Notes: Abstract Acute rejection is a frequent consequence after heart transplantation. To expand our knowledge of the rejection process and to investigate some intragraft events during acute rejection, the following experimental transplantation model was designed. Right cervical heart transplantation was performed in 12 mongrel dogs. Two experimental groups of six animals each received different immunosuppressive regimens. All animals were treated with daily triple drug therapy. In contrast to group 1, the animals in group 2 received high-dose steroids during rejection. The condition of the hearts was examined by daily transmural biopsies, graded according to the Billingham classification. To detect and quantify alterations in the mononuclear cell subsets of the myocardial venous return, blood samples from the coronary sinus blood (CS) and from peripheral blood (PB) were taken simultaneously with the biopsy. The total number of lymphoblasts and activated lymphocytes was determined and an activation index (AI) was calculated. The data referred to was established from 337 transmural biopsies. The AI of PB (n=287) correlated well with the different stages of acute rejection (grade B0: AI=2.2±2.1; grade B1+2: AI=6.3±1.7; grade B3: AI=10.0±4.7; P〈0.001). The rejection kinetics of both groups, including the rejection-free interval following high-dose steroid administration in group 2, could be expressed accurately by the AI. The time course of the total number of lymphoblasts in CS versus PB demonstrated that the lymphoproliferative response started 4 days prior to the first intramyocardial signs of rejection (x = 3.8 ± 0.7; n=12). The maximum number of lymphoblasts was seen on the day of rejection in group 1 and 1 day after the onset of histologically proven rejection in group 2 (group 1: n=6: CS x = 40.1 ± 7.5; PB x = 12.2 ± 4.1; P〈0.001; group 2: n=6: CS x = 39.4 ± 8.8; PB x = 12.9 ± 3.7; P〈0.001). Under rejection therapy in group 2 these cells decreased immediately, followed by a short rejection-free interval. In group 1 the total number of lymphoblasts diminished continuously, almost reaching the nuber in PB at the time of final rejection. In contrast, activated lymphocytes did not render adequate results. Comparison of daily histology and the data of PB proved there is a good correlation between the AI and the different histologic stages of acute rejection. The total number of lymphoblasts in CS during rejection is significantly higher than in PB. Acute rejection seems to be detectable almost 4 days before histology and PB cytology by cytologic evaluation of the CS. Therefore, we speculate that the differentation and proliferation of lymphoblasts during the initial phase of acute rejection takes place within the graft itself.