Commentary: Juvenile idiopathic arthritis—issues of definition and causation

immunological abnormality in the child, which later leads to arthritis. Second the role of maternal smoking could be via low birth weight and small-for-gestational age. This in turn might lead to an increased susceptibility to childhood infections, which might trigger arthritis. Jaakola and Gissler considered this possibility but felt it could not explain the size of the observed effect. Third the maternal smoking might be a marker for what is actually the true risk factor. For example, if a mother smokes during pregnancy she is also likely to smoke after pregnancy— and it may be the exposure to environmental smoke as an infant that is the true culprit. Smoking is an important contributor to autoimmunity.6 Smokers have been reported to have abnormalities in T-lymphocyte function, a reduction in the number of natural killer cells and abnormalities in humoral and cellular immunity. This might be conveyed to the child in utero or might be acquired by passive smoking. Finally, an earlier study showed that the risk of JIA was 40% lower in breast-fed than in non-breast-fed children.7 It is possible that mothers who smoke are less likely to breast-feed and so this might be the link between smoking and arthritis onset. In conclusion further studies are needed to confirm these results in other populations and to try and understand the mechanism underlying the association.