The predictive value of serum enzymes for perioperative myocardial infarction after cardiac operations. An autopsy study
Van Lente, F.; Martin, A.; Ratliff, N.B.; Kazmierczak, S.C.; Loop, F.D.
Journal of Thoracic and Cardiovascular Surgery 98(5 Pt 1): 704-710
1989
ISSN/ISBN: 0022-5223 PMID: 2811407 Document Number: 330448
We evaluated the utility of serum enzyme and isoenzyme activities for detecting autopsy-proved perioperative myocardial infarction in patients who died after cardiac operations. We studied 79 patients who had autopsies performed after coronary artery bypass grafting or value replacement, or both. Thirty-seven had histologic evidence of a perioperative myocaridal infarction. We found statistically significant differences between the group of patients with infarction and the group without infarction when we compared the mean activities of creatine kinase, creatine kinase MB, aspartate aminotransferase, and the lactate dehydrogenase-1/lactate dehydrogenase-2 ratio. The postoperative changes in serum enzymes were analyzed by logistic regression for their relation to perioperative myocardial infarction. Creatine kinase MB exhibited the best diagnostic association with the presence of perioperative myocardial infarction. The lactate dehydrogenase-1/lactate dehydrogenase-2 ratio correlated to a lesser extent with infarction. Adjustment of the diagnostic cutoff to 133 U/L for creatine kinase-MB measured 15 hours after operation yielded a sensitivity of 0.60 and a specificity of 1.0. This study demonstrates that no combination of enzyme activity changes after operation can completely discriminate all patients with perioperative myocardial infarction from these without. Nonetheless, measurement of creatine kinase MB activity provide 96% accuracy for diagnosing infarction at a prevalence of 10%.