Delta cardiac troponin values in practice: are we ready to move absolutely forward to clinical routine?

The improved analytical characteristics of both sensitive and high-sensitivity assays for cardiac troponins I (cTnI) and T (cTnT) have substantially increased the diagnostic sensitivity for early detection of acute myocardial infarction (AMI) (1–3 ). With the ability to detect small increases in circulating cardiac troponins, any cause of myocardial injury will now produce a substantially greater number of analytically true positive findings not detectable by earlier generations of cardiac troponin assays (3– 6 ). This evolution has led to a decrease in diagnostic specificity for the diagnosis of AMI, a concern to many clinicians who may incorrectly equate any increased cardiac troponin value to an AMI (7, 8 ). Observational studies, although conducted predominately with prior-generation assays, have consistently demonstrated that any increase in cardiac troponin due to almost any mechanism of myocardial injury carries an association with worse outcomes (3, 7 ). Nevertheless, because diagnostic specificity is pivotal to appropriate therapy, pragmatic approaches are needed to sustain diagnostic specificity with sensitive and high-sensitivity assays for cardiac troponin. In this issue of Clinical Chemistry, Mueller and colleagues (9 ) report on their experience with one such approach that uses the change in cardiac troponin over time ( ). First proposed by Fesmire in 2000 (10 ), diagnostic algorithms based on the cardiac troponin have now been shown in several studies to improve diagnostic specificity, but not necessarily diagnostic sensitivity (11–13 ). When a sensitive contemporary assay (Ortho Clinical Diagnostics Vitros ES cTnI) was used, a 6-h based on an optimized 30% change improved diagnostic specificity from 77% to 91%, whereas diagnostic sensitivity changed from 94% based on the 99thpercentile value, compared with 75% based on the value (11 ). More recently, using the high-sensitivity cTnT assay (hs-cTnT), Reichlin and coworkers observed that a criterion of an absolute over a 2-h period of 0.007 g/L (7 ng/L) significantly improved the diagnostic specificity for AMI from 93% to 95% (13 ). A similar result was obtained with a sensitive contemporary cTnI assay ( , 0.020 g/L; diagnostic specificity, from 91% to 96%; not significantly different compared with the hs-cTnT assay). Reichlin et al. found that a diagnostic criterion that used the absolute provided a superior overall diagnostic accuracy for AMI, compared with relative (percentage) (0.95 vs 0.72, P 0.001). Together, these data strongly suggest that a diagnostic approach that uses serial cardiac troponin values measured with an analytically robust assay over a very short time period substantially mitigates the major concern of clinicians about high-sensitivity cardiac troponin assays. Despite these encouraging findings, several challenges and unanswered questions regarding the clinical implementation of -based strategies remain, and several are illustrated in the work by Mueller and colleagues (9 ). In contrast to previous diagnostic studies, Mueller et al. selected patients with definite acute coronary syndrome (ACS) and those with an increased hs-cTnT result unrelated to ACS. Patients without ACS and with an hs-cTnT result 99th percentile were not included in their analysis. The authors then tried to discriminate patients with AMI from those with nonACS increases in hs-cTnT values and found that, analogously to Reichlin et al., an absolute of 0.0092 g/L (9.2 ng/L) from 2 samples obtained within 6 h performed better than using a percentage change (0.90 vs 0.75). Interestingly, use of the peak hs-cTnT alone from 0 – 6 h (area under the ROC curve, 0.83) performed better than the percentage and offered perfect diagnostic sensitivity (100% at the 99th percentile). In addition, when categorized by the baseline concentration of hs-cTnT, the absolute was superior only at the extremes, where either very low initial concentrations produced very large percentage changes or very high initial concentrations required large absolute changes to meet the relative criterion. In contrast to previous 1 Hennepin County Medical Center, Minneapolis, MN; 2 Department of Laboratory Medicine and Pathology, University of Minnesota, Minneapolis, MN; 3 TIMI Study Group, and 4 Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, Boston, MA. * Address correspondence to this author at: Hennepin County Medical Center, Clinical Laboratories P4, 701 Park Ave., Minneapolis, MN 55415. Fax 612-9044229; e-mail [email protected]. Received October 31, 2011; accepted November 1, 2011. Previously published online at DOI: 10.1373/clinchem.2011.175414 3 Nonstandard abbreviations: cTnI, cardiac troponin I; cTnT, cardiac troponin T; AMI, acute myocardial infarction; hs-cTnT, high-sensitivity cTnT (assay); ACS, acute coronary syndrome; BV, biological variability. Clinical Chemistry 58:1 8–10 (2012) Editorials