Probabilistic classification of acute myocardial infarction from multiple cardiac markers

Logistic regression and Gaussian mixture model (GMM) classifiers have been trained to estimate the probability of acute myocardial infarction (AMI) in patients based upon the concentrations of a panel of cardiac markers. The panel consists of two new markers, fatty acid binding protein (FABP) and glycogen phosphorylase BB (GPBB), in addition to the traditional cardiac troponin I (cTnI), creatine kinase MB (CKMB) and myoglobin. The effect of using principal component analysis (PCA) and Fisher discriminant analysis (FDA) to preprocess the marker concentrations was also investigated. The need for classifiers to give an accurate estimate of the probability of AMI is argued and three categories of performance measure are described, namely discriminatory ability, sharpness, and reliability. Numerical performance measures for each category are given and applied. The optimum classifier, based solely upon the samples take on admission, was the logistic regression classifier using FDA preprocessing. This gave an accuracy of 0.85 (95% confidence interval: 0.78-0.91) and a normalised Brier score of 0.89. When samples at both admission and a further time, 1-6 h later, were included, the performance increased significantly, showing that logistic regression classifiers can indeed use the information from the five cardiac markers to accurately and reliably estimate the probability AMI. © Springer-Verlag London Limited 2008.

The impedance cardiogram recorded through two electrocardiogram/defibrillator pads as a determinant of cardiac arrest during experimental studies

OBJECTIVE: Laypersons are poor at emergency pulse checks (sensitivity 84%, specificity 36%). Guidelines indicate that pulse checks should not be performed. The impedance cardiogram (dZ/dt) is used to assess stroke volume. Can a novel defibrillator-based impedance cardiogram system be used to distinguish between circulatory arrest and other collapse states?

DESIGN: Animal study.

SETTING: University research laboratory.

SUBJECTS: Twenty anesthetized, mechanically ventilated pigs, weight 50-55 kg.

INTERVENTIONS: Stroke volume was altered by right ventricular pacing (160, 210, 260, and 305 beats/min). Cardiac arrest states were then induced: ventricular fibrillation (by rapid ventricular pacing) and, after successful defibrillation, pulseless electrical activity and asystole (by high-dose intravenous pentobarbitone).

MEASUREMENTS AND MAIN RESULTS: The impedance cardiogram was recorded through electrocardiogram/defibrillator pads in standard cardiac arrest positions. Simultaneously recorded electro- and impedance cardiogram (dZ/dt) along with arterial blood pressure tracings were digitized during each pacing and cardiac arrest protocol. Five-second epochs were analyzed for sinus rhythm (20 before ventricular fibrillation, 20 after successful defibrillation), ventricular fibrillation (40), pulseless electrical activity (20), and asystole (20), in two sets of ten pigs (ten training, ten validation). Standard impedance cardiogram variables were noncontributory in cardiac arrest, so the fast Fourier transform of dZ/dt was assessed. During ventricular pacing, the peak amplitude of fast Fourier transform of dZ/dt (between 1.5 and 4.5 Hz) correlated with stroke volume (r2 = .3, p < .001). In cardiac arrest, a peak amplitude of fast Fourier transform of dZ/dt of < or = 4 dB x ohm x rms indicated no output with high sensitivity (94% training set, 86% validation set) and specificity (98% training set, 90% validation set).

CONCLUSIONS: As a powerful clinical marker of circulatory collapse, the fast Fourier transformation of dZ/dt (impedance cardiogram) has the potential to improve emergency care by laypersons using automated defibrillators.

Demographic and temporal trends in out of hospital sudden cardiac death in belfast

Objective: To determine the epidemiology of out of hospital sudden cardiac death (OHSCD) in Belfast from 1 August 2003 to 31 July 2004.

Design: Prospective examination of out of hospital cardiac arrests by using the Utstein style and necropsy reports. World Health Organization criteria were applied to determine the number of sudden cardiac deaths.

Results: Of 300 OHSCDs, 197 (66%) in men, mean age (SD) 68 (14) years, 234 (78%) occurred at home. The emergency medical services (EMS) attended 279 (93%). Rhythm on EMS arrival was ventricular fibrillation (VF) in 75 (27%). The call to response interval (CRI) was mean (SD) 8 (3) minutes. Among patients attended by the EMS, 9.7% were resuscitated and 7.2% survived to leave hospital alive. The CRI for survivors was mean (SD) 5 (2) minutes and for non-survivors, 8 (3) minutes (p < 0.001). Ninety one (30%) OHSCDs were witnessed; of these 91 patients 48 (53%) had VF on EMS arrival. The survival rate for witnessed VF arrests was 20 of 48 (41.7%): all 20 survivors had VF as the presenting rhythm and CRI ? 7 minutes. The European age standardised incidence for OHSCD was 122/100 000 (95% confidence interval 111 to 133) for men and 41/100 000 (95% confidence interval 36 to 46) for women.

Conclusion: Despite a 37% reduction in heart attack mortality in Ireland over the past 20 years, the incidence of OHSCD in Belfast has not fallen. In this study, 78% of OHSCDs occurred at home.