Localization of recurrent lesions ileocolic resection for Crohn's disease


 Background

Crohn’s disease (CD) recurrence can occur not only at the site of anastomosis but also elsewhere in the bowel following ileocolic resection (ICR). The aims of the present study were to assess long-term outcomes of a primary ICR procedure for CD in consecutive patients and examine the location of the reoperation causative lesion.
Methods

We examined cases of surgery with ICR initially performed at our institution. Those with a simultaneous multiple bowel resection or bowel resection with strictureplasty were excluded.
Results

A total of 169 patients who underwent ICR due to CD were enrolled. The median follow-up period was 12.6 years (range 4–27 years). A reoperation was needed in 45 (26.6%), of whom 14 had lesions causative of the reoperation at other than the anastomotic site. The most common causative lesion location was in the colon rather than the oral side of the small intestine. Furthermore, we investigated the relationship between presence of residual lesions following the initial surgery and lesions causative of reoperation. In the group without residual disease (n = 31), 29.0% (n = 9) had non-anastomotic lesions involved in indications for reoperation, while that was 35.7% (n = 5) in the group with residual disease (n = 14).
Conclusions

Anastomotic site lesion is not the only causative factor for reoperation following ICR. Regular examinations and applicable treatment with awareness that the cause of reoperation is not limited to the site of anastomosis are important in these cases.


Abstract Background
Crohn's disease (CD) recurrence can occur not only at the site of anastomosis but also elsewhere in the bowel following ileocolic resection (ICR). The aims of the present study were to assess long-term outcomes of a primary ICR procedure for CD in consecutive patients and examine the location of the reoperation causative lesion.

Methods
We examined cases of surgery with ICR initially performed at our institution. Those with a simultaneous multiple bowel resection or bowel resection with strictureplasty were excluded.

Results
A total of 169 patients who underwent ICR due to CD were enrolled. The median follow-up period was 12.6 years (range 4-27 years). A reoperation was needed in 45 (26.6%), of whom 14 had lesions causative of the reoperation at other than the anastomotic site. The most common causative lesion location was in the colon rather than the oral side of the small intestine. Furthermore, we investigated the relationship between presence of residual lesions following the initial surgery and lesions causative of reoperation. In the group without residual disease (n = 31), 29.0% (n = 9) had non-anastomotic lesions involved in indications for reoperation, while that was 35.7% (n = 5) in the group with residual disease (n = 14).

Conclusions
Anastomotic site lesion is not the only causative factor for reoperation following ICR. Regular examinations and applicable treatment with awareness that the cause of reoperation is not limited to the site of anastomosis are important in these cases.

Background
Crohn's disease (CD) is a chronic in ammatory disease potentially involving the entire gastrointestinal tract. Long-standing in ammation causes irreversible damage to the bowel wall, resulting in either brotic strictures or penetrating disease. Conservative medical treatment is considered to be the principal therapeutic approach to treat patients with CD, though up to 80% will ultimately require surgery [1].
Moreover, a repeat operation for recurrent CD has been estimated to be required in up to 40% of surgically treated cases after 10 years [2].
The anastomotic site following CD surgery is where recurrence is most likely to occur. Recently, there has been an increasing number of reports showing that results of treatment were improved by providing stepup therapy based on endoscopic ndings of the site of anastomosis. On the other hand, Fichera A et al.
reported that even though it is commonly believed that Crohn's recurrence is always at the site of a previous intestinal anastomosis, such a concept is probably not true nor mindful of the pan-intestinal nature of the disease [3]. They went on to state that while the site of original operative intervention is the most common for recurrence, as many as one-third of recurrence cases appear in an area separate from that. It is important to note that in CD cases there may be gaps between endoscopic, clinical, and surgical recurrence. Most published studies have not addressed surgical recurrence as the end-point and enrolled a low number of patients, thus surgical recurrence has not been addressed.
The aim of the present study was to assess long-term outcomes of a primary ileocolic resection (ICR) for CD in consecutive patients and clarify the location of the causative lesion related to the reoperation.

Patients and data collection
The authors developed, maintained, and upgraded a database containing clinical ndings of all CD patients who underwent an operative procedure at our institution, with data entered on a prospective basis and updated with each new instance of recurrence. This database was used to investigate CD recurrence sites for the present study.
From January 1990 to December 2015, 1143 patients with CD underwent surgery at the Department of Gastroenterological Surgery, Hyogo College of Medicine. Of those, cases of initial surgery with ICR were selected. Patients with simultaneous multiple bowel resection and bowel resection with strictureplasty were excluded. Altogether, 169 patients undergoing ICR due to CD were included in the study. The median follow-up period was 12.6 years (range 4-27 years).
All surgeries were performed by colorectal surgeons experienced in surgical management of CD. A handsewn end-to-end ileocolonic anastomosis was constructed as a two-layer anastomosis using a continuous inner layer reinforced with interrupted sero-muscular sutures. Postoperative prophylaxis with immunomodulators and/or biologic agents or other treatments were started 2-4 weeks after surgery at the discretion of the attending gastroenterologist.
De nitions CD was con rmed by a histological examination of the resected specimen. Surgical recurrence was de ned as the necessity of a reoperation because of recurrent CD during the observation period. Lesions considered responsible for the reoperation were classi ed as anastomotic, non-anastomotic, or both.
When the preoperative examination results showed no obstruction but intraoperative ndings revealed CD lesions with creeping fat, that case was de ned as residual disease ( Figure 1).

Statistical analyses
Descriptive statistics are reported as median values (range) and frequency (percent) for categorical variables. Survival data are presented as Kaplan-Meier curves and described by 5-year probability for surgical recurrence survival. All statistical analyses were performed using JMP version 11 (SAS Institute Inc., Cary, NC, USA).

Ethical considerations
The study protocol was approved by the Ethical Committee of Hyogo College of Medicine.

Patient characteristics
Of the 169 patients enrolled, 117 (69.2%) were male and 52 (30.8%) were female. Mean age at onset of CD was 22 years  and mean age at the initial operation was 30 years (14-76). Patient demographics are presented in Table 1.

Surgical indications
Indications for the initial procedure for all cases are shown in Table 2, with stenosis found to be most common.

Postoperative complications
Postoperative complications were noted in 10 patients (5.9%), with details shown in Table 3. Wound infection and intra-abdominal abscess were the most common following the initial operation, though no cases required a reoperation due to a postoperative complication.

Localization of lesions causing reoperation
Forty-ve patients required a reoperation due to recurrence of CD. Table 5 shows details of the causative lesion. Fourteen of these patients had causative lesions in other than the anastomotic site.
Localization of causative lesions other than anastomosis Table 6 shows localization of lesions in the 14 cases in which lesion location other than the anastomotic region caused the reoperation. There were some overlaps. The most common was a lesion in the colon rather than the oral side small intestine. Intraoperative ndings obtained at the initial surgical procedure in these reoperation cases showed no residual lesion in the colon.

Presence of residual lesion and localization of lesion causing reoperation
Relationships between presence or absence of residual lesions in the proximal small intestine at the time of the initial surgery and localization of lesions responsible for reoperation are shown in Table 7. Nonanastomotic lesions were involved in the surgical indication for reoperation in 9 (29.0%) of 31 patients without residual disease and 5 (35.7%) of 14 with residual disease.

Cumulative 5-year reoperation rate
The cumulative reoperation rate is presented in Figure 2. That after 5 years was 11.3% in the present cohort.

Discussion
Aaltonen G et al. reported that the majority of cases of surgical recurrence (71.4%) were because of a lesion located in the proximal portion of the anastomosis and required a new ileocolic resection [4], while other studies also concluded that surgical recurrence after an ileocolic resection is more common in the anastomosis [5,6]. Furthermore, the reoperation rate after stoma construction is high in CD patients. In a recent study, Aaltonen reported a surgical recurrence rate of 38% with a median follow-up of 2.4 years [7].
Based on these results, step-up therapy based on endoscopic ndings of the anastomosis site or oral lesions of the stoma is generally given to affected patients.
In a study performed approximately 30 years ago, Rutgeerts and colleagues showed that the severity of mucosal lesions was correlated with likeliness to develop clinical recurrence [8]. Thereafter, the so-called Rutgeerts endoscopic index has been used to de ne the primary endpoint in most randomized controlled trials conducted in the recent decade to evaluate preventive strategies for postoperative recurrence (POR) as a surrogate for clinical recurrence. Although POR is usually seen in the neoterminal ileum after ICR, some patients may develop mucosal lesions in another intestinal area. This is not considered when determining Rutgeerts score.
On the other hand, Fichera A et al. noted that even though it is commonly believed that Crohn's recurrence is always located at the site of a previous intestinal anastomosis, that concept is probably not true and not mindful of the pan-intestinal nature of the disease [3]. In that study, the authors reported that while the site of original operative intervention is the most common for recurrence, as many as one-third of recurrence cases occur in a location separate from that. In the present study as well, 14 (31.1%) cases had lesions away from the anastomosis site causative of the reoperation. In addition, abdominal ndings obtained at the rst operation, even in cases with no macroscopic residual lesion, showed that 9 (29.0%) of 31 of the lesions causative of the need for a reoperation were located in other than the anastomotic region.
Recently, in order to reduce the number of reoperations caused by anastomosis, improvements in anastomosis treatment methods have been made, with endoscopic balloon dilation of the stenosis site commonly performed. To prevent surgical recurrence at the anastomotic site, Kono et al. developed a new antimesenteric functional end-to-end anastomosis technique in 2003 termed Kono-S anastomosis [9]. With this technique, the supporting column prevents distortion and keeps the intestinal tract in the anastomosis region straight. A straight structure also makes endoscopic examination and balloon dilatation following surgery easier to perform. Shimada N et al. reported that the 5-year surgery-free survival rate at the anastomosis site with Kono-S anastomosis (95%) was signi cantly higher as compared to end-to-end anastomosis (81.3%; p<0.001) [10]. Furthermore, Luglio G et al. reported the rst randomized clinical trial to compare Kono-S anastomosis and conventional side-to-side anastomosis in CD patients. Those results demonstrated a signi cant reduction in postoperative endoscopic and clinical recurrence rates for patients who underwent Kono-S anastomosis [11].
Reports of the usefulness of balloon dilatation for stenosis in CD cases are increasing. Shivashankar R noted that endoscopic stricture dilatation in CD was safe and effective, and that the most common stricture location was ileocolonic anastomosis [12]. Ding NS as well reported that endoscopic balloon dilatation of an anastomotic stricture in Crohn's cases is safe and effective over the long term [13]. Based on these results, it is considered possible that lesions in locations other than the anastomotic site will cause a relatively high rate of reoperation.
Alternative non-invasive imaging techniques used to assess postoperative recurrence have been evaluated in recent years [14][15][16]. Wireless capsule endoscopy has potential advantages over ileocolonoscopy, as it is more comfortable and better tolerated by patients, does not require sedation, and is less in uenced by technical limitations. Wireless capsule endoscopy was shown able to detect mucosal lesions one year after surgery in the upper segment of the gastrointestinal tract, outside the area of visualization by ileocolonoscopy, in 60% to 70% of examined patients. Furthermore, in recent years, the number of reports showing the usefulness of minimally invasive magnetic resonance (MR) enterography for evaluation of lesions of CD has been increasing [17,18].
In future examinations, not only the anastomotic site, but all remaining intestinal tracts should be checked regularly using a minimally invasive method such as capsule endoscopy or MR enterography. In cases with a newly diagnosed lesion, it may be possible to avoid a reoperation by performing step-up therapy.
The present study has some limitations. The gastroenterologist in charge of the follow-up examinations assessed the need for therapeutic modi cation on an individual patient basis, without reference to a prede ned protocol.

Conclusion
It is important to recognize that there may be a discrepancy between endoscopic or clinical and surgical recurrence in CD patients. In such cases, it is expected that the incidence of reoperations due to an anastomotic stricture will be reduced by improving the anastomosis method and advancing balloon dilatation. We consider that greater attention should be given to relapse in the remaining intestinal tract, thus recommend regular examinations and treatments with due recognition that the cause of reoperation may not be limited to the anastomosis. This study protocols were approved by the institutional review board at Hyogo College of Medicine, and informed consent for permission to use patient data was required prior to admission.

Consent for publication
Not applicable

Availability of data and materials
The data sets supporting the conclusions of this article are included within the article. Data are available from the corresponding author upon reasonable request.  Figure 1 The cumulative 5-year reoperation rate