Abstract
Introduction: Magnetic resonance imaging (MRI) is increasingly recommended by international opinion for the evaluation of perianal fistula, but whether MRI is associated with improved operative outcomes remains unclear. We aimed to evaluate the association between MRI utilization and operative burden in patients with perianal fistula.
Methods: This retrospective cohort study included all adult patients who underwent operative management for perianal fistula between 2015 and 2021 at Wellington Hospital in Aotearoa New Zealand. Fistulae were classified as simple (intersphincteric or low trans-sphincteric crossing <30% of the external anal sphincter) or complex (high trans-sphincteric involvement, recurrent, inflammatory bowel disease-associated, anterior location in females, or involving multiple tracts). The primary outcome was the total number of operations required until healing (minimum 12 months without symptom recurrence). Stratified analyses based on the timing of MRI and comparisons with versus without MRI were performed. A multivariable regression analysis was also undertaken to further account for confounding.
Results: In total, 232 participants were included, of which 86 (37%) were investigated with MRI during their clinical course, with the majority in complex fistula (76%). Patients investigated with MRI underwent more operations than those managed without MRI (median 4 versus 2; P < 0.001). In stratified univariate analyses, MRI use for complex fistulae was associated with fewer operations to achieve healing (median 2 versus 3; P < 0.001), and preoperative MRI was associated with fewer operations compared with its use later in the management pathway (median 2 versus 3 versus 4; P < 0.05). In the adjusted analysis, however, operative burden was mostly explained by fistula complexity, whereas MRI use was not independently associated with fewer operations required to achieve healing.
Conclusions: MRI was most frequently used in complex perianal fistula. While unadjusted analysis demonstrated an association between MRI utility and operative burden when used preoperatively, adjusted analyses demonstrated that operative burden was instead driven predominantly by fistula complexity.