The father figure in coeliac disease.

The previous story It is surprising that although clear reports were given in the pilot work of Ferguson and colleagues, 1 Farthing and colleagues , 2 and others about 20 years ago, coeliac disease (CD) has not yet gained popularity among obstetricians. In the subsequent two decades, many epidemiological studies clearly showed that it is a very common disease, that it aVects women more than men, and that it has to be considered in relation to reproductive function. The actual story In the past few years several groups have shown that untreated CD is an important cause of abortion, poor outcome of pregnancy, and intrauterine growth retardation. 3 4 Most authors agree that malnutrition could not have explained the reproductive diYculties of many, now most, coeliac women. Martinelli and colleagues 5 recently found that 1 in every 70 pregnant women admitted to a major city hospital suVered from untreated CD: 70% had a poor outcome of pregnancy, and 8/9 women had a second healthy baby after one year on a gluten free diet. The issue But the paper of Ludvigsson and Ludvigsson in this issue of Gut 6 raises new questions on a well described story: an unfavourable neonatal outcome was not only associated with maternal CD but also with paternal CD, which is a new finding (see page 169). Infants of coeliac mothers weighed 222 g less than the population average, and infants of coeliac fathers weighed 266 g less than the population average. The risk of a low birth weight baby to coeliac fathers was five times higher than that in the general population (11% v 2.5%). The paper Ludvigsson and Ludvigsson 6 report on a large (10 597 births) population based study with the drawback that it was limited to pregnant women who delivered a live baby: the results, although impressive, tend to underestimate the actual prevalence of untreated CD and the damage induced by undiagnosed disease. There is little doubt that more women who did not become pregnant, had an abortion , or who did not deliver a live baby need to be identified and inclusion of these women would change the results. Similarly, men who could not father a child need to be identified. The other debatable point is the assumption that mothers with CD were all " treated " , which means free from gluten. Unfavourable neonatal outcome would then have not been corrected, …