Statistics from Altmetric.com
Editor,—I was interested to read the review on the role of inherited factors in inflammatory bowel disease by Satsangi et al (Gut1997;40:572–4), but dismayed by the evidence provided. They claim that twin pairs provided strong evidence for genetic predisposition and that the most pertinent was that reported by Tysket al using data from the Swedish Twin Registry.1 However, the claim for ulcerative colitis rests on a single twin of the 50 000 registered. The probands were identified as hospitalised cases in Sweden. The 50 pairs of twins where one twin had suspected ulcerative colitis were whittled down to 36 pairs: 16 monozygotic and 20 dizygotic or of unknown zygosity. No hospitalised pairs were found. Twins of probands were interviewed and the occurrence of any gastrointestinal complaint was thoroughly discussed in order not to miss a mild non-hospitalised case. In spite of this approach only one twin of a proband could be claimed as a case of ulcerative colitis and he was one of the monozygotic group. This pair had only distal ulcerative colitis. From this one twin a heritability coefficient of 0.53 was calculated (with 95% confidence interval 0.24 to 0.82) and not 0.24 to 1.0 as stated by Satsangiet al. Thus, the 53% heritability claimed in both papers is based on one man with mild ulcerative colitis. The definition of his disease differed from that in the probands and the authors admit to deficiencies in their selection procedure. Tysk et al also admit that the prevalence of ulcerative colitis in the twins differs from that of the general population in Stockholm (mostly singletons). Twin studies cannot inform about disease in singletons if the prevalences are not the same. As the mean observation period was almost 20 years, it is unlikely that many more healthy twins will develop ulcerative colitis.1 With 94% discordance in monozygotic twins, it is difficult to accept that most ulcerative colitis is inherited in either twins or singletons.
Editor,—We are grateful to Dr Harding for his comments, and we share his reservations regarding interpretation of twin studies in common diseases. Indeed, we have pointed out on a number of occasions that the genetic contribution in ulcerative colitis is less strong than that in Crohn’s disease. Nevertheless, the model of disease inheritance which is most consistent with the epidemiological and molecular genetic data currently available is that Crohn’s disease and ulcerative colitis are related polygenic disorders. The twin data are supported by epidemiological studies of familial prevalence from Scandinavia, the UK, and the USA. In recent years these epidemiological data have been supplemented by molecular genetic studies from a number of independent investigators. These studies have provided evidence for the presence of a number of genetic determinants of susceptibility and disease behaviour in ulcerative colitis. There is great optimism that these studies will result in an advance in the understanding of the pathogeneis of these common disorders. We stongly agree that both environmental as well as genetic factors are important in the pathogenesis of ulcerative colitis.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.