Using receiver-operating characteristic (ROC) curve and likelihood ratio analysis, we examined the diagnostic utility of total lactate dehydrogenase (LD; EC 1.1.1.27) activity (I). LD isoenzyme-1 activity (II), and the LD-1 percentage of total LD activity (III), LD-1 LD-2 (IV), and LD-1/LD-4 (V) in 347 persons admitted to the Cardiac Care Unit (of whom 173 were subsequently proven to have had myocardial infarction). Blood was sampled from these subjects at about 6-h intervals for up to 96 h from the onset of chest pain. Defining an "effective" test as one having an area under the ROC curve of greater than or equal to 0.9, we determined the ranked utility (greatest to least) of these tests as V = IV greater than III greater than II greater than I. Tests III, IV, and V had by this criterion, diagnostic effectiveness equivalent to measurements of creatine kinase-2 in serum but in samples obtained at later time intervals. The decision thresholds for both high (constant) test sensitivity and specificity varied with time, to differing extents, over the entire 96-h period, a finding with important diagnostic implications. We document positive and negative likelihood ratio values for each of these tests throughout the entire period of study.