[Multicenter evaluation of h-FABP semi-quantitative assay (Cardio Detect) in central laboratory: the point in acute myocardial infarction diagnosis].

The diagnostic performance of heart-Fatty Acid Binding Protein (h-FABP) (semi-quantitative CardioDetect test) and cardiac troponin I (TnIc) blood assays were compared in one hundred patients presenting with suspicion of acute coronary syndrome. Final patient diagnosis was "acute myocardial infarction" in 36 cases, "non ST myocardial infarction" in 25 cases and "non ischemic pathologies" in 39 cases. h-FABP results were positive in 26 patients, negative in 57 patients and ambiguous in 17 patients, the latter corresponding to the final diagnosis of "acute myocardial infarction" in 5 cases, "non ST myocardial infarction" in 2 cases and "non ischemic pathologies " in 10 cases. At admission, h-FABP and TnIc exhibiteda sensitivity of 54% an 66%, respectively and a specificity of 86% and 95%, respectively. Positive and negative predictive values were 81% and 64% for h-FABP, respectively and 92% and 75% for cTnI, respectively. h-FABP and cTnI demonstrated a similar diagnostic efficiency if admission delay is less than 4 hours after onset of chest pain (area under ROC curve TnIc = 0.767 +/- 0.091 ; area under ROC curve h-FABP = 0.622 +/- 0.109 ; p = 0.144). On the contrary, cTnI assay demonstrated a better efficiency than h-FABP (p< 0.005) for patients admitted in a delay of 4 to 12 hours after the onset of chest pain. If chosen cTnI cut-off corresponded to the recent consensus definition used for monitoring acute coronary syndrome patients, h-FABP semi-quantitative assay realized within central laboratory did not demonstrated a better diagnostic efficiency than cTnI.