Genetics in twins with Crohnʼs disease: Less pronounced than previously believed?

BACKGROUND: The influence of genetics in inflammatory bowel disease is emphasized by twin concordance. Previous studies have methodological limitations. The aims were to establish reliable concordance rates and compare phenotypic characteristics in concordant and discordant monozygotic pairs, anticipating the former reflects a genetically determined subgroup. METHODS: By re-running the Swedish twin registry with the Swedish hospital discharge register, observation time was extended. Diagnoses and phenotype were based on medical notes. Pairs with unknown zygosity and where both twins were not alive or not responding to the questionnaire were excluded. In all, 149 new twin pairs of the same sex, born 1909-1980 were identified. RESULTS: Of new pairs, 4/29 monozygotic, 0/38 dizygotic, and 0/1 twin pairs with unknown zygosity were concordant for Crohn's disease (CD). In ulcerative colitis (UC), 4/31 monozygotic, 4/48 dizygotic, and 0/1 twin pairs with unknown zygosity were concordant. New pairs were added to the original cohort. Restricting analyses to pairs born 1886-1958, the time period used in the original cohort, 9/33 monozygotic and 1/50 dizygotic pairs were concordant for CD (P = 0.008), 6/41 and 3/49, correspondingly, for UC (P = 0.29). There was a trend for concordant twins to have less colonic CD than discordant twins, 15% versus 35% (P = 0.09) in twins born 1886-1980. CONCLUSIONS: Previous twin studies have overestimated the influence of genetics in CD. A trend for phenotypic difference between concordant and discordant pairs was observed, suggesting that the clinical entity represents diseases with different pathophysiological backgrounds.