[Skip to Content]
[Skip to Content Landing]
August 3, 2011

Sensitive Troponin I Assay in Patients With Suspected Acute Coronary Syndrome

Author Affiliations
Author Affiliations: Mayo Clinic, Rochester, Minnesota (Dr Jaffe) (jaffe.allan@mayo.edu); Uppsala Clinical Research Center, Uppsala, Sweden (Dr Lindahl); and University Hospital, Heidelberg, Germany (Dr Katus).
JAMA. 2011;306(5):488-489. doi:10.1001/jama.2011.1062

To the Editor: The study on the use of a contemporary sensitive troponin assay and its association with reductions in death and recurrent MI used a diagnostic threshold of troponin of 0.05 ng/mL, based on a coefficient of variation (CV) of less than 10% at that level.1 However, the 99th percentile reference value of 0.012 ng/mL would have been preferable as the diagnostic threshold because this is the value recommended by guidelines groups.2 This cutoff value optimizes the identification of patients at risk with ACS in all studies that have explored this issue.2 There are many reasons for the reluctance of clinicians to use the 99th percentile value, but one reason often cited is that the assay does not have imprecision less than 10% at that value. The Biochemistry Group of the Global Task Force for the Universal Definition of Acute Myocardial Infarction and the Global Task Force itself have for years advocated for this level of imprecision at the 99th percentile value but have never suggested using a less sensitive value when this level of imprecision is not met2,3 because the 99th percentile value maximizes clinical outcomes. In addition, modest increases in imprecision do not cause false-positive values whether used for diagnosis4 or prognosis.5 However, increased imprecision can affect the ability to observe serial changes.3