Background: Treatment paradigms for Crohn's disease with perianal fistula (CD-PAF) are still evolving. We aimed to study the impact of multidisciplinary multimodality treatment approach in patients with CD-PAF on the recurrence rates of fistula and the need for re-interventions. We also aimed to study the predictive factors for the need for re-intervention
Methods: This was a multinational multicentre (11 centres) retrospective cohort study with data collected in CD patients with CD-PAF from 2010 to 2015.Multidisciplinary multimodality approach was defined as using a combination of medical treatments (antibiotics, immunomodulators, and biologics) along with surgical approach (examination under anaesthesia (EUA) +/− Seton drainage) at diagnosis. Univariate was analysis done for variables impacting fistula recurrence and re-intervention a logistic regression adjusting for age to identify significant predictors of re-intervention.
Results: 253 adult onset CD-PAF patients were included (161 M, 92 F). 53% had complex fistula. 70% of the patients had EUA at presentation with 136 patients (53.8%) needing a median of 1 Seton insertion (range 1–6.84% of the patients had anti TNF therapy. There was significant difference in fistula healing rates between simpleand complex fistulae (complete healing 60% vs 41%, p=0.015.52% of patients who received multimodality treatment had complete fistula healing. 27% of simple fistula and 40.3% of the complex patients had recurrent fistula needing re-intervention. There was a significant difference in the need for re-intervention based on fistula healing with 22% of those with complete healing needing repeat surgery compared to 49% with partial healing and 71% in those with no healing (p≤0.001. Only 26% of the 141 patients having multidisciplinary multimodal treatment needed surgical re-intervention when compared to 59% without this (p≤0.001. Univariate analysis showed complex (p=0.008), absence of multidisciplinary approach (p≤0.001, EUA (p=0.005), combined immunosuppression (p=0.032, presence of proctitis (p≤0.001) as factors impacting need for re-intervention but there was no impact of age, gender, smoking status, mode of presentation, Montreal class, presence of anal stenosis and thiopurine use alone. On logistic regression, absence of multi-disciplinary approach (OR 2.8,95% CI: 1.4–5.6)and presence of proctitis (OR2.2, 95% CI: 1.2, 3.9) were predictors for re-intervention.
Conclusions: This large multicentre cohort study describes outcomes in CD-PAF patients receiving multidisciplinary multimodality treatment approach. In this cohort, complete fistula healing rates were higher and the recurrence rates lower than previously reported. Presence of proctitis and lack of multidisciplinary approach are predictors for recurrence and re-intervention for CD-PAF.