Search in the Abstract Database

Abstracts Search 2019

P232 Oesophageal Crohn’s disease: diagnosis and outcome of an ECCO-CONFER case series

R. Rodrigues*1, M. Sladek2, K. Katsanos3, C. J. Van der Woude4, J. Wei5, N. Teich6, P. Ellu7, E. Savarino8, M. Chaparro9, D. Beaton10, A. M. Oliveira11, M. Fragaki12, A. Bar-Gil Shitrit13, L. Ramos14, K. Karmiris12

1Instituto Português de Oncologia de Lisboa, Gastroenterology, Lisbon, Portugal, 2Jagiellonian University Medical College, Pediatrics, Gastroenterology and Nutrition, Krakow, Poland, 3School of Health Sciences and University Hospital of Ioannina, Gastroenterology, Ioannina, Greece, 4Erasmus Medical Center, Gastroenterology and Hepatology, Rotterdam, The Netherlands, 5Nanjing University Affiliated Jinling Hospital, Gastroenterology and Hepatology, Nanjing, China, 6Gastroenterology Outpatients Clinic, Leipzig, Germany, 7Mater Dei Hospital, Medicine, Msida, Malta, 8University of Padua, Surgery, Oncology and Gastroenterology – DiSCOG, Padova, Italy, 9Hospital Universitario de La Princesa, Gastroenterology, Madrid, Spain, 10Royal Victoria Infirmary, Gastroenterology, Newcastle Upon Tyne, UK, 11Hospital Prof. Doutor Fernando Fonseca, Gastroenterology, Amadora, Portugal, 12Venizeleio General Hospital, Gastroenterology, Heraklion, Greece, 13Digestive diseases Institute, Shaare Zedek Medical Center, Jerusalem, Israel, 14Hospital Universitario de Canarias, Tenerife, Spain

Background

Crohn’s disease (CD) can involve any part of the gastrointestinal tract. We aimed to characterise clinical, endoscopic, histological features and treatment outcomes of CD patients with oesophageal involvement.

Methods

This was a European Crohn’s and Colitis Organization (ECCO) retrospective observational study performed as part of CONFER [COllaborative Network For Exceptionally Rare case reports] project. A call was made to all ECCO members to report CD patients with oesophageal involvement. Clinical data were recorded in a standardised case report form.

Results

Forty patients were reported [24 males, mean age at oesophageal CD diagnosis: 22 years (10–46) and mean time of follow-up: 61 (3–240) months]. Oesophageal involvement was established at CD diagnosis in 25 patients (62.5%) and during follow-up in 15. CD was exclusively located in the oesophagus in 2 patients while in the rest, small bowel involvement was present in 81.5%, colonic in 73.6% and gastric in 50%. Twenty-three patients (57.5%) presented with non-stricturing, non-penetrating behaviour, 18 (45%) had perianal disease, and 12 (30%) extra intestinal manifestations. Nine patients (22.5%) were asymptomatic at oesophageal disease diagnosis. Distal oesophagus was the most common site of involvement (62.5%). Oesophageal strictures were present in six patients and fistulising oesophageal disease in one. Eight patients exhibited granulomas on biopsies. Medical treatment: proton-pump inhibitors (PPIs, 87.5%), steroids (52.5%), thiopurines (52.5%), anti-TNFs (52.5%) and exclusive enteral nutrition (20%). Three patients underwent endoscopic dilation for symptomatic strictures and none oesophageal surgery. Oesophageal disease diagnosed during follow-up for CD (15/40) resulted in treatment modifications in 9 patents. Remission or improvement of oesophageal disease was seen in 38/40 (95%) patients after a mean time of 7 (1–18) months while in overall CD these outcomes were reported in 29/38 patients (76.3%). Follow-up endoscopy was performed in 30/40 patients and 27/30 (90%) achieved mucosal healing.

Conclusion

Oesophageal CD involvement can be detected either at CD diagnosis or during follow-up, manifesting as the only site of CD location in rare cases. Phenotypic characteristics are similar to those of other sites of involvement and diagnosis can be done even during overall CD remission. Optimal treatment is conservative but not consensual depending also on extra oesophageal sites of involvement, with PPIs administered in the majority of patients and treatment modifications occurring infrequently when diagnosed at a later phase. These results should be interpreted with caution due to the small sample size and the design of this project.