J Pediatr Gastroenterol Nutr. 2022 Oct 1;75(4):480-484. doi: 10.1097/MPG.0000000000003574.Epub 2022 Aug 18.

Åsa H Everhov ^1 2, Jonas F Ludvigsson ^3 4 5, Jacob Järås ², Rune Erichsen ^6 7, Lars Pedersen ⁶, Jonas Halfvarson ⁸, Johan Askling ², Anders Ekbom ², Henrik Toft Sørensen ⁶, Ola Olén ^1 2 9

Author information

¹From the Department of Clinical Science and Education Södersjukhuset, Karolinska Institutet, Stockholm, Sweden.

²the Clinical Epidemiology Division, Department of Medicine Solna, Karolinska Institutet, Stockholm, Sweden.

³the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.

⁴the Department of Pediatrics, Orebro University Hospital, Orebro, Sweden.

⁵the Department of Medicine, Columbia University College of Physicians and Surgeons, New York, NY.

⁶the Department of Clinical Epidemiology, Aarhus University Hospital and Aarhus University, Aarhus, Denmark.

⁷the Department of Surgery, Randers Regional Hospital, Randers, Denmark.

⁸the Department of Gastroenterology, Faculty of Medicine and Health, Örebro University, Örebro, Sweden.

⁹the Sachs' Children and Youth Hospital, Stockholm South General Hospital, Stockholm, Sweden.

Abstract

Through linkage of data from Danish and Swedish national registers we identified 6937 patients with childhood (<18 years)-onset Crohn disease (CD), 8514 patients with childhood-onset ulcerative colitis (UC) and up to 10 times as many matched (sex, age, residence) reference individuals 1969-2017. During follow-up to a median age of 27 (interquartile range = 21-39) years, 25 (0.36%) CD patients were diagnosed with colorectal cancer (CRC) versus 43 (0.06%) reference individuals, and 113 (1.33%) UC patients versus 45 (0.05%) reference individuals. The hazard ratio (HR) for CRC was 6.46 (95% CI = 3.95-10.6) in CD and 32.5 (95% CI = 23.0-45.9) in UC and increased with decreasing age at diagnosis. The HR for CRC was increased for all phenotypes, but with higher estimates for colonic CD [17.9 (95% CI = 7.43-43.3)] and UC with extensive/pancolitis [36.3 (95% CI = 22.8-57.8)]. The relative risk of CRC was increased for all phenotypes of childhood-onset inflammatory bowel disease. Age at onset may be considered an additional risk factor when implementing surveillance programs.