ECG analysis in patients with acute coronary syndrome undergoing invasive management: rationale and design of the electrocardiography sub-study of the MATRIX trial

Electrocardiogram-based deep learning algorithm for the screening of obstructive coronary artery disease

BACKGROUND

Information on electrocardiogram (ECG) has not been quantified in obstructive coronary artery disease (ObCAD), despite the deep learning (DL) algorithm being proposed as an effective diagnostic tool for acute myocardial infarction (AMI). Therefore, this study adopted a DL algorithm to suggest the screening of ObCAD from ECG.

METHODS

ECG voltage-time traces within a week from coronary angiography (CAG) were extracted for the patients who received CAG for suspected CAD in a single tertiary hospital from 2008 to 2020. After separating the AMI group, those were classified into ObCAD and non-ObCAD groups based on the CAG results. A DL-based model adopting ResNet was built to extract information from ECG data in the patients with ObCAD relative to those with non-ObCAD, and compared the performance with AMI. Moreover, subgroup analysis was conducted using ECG patterns of computer-assisted ECG interpretation.

RESULTS

The DL model demonstrated modest performance in suggesting the probability of ObCAD but excellent performance in detecting AMI. The AUC of the ObCAD model adopting 1D ResNet was 0.693 and 0.923 in detecting AMI. The accuracy, sensitivity, specificity, and F1 score of the DL model for screening ObCAD were 0.638, 0.639, 0.636, and 0.634, respectively, while the figures were up to 0.885, 0.769, 0.921, and 0.758 for detecting AMI, respectively. Subgroup analysis showed that the difference between normal and abnormal/borderline ECG groups was not notable.

CONCLUSIONS

ECG-based DL model showed fair performance for assessing ObCAD and it may serve as an adjunct to the pre-test probability in patients with suspected ObCAD during the initial evaluation. With further refinement and evaluation, ECG coupled with the DL algorithm may provide potential front-line screening support in the resource-intensive diagnostic pathways.

Electrocardiographic patterns predict the presence of collateral circulation and in-hospital mortality in acute total left main occlusion

Background

Data on the clinical characteristics, electrocardiogram (ECG) findings and outcomes of patients with acute myocardial infarction (AMI) due to total unprotected left main (ULM) artery occlusion is limited.

Methods

Between 2009 and 2021, 44 patients with AMI due to total ULM occlusion underwent primary percutaneous coronary intervention (PCI) at our institution. The ECG, collateral circulation, clinical and procedural characteristics, and in-hospital mortality were retrospectively evaluated.

Results

Twenty five patients presented with shock and 18 patients had in-hospital mortality. Nineteen patients presented with ST-segment elevation myocardial infarction (STEMI), while 25 presented with non-ST-segment elevation myocardial infarction (NSTEMI). ST-segment elevation (STE) in I and STEMI were associated with the absence of collateral circulation, while STE in aVR was associated with its presence. In the NSTEMI group, patients with STE in both aVR and aVL showed more collateral filling of the left anterior descending coronary artery (LAD) territory, while patients with STE in aVR showed more collateral filling of the LAD and the left circumflex artery territory. Compared with total ULM occlusion, patients with partial ULM obstruction presented with more STE in aVR, less STE in aVR and aVL, and less STEMI. Shock, post-PCI TIMI 0–2 flow, non-STE in aVR, STEMI, and STE in I predicted in-hospital mortality. STEMI and the absence of collateral flow were significantly associated with shock.

Conclusions

STE in the precordial leads predicted the absence of collateral circulation while STE in aVR and STE in both aVR and aVL predicted different collateral filling territories in ULM occlusion. STE in I, non-STE in aVR, and STEMI predicted in-hospital mortality in these patients.

Clinical utility of aVR lead T-wave in electrocardiogram of patients with ST-elevation myocardial infarction

Background

aVR lead is often neglected in routine clinical practice largely because of its undefined clinical utility specifications. Nevertheless, positive T-wave in aVR lead has been reported to be associated with poor clinical outcomes in some cardiovascular diseases. This study aimed to prospectively investigate the prognostic value and clinical utility of T-wave amplitude in aVR lead in patients with acute ST-elevation myocardial infarction (STEMI).

Methods

A total of 340 STEMI patients admitted to a tertiary heart center were consecutively included. Patients were categorized into four strata, based on T wave amplitude in aVR lead in their admission ECG (i.e. < − 2, − 1 to − 2, − 1 to 0, and ≥ 0 mV). Patients’ clinical outcomes were also recorded and statistically analyzed.

Results

In-hospital mortality, re-hospitalization, and six-month-mortality significantly varied among four T wave strata and were higher in patients with a T wave amplitude of ≥ 0 mV (p 0.001–0.002). The groups of patients with higher T wave amplitude in aVR, had progressively increased relative risk (RR) of in-hospital mortality (RRs ≤ 0.01, 0.07, 1.00, 2.30 in four T wave strata, respectively). T wave amplitude in the cutoff point of − 1 mV exhibited a sensitivity and specificity of 95.83 (95% CI 78.88–99.89) and 49.68 (95% CI 44.04–55.33).

Conclusion

Our study demonstrated a significant association of positive T wave in aVR lead and adverse clinical outcomes in STEMI patients. Nevertheless, the clinical utility of T-wave amplitude at aVR lead is limited by its low discriminative potential toward prognosis of STEMI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12872-021-02335-5.

ST-segment resolution as a marker for severe myocardial fibrosis in ST-segment elevation myocardial infarction

Objective

To investigate the relationship between ST-segment resolution (STR) and myocardial scar thickness after percutaneous coronary intervention (PCI) in patients with ST-segment elevation myocardial infarction (STEMI).

Methods

Forty-two STEMI patients with single-branch coronary artery stenosis or occlusion were enrolled. ST-segment elevations were measured at emergency admission and at 24 h after PCI. Late gadolinium-enhanced cardiac magnetic resonance imaging (CMR-LGE) was performed 7 days after PCI to evaluate myocardial scars. Statistical analyses were performed to assess the utility of STR to predict the development of transmural (> 75%) or non-transmural (< 75%) myocardial scars, according to previous study.

Results

The sensitivity and specificity of STR for predicting transmural scars were 96% and 88%, respectively, at an STR cut-off value of 40.15%. The area under the curve was 0.925. Multivariate logistic proportional hazards regression analysis disclosed that patients with STR < 40.15% had a 170.90-fold higher probability of developing transmural scars compared with patients with STR ≥ 40.15%. Pearson correlation and linear regression analyses showed STR percentage was significantly associated with myocardial scar thickness and size.

Conclusion

STR < 40.15% at 24 h after PCI may provide meaningful diagnostic information regarding the extent of myocardial scarification in STEMI patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12872-021-02269-y.

Predictive Value of Serial ECGs in Patients with Suspected Myocardial Infarction

The electrocardiogram (ECG) is an important diagnostic tool for patients with suspected acute myocardial infarction (AMI). Current guidelines recommend serial ECGs in case of persisting symptoms. We aimed to analyze the predictive value of ischemic ECG-signs in patients with suspected AMI. Patients presenting to the emergency department with suspected AMI were included. All patients with ST-elevation AMI were excluded from analyses. Patients received 12-lead-ECG and high-sensitive Troponin T (hs-TnT)-measurement at admission and after 3 h. Four groups were defined: no ischemic signs in either ECG; new ischemic signs in the second ECG; resolved ischemic signs in the second ECG; and persistent ischemic signs in both ECGs. Patients were followed for 2 years to assess the composite endpoint of all-cause-mortality, AMI, and coronary revascularization. Using a 30-day landmark analysis, a Cox regression with ischemic signs as the variable of interest, adjusted by cardiovascular risk factors, was calculated. Of 1675 patients, 1321 showed no ischemic signs, in 25 new-, in 92 resolved- and in 237 patients, persistent ischemic signs were documented. Patients with persistent ischemic signs had significantly worse outcomes, compared to those without. Compared to no ischemic signs, adjusted hazard ratios for the combined endpoint were 0.81 (95% CI 0.20, 3.31; p-value = 0.77) for new-, 0.59 (95% CI 0.26, 1.34; p-value = 0.21) for resolved-, and 1.47 (95% CI 1.102, 2.13; p-value = 0.041) for persistent ischemic signs. In patients with suspected AMI, persistent ischemic ECG-signs are predictive of a higher rate of all-cause-mortality, AMI, and revascularization.

Facilitation Through Aggrastat or Cangrelor Bolus and Infusion Over PrasugreL: a MUlticenter Randomized Open-label Trial in PatientS with ST-elevation Myocardial InFarction Referred for PrimAry PercutaneouS InTERvention (FABOLUS FASTER) Trial: Design and Rationale : The FABOLUS FASTER Trial

Antithrombotic therapy is a critical component of the management of ST-elevation myocardial infarction (STEMI) patients treated with primary percutaneous coronary intervention (PCI). Rapid and profound inhibition of platelet reactivity has been shown to mitigate the ischemic risks and improve myocardial salvage. High residual platelet reactivity (HRPR) has been reported up to 4 or 6 h after loading dose of prasugrel or ticagrelor; therefore, multiple alternative strategies, including crushed or chewed oral tables or intravenous agents, have been investigated to provide a more rapid and sustained inhibition of platelet function and bridge the initial treatment gap. The FABOLUS FASTER is the first investigator-initiated, multicentre, open-label, prospective, randomized study to directly compare the pharmacodynamics effects of cangrelor, tirofiban, chewed or integer prasugrel. This study will add new insights in the management of antiplatelet therapy in patients with STEMI undergoing primary PCI and might be hypothesis-generating for future clinical trials in this field. The trial is registered on clinicaltrials.gov NCT02978040, and EudraCT 2017-001065-24.

Clinical Utility of the Inflammatory Factors Combined With Lipid Markers in the Diagnostic and Prognostic Assessment of Ischemic Stroke: Based on Logistic Regression Models

BACKGROUND

In this study, we developed novel logistic regression models for the diagnostic and prognostic assessment of ischemic stroke.

METHODS

A total of 288 ischemic stroke patients and 300 controls admitted to The First Affiliated Hospital of Soochow University were included in the testing group. Two validation groups from The Affiliated Kunshan Hospital of Jiangsu University and The Second Affiliated Hospital of Soochow University were included to assess our novel assessment models.

RESULTS

Results from the testing group indicated that the diagnostic assessment model for ischemic stroke prediction was: Logit(P) = 437.116 - 87.329 (Hypertension) - 89.700 (Smoking history) - 87.427 (Family history of ischemic stroke) - .090 (high-density lipoprotein cholesterol [HDL-C]) - 1.984 (low-density lipoprotein cholesterol [LDL-C]) - 17.005 (Lp(a)) - 15.486 (Apo A/Apo B), and the final prognostic assessment model of ischemic stroke was: Logit(P) = 458.437-92.343 (Hypertension) - 89.763 (Smoking history) + .251 (NLR) - .088 (HDL-C) - 1.994 (LDL-C) - 2.883 (hs-CRP) - .058 (IL-6) - 6.356 (TNF-α) - 16.485 (Lp(a)) - 17.658 (Apo A/Apo B). In the validation groups, our novel diagnostic assessment model showed good identification (with 87.5% sensitivity and 84.2% specificity in The Affiliated Kunshan Hospital of Jiangsu University, with 85.5% sensitivity and 89.0% specificity in The Second Affiliated Hospital of Soochow University). Moreover, our novel prognostic assessment model has a high value in identifying poor prognosis patients in the validation groups from The Affiliated Kunshan Hospital of Jiangsu University (χ² = 8.461, P = .004), and The Second Affiliated Hospital of Soochow University (χ² = 7.844, P = .005).

CONCLUSIONS

The diagnostic and prognostic assessment models we have established are of great value in the diagnosis and prognostic evaluation of ischemic stroke.