Abstract: BACKGROUND: The myocardial-ischaemic-injury-index (MI(3)) is a novel machine learning algorithm for the early diagnosis of type 1 non-ST-segment elevation myocardial infarction (NSTEMI). The performance of MI(3), both when using early serial blood draws (eg, at 1 h or 2 h) and in direct comparison with guideline-recommended algorithms, remains unknown. Our aim was to externally validate MI(3) and compare its performance with that of the European Society of Cardiology (ESC) 0/1h-algorithm. METHODS: In this secondary analysis of a multicentre international diagnostic cohort study, adult patients (age >18 years) presenting to the emergency department with symptoms suggestive of myocardial infarction were prospectively enrolled from April 21, 2006, to Feb 27, 2019 in 12 centres from five European countries (Switzerland, Spain, Italy, Poland, and Czech Republic). Patients were excluded if they presented with ST-segment-elevation myocardial infarction, did not have at least two serial high-sensitivity cardiac troponin I (hs-cTnI) measurements, or if the final diagnosis remained unclear. The final diagnosis was centrally adjudicated by two independent cardiologists using all available medical records, including serial hs-cTnI measurements and cardiac imaging. The primary outcome was type 1 NSTEMI. The performance of MI(3) was directly compared with that of the ESC 0/1h-algorithm. FINDINGS: Among 6487 patients, (median age 61.0 years [IQR 49.0-73.0]; 2122 [33%] female and 4365 [67%] male), 882 (13.6%) patients had type 1 NSTEMI. The median time difference between the first and second hs-cTnI measurement was 60.0 mins (IQR 57.0-70.0). MI(3) performance was very good, with an area under the receiver-operating-characteristic curve of 0.961 (95% CI 0.957 to 0.965) and a good overall calibration (intercept -0.09 [-0.2 to 0.02]; slope 1.02 [0.97 to 1.08]). The originally defined MI(3) score of less than 1.6 identified 4186 (64.5%) patients as low probability of having a type 1 NSTEMI (sensitivity 99.1% [95% CI 98.2 to 99.5]; negative predictive value [NPV] 99.8% [95% CI 99.6 to 99.9]) and an MI(3) score of 49.7 or more identified 915 (14.1%) patients as high probability of having a type 1 NSTEMI (specificity 95.0% [94.3 to 95.5]; positive predictive value [PPV] 69.1% [66.0-72.0]). The sensitivity and NPV of the ESC 0/1h-algorithm were higher than that of MI(3) (difference for sensitivity 0.88% [0.19 to 1.60], p=0.0082; difference for NPV 0.18% [0.05 to 0.32], p=0.016), and the rule-out efficacy was higher for MI(3) (11% difference, p<0.0001). Specificity and PPV for MI(3) were superior (difference for specificity 3.80% [3.24 to 4.36], p<0.0001; difference for PPV 7.84% [5.86 to 9.97], p<0.0001), and the rule-in efficacy was higher for the ESC 0/1h-algorithm (5.4% difference, p<0.0001). INTERPRETATION: MI(3) performs very well in diagnosing type 1 NSTEMI, demonstrating comparability to the ESC 0/1h-algorithm in an emergency department setting when using early serial blood draws. FUNDING: Swiss National Science Foundation, Swiss Heart Foundation, the EU, the University Hospital Basel, the University of Basel, Abbott, Beckman Coulter, Roche, Idorsia, Ortho Clinical Diagnostics, Quidel, Siemens, and Singulex.
Tags: *Algorithms, *Early Diagnosis, *Machine Learning, *Non-ST Elevated Myocardial Infarction/diagnosis, Academy of Medical Sciences, and the Gottfried and Julia, Aged, and speaker honoraria from Roche Diagnostics, Abbott, and Siemens, paid to their, and speaker honoraria from Siemens, outside the, and speaker honoraria or consulting honoraria, and speaker or consulting honoraria or research support from Edwards, Bangerter-Rhyner-Foundation, Bangerter-Rhyner-Foundation, and the Freiwillige Akademische Gesellschaft Basel, Biomarkers/blood, Clinical Diagnostics, and Orion Pharma, outside the submitted work. LK has, Cohort Studies, Emergency Service, Hospital, Europe, Female, Foundation (FF20079 and FF21103) and speaker's honoraria from Quidel, paid to, Foundation (P400PM_191037/1), the Prof Dr Max Cloetta Foundation, the Margarete, from Abbott, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol Myers, from the University of Basel and the Division of Internal Medicine, the Swiss, Humans, Idorsia, LSI Medience Corporation, Novartis, Ortho Diagnostics, Quidel, Roche,, institution and outside the submitted work. PB has received a research grant from, institution. All other authors declare no competing interests., Lifesciences, Boston Scientific, Medtronic, Abbott, Beckman Coulter, Bayer, Ortho, Male, Middle Aged, Myocardial Infarction/diagnosis, of Basel, the University Hospital Basel, Abbott, Beckman Coulter, Brahms,, Prospective Studies, received a research grant from the Swiss Heart Foundation, University of Basel,, Siemens, Singulex, and Sphingotec, Squibb, Idorsia, Novartis, Osler, Roche, and Sanofi, all paid to their, submitted work. TN has received research support from the Swiss National Science, the Swiss Academy of Medical Sciences, the Gottfried and Julia, the Swiss Heart Foundation (FF23062). CM reports receiving research support from, the Swiss National Science Foundation, the Swiss Heart Foundation, the University, their institution and outside the submitted work. JB has received research grants, Troponin I/blood, und Walter Lichtenstein-Stiftung (3MS1038), and the University Hospital Basel.
RSS 2.0