Roger Dzwonczyk

Median frequency — a new parameter for predicting defibrillation success rate

STUDY HYPOTHESIS Current American Heart Association guidelines recommend immediate defibrillation of ventricular fibrillation. When this is unsuccessful, there are no guidelines to help determine the optimum time at which to defibrillate after the administration of an alpha-adrenergic agonist. Previous studies have shown that the median frequency of the ventricular fibrillation ECG signal correlates with myocardial perfusion during CPR. We hypothesized that median frequency could predict the success of defibrillation and thus accurately determine the most appropriate time at which to defibrillate during ventricular fibrillation. STUDY POPULATION Twenty-two mixed-breed swine weighing more than 15 kg were studied. METHODS Ventricular fibrillation was induced electrically, and the ventricular fibrillation ECG signal was analyzed using fast Fourier analysis. After ten minutes of ventricular fibrillation, mechanical CPR was begun. After three minutes of CPR, the animals received one of three alpha-adrenergic agonists and CPR was continued. Defibrillation was attempted three and one-half minutes after drug administration. The average median frequency 20 seconds before defibrillation was calculated. Sensitivity and specificity of median frequency with respect to defibrillation success were determined. RESULTS A median frequency of 9.14 Hz had a sensitivity of 100% and a specificity of 92.31% in predicting the results of defibrillation in this model. CONCLUSION The median frequency may serve as a valuable parameter to guide defibrillation therapy during ventricular fibrillation.

Signal analysis of the human electrocardiogram during ventricular fibrillation : frequency and amplitude parameters as predictors of successful countershock

STUDY OBJECTIVE To determine whether there is information in the human ventricular fibrillation (VF) ECG signal that is predictive of successful countershock. METHODS We carried out a retrospective analysis of ECG signals recorded during out-of-hospital treatment of adult patients in VF. Four parameters--centroid frequency (FC), peak power frequency (FP), average segment amplitude (SA), and average wave amplitude (WA)--were extracted from the recorded ECG signal immediately before each countershock and compared with countershock outcome. RESULTS The outcome of each countershock (total, 128 countershocks) administered to 55 patients in VF was determined from available emergency medical services data sheets and time-domain ECG signal and voice recordings. The original 4-second time-domain ECG segment immediately before the countershock was used to extract SA and WA. The 4-second ECG segment immediately before each countershock was transformed into the frequency domain by means of Fourier analysis, and the parameters FC and FP were extracted from the result. These parameters were compared with countershock outcome by means of Kolmogrov-Smirnov analysis. Sensitivity and specificity of these parameters, as well as receiver operating characteristic curves, were constructed. FC was statistically higher for successful countershocks (FC, 5.48 +/- .67 Hz) than for successful countershocks (FC, 4.85 +/- 1.16 Hz; P=.012). We found no statistical difference for FP (P=.066), SA (P=.549), and WA (P =.337). FP and FC, when used in combination and in certain ranges (3.5 Hz < or = FP < or = 7.75 Hz and 3.86 Hz < or = FC < or = 6.12 Hz) had a sensitivity of 100% and a specificity of 47.1% in predicting successful countershock. The probabilities of predicting countershock outcome for FC, FP, SA, and WA were .72, .70, .52, and .53, respectively. CONCLUSION FC and FP are predictive of countershock outcome for patients in VF and hold the potential to guide therapy during cardiac arrest.

Estimating the duration of ventricular fibrillation

As the duration of time between the onset of ventricular fibrillation and the application of defibrillation (downtime) increases, the rate of successful resuscitation decreases. Results of recent animal studies suggest that the rate of successful resuscitation may be increased after a prolonged cardiorespiratory arrest when pharmacologic therapy is instituted before defibrillation. An accurate estimation of downtime could be critical in selecting the most appropriate therapeutic intervention. The purpose of our study was to determine whether changes in the frequency or amplitude of the ventricular fibrillation ECG signal during cardiac arrest could be used to estimate downtime. We characterized the dynamics of both total power and frequency distribution of the power in the ECG during ventricular fibrillation in 11 swine to determine whether enough information existed in either parameter to estimate downtime. The median frequency of the power spectrum was used to track power distribution. Both parameters followed a dynamic, repeatable pattern. However, median frequency showed less intersubject variability than did total power. A mathematical model of median frequency was developed and used with data obtained from ten additional swine to estimate downtime. The model estimated downtime to within 1.3 minutes of actual downtime between one and ten minutes of ventricular fibrillation. Our study has identified a new, potentially useful parameter for studying various management strategies in ventricular fibrillation as a function of downtime.

The median frequency of the ECG during ventricular fibrillation: its use in an algorithm for estimating the duration of cardiac arrest

The dynamics of the frequency distribution in the power spectrum of the ECG recorded form 11 swine during ventricular fibrillation (VF) are characterized to determine whether enough information exists in this domain to estimate downtime (DT). The median frequency (FM) of the power spectrum is used to track the frequency distribution. The FM follows a dynamic repeatable course during the first 10 min of VF. Intersubject variability is small. The FM data of the 11 subjects are modeled with a set of first-order polynomial equations, and the algorithm is tested with data from an additional ten subjects. The algorithm predicts VF duration with an average error of -0.86 min; 71.5% of the predictions fell within the 95% confidence limits of the model. A signal processing tool is identified which may be useful in the prehospital treatment of VF.<<ETX>>

Frequency analysis of the human and swine electrocardiogram during ventricular fibrillation

Recent studies in swine have suggested that estimating the duration of ventricular fibrillation (VF) could have important implications regarding the selection of the best therapeutic intervention during cardiopulmonary resuscitation (CPR). Successful defibrillation resulting in a pulsatile rhythm is more likely with VF of short duration, whereas VF of longer duration may require interventions designed to augment myocardial blood flow prior to defibrillation. Duration of VF has been estimated in a swine model by modelling the median frequency (FM) of the VF ECG signal. The purpose of this study was to characterize the time course of the FM ECG signal in humans during VF and compare the characteristics of the human VF ECG signal to that previously described in swine. Seven two-channel human VF ECG recordings were analyzed via fast Fourier transformation. The FM was extracted from each four-second segment of the recordings and plotted versus time. The FM in humans followed a repeatable time course during VF. The human data revealed an FM which had two peaks with subsequent gradual decline. The data in swine revealed an FM during VF which decreased initially then increased to a peak followed by a gradual decline. Our preliminary results demonstrate that a characteristic median frequency exists in humans which could be used to estimate the duration of VF.

Myocardial electrical impedance responds to ischemia and reperfusion in humans

Myocardial electrical impedance (MEI) is correlated to ischemia and reperfusion of the heart muscle. The entire body of work with MEI to this point has been carried out in animal subjects in vivo and excised tissue samples. In this study, we measured MEI clinically for the first time in human patients who were undergoing off-pump coronary artery bypass (OPCAB) surgery. MEI was measured with a monitor designed in this laboratory and approved by the FDA for use on human subjects. Our patient population (n=18) had a 70%-100% stenosis of the diseased coronary artery targeted for bypass. We measured MEI continuously during surgery and at 3, 6, 24, and 72 h postoperatively from two temporary pacing electrodes attached to the heart muscle. Absolute baseline impedance ranged from 173 to 729 /spl Omega/. MEI increased with occlusion of the diseased artery prior to bypass. The percent increase from baseline was inversely correlated to the percent stenosis of the diseased artery. MEI decreased below baseline immediately on reperfusion following bypass in all patients and continued decreasing over the measurement period. MEI is a reliable clinical indicator of ischemia and reperfusion in humans and may indicate the effectiveness of coronary artery surgery. The parameter may have monitoring and diagnostic value in heart disease in humans.

An evaluation of a new two-electrode myocardial electrical impedance monitor for detecting myocardial ischemia.

The objective of this study was to determine the efficacy of a two-electrode myocardial electrical impedance (MEI) monitor in reproducibly detecting induced myocardial ischemia by comparing MEI changes with hemodynamic changes, including sonomicrometric changes. With institutional approval, 80 dogs were anesthetized with sodium thiamylal, intubated, ventilated, and had venous, arterial, and pulmonary artery catheters placed. Medial sternotomy was performed to facilitate myocardial exposure and allow the left anterior descending coronary artery (LAD) to be isolated. Two pacing electrodes were attached to the myocardium to measure MEI with a monitor. Seventy dogs were randomly assigned to the 15, 30, 45, 60, or 120 min LAD occlusion group. Sonomicrometric transducers were attached to the myocardium of the ten remaining dogs and their LAD was occluded for 36 min. MEI increased immediately after LAD occlusion to a level significantly more (P <0.05) than baseline and returned to the baseline level upon reperfusion. Twenty dogs developed ventricular fibrillation with no attempts at resuscitation. MEI changes paralleled the sonomicrometric changes expected with ischemia. No significant cardiovascular hemodynamic changes were found with less than 45 min of LAD occlusion. Sixty and 120 min LAD occlusion resulted in significant decreases in cardiac output. The results of these experiments demonstrate that the two-electrode MEI monitor reproducibly changes in response to myocardial ischemia. Implications We used dogs to determine if we could measure myocardial ischemia using a device that measures impedance and demonstrated that a two-electrode myocardial electrical impedance monitor reliably reflected changes induced by myocardial ischemia.

Description and validation of a technique for the removal of ECG contamination from diaphragmatic EMG signal

THE DIAPHRAGMATIC electromyogram (EMG) conveys important information regarding the respiratory control mechanisms. For example, the intensity of the EMG signal, measured by variables such as the mean rectified EMG (MREMG) and power (PWR), is a good indicator of the magnitude of the neural drive to the diaphragm (EVANICH et aL, 1977; LYNNEDAVIES, 1979). The frequency content of the signal, measured by variables such as the zero crossing frequency (ZC), median frequency (MF), mean power frequency (MPF) and expected zero crossing frequency (E[ZC]), conveys information regarding changes in the activation patterns of the diaphragmatic motor units (ScHWEITZER et al., 1979; HAGG, 1991; 1992). A major obstacle encountered during analysis of the diaphragmatic EMG signal is electrocardiogram (ECG) contamination. This causes an increase in the power content of the signal and a distortion of its frequency content (SCHWEITZER e t al., 1979). Early techniques employed to reduce the level of ECG contamination include amplitude clipping and highpass filtering, neither of which have proved effective (SCHWEITZER et al., 1979; MUELLER and LOURENqO, 1968). At present, the method most frequently used is the gating technique (SCHWEITZER et al., 1979; CHOUKROUN et aL, 1990; LANTZu et aL, 1991; SHARP et aL, 1993), which involves the removal of a section of the EMG signal centred around each QRS

Clinical and economic outcomes of low-field intraoperative MRI-guided tumor resection neurosurgery

To compare low‐field (0.15 T) intraoperative magnetic resonance imaging (iMRI)‐guided tumor resection with both conventional magnetic resonance imaging (cMRI)‐guided tumor resection and high‐field (1.5 T) iMRI‐guided resection from the clinical and economic point of view.

Magnetic field threshold for accurate electrocardiography in the MRI environment.

Although the electrocardiogram is known to be nondiagnostic within the bore of any high‐field magnet due to the magnetohydrodynamic effect, there are an increasing number of applications that require accurate electrocardiogram monitoring of a patient inside the MRI room but outside of the magnet bore. Magnetohydrodynamic effects on the ST segment of the electrocardiogram waveform were investigated in six subjects at magnetic field strengths ranging from 6.4 mT to 652 mT at the aortic midarch, and the electrocardiogram was found to be accurate at magnetic fields below 70 mT. This corresponds to a distance of 160 cm from the isocenter and 80 cm from the bore entrance for the 1.5‐T MRI system used in this study. These results can be translated to any MRI system, with knowledge of the fringe field. Accurate electrocardiogram monitoring is feasible in close proximity to the MRI magnet, such as during and after pharmacologic or exercise stress, or interventional or surgical procedures performed in the MRI room. Magn Reson Med, 2010.