We read the report by Pfortmueller et al, who showed that the diagnostic value for acute myocardial infarction (AMI) of high-sensitivity (HS) troponin T (TnT) was poor in patients with impaired renal function at emergency department (ED) admission. Because the renal metabolism of the different troponin isoforms is markedly different, we carried out a clinical investigation to establish whether renal function may influence the values of TnT and troponin I (TnI) measured with HS immunoassays in patients admitted to the ED with suspected AMI but without a final diagnosis of myocardial ischemia.
The study population consisted of 48 consecutive patients admitted with chest pain to the local ED and for whom a final diagnosis of myocardial ischemia could be ruled out according to the third universal definition of AMI. Cardiac troponins were assessed at patient admission with 2 highly sensitive methods, that is, the prototype Beckman Coulter HS-AccuTnI (Beckman Coulter, Chaska, Minnesota) and the Roche HS-TnT (Roche Diagnostics, Basel, Switzerland), which are characterized by a ninety-ninth percentile of the upper reference limit of 32 ng/L and 14 ng/L, respectively. Serum creatinine level was also measured at patient admission, and glomerular filtration rate (GFR) was calculated according to the currently recommended Chronic Kidney Disease Epidemiology Collaboration equation.
The results of this study are listed in Table 1 . In 19 of 48 patients (∼40%), the GFR was <60 ml/min. In agreement with the data of Pfortmueller et al, we found that HS-TnT values were up to 10-fold higher in patients with impaired renal function than in those with normal GFR. A lower and not significant increase (approximately threefold) was also observed for HS-TnI in patients with GFR <60 ml/min than those with GFR >60 ml/min. Comparing subjects with preserved and impaired renal function, a significant difference of values exceeding the ninety-ninth percentile of the upper reference limit was however found for HS-TnT but not for HS-TnI. Even more interestingly, although the frequency of HS-TnT and HS-TnI values exceeding the ninety-ninth percentile of the upper reference limit was identical (i.e., 21%), in patients with normal GFR, the frequency of values above the relative thresholds of the 2 immunoassays was nearly double for HS-TnT than for HS-TnI in patients with GFR <60 ml/min (63% vs 37%, respectively, p <0.001). Finally, a significant Pearson correlation was found between GFR and HS-TnT (r = −0.37, p = 0.01) but not between GFR and HS-TnI (r = −0.24, p = 0.10).
