Genetic factors undoubtedly play a part at least in some forms of juvenile idiopathic arthritis (JIA). Yet it is commonly believed that the risk to a sibling of a patient with JIA is not particularly strong as shown by the rarity of reported multicase families.1

Multicase JIA families have been traced systematically at the Rheumatism Foundation Hospital in Heinola, Finland, over a period of 15 years. A total of 41 families with 88 affected siblings (34 boys, 54 girls) were found fulfilling the Durban criteria for JIA.2In 60 (68%) of these 88 patients the disease was pauciarticular and, in most instances, it ran a mild course. The mean age at JIA diagnosis was 4.6 years.

Over the same period, eight sets of MZ twins were found; two twin pairs were concordant for JIA.

The incidence of JIA in Finland is 14/100 000 in the paediatric population.3 This corresponds to about 150 cases a year. Because the mean age at diagnosis was 7 years, the total number of cases of JIA in the paediatric population is about 1200, and the prevalence of the disease is about 1 per 1000, in a population of 1.2 million under 16 years of age in the country. The incidence and prevalence figures are similar to those reported from other countries for which data are available.4 For instance, it has been estimated that in the United States with a population 50 times larger than that of Finland there are about 71 000 cases of JIA.1

Over the 15 year period of study 2300 patients were seen at the paediatric department of the Rheumatism Foundation Hospital. Of these, about 90% had JIA. It can thus be estimated that the population of JIA cases from which all the recorded multicase families were derived amounted to about 2000. This corresponds to 60% of JIA cases in the country.

There are about four MZ births per 1000 in the population. Thus our eight MZ pairs indicated that most, if not all, such pairs had been traced in the study group. Considering the population prevalence of JIA (1 per 1000) the concordance rate of 25% implies a relative risk of about 250 for a MZ twin. In adult rheumatoid arthritis the risk was found to be nine.5

The average number of children in Finnish families is 1.8; 45% of families have only one child, 38% have two children, 13% have three children, and 4% four or more children. In adult rheumatoid arthritis the relative risk for a sibling is about three. In view of the population prevalence of JIA, this finding of 41 multicase families in the basic population of 2000 indicates that the relative risk in JIA is much higher.

These findings cannot perhaps be extended to other populations, though there is no reason to believe that this would not be the case. In any event, genome wide screening of Finnish patients offers a promise that new susceptibility loci outside the HLA region may be found.