Reproducibility of the signal-averaged electrocardiogram

Measurement and reproducibility of magnetocardiographic filtered atrial signal in patients with paroxysmal lone atrial fibrillation and in healthy subjects

Magnetocardiography (MCG) is a method complementary to electrocardiography (ECG). We examined recording and reproducibility of atrial depolarization signal by MCG. Multichannel MCG over anterior chest and orthogonal 3-lead ECG were recorded in 9 patients who had paroxysmal lone atrial fibrillation and in 10 healthy subjects in duplicate at least 1 week apart. Data were averaged using atrial wave template and high-pass filtered at 25, 40, and 60 Hz. Atrial signal duration with automatic detection of onset and offset and root mean square amplitudes of the last portion of atrial signal were determined. Coefficient of variation of atrial signal duration by MCG at 40 Hz was 3.3% and difference between the measurements was 3.5 milliseconds on average. The corresponding figures obtained by signal-averaged ECG (SAECG) were 6.1% and 6.9 milliseconds. Coefficient of variation for root mean square of the last 40 milliseconds of atrial signal were 16% in MCG and 17% in SAECG. Reproducibility was best at 40-Hz filter and similar in patients and healthy subjects. In conclusion, the reproducibility of atrial signal variables in MCG is adequate and somewhat better than in SAECG and equal in patients with lone atrial fibrillation and healthy subjects. Magnetocardiography seems to be a potentially valuable method to evaluate features of atrial depolarization in patient studies.

Therapeutic monitoring of class I antiarrhythmic agents using high-resolution electrocardiography instead of blood samples

Antiarrhythmic therapy requires monitoring of serum drug concentrations to determine a patient's optimal oral dose of medication. Repeated examination of blood samples, however, is costly and time-consuming, so the present study evaluated whether changes in serum concentrations could be estimated from changes in electrocardiographic (ECG) parameters. Of 36 patients receiving antiarrhythmic drugs for supraventricular or ventricular arrhythmias, 12 were treated with flecainide, 12 with pilsicainide, and 12 with pirmenol. Signal-averaged ECG (SAECG) were recorded before starting drug administration, 1 month later, and twice during ongoing therapy. At the time of the 2nd to the 4th recordings, serum concentrations of the drugs were also measured. As previously reported, all agents, but especially flecainide and pilsicainide, prolonged the filtered QRS (f-QRS) and the duration of low-amplitude signals at the terminal portion of the QRS complex. The SAECG parameters varied between the recordings made during therapy. Differences in the duration of the f-QRS between 2 recordings correlated significantly with differences in serum drug concentrations (r=0.91 for flecainide, r=0.70 for pilsicainide, and r=0.61 for pirmenol). No significant correlation between drug concentration and other SAECG parameters was found. Changes in the serum concentration of flecainide, pilsicainide and pirmenol can be estimated from changes in the duration of the f-QRS on the SAECG and periodic monitoring of such could help reduce the number of repeat measurements of drug concentrations in blood samples.

Uncertainty principle of signal-averaged electrocardiography

BACKGROUND

Signal-averaged ECG (SAECG) reproducibility is reported to have a component that is independent of residual noise. Methods and Results-In group 1, multiple paired SAECGs were obtained to noise levels of 0.3+/-0.1 and 0.5+/-0.2 microV. For the 0.5- and 0. 3-microV noise recordings, QRS duration (QRSd) was 101.2+/-11.3 and 104.6+/-9.6 ms, respectively (P<0.0001), and the differences in paired QRSd (DeltaQRSd) were normally distributed, with variances of 11.4 and 26.2 ms(2) (P<0.0001). Paired SAECGs were obtained in group 2 patients without and with late potentials; DeltaQRSd variance was 3.3 and 217.9 ms(2) (P<0.0001). In group 3, >/=10 SAECGs were acquired at noise levels of 0.2 to 0.8 microV, in 0.1-microV increments. QRSd increased as noise level decreased. The variance was greater in low-noise (0.2 to 0.4 microV) versus higher-noise (0. 5 to 0.8 microV) recordings. In group 4, SAECGs were analyzed with bidirectional and Bispec filters, with no difference in QRSd between the 2 filters and a normally distributed DeltaQRSd. A computer simulation demonstrated that alterations in the phase relationship of noise to signal results in a normal distribution of signal end points.

CONCLUSIONS

Within the acceptable noise range for SAECG, lower noise results in longer QRSd and larger variance, suggesting that more accurate recordings may have less reproducibility. The random timing of noise relative to signal results in the distribution/variance of repeated measurements. Statistical strategies may be used to reduce some of this variance and may enhance the diagnostic utility of SAECG.

Reproducibility of time-domain and three different frequency-domain techniques for the analysis of the signal-averaged electrocardiogram

Because time-domain (TD) analysis of the signal-averaged ECG (SAECG) has some limitations that limit its use, several frequency-domain analysis techniques were developed in an attempt to improve the diagnostic ability of the SAECG. However, it is not known how reliable these techniques are at detecting late potentials. This prospective study compares the short-term reproducibility of 4 analysis techniques: TD analysis, spectral temporal mapping (STM), spectral turbulence analysis (STA), and acceleration spectrum analysis (ASA), in a large series of normal patients and post-myocardial infarction (MI) patients. Two consecutive SAECGs were recorded in 634 patients that were divided into 3 groups: 117 remote MI patients undergoing programmed electrical stimulation for the inducibility of ventricular tachycardia (Group 1), 407 consecutive acute MI survivors (Group 2), and in 110 healthy volunteers (Group 3). The diagnostic reproducibility of the 4 techniques was evaluated by comparing rates of inconsistent results (1 normal and the other abnormal). The numeric reproducibility for each technique was assessed by comparing the normalized differences of each single SAECG parameter between the 2 recordings. Inconsistent results of diagnostic reproducibility were observed in 4.1%, 6.9%, 9.8%, and 18.0%, with TD, STA, ASA, and STM, respectively. Comparisons of these rates were significantly different (P < .05) except between STA and ASA (P = .07). The numeric reproducibility was highest for TD parameters, lowest for STM factors of normality, and intermediate for STA and ASA indices. TD analysis remains the most reproducible SAECG analysis technique, whereas STM showed the worst reproducibility, which limits its clinical applicability. STA and ASA provide an acceptable intermediate reproducibility, the former being slightly, although not significantly, more reproducible than the latter.

Short-, mid-, and long-term reproducibility of the atrial signal-averaged electrocardiogram in healthy subjects: comparison with the conventional ventricular signal-averaged electrocardiogram

Although atrial signal-averaged electrocardiogram (SAECG) has been proposed for noninvasive identification of patients with atrial tachyarrhythmias, the substantial variability of the measurement limits the clinical value. The aim of the study was to assess the short- to long-term reproducibility of atrial SAECG and to compare it to that of the conventional ventricular SAECG in 51 healthy volunteers (30 men; age 32 +/- 8 years). In each subject, SAECG recordings were obtained using MAC-VU electrocardiograph and HiRES and PHiRES software (Marquette Medical Systems) and repeated after 5 minutes, 1 day, 1 week, and 1 month. Automatically detected onset and offset of the filtered QRS complex and P wave were subsequently corrected by two independent observers, and the averaged values were used for the analysis. Conventional ventricular SAECG parameters: filtered QRS duration (QRStot), low amplitude signal duration, and root mean square voltage (RMS) of the terminal 40 ms of QRS, and 5 atrial parameters: filtered P wave duration (Ptot), RMS of the terminal 40, 30, 20 ms, and of the entire P wave were obtained. Relative errors of different pairs of measures were used to assess the intrasubject reproducibility. QRStot and Ptot were the most reproducible parameters. The relative errors after 5 minutes, 1 day, 1 week, and 1 month were 0.8% to 2.4% for QRStot, and 1.3% to 4.2% for Ptot. For RMS voltages, the relative errors exceeded 15% in short-term and 20% in long-term recordings. Although Ptot was statistically less reproducible than QRStot, the reproducibility of the former was good and comparable to that of the QRStot. The reproducibility of the voltage parameters was significantly poorer than that of the duration parameters. The study showed a satisfactory short- and long-term reproducibility of Ptot in the atrial SAECG in healthy subjects. However, low reproducibility of the voltage parameters should be considered in clinical applications.

Improved diagnostic value of combined time and frequency domain analysis of the signal-averaged electrocardiogram after myocardial infarction

BACKGROUND

Time domain analysis (TD) of the signal-averaged electrocardiogram (SAECG) presents a higher incidence of false positives in inferior myocardial infarction (MI), whereas spectral turbulence analysis (STA) suffers from a higher incidence of false positives in anterior MI. We investigated the hypothesis that a combined TD and STA (TD+STA) analysis of the SAECG could improve its predictive accuracy for major arrhythmic events (MAE) after MI.

METHODS

Signal-averaged electrocardiograms were prospectively recorded 10.1 +/- 2.6 days after acute MI in 602 patients. Time domain analysis and STA were performed using standard parameters and criteria for abnormality. For the combined TD+STA model, stepwise discriminant analysis was utilized to optimize prediction of MAE. Receiver operating characteristic curves were utilized to optimize cutoff values for each SAECG parameter separately, and also for the combined TD+STA model.

RESULTS

During a one-year follow-up period, 38 patients had MAE: 14 sustained ventricular tachycardia, 2 resuscitated ventricular fibrillation and 22 sudden cardiac deaths. The total predictive accuracy of combined TD+STA (89.9%) was significantly higher than TD (75.1%) or STA (77.6%). The negative predictive accuracy of all three analyses was high (98%). The positive predictive accuracy of TD (19.6%) or STA (18.3%) was quite low, and significantly improved to 35.8% by combined TD+STA analysis. The positive predictive accuracy of TD+STA improved to 51.2% in patients with left ventricular ejection fraction <40%.

CONCLUSIONS

Combined TD + STA analysis of the SAECG significantly improves its prognostic ability for MAE in post-MI patients compared with TD or STA analyzed separately.

Prognostic value of thrombolysis, coronary artery patency, signal-averaged electrocardiography, left ventricular ejection fraction, and Holter electrocardiographic monitoring for life-threatening ventricular arrhythmias after a first acute myocardial infarction

Prognostic studies after acute myocardial infarction (AMI) have mainly been performed in the prethrombolytic era. Despite the fact that modern management of AMI has reduced mortality rates, the occurrence of malignant ventricular arrhythmias in the late phase of AMI remains an important issue. We prospectively studied 244 consecutive patients (97 treated with thrombolytics) who survived a first AMI. All patients underwent time domain signal-averaged electrocardiography (vector magnitude: measurements of total QRS duration, terminal low [<40 microV] amplitude signal duration, and root-mean-square voltage of the last 40 ms of the QRS complex), Holter electrocardiographic monitoring, and cardiac catheterization. Late life-threatening ventricular arrhythmias were recorded. Eighteen arrhythmic events occurred during a mean follow-up period of 57 +/- 18 months. Three independent factors were associated with a higher risk of arrhythmic events: (1) left ventricular ejection fraction (odds ratio 1.9/0.10 decrease), (2) terminal low-amplitude signal duration (odds ratio 1.5/5 ms increase), and (3) absence of thrombolytic therapy (odds ratio 3.9). Low-amplitude signal duration sensitivity for sudden cardiac death was low (30%). Left ventricular ejection fraction had the highest positive predictive value for sudden cardiac death (10%). Thus, thrombolysis decreases both the incidence of ventricular tachycardia and sudden cardiac death with a higher reopening rate of the infarct-related vessel. Signal averaging predicts the occurrence of ventricular tachycardia and an impaired left ventricular ejection fraction predicts the occurrence of sudden cardiac death.

Autonomic effects on noise recorded during signal-averaged electrocardiography

The purpose of this study was to assess the effects of autonomic stimulation and blockade on noise levels and to compare the noise measurements in the ST and TP segments of the signal-averaged ECG. Five-minute electrocardiographic data were recorded in 14 normal volunteers (8 males and 6 females; mean age 28.5 +/- 5.0 years) on two separate days (day 1-baseline, epinephrine infusion, isoproterenol infusion, beta-blockade, and combined adrenergic and parasympathetic blockade; day 2-baseline, phenylephrine infusion, parasympathetic blockade, and during phenylephrine infusion following atropine). Signal averaging was done off-line on 100 beats and noise was measured in both the ST and TP segments as the standard deviation of voltage in the segment of interest. For all conditions tested, the mean noise level measured in the ST segment (0.46 +/- 0.16 microV) was significantly less than that measured in the TP segment (0.52 +/- 0.24 microV; P = 0.0003), but there was good correlation between the noise measured in the ST and the TP segment (R2 = 0.62, P < 0.0001). Noise increased with isoproterenol infusion and decreased following adrenergic blockade. In addition, day 2 baseline noise was less than baseline noise on day 1. Finally, neither parasympathetic stimulation or blockade nor alpha-adrenergic stimulation significantly affected signal-averaged electrocardiography (SAECG) noise levels. Thus, the data support the notion that enhanced sympathetic tone increases noise levels and beta-adrenergic blockade may decrease noise levels, likely due to effects from muscle sympathetic nerve activity. These findings are important since the target population for the SAECG are patients with myocardial infarction and congestive heart failure, conditions associated with increased sympathetic tone, which may in turn impact on the reproducibility or technical aspects of the SAECG. In addition, because noise in the ST and TP segments are highly correlated and the noise measured in the ST segment is less than that in the TP segment, uniform adoption of noise measurement in the ST segment seems most appropriate.

Immediate and short-term reproducibility of the P wave signal-averaged electrocardiogram

While abnormalities in the P wave SAECG have been associated with the occurrence of AF, its reproducibility has never been documented. The purpose of this study was to evaluate the immediate and short-term reproducibility of measurements from the P wave SAECG. P wave SAECGs were obtained using well-described techniques that utilize the QRS complex as the trigger and the P wave as template for averaging. In 28 subjects (8 controls, 11 with cardiac disease, 9 with prior AF), 3 P wave SAECGs were obtained: an initial study; on immediate reacquisition; and reacquisition after 4-5 days. Vector duration and RMS voltage of the terminal 20 ms of the P wave SAECG were measured and compared. The mean P wave duration was 152 +/- 14 ms on initial SAECG, 152 +/- 14 ms and 152 +/- 15 ms at immediate and short-term reacquisitions, respectively (both P = NS vs initial). The mean terminal RMS voltage was 6.4 +/- 6.0 mcV on initial SAECG, 6.4 +/- 5.9 mcV and 6.5 +/- 5.8 mcV at immediate and short-term reacquisitions, respectively (both P = NS vs initial). Linear regression analysis showed high reproducibility for both P wave duration (r = 0.94 for immediate and r = 0.96 for short-term reacquisition vs initial) but slightly less for terminal RMS voltage (r = 0.92 for immediate and r = 0.84 for short-term reacquisition vs initial). In subgroup analysis, P wave duration measurements were highly reproducible in controls, in subjects with cardiac disease, and in those with a history of AF. P wave duration was also reproducible for both males and females, as well as for subjects age > 65 years (r = 0.96 and 0.89 for immediate and short-term reacquisition, respectively). Terminal RMS voltage measurements were reproducible for controls, but less reproducible in other subgroups. In conclusion, P wave duration measurements on SAECG are reproducible when evaluated at immediate and short-term reacquisition regardless of age, sex, cardiac disease, or prior AF. Terminal RMS voltages were less reproducible, especially in patients with cardiac disease and/or prior AF. These findings may explain conflicting observations regarding the clinical utility of terminal P wave measurements.

Reproducibility of the signal averaged P wave: time and frequency domain analysis

OBJECTIVE

To assess the reproducibility of time and frequency domain variables derived from the signal averaged P wave.

DESIGN

Longitudinal within patient study.

SETTING

Regional cardiothoracic centre.

PATIENTS

20 patients (10 with documented paroxysmal atrial fibrillation and 10 normal controls) were studied on three occasions to assess the reproducibility of repeated signal averaged P wave recordings. Digital P wave recordings were made on a further 10 patients on a single occasion and the recordings signal averaged twice in order to assess the reproducibility of the averaging system itself in the absence of biological variation.

MAIN OUTCOME MEASURES

P wave duration, spatial velocity, and energies contained in frequency bands from 20, 30, and 60-150 Hz of the P wave spectrum were measured after P wave specific signal averaging. Coefficients of reproducibility were calculated for paired signal averaged P waves derived by signal averaging the same digital recordings on two separate occasions, for recordings performed in the same patients immediately after each other ("back to back") and those performed one week apart.

RESULTS

System reproducibility when the same digital P wave recordings were signal averaged on two separate occasions was high (< 11% for all variables). For P wave duration the coefficient of reproducibility was 11.4% for back to back recordings and 13.1% for those one week apart. The reproducibility of spatial velocity and P wave energy was low. Variation in P wave morphology was noted when successive P waves from the same subject were examined. If recordings with the same P wave morphology were analysed the reproducibility of spatial velocity and P wave energy improved but remained significantly poorer than that for P wave duration.

CONCLUSIONS

P wave duration is reproducible within subjects in the short and medium term. Frequency domain and spatial velocity analysis are poorly reproducible, due more to spontaneous variation in P wave morphology than to instability of the signal averaging process. This may limit the utility of signal averaged P wave variables other than duration for the prediction of atrial arrhythmia.

Evaluation of myocardial ischemia and infarction by signal-averaged electrocardiographic late potentials in children with Kawasaki disease

We investigated myocardial ischemia and old myocardial infarction noninvasively using signal-averaged electrocardiographic late potentials (LPs) in patients with Kawasaki disease. Patients were divided into 4 groups: a noncoronary artery lesion group (n=136), a coronary artery lesion group (without myocardial ischemia and an old myocardial infarction; n=33), an ischemia group (n=16), and an old myocardial infarction group (n=13). Grouping was based on exercise thallium-201 myocardial scintigraphy, thallium-201 myocardial scintigraphy, exercise electrocardiography, coronary angiography, left ventriculography, and echocardiography. Signal-averaged electrocardiograms were recorded using a high-resolution system. Values of filtered QRS duration (f-QRSd), root-mean-square voltage, and duration of low-amplitude signal were judged using our own body surface area-related criteria (n=205) to determine positive rates of LPs and sensitivities and specificities to ischemia and infarction. These data were also interpreted using published criteria for adults and compared with those interpreted by our criteria. Positive rates by our criteria were 0% in the noncoronary artery lesion group, 9.1% in the coronary lesion group, 56.3% in the ischemia group, and 69.2% in the old myocardial infarction group. However, using the criteria for adults, these values were 0%, 3.0%, 25%, and 46.2%, respectively. Sensitivities to ischemia and infarction using our criteria were significantly higher (56.3% and 69.2%) than those using the criteria for adults (p < 0.05). Moreover, specificities to ischemia and infarction were very high (93.4% and 93.5%, respectively) using our criteria, and there were no significant differences from specificities using the criteria for adults. Also, we examined the reproducibility of values of LPs and LP parameters. The values of filtered QRS duration showed a high reproducibility in both LP-positive and -negative groups, followed by low-amplitude signal and then root-mean-square voltage. The results of LP presence or absence showed 100% reproducibility for both the LP-positive and -negative groups, supporting the utility of LPs for clinical applications. Thus, LPs provide useful information in a noninvasive manner for clarifying ischemia and infarction in patients with Kawasaki disease.

Effect of residual noise level on reproducibility of the signal-averaged ECG

Late potentials are detected at various noise levels in clinical studies. The aim of this study was to assess the effect of residual noise level on the reproducibility of the signal-averaged electrocardiogram (ECG). Two consecutive raw 15-minute ECG from each of 188 patients with coronary artery disease were digitized and stored on optical discs. Each raw ECG was analyzed by two signal-averaging procedures to noise level 0.2 microV or 0.4 microV. Standard time-domain parameters were measured: QRS duration (SA-QRS), late potential (LP) duration, and root-mean-square voltage of the terminal 40 ms of the filtered QRS (RMS40). The SA-QRS was prolonged by 12 +/- 14 ms by the reduction in noise level from 0.4 microV to 0.2 microV, LP duration was prolonged by 10 +/- 10 ms, and RMS40 was reduced by 19 +/- 22 microV. The temporal variation of the measured SA-QRS from ECG1 to ECG2 was significantly lower at noise level 0.2 microV (9 +/- 13 ms) than at noise level 0.4 microV (13 +/- 14 ms) (P < .001). The LP duration was also more stable at noise level 0.2 microV than at noise level 0.4 microV (0.5 +/- 11 ms vs 2 +/- 13 ms, P < .05). The presence of any two of three abnormal parameters (SA-QRS > 120 ms, RMS40 < 25 microV, LP duration > 40 ms) was used as the criterion for the presence of LPs. At noise level 0.4 microV, the proportion of patients with diagnosed LPs in ECG1 was 25% and at noise level 0.2 microV it was 62%. At noise level 0.4 microV, 20% were reclassified from LP-negative in ECG1 to LP-positive in ECG2, and 7% were reclassified from LP-positive in ECG1 to LP-negative in ECG2. At noise level 0.2 microV, 20% were reclassified from LP-negative in ECG1 to LP-positive in ECG2, and 9% were reclassified from LP-positive in ECG1 to LP-negative in ECG2. It was concluded that (1) the diagnosis of LPs is significantly dependent on the extent of noise reduction by signal averaging: and (2) the numerical reproducibility of signal-averaged QRS duration and LP duration is lower at noise level 0.4 microV then at noise level 0.2 microV; and the diagnostic reproducibility of LPs is similar at both noise levels.

Immediate and day-to-day reproducibility of the signal-averaged electrocardiogram in patients with coronary artery disease

The purpose of this study was to prospectively evaluate the immediate (5 minutes) and short-term (1 day) reproducibility of the signal-averaged electrocardiogram (SAECG) in patients with coronary artery disease. A total of 59 consecutive patients with coronary artery disease (50 male, 9 female, mean age 59 +/- 13 years) were included. Analysis was performed using a commercially available system with high-pass filters of 25, 40, and 80 Hz without any change in electrode position. The following time-domain parameters were measured: (1) total filtered QRS duration; (2) duration of high-frequency low-amplitude signals (HFLA); and (3) root-mean-square voltage of the terminal 40 ms of the QRS complex (RMS40). Correlation between pairs of measurements was excellent, ranging from 0.91-0.99 for QRS and HFLA duration, and from 0.83-0.98 for RMS40 value. However, the range of differences was wide for all parameters, especially at 25 and 80 Hz, revealing marked individual variations in selected cases. The immediate reproducibility of an initially normal SAECG was 95% at 25 Hz, 100% at 40 Hz, and 81% at 80 Hz; figures for an initially abnormal SAECG were 89%, 91%, and 93%, respectively, and the coefficient of agreement kappa was highest at 40 Hz than at 25 or 80 Hz. Discordant results were most frequently observed in patients with borderline values, or in cases with nonsimilar residual noise levels. In conclusion, the immediate and day-to-day reproducibility of the SAECG is excellent in patients with chronic stable coronary artery disease. However, individual variations are observed in selected cases, leading to modification in interpretation of the result. For this reason, individual reproducibility should be considered before evaluating the effect of any therapeutic intervention on the SAECG, especially in patients with borderline values of the quantitative parameters of the SAECG.

Maximum likelihood analysis of cardiac late potentials

This study presents a new time-domain method for the detection of late potentials in individual leads. Basic statistical properties of the ECG samples are modeled in order to estimate the amplitude and duration of late potentials. The signal model accounts for correlation in both time and across the ensemble of beats. Late potentials are modeled as a colored process with unknown amplitude which is disturbed by white, Gaussian noise. Maximum likelihood estimation is applied to the model for estimating the amplitude of the late potentials. The resulting estimator consists of an eigenvector-based filter followed by a nonlinear operation. The performance of the maximum likelihood procedure was compared to that obtained by traditional time-domain analysis based on the vector magnitude. It was found that the new technique yielded a substantial improvement of the signal-to-noise ratio in the function used for endpoint determination. This improvement leads to a prolongation of the filtered QRS duration in cases with late potentials.

Short-term reproducibility of time domain, spectral temporal mapping, and spectral turbulence analysis of the signal-averaged electrocardiogram in normal subjects and patients with acute myocardial infarction

The aim of this prospective study was to compare the short-term reproducibility of the signal-averaged ECG (SAECG) with three analysis techniques, conventional time-domain analysis, spectral-temporal mapping (STM), and spectral-turbulence analysis (STA), in a large series of normal subjects and patients with acute myocardial infarction (AMI). Two consecutive SAECGs were recorded in 225 consecutive patients 10.2 +/- 2.7 days after AMI and in 85 healthy volunteers. The visual, diagnostic, and quantitative reproducibility of the three techniques was compared. Time-domain analysis was the most reproducible method, having high R2 correlations, statistically fewer inconsistent diagnostic recordings, and statistically smaller differences compared with other techniques. STM was the least reproducible, justifying caution in its current form. Although STA was significantly less reproducible than time-domain analysis, it was also significantly better than STM. Two STA parameters, spectral entropy and interslice correlation mean, showed good reproducibility, suggesting that modification of this analysis technique could be useful in risk stratification.

Is the time domain signal-averaged electrocardiogram helpful in patients with ventricular tachycardia without apparent structural heart disease?

The signal-average electrocardiogram (SAECG) has been a screening method for identifying patients at risk for ventricular tachycardia (VT) in the setting of coronary artery disease (CAD). Its significance in patients with VT unrelated to CAD or left ventricular dysfunction is undetermined. In order to define the value of SAECG in this patient population further, we compared the time domain SAECG at 25, 40, and 80 Hz filters in 35 patients with clinically symptomatic VT in the absence of structural heart disease was compared with 10 normal controls and 10 patients with CAD and inducible VT. SAECG data in patients without structural heart disease were intermediate between normal controls and patients with CAD. No single or combined SAECG criterion helped to differentiate between patients with inducible and noninducible VT. There was no concordance to other arrhythmia testing. It was concluded that signal-averaged electrocardiography may have little screening value in VT unrelated to CAD or left ventricular dysfunction.

Assessment of effects of autonomic stimulation and blockade on the signal-averaged electrocardiogram

BACKGROUND

Signal-averaged ECG is a noninvasive test designed to detect "late potentials." The effects of alterations in autonomic tone on the signal-averaged ECG have not been evaluated systematically.

METHODS AND RESULTS

The effects of autonomic stimulation and blockade on the signal-averaged ECG were evaluated in 14 healthy subjects (8 men and 6 women; age, 28.5 +/- 4.8 years) on 2 separate days. The signal-averaged ECG was recorded at baseline and after physiological and pharmacologic beta-adrenergic stimulation (tilt, exercise, and epinephrine and isoproterenol infusions), sequential and combined beta-adrenergic and parasympathetic blockade, and alpha-adrenergic stimulation before and after parasympathetic blockade. Analysis was performed with a bidirectional filter (40-Hz high-pass). Significant changes in the signal-averaged QRS duration from baseline (105.1 +/- 12.0 milliseconds) were noted with tilt (96.8 +/- 8.8 milliseconds), tilt after double blockade (97.5 +/- 9.0 milliseconds), epinephrine (110.5 +/- 11.8 milliseconds), and isoproterenol (99.6 +/- 12.6 milliseconds). Changes in the root-mean-square voltage of the terminal 40 milliseconds and the low-amplitude (< 40 microV) signal duration paralleled the changes in the QRS duration.

CONCLUSIONS

The signal-averaged ECG does not measure only "fixed" parameters but rather is altered under a variety of physiological and pharmacologic conditions. Upright tilt leads to shortening of the QRS duration before and after autonomic blockade; thus, the decrease in QRS duration with tilt may be related to factors other than changes in autonomic tone. These findings have implications for interpretation of the results of signal-averaged ECG.

The signal-averaged P wave duration: a rapid and noninvasive marker of risk of atrial fibrillation

OBJECTIVES

This study was undertaken to determine the ability of the signal-averaged electrocardiogram (ECG) to identify evidence of delayed atrial activation in patients with a history of atrial fibrillation.

BACKGROUND

Atrial fibrillation is a reentrant rhythm and depends on atrial conduction delay for its development. The signal-averaging technique is useful for accurately measuring total cardiac activation times, including delayed low amplitude signals, and thus can help identify the substrate for reentrant arrhythmias.

METHODS

Standard 12-lead and signal-averaged ECGs were recorded from 15 patients with a documented history of prior paroxysmal or chronic atrial fibrillation and 15 age- and disease-matched control subjects without a history of atrial fibrillation. Signal averaging was performed using an orthogonal lead system with the QRS complex as a trigger and the P wave as a template for the signal-averaging process. Total P wave duration was measured before and after filtering with a least squares fit filter. The P wave complexes on the three bipolar leads were combined into a vector combination of orthogonal leads. The total P wave duration of the individual unfiltered and filtered leads and the vector combination of filtered leads were calculated and used for analysis.

RESULTS

The P wave duration by standard ECG was not significantly different in patients with a history of atrial fibrillation and control subjects. Signal-averaged P wave durations were measured from orthogonal leads before and after digital filtering. Mean unfiltered P wave duration was significantly longer in patients with a history of atrial fibrillation than in control subjects (132 +/- 22 vs. 114 +/- 14 ms [p < 0.03] in the X lead, 135 +/- 21 vs. 115 +/- 15 ms [p < 0.03] in the Y lead and 133 +/- 23 vs. 114 +/- 14 ms [p < 0.03] in the Z lead). Mean filtered P wave duration was also longer in patients with atrial fibrillation than in control subjects (151 +/- 23 vs. 130 +/- 19 ms [p < 0.01] in the X lead, 157 +/- 22 vs. 136 +/- 17 ms [p < 0.01] in the Y lead and 154 +/- 23 vs. 135 +/- 15 ms [p < 0.01] in the Z lead). After filtering, a vector composite of orthogonal leads was determined. Again, P wave duration in patients with a history of atrial fibrillation exceeded that in the control subjects (162 +/- 15 vs. 140 +/- 12 ms [p < 0.01]). Using the vector composite of filtered orthogonal leads, a P wave duration > or = 155 ms was associated with a sensitivity of 80%, a specificity of 93% and a positive predictive value of 92% for identifying patients with history of atrial fibrillation.

CONCLUSIONS

A prolonged signal-averaged P wave duration may be a simple noninvasive marker of the risk for development of atrial fibrillation.

Effects of procainamide on the signal-averaged electrocardiogram in relation to the results of programmed ventricular stimulation in patients with sustained monomorphic ventricular tachycardia

OBJECTIVES

The aim of this study was to assess the ability of the signal-averaged electrocardiogram (ECG) to predict the efficacy of procainamide.

BACKGROUND

The main role of the signal-averaged ECG has been the identification of postinfarction patients at risk of sudden death. Prediction of the efficacy of antiarrhythmic drugs represents another potential clinical application of this technique.

METHODS

The study examined the effects of procainamide on the time domain and spectral temporal analysis of the signal-averaged ECG in relation to the results of programmed ventricular stimulation studies in 31 patients with inducible sustained monomorphic ventricular tachycardia.

RESULTS

Procainamide significantly prolonged the total and the initial QRS complex and low amplitude signal durations (mean +/- SD 135 +/- 30 vs. 161 +/- 46 ms, p < 0.0001; 87 +/- 16 vs. 98 +/- 20 ms, p < 0.0001, and 48 +/- 23 vs. 63 +/- 36 ms, p < 0.001, respectively) whereas the root-mean-square voltage of the total QRS complex and of the last 40 ms of the QRS complex was significantly reduced (mean +/- SD 112 +/- 36 vs. 87 +/- 36 microV, p < 0.0001; 21 +/- 19 vs. 13 +/- 12 microV, p < 0.002, respectively). The results of spectral temporal mapping of the signal-averaged ECG were similar before and after procainamide administration. Procainamide prevented the inducibility of sustained ventricular tachycardia or prolonged the cycle length of ventricular tachycardia by > or = 100 ms in 16 patients (52%) (responders). The fractional prolongation of the total QRS duration was significantly greater in responders (26 +/- 15%) than in nonresponders (10 +/- 10%) (p < 0.002) and, when this prolongation was > or = 15%, identified responders with a sensitivity of 94%, a specificity of 87% and an overall predictive accuracy of 90%.

CONCLUSIONS

The effects of procainamide on inducibility of ventricular tachycardia during programmed ventricular stimulation can be predicted by the degree of drug-induced prolongation of the signal-averaged QRS complex.

Frequency versus time domain analysis of the signal-averaged electrocardiogram: reproducibility of the spectral turbulence analysis

Reproducibility of the Spectral Turbulence Analysis. Spectral turbulence analysis (STA) of the signal-averaged electrocardiogram (ECG) is a new frequency domain method that analyzes the total high gain QRS complex and not only its terminal portion. This study examined the qualitative and quantitative short-term reproducibility of this technique (three recordings made within 25 min) in 68 subjects: 16 healthy volunteers; 22 patients with ventricular tachycardia and no evidence of heart disease; and 30 postinfarction patients with sustained ventricular tachycardia. The reproducibility of diagnosis of the STA was compared with that of the conventional time domain analysis of the signal-averaged ECG using standard criteria of abnormality. The reproducibility of numeric values of the spectral turbulence and of the time domain indices was performed by computing the ratios between standard deviation of measurements in individual subjects and standard deviations of all measurements. The reproducibility of diagnostic conclusions of the time domain analysis was slightly better than that of the STA but the differences were not significant (88%-91% of consistent time domain results vs 84% of consistent STA results). The numeric reproducibility of three STA parameters was slightly but not significantly inferior to that of the time domain indices whereas the reproducibility of the fourth STA variable, the intersegment correlation standard deviation (ISCSD), was significantly worse than that of the other indices. Of the two different ECG segments analyzed, the reproducibility of the STA variables calculated for the total QRS region was significantly better than that of the terminal low power QRS region. In conclusion, the qualitative and quantitative reproducibility of the STA is slightly but not significantly worse than that of the time domain analysis with the exception of the ISCSD, which is significantly less reproducible than all other parameters.