Optimal cutoff-values of roche high-sensitivity cardiac troponin T in patients with kidney disease for the early diagnosis of acute myocardial infarction

Abstract

Purpose: The recent introduction of more sensitive cardiac troponin assays improved the early diagnosis of acute myocardial infarction (AMI). However, its diagnostic utility has never been tested in patients with kidney disease (KD), who are known to have elevated levels of cardiac troponins (cTn) already in the absence of AMI, which may lead to a lower diagnostic value of more sensitive cTn in this high-risk subgroup. Methods: We conducted an international multicenter study to examine the diagnostic accuracy of the Roche high-sensitivity (hs) cTnT assays in 2351 consecutive patients presenting to the emergency department with symptoms suggestive of AMI, of whom 396 (17%) were determined to have KD (MDRD GFR <60ml/min/1.73m2) and to derive the optimal cutoff value for the diagnosis of AMI in patients with KD. The diagnostic accuracy was further compared to a standard cTn assay (Roche Troponin T fourth generation). The final diagnosis was adjudicated by two independent cardiologists based on hs-cTnT using all available data. Results: AMI was the final diagnosis in 36% (n=143) of all KD-patients as compared to 18% in patients with normal kidney function (p<0.001). Among KD-patients with other diagnoses than AMI, baseline hs-cTnT-levels were elevated above the 99thpercentile in 68%, In patients with KD the diagnostic accuracy at presentation, quantified by the area under the receiver-operator-characteristic curve (AUC), was significantly greater for hs-cTnT as compared to the standard assay (AUC for hs-TnT, 0.87 vs. AUC for the standard assay, 0.82, p=0.002). In patients presenting within three hours after the onset of chest pain, hs-cTnT was still superior compared to the conventional cTnT (AUC 0.81 vs. 0.72, p=0.02). In KD, the optimal hs-cTnT cutoff derived from the ROC curve was 36 ng/l compared to 16 ng/l in patients with normal kidney function (official 99th percentile cutoff-value 14 ng/l, provided by the manufacturer). Conclusions: The Roche hs-cTnT assay has a high diagnostic accuracy also in KD-patients and is superior to conventional cTn-assays. Mild elevations are common in non-AMI patients. Of note, the test-specific optimal cutoff-level in KD-patients seems to be nearly three times as high as the standard 99th percentile level. ClinicalTrials.gov number, [NCT00470587][1]. [1]: /lookup/external-ref?link_type=CLINTRIALGOV&access_num=NCT00470587&atom=%2Fehj%2F34%2Fsuppl_1%2FP3128.atom