Letter: Re-Examining Seton Efficacy in Perianal Crohn's Disease—Critical Considerations for Outcome Measurement and Clinical Interpretation. Authors' Reply

We thank Dr. Liu and colleagues for their interest in our study [1, 2]. Three important considerations were raised in their letter that require further explanation.

We did not capture seton duration prior to initiating anti-TNF therapy. Our study reflected real-world practice. Therefore, it is likely that there was variability in the duration of setons prior to initiating anti-TNF therapy. Contrary to Dr. Liu and colleagues' recommendation to delay anti-TNF for greater than 8 weeks after seton placement, we recommend against this practice for a number of reasons. First, delays in initiating anti-TNF therapy are associated with worsening luminal Crohn's disease [3] and persistent perianal fistula Crohn's disease (PFCD) activity (OR, 2.98; 95% CI, 1.30–6.80) [4]. Second, animal models have demonstrated that setons promote fistula tract epithelialization, a phenomenon believed to prevent fistula closure [5, 6]. Finally, surgical guidelines recommend initiating biologic therapy soon after the control of sepsis to avoid unnecessary delays in healing [7]. Therefore, we believe that anti-TNF therapy should be initiated as soon as possible for PFCD, after source control has been achieved.

We did not omit the key aspects of seton technical variables influencing seton efficacy. Indeed, our propensity score-weighted analysis adjusted for supra/transsphincteric anatomy. This was chosen based on previous work where we demonstrated a trend towards anti-TNF treatment failure in patients with supra/transsphincteric anatomy (OR, 2.28; 95% CI, 0.97–5.35) [4]. In contrast, we did not adjust for transsphincteric anatomy (versus intersphincteric anatomy) since this was not associated with anti-TNF treatment failure (OR,1.04; 95% CI, 0.57–1.90) [4]. Importantly, cutting setons were not used in our study. This is in accordance with guidelines that recommend against their use for complex PFCD due to the risk of anal sphincter injury [7, 8]. Similar to our practice, 90% of surgeons from the United Kingdom reported they would never use cutting setons [9].

We agree with Dr. Liu and colleagues that, when assessed clinically, fistula remission lacks objectivity. This is why we chose major adverse fistula outcomes, a composite of exams under anaesthesia, hospitalisation for PFCD and faecal diversion for PFCD, as our primary outcome. It was our perception that these elements represented the most objective measures to capture major adverse outcomes related to PFCD that could be reasonably obtained retrospectively. We also agree with Dr. Liu and colleagues that patient-centric outcome measures such as patient-reported outcomes and quality-of-life measures would have added valuable information to our study. These will be important to assess in prospective studies since it has yet to be established in a clinical trial whether setons truly improve symptoms and quality of life above and beyond what is achieved by anti-TNF therapy alone.