Author + information
- Mikkel M. Schoos,
- Martin Nielsen,
- Jørgen Thode,
- Steen Hansen,
- Kasper Iversen,
- Peter Clemmensen and
- Rolf Steffensen
Creatine-kinase myocardial band (CK-MB) is currently widely used as a second ‘rule in’ marker for acute myocardial infarction (MI) in order to increase diagnostic accuracy of standard troponins. With the advent of high-sensitive cardiac troponins (hscTn), we investigated whether this two marker strategy is still necessary.
In 323 consecutive unselected chest pain patients with 49 (15.2%) MI, we investigated if CK-MBR (Elecsys, Roche Diagnostics) or CK-MBS (Siemens Dimension Vista) add information to the diagnostic accuracy of hscTnT (Roche Diagnostics) and hscTnI (Siemens Dimension Vista). HscTn positive was defined as the combination of >99 percentile (14 mg/L and 45 ng/L) and a ROC optimized absolute Δ value (5.7 ng/L and 21 ng/L) for hscTnT and hscTnI, respectively.
Diagnostic accuracy of hscTnT (AUC 0.935) and hscTnI (AUC 0.911) did not differ significantly (p=0.344), nor did diagnostic accuracy of peak CK-MBR (AUC 0.842) and peak CK-MBS (AUC 0.830) (p=0.199). CK-MBR and CK-MBS demonstrated 49 % and 34.7 % sensitivity, 90.5 % and 95.5% specificity, 41.4% and 56.7% PPV and 90.5 % and 89 % NPV, respectively. This confirms CKMB as a marker with moderate ‘ruling out’, but excellent ‘ruling in’ capacity. In comparison with hscTnI, reclassifying cases to non-cases by normal peak CK-MBS (<4 µg/L for women & <7 µg/L for men) over 2 samples, resulted in a net reclassification index for cases (NRIC) of 30.6% in favor of hscTnI and a NRI for non cases (NRINC) of 3.6 % in favor of CK-MB. Regarding hscTnT, net reclassification was in favor for hscTnT regarding both cases (NRIC = 22.45%) and non-cases (NRINC = 2.55%). The diagnostic accuracy of the combination of hscTnT and CK-MBR (AUC 0.898) was inferior to hscTnT alone (p=0.006), whereas hscTnI and CK-MBS (AUC 0.915) did not add to hscTnI alone.
Our results indicate that CK-MB does not add incremental diagnostic accuracy to a two sample strategy with hscTn > 99 percentile + significant absolute Δ value.
Poster Sessions, Expo North
Monday, March 11, 2013, 9:45 a.m.-10:30 a.m.
Session Title: Cardiac Biomarkers and Cardiovascular Risk
Abstract Category: 1. Acute Coronary Syndromes: Clinical
Presentation Number: 1303-212
- 2013 American College of Cardiology Foundation