Abstract Background Antibiotics are widely used for perianal fistulising complications (PFC) of Crohn’s disease (CD). We sought to characterize antibiotic use for PFCs in a multicentre cohort of paediatric patients with CD. We hypothesized prolonged antibiotic therapy would not improve healing or reduce recurrence. Methods We conducted a retrospective study at 7 US paediatric sites. Patients (age 2-21yr) with new PFC from CD (1/1/2014–31/12/2019), and minimum 3 yr follow-up were included. PFC was defined as perianal/perirectal abscess and/or fistula. Antibiotic exposure was measured from PFC diagnosis through 3 yr; duration was calculated from all prescriptions by start-day and quantity dispensed. The primary outcome was any documented PFC healing within 1 yr. Secondary outcomes were PFC healing by imaging, and PFC recurrence. Analyses included chi-square, Kruskal-Wallis, and logistic regression. Results 291 patients with PFC were included (Table 1). 80 (27%) had abscess on imaging or exam, and 53 (66%) had drainage/seton procedure within 90 days. Antibiotics were used by 88% of those with and 70% of those without abscess (p = 0.002). Among those with abscess, 91% with and 81% without antibiotics had drainage procedures (p = 0.25). Antibiotic prescriptions over 3 yr ranged from 0-33 per patient: median cumulative duration 89 days (interquartile range IQR 36-236). 153 (53%) had PFC healing within 1 yr: 24 (16%) by symptom resolution alone, 90 (59%) on exam, 37 (24%) by imaging. 65% started anti-tumour necrosis factor (TNF) medications within 90 days. Healing by 1 yr was similar for those treated with and without antibiotics (56% vs 44%; p = 0.084). Healing was the same with and without abscess (57% vs 50%; p = 0.23), and among those with abscess the same with and without antibiotics (57% vs. 55%; p = 0.87). When requiring imaging (N = 125), there was no difference in antibiotic duration between those with and without healing (median 152, IQR 69-298 vs. 87, IQR 31-256; p = 0.103). In adjusted analyses (Table 2), duration of antibiotic use was not associated with PFC healing, but early anti-TNF improved odds of healing. Among all with PFC healing by 1 yr (N = 153), 38 (25%) developed recurrent PFCs. More antibiotics for 1st PFC was associated with PFC recurrence (median 108 days, IQR 35-369 vs no recurrence 30, IQR 0-114; p = 0.001). Conclusion Children with PFC on average received ∼3 mo antibiotics, yet there was no difference in PFC healing associated with duration. Prolonged antibiotic use may reflect severity. These findings may suggest limited benefit of prolonged antibiotic therapy for PFC healing in CD. Further studies must address patient symptoms and QOL to guide treatment decisions. Conflict of interest: Prof. Dr. Adler, Jeremy: My institution has received grant funding from the Agency for Healthcare Research and Quality to fund this current research, and grant funding from Janssen Research & Development LLC for studies of the SEMA-CD colonoscopy score. I have received no personal funding. Tribble, Alison: Grant funding to my institution from the Agency for Healthcare Research and Quality (AHRQ) to fund this current research. Grant from the Cystic Fibrosis Foundation. Grant support from another AHRQ grant. McDonnell, Erin: No conflict of interest Chavannes, Mallory: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research. I have a CDA from the Robert A Winn Foundations, which funding partially comes from a collaboration between the VCU Foundation and the BMS Foundation. I receive consultation fees from CapsoVision Inc. Egberg, Matthew: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research. Kandavel, Prashanthi: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research. Maltz, Ross: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research. Minar, Phillip: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research. Pasternak, Brad: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research. Kimberly, Armanda: No conflict of interest Manning, Lauren: No conflict of interest Sedjiu, Anita: No conflict of interest Valenzuela, Isabella: No conflict of interest Gayer, Christopher: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research McMahon, Lisa: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research Mollen, Kevin: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research Nwomeh, Benedict: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research Phillips, Michael: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research Gadepalli, Samir: Grant funding to my institution from the Agency for Healthcare Research and Quality to fund this current research
Adler et al. (2026) studied this question.