The autoimmune nature of multiple sclerosis introduces cytokine genes as logical candidates for the loci determining susceptibility to the disease, and/or influencing disease progression. Working on this principle, several groups have investigated the relevance of polymorphism in the interleukin 1 receptor antagonist gene (IL1RN) but with conflicting results. In an effort to clarify this situation, we typed the functionally significant variable number of tandem repeat (VNTR) polymorphism from intron 2 of IL1RN in 536 simplex families with multiple sclerosis. In order to improve the information extracted from these families, we also typed a closely mapped single nucleotide polymorphism (SNP) from the promoter of IL1B (the gene for IL-1beta). Disease associations were assessed by transmission disequilibrium testing (TDT), alone and after haplotype construction. There was highly significant (P</=2.48.10(-16)) linkage disequilibrium (LD) between the two polymorphisms studied, illustrating that LD adjacent to an SNP can be considerably more extensive than has recently been suggested. None of the alleles from the VNTR, the SNP or their haplotype showed statistically significant evidence for association. We stratified patients for current disability status but using this manoeuvre found no evidence that either of the polymorphisms influences disease severity. Combining the available data on the IL1RN VNTR suggests that any effect of this gene on susceptibility to multiple sclerosis, or its progression is, at best, small.