Recent advances in imaging, particularly the widespread use of magnetic resonance imaging (MRI), have been effective in preoperatively identifying pus cavities and fistulas that might otherwise be missed. Preoperative MRI findings have been shown to significantly influence surgical outcomes, reduce postoperative fistula recurrence, and improve anal control. MRI provides ideal images from sagittal, coronal, and transverse views to adequately visualize the perirectal muscles and accurately distinguish between fistulas and scars with different imaging signals. Preoperative MRI of anal fistulas has become the gold standard for evaluating complex anal fistulas in most medical centers.
I. Coils and sequences
All three coils of MRI, body coil, intracavitary coil, and phased array coil, can be used to evaluate complex anal fistulas. Although Halligan et al. first reported that better imaging results could be obtained by applying MRI body coils. However, in a subsequent series of studies, the authors showed that the clinical results of body coils were significantly lower than those of intracavitary and phased-array coils. The use of rectal endoluminal coils has provided more detailed imaging of anal fistulae, perirectal muscle injuries, and rectal tumors (Figure 1). desouza et al. reported that MRI rectal endoluminal coils could show the perirectal muscle and surrounding fat, and their diagnostic accuracy for anal fistulae and perirectal abscesses was 100%. However, endorectal coils are difficult to place in patients with anal stenosis or severe pain due to local inflammation, and Halligan et al. reported that endorectal coils could not be placed in 17% of patients. Also, endoluminal coils are expensive and limited by the number of uses. The improved signal-to-noise ratio and spatial resolution of surface phased array coils have resulted in significantly better imaging results and can achieve the same results as endoluminal coils for anal fistulae. However, the clinical application of surface phased array coils is simpler and more convenient. In fact, phased array coils have become the standard coils for clinical anal fistula MRI examinations. Since the normal lower and middle rectum is in a closed or semi-closed state, it is difficult to show the relationship between the anorectal rectum and the surrounding tissue structures, the authors used the placement of an intraluminal rectal water bladder to dilate the intestinal canal, which proved to be beneficial in showing the tissue structures around the lesion and improving the imaging contrast.
MRI scan sequences vary, but the more frequently reported sequence is the spin-echo sequence (SE) and the axial position is used as the routine examination method, which can assess the relationship between the anal fistula canal and the sphincter, and the coronal image can improve the display of the internal fistula opening, while the sagittal position has limited value.
①Spin-echo T1-weighted and T2-weighted sequences:T1-weighted sequences can show the external sphincter and anal levator muscle, and the anal fistula tube is low signal, but cannot show the anal canal mucosa, submucosa and internal sphincter.
The T1-weighted sequence is not very helpful for the diagnosis of anal fistula because the fistula and perianal structures are low-signal, and sometimes it is difficult to distinguish between the two on T1-weighted plain scan.
②Short time reversal recovery sequence (STIR)
The scan time of the STIR sequence is significantly shorter than that of the SE- T1-weighted parameter, but the STIR sequence has some shortcomings in the display of anal fistulae, because the STIR sequence is a more water-sensitive sequence. sequence is a more water-sensitive sequence and is not sensitive to inactive fistulas with little secretion and postoperative scar-forming fistulas.
(iii) Fast small-angle excitation imaging ((3D-FLASH)
This sequence is a gradient echo sequence, which takes layer block acquisition, no signal loss, shorter scan time than SE-T1-weighted and STIR, and high image resolution. The application of T2-weighted 3DFLASH sequence flat scan with enhanced image subtraction technique can improve the signal intensity of fistula and reduce the signal of surrounding soft tissues, so that the fistula can be displayed more prominently. This sequence combined with the STIR sequence can be used as a routine method for anal fistula examination, which not only improves the detection rate of anal fistula, but also significantly shortens the examination time.
Reading
The sensitivity of MR to fistula ducts and abscesses, the high definition anatomy, and the ability to show surgically relevant anatomic planes directly determine the success of preoperative diagnostic classification of anal fistulas with MRI. An accurate preoperative classification should include images of the associated fistula ducts and sphincters.
1. Primary canal
Active fistulas are filled with pus and granulation tissue and show up as long high-signal structures in T2-weighted or STIR sequences (Figure 3). In some patients with recurrent episodes or multiple surgeries, the fistula wall is correspondingly thickened, showing an active fistula tract encapsulated by a low-signal fibrous tissue wall. Occasionally, some high signal shadows are seen in these fibrous tissues, which are mainly due to tissue edema. Similarly, high signal shadows may appear outside the fistula or fibrous canal wall, which represents an inflammatory response in the adjacent tissue.
MRI is able to clearly demonstrate the external sphincter. It is a low-signal structure in T2-weighted or STIR sequences, with high-signal sciatic rectal fossa fat on the lateral side (Figure 1). Therefore, it is easy to analyze whether the fistula crosses the external sphincter or spans it. If the primary main tube is completely confined to the medial aspect of the external sphincter, this should be an intersphincteric fistula (Figure 4). Conversely, any evidence of a fistula in the colorectal fossa suggests a non-intersphincteric fistula. However, trans-sphincteric fistulas, suprasphincteric fistulas, and lateral sphincteric fistulas have similar MRI images and all breach the external sphincter. These three can only be distinguished from each other by the location of the internal orifice, as well as the lineage of the primary fistula (Figure 5, Figure 6, Figure 7).
2. Internal orifice
Regardless of the imaging morphology, the correct localization of the internal orifice is more difficult. How to determine the true location of the internal orifice and its height? According to the theory of adenogenic anal fistula, the majority of endografts are located at the posterior median dentition of the anal canal and most are located at 6 points of the posterior median truncus. However, even with the application of MRI intracavitary coils, the dentate line cannot be determined as a separate anatomical entity on MRI imaging and can only be assessed with other imaging landmarks. The dentate line is located approximately in the middle of the anal canal, usually midway between the superior border of the puborectalis muscle and the lower part of the external sphincter skin.
Both suprasphincter fistulas and external sphincter fistulas may cross the puborectalis muscle into the pelvic floor. However, the location of the internal orifice is completely different for both. Usually the internal opening of a suprasphincter fistula is located in the anal canal area, whereas a lateral sphincter fistula is located in the rectum (Figure 8, Figure 9). Because sphincter fistulas cross the external sphincter, they have typical features in cross section (Figure 5). However, for some patients, MRI cannot be traced along the fistula to the anal canal, in which case the possible site of the internal opening can only be rationally speculated based on the morphology of the fistula.
3. Branches and pus cavities
Another important aspect of MRI is its ability to accurately detect and localize the branched canal and residual pus cavity of an anal fistula. Branches and remnant cavities appear as high-signal structures present around the primary main tube in T2-weighted and STIR sequences, and intravenous application of control agents results in local signal enhancement. The most common branched form is the trans-sphincteric anal fistula, where the main tube crosses the external sphincter into the anal canal and the branched tube enters the top of the colorectal fossa.MRI is more important for the branched tubes above the levator muscle, which are not only difficult to detect but also extremely difficult to manage. For recurrent anal fistulas and Crohn’s fistulas, it is even more important to use MRI to diagnose complex branched canals and residual pus cavities.
Impact of preoperative MRI on surgical and treatment outcomes
Over the past 10 years, the development of MRI has revolutionized the treatment of complex anal fistulas. Preoperative MRI provides ideal images from sagittal, coronal and transverse views, fully revealing the perianal rectal muscles, and accurately distinguishing between fistula and scar with different imaging signals, providing more detailed imaging information for accurate preoperative localization of the internal orifice, branches and pus cavity, and the anatomy of the perianal rectal sphincter, enabling clinicians to fully define the extent of the lesion before surgery and develop the best treatment plan. Desouza et al [7] reported that MRI rectal endoluminal coil could show the anorectal muscle and surrounding fat, and its diagnostic accuracy for anal fistula and perirectal abscess was 100%. In a series of studies, Halligan and Stoker showed that MRI had a higher detection rate of distant branches and pus cavities than any other method (including intraoperative exploration), and that MRI-guided surgery for complex fistulas reduced postoperative recurrence rates by 75