In cardiac transplant recipients the release of soluble cellular adhesion molecules intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) and E-Selectin into serum is pronounced during immune activation. It is uncertain whether there is a specific pattern of release during infection or cardiac allograft rejection. In a prospective study, 30 consecutive cardiac allograft recipients were followed for a median period of 11.4 months (range 1-34). Soluble ICAM-1 (sICAM-1), soluble VCAM-1 (sVCAM-1) and soluble E-Selectin (sE-Selectin) were measured in addition to acute phase proteins (C-reactive protein, alpha1-antitrypsin), complement factors (C3, C4) and beta2-microglobulin. The measured serum levels were correlated with the clinical status of the transplant recipient: 1) uneventful clinical status; 2) asymptomatic infection; 3) symptomatic infection and 4) rejection. Forty age-matched healthy subjects served as controls. Six days before biopsy-proven cardiac allograft rejection sICAM-1-release started to increase (p < 0.05) as compared to uneventful clinical status. The peak concentration of sICAM-1 was measured three days before rejection. On the day of rejection, serum concentrations of sICAM-1 (p < 0.001) and sVCAM-1 (p < 0.05) were increased, whereas sE-Selectin was not markedly elevated. In symptomatic infections, the serum concentrations of sICAM-1 (p < 0.001) and sVCAM-1 (p < 0.05) were elevated at the day of diagnosis and both parameters reached peak levels three days after onset of chemotherapy. In multivariate analysis soluble adhesion molecules only weakly discriminated between rejection and infection (sensitivity: 13%, specificity: 95%). Although, in combination with routine blood parameters the discriminatory power could be improved (sensitivity: 85%, specificity: 85%) the clinical utility of these markers in non-invasive monitoring is limited.