Author + information
- Received September 21, 1994
- Revision received March 16, 1995
- Accepted March 31, 1995
- Published online August 1, 1995.
- John F. Carlquist, PhDa,b,
- Richard H. Ward, PhDa,
- Karen J. Meyer, BSb,
- Dee Husebyea,
- Michael Feoloa and
- Jeffrey L. Anderson, MD, FACC1,a,b
- ↵1Address for correspondence: Dr. Jeffrey L. Anderson, LDS Hospital, Division of Cardiology, 8th Avenue and C Street, Salt Lake City, Utah 84143.
Objectives. This study used a meta-analysis to examine HLA-DR frequencies in rheumatic heart disease and prospectively examined other class II allelic disease associations.
Background. Studies of rheumatic heart disease have reported HLA class II allelic associations, but these are inconsistent.
Methods. A meta-analysis combined all known (n = 10) studies to determine disease risk associated with HLA-DR antigen expression. Meta-analysis of studies grouped by ethnic derivation of subjects was also performed. The present study also examined DQA, DQB and DPB allele frequencies by DNA-based strategies.
Results. Meta-analysis showed a significant negative disease association with DR5 (odds ratio [OR] 0.67, p < 0.00003) for all combined studies. Among black patients, DR1 was increased (OR 2.80, p < 0.004); DR6 was increased (OR, 2.03, p < 0.003); and DR 8 was decreased (OR 032, p < 002). Among Eastern Indian patients, DR3 was increased (OR 244, p < 0.00003), with decreased expression for DR2 (OR 0.31, p < 0.00001) and DR5 (OR 0.52, p < 0.05). DR4 was increased among American whites (OR 1.74, p < 0.03), although there was significant heterogeneity among studies of whites. DQA, DQB and DPB allele frequencies were similar for control subjects and patients.
Conclusions. Our findings support an association between major histocompatibity complex class II alleles and risk for rheumatic heart disease. However, heterogeneity in associations was observed among different ethnic and racial groups; regional and temporal differences in streptococcal outbreaks may compound this heterogeneity. Further studies are necessary to elucidate therespective contributions of these variables.
☆ This work was supported in part by Biomedical Research Support Grant S07 RR-05804-09 to Dr. Carlquist from the Division of Research Resources, National Institutes of Health, Bethesda, Maryland and a grant from the Nora Eccles Treadwell Foundation, Salt Lake City, Utah to Dr. Ward.
- Received September 21, 1994.
- Revision received March 16, 1995.
- Accepted March 31, 1995.