Guidelines for diagnosis of acute myocardial infarction recommend that, if acute myocardial infarction is suspected, creatine kinase (CK)-MB levels should be measured on admission and again at 12 and 24 hours. In light of these recommendations, we conducted a quality assurance study to determine whether utilization of CK-MB tests in our institution, a large, university-affiliated teaching hospital, was consistent with current guidelines. Also, several years ago, we had established a policy of cancelling lactate dehydrogenase isoenzyme orders if the request originated from an unauthorized location, unless it was approved by a laboratory staff. Since this policy led to a greater than 90% reduction in the requests for lactate dehydrogenase isoenzyme testing, an additional objective was to reevaluate this policy. Of 774 patients evaluated with CK-MB tests, 294 (38%) received only a single test. Of these single tests, 277 had normal results (CK-MB < 5%). For the remaining 17 patients, the single CK-MB test findings were abnormal (CK-MB > 5%) without follow-up testing. Only two CK-MB tests were ordered for 187 patients (24%). Three or more CK-MB tests were obtained in 293 cases (38%). When two or more CK-MB tests were ordered, the time interval between the first and second tests was inappropriately short in 70% and long in 24%. The recommended timing for the third CK-MB was followed in only 4% of cases. Review of 32 cancellations of lactate dehydrogenase isoenzyme tests disclosed that lactate dehydrogenase isoenzyme tests were requested when unnecessary in 26 cases. Despite published guidelines for use of CK-MB for acute myocardial infarction, physicians at our institution continue to use these tests inappropriately by ordering only single CK-MB tests or by ordering repetitions of CK-MB tests at excessively short or long intervals.