Comparison of the diagnostic utility of timed serial (slope) creatine kinase measurements with conventional serum tests in the early diagnosis of myocardial infarction.

We compared the diagnostic utility of recently proposed slope assays for serum creatine kinase and creatine kinase-2 with the optimized decision threshold assays. The former approach has been claimed to be superior to any other single diagnostic technique. We show, by ROC curve and likelihood ratio analyses, that the total creatine kinase slope assay possesses the same diagnostic power, when confidence intervals are used, as the optimized decision threshold assay. Moreover, slope assays of creatine kinase-2 were diagnostically inferior to the optimized decision threshold assays. Indeed, these latter, optimized, assays have the highest likelihood ratios for a positive test result of the available assays, and they should always be used in situations of diagnostic doubt.