Persistent increases in cardiac troponin concentrations as measured with high-sensitivity assays after acute myocardial infarction.

We read with interest the report by Koenig et al. (1 ) showing the value of high-sensitivity cardiac troponin T (hs-cTnT) assays for predicting long-term cardiovascular events in patients after an acute coronary syndrome (ACS) or coronary artery bypass grafting. Although studies have demonstrated long-term cardiovascular events with measurements of hs-cTnI or hs-cTnT in patients who present with symptoms suggestive of ACS (2, 3), Koenig et al. found a relationship with hs-cTnT concentrations after patients had been discharged from a cardiac rehabilitation program (mean time to discharge and hs-cTnT measurement was 43 days from the index event) (1 ). hs-cTnT concentrations 3 ng/L (limit of blank) remained detectable in 84% of patients and were 14 ng/L (99th percentile) in 37.1% of the patients. These findings raise important questions about how gradually hs-cTnT concentrations decrease in the absence of recurrent ACS and have implications regarding the use of highsensitivity cardiac troponin assays for diagnosing subsequent events in patients who are stable after an ACS. To address these questions, we conducted a prospective observational study to assess whether cardiac troponin concentrations normalize by 30 days after an acute myocardial infarction (AMI). After obtaining ethics approval, we recruited a convenience sample of Hamilton General Hospital patients who consented to return 30 days after their first AMI (index event) to provide a follow-up sample (plasma from EDTA-anticoagulated blood) for cTnT analyses (Roche fourthgeneration assay, used clinically). Samples were frozen at 80 °C and were subsequently thawed for the first time for hs-cTnT analysis (Roche Diagnostics E Modular Analytics assay; CV, 2.4% at 30 ng/L) and thawed a second time for the hs-cTnI assay (Abbott Diagnostics precommercial prototype assay, ARCHITECT i1000SR; CV, 6.1% at 21 ng/L). We used the following thresholds, which were obtained from the literature for cTnT and from the manufacturer for hscTnI, as evidence of a persistent abnormal increase: cTnT 10 ng/L, hs-cTnT 99th percentiles [ 14 ng/L (overall), 9 ng/L (women), and 16 ng/L (men)], and hs-cTnI 99th percentiles [ 26 ng/L (overall), 16 ng/L (women), and 34 ng/L (men)]. We collected samples from 46 patients, 38 with STelevation myocardial infarction (STEMI) and 8 non-STEMI patients. Two of the patients had a recurrent myocardial infarction during the 30-day follow-up. We assessed the remaining 44 patients [38 men and 6 women; mean (SD) age, 60 (12) years] who had remained event free. Similar to the results of Koenig et al., we found that 80% of the patients had hscTnT values 3 ng/L and 27% had hs-cTnT values 14 ng/L by 30 days after their AMI (Fig. 1). In addition, 84% of the patients had a detectable hs-cTnI concentration ( 1.3 ng/L, limit of blank). Persistent increases, however, were not as common for cTnT values 10 ng/L (2% of patients) or for hscTnI values greater than the overall 99th percentile (7% of patients had values 26 ng/L) by 30 days after an AMI. The use of sex-specific 99th percentile cutoffs further reduced the overall prevalence of persistent hs-cTnT increases (20% of patients) and hs-cTnI (2% of patients). When overall or sexspecific 99th percentile cutoffs were used, the hs-cTnI assay classified participants differently than the hs-cTnT assay (P 0.007, and P 0.011, respectively, by the McNemar test). The probabilities of persistent increases in hs-cTnT or hs-cTnI were not affected by sex (male, female), age ( 60 years, 60 years), or type of index event (STEMI, non-STEMI) (P 0.10, 2 test). There is evidence that increases in hs-cTnT 99th percentile may last up to 7 weeks after an ACS. Another recent cohort study showed that a similar proportion of patients (31.3%) had persistent hs-cTnT increases 14 ng/L at 7 weeks after an ACS (4 ). The risk of a cardiovascular event in patients who had a persistent hs-cTnT increase was 3-fold higher than in patients whose hs-cTnT values normalized by 7 weeks after an ACS (4 ). It remains unclear, however, how long hs-cTnT increases persist beyond 7 weeks or whether this effect is observed with the hs-cTnI assays, because it appears that cTnI concentrations may fall below the 99th percentile cutoffs during the healing phase (approximately 5 to 6 weeks) after an AMI (5 ). In conclusion, although there may be prognostic value for cardiac troponin concentrations measured by high-sensitivity assays during the convalescence phase of stable patients after an ACS, our data suggest that diagnosis of subsequent events (reinfarction) is not possible with a single hs-cTnT measurement. A substantial number of patients will have increased hs-cTnT concentrations for at least 30 days after an AMI without showing increases with either the 7 Nonstandard abbreviations: hs-cTnT, high-sensitivity cardiac troponin T (assay); ACS, acute coronary syndrome; hs-cTnI, high-sensitivity cardiac troponin I (assay); AMI, acute myocardial infarction; STEMI, ST-elevation myocardial infarction. Clinical Chemistry 59:2 000 – 000 (2013) Letters to the Editor