1
|
Rahman SSMM, Chen Z, Lalande A, Decourselle T, Cochet A, Pommier T, Cottin Y, Salomon M, Couturier R. Automatic classification of patients with myocardial infarction or myocarditis based only on clinical data: A quick response. PLoS One 2023; 18:e0285165. [PMID: 37146017 PMCID: PMC10162556 DOI: 10.1371/journal.pone.0285165] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/07/2022] [Accepted: 04/17/2023] [Indexed: 05/07/2023] Open
Abstract
BACKGROUND In acute cardiovascular disease management, the delay between the admission in a hospital emergency department and the assessment of the disease from a Delayed Enhancement cardiac MRI (DE-MRI) scan is one of the barriers for an immediate management of patients with suspected myocardial infarction or myocarditis. OBJECTIVES This work targets patients who arrive at the hospital with chest pain and are suspected of having a myocardial infarction or a myocarditis. The main objective is to classify these patients based solely on clinical data in order to provide an early accurate diagnosis. METHODS Machine learning (ML) and ensemble approaches have been used to construct a framework to automatically classify the patients according to their clinical conditions. 10-fold cross-validation is used during the model's training to avoid overfitting. Approaches such as Stratified, Over-sampling, Under-sampling, NearMiss, and SMOTE were tested in order to address the imbalance of the data (i.e. proportion of cases per pathology). The ground truth is provided by a DE-MRI exam (normal exam, myocarditis or myocardial infarction). RESULTS The stacked generalization technique with Over-sampling seems to be the best one providing more than 97% of accuracy corresponding to 11 wrong classifications among 537 cases. Generally speaking, ensemble classifiers such as Stacking provided the best prediction. The five most important features are troponin, age, tobacco, sex and FEVG calculated from echocardiography. CONCLUSION Our study provides a reliable approach to classify the patients in emergency department between myocarditis, myocardial infarction or other patient condition from only clinical information, considering DE-MRI as ground-truth. Among the different machine learning and ensemble techniques tested, the stacked generalization technique is the best one providing an accuracy of 97.4%. This automatic classification could provide a quick answer before imaging exam such as cardiovascular MRI depending on the patient's condition.
Collapse
Affiliation(s)
| | - Zhihao Chen
- FEMTO-ST Institute, CNRS UMR 6174, Univ. Bourgogne Franche-Comté, Belfort, France
| | - Alain Lalande
- ICMUB Laboratory, CNRS UMR 6302, Faculté de Médecine, Univ. Bourgogne Franche-Comté, Dijon, France
- Dijon Bourgogne University Hospital, Dijon, France
| | | | - Alexandre Cochet
- ICMUB Laboratory, CNRS UMR 6302, Faculté de Médecine, Univ. Bourgogne Franche-Comté, Dijon, France
- Dijon Bourgogne University Hospital, Dijon, France
| | | | - Yves Cottin
- Dijon Bourgogne University Hospital, Dijon, France
| | - Michel Salomon
- FEMTO-ST Institute, CNRS UMR 6174, Univ. Bourgogne Franche-Comté, Belfort, France
| | - Raphaël Couturier
- FEMTO-ST Institute, CNRS UMR 6174, Univ. Bourgogne Franche-Comté, Belfort, France
| |
Collapse
|
2
|
Wieczorkiewicz P, Supel K, Przybylak K, Kacprzak M, Zielinska M. Acute coronary syndrome versus acute myocarditis in young adults–value of speckle tracking echocardiography. PLoS One 2022; 17:e0271483. [PMID: 35939417 PMCID: PMC9359587 DOI: 10.1371/journal.pone.0271483] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/20/2022] [Accepted: 07/04/2022] [Indexed: 11/19/2022] Open
Abstract
Purpose Comparing myocarditis with an acute coronary syndrome (ACS)-like presentation and acute myocardial infarction (AMI) poses an important clinical challenge. The purpose of the study was to investigate the diagnostic value of the clinical, laboratory and especially echocardiographic characteristics including speckle tracking echocardiography (STE) of patients with ACS-like myocarditis and AMI. Methods We conducted a retrospective analysis comparing 69 symptomatic patients (≤ 45 years old), hospitalized at the Department of Interventional Cardiology (Medical University of Lodz, Poland) between April 2014 and June 2021 with an initial diagnosis of ST-segment elevation myocardial infarction. Results 37 patients with the cardiac magnetic resonance–confirmed acute myocarditis and 32 patients diagnosed with AMI based on the clinical presentation, electrocardiogram and the presence of a culprit lesion on the coronary angiography were analysed including echocardiography parameters. On STE analysis an average global longitudinal (GLS), radial and circumferential strain including three—layers observation were significantly lower (absolute value) in patients with AMI versus acute myocarditis (p<0.05). There was no significant difference in Endo/Epi ratio (p = 0.144) between the groups. An average GLS < (-17.5) represented the optimal cut-off value for the myocarditis diagnosis. Conclusion In patients with AMI a significant reduction of global and three-layers strains compared to patients with myocarditis was detected. Furthermore, our analysis also confirmed the discriminative pattern of myocardial injury between the groups.
Collapse
Affiliation(s)
- Paulina Wieczorkiewicz
- The Department of Interventional Cardiology, Medical University of Lodz, Lodz, Poland
- * E-mail:
| | - Karolina Supel
- The Department of Interventional Cardiology, Medical University of Lodz, Lodz, Poland
| | - Katarzyna Przybylak
- The Department of Interventional Cardiology, Medical University of Lodz, Lodz, Poland
| | - Michal Kacprzak
- The Department of Interventional Cardiology, Medical University of Lodz, Lodz, Poland
| | - Marzenna Zielinska
- The Department of Interventional Cardiology, Medical University of Lodz, Lodz, Poland
| |
Collapse
|
3
|
Wieczorkiewicz P, Przybylak K, Supel K, Kidawa M, Zielinska M. Symptomatic Young Adults with ST-Segment Elevation-Acute Coronary Syndrome or Myocarditis: The Three-Factor Diagnostic Model. J Clin Med 2022; 11:jcm11040916. [PMID: 35207189 PMCID: PMC8877187 DOI: 10.3390/jcm11040916] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/17/2021] [Revised: 01/29/2022] [Accepted: 02/07/2022] [Indexed: 02/04/2023] Open
Abstract
Myocarditis may mimic myocardial infarction (MI) due to a similar clinical presentation, including chest pain, electrocardiography changes, and laboratory findings. The purpose of the study was to investigate the diagnostic value of clinical, laboratory, and electrocardiography characteristics of patients with acute coronary syndrome - like myocarditis and MI. We analysed 90 patients (≤45 years old) with an initial diagnosis of ST-segment elevation myocardial infarction; 40 patients (44.4%), through the use of cardiac magnetic resonance, were confirmed to have myocarditis, and 50 patients (55.6%) were diagnosed with MI. Patients with myocarditis were younger and had fewer cardiovascular risk factors than those with MI. The cutoff value distinguishing between myocarditis and MI was defined as the age of 36 years. The history of recent infections (82.5% vs. 6%) and C-reactive protein (CRP) levels on admission (Me 45.9 vs. 3.4) was more associated with myocarditis. Further, the QTc interval was inversely correlated with the echocardiographic ejection fraction in both groups but was significantly longer in patients with MI. Non-invasive diagnostics based on clinical features and laboratory findings are basic but still essential tools for differentiation between MI and myocarditis. The three-factor model including the criteria of age, abnormal CRP, and history of recent infections might become a valuable clinical indication.
Collapse
|
4
|
Role of thrombin generation assays in the diagnosis of acute myocarditis and non-ST myocardial infarction. J Thromb Thrombolysis 2021; 50:144-150. [PMID: 31754904 DOI: 10.1007/s11239-019-01996-6] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 10/25/2022]
Abstract
Myocarditis and myocardial infarction share a common clinical characteristics despite significant differences in etiology and pathogenesis. Current guidelines recommend using cardiac magnetic resonance imaging (MRI) or endocardial biopsy for a definite diagnosis; however, these guidelines are not fully implemented due to the high cost and low availability. We used a thrombin generation assay and simple blood test to characterize both diseases. We conducted a cross-sectional study from April to December 2018. Patients with initial clinical suspicions of non-ST elevation myocardial infarction (NSTEMI) or myocarditis were eligible. All patients were recruited prior to anticoagulant treatment. Patients in both groups underwent acceptable standard clinical evaluation. Twenty-eight patients were enrolled; 12 patients in the NSTEMI group and 16 in the myocarditis group. Patients in the NSTEMI group were significantly older than those in the myocarditis group (64.25 ± 9.67 vs. 37.94 ± 19.66 years, p < 0.01, respectively) with a higher prevalence of hyperlipidemia, diabetes mellitus, and ischemic heart disease (p < 0.01 for all). There was no difference between the groups regarding INR, PT, aPTT, and serum levels of creatinine, urea, CPK, troponin, and fibrinogen. Endogenous thrombin potential (ETP), which represents the total thrombin concentration in the plasma, was significantly higher in the myocarditis group than in the NSTEMI group (2091.88 ± 336.41 vs. 1860.75 ± 438.02 nM × min, p < 0.03). Myocarditis and myocardial infarction have a different pattern of thrombin generation Thrombogram. The myocarditis group had significantly higher plasma ETP than the NSTEMI group. This finding requires further evaluation to define a numerical threshold, thus avoiding invasive or expensive assessment of myocarditis.
Collapse
|
5
|
Haouzi A, Ahmed A. ‘Clinically suspected myocarditis with pseudoinfarct presentation’ complicated with left ventricular aneurysm. BMJ Case Rep 2018; 2018:bcr-2017-222114. [DOI: 10.1136/bcr-2017-222114] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/04/2022] Open
|
6
|
Kytö V, Sipilä J, Rautava P. Acute myocardial infarction or acute myocarditis? Discharge registry-based study of likelihood and associated features in hospitalised patients. BMJ Open 2015; 5:e007555. [PMID: 26009575 PMCID: PMC4452743 DOI: 10.1136/bmjopen-2014-007555] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/21/2022] Open
Abstract
OBJECTIVE To evaluate the likelihood of and patient features associated with acute myocardial infarction (AMI) versus acute myocarditis in different population segments. DESIGN Nationwide, multihospital observational retrospective registry study of 9.6 years in Finland. PARTICIPANTS All consecutive patients aged ≥18 years hospitalised with a primary diagnosis of AMI (n=89 399) or acute myocarditis (n=2131) in 22 hospitals with a coronary catheterisation laboratory. PRIMARY OUTCOME MEASURES Likelihood of AMI versus acute myocarditis and associated patient features. RESULTS Men were over-represented in patients with AMI (59.8%) and in patients with acute myocarditis (76.1%). Age distributions of AMI and acute myocarditis were opposite as a majority of patients with myocarditis were aged 18-29 years, while the number of patients with AMI increased gradually up to 80 years of age. Patients aged 18-29 years were more likely to have acute myocarditis as the cause of hospitalisation (relative risk (RR)=11.4; 95% CI 7.6 to 16.1 for myocarditis, p<0.0001), but after 30 years of age the likelihood of infarction was higher with exponentially increasing RR for AMI. In youngest patients (18-29 years), the likelihood of AMI was higher in women, but men had higher odds for AMI after 40 years of age. Overall, men had OR of 1.97 (95% CI 1.74 to 2.23, p<0.0001) for AMI versus myocarditis when compared with women. Hypercholesterolaemia, chronic coronary artery disease, diabetes and hypertension predicted AMI in multivariate analysis. Odds for myocarditis were significantly higher if the patient had an otolaryngeal infection (OR 18.13; 95% CI 8.96 to 36.67, p<0.0001). CONCLUSIONS Acute myocarditis is more common than AMI in hospitalised patients aged 18-29 years, but the risk of AMI increases exponentially thereafter. Hypercholesterolaemia, diabetes and hypertension predict AMI regardless of age and gender.
Collapse
Affiliation(s)
- Ville Kytö
- Heart Center, Turku University Hospital, Turku, Finland
- PET Center, University of Turku, Turku, Finland
- Department of Medicine, University of Turku, Turku, Finland
| | - Jussi Sipilä
- Division of Clinical Neurosciences, Department of Neurology, Turku University Hospital, Turku, Finland
- Department of Neurology, University of Turku, Turku, Finland
| | - Päivi Rautava
- Clinical Research Center, Turku University Hospital, Turku, Finland
- Department of Public Health, University of Turku, Turku, Finland
| |
Collapse
|
7
|
Kawecki D, Morawiec B, Monney P, Pellaton C, Wojciechowska C, Jojko J, Basiak M, Przywara-Chowaniec B, Fournier S, Nowalany-Kozielska E, Schwitter J, Muller O. Diagnostic contribution of cardiac magnetic resonance in patients with acute coronary syndrome and culprit-free angiograms. Med Sci Monit 2015; 21:171-80. [PMID: 25604184 PMCID: PMC4345916 DOI: 10.12659/msm.892296] [Citation(s) in RCA: 13] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/21/2014] [Accepted: 09/23/2014] [Indexed: 01/17/2023] Open
Abstract
BACKGROUND In spite of robust knowledge about underlying ischemic myocardial damage, acute coronary syndromes (ACS) with culprit-free angiograms raise diagnostic concerns. The present study aimed to evaluate the additional value of cardiac magnetic resonance (CMR) over commonly available non-CMR standard tests, for the differentiation of myocardial injury in patients with ACS and non-obstructed coronary arteries. MATERIAL/METHODS Patients with ACS, elevated hs-TnT, and a culprit-free angiogram were prospectively enrolled into the study between January 2009 and July 2013. After initial evaluation with standard tests (ECG, echocardiography, hs-TnT) and provisional exclusion of acute myocardial infarction (AMI) in coronary angiogram, patients were referred for CMR with the suspicion of myocarditis or Takotsubo cardiomyopathy (TTC). According to the result of CMR, patients were reclassified as having myocarditis, AMI, TTC, or non-injured myocardium as assessed by late gadolinium enhancement. RESULTS Out of 5110 patients admitted with ACS, 75 had normal coronary angiograms and entered the study; 69 of them (92%) were suspected for myocarditis and 6 (8%) for TTC. After CMR, 49 patients were finally diagnosed with myocarditis (65%), 3 with TTC (4%), 7 with AMI (9%), and 16 (21%) with non-injured myocardium. The provisional diagnosis was changed or excluded in 23 patients (31%), with a 9% rate of unrecognized AMI. CONCLUSIONS The study results suggest that the evaluation of patients with ACS and culprit-free angiogram should be complemented by a CMR examination, if available, because the initial work-up with non-CMR tests leads to a significant proportion of misdiagnosed AMI.
Collapse
Affiliation(s)
- Damian Kawecki
- 2 Department of Cardiology, Silesian Medical University, Zabrze, Poland
| | - Beata Morawiec
- 2 Department of Cardiology, Silesian Medical University, Zabrze, Poland
- Department of Cardiology, University Hospital, Lausanne, Switzerland
| | - Pierre Monney
- Department of Cardiology, University Hospital, Lausanne, Switzerland
- Center of Cardiac Magnetic Resonance, University Hospital, Lausanne, Switzerland
| | - Cyril Pellaton
- Department of Cardiology, University Hospital, Lausanne, Switzerland
- Center of Cardiac Magnetic Resonance, University Hospital, Lausanne, Switzerland
| | | | - Joanna Jojko
- 2 Department of Cardiology, Silesian Medical University, Zabrze, Poland
| | - Marcin Basiak
- Department of Internal Medicine and Clinical Pharmacology, Silesian Medical University, Katowice, Poland
| | | | - Stephane Fournier
- Department of Cardiology, University Hospital, Lausanne, Switzerland
| | | | - Juerg Schwitter
- Department of Cardiology, University Hospital, Lausanne, Switzerland
- Center of Cardiac Magnetic Resonance, University Hospital, Lausanne, Switzerland
| | - Olivier Muller
- Department of Cardiology, University Hospital, Lausanne, Switzerland
| |
Collapse
|
8
|
Florian A, Schäufele T, Ludwig A, Rösch S, Wenzelburger I, Yildiz H, Sechtem U, Yilmaz A. Diagnostic value of CMR in young patients with clinically suspected acute myocarditis is determined by cardiac enzymes. Clin Res Cardiol 2014; 104:154-63. [DOI: 10.1007/s00392-014-0770-7] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/04/2014] [Accepted: 10/01/2014] [Indexed: 11/28/2022]
|