AIM: Chronic peri-pouch sepsis (CPPS) may be mistaken for antibiotic-dependent or refractory primary idiopathic pouchitis (ADRP), but requires different treatment such as drainage. The study aimed to identify the prevalence of CPPS in patients thought to have ADRP. The secondary aims were to identify any specific features on pouchoscopy suggesting CPPS and to determine the results of treatment for CPPS. METHOD: The records of patients who had been treated for ADRP between March 2006 and June 2015 were reviewed retrospectively. Only those with endoscopic evidence of pouch inflammation who had also undergone magnetic resonance imaging (MRI) of the pelvis were included. The findings on pouchoscopy and the outcome of treatment were determined. RESULTS: Sixty-eight (43 (63%) male) patients were identified with apparent ADRP between March 2006 and June 2015. MRI of the pelvis showed CPPS in 26 (38%). In those with CPPS, the inflammation was more often located in the upper pouch alone (15%) compared with patients without CPPS (0%) (p = 0.0184). Examination under anaesthesia was performed in 13 of those with CPPS. In five a collection was identified and drained; symptoms improved in only one (4%). Eighteen patients (69%) remained on antibiotics and 7 (27%) had a defunctioning stoma or underwent pouch excision. CONCLUSION: In patients thought to have ADRP, 38% had CPPS on MRI. There was no clinically relevant specific feature on pouchoscopy suggestive of CPPS. The possibility of CPPS should be considered early in patients with apparent ADRP and pelvic MRI performed. This might lead to earlier detection of CPPS and appropriate treatment. This article is protected by copyright. All rights reserved.