OBJECTIVE The prognostic significance of elevations in creatine kinase-MB and troponin T (cTnT), which have been conventionally measured 6 to 8 h after percutaneous coronary intervention (PCI), has been established. However, the time to peak biomarker appearance in the circulation has not been defined and is the purpose of this pilot study. METHODS Nonrandomized, nonconsecutive patient cohort. METHODS Clinical practice, Mayo Clinic, Rochester, MN. METHODS Cohort (n = 57) undergoing elective PCI. METHODS cTnT and creatine kinase (CK)-MB measured at baseline, 2 h, 4 h, 8 h, and > or = 2 h (mean +/- SEM, 18 +/- 5 h) after PCI. RESULTS Postprocedure cTnT elevations were detected in 30 of 57 patients (53%). Of these, 4 of 30 patients (13%) had peak cTnT at 4 h (0.80 +/- 0.40 ng/mL), 5 of 30 patients (17%) had peak cTnT at 8 h (1.07 +/- 0.48 ng/mL), and 21 of 30 patients (70%) had peak cTnT at > or = 12 h (0.21 +/- 0.06 ng/mL); 22 of 30 patients received abciximab. Elevations in CK-MB occurred in 14 of 57 patients (25%). Of these, 3 of 14 patients (21%) demonstrated peak CK-MB at 2 h (18.5 +/- 7.9 ng/mL) and the remainder (11 of 14 patients, 79%) during the 12- to 20-h interval (20.2 +/- 4.4 ng/mL); 12 of 14 patients received abciximab. CONCLUSIONS More cTnT than CK-MB elevations occur after PCI; however, both biomarkers demonstrate a longer time to peak value than anticipated in clinical practice. Early surveillance monitoring (< 12 h) does not detect peak biomarker levels, especially in patients with normal baseline values. If peak levels are to be used to determine prognosis, then longer time intervals should be used for post-PCI surveillance. The timing of peak elevations appears to be influenced by baselines values as well. Early elevations may reflect the conjoint effects of injury associated with the disease process and the intervention itself. These data suggest that a re-evaluation of surveillance monitoring to account for the variability reported and the influence of baseline elevations of biomarkers may improve the prognostic power of the measurements.