Effect of human leukocyte antigen-C and -DQ matching on pediatric heart transplant graft survival.

BACKGROUND: A higher degree of human leukocyte antigen (HLA) matching at the A, B, and DR loci has been associated with improved long-term survival after pediatric heart transplantation in multiple International Society for Heart and Lung Transplantation registry reports. The aim of this study was to investigate the association of HLA matching at the C and DQ loci with pediatric graft survival.
METHODS: The United Network of Organ Sharing database was queried for isolated heart transplants that occurred from 1988 to 2012 with a recipient age of 17 or younger and at least 1 postoperative follow-up encounter. When HLA matching at the C or DQ loci were analyzed, only transplants with complete typing of donor and recipient at the respective loci were included. Transplants were divided into patients with at least 1 match at the C locus (C-match) vs no match (C-no), and at least 1 match at the DQ (DQ-match) locus vs no match (DQ-no). Primary outcome was graft loss. Univariate analysis was performed with the log-rank test. Cox regression analysis was performed with the following patient factors included in the model: recipient age, ischemic time; recipient on ventilator, extracorporeal membrane oxygenation, ventricular assist device, or inotropes at transplant; recipient serum bilirubin and creatinine closest to transplant, ratio of donor weight to recipient weight, underlying cardiac diagnosis, crossmatch results, transplant year, and HLA matching at the A, B, and DR loci.
RESULTS: Complete typing at the C locus occurred in 2,429 of 4,731 transplants (51%), and complete typing at the DQ locus occurred in 3,498 of 4,731 transplants (74%). Patient factors were similar in C-match and C-no, except for year of transplant (median year, 2007 [interquartile range, 1997-2010] vs year 2005 [interquartile range, 1996-2009], respectively; p = 0.03) and the degree of HLA matching at the A, B, and DR loci (high level of HLA matching in 11.9% vs 3%, respectively; p < 0.01). Matching at the C locus was not associated with a decreased risk of graft loss (median graft survival: 13.1 years [95% confidence interval {CI}, 11.5-14.8] in C-no vs 15.1 years [95% CI, 13.5-16.6) in C-match, p = 0.44 log-rank; hazard ratio, 0.93; 95% CI, 0.76-1.15; p = 0.52). DQ-match did not differ from DQ-no in any of the analyzed patient factors, except DQ-match was more likely to have high degree of matching at the A, B, and DR loci vs DQ-no (9.8% vs 3.2%, p < 0.01). Matching at the DQ locus was not associated with decreased risk of graft loss (median graft survival: DQ-no, 13.1 years [95% CI, 11.7-14.6) vs DQ-match, 13.0 years [95% CI, 11.4-14.6], p = 0.80, log-rank; hazard ratio, 0.95; 95% CI, 0.81-1.1; p = 0.51.
CONCLUSIONS: Complete typing at the C locus of both donor and recipient occurs less often then typing at the DQ locus. A higher degree of donor-recipient HLA matching at the C locus or the DQ locus appears not to confer any graft survival advantage.