A rapid non-invasive wearable device for assessing cardiac troponin.

William Franklin Peacock, Karina Melissa Soto-Ruiz, Allan S Jaffe, Brian R Tiffany, Simon A Mahler, Brian W Patterson, Alan Hb Wu, Robert Christenson
Author Information
  1. William Franklin Peacock: Department of Emergency Medicine, Baylor College of Medicine, Houston, TX.
  2. Karina Melissa Soto-Ruiz: Comprehensive Research Associates, LLC, Houston, TX.
  3. Allan S Jaffe: Department of Cardiology, Mayo Clinic, Rochester, MN.
  4. Brian R Tiffany: Department of Emergency Medicine, Chandler Regional Medical Center, Chandler, AZ.
  5. Simon A Mahler: Departments of Emergency Medicine, Implementation Science, and Epidemiology and Prevention, Wake Forest University School of Medicine, Winston-Salem, NC.
  6. Brian W Patterson: Berbee Walsh Department of Emergency Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI.
  7. Alan Hb Wu: Department of Laboratory Medicine, University of California, San Francisco, CA.
  8. Robert Christenson: Department of Laboratory Medicine, University of Maryland, Baltimore, MD.

Abstract

BACKGROUND: Chest pain, a common emergency department 35 (ED) presentation, requires rapid evaluation. Optical technology-based non-invasive wearable devices (Infrasensor, RCE, Carlsbad, CA) rapidly and transcutaneously assesses cardiac Troponin I (cTnI).
OBJECTIVES: To perform a pilot study describing the performance of the Infrasensor in cTnI defined cohorts.
METHODS: This was a 10-hospital prospective observational study in healthy US subjects with a normal cTnI, and in patients with an elevated local cTnI. Healthy subjects were without disease, defined by a negative questionnaire and bloodwork, had a 3-minute Infrasensor measurement and blood samples for high-sensitivity cardiac Troponin I (hs-cTnI), n-terminal pro-B-type natriuretic peptide (NTproBNP), creatinine, and glycosylated hemoglobin (HbA1c). Elevated cTnI's patients had the same Infrasensor and blood sample measurements. Using a cross validation technique, a cTnI based binary classification model that did, and did not, include age was trained with 80%, and validated on 20% (n=168; elevated hs-cTnI equally distributed across 5 folds) of the overall cohort.
RESULTS: Of 840 patients, 727 (87.5%) were non-elevated cTnI controls and the remainder, n=113, had elevated cTnI. Median (25th, 75th percentiles) age was 61 (52, 71) and 48 (32, 57) years for the elevated and healthy control cohorts, respectively. Overall, 50.5% were female, with 29.2% and 52.7% in the elevated and non-elevated troponin cohorts respectively. Overall, the sensitivity, specificity, negative and positive predictive values of the Infrasensor for identifying an elevated cTnI were 0.9, 0.7, 0.98 and 0.48 respectively, with a C-statistic of 0.90 (0.89-0.99).
CONCLUSIONS: The Infrasensor identifies elevated cTnI within 3 minutes of application.

Keywords

Word Cloud

Created with Highcharts 10.0.0cTnIelevatedInfrasensor0cardiaccohortspatientstroponinrespectivelyrapidnon-invasivewearableTroponinstudydefinedhealthysubjectsnegativebloodhs-cTnIage5%non-elevated5248OverallAcuteBACKGROUND:Chestpaincommonemergencydepartment35EDpresentationrequiresevaluationOpticaltechnology-baseddevicesRCECarlsbadCArapidlytranscutaneouslyassessesOBJECTIVES:performpilotdescribingperformanceMETHODS:10-hospitalprospectiveobservationalUSnormallocalHealthywithoutdiseasequestionnairebloodwork3-minutemeasurementsampleshigh-sensitivityn-terminalpro-B-typenatriureticpeptideNTproBNPcreatinineglycosylatedhemoglobinHbA1cElevatedcTnI'ssamplemeasurementsUsingcrossvalidationtechniquebasedbinaryclassificationmodelincludetrained80%validated20%n=168equallydistributedacross5foldsoverallcohortRESULTS:84072787controlsremaindern=113Median25th75thpercentiles61713257yearscontrol50female292%7%sensitivityspecificitypositivepredictivevaluesidentifying9798C-statistic9089-099CONCLUSIONS:identifieswithin3minutesapplicationdeviceassessingCoronarySyndromesMyocardialInfarctionRiskStratification

Similar Articles

Cited By

No available data.