Personalized diagnosis in suspected myocardial infarction
Standard
Personalized diagnosis in suspected myocardial infarction. / Neumann, Johannes Tobias; Twerenbold, Raphael; Ojeda, Francisco; Aldous, Sally J; Allen, Brandon R; Apple, Fred S; Babel, Hugo; Christenson, Robert H; Cullen, Louise; Di Carluccio, Eleonora; Doudesis, Dimitrios; Ekelund, Ulf; Giannitsis, Evangelos; Greenslade, Jaimi; Inoue, Kenji; Jernberg, Tomas; Kavsak, Peter; Keller, Till; Lee, Kuan Ken; Lindahl, Bertil; Lorenz, Thiess; Mahler, Simon A; Mills, Nicholas L; Mokhtari, Arash; Parsonage, William; Pickering, John W; Pemberton, Christopher J; Reich, Christoph; Richards, A Mark; Sandoval, Yader; Than, Martin P; Toprak, Betül; Troughton, Richard W; Worster, Andrew; Zeller, Tanja; Ziegler, Andreas; Blankenberg, Stefan; ARTEMIS study group.
In: CLIN RES CARDIOL, Vol. 112, No. 9, 09.2023, p. 1288-1301.Research output: SCORING: Contribution to journal › SCORING: Journal article › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Personalized diagnosis in suspected myocardial infarction
AU - Neumann, Johannes Tobias
AU - Twerenbold, Raphael
AU - Ojeda, Francisco
AU - Aldous, Sally J
AU - Allen, Brandon R
AU - Apple, Fred S
AU - Babel, Hugo
AU - Christenson, Robert H
AU - Cullen, Louise
AU - Di Carluccio, Eleonora
AU - Doudesis, Dimitrios
AU - Ekelund, Ulf
AU - Giannitsis, Evangelos
AU - Greenslade, Jaimi
AU - Inoue, Kenji
AU - Jernberg, Tomas
AU - Kavsak, Peter
AU - Keller, Till
AU - Lee, Kuan Ken
AU - Lindahl, Bertil
AU - Lorenz, Thiess
AU - Mahler, Simon A
AU - Mills, Nicholas L
AU - Mokhtari, Arash
AU - Parsonage, William
AU - Pickering, John W
AU - Pemberton, Christopher J
AU - Reich, Christoph
AU - Richards, A Mark
AU - Sandoval, Yader
AU - Than, Martin P
AU - Toprak, Betül
AU - Troughton, Richard W
AU - Worster, Andrew
AU - Zeller, Tanja
AU - Ziegler, Andreas
AU - Blankenberg, Stefan
AU - ARTEMIS study group
N1 - © 2023. The Author(s).
PY - 2023/9
Y1 - 2023/9
N2 - BACKGROUND: In suspected myocardial infarction (MI), guidelines recommend using high-sensitivity cardiac troponin (hs-cTn)-based approaches. These require fixed assay-specific thresholds and timepoints, without directly integrating clinical information. Using machine-learning techniques including hs-cTn and clinical routine variables, we aimed to build a digital tool to directly estimate the individual probability of MI, allowing for numerous hs-cTn assays.METHODS: In 2,575 patients presenting to the emergency department with suspected MI, two ensembles of machine-learning models using single or serial concentrations of six different hs-cTn assays were derived to estimate the individual MI probability (ARTEMIS model). Discriminative performance of the models was assessed using area under the receiver operating characteristic curve (AUC) and logLoss. Model performance was validated in an external cohort with 1688 patients and tested for global generalizability in 13 international cohorts with 23,411 patients.RESULTS: Eleven routinely available variables including age, sex, cardiovascular risk factors, electrocardiography, and hs-cTn were included in the ARTEMIS models. In the validation and generalization cohorts, excellent discriminative performance was confirmed, superior to hs-cTn only. For the serial hs-cTn measurement model, AUC ranged from 0.92 to 0.98. Good calibration was observed. Using a single hs-cTn measurement, the ARTEMIS model allowed direct rule-out of MI with very high and similar safety but up to tripled efficiency compared to the guideline-recommended strategy.CONCLUSION: We developed and validated diagnostic models to accurately estimate the individual probability of MI, which allow for variable hs-cTn use and flexible timing of resampling. Their digital application may provide rapid, safe and efficient personalized patient care.TRIAL REGISTRATION NUMBERS: Data of following cohorts were used for this project: BACC ( www.CLINICALTRIALS: gov ; NCT02355457), stenoCardia ( www.CLINICALTRIALS: gov ; NCT03227159), ADAPT-BSN ( www.australianclinicaltrials.gov.au ; ACTRN12611001069943), IMPACT ( www.australianclinicaltrials.gov.au , ACTRN12611000206921), ADAPT-RCT ( www.anzctr.org.au ; ANZCTR12610000766011), EDACS-RCT ( www.anzctr.org.au ; ANZCTR12613000745741); DROP-ACS ( https://www.umin.ac.jp , UMIN000030668); High-STEACS ( www.CLINICALTRIALS: gov ; NCT01852123), LUND ( www.CLINICALTRIALS: gov ; NCT05484544), RAPID-CPU ( www.CLINICALTRIALS: gov ; NCT03111862), ROMI ( www.CLINICALTRIALS: gov ; NCT01994577), SAMIE ( https://anzctr.org.au ; ACTRN12621000053820), SEIGE and SAFETY ( www.CLINICALTRIALS: gov ; NCT04772157), STOP-CP ( www.CLINICALTRIALS: gov ; NCT02984436), UTROPIA ( www.CLINICALTRIALS: gov ; NCT02060760).
AB - BACKGROUND: In suspected myocardial infarction (MI), guidelines recommend using high-sensitivity cardiac troponin (hs-cTn)-based approaches. These require fixed assay-specific thresholds and timepoints, without directly integrating clinical information. Using machine-learning techniques including hs-cTn and clinical routine variables, we aimed to build a digital tool to directly estimate the individual probability of MI, allowing for numerous hs-cTn assays.METHODS: In 2,575 patients presenting to the emergency department with suspected MI, two ensembles of machine-learning models using single or serial concentrations of six different hs-cTn assays were derived to estimate the individual MI probability (ARTEMIS model). Discriminative performance of the models was assessed using area under the receiver operating characteristic curve (AUC) and logLoss. Model performance was validated in an external cohort with 1688 patients and tested for global generalizability in 13 international cohorts with 23,411 patients.RESULTS: Eleven routinely available variables including age, sex, cardiovascular risk factors, electrocardiography, and hs-cTn were included in the ARTEMIS models. In the validation and generalization cohorts, excellent discriminative performance was confirmed, superior to hs-cTn only. For the serial hs-cTn measurement model, AUC ranged from 0.92 to 0.98. Good calibration was observed. Using a single hs-cTn measurement, the ARTEMIS model allowed direct rule-out of MI with very high and similar safety but up to tripled efficiency compared to the guideline-recommended strategy.CONCLUSION: We developed and validated diagnostic models to accurately estimate the individual probability of MI, which allow for variable hs-cTn use and flexible timing of resampling. Their digital application may provide rapid, safe and efficient personalized patient care.TRIAL REGISTRATION NUMBERS: Data of following cohorts were used for this project: BACC ( www.CLINICALTRIALS: gov ; NCT02355457), stenoCardia ( www.CLINICALTRIALS: gov ; NCT03227159), ADAPT-BSN ( www.australianclinicaltrials.gov.au ; ACTRN12611001069943), IMPACT ( www.australianclinicaltrials.gov.au , ACTRN12611000206921), ADAPT-RCT ( www.anzctr.org.au ; ANZCTR12610000766011), EDACS-RCT ( www.anzctr.org.au ; ANZCTR12613000745741); DROP-ACS ( https://www.umin.ac.jp , UMIN000030668); High-STEACS ( www.CLINICALTRIALS: gov ; NCT01852123), LUND ( www.CLINICALTRIALS: gov ; NCT05484544), RAPID-CPU ( www.CLINICALTRIALS: gov ; NCT03111862), ROMI ( www.CLINICALTRIALS: gov ; NCT01994577), SAMIE ( https://anzctr.org.au ; ACTRN12621000053820), SEIGE and SAFETY ( www.CLINICALTRIALS: gov ; NCT04772157), STOP-CP ( www.CLINICALTRIALS: gov ; NCT02984436), UTROPIA ( www.CLINICALTRIALS: gov ; NCT02060760).
KW - Humans
KW - Angina Pectoris
KW - Biomarkers
KW - Myocardial Infarction/diagnosis
KW - ROC Curve
KW - Troponin I
KW - Troponin T
KW - Clinical Studies as Topic
U2 - 10.1007/s00392-023-02206-3
DO - 10.1007/s00392-023-02206-3
M3 - SCORING: Journal article
C2 - 37131096
VL - 112
SP - 1288
EP - 1301
JO - CLIN RES CARDIOL
JF - CLIN RES CARDIOL
SN - 1861-0684
IS - 9
ER -