ABC | Volume 110, Nº1, January 2018

Original Article Soeiro et al Troponin vs. coronary lesion Arq Bras Cardiol. 2018; 110(1):68-73 Table 2 – Multivariate analysis comparing patients with versus without significant coronary lesion: A. Using the 99 th percentile of the troponin assay; B. using the best cutoff point for troponin found in the study A OR 95% CI p Male sex (%) 1.32 0.99 - 1.76 0.052 Arterial hypertension (%) 0.81 0.55 - 1.18 0.272 Smoking (%) 1.58 1.18 - 2.14 0.002 ST depression/T-wave inversion 2.05 1.47 - 2.88 < 0.0001 Troponin + / 99 th percentile 3.39 2.32 - 4.94 < 0.0001 B OR 95% CI p Male sex (%) 1.35 1.02 - 0.180 0.039 Arterial hypertension (%) 0.89 0.60 - 1.31 0.548 Smoking (%) 1.64 1.21 - 2.22 0.001 ST depression/T-wave inversion 2.22 1.58 - 3.12 < 0.0001 Troponin + / Best cutoff point 3.39 2.53 - 4.54 < 0.0001 OR: odds ratio; CI: confidence interval. Irfan et al. 14 have conducted an observational multicenter study with 1181 patients hospitalized because of non-cardiac causes, 15% of whom had positive high-sensitivity troponin T. Of the major factors related to that unexpected elevation, the presence of kidney dysfunction was identified as a significantly influencing factor. In addition, once again, patients with elevated troponin were at higher risk for death (HR = 3.0; p = 0.02). 14 In individuals older than 75 years, high-sensitivity troponin T was assessed in the context of chest pain, being measured at baseline and 3-4 hours. Approximately 27% of the patients were classified as having acute coronary syndrome. The sensitivity and specificity found in that population were 88% and 38%, respectively. The greater the initial level or the increase (mainly absolute) in the subsequent measures, the higher the specificity found. 15 That specificity value can be greater than ours found in the general population, probably because of the inclusion of more patients with other heart diseases, because we belong to a referral tertiary cardiology center. The concept of variation in the levels of sensitive troponin and high-sensitivity troponin in different measurements has been studied, and establishing a correlation between the amplitude of variability and the probability of coronary artery disease has been consecutively attempted. In addition, amplitude can be relative (expressed as percentages) or absolute, with possible implications and distinct interpretations. 1 A retrospective study published in 2014, including 1054 patients with chest pain, assessed the variability related to high-sensitivity troponin T. Approximately 40% of the patients showed alteration in at least one measurement. Even with a variation greater than 20% as compared to the initial level, the specificity did not exceed 70%. 16 Assessing specifically the same current sensitive troponin assay used in this study, in 2013 Bonaca et al. 17 published a study comparing current sensitive troponin I versus high-sensitivity troponin I in 381 patients with chest pain at the emergency department. Those authors found sensitivity values for the two assays of 94% and 97%, and negative predictive values of 98% and 99%, respectively, with no significant difference. 17 Another similar study of 1807 patients with non-ST-segment elevation acute coronary syndrome has shown no significant difference regarding prognosis when comparing the positivity of current sensitive troponin I versus high-sensitivity troponin I. 18 Differently from the findings of those studies and using the same assay, ours showed lower sensitivity and specificity of 23% when using the 99 th percentile of the method. That shows the importance of assessing each center’s population, respecting their specific individualities. In alignment with that, the meta-analysis published in 2014 with 17 studies and 8644 patients with chest pain compared the use of high-sensitivity troponin with that of conventional troponin. There were differences regarding sensitivity (88.4% vs. 74.9%; p < 0.001) and specificity (81.6% vs. 93.8%; p < 0.001), respectively. Despite that increase in sensitivity with high-sensitivity troponin, the number of patients with the final diagnosis of myocardial infarction and the need for additional tests for ischemia did not differ between the groups, showing no additional clinical advantage with the use of high‑sensitivity troponin. 2 71