance[45]. Over vigorous endoscopy may result in perforation due to an increase in intraluminal pressure or direct scope trauma. Thus all endoscopists must be aware of this potential risk when performing surveillance of the rectal stump, particularly given the coexisting inflammation that is often present, and the endoscopist must be aware of the amount of rectum left in situ.

Traditionally the rectal stump was placed above the lower abdominal wall fascia. In this case, ensuing perforation/dehiscence which tends to occur at the apex of the staple line would result in a superficial wound infection. However, with the advent of minimally invasive surgery the trend now is to staple the remnant rectum above the peritoneal reflection intraperitoneally (Figure 2). However in this scenario any breakdown of the staple line may result in significant intra-abdominal or pelvic sepsis with systemic manifestations, carrying a risk of morbidity and mortality, prolonged hospital stay and the potential need for an emergent surgery. Severe non-resolving pain post endoscopy should alert the operator to the potential for perforation. An erect chest X-ray or computed tomography (CT) scan will show free air.

Figure 2 Options for placement of the rectal stump. A: Placement outside of the abdominal wall fascia; B: Intraperitoneal placement.

Clinically these patients may be difficult to manage. In the authors' experience this patient cohort may get quite sick despite having an end ileostomy in place. We hypothesize that the remnant rectum has a significant bacterial load which contaminates the sterile pelvic field post perforation. If the patient has ongoing infection that is not responding to conservative management then a CT abdomen and pelvis is indicated. If scanning shows a well-defined pelvic abscess then this may be drained percutaneously under radiological guidance. A tubogram will often show communication with the rectum. If radiological drainage is not feasible then a laparotomy may be required. If the patient has a long rectal stump then a staple line may be placed distal to the perforation site to healthy tissue. If there is only a short rectal stump that cannot be closed then an irrigation system may be used where a tube is placed proximally and via the anal canal and irrigation continued for 5-7 d[46].

Medical management of diversion proctitis

Following colectomy, a number of patients may develop diversion proctitis or recurrent ulcerative proctitis in the retained rectum. In fact, endoscopic studies have shown that nearly all patients will go on to develop some level of inflammation, though less than 50% of these will be symptomatic[47-50]. For these patients, symptoms may include cramping abdominal pain, mucous or bloody discharge and tenesmus or anorectal pain[51]. In patients requiring treatment, nonsurgical management options include the use of short chain fatty acids (SCFA), topical 5-ASAs (American Society of Anesthesiologists), and topical glucocorticoids delivered via enema[21]. Despite early evidence for the efficacy of SCFA[52], more recent studies have doubted their efficacy[53,54], though butyrate enemas in particular may have an impact on tissue recovery[55] they are not widely used or available in practice. Topical 5-ASAs and steroids have also exhibited varying efficacy[56-60]. More experimental methods of management include fibre irrigation[61], endoscopic dextrose spray[62], leukocytapher-esis[63] and faecal transplantation[64], however the evidence for each of these methods is based on limited studies and case reports. If severe proctitis in a diverted rectum was refractory to 5ASA and topical steroids, consideration could be given to the addition of immunomodulators or biologics. However, the efficacy of medical management is highly variable and a proportion of patients will progress to requiring surgical input[21].

Options for completion proctectomy

The authors generally favour removal of the anal canal at time of proctectomy, however it must be individualized to patient factors. Exceptions may include a patient with high ASA grades in whom reduced operative time is important, patients with associated fistulizing disease, hidradenitis suppurativa or other co-morbidities in which an anusectomy may be considered at a later stage when patient factors are more favourable. An early multidisciplinary approach in patients potentially needing surgery for ulcerative colitis is critical. In older patients, those with comorbidities or those who have a good understanding of the options who are not interested in reconstruction (IPAA) then the operating surgeon may consider a total proctocolectomy with end ileostomy as a single stage procedure, and avoid the need for a second intervention at a later stage. If clinically appropriate and the surgeon has the skill set, the entire procedure may be performed using a minimally invasive technique with the specimen removed via the perineum. However, in the emergency setting the majority of patients undergo a STC with end ileostomy. In the cohort of patients who elect not to undergo reconstruction with IPAA or IRA, a proportion (7%-14%) will eventually require completion proctectomy either as a result of ongoing proctitis or due to the development of dysplasia or CRC[65-67]. In these patients there are a number of options available in terms of surgical excision of the remaining rectum. While there have been no large scale, randomised trials encompassing all of these methods, there is adequate individual evidence supporting each one, and which method is employed often comes down to the operating surgeons training and preference.

The traditional gold standard approach would have consisted of an open abdominoperineal approach in the lithotomy position, with the perineal portion completed from below. The small bowel must be extruded from the pelvic field before starting proctectomy. Most surgeons follow the TME planes. Intramesorectal dissection has not been shown to reduce the nerve injury rate and is associated with increased blood loss. However this procedure is known to be associated with significant morbidity (41%) and mortality (up to 6%) of its own, as well as issues with perineal wound healing[65,68,69]. Further surgical advancement in this area has led to the evolution of hybrid approaches involving the use of laparoscopic, hand assisted laparoscopic/laparoscopic assisted surgery and robotic procedures for the abdominal component of dissection where required. Though recent studies in the use of laparoscopic surgery in ulcerative colitis for restorative proctocolectomy have found no major evidence for significant benefit of laparoscopic techniques[70,71], laparoscopic studies for proctectomy in oncology patients have indicated potential for shorter inpatient stays as well as faster return of bowel function[72] and there is also encouraging evidence for laparoscopic proctocolectomy and end ileostomy in ulcerative colitis patients[73,74]. There is also emerging evidence on the use of robotic technology, although this is still under investigation and there are issues surrounding accessibility for many institutions[73,75,76]. Irrespective of technique used completion proctectomy should be undertaken by a colorectal surgeon with considerable expertise given the associated morbidity.

An exclusively perineal approach can also be used, involving intersphincteric dissection from either a lithotomy or a prone, jack knife position. While not a commonly described approach for completion proctectomy in this patient cohort, in studies investigating the use of a jack knife approach for low rectal cancer it was associated with a significant reduction in operative complications including perineal infection and wound dehiscence as well as pelvic sepsis when compared to traditional lithotomy position[77,78]. Unlike an abdominoperineal resection for cancer there is no need for wide margins/extralevator dissection. We perform our dissection in the intersphincteric plane removing the internal sphincter complex and retaining the external sphincter complex. This reduces the size of the perineal defect reducing the potential for perineal wound breakdown and perineal herniation. Intersphincteric dissection may also have the benefit of reduced risk of impact on sexual function, as well as improved perineal healing. One of the draw-backs of this technique however is that it is difficult to employ in patients with longer rectal stumps[65], and may be technically more difficult in patients with significant adhesions[68,79,80].

Other, newer approaches are still being explored. This includes options such as transanal endoscopic microsurgery (TEMS)[81]. This is a relatively new technique for completion proctectomy, which may be considered in patients with longer rectal stumps which are not amenable to intersphincteric dissection. TEMS was initially described in the 80’s and involves the use of 10-20 cm proctoscopes that include a camera, suction and insufflation as well as ports for dissecting instruments. Generally it has been shown to be associated with less morbidity and shorter inpatient stays than open techniques[65]. In a small case series, Liyanage et al[81] describe the use of the technique in a cohort of 12 patients, nine of whom were undergoing the procedure due to IBD related complications. They found the technique to be particularly useful in those patients presenting with longer stumps and a potentially hostile abdomen, and overall found it to be both safe and effective. It is worth noting however that this is a highly specialised technique which requires extensive and specific training.

CONCLUSION

In conclusion, there is substantial evidence that patients with retained rectal tissue following STC have ongoing symptoms that interfere with their quality of life. In addition they require long term surveillance for dysplasia and CRC. While there are no clear guidelines produced by any governing body to guide surveillance intervals, stratification of patients into risk categories and basing intervals on this may be an appropriate solution for guiding long term follow up in this patient cohort. Great care must be taken when performing endoscopic surveillance and biopsies to avoid rectal stump perforation and ensuing pelvic sepsis. Patients with proctitis whose symptoms fail to respond to medical management may require completion proctectomy. A number of options are available for this, and again no strong evidence exists for one method over another. Similar to screening guidelines, in an era in which more of a focus is being placed on patient centred care and biopsychosocial models of health, the operating surgeon may have to decide on a case by case basis which intervention may be most appropriate, taking into account the patients co-morbidities, disease status, functional status and other factors such as age and lifestyle. When completion proctectomy is indicated the authors favour removal of the anal canal using intersphincteric dissection to reduce the potential for perineal wound breakdown or herniation.

ARTICLE HIGHLIGHTS

Research background

Retained rectal stumps in ulcerative colitis carry long term risks including malignancy and recurrent disease. No clear body of literature exists on their long term management and surveillance.

Research motivation

To explore the current literature and provide a concise overview of the current evidence, as well as recommendations.

Research objectives

To provide an overview of options for the surgical management of remnant rectum and anal canal.

Research methods

Systematic and narrative review of the literature.

Research results

All studies agreed surveillance should be carried out via endoscopy and biopsy. Increased vigilance is needed in endoscopy in these patients. Literature review revealed a number of options for surgical management of the remnant rectum.

Research conclusions

Surveillance is necessary and should be risk stratified. The Authors favor intersphincteric dissection for removal of the rectal stump.

Research perspectives

This is an important issue which requires ongoing research.

ACKNOWLEDGEMENTS

The authors would like to acknowledge Hennessy É for her contribution in the form of the illustrations in Figure 2.