High-frequency QRS Research Articles

Abstract 19310: High Frequency QRS Analysis has Incremental Diagnostic Accuracy Over ST-Segment Analysis Alone for Ischemia Identification in Patients Referred for Exercise Myocardial Perfusion Imaging

Introduction: Myocardial perfusion imaging (MPI) to identify ischemia is expensive, but existing electrocardiographic techniques have poor sensitivity. A novel method of high-frequency QRS analysis (HF-QRS) may improve ischemia detection. We sought to compare HF-QRS to standard ST-segment analysis for the identification of any and significant (≥10%) left ventricular (LV) ischemia by exercise SPECT MPI. Methods: We analyzed 292 consecutive patients. HF-QRS was determined within a bandwidth of 150 -250 Hz. A 50% relative reduction and ≥1 µV drop in the HF-QRS signal in ≥3 leads and ≥1mm ST-depression were the criteria for positivity for HF-QRS and ST-analysis, respectively. Quantitative, gated-SPECT MPI was performed according to standard protocols. Sensitivities and specificities and incremental diagnostic accuracy were assessed using chi-square and logistic regression analyses. Results: The mean cohort age was 59.6 years and 64.7 were male. Twenty nine percent had known coronary artery disease. LV ischemia was present in 54 patients (18.4%) and significant (≥10%) LV ischemia in 17 (5.8%). ST-analysis and HF-QRS analyses were positive in 42 (14.4%) and 70 (23.9%) respectively. As shown in the figure, ST-analysis had poor sensitivity for any ischemia that improved but remained low for ≥10% LV ischemia. HF-QRS had substantially higher sensitivities for both any and ≥10% LV ischemia. The 84.9% specificity of HF-QRS analysis for any ischemia was similar to the 86.6% specificity of ST-analysis. ST depression failed to add incremental diagnostic information to clinical risk factors (p=0.519), while HF-QRS increased the model Χ2 from 39.3 to 54.2 (p<0.001). Conclusions: High-frequency QRS analysis substantially improves the detection of any and significant ischemia over ST-segment analysis alone in patients undergoing exercise stress. This novel method has the potential to decrease the high rate of MPI for CAD risk stratification.

Cardiac Resynchronization Therapy Outcomes in Clinical Practice: Comparison With the COMPANION Study

Introduction: Large multicenter studies have shown that CRT improves symptoms, LV size and function, hospitalization rate and mortality in patients with advanced HF, decreased EF and wide QRS. The objective of this study was to determine if these beneficial effects are seen in clinical practice. Methods: We studied all patients who received a first CRT device at St. Paul Heart Clinic with EF ≤ 35%, QRS ≥ 120 ms and NYHA class III/IV HF during years 2003-2008 (Group 1). A subset of patients (Group 2) who met entrance criteria for the COMPANION CRT study, (hospitalization for HF in the year prior to CRT and not previously RV paced) were studied.

Detection of stress-induced myocardial ischemia from the depolarization phase of the cardiac cycle—a preliminary study

Background Electrocardiogram (ECG)-based detection of ischemia is typically dependent on identifying changes in repolarization. Analysis of high-frequency QRS (HFQRS) components, related to the depolarization phase of the cardiac action potential, has been reported to better identify ischemia. Our aim was to test the hypothesis that HFQRS analysis is both more sensitive and specific than standard ECG for detecting exercise-induced ischemia in patients undergoing exercise myocardial perfusion imaging (MPI). Methods Exercise MPI was performed in 133 consecutive patients (age, 63 ± 12; 100 males) and used as the gold standard for ischemia. Patients with QRS duration more than 120 milliseconds (n = 20), technical problems (n = 8), or inconclusive MPI (n = 4) were excluded, leaving 101 patients for analysis. Conventional ECG was combined with high-resolution ECG acquisition that was digitized and analyzed using the HyperQ System (BSP, Tel Aviv, Israel). The relative HFQRS intensity change during exercise was used as an index of ischemia. Results Of the 101 patients who were included in the analysis, 19 exhibited MPI ischemia. The HFQRS index of ischemia was found to be more sensitive (79% vs 41%; P < .05) and more specific (71% vs 57%; P < .05) than conventional ST analysis. Conclusions The HFQRS analysis was more sensitive and specific than conventional ECG interpretation in detecting exercise-induced ischemia and exhibited enhanced diagnostic performance in both women and men. Thus, it may aid in the noninvasive diagnosis of ischemic heart disease.

High-frequency QRS electrocardiogram analysis during exercise stress testing for detecting ischemia

Introduction ECG stress testing is an inexpensive and non-invasive detector of myocardial ischemia; addition of high-frequency QRS analysis (HFQRS) may improve accuracy. This study compared HFQRS during exercise in patients with and without ischemia as defined by multiple criteria. Material and methods High-resolution ECGs were recorded for 139 patients undergoing T99-sestamibi/T201-thallium stress testing. Twenty-three were positive by at least two and 37 were negative for ischemia by all three of the following criteria: nuclear scan, ST-segment analysis and typical angina. Sixty-four not meeting criteria for positive or negative, six with adenosine test and nine patients with ECG recording artifacts were excluded. Mean age of the study group was 62 ± 10 years, 83% were male. Ischemic patients had a higher incidence of previous myocardial infarction and coronary intervention than non-ischemic patients (74% vs. 46%; P = 0.03 and 70% vs. 43%; P = 0.05, respectively), but had a lower body mass index (28.7 ± 5 vs. 33.0 ± 8; P = 0.015). HFQRS analysis consisting of signal averaging (150–250 Hz) and calculation of root mean squared values for each lead at different time points was performed and was similar between the groups. The relative change in HFQRS (RCQ) was calculated for each lead: {(maxHFQRS − minHFQRS) / maxHFQRS}. For each patient an RCQ index was calculated by averaging the two leads with the greatest RCQ value. The RCQ index was greater in ischemic vs. non-ischemic patients (45% vs. 34%; P = 0.0069). Conclusion Maximum decrease in HFQRS, as quantified by RCQ index, was greater in ischemic vs. non-ischemic patients. Use of the RCQ index may improve the diagnosis of ischemia during exercise stress testing.

HyperQ—new horizons in ischemia detection

Fig. 1. A schematic flow of the production of the HyperQ signal. First, QRS complexes are detected, while rejecting arrhythmias and noisy complexes. The detected QRS complexes are carefully aligned and averaged. The averaged QRS complex is then filtered in the 150to 250-Hz band to Electrocardiogram (ECG)-based ischemia detection is typically based on identifying changes in the ST-T segment, yet with limited accuracy. Analysis of high-frequency QRS components, which quantifies changes in the depolarization phase of the cardiac cycle, has been previously reported to better identify ischemia. BSP Ltd (Tel Aviv, Israel) presented a new, Food and Drug Administration–approved technology to monitor and quantify changes in the highfrequency QRS components (HyperQ) during exercise, based on sophisticated ECG signal analysis (Fig. 1). Several studies aimed at assessing the clinical diagnostic value of the HyperQ System in detecting demand-induced ischemia have been performed. A large-scale clinical study aimed at evaluating the HyperQ during routine treadmill exercise test to detect induced myocardial ischemia. Exercise myocardial perfusion single-photon emission computed tomography (SPECT) was used as the gold standard for comparison. The study was performed in 885 consecutive patients. The relative intensity change of HyperQ during exercise was used as an index of ischemia (see Figs. 2 and 3). Logistic regression was used to assess the incremental diagnostic value of HyperQ data over conventional ECG analysis and clinical parameters. The HyperQ index of ischemia was found more sensitive than conventional ST analysis (78% vs 56%, P b .01) with similar specificity (74% vs 78%, P = NS). The HyperQ index offered a significant incremental diagnostic value over clinical symptoms and stress test data. Another important finding was that changes in the HyperQ signal were observable before ST changes were evident, indicating an early detection of ischemia by the HyperQ. In a research project performed at the Rabin Medical Center, Israel, HyperQ and conventional ECG were obtained in 95 patients undergoing exercise myocardial perfusion SPECT. The HyperQ index of ischemia exhibited enhanced sensitivity and specificity relative to conventional ST analysis (76% vs 59% [P b .01] and 85% vs 57% [P b .01], respectively). Data from a study that was carried out at the Charleston Area Medical Center, West Virginia, were reanalyzed and

CardioStudy: a host/client software suit for computerized electrocardiography on terabyte databases

Medical researchers who need to test novel computational algorithms on recorded electrocardiogram (ECG) data are faced by several problems. Although data have been stored in the form of computer files for a couple of decades now and many groups have large collections of old ECG data stored in their original formats, it is by no means easy to access the raw data in these files. Commercialized data analysis software packages that focus on day-to-day clinical needs are not flexible enough to be used for testing new computational protocols. The Drew laboratory has more than a decade’s worth (many thousands) of 24-hour ECG recordings in several different formats adding up to several terabytes, together with medical records for each patient. We wanted to make it practical to test newly invented computation algorithms and do new clinical studies on this unique database. New Physics Devices is developing CardioStudy, a suit of software for the storage, manipulation, and analysis of data in extensive (in terabytes and upsized) physiologic databases consisting of a very large number of individual patient records. The CardioStudy suit consists of a hosting component and client components. The host component provides tools for the creation, update, and management of very large physiologic databases, together with tools for performing data analysis and communicating with client processes. The client components provide tools for the composition and performance of clinical studies that involve computerized analysis on whole populations of the host database. In this presentation, we outline our progress to date by describing the host database and features of the client architecture. We illustrate the operation of the whole by performing several novel computerized studies on highfrequency QRS algorithms as applied to the database.

Temporal and postural variation of 12-lead high-frequency QRS electrocardiographic signals in asymptomatic individuals

Because changes in the 12-lead high-frequency QRS electrocardiogram (HF QRS ECG) more sensitively identify myocardial ischemia than do changes in the ST segments of the conventional ECG, it is important that changes in HF QRS signals that are merely physiological be distinguishable from those that are potentially pathological. We therefore studied the temporal variation of HF QRS measures such as root mean square (RMS) voltage and the presence vs absence of reduced amplitude zones (RAZs) in 107 asymptomatic individuals in the supine position during a brief period of ECG monitoring. In addition, to ascertain the effects of posture on the 12-lead HF QRS ECG, we collected additional seated data from 25 of these individuals and estimated the fifth and 95th percentile of the percent relative change between the supine and seated measurements. In all cases, variation of HF QRS parameters decreased as the number of beats in the signal average increased. For example, in the supine position, the 95th percentile of the percent relative change between consecutive within-lead measurements of RMS voltage for a 50-beat signal average was 12.3% but decreased to 11.7%, 11.2%, and 10.7% for 75, 100, and 150 beat signal averages, respectively ( P < .01). After transition from the supine to the seated upright position, changes in some measures of HF QRS were statistically significant, with RMS voltage decreasing significantly in lead V 3 and with the number of RAZs lost in the 12-lead HF QRS ECG significantly exceeding the number of RAZs gained. We conclude that most measures of HF QRS ECG are sufficiently stable for routine continuous monitoring.

High-frequency QRS electrocardiogram predicts perfusion defects during myocardial perfusion imaging

Background Changes in high-frequency (HF) QRS components of the electrocardiogram (ECG) (150-250 Hz) are more sensitive than changes in conventional ST segments for detecting myocardial ischemia. We investigated the accuracy of 12-lead HF QRS ECG in detecting perfusion defects during adenosine tetrofosmin myocardial perfusion imaging (MPI). Methods and Results 12-lead HF QRS ECG recordings were obtained from 45 patients before and during adenosine technetium Tc 99m tetrofosmin MPI tests. Before the adenosine infusions, recordings of HF QRS were analyzed according to a morphologic score that incorporated the number, type, and location of reduced amplitude zones (RAZs) present in the 12 leads. During the adenosine infusions, recordings of HF QRS were analyzed according to the maximum percentage changes (in both the positive and negative directions) that occurred in root mean square voltage amplitudes within the 12 leads. The best set of prospective HF QRS criteria had a sensitivity of 94% and a specificity of 83% for correctly identifying the MPI result. The sensitivity of simultaneous ST-segment changes (18%) was significantly lower than that of any individual HF QRS criterion ( P < .001). Conclusions Analysis of 12-lead HF QRS ECG is highly sensitive and reasonably specific for detecting perfusion defects during adenosine MPI stress tests and significantly more sensitive than analysis of conventional ST segments.

High-frequency QRS electrocardiogram predicts ischemic defects during myocardial perfusion imaging

High-frequency QRS electrocardiogram predicts ischemic defects during myocardial perfusion imaging

Twelve-Lead High-Frequency QRS Electrocardiography During Anesthesia in Healthy Subjects

Analysis of the high-frequency (HF) components of the QRS complex has been shown to be a more sensitive indicator of myocardial ischemia and infarction than conventional ST segment analysis in settings outside of the operating room. In this study, we documented the effect of general anesthesia on HF QRS analysis in healthy patients as the first step in determining the potential of this technique for monitoring anesthetized patients. HF QRS electrocardiograms (ECGs) were obtained from all 12 ECG leads in 30 healthy subjects before and after the induction of anesthesia. When compared with preinduction values, there were significant postinduction changes in multiple variables of the HF QRS in many leads studied that were within previously described normal limits. Additional study is needed to understand the potential of this monitoring technique for enhancing detection of myocardial ischemia in the anesthetized population.

Month-to-month and year-to-year reproducibility of high frequency QRS ECG signals

High frequency electrocardiography analyzing the entire QRS complex in the frequency range of 150 to 250 Hz may prove useful in the detection of coronary artery disease, yet the long-term stability of these waveforms has not been fully characterized. Therefore, we prospectively investigated the reproducibility of the root mean squared voltage, kurtosis, and the presence versus absence of reduced amplitude zones in signal averaged 12-lead high frequency QRS recordings acquired in the supine position one month apart in 16 subjects and one year apart in 27 subjects. Reproducibility of root mean squared voltage and kurtosis was excellent over these time intervals in the limb leads, and acceptable in the precordial leads using both the V-lead and CR-lead derivations. The relative error of root mean squared voltage was 12% month-to-month and 16% year-to-year in the serial recordings when averaged over all 12 leads. Reduced amplitude zones were also reproducible up to a rate of 87% and 81%, respectively, for the month-to-month and year-to-year recordings. We conclude that 12-lead high frequency QRS electrocardiograms are sufficiently reproducible for clinical use.

Temporal and postural variation of 12-lead high frequency QRS ECG signal

Temporal and postural variation of 12-lead high frequency QRS ECG signal

Cardiac resynchronization therapy is effective in patients with atrial fibrillation: interim results from the RESTORE-U.S. study

Introduction: Cardiac resynchronization therapy (CRT) has been shown to improve symptoms, exercise tolerance, and ventricular function in patients with advanced HF (NYHA III, IV), LV dysfunction and prolonged QRS duration in sinus rhythm (SR). There are little data demonstrating the effectiveness of CRT in atrial fibrillation (AF) patients. Small studies, including MUSTIC AF, studied less than 100 AF patients. Also, most controlled clinical trials excluded chronic AF patients in order to avoid confounding results. The Registry of Cardiac Resynchronization Therapy-U.S. (RESTORE-U.S.) provides an opportunity to assess the largest subgroup of AF patients receiving CRT to date. Methods: RESTORE-U.S. is a prospective, observational study to assess the responses of patients to CRT in a standard care setting with follow-up of 12 months. Enrollment criteria are the same as the current clinical indications for CRT. The primary endpoints are changes in NYHA class, QOL as measured by the Minnesota Living with Heart Failure Questionnaire, physician and patient global assessments, and echocardiographic indices of cardiac function and dimension. Approximately 2000 patients will be enrolled at 200 sites. All patients are implanted with a Medtronic CRT or CRT+D system. Patients are evaluated at 1, 3, 6, and 12 months postimplantation. Echocardiographic evaluations are only performed at 12 months. Results: To date, 185 AF patients (paroxysmal, chronic or ablated) have received CRT. Baseline characteristics of AF patients are similar to the study population, although AF patients were more commonly in class IV (24% vs. 20%). At 1 month, 54% (N=155) of the AF patients improved from a NYHA class III/IV to a class I/II, and 68% (N=89) at 3 months. AF patients also experienced a 40% (N=158) reduction in QOL score at 1 month, and 44% (N=92) reduction at 3 months, indicating improvements in QOL. These data are consistent with the results of the physician and patient global assessments. The results seen in the AF subgroup are similar to the improvements seen in patients in SR. To date, the rate of adverse events was similar in the AF subgroup compared with the study population as a whole. Conclusions: CRT resulted in a significant improvement in functional status and QOL in HF patients with ventricular dyssynchrony and AF.

High-frequency QRS potentials as a marker of myocardial dysfunction after cardiac surgery

Background High-frequency QRS potentials are sensitive to myocardial ischemia. The aim of this study was to evaluate the usefulness of high-frequency QRS potentials as a marker of myocardial dysfunction after cardiac surgery. Methods Seventy patients undergoing coronary artery bypass grafting or heart valve surgery were involved. High-frequency QRS potentials were measured by signal-averaged electrocardiogram, and calculated as the root-mean-square voltage of the total QRS duration (RMST). The postoperative RMST was expressed as a percentage of the preoperative RMST. The mean RMST at 1 to 2 hours after removing the aortic cross-clamp was compared with the cardiac index, inotropic agents, and aortic cross-clamping time. The occurrence of ventricular tachycardia within 24 hours and the RMST at 2 postoperative days were also evaluated. Patients were divided into quartile groups from highest to lowest at postoperative RMST (groups 1, 2, 3, and 4, respectively, from maximum to minimum). Results In postoperative states, cardiac index significantly decreased in accordance with the RMST decrease in a stepwise manner, although there were no differences in cardiac index among the four groups preoperatively. Inotropic agents and aortic cross-clamping time increased as RMST decreased. A high rate of ventricular tachycardia within 24 hours and delayed RMST recovery at 2 postoperative days were seen in group 4. The curve of sensitivity and specificity showed that severe reduction (threshold, 35%) of RMST indicated low-output syndrome. Conclusions The severe reduction of filtered high-frequency QRS potentials was related to myocardial dysfunction. Measurement of filtered high-frequency QRS potentials could become a useful, noninvasive, real-time monitor of myocardial dysfunction after surgery.

1126-116 The utility of high frequency QRS electrocardiogram in the diagnosis of cardiomyopathy

1126-116 The utility of high frequency QRS electrocardiogram in the diagnosis of cardiomyopathy

Newly developed high frequency QRS electrocardiograph in the detection of coronary disease

Newly developed high frequency QRS electrocardiograph in the detection of coronary disease

Changes in 12-Lead High Frequency QRS Electrocardiography Following Induction of Anesthesia in Healthy Individuals

Changes in 12-Lead High Frequency QRS Electrocardiography Following Induction of Anesthesia in Healthy Individuals

Changes in High Frequency QRS Energy in the Signal‐Averaged Electrocardiogram After Acute Experimental Infarction

Background: The effect of Ml on the high frequency content of the QRS complex in the SAECG remains a matter of controversy. The purpose of this study was (1) to characterize the changes in the high frequency QRS energy that occur after experimental Ml in dogs, (2) to evaluate the change in QRS energy in a number of frequency bandwidths, and (3) to evaluate the relationship between the change in QRS energy and infarct size and location, as well as the inducibility of VT.Methods: MIs were created in 21 dogs by balloon occlusion of the left circumflex (n = 10) or left anterior descending (n = 11) coronary arteries. SAECGs, using three orthogonal leads (X, Y, Z), were obtained before and 4–6 days after the Ml. The vector magnitude was calculated. Using a spectral filter, we calculated the energy in û Vs) of the entire QRS as the integrated area of the signal voltage within a low frequency bandwidth (15–40Hz) and two high frequency bandwidths (40–80 Hz, 80–300 Hz). QRS energy was log transformed to achieve a normal distribution.Results: There was a significant decrease in QRS energy after Ml in the low and high frequency bandwidths in the 21 dogs that was primarily detected in the Y lead. This decrease in high frequency energy after Ml occurred primarily in the dogs with anterior infarction where there was a 13–21% drop in QRS energy in all three bandwidths in the vector magnitude as well as the Y and Z leads (all P &gt; 0.05). This drop in QRS energy after anterior wall Ml was moderately correlated with increasing infarct size in all three bandwidths. There were no consistent changes in QRS energy in the dogs with posterior infarction. The drop in QRS energy after Ml was not associated with inducible VT. The change in QRS energy after Ml in all three bandwidths were highly correlated.Conclusions: The process of Ml produced a decrease in QRS energy, in both low and high frequency bandwidths, that was most consistent in dogs with anterior wall Ml. The decrease in QRS energy after Ml was not associated with inducible VT. Ml produced similar changes in QRS energy in low and high frequency bandwidths. A.N.E. 1999;4(1):72–82

Signal-averaged electrocardiograms in normal Doberman pinschers.

Signal-averaged electrocardiograms (SAECGs) were performed on nonsedated normal dogs in left-lateral recumbency. Following signal averaging, both time-domain and 3-dimensional frequency-domain analyses were performed. For time-domain analysis, the high-frequency QRS (HFQRS) duration, duration of the terminal QRS complex less than 40 microV (LAS40), and root mean square (RMS) voltages (microV) of the terminal 40 milliseconds (RMS40) and 30 milliseconds of the QRS complex were calculated. For frequency-domain analysis, correlation ratios were calculated for 30-, 40-, 50-, and 60-millisecond segment lengths begun 10, 15, or 20 milliseconds before the end of the QRS complex. Spectro-temporal mapping was also performed. All of the parameters of the SAECGs analyzed in the time domain were associated with each other. LAS40 and RMS voltages regressed significantly (P < .0000) on the HFQRS duration. Ninety-five percent of the HFQRSs were 55-75 milliseconds, 95% of the LAS40s were 9-26 milliseconds, and 95% of the RMS40 voltages were 177-444 microV. None of the SAECGs contained evidence of ventricular late potentials. Spectro-temporal maps were similar in each dog when the same segment lengths and starting points were compared. No evidence of ventricular late potentials was observed. Correlation ratios were lower when windowed segments included 15 or 20 milliseconds (versus 10 milliseconds) of the terminal QRS complex. When only 10 milliseconds of the terminal QRS complex were included in windowed segments, the mean correlation ratios for 30- and 40-millisecond segment lengths were > 0.8 and > 0.61 in 67% of all analyses, respectively.

Slowed Conduction with Antiarrhythmic Drugs Reduces High Frequency Energy of the Signal‐Averaged Electrocardiogram

Background: We have previously demonstrated that high frequency QRS energy decreases after experimental myocardial infarction (Ml) and is lower in patients who have had MI compared to controls. The mechanism for this decrease in high frequency QRS energy is unclear. The objective of this study was to evaluate the effect of slowed myocardial conduction on high frequency QRS energy measured on the signal‐averaged SAECG. We hypothesized that slowed conduction in the infarcted ventricle may be responsible for the decrease in high frequency QRS energy. In order to test this hypothesis, we examined antiarrhythmic drug therapy known to slow conduction (propafenone) compared to drug therapy with minimal effects on conduction (sotalol).Methods: In patients with sustained ventricular tachyarrhythmias undergoing serial drug testing, SAECGs were obtained before and after antiarrhythmic therapy. After filtering the leads with a spectral band‐pass (15–40 Hz, 40–80 Hz, and 80–300 Hz) filter, the vector magnitude was con structed, and the energy (in μV sec) was calculated for the entire QRS by integrating the area under the filtered QRS complex.Results: Propafenone significantly prolonged QRS duration (+17%, P &gt; 0.001) and significantly reduced QRS energy (‐16.1% to 21.8%, P &gt; 0.0001) in all three band widths. Sotalol did not have either effect (P = ns). There was a strong correlation between the prolongation of the filtered QRS and the drop in QRS energy for all three band widths (r values ranging from 0.64–0.90, all P &gt; 0.05). When the changes in QRS energy were corrected for QRS duration, the results did not change.Conclusion: The Class IC antiarrhythmic drug propafenone, known to decrease myocardial conduction velocity, significantly reduced QRS energy in all three band widths, whereas the Class III drug sotalol did not. These data are consistent with the hypothesis that decreases in low and high frequency QRS energy after Ml result in part from slowed conduction.