This was a retrospective, single-center, institutional review board approved study. CD patients treated by ileocolic resection and ileocolic anastomosis were identified by performing a search in our single-institution radiology database for the association of the keywords "Crohn's disease" and "anastomotic recurrence" in all CTe-WE examinations from January 2007 through December 2011.

A total of 75 CTe-WEs were obtained, and if one of the selected patients underwent more than one CTe-WE, only the more recent examination was considered for further evaluation, thus resulting in 51 CTe-WEs of CD patients with ileocolic anastomosis. The following clinical and laboratory data were retrieved from the medical records of each patient, when available: 1) age of onset of CD symptoms; 2) time interval between onset of symptoms and surgery; 3) Crohn's Disease Activity Index (CDAI), C-reactive protein values, and fecal markers of inflammation – faecal calprotectin and lactoferrin – performed within one month of the CTe-WE; 4) clinical indication for the CTe-WE. Surgical records were also reviewed, and ileocolic anastomoses were divided in three main types: “wide-lumen” stapled isoperistaltic side-to-side (functional end-to-end), end-to-side (including both stapled and hand-sewn subtypes), end-to-end (including both stapled and hand-sewn subtypes). Reports of endoscopy studies performed within one month of the CTe-WE examination were analyzed, and Rutgeerts' scores [14] (0: no lesions; 1: less than 5 aphthous lesions; 2: more than 5 aphthous lesions with normal mucosa between the lesions, or lesions confined to the ileocolonic anastomotic lining (<1 cm); 3: diffuse aphthous ileitis with diffusely inflamed mucosa; 4: diffuse ileal inflammation with larger ulcers, nodules, or narrowing) were considered, when available. When the Rutgeerts' score was not explicitly written, it was retrieved from the endoscopy report. Mucosal hyperemia and edema of the anastomosis were not considered as signs of recurrence [15].

CTe-WE technique

CTe-WE technique was performed as previously described [13, 16]. Large bowel cleansing was obtained with a low fiber diet for 3 days before the CTe-WE, and oral administration of 2.000 mL of an isotonic non-absorbable electrolyte solution containing polyethylene glycol (Isocolan, Bracco, Milan, Italy) was given the afternoon before examination. At the time of examination, both small and large bowels were distended with neutral enteral contrast material. Small bowel lumen distension was achieved by oral administration of 1.000–2.000 mL of the previously mentioned isotonic solution (Isocolan) 45 min prior to CTE-WE; large bowel lumen was distended by trans-rectal introduction of 2,000 mL of tap water with the patient placed on the CT table. Just before the CT acquisition, bowel hypotonia was obtained with intravenous injection of 2 mL hyoscine-N-butylbromide 20 mg/mL (Buscopan, Boehringer Ingelheim). CT images were acquired with the patient in supine position. Bowel wall enhancement was produced by i.v. injection of iodinated contrast medium with an iodine concentration ranging between 350 and 370 mg/mL (iobitridol, Xenetix 350, Guerbet, or iopamidol, Iopamiro 370, Bracco). The iodine load was 1.5 mg per kg of body weight. The flow rate was set at 3.2–3 mL/s with an automatic injector and CT acquisition was started in the portal phase 45s after contrast media bolus detection in the upper abdominal aorta, using a bolus-tracking software.

Description of Reading Criteria for CTe-WE Images

CTe-WEs were reviewed in consensus by two gastrointestinal radiologists (M.R. and C.P.) on dedicated workstations, and 1.25-mm thickness reconstructed images were used to obtain Multi Planar Reconstructions (MPR). The distension of the ileal and colonic sides of each anastomosis was qualitatively classified according to a four-point scale (0=absent, 1=incomplete, 2=good, 3=optimal), as previously proposed by Minordi et al [11]. A score of 3 meant that the anastomosis was optimally distended, and the wall was uniformly visualized. A score of 0 meant that the anastomosis was collapsed without any luminal separation and the wall could not be seen. For this purpose axial and multiplanar reformatted images were analyzed. The number of CTe-WEs with a score ≥2 (i.e. luminal filling adequate for a correct evaluation) was calculated. Obviously, all cases of stricturing CD recurrence may have determined lower scores of distension. Readers had to indicate what side of the ileocolic anastomosis (ileal, colonic or both) was involved by the recurrence.  

The following CT signs were considered indicative of CD anastomotic recurrence [10]: mucosal hyper-enhancement of the neoterminal ileum/anastomotic wall compared to that of a presumably normal adjacent bowel loop; parietal stratification, which was defined as bilaminar (hyper-enhancing mucosa with hypoattenuating thickened submucosal layer) or trilaminar (enhanced outer serosal and inner mucosal layers with an interposed hypoattenuating submucosal layer) appearance of the neoterminal ileum/anastomotic wall; preanastomotic ileal wall thickening >3 mm; preanastomotic ileal dilatation, which was considered to be present when the maximal luminal diameter of the preanastomotic ileal loop was >25 mm; perianastomotic comb sign (engorged vasa recta separated more than usual and particularly parallel because of fibrofatty proliferation [13]). The presence of perianastomotic lymphadenopathies, fistulas and abscesses was also recorded. The CT appearance of a perianastomotic fistula was that of an extraenteric tract (hyper-enhancing, iso-enhancing, or hypo-enhancing relative to nearby bowel loops) with or without internal air or fluid [13]. Abscesses were defined as fluid collections with hyper-enhancing rims. The behavior of each CD recurrence (i.e. inflammatory, stricturing, penetrating) was further characterized according to the most evident CT features of the disease. In particular, preanastomotic ileal dilatation associated to anastomotic stenosis were considered CT findings of a stricturing recurrence, while the presence of perianastomotic abscesses and/or fistulas was used to identify a recurrence with penetrating complications. A CTe-WE examination was considered positive for CD recurrence when one or more of the above-mentioned CTe-WE findings were present. CTe-WEs were also analyzed searching for CD lesions involving other gastrointestinal sites distinct from the ileocolic anastomosis. Only when CD anastomotic recurrence was associated to another gastrointestinal CD-related inflammatory lesion, the latter one was defined SIL (synchronous inflammatory lesion).

A bowel segment was considered to be affected by a SIL when the following CT signs were present: bowel wall thickening (normal thickness of the wall of the small intestine and colon is 1–2 mm and 3 mm, respectively, when the lumen is distended [17]); mucosal hyperenhancement; multilayered appearance of the bowel wall (bilaminar or trilaminar); the comb sign [13]; increased density in the perienteric fat (creeping fat); stricture with prestenotic dilatation (a prestenotic dilatation was considered significant when the small bowel lumen exceeded a 25-mm diameter and the large bowel lumen exceeded a 50-mm diameter); extraenteric complications (fistulas and abscesses). SILs were further characterized according to their location (upper gastrointestinal tract – gastro-duodenal –, small bowel, colon, perianal) and behavior (i.e. inflammatory, stricturing, penetrating).

Reference Standard

The reference standard for determining the real status of the anastomoses was established by the study coordinator (GA. R.) according to the available data of each patient, and it was represented by endoscopy in 30 patients; results of the histopathologic analysis of surgical specimens in 7 patients; Crohn’s disease activity index (CDAI) in 5 patients; physical examination and laboratory tests (C-reactive protein [normal values <1 mg/dL], faecal markers of inflammation – faecal calprotectin [normal values >50 mg/L], lactoferrin [positive/negative] –) in the remaining 9 patients. Patients were considered to have recurrent CD when: 1) endoscopy or histopathologic analysis of surgical specimens revealed ulcerations in the neoterminal ileum (Rutgeerts score >1); 2) CDAI was >150 [15]; 3) clinical and laboratory data were consistent with recurrent inflammation (C-reactive protein values >1 mg/dL, fecal calprotectin values >50 mg/L, or positive fecal lactoferrin). CD recurrence was excluded when: 1) endoscopy or histopathologic findings indicated no abnormalities (Rutgeerts score = 0); 2) CDAI was <150; 3) clinical and laboratory data were not coherent with a recurrence.

When perianal disease was detected on CTe-WE images, it was confirmed using MRI and/or fistulography (performed within 2 months of the CTe-WE) as reference standards.

Statistical analysis

Descriptive statistics included frequencies and proportions for the binary and semiquantitative data. Quantitative data were expressed as mean ± standard deviation. Sensitivity, specificity, accuracy, positive and negative likelihood ratio, positive and negative predictive values – with corresponding 95% confidence intervals – of CTe-WE in the detection of anastomotic recurrence were calculated. The chi-square test was used to assess the presence of a significant association between the type of surgical ileocolic anastomosis (side-to-side, end-to-side or end-to-end anastomosis) and CD recurrence behavior (i.e. inflammatory, stricturing, penetrating). Patients were further divided in two groups: group A including “wide-lumen” isoperistaltic side-to-side (functional end-to-end) ileocolic anastomoses; group B including both end-to-side and end-to-end anastomoses. The Fisher's exact test was used to assess the presence of a significant difference in CD recurrence behavior between group A and B.