Resuscitation from out-of-hospital cardiac arrest: implications for cardiac enzyme estimation.

BACKGROUND: diagnosis of precipitating myocardial infarction is essential for management of victims of out-of-hospital cardiac arrest, since investigations and treatment are determined by the underlying cause. Skeletal muscle and myocardial damage from external cardiac massage and defibrillation may complicate biochemical diagnosis of myocardial infarction.
OBJECTIVES: (a) to examine the relationship between cumulative defibrillation energy and serum levels of cardiac troponin T and MB creatine kinase (MB-CK) mass in out-of-hospital cardiac arrest survivors without electrocardiographic evidence of myocardial infarction; (b) to reassess diagnostic thresholds for myocardial infarction using MB-CK mass and troponin T in this setting.
METHODS: 77 victims of out-of-hospital cardiac arrest were studied. Serum was obtained for MB-CK mass, CK and troponin T estimation on the first 4 days of admission. Patients were divided into three groups using electrocardiographic criteria: group 1, myocardial infarction; group 2, no evidence of infarction; and group 3, equivocal electrocardiograms. Correlation coefficients were calculated for highest recorded levels of the biochemical markers versus defibrillation energy. Receiver-operating characteristic plots were used to determine optimum biochemical diagnostic thresholds for subjects in groups 1 and 2.
RESULTS: using predefined criteria, 27 patients had myocardial infarction, 34 did not have myocardial infarction and 16 had equivocal electrocardiograms. Significant correlations were found for defibrillation energy versus log troponin T (r = 0.42, P < 0.05), log MB-CK mass (r = 0.51, P < 0.01) and total CK (r = 0.68, P < 0.001) in group 2. Within groups 1 and 2, MB-CK mass and troponin T provided additional diagnostic value over MB-CK fraction (P < 0.001). Diagnostic accuracy was not improved by adjusting for shock energy. The optimum threshold value was 4 ng/ml for troponin T (sensitivity 88%, specificity 95%), 60 ng/ml for MB-CK mass (sensitivity 88%, specificity 88%) and 8% of total CK for MB-CK fraction (sensitivity 74%, specificity 82%). These values should be interpreted with caution, since this study is limited by the exclusion of patients with uncertain electrocardiographic diagnoses into group 3.
CONCLUSIONS: skeletal muscle and myocardial damage occurs in survivors of out-of-hospital cardiac arrest and is related to the duration of resuscitation. This complicates biochemical diagnosis of underlying myocardial infarction. Specific high diagnostic threshold values for MB-CK and troponin T are needed to optimise diagnostic accuracy. The use of MB-CK fraction leads to greater diagnostic error because of the variability of muscle CK release after resuscitation.