To the Editor.— In a recent BRIEF REPORT (1980;244:580) Siegel and Dawson correctly point out that small levels of the MB fraction of creatine kinase (CK) can be detected in the undiluted sera of patients without myocardial involvement if the total CK levels are substantially elevated. Since numerous studies, including those they cite, demonstrate that the MB fraction in noncardiac patients is typically 0% to 2% of the total CK activity and if we accept the authors' assessment that about 5 IU/L is the minimum detectable level of CK by the fluorescent method, then it is clear that many noncardiac patients with total CK activity of about 500 IU/L will possess detectable levels of CK-MB in their undiluted sera. This underscores the need for a percentage quantitation of the CK-MB as a fraction of the total CK level, although the absolute level of CK-MB is also generally regarded as important