Electrocardiogram-based mortality prediction in patients with COVID-19 using machine learning.

BACKGROUND AND PURPOSE: The electrocardiogram (ECG) is frequently obtained in the work-up of COVID-19 patients. So far, no study has evaluated whether ECG-based machine learning models have added value to predict in-hospital mortality specifically in COVID-19 patients. METHODS: Using data from the CAPACITY-COVID registry, we studied 882 patients admitted with COVID-19 across seven hospitals in the Netherlands. Raw format 12-lead ECGs recorded within 72 h of admission were studied. With data from five hospitals (n = 634), three models were developed: (a) a logistic regression baseline model using age and sex, (b) a least absolute shrinkage and selection operator (LASSO) model using age, sex and human annotated ECG features, and (c) a pre-trained deep neural network (DNN) using age, sex and the raw ECG waveforms. Data from two hospitals (n = 248) was used for external validation. RESULTS: Performances for models a, b and c were comparable with an area under the receiver operating curve of 0.73 (95% confidence interval [CI] 0.65-0.79), 0.76 (95% CI 0.68-0.82) and 0.77 (95% CI 0.70-0.83) respectively. Predictors of mortality in the LASSO model were age, low QRS voltage, ST depression, premature atrial complexes, sex, increased ventricular rate, and right bundle branch block. CONCLUSION: This study shows that the ECG-based prediction models could be helpful for the initial risk stratification of patients diagnosed with COVID-19, and that several ECG abnormalities are associated with in-hospital all-cause mortality of COVID-19 patients. Moreover, this proof-of-principle study shows that the use of pre-trained DNNs for ECG analysis does not underperform compared with time-consuming manual annotation of ECG features.

Clinical characteristics of patients with suspected cardiac chest pain and angiographically normal coronary arteries in a secondary care hospital.

BACKGROUND: An important number of patients with suspected cardiac chest pain have non-obstructive coronary artery disease. Our purpose was to describe the clinical characteristics of patients with normal or near-normal coronary arteries in routine cardiological practice in a secondary care hospital. METHODS: In 2013, consecutive patients referred for invasive coronary angiography with suspected cardiac chest pain were analysed at a single-centre (Westfriesgasthuis, Hoorn, the Netherlands). Coronary arteries were defined as normal or near-normal if they showed no stenosis or only slight wall irregularities on visual assessment. Patients with a final non-cardiac diagnosis for the chest pain were excluded. RESULTS: A total of 558 patients were included. Of these, 151 (27%) showed normal or near-normal coronary arteries on visual assessment. This group of patients were significantly more often female (p < 0.001), younger (p < 0.001) and non-diabetic (p = 0.002). Forty percent of hospitalised patients who had normal or near-normal coronary arteries at coronary angiography showed an elevated troponin. CONCLUSION: In routine cardiological practice, around 1 out of 4 patients with suspected cardiac chest pain undergoing invasive angiography had normal or near-normal coronary arteries. We suggest that premenopausal women with suspected cardiac chest pain could be considered for non-invasive coronary imaging as a first step in clinical practice.

Takotsubo cardiomyopathy shortly following pacemaker implantation-case report and review of the literature.

We describe a patient with acute heart failure shortly after pacemaker implantation. With the documentation of typical dyskinesia of the apical segments with hyperdynamic contractility of the basal segments and a normal coronary angiogram, pacemaker implantation-induced Takotsubo cardiomyopathy was diagnosed. Supportive care was administered and within several days the patient's symptoms resolved. After several weeks, the left ventricular function had fully recovered. A review of the literature on Takotsubo cardiomyopathy after pacemaker implantation is presented.

Prognostic significance of nonfatal myocardial reinfarction in survivors of a first infarction.

BACKGROUND: Patients who develop a reinfarction are at increased risk for subsequent reinfarctions and death. However, follow-up studies in these patients are rare. OBJECTIVE: The purpose of this study was to examine the risk of mortality after a first myocardial reinfarction and to determine the independent contribution of nonfatal reinfarction to the risk of subsequent mortality. METHODS: The prognostic value of nonfatal reinfarction was assessed in a large series (n=3097) of patients with a first myocardial infarction who participated in the ASPECT trial, comparing coumarin or matching placebo. RESULTS: A second myocardial infarction was documented in 299 patients (82% Q-wave infarctions), 45 (15%) of which were fatal. Of the 254 nonfatal reinfarctions, 31 patients (12%) died during subsequent follow-up. After adjustment for baseline characteristics, the relative risks of nonfatal reinfarction for subsequent cardiac mortality at one month were: 2.90 (1.49-5.64), at one year: 2.50 (1.47-4.23) and at three years: 2.71 (1.77-4.17). Rates of death or a second reinfarction in patients who did not undergo a revascularisation procedure after a first reinfarction were almost three times higher than in patients who did have PTCA or bypass surgery after a reinfarction (38% versus 14%; p<0.0001). CONCLUSION: This study population with three-year follow-up confirms that nonfatal reinfarction carries a strong and independent risk for recurrent reinfarction and subsequent mortality. Thus, prevention of reinfarction by intensive treatment might contribute in reduction of mortality.