Anastomotic Considerations in Crohn's Disease

Abstract

A number of factors should be considered when performing an intestinal anastomosis in the setting of surgery for Crohn's disease. Preoperative risk factors, such as malnutrition, abdominal sepsis, and immunosuppressive medications, may increase the risk of postoperative anastomotic complications and alter surgical decision-making. The anatomical configuration and technique of constructing the anastomosis may have an impact on postoperative function and risk of recurrence, particularly in the setting of ileocolic resection, where the Kono-S anastomosis has gained popularity in recent years. There may be circumstances in which it may be more appropriate to perform an ostomy either without an anastomosis or to temporarily divert an anastomosis when the risk of anastomotic complications is felt to be high. In the setting of total abdominal colectomy or proctocolectomy for Crohn's colitis, restorative procedures may appropriate in lieu of a permanent stoma in certain scenarios.

Crohn's disease is a chronic, granulomatous inflammatory bowel disease that can affect any portion of the gastrointestinal tract from mouth to anus, as well as extraintestinal organs. It typically presents with either structuring or transmural/penetrating disease of the gastrointestinal tract, resulting in a spectrum of symptoms. The incidence and prevalence of Crohn's disease are highest in western industrialized countries, with a peak incidence in the second and third decades of life. It is characterized by an intermittent and relapsing course, and the negative impact on the quality of life of those afflicted can be profound. Medical therapy is typically employed as first line treatment, with surgery reserved for failed medical treatment or other complications of disease, such as perforation, obstruction, or fistulae. Medical therapy and our understanding of the pathophysiology of Crohn's disease have made significant advances in recent decades, and the advent of immunomodulators and biologic therapies have reduced the need for surgery based on time from diagnosis. In a cohort study of patients diagnosed with Crohn's disease between 1955 and 1989, 73% required surgery: 44% at 1 year, 61% at 5 years, and 71% at 10 years from diagnosis. ¹ However, a more recent cohort study of patients in the era of biologics, diagnosed from 2003 to 2004 and followed out to 2011, showed that only 29% required surgery: 14.6% at 1 year, 24.6% at 5 years, and 28.5% at 7 years from diagnosis. ²

While the incidence of surgery for Crohn's disease seems to be gradually decreasing, surgery continues to play an important role in its management. More than ever, a multidisciplinary approach, as well as specialized colorectal surgeons with a special interest in inflammatory bowel disease, have been used to better optimize the planning and timing of surgery. While surgery does not cure Crohn's disease, proper surgical intervention is key in achieving goals such as relief from symptomatic complications, preservation of small bowel length, improved quality of life, and minimization of treatment interruptions to reduce the risk of postoperative recurrence. ³ A further goal of surgical therapy is avoidance of a permanent stoma and maintenance of intestinal continuity when appropriate – central to this is the proper construction of an intestinal anastomosis. This article will discuss the key considerations when performing an intestinal anastomosis in the setting of Crohn's disease, focusing on minimizing the risk of postoperative complications and disease recurrence, as well as optimizing postoperative functional outcomes.

Preoperative Risk Factors for Anastomotic Complications

Nutritional Status

By the time most patients with Crohn's disease require surgery, there is often an element of malnutrition present, which can significantly increase the risk of adverse clinical outcomes. A recent study from China found that 59% of screened patients with inflammatory bowel disease were felt to be at risk for malnutrition. ⁴ While exclusive enteral nutrition has been shown to induce remission in the pediatric Crohn's population, ⁵ it is not as effective in inducing remission in the adult population, though there may be some benefit in terms of maintaining remission in patients with clinically quiescent disease. ⁶

Most studies in the adult literature have sought to evaluate nutritional support for preoperative optimization in Crohn's disease. While there is no gold standard for nutritional assessment in Crohn's disease, body mass index (BMI) and unintentional weight loss are generally agreed upon as acceptable nutritional measures. ^{7 8} Hypoalbuminemia has typically been associated with poor nutritional status, though during an acute phase response, such as a Crohn's disease flare, albumin levels will typically decrease. For this reason, hypoalbuminemia could be an indicator of the disease state rather than nutritional status alone. A number of studies have demonstrated conflicting results in the context of using preoperative serum albumin levels to predict postoperative anastomotic complications. ^{9 10 11 12 13}

A review published in 2019 reported that intensive preoperative enteral or parenteral nutritional support provided significant benefit for patients who Crohn's disease suffering from serious nutritional deficits, such as weight loss >10% in the last 3 to 6 months, BMI <18.5 kg/m ² , or albumin levels <30 g/L, reducing the risk of surgical site infections and other postoperative septic complications such as anastomotic leak. ¹⁴ A personalized prehabilitation program for high-risk patients with Crohn's disease requiring ileocolic resection, including enteral or parenteral nutrition when appropriate, has also been shown to reduce the rate of both anastomotic complications and re-operation. ¹⁵ Given that many Crohn's disease patients undergoing surgery are also being treated with immunosuppressive and biologic therapies, it may be hard to parse out the specific effects of malnutrition itself. However, it has been shown that malnutrition alone is an independent risk factor for postoperative morbidity and mortality irrespective of these medications. ¹⁶

The prescribed duration of preoperative nutritional support varies, depending on a number of factors, such as the urgency of the operation, the degree of preoperative malnutrition, the suitability of the gastrointestinal tract for enteral administration, and how the patient responds to nutritional support. Unfortunately, the protein-losing enteropathy associated with active Crohn's disease is often too profound to be overcome by aggressive nutritional support. If the patient's nutritional status cannot be properly optimized prior to surgery, an intraoperative decision must be made as to whether to create an end ostomy, perform a primary anastomosis, or potentially divert a new anastomosis. When in doubt, a conservative approach is often the most judicious approach.

Abdominal Sepsis

The presence of intraabdominal abscess or fistula at the time of surgery can have a significant impact on the risk of postoperative anastomotic complications. One retrospective study of 173 primary anastomoses for Crohn's disease found that the presence of a preoperative abscess was an independent predictor of anastomotic-related complications (OR 3.4, 95% confidence interval [CI] 1.2–9.8). ¹⁷ In this study, the risk of an anastomotic complications was 14% in the presence of an intrabdominal abscess; this increased to 40% when patients were also undergoing active treatment with corticosteroids. Alves et al reported that in patients undergoing elective ileocecal resection with primary anastomosis for Crohn's disease, an intraabdominal abscess noted at the time of surgery was an independent risk factor for postoperative intraabdominal septic complications (OR, 7.47; 95% CI 1.5–37.69). ¹⁸ Another study from St. Mark's Hospital found that the presence of intraabdominal sepsis at the time of ileocolic resection and primary anastomosis for Crohn's disease was associated with postoperative intraabdominal septic complications on both univariate ( p  = 0.005) and multivariate analysis (hazards ratio [HR] 8.6, 95% CI 1.2–60.1). ¹⁹ However, other studies that included abscesses drained preoperatively have reported no association between preoperative abscess and postoperative intraabdominal septic complications. ^{20 21} A number of cofactors should be taken into account when deciding whether or not to perform primary anastomosis when an intraabdominal abscess is present, including nutritional status, whether or not the patient is receiving corticosteroids or biologic therapy, the extent of intraabdominal contamination, the condition of the bowel to be anastomosed, and the surgeon's experience.

Medications – Corticosteroids, Immunomodulators, and Biologics

In the current era, a large percentage of patients with Crohn's disease undergoing surgery are being actively treated with corticosteroids at the time of surgery. While the overall deleterious impact of corticosteroids on wound healing (and by proxy anastomotic healing) is well-established, the exact dose and duration of therapy at which these agents are likely to have a negative impact are less well-defined. Unjustified concern over postoperative adrenal insufficiency and adrenal crisis has led to overuse of “stress-dose” steroids in the immediate postoperative period, a practice not supported by current literature. Given that a number of studies have implicated perioperative steroid use as an independent risk factor for postoperative intraabdominal septic complications, ^{9 11 18 21 22 23 24} reducing or eliminating steroids in the perioperative period seems justified. A meta-analysis published by Huang reported that preoperative use of steroids was clearly a risk factor for postoperative intraabdominal septic complications in patients undergoing abdominal surgery for Crohn's disease (OR: 1.99; 95% CI 1.54–2.57). ²⁵ Another retrospective analysis from Italy of 307 consecutive patients undergoing primary ileocolic resection for Crohn's disease found that preoperative use of steroids was an independent risk factor for major complications (OR 5.45, p <0.001), and preoperative treatment with steroids (OR 6.59, p  = 0.003) was independently associated with anastomotic leak. ²⁶ The exact dose at which one should consider a diverting stoma or avoidance of an anastomosis altogether is not well established, and likely should not be used as a sole determinant in making this decision, given the host of other patient factors that likely also affect the risk of anastomotic leak.

Immunomodulators, including 6-mercaptopurine, azathioprine, and methotrexate, are often used either as monotherapy or in combination with corticosteroids or biologics in the medical management of patients with Crohn's disease. While there have been very few studies looking specifically at their potentially deleterious effects on postoperative infectious complications and anastomotic leak in the setting of surgery for Crohn's disease, this risk is generally felt to be very low. A retrospective study by Aberra reported that 6-mercaptopurine or azathioprine alone and the addition of 6-mercaptopurine or azathioprine for patients receiving corticosteroids were not found to significantly increase the risk of postoperative infectious complications. ²⁷ Similarly, there is no existing literature to support an increased risk of complications with methotrexate. Generally, these medications are not held for a prescribed period of time prior to surgery and can be resumed relatively soon postoperatively in the absence of any obvious immediate infectious postoperative complications.

Biologic therapy for Crohn's disease includes anti-TNF-α agents (infliximab, adalimumab, certolizumab pegol, and golimumab), anti-integrins (natalizumab and vedolizumab), and anti-interleukins (ustekinumab and risankizumab). The influence of these agents on postoperative complications and surgical timing has been an area of interest, though it is a difficult topic to study given the heterogeneity of the patient populations and treatment modalities.

The medical management of Crohn's disease changed dramatically with the approval of infliximab in 1998. TNF-α is a key cytokine in the pathogenesis of Crohn's disease, and the available anti-TNF-α agents have proven to be some of the most successful treatments available for Crohn's disease. While prior studies have shown conflicting results, more recent studies have suggested that perioperative treatment with infliximab and other anti-TNF-α agents in patients with Crohn's disease are generally not felt to be associated with an increased rate of postoperative complications, including anastomotic leak. ^{28 29 30 31 32 33} The recently published multicenter prospective PUCCINI study found that there was no increase in overall infection (OR, 1.050; 95% CI, 0.716–1.535) or surgical site infection (OR, 1.249; 95% CI, 0.793–1.960) in patients with inflammatory bowel disease undergoing intraabdominal surgery with preoperative anti-TNF-α exposure, compared with those without, though they did not look specifically at anastomotic leak rates. ³⁴

Newer immunosuppressive agents target different pathways in the inflammatory process. Vedolizumab is a gut-selective antibody that selectively targets the heterodimer α4β7 to VCAM-1. Given its gut-specificity, the systemic immunosuppressive effects seen with anti-TNF-α agents are not present. There have been three robust meta-analyses evaluating the risk of postoperative complications in patients with inflammatory bowel disease treated preoperatively with vedolizumab. Law reported no significant risk of both postoperative infectious complications and overall complications, when comparing treatment with vedolizumab with both no biologic treatment and treatment with anti-TNF agents. ³⁵ Similarly, Yung found no difference in the risk of infectious complications, surgical site infection, need for reoperation, or major surgical complications in patients exposed to vedolizumab compared with both those receiving anti-TNF-α therapy and those with no biologic exposure. ³⁶ In another metanalysis, Moovsi found that preoperative treatment with vedolizumab in patients with inflammatory bowel disease undergoing abdominal surgery was not associated with an increase in overall or infectious postoperative complications compared with both preoperative anti-TNF-α treatment and no preoperative biologic treatment. ³⁷

Natalizumab is a recombinant monoclonal antibody that binds to the α4-subunit of α4β1 and α4β7 integrins, expressed on the surface of all leukocytes except neutrophils, and inhibits the α4-mediated adhesion of leukocytes to their counter-receptors. There is little published literature regarding the risk of postoperative complications in patients receiving natalizumab.

Ustekinumab is a fully human monoclonal antibody that functions by targeting the p40 subunit of IL-12 and IL-23. One multicenter retrospective study comparing postoperative infectious complications in patients with Crohn's disease who received ustekinumab within 12 weeks prior to surgery with patients who received anti-TNF agents prior to surgery found no significant differences in postoperative complications or readmissions, though ustekinumab-treated patients had a higher rate of return to the operating room (16 vs. 4%, p <0.001), most of which were due to anastomotic leak or surgical site infection. ³⁸ Another case-matched analysis comparing patients with Crohn's disease who received preoperative ustekinumab with those who received preoperative vedolizumab found that vedolizumab-treated patients had a higher postoperative complication rate ( p  = 0.009) and ileus rate ( p  = 0.015) on univariate analysis, though after case matching, logistic regression models demonstrated no significant differences. ³⁹ Risankizumab is another interleukin antagonist that targets IL-23. Its impact on the risk of postoperative complications and anastomotic leak has not been studied.

The most recent Clinical Practice Guidelines for the Surgical Management of Crohn's Disease published by the American Society of Colon and Rectal Surgeons state that whether or not preoperative exposure to monoclonal antibody therapy influences outcomes remains controversial, but delaying surgical intervention based on monoclonal antibody therapy alone is not typically recommended. ⁴⁰ However, this is a weak recommendation based on low-quality evidence, and the guidelines do go on to state that, when possible, if a patient is on a dosing regimen of every 8 weeks, the optimal time to perform surgery may be approximately 4 weeks after the last monoclonal antibody dose to allow for a washout period of one half-life.

Anastomotic Technique

There has been much controversy surrounding the optimal anastomotic technique when performing surgery for ileocolic Crohn's disease. Numerous variations – end-to-end versus side-to-side versus end-to-side, handsewn versus stapled, isoperistaltic versus anti-peristaltic – have been studied in relation to their impact on both recurrence rates and functional outcomes, without a clear consensus as to the optimal approach.

Given that there is a paucity of prospective randomized data regarding optimal anastomotic technique and configuration, and that the majority of published literature consists of retrospective studies and case series, a number of meta-analyses examining this topic have been published. One meta-analysis, published in 2007, evaluated 661 patients in eight studies undergoing a total of 712 anastomosis, and compared end-to-end anastomoses with other anastomotic configurations, which included stapled side-to-side anastomoses. ⁴¹ No differences were seen between the groups in terms of overall complications, anastomotic recurrence, or surgical anastomotic recurrence. When side-to-side anastomoses were compared with only end-to-end anastomoses, a reduced leak rate (OR, 4.37; p  = 0.02) and a reduction in overall postoperative complications (OR, 2.64; p <0.001) were seen in the side-to-side group. However, there was no significant difference between the groups in terms of perianastomotic recurrence and the need for reoperation due to perianastomotic recurrence.

Another meta-analysis of studies comparing stapled side-to-side anastomosis and handsewn end-to-end anastomosis after ileocolic resection in patients with Crohn's disease found stapled side-to-side anastomoses to be superior in terms of overall postoperative complications (OR, 0.54; 95% CI 0.32–0.93), anastomotic leak (OR 0.45; 95% CI 0.20–1.00), recurrence (OR 0.20; 95% CI 0.07–0.55), and re-operation for recurrence (OR 0.18; 95% CI 0.07–0.45). ⁴² Another meta-analysis published by Guo compared side-to-side with other anastomotic configurations, including data from 11 trials, and found that there was a significant reduction in the risk of overall postoperative complications (OR, 0.60; p  = 0.01) with side-to-side anastomoses. However, side-to-side anastomoses did not reduce the anastomotic leak rate (OR, 0.48; p  = 0.07), endoscopic recurrence rates (HR, 0.73; p  = 0.07), or reoperation rates for recurrence (HR, 0.37; p  = 0.06). ⁴³

A retrospective, multicenter observational study published by the Italian Society of Colorectal Surgery found that a side-to-side anastomotic technique was the chosen configuration in 89% of patients undergoing ileocecal resection for Crohn's disease, with a stapled technique being used in 67%. Anastomotic leak rate was 3.7% and was found to be independent of the type of anastomosis performed. Anastomotic leak was associated with ASA grade≥ 3, a penetrating phenotype of disease and ileo-colonic distribution of Crohn's disease. ⁴⁴ Another retrospective observational study reported better quality of life scores and less utilization of health care resources postoperatively in Crohn's disease patients undergoing end-to-end anastomosis compared with an anti-peristaltic side-to-side configuration, with no difference in 30-day surgical complications, 2-year pattern of disease activity, and endoscopic recurrence. No anastomotic leaks occurred in either group. ⁴⁵

Muñoz-Juárez published one of the first case-controlled studies to compare wide-lumen stapled side-to-side anastomoses with hand-sewn end-to-end anastomoses, reporting that clinical recurrence was seen less frequently in the side-to-side group compared with the end-to-end group (24 vs. 57%), and the cumulative surgical recurrence rate was 11% at 5 years in the side-to-side group compared with 20% in the end-to-end group ( p  = 0.017). ⁴⁶ Resegotti reported a retrospective non-randomized series of 122 consecutive patients who underwent elective resection with ileocolonic anastomosis for Crohn's disease with either handsewn end-to-end anastomosis or side-to-side stapled anastomosis and found that the anastomotic leak rate in the handsewn group was 14.1%, compared with 2.0% in the stapled group (risk difference, +12.1%; 95% CI, 1.7–22.2; p  = 0.02). While more patients in the handsewn group were on corticosteroids at the time of surgery, anastomotic configuration was found to be the sole variable that influenced anastomotic leak rates on univariate analysis. Additionally, complications other than anastomotic leak developed in 15.5% of patients in the hand-sewn group, compared with 11.8% of patients in the stapled group. ²⁴ Scarpa reported a retrospective series of 141 consecutive patients who had undergone ileocolonic resection for Crohn's disease, and found that those with a hand-sewn side-to-side anastomosis had a significantly lower surgical recurrence rate than stapled end-to-side anastomosis ( p <0.05); on multivariate analysis, anastomosis type, surgical and intestinal complications ( p <0.01), and age at Crohn's disease onset ( p <0.05) were found to be predictors for re-operation for Crohn's disease recurrence. ⁴⁷

In 2011, Kono reported a novel anastomotic technique utilizing a hand-sewn antimesenteric functional end-to-end anastomosis, centered on preservation of the mesenteric vascularization and innervation with a posterior supporting column created by suturing the two staple lines together. ⁴⁸ With this technique, a small window is created in the mesentery at the level of the proximal and distal resection margins, and the mesentery is divided using a tissue-device close to the intestinal wall to preserve vascularization and innervation. The bowel is then divided by placing a linear stapler perpendicular to the intestinal lumen and the mesentery, so that the mesentery is located in the middle of staple lines. The two stapled lines then sewn together with interrupted silk sutures, creating the so-called “supporting column.” An antimesenteric longitudinal enterotomy/colotomy is created on each stump to allow a transverse lumen of 7 to 8 cm, and the anastomosis is completed by closing the longitudinal openings transversely in two layers. In addition to preserving mesenteric vascularization and innervation, the supporting column essentially excludes the anastomosis from the mesentery. This technique has been termed the “Kono-S anastomosis,” with the “S” signifying the use of a supporting column. In their original paper, the authors performed the Kono-S anastomosis in 69 Crohn's disease patients and compared this group with a historical cohort of 73 Crohn's disease patients undergoing hand-sewn or stapled end-to-end anastomoses or stapled side-to-side anastomoses and reported significantly lower endoscopic recurrence rates in the Kono-S group than in the conventional one, with a lower probability of anastomotic surgical recurrence in the Kono-S group at 5 years (0 vs. 15%; p <0.0013) ⁴⁸ ( Fig. 1 )

Fig. 1.

Kono-S anastomosis for ileocolic resection for Crohn's. (Source: Reproduced with permission of Scott R. Steele, MD, Cleveland Clinic, all rights reserved).

This anastomotic configuration has been evaluated in two large multicenter studies. In the first, only two surgical anastomotic recurrences were noted in 187 patients undergoing a Kono-S anastomosis, both in Japan and in the United States, with a follow-up of 65 months and 5- and 10-year surgical recurrence-free survival rates of 98.6%. ⁴⁹ In the second, ⁵⁰ patients undergoing Kono-S anastomosis were found to have a surgical recurrence rate of 3.4%, compared with 24.4% in those undergoing sutured end-to-end anastomosis. Additionally, the Kono-S group was found to have a lower anastomotic leak rate (5.1 vs. 17.3%), a lower risk of anastomotic surgical recurrence at 1 year (OR 0.14), and a higher 5-year surgery-free survival rate on the anastomosis site (95 vs. 81.3%; p <0.001).

Luglio published the first randomized controlled trial comparing Kono-S and stapled side-to-side anastomoses in 2020. ⁵¹ After 6 months, endoscopic recurrence was seen in 22.2% in the Kono-S group compared with 62.8% in the conventional group ( p <0.001, OR 5.91). Clinical recurrence was seen in 8% in the Kono-S group versus 18% in the conventional group after 12 months ( p  = 0.2) and in 18% versus 30.2% after 24 months ( p  = 0.04, OR 3.47). The surgical recurrence rate after 24 months was 0% in the Kono-S group compared with 4.6% in the conventional group ( p  = 0.3). Patients in the Kono-S group also presented with clinical recurrence at a longer time interval. A systematic review that included 896 patients from nine studies found that the Kono-S anastomosis was associated with a lower incidence of endoscopic recurrence (0 vs. 3.4%), surgical recurrence (15 vs. 24.4%), and anastomotic leak (1.8 vs. 9.3%). ⁵²

There are several theoretical advantages to this anastomotic configuration. It has been proposed that creation of the supporting column immediately behind the posterior wall of the anastomosis prevents flow-limiting alterations of the fecal stream and maintains the orientation and large diameter of the anastomosis, potentially limiting fecal stasis, which is felt to play a part in the pathogenesis of recurrent Crohn's disease. Additionally, the mesenteric side of the intestine, which is usually the initial site of macroscopic anastomotic recurrence, is positioned in the center of the posterior wall of the anastomosis. The supporting column fixes it in place, reinforcing the central portion of the posterior wall suture line, so that if macroscopic recurrence occurs on the mesenteric side of the anastomosis, the supporting column prevents distortion of the lumen of the anastomosis.

Preservation of the innervation and blood supply, which are crucial for healing of the anastomosis is also felt to play a role in the prevention of endoscopic recurrence. It has been shown in both animal models and patients with Crohn's disease that blood flow to the bowel can be decreased by more than 50% in patients with Crohn's disease due to depletion of calcitonin gene-related peptide, which functions as a potent vasodilator. ⁵³ Decreased blood flow, which can be further compromised at an anastomotic site, has also been shown to be associated with recurrent disease at anastomotic sites. ⁵⁴ By dividing the mesentery as close as possible to the bowel wall at the anastomotic site, unnecessary denervation or devascularization is minimized. Mesenteric adipose tissue has been shown to clearly play a role in progression of Crohn's disease, and excluding it from the anastomotic lumen, as the Kono-S anastomosis does, may contribute to lower recurrence rates.

The mesentery has long been thought by some to be involved in the initiation and recurrence of Crohn's disease, as early ulcers develop typically on the mesenteric side of the bowel with the corresponding “creeping fat.” The role of wide mesenteric resection to remove all visibly diseased mesentery has been debated. While this would seem to reduce this risk of recurrence, the technical challenge of removing grossly thickened, inflamed, edematous mesentery down to the base of the mesentery is often accompanied by significant blood loss and risk of compromising the vascular supply to the remaining bowel. A recent report from Ireland compared wide excision of the mesentery with conventional closer division in 64 patients. They reported surgical recurrence rates of 2.9 versus 40% in favor of the wide excision group ( p  = 0.003). ⁵⁵

In summary, there is no definitive evidence supporting superior safety of one anastomotic technique over the other following ileocolic resection for Crohn's disease. More recent studies seem to favor a side-to-side configuration over an end-to-end configuration, though the Kono-S anastomosis does seem to challenge this. With regards to the risk of recurrence, the role of radical mesenteric excision and the promising results reported with the Kono-S anastomosis will require further study.

Ileorectal Anastomosis

For patients with isolated colonic Crohn's disease not suitable for segmental colectomy, total abdominal colectomy with ileorectal anastomosis may be an option to maintain intestinal continuity, assuming the absence of active rectal or perianal disease. This option avoids the risks of bladder and sexual dysfunction after the pelvic dissection while performing proctectomy and may offer an improved postoperative quality of life while potentially avoiding a permanent stoma. The main drawbacks include increased risk of anastomotic leak, unacceptable functional outcomes, and recurrent disease necessitating completion proctectomy and a permanent ileostomy.

The majority of the literature surrounding the role of ileorectal anastomosis in the setting of Crohn's disease consists of retrospective case series and cohort studies, with only a handful of prospective studies published in the pre-biologic era. ^{56 57 58 59 60} Prior to the use of biologics, patients with Crohn's colitis were often offered proctocolectomy with permanent ileostomy due to the suboptimal outcomes with segmental resection or ileorectal anastomosis. During the pre-biologic era, proctectomy rates following failed ileorectal anastomosis were reported to be as high as 80%. ^{58 61}

With the advent of biologics and their role in postoperative prophylaxis for clinical recurrence, the use of ileorectal anastomosis in properly selected patients with Crohn's colitis has become more widespread. Sciaudone published a prospective series in 2011 in which long-term infliximab was shown to result in rectal preservation in 83.3% of patients with recurrent disease who had previously undergone an ileorectal anastomosis. ⁶² Anorectal function, quality of life, continence, and the number of daily defecations were shown to improve as well after initiation of treatment with infliximab. McKenna reported a retrospective cohort study in 2020 aimed at evaluating ostomy-free survival and predictors of ileostomy requirement in patients who had undergone prior ileosigmoid or ileorectal anastomosis for Crohn's disease in the biologic era. ⁶³ At a mean follow-up of 3 years, 14.7% had undergone either proctectomy or diversion with the rectum in situ, resulting in ostomy-free survival estimates at 5 and 10 years of 78% (95% CI, 68–90) and 58% (95% CI, 35–94). A positive distal microscopic margin was the only risk factor predicting a later requirement for permanent ileostomy (OR, 5.4; 95% CI, 1.7–17.2).

Careful patient selection is critical in deciding whether to perform an ileorectal anastomosis in the setting of Crohn's disease. Salice published a retrospective study in 2020 that stratified the risk of recurrence in patients with a history of colectomy and ileorectal anastomosis for refractory Crohn's colitis according to the presence or absence of rectal/ileal/perineal disease. ⁶⁴ Disease recurred in 89% of patients with Crohn's disease with rectal/ileal/perianal involvement (in spite of postoperative biologic therapy in 65.6%), compared with only 6.3% of patients without rectal/ileal/perineal disease.

A systematic review published in 2021 that included 37 studies assessing the outcome of ileorectal anastomosis in Crohn's disease found a cumulative anastomotic leak rate of 6.4%, cumulative rates of clinical recurrence of 43% at 5 years and 67% at 10 years, and an overall rate of proctectomy of 18.9%. ⁶⁵ The authors also reported only minor improvements in outcomes for ileorectal anastomosis in the setting of Crohn's disease over the past 50 years, in spite of the advances in the medical management of Crohn's disease over this time frame.

Ileal Pouch-Rectal Anastomosis

In certain instances, a patient with Crohn's disease will have inflammation of the colon and upper rectum, and resection of the diseased large bowel will leave the patient with only a short segment of healthy rectum. Decreased storage capacity of the short rectal remnant will result in the need for frequent bowel movements. In this situation, construction of a small (8–10 cm) ileal pouch reservoir may be created and anastomosed to the residual rectum. A report from the Cleveland Clinic found that of 22 patients who underwent an ileal pouch-rectal anastomosis with long-term follow-up, 50% developed perianastomotic disease recurrence and 10% ultimately required pouch-rectal excision with permanent ileostomy or permanent defunctioning loop ileostomy. ⁶⁶ When patients with Ileal pouch-rectal anastomosis were compared with case-matched patients who underwent straight ileorectal, anastomosis, bowel movement frequency, fecal urgency and incontinence, and quality of life were similar, though night-time fecal seepage and protective pad usage were more frequent after ileal pouch-rectal anastomosis. Given that the alternative in these cases is most often an ileal pouch-anal anastomosis (IPAA) or permanent ileostomy, the authors felt that ileal pouch-rectal anastomosis could be considered in carefully selected patients.

Ileal Pouch-Anal Anastomosis

A number of patients who undergo total proctocolectomy and IPAA for suspected ulcerative colitis are ultimately found to have Crohn's disease, as evidenced by pouch-related complications, such as anal abscesses, fistulae, stricture, and stenosis. Retrospective studies prior to the widespread use of biologics evaluating patients thought to have ulcerative colitis who ultimately proved to have Crohn's disease reported high complication and pouch failure rates, with a high incidence of pouch excision or indefinite diversion. A retrospective study from the Lahey Clinic published in 2004 found a pouch failure rate of 29% and a perioperative complication rate of 93% in patients with an IPAA in the setting of Crohn's disease. ⁶⁷ IPAA-related perineal abscess (67 vs. 26%, p  = 0.05) and pouch fistula (89 vs. 30%, p  = 0.01) were associated with pouch failure, while preoperative clinical, endoscopic, and pathologic features were not predictive of pouch failure. A similar study from the Cleveland Clinic reported a pouch failure rate of 33% for those with recrudescent Crohn's disease. ⁶⁸

Fistulizing disease of the ileal pouch is one of the most common causes of pouch failure. Among patients diagnosed with fistulizing Crohn's disease of the pouch, 30 to 80% will eventually require fecal diversion or pouch excision. ⁶⁹ It is sometimes difficult to distinguish fistulas related to technical surgical issues from fistulizing Crohn's disease of the pouch. The anatomical location of the fistula, timing of fistula formation in relation to when surgery was performed, and degree of response to treatment can help distinguish the two. Anatomically, a fistula associated with the anastomosis or with the efferent limb is more likely to suggest a surgical etiology, while a fistula arising from the anal canal distal to the anastomosis is more likely to be a result of Crohn's disease. A fistula in the immediate postoperative period is more likely to be related to an anastomotic leak; a fistula occurring more than 6 to 12 months after surgery without evidence of a postoperative leak, abscess, or pelvic sepsis may suggest Crohn's disease of the pouch. ^{70 71} Additionally, a good clinical response to anti-TNF agents favors a diagnosis of Crohn's, whereas a lack of response to medical management suggests that the fistula may be related to a technical or surgical issues. ⁶⁹

More recent studies have suggested lower rates of pouch failure in patients with Crohn's disease. A prospective series from the Cleveland Clinic published in 2013 reported a 13.3% rate of pouch failure in Crohn's patients, compared with 5.1% in those with ulcerative colitis. ⁷² Li reported a comparable 15.5% indefinite diversion rate in patients with Crohn's disease undergoing intentional IPAA. ⁷³ Based on a number of similar studies, there is growing support and an evolving body of literature surrounding restorative proctocolectomy utilizing IPAA in patients with established Crohn's colitis and the absence of perianal and small bowel disease, mainly at a number of high-volume centers. Obviously, this requires that both the patient and surgeon assume some risk tolerance and that their shared decision-making priorities are aligned.

Le reported a series of 17 patients with known preoperatively diagnosed Crohn's disease who underwent planned IPAA, only one of which required pouch excision. ⁷⁴ The majority of patients were treated with biologics postoperatively, yet 41% of patients developed postoperative recurrent Crohn's disease in the afferent limb or pouch fistulizing disease. Interestingly, afferent limb inflammation developed in 50% of patients with pANCA +/OmpC- expression compared to none of the patients without this serologic profile ( p  = 0.03). Other studies have suggested that IPAA can be performed in patients with Crohn's disease in the absence of perianal or small bowel disease with comparable outcomes to those undergoing surgery for ulcerative colitis. ^{75 76} A meta-analysis published in 2021 found an overall pouch failure rate of 15% in patients undergoing intentional IPAA with a preoperative diagnosis of Crohn's disease. ⁷⁷ The authors reported that pelvic sepsis occurred in 13% of patients, pouchitis in 31%, anal stricture in 18%, and chronic sinus tract in 28%. Mean 24-hour stool frequency was 6.3, and no significant risk factors for pouch failure were identified.

Divert or Not?

When performing surgery for Crohn's disease, one of the most feared complications is an anastomotic leak. This can be a significant, life-threatening morbidity resulting in interventions such as prolonged antibiotic administration, percutaneous drainage, need for prolonged parenteral nutrition, and emergency laparotomy with potential for additional bowel resection and fecal diversion. Given the complexity and number of disease-specific risk factors, the clinical decision about when to perform a primary anastomosis as opposed to a primary anastomosis with diverting loop ileostomy or an end ileostomy at the time of ileocolic resection for Crohn's disease can be complex and multifactorial. This decision is typically made based on very subjective assessments, estimating the risk and consequences of an anastomotic leak compared with the morbidity associated with having a diverting loop ileostomy, including pouching difficulty, readmission for dehydration and electrolyte derangements, morbidity associated with the reversal surgery, and the potential for non-reversal. In general, this decision is often made based on the surgeon's education, experience, and preferences, estimating the risk of anastomotic leak and utilizing a diverting loop ileostomy as deemed necessary. Unfortunately, this traditional method has been shown to be inaccurate and have the potential to result in avoidable adverse outcomes. ⁷⁸

Neary published a study that used data from the National Surgical Quality Improvement Program Colectomy Module and performed multivariable logistic regression analysis and propensity-score matching to identify independent risk factors for leak. ⁷⁹ Multivariate analysis found that omission of an ileostomy, emergency surgery, smoking, inpatient status, wound classification 3 or 4, weight loss, steroid use, and prolonged operative time were all independently associated with leak. Patients with 0 to 6 risk factors had leak rates of 1.6, 2.7, 4.3, 6.7, 8.8, 11.5, and 14.3% ( p ≤0.001), respectively. After propensity-score matching, creation of an ileostomy was found to reduce the risk of leak rate by 55% ( p  = 0.005). Additionally, patients with primary anastomosis who leaked most frequently required reoperation (57.8%), while those with an anastomosis protected by a diverting ileostomy who leaked were most frequently managed by percutaneous drainage (70%), p  = 0.04. The authors concluded that fecal diversion should be considered when ≥3 of the risk factors listed above are present. The same authors published a report using their own institutional data, looking at patients who underwent laparoscopic ileocecal resection for Crohn's disease from 2005 to 2015, and found that a diverting stoma was created in 22% of cases; this was independently associated with BMI <18.5 kg/m ² ( p  = 0.001), low serum albumin levels ( p  = 0.006), and longer operative time ( p  = 0.003). Use of any immunosuppressive medication was the only variable independently associated with septic complications, both in the overall population (OR 2.7, p  = 0.036) and in the subgroup of undiverted patients (OR 3.1, p  = 0.031). They concluded that their data are unable to prove that a diverting stoma is associated with reduced morbidity. ⁸⁰

Another retrospective study published by Tzivanakis found that in patients undergoing ileocecal or ileocolic resection for Crohn's disease, steroid usage (OR 2.67, 95% CI 1.0–7.2) and the presence of preoperative abscess (OR 3.4, 95% CI 1.2–9.8) were found to be independent predictors of anastomotic-associated complications. In the absence of both steroids and intraabdominal abscess, the risk of anastomotic complications was 6%, which increased to 14% if either risk factor was present and 40% when both risk factors were present. The authors advocated for avoidance of a primary anastomosis when both risk factors are present. ¹⁷

Instead of a protective loop ileostomy upstream from an ileocolic anastomosis, other authors have advocated for the use of a “split stoma,” with excision of the disease and exteriorization of both ends of the bowel through a common trephination, followed by an anastomosis at some point in the future. Myrelid found that in high-risk patients (two or more preoperative risk factors) undergoing surgery for ileocolonic Crohn's disease, early anastomotic complications were diagnosed in 19% of patients undergoing primary anastomosis compared with 0% of patients after split stoma creation and delayed anastomosis ( p  = 0.038). ⁸¹ A multicenter retrospective study from Europe ⁸² took a different approach to evaluating temporary fecal diversion after ileocolic resection for Crohn's disease in terms of reducing long-term surgical recurrence. They reported that 20% of patients undergoing ileocolic resection for Crohn's disease underwent fecal diversion at the time of surgery with a median time to ileostomy closure of 9 months (IQR 5–12 months). At a median follow-up of 105 months, they found no significant overall difference in surgical recurrence between the one- and two-stage groups (18 vs. 16%, p  = 0.94), though a non-significant reduced recurrence rate was observed in a small high-risk group of patients who were smokers with penetrating disease treated with a two-stage strategy (0/10 vs. 4/7, p  = 0.12).

Conclusion

Several considerations should come to mind when performing an intestinal anastomosis in the setting of surgery for Crohn's disease. A number of preoperative risk factors may increase the possibility of postoperative anastomotic complications and alter surgical decision-making. While malnutrition, abdominal sepsis, and preoperative steroid use have been shown to increase the risk of postoperative complications, including anastomotic leak, the literature surrounding the impact of other preoperative immunosuppressive medications such as anti-TNF-α agents is less clear. A recently published multicenter prospective trial found no increase in the risk of overall infection or surgical site infection in patients with inflammatory bowel disease undergoing intraabdominal surgery with preoperative anti-TNF-α exposure, and the most recent Clinical Practice Guidelines for the Surgical Management of Crohn's Disease published by the American Society of Colon and Rectal Surgeons recommend against delaying surgical intervention based on monoclonal antibody therapy alone. The impact of anatomical configuration and technique of anastomotic construction on postoperative function and risk of recurrence varies widely in the reported literature, though in recent years, the Kono-S anastomosis has been used with increasing frequency with lower rates of post-surgical disease recurrence. In the setting of total abdominal colectomy or proctocolectomy for Crohn's colitis, restorative procedures may be utilized in properly selected patients, though both the patient and surgeon must be willing the possibility of the eventual need for additional surgery with proctectomy and a permanent stoma. Lastly, one should have a well-defined thought-process to help guide the intraoperative decision as to whether or not to perform a temporary ostomy when performing surgical resection for Crohn's disease.