Measures of cardiac troponin (cTn) may have lower specificity for myocardial infarction in patients with CKD. We examined the diagnostic accuracy of baseline and serial high-sensitivity cTnI (hs-cTnI) measurements for myocardial infarction and 30- and 180-day mortality according to renal function. hs-cTnI was measured (Abbott assay) using sex-specific 99th percentiles (women, 16 ng/L; men, 34 ng/L) in 1555 adults presenting to the emergency department with symptoms suggesting ischemia (NCT02060760). Myocardial infarction was adjudicated along universal definition classification. Renal function did not significantly affect sensitivity or negative predictive values. Specificity decreased with impaired renal function from 93%-95% with normal function (eGFR$90 ml/min per 1.73 m2; n=722) to 57%-61% with severely impaired renal function (eGFR,30 ml/min per 1.73 m2; n=81) and 40%-41% on dialysis (n=78). Positive predictive values decreased with decreasing renal function from 51%-57% with normal function to 27%-42% with severely impaired function and 15%-32% on dialysis. Receiver operating characteristic curve areas trended lower at baseline and 3 hours with renal impairment. Mortality increased significantly with increasing hs-cTnI tertile (1.3%, 6.0%, and 10.4%, respectively). Patients with hs-cTnI concentration exceeding concentrations in the 99th percentiles had a mortality rate (11.7%) significantly higher than that of patients with concentrations between 99th percentile concentrations and limit of detection (6.2%) or below limit of detection (1.1%). Renal dysfunction and dialysis reduced the rule-in performance but not the rule-out performance of hs-cTnI for myocardial infarction, and mortality increased in patients with higher hs-cTnI concentrations and any level of renal dysfunction.
Bibliographical noteFunding Information:
This study was supported in part by a grant from Abbott Diagnostics and the Minneapolis Medical Research Foundation.
Copyright © 2018 by the American Society of Nephrology