Background: Acute cellular rejection in cardiac allografts is a major cause of graft loss, and is associated with activation of the coagulation system. We investigated whether plasma markers of coagulation predict the presence of allograft rejection.
Methods: A total of 132 blood specimens and endomyocardial biopsies were collected from 35 patients, between February of 1997 and May of 1998. We measured plasma prothrombin fragment 1.2 (PF1.2) and p-selectin, fibrinogen, thrombomodulin, and d-dimer. Biopsies were graded according to the International Society of Heart and Lung Transplantation system, with a range of 0 to 4. Grades 0 and 1A were grouped as "no rejection," and the higher grades as "rejection." Linear and logistic regression, accounting for longitudinal data, were the principal analytic tools.
Results: p-Selectin level increased progressively with increasing rejection grade (P<0.001). With multivariate analysis, both p-selectin and prothrombin fragment levels significantly predicted rejection. p-Selectin levels were predictive of prothrombin fragment levels (P<0.0001) but not of d-dimer, fibrinogen, or thrombomodulin levels. This model allowed correct prediction of rejection, based on p-selectin and prothrombin fragment values, up to 85% of the time. Dichotomizing patients by a p-selectin level of 65 ng/ml resulted in an odds of rejection of 21.4 [95% C.I. 7.1-64.7] for the patients in the high- compared with the lower risk group.
Conclusions: In heart transplant recipients, p-selectin levels and PF 1.2 levels are highly predictive of organ rejection. The elevation of PF 1.2 suggests that there is systemic generation of thrombin generation. These markers may be useful for noninvasively monitoring patients for organ rejection or for after response to treatment.