Imaging features of surgical complications of Crohn′s disease

Objective To investigate the imaging features of surgical complications of Crohn′s disease (CD). Methods The retrospective cross-sectional study was conducted. The clinical data of 128 CD patients with surgical complications who were admitted to the Sixth Affiliated Hospital of Sun Yat-sen University from January 2014 to December 2015 were collected. All the patients underwent computed tomography enterography (CTE), magnetic resonance enterography (MRE), magnetic resonance imaging (MRI) of anal tube and X-ray examination. The patients underwent surgical therapies after examinations. Observation indicators: (1) clinical characteristics of surgical complications of CD: occurrence of surgical complications of CD, correlation between preoperative complications and Montreal types of CD, (2) diagnostic accuracy of surgical complications and perianal lesions through CTE, MRE and MRI of anal tube, (3) imaging findings of preoperative complications of CD, (4) treatment and follow-up situations. Follow-up using telephone interview and outpatient examination was performed to detect the postoperative recovery and reoperations of patients up to March 2016. Count data were represented as frequency or percentage. The correlation between preoperative complications and Montreal types of CD was represented as the odds ratio (OR) and 95% confidence interval (CI). The features of diagnostic indexes were described as the sensitivity and specificity. Results (1) Clinical characteristics of preoperative complications of CD: all the 128 patients had preoperative complications, including 71 with anal fistula or anal fistula combined with perianal abscess, 26 with intestinal fistula, 24 with intestinal obstruction, 23 with abdominal abscess, 3 with digestive tract perforation, 1 with kidney stone and 1 with hydronephrosis. Of 128 patients, 12 had intestinal fistula combined with abdominal abscess, 1 had intestinal fistula combined with intestinal obstruction, 3 had intestinal fistula combined with abdominal abscess and intestinal obstruction, 1 had intestinal fistula combined with digestive tract perforation and 1 had intestinal fistula combined with hydronephrosis. There was a correlation between lesion location of CD and type of complications. Thirty-five of 51 patients had strictures or penetration-type lesions, with a correlation between strictures or penetration-type lesions and ileal CD [OR=6.23, 95% confidence interval (95%CI) : 2.86-13.61, P<0.05]. Fifty-six of 77 patients had anal fistula, showing a correlation between combined anal fistula and colonic lesions of CD (OR=6.40, 95%CI: 2.92-14.01, P<0.05). (2) Diagnostic accuracy of CTE, MRE and MRI of anal tube: with intraoperative exploration findings as the standard, the sensitivity and specificity of surgical complications of CD were 84% and 95% through CTE or MRE, and the sensitivity and specificity of anal fistula were 100% and 100% through MRI of anal tube. (3) Imaging findings of surgical complications of CD. ① Of 71 patients with anal fistula, 65 had 2 or more internal openings or fistula tract of complex anal fistula. The internal opening was a starting point of the fistula tract, and enhanced scans of MRI displayed punctate, shredded or small round abnormal strengthening signals between under mucous membrane of the anal canal and sphincter. The tubular and striped fistula tract was horizontal or vertical distribution, with a lightly high-high signal on T2 weighted-images (WI). The results of enhanced scans of MRI showed that there was an obvious homogeneous enhancement in the fistula tract or enhancement in the fistula tract wall, and no enhancement in the cavity of fistula tract. MRI findings in 38 patients combined with perianal abscess included a obvious high-signal on T2WI, and enhanced scans of MRI showed circular enhancement and no internal enhancement. ② Of 26 patients with intestinal fistula, 17 had intestinal fistula, imaging findings included multiple thickened intestinal walls and more obvious enhancement compared with normal intestinal canal. There was gathering and adhesions among intestinal canals, showing mash connections and petal-like changes. Fourteen patients had enterocutaneous fistula (6 combined with intestinal fistula and 1 patient combined with intestinal fistula and intestine-bladder fistula). Four patients had intestine-bladder fistula (2 combined with intestinal fistula and 1 combined with intestinal fistula and enterocutaneous fistula). One patient had intestine-vagina fistula. The results of CTE and MRE examinations showed that thickened intestinal canals and intestinal walls in the lesions, and shadows of intestinal canals communicated with the abdominal, bladder wall and vagina, with a high signal on T2WI and enhancement of intestinal wall by enhanced scan. The partial intestinal canals were physically close to other organs, without a connection between them, and anomalous enhancement or local pneumatosis among the adjacent organs were detected. ③ CTE findings of intestinal obstruction included constriction of intestinal canal combined with dilatation of proximal intestinal canal. There were 3 enhancement methods of thickened intestinal wall in 24 patients with intestine obstruction. Imaging findings of 12 patients included enhancement in the intestinal mucosa and no enhancement in the submucosa and muscularis mucosa. Imaging findings of 4 patients included enhancement of intestinal mucosa and muscularis mucosa and no enhancement in the submucosa. Imaging findings of 8 patients included homogenous and heterogeneous enhancements in the intestinal walls. ④ Twenty-three patients were complicated with abdominal abscess, including 15 combined with intestine fistula. The scans of CTE showed that there was a round-like or oval mass in the abdomen, with a high signal on T2WI, fluid-dominated inflammatory exudation around the mesentery, the enhancement of annular wall of mesentery and no enhancement of pus within the mesentery. ⑤ Three patients were combined with digestive tract perforation, including 1 combined with intestine fistula. CTE and X-ray detections showed there was a shadow of free gas in the intestinal mesentery and under abdominal diaphragm. ⑥ Two patients were combined with kidney stone and hydronephrosis. X-ray findings of kidney stone included the deposition of multiple and sharp-edged dense shadows within the renal calices. CTE findings of hydronephrosis included inflammatory thickening of ureteric wall with proximal ureter dilatation. (4) Treatment and follow-up situations: 128 patients underwent successful operation and were followed up for 4-27 months. Of 10 patients undergoing reoperations due to postoperative complications, MRI detection of 7 patients with recurrence of anal fistula showed fistula tract or abscess located at the previous loci was incompletely healed or progressed, morphous and location of lesions were roughly the same as the preoperative situations. The scans of CTE in 2 patients with anastomotic stricture showed that there were the thickening of intestinal wall and obstruction and dilatation at the proximal anastomotic-site. The enhanced scan of CTE in 1 patient with anastomotic fistula showed that there were irregularly cavity-like lesion beside the metal anastomotic ring, and effusion was seen within the lesions, with an edge enhancement. The other 118 patients recovered well without intestinal fistula or intestinal obstruction on CTE or MRE examination. Conclusions Anal fistula is the most common surgical complication of CD, and intestinal fistula, intestinal obstruction and abdominal abscess are also relatively common. The early postoperative complications consist of the recurrence of anal fistula. Location of lesions in CD is associated with the type of complications. CTE or MRE and anal MRI findings have different imaging characteristics for CD combined with different complications, with a certain value in the assessment of abdominal and perianal complications. Key words: Crohn′s disease; Surgical complications; Tomography, X-ray computed; Magnetic resonance imaging

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.