BACKGROUND: Considering the controversial results published in the literature concerning associations between human leukocyte antigens (HLA) and rheumatic fever (RF), the purpose of the present study was to investigate by means of cosegregation analysis the participation of HLA genes in susceptibility to RF. METHODS AND RESULTS: The sample reported here was composed of 51 affected and 66 healthy individuals belonging to 22 genetically informative families. The comparison (chi 2 goodness-of-fit test) of the observed numbers of identical-by-descent (IBD) HLA haplotypes among all affected individuals (siblings, cousins, and uncle/nephew and grandparent/grandchild type of pairs) with the expected ones under the assumption of independent segregation of HLA alleles and the presumptive RF susceptibility gene gave a value of P = .088. Since the number of subjects studied was relatively small and the rejection level obtained was near the usual .05 significance level, we calculated the expected HLA IBD scores in the 13 pairs of affected sibs of our sample for all possible frequencies of the presumptive RF susceptibility gene. This analysis allowed clear rejection of a recessive mode, considering susceptibility gene frequencies lower than 20%, whereas the observed values fitted very well a dominant mode of inheritance, with penetrance (K) values varying between 0.5 and 0.9 and a frequency of the susceptibility gene of at least 1%. CONCLUSIONS: The present data support the hypothesis of an RF susceptibility gene within or very near the HLA complex.
BACKGROUND: Considering the controversial results published in the literature concerning associations between human leukocyte antigens (HLA) and rheumatic fever (RF), the purpose of the present study was to investigate by means of cosegregation analysis the participation of HLA genes in susceptibility to RF. METHODS AND RESULTS: The sample reported here was composed of 51 affected and 66 healthy individuals belonging to 22 genetically informative families. The comparison (chi 2 goodness-of-fit test) of the observed numbers of identical-by-descent (IBD) HLA haplotypes among all affected individuals (siblings, cousins, and uncle/nephew and grandparent/grandchild type of pairs) with the expected ones under the assumption of independent segregation of HLA alleles and the presumptive RF susceptibility gene gave a value of P = .088. Since the number of subjects studied was relatively small and the rejection level obtained was near the usual .05 significance level, we calculated the expected HLA IBD scores in the 13 pairs of affected sibs of our sample for all possible frequencies of the presumptive RF susceptibility gene. This analysis allowed clear rejection of a recessive mode, considering susceptibility gene frequencies lower than 20%, whereas the observed values fitted very well a dominant mode of inheritance, with penetrance (K) values varying between 0.5 and 0.9 and a frequency of the susceptibility gene of at least 1%. CONCLUSIONS: The present data support the hypothesis of an RF susceptibility gene within or very near the HLA complex.