MRI- enema for the assessment of pelvic intestinal anastomotic integrity

Aim: Anastomotic leak causes significant morbidity for patients undergoing pelvic intes tinal surgery. Fluoroscopic assessment of anastomotic integrity using water- soluble con trast enema (WSCE) is of questionable benefit over examination alone. We hypothesized that MRI- enema may be more accurate. The aim of this study was to compare MRI- enema with fluoroscopic WSCE. Method: Patients referred for WSCE with pelvic intestinal anastomosis and defunction ing ileostomy (including patients with suspected or known leaks) were invited to partici -pate. WSCE and MRI- enema were undertaken within 48 h of each other. MRI sequences were performed before, during and immediately after the introduction of 400 ml of 1% gadolinium contrast solution per anus. MRI examinations were reported to protocol by two blinded gastrointestinal radiologists.


INTRODUC TI ON
Anastomotic leak is 'a communication between the intra-and extraluminal compartments owing to a defect of the integrity of the intestinal wall at the anastomosis between the colon and rectum or colon and anus' [1]. Leaks have a significant and detrimental impact on patient outcome after colorectal surgery. Following restorative proctocolectomy, sepsis complicating anastomotic leak is a major predictor of ileoanal pouch failure [2][3][4]. In rectal cancer surgery, leak increases the likelihood of low anterior resection syndrome (LARS) and local recurrence and delays adjuvant therapy [5,6]. In order to minimize the consequences of a potential anastomotic leak, a defunctioning ileostomy is frequently formed in patients undergoing low pelvic anastomosis.
Several imaging investigations are available to help clinicians assess anastomotic integrity prior to the reversal of ileostomy and restoration of intestinal continuity [7,8]. Fluoroscopy with water-soluble contrast enema (WSCE) is widely available and most commonly used to assess the integrity of the anastomosis, colonic cul-de-sac (when a side-to-end anastomosis is formed) or seam of an ileoanal pouch [9,10]. A leak is diagnosed by the examining radiologist identifying extraluminal iodinated contrast first introduced through the anus via a catheter. WSCE utilizes x-ray fluoroscopy to investigate the flow of contrast through the anastomosis with representative fluoroscopic images acquired, supplemented by higher quality radiographs when required [11]. WSCE using x-ray fluoroscopy has a number of limitations, including limited accuracy with a sensitivity of 78% and a positive predictive value of 62% [10], use of ionizing radiation (particularly relevant for younger patients), limited two-dimensional images, paucity of anatomical information and diminishing fluoroscopy capacity in the UK National Health Service (NHS) and North America, with associated decrease in fluoroscopic skills and interpretative expertise [12]. Unsurprisingly, therefore, some surgeons consider clinical assessment superior and sufficient, questioning the need for routine WSCE [10,11,13]. Where WSCE has been requested, presacral widening is a relatively frequent finding, with MRI or CT required to interrogate the cause. Consequently, the authors of this study hypothesized that an MRI-enema technique could feasibly offer a more accurate assessment of anastomotic integrity and provide additional relevant information compared with WSCE and clinical examination alone.

ME THOD
Regional ethical approval was obtained (REC reference 17/LO/0629, IRAS ID 191029) and the study registered with ClinicalTrials.gov (ID NCT04719169). The study did not meet the criteria for reporting according to the Standards for Reporting Diagnostic accuracy studies (STARD) guideline as calculating sensitivity and specificity was not appropriate in this cohort [14].

Inclusion criteria
Adult patients with diverting ileostomy created after low to mid rectal resection and pelvic anastomosis (ileoanal pouch reconstruction or colorectal/coloanal anastomosis) were recruited prospectively and consecutively from the waiting list for fluoroscopic WSCE assessment of anastomotic integrity between September 2017 and March 2019. Recruitment was in a single NHS hospital setting and patients were excluded if they had a contraindication to MRI.
Patients were included whether they had undergone uncomplicated primary surgery or had a known leak undergoing surveillance.

Fluoroscopic WSCE protocol
Fluoroscopic WSCE was performed according to standard practice in the host institution (see the WSCE protocol in Appendix S1 in the Supporting Information).

What does this paper add to the literature?
This paper is the first to describe MRI-enema as a feasible alternative radiological technique for assessing pelvic intestinal anastomoses. Compared with water-soluble contrast enema, MRI-enema offers more detailed assessment of anastomotic leak, with additional pelvic information and equivalent patient experience.

MRI protocol
Multiplanar MRI sequences of the pelvis were acquired before, during (dynamic sequences) and after introduction of dilute MRI contrast, introduced through a Foley catheter inserted through the anus. Buscopan® (hyoscine butylbromide, Sanofi, UK) was used routinely to limit artefacts from bowel peristalsis. The detailed MRI protocol is given in Appendix S1.
Participants were invited to complete a nonvalidated

Postimaging outcome
The median time from radiological assessment to completion of   Table 2).

Individual case studies
Patient J One male patient developed pelvic sepsis diagnosed on CT following ileoanal pouch surgery for FAP ( Table 2

Patient experience
Both investigations were similarly well tolerated, with no statistically significant difference between Likert scores for overall experience and its component parts (Table 4).

DISCUSS ION
This study has shown that MRI-enema is technically feasible and compares favourably to WSCE for detection and characterization of anastomotic leak in our patient cohort.
MRI-enema detected all the leaks present in our patient cohort and also identified a leak not reported or identifiable on WSCE. This finding was confirmed by endoscopy and clinical EUA. A further advantage of MRI is avoidance of ionizing radiation, which is particularly relevant to younger patients with chronic disease (such as inflammatory bowel disease) who may undergo frequent imaging and regular exposure to ionizing radiation for disease assessment and monitoring.
Limitations of MRI-enema include the requirement for Buscopan™ and MRI contraindications, including patient claustrophobia and metallic implants.
CT with rectal contrast is described for the detection of anastomotic leak in the emergency postoperative setting but is rarely utilized for routine assessment before ileostomy reversal [7,15,16].
A comparative study of CT with rectal contrast enema (CT RCE) and MRI-enema would be of interest. However, a 2017 study showed that even high-quality CT RCE used in routine practice in a high-volume institution has relatively poor sensitivity (72%) for diagnosing leaks, albeit this included both rectal and abdominal anastomoses [17].
Limitations of this study include the small number of patients with associated recruitment bias and the single-centre design, which are in keeping with a developmental study of a new technique rather than a diagnostic impact assessment. Although MRIenema provides inherently more anatomical information than fluoroscopy, a more detailed comparison of performance characteristics is required before we can recommend its wider implementation. To assist such research, we recommend a consensus amongst expert groups to agree an optimal reference standard for anastomotic leak [18,19].

ACK N OWLED G EM ENTS
The research team would like to thank the managerial and MR radi- analysed the results and drafted the manuscript. All co-authors approved the final version of the manuscript before submission.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

E TH I C A L S TATEM ENT
Regional ethical approval was obtained (REC reference 17/LO/0629, IRAS ID 191029) and the study registered with ClinicalTrials.gov (ID NCT04719169).