Robert H Christenson1, Show-Hong Duh2, Fred A Apple3, Richard Nowak4, W Frank Peacock5, A T Limkakeng6, Zohrab Bostanian7, Amin Mohammad8, James McCord9, Christopher R deFilippi10. 1. Department of Pathology, University of Maryland School of Medicine, Baltimore, MD, United States. Electronic address: rchristenson@umm.edu. 2. Department of Pathology, University of Maryland School of Medicine, Baltimore, MD, United States. 3. Department of Laboratory Medicine and Pathology, Hennepin County Medical Center of Hennepin Healthcare and University of Minnesota Minneapolis, Minneapolis, MN, United States. 4. Henry Ford Health System, Detroit, MI, United States. 5. Department of Emergency Medicine, Baylor College of Medicine, Houston, TX, United States. 6. Division of Emergency, Medicine, Duke University, School of Medicine, Durham, NC, United States. 7. Research & Development Institute, Inc., Calabasas, CA, United States. 8. Baylor Scott & White Healthcare, Texas A&M Health Science Center, Temple, TX, United States. 9. Henry Ford Hospital, Detroit, MI, United States. 10. Inova Heart and Vascular Institute, Falls Church, VA, United States.
Abstract
BACKGROUND: Cardiac troponin (cTn) is the keystone for diagnosis of acute myocardial infarction (AMI). We examined the analytical and diagnostic accuracy of the Atellica IM TnIH assay to determine high-sensitivity performance and appropriate diagnostic performance for clinical use. METHODS: Sex-specific 99th percentile upper reference limits (URLs) were determined for a healthy cohort of 1007 women and 1000 men using non-parametric statistics. High-sensitivity performance was assessed by examining if imprecision was ≤10% at sex-specific URLs and if ≥50% of cTnI values for each sex exceeded the assay's limit of detection (LoD) with the AACC Universal Sample Bank. Precision, high-dose hook effect, endogenous/exogenous interferences were examined with CLSI guidance. Clinical characterization was with 2494 suspected AMI subjects presenting to emergency departments across the United States. AMI was adjudicated by expert cardiologists and emergency medicine physicians. There were no comorbidity exclusions. RESULTS: 99th percentile URLs were 34 ng/L, 53 ng/L and 45 ng/L for the female, male and overall populations, respectively. Total imprecision was <5% from 12 ng/L to 16,000 ng/L; ≥55% of cTnI values for each sex exceeded the LoD. No high-dose hook or endogenous/exogenous interferences were identified. After 2.5-3.5 h post presentation the sensitivity and specificity were >90%; negative and positive predictive value were ≥98% and >60%, respectively. Non-AMI subjects with comorbidities and values exceeding 99th percentile URLs had absolute and percent change at 2-4 h that were lower than AMI patients with comorbidities (p = 0.001). CONCLUSION: The Atellica IM TnIH assay is a high-sensitivity method and demonstrates clinical performance appropriate for AMI diagnosis.
BACKGROUND: Cardiac troponin (cTn) is the keystone for diagnosis of acute myocardial infarction (AMI). We examined the analytical and diagnostic accuracy of the Atellica IM TnIH assay to determine high-sensitivity performance and appropriate diagnostic performance for clinical use. METHODS: Sex-specific 99th percentile upper reference limits (URLs) were determined for a healthy cohort of 1007 women and 1000 men using non-parametric statistics. High-sensitivity performance was assessed by examining if imprecision was ≤10% at sex-specific URLs and if ≥50% of cTnI values for each sex exceeded the assay's limit of detection (LoD) with the AACC Universal Sample Bank. Precision, high-dose hook effect, endogenous/exogenous interferences were examined with CLSI guidance. Clinical characterization was with 2494 suspected AMI subjects presenting to emergency departments across the United States. AMI was adjudicated by expert cardiologists and emergency medicine physicians. There were no comorbidity exclusions. RESULTS: 99th percentile URLs were 34 ng/L, 53 ng/L and 45 ng/L for the female, male and overall populations, respectively. Total imprecision was <5% from 12 ng/L to 16,000 ng/L; ≥55% of cTnI values for each sex exceeded the LoD. No high-dose hook or endogenous/exogenous interferences were identified. After 2.5-3.5 h post presentation the sensitivity and specificity were >90%; negative and positive predictive value were ≥98% and >60%, respectively. Non-AMI subjects with comorbidities and values exceeding 99th percentile URLs had absolute and percent change at 2-4 h that were lower than AMIpatients with comorbidities (p = 0.001). CONCLUSION: The Atellica IM TnIH assay is a high-sensitivity method and demonstrates clinical performance appropriate for AMI diagnosis.
Authors: Austin G Milton; Stephan Lau; Karlea L Kremer; Sushma R Rao; Emilie Mas; Marten F Snel; Paul J Trim; Deeksha Sharma; Suzanne Edwards; Mark Jenkinson; Timothy Kleinig; Erik Noschka; Monica Anne Hamilton-Bruce; Simon A Koblar Journal: BMJ Open Date: 2022-04-01 Impact factor: 2.692