Optimizing MRI for Perianal Fistula Assessment: A Head-to-Head Comparison of Post-Contrast 3D LAVA Sequence and Post-Contrast T1 Weighted Fast Spin Echo Fat Saturation Sequences in Identifying Internal Openings

Background: Perianal fistulas have been evaluated with a variety of Magnetic Resonance Imaging (MRI) sequences, each with advantages and limitations. Given the importance of identifying the internal opening for surgical planning and the lack of consensus on the optimal MRI sequence, a direct comparison of post-contrast T1-weighted Fast Spin-Echo Fat Saturated (T1FS) and post-contrast 3D liver acquisition with volume acceleration (LAVA) sequences is warranted. Objective: To compare the post-contrast T1-weighted Fast Spin-Echo Fat Saturated (T1FS) and post-contrast 3D liver acquisition with volume acceleration (LAVA) sequences to identify the internal openings of perianal fistulas using MRI. Methods: This retrospective study was conducted from February 2024 to January 2025 in the Radiology Department of the Rehman Medical Institute in Peshawar. A total of 44 patients with perianal fistulas, aged 26 to 70, were selected based on their perianal discharge. 2D T2 PROPELLER, 2D T2 PROPELLER with fat saturation, 2D T1 Fast spin echo fat saturation, and 2D post-contrast T1 Fast spin echo fat saturation (T1FS) are among the MRI techniques that are frequently used to treat perianal fistulas. In the axial plane, additional imaging was performed using post-contrast 3D liver acquisition with volume acceleration (LAVA) and chemical-shift technique. An impartial expert evaluated each sequence independently to determine the internal opening. Clock positions, 12 o’clock anterior and 6 o’clock posterior, were used to determine the radial location of the internal opening. Results: There were 42 inter-sphincteric and 2 trans-sphincteric fistula tracts discovered. There was only one instance of a secondary ramification with a distinct internal opening. A total of 45 internal openings were identified, of which 7 were visible only on post-contrast LAVA (15.6%) and 6 on post-contrast T1FS (13.3%). On both post-contrast T1FS and LAVA, 32 internal openings were visible (71.1%). Post-contrast T1FS, with 13.3% of openings detected only on this sequence, and post-contrast LAVA, with 15.5% of internal openings visualized only on this sequence, were the best sequences for depicting internal openings. A comparison between the post-contrast T1FS and LAVA sequences revealed that both sequences were required for optimal outcomes. Conclusion: We found that a fair percentage of internal openings were detected by only one of the sequences, indicating that both post-contrast T1FS and LAVA sequences were required for better detection of internal openings.