Abstract
The Tokyo Metropolitan Matsuzawa Hospital has 898 beds (808 beds for psychiatric patients), and conducts high-sensitivity troponin I (hs-TnI) examination at the time of emergency consultation. In the past, the hs-TnI positive rate after introduction is at least three times higher in odds ratio than the prevalence rate of acute myocardial infarction (AMI) calculated in our hospital. Therefore, to confirm the reliability of the hs-TnI value, measurements with other analyzers and nonspecific reactions using heterophilic antibody inhibitors were first examined, but no particular problems were observed. Next, the AMI diagnostic specificity of myocardial markers was evaluated. As myocardial markers, hs-TnI, CK-MB mass, CK-MB activity, and human-heart-type fatty acid binding protein were used. Then, 27 items of blood and biochemical data were added, and an AMI diagnosis guide using a decision tree was made on a trial basis. Only hs-TnI showed a significant difference for the presence or absence of AMI diagnosis and was a myocardial marker independent of other factors. In the decision tree, chloride, hs-TnI concentration, and CK-MB mass value at the time when a positive hs-TnI result was obtained were derived as branch nodes. The combination of these extraction factors showed improvement in AMI diagnostic accuracy. However, the positive predictive value of less than 50% is insufficient, suggesting the need for diagnostic guidelines to which qualitative data such as the blood collection results and electrocardiograms can be added over time.
