A novel breakthrough in wrist-worn transdermal troponin-I-sensor assessment for acute myocardial infarction

Abstract

Aims Clinical differentiation of acute myocardial infarction (MI) from unstable angina and other presentations mimicking acute coronary syndromes (ACS) is critical for implementing time-sensitive interventions and optimizing outcomes. However, the diagnostic steps are dependent on blood draws and laboratory turn-around-times. We tested the clinical feasibility of a wrist-worn transdermal infra-red spectrophotometric sensor (Transdermal-ISS) in clinical practice and assessed the performance of a machine-learning algorithm for identifying elevated high-sensitivity cardiac Troponin-I (hs-cTnI) levels in patients hospitalized with ACS. Methods and Results We enrolled 238 patients hospitalized with ACS at five sites. The final diagnosis of MI (with or without ST-elevation) and unstable angina was adjudicated using ECG, cTn, echocardiography (regional wall motion abnormality), or coronary angiography. A Transdermal-ISS-derived deep-learning model was trained (three sites) and externally validated with hs-cTnI (one site), and echocardiography and angiography (two sites), respectively. The Transdermal-ISS model predicted elevated hs-cTnI with the area under the receiver operator characteristics of 0.90 (95% confidence interval [CI], 0.84-0.94; sensitivity, 0.86; specificity, 0.82) and 0.92 (95% CI, 0.80-0.98; sensitivity, 0.94; specificity, 0.64), for internal and external validation cohorts, respectively. In addition, the model predictions were associated with regional wall motion abnormalities (Odds Ratio [OR], 3.37; CI, 1.02-11.15; p=0.046) and significant coronary stenosis (OR, 4.69; CI, 1.27- 17.26; p = 0.019). Conclusion A wrist-worn transdermal infrared spectrophotometric sensor is clinically feasible for rapid, bloodless prediction of elevated hs-cTnI in real-world settings. It may have a role in establishing a point-of-care biomarker diagnosis of myocardial infarction and impact triaging patients with suspected ACS.

Clinical Biochemistry of Serum Troponin

“…Because the increased clinical sensitivity comes at the cost of poorer clinical specificity, interpreting test results should involve the appropriate application of conditional probability, based on differing pre-test and post-test probability in different settings of care [74], use of contextual clinical information such as echocardiographic, electrocardiographic, and physical exam data, and use of more population-specific cutoffs to define elevated Tn such as those that take age or race into consideration [102,103]. Emerging technologies for Tn measurement include those that incorporate new nanomaterials into the biosensor, make use of in vivo measurement of biomarkers, and those that are deployed in the form of Point-of-Care Testing devices [104] and wearable Tn sensors [87]. Understanding the molecular nature of Tn and its assays is essential for the evaluation and management of patients with and without chest pain, especially as the development of advanced Tn assays promises to herald increasingly sensitive and rapid detection.…”

“…Recently, a technique known as mid-infrared attenuated total reflection spectroscopy has enabled noninvasive measurements of TnI through the skin, achieving a sensitivity of 96.3% and a specificity of 60% for predicting elevated vs. non-elevated Tn, with the gold standard being an hs-TnI assay (Advia Centaur, Siemens Healthineers, Milan, Italy) [59]. Moreover, in a prospective trial of 238 patients diagnosed with MI by conventional means, the transdermal, noninvasive technology yielded an AUC of 0.90-0.92 and also predicted the presence of significant coronary stenosis (OR 4.69, 95% CI 1.27-17.26, p = 0.019) [87]. Whether these emerging technologies reach the market in the coming years will depend on further validation studies in diverse populations and regulatory clearance.…”

Infrared spectral analysis for the classification of patients with acute coronary syndrome. The questions run so deep

“…The performance of the ML algorithm was assessed using the standard descriptive statistical measures in an internal and external validation cohort of 134 and 45 patients, respectively, elegantly summarized in Table 2. 1 Overall an AUC-ROC of 90% and higher was observed. In addition, some ancillary analyses have been executed relating the elevated troponin levels with regional wall motion abnormalities on echo and coronary artery stenosis >50% diameter stenosis plus the effect of a number of clinical variables potentially affecting algorithm performance.…”

“…In the spirit of what DH may offer, the authors took a next logic step, namely the introduction of automated analysis of cardiac biomarkers using a ML algorithm. 1 In this paper, algorithm performance was assessed by a five-centre pilot study encompassing 238 patients with ACS of whom 20% with unstable angina, 21% non-ST segment elevation acute myocardial infarction (STEMI), and 57% STEMI ( guideline criteria ). The definition of elevated troponin level at a single time point ( yes/no ) was based upon a central lab reference value.…”

“…signal-to-noise ratio, the effect of stray light ) and those related to the algorithm(s) used for the data generation and processing. 1 , 2 , 4 These data are then transferred for stockage and analysis. The task of the ML algorithm is to discern in the multitude of data within this spectrum, which is the relation of absorbance ( Y -axis) with the wavelength or its inverse (wavenumber, X -axis), the differences between patients with disease or different diseases or states, which is not a simple task.…”