Estimates of myocardial infarct (MI) size based on plasma creatine kinase (CK) are used widely for prognosis and in the assessment of therapy designed to salvage ischemic myocardium. However, if the initial plasma CK activity is elevated, MI size will be underestimated. To determine the impact of loss of early CK values on estimates of MI size and to develop a procedure to compensate for it, estimates of MI size based on complete and incomplete MB and total CK time-activity curves from 120 patients (experimental group) were compared. Estimates of MI size based on data inclusion intervals beginning at 24, 12, 8, and 4 hours before peak CK were 11, 14, 23, and 47% smaller than values based on complete CK curves, but the correlation was good between complete and incomplete estimates of MI size at any given interval, with r values ranging from 0.91 to 0.98. The derived correction factors were then prospectively applied to a new population (n = 25) with complete CK curves to compensate for purposely omitted early CK values. The corrected estimates of MI size were within 7% of those based on the complete CK curves. Similar results were obtained for transmural and nontransmural and anterior or inferior MI. Thus, if peak plasma CK is known, underestimation of MI size can be compensated for despite the unavailability of early CK values. Since greater than 90% of patients present before plasma CK has reached its peak (24 hours), MI size can be obtained in nearly all patients. Thus, being able to correct for unavailable early CK values makes MI size a more widely applicable endpoint for use in clinical trials and patient management.