The role of human leukocyte antigen genes in the formation of abdominal aortic aneurysms.

BACKGROUND

Increasing evidence suggests an autoimmune component to abdominal aortic aneurysm (AAA) formation. This study was conducted to determine if a difference exists in human leukocyte antigen (HLA) allele distribution between patients with AAA and population controls, and between patients with small and large AAA.

METHODS

Patients with known AAA attending the vascular unit were consented for recruitment. HLA-A, HLA-B and HLA-DR was determined by polymerase chain reaction and sequence-specific oligonucleotide probes. The distribution of these alleles in the Northern Ireland general population was obtained from the histocompatibility and immunogenetics database. The chi(2) test was used for statistical analysis with Bonferroni correction.

RESULTS

A total of 241 AAA patients were recruited, with a wide range of aneurysm size. In class I, the most frequent allele families were HLA-A*02 and 01 and HLA-B07, 08, and 44. In class II, HLA-DRB103, 04, 07, and 15 were the most frequent. HLA-A11 was lower in AAA cases (10.4% vs 15.0%; P = .08), whereas HLA-B08 was lower in the controls (29.8% vs 36.5%; P = .05) and HLA-DRB111 was lower in cases (4.2% vs 8.1%; P = .05). After Bonferroni correction, however, the proportion of allele families was not significantly different in AAA patients compared with the proportion seen in controls. HLA-DRB111 and 14 had a lower prevalence in large AAAs (0.9% vs 6.7% [P = .05]; 0.0% vs 5.9% [P = .03]). HLA-A68 was also lower in large AAA (1.9% vs 11.9%; P = .0075). After Bonferroni correction, however, no difference was demonstrated between small and large aneurysms.

CONCLUSION

This study provides more definitive results on this important subject and has failed to demonstrate the risk association between AAA and these alleles as reported by others. Therefore, the role of these particular genes and the autoimmune component in AAA etiology does not appear to be as crucial as previously proposed.