Background : Despite the large body of literature, no clear causal relationship between chronic CMV infection of kidney transplants and long-term graft dysfunction has yet been established in humans and factors that may modulate these interactions have never been documented. Methods : In this retrospective study in 1279 kidney transplant patients, we analyzed the 25-year actuarial graft survival and the evolution of the CD8+ cell population according to donor and recipients CMV serology and occurrence of antigenemia. Results : CMV latency in the donor was an independent risk factor of kidney graft loss (HR D+vsD-= 1.392 [1.059-1.830], p=0.018), especially in R+ recipients (HR D+ vs D-= 1.493 [1.008-2.211], p=0.046). This detrimental effect was not correlated with the occurrence of a post-transplant antigenemia, since kidneys allograft issued from CMV-positive donors had also a worse outcome in antigenemia-free recipients (HR D+ vs D-= 2.238 [1.300-3.853], p=0.004), especially in R+ recipients (HR D+ vs D-= 3.357 [1.365-8.257], p=0.008). Impact of donor’s CMV serostatus on graft survival was restricted to patients with a full HLA-I mismatch (HR D+ vs D- in full mismatched patients = 2.531 [1.484-4.316], p=0.0006; HR D+ vs D- in full mismatched patients = 1.194 [0.918-1.552], p=0.185). Further, we assessed CD8+-cell compartment according to the recipient’s CMV serostatus. CMV viremia induced a long-lasting CD8 cells expansion in R- (402±169, 836±481 and 1112±672/mm3 at D0, 1 and 2 years respectively; D0 vs 1 and 2 years p<0.0001), but not in R+ (502±245/mm3 at D0 and 587±284 at 2 years, p=NS). Finally a CD8+ T-cell attrition of more than 40% at 2 years (compared to pre-transplant values) was associated with an increased risk of graft loss in CMV-exposed patients (R+ and/or D+)(p=0.003). Conclusion : We identified risk factors of the development of CMV-associated chronic allograft dysfunction opening the way for preventive strategies.

« Back to 2013 American Transplant Congress