The aims of our study were to evaluate the plasma carboxypeptidase N activity in normal subjects and in patients with acute myocardial infarction and to delineate its relationship with creatine kinase-MB isoforms in monitoring of acute myocardial infarction, carboxypeptidase N being the major determinant of creatine kinase isoform conversion in plasma. The study was carried out in 34 healthy subjects and 19 patients with acute myocardial infarction diagnosed according to the World Health Organization (WHO) criteria in which the blood samples were collected immediately upon admission to the coronary care unit (median time 3.5 hours), every 4 to 6 hours for 24 hours, and every 12 hours until the third day post admission. Carboxypeptidase N activity, total creatine kinase, creatine kinase-MB mass concentration and creatine kinase-MB isoforms were determined in each sample from acute myocardial infarction patients, whereas only carboxypeptidase N and total creatine kinase activities were assayed in samples from healthy subjects. The results showed a high variability in carboxypeptidase N values among healthy subjects (median = 200 U/l; interquartile range = 190-247 U/l) and in the first available samples from acute myocardial infarction patients (median = 213 U/l; interquartile range = 234 U/l) without significant differences between groups and without a correlation between carboxypeptidase N and creatine activities either in healthy subjects or in acute myocardial infarction patients; in the latter group, however, a significant correlation (p < 0.01) with creatine kinase-MB calculated on all samples, was observed. In acute myocardial infarction patients carboxypeptidase N showed time-related variations, reaching the highest levels about 48 h after onset of chest pain. A statistically significant difference in carboxypeptidase N values (p = 0.0001) was found before and after creatine kinase-MB peak values as well as before and after MB2/MB1 normalization. Worthy of note is the finding that in two acute myocardial infarction patients presenting MB2/MB1 ratios lower than the cutoff value (1.5) throughout the period of observation, the baseline values for carboxypeptidase N were higher than in other patients studied. Our results suggest that the increase of carboxypeptidase N activity after infarction could be induced by an increase in endogenous substrate concentrations, in particular creatine kinase-MB released from damaged myocardium. Furthermore, high baseline levels of carboxypeptidase N will reduce the diagnosis efficiency of creatine kinase-MB isoforms in the diagnosis of acute myocardial infarction.