M. Leiniö

Identification of patients with ventricular tachycardia after myocardial infarction by high-resolution magnetocardiography and electrocardiography

The value of time domain analysis of late fields in the high-resolution magnetocardiogram in the identification of myocardial infarction patients with ventricular tachycardia was investigated in 30 subjects: 10 patients with documented sustained ventricular tachycardia and old myocardial infarction, 10 patients with old myocardial infarction without complex ventricular arrhythmias, and 10 normal volunteers. The duration of the QRS complex in the magnetocardiogram was significantly longer in ventricular tachycardia patients compared to myocardial infarction patients (144 (SD, 33) vs 109 (SD, 8) ms; p = 0.004). The root-mean-square field of the last 60 ms of the QRS complex was smaller in ventricular tachycardia patients than in myocardial infarction patients (830 (SD, 650) vs 1,480 (SD, 730) fT, respectively; p = 0.047). Also, the duration of the low-amplitude signal less than 700 fT was longer in ventricular tachycardia patients than in myocardial infarction patients (47 (SD, 28) vs 28 (SD, 8) ms, respectively; p = 0.048). The sensitivity and specificity in identifying ventricular tachycardia patients were both 80%, and the positive and negative predictive values were 78% and 86%, respectively. High-resolution electrocardiography recorded during the same session performed slightly better: sensitivity 90%, specificity 90%, and positive and negative predictive values 90%. The signal-to-noise ratio of electrocardiogram was higher (approximately 2 x) than that of magnetocardiogram. It is concluded that the new magnetocardiographic technique seems helpful in screening patients at risk of ventricular arrhythmias after myocardial infarction. The results encourage further refinement of the technique and application in prospective studies.

Localization of accessory pathways in Wolff-Parkinson-White syndrome by high-resolution magnetocardiographic mapping

Fifteen patients with Wolff-Parkinson-White syndrome were studied with standard 12-lead electrocardiogram, invasive electrophysiologic study, and high-resolution magnetocardiographic (MCG) mapping. In addition, intraoperative epicardial mapping was performed in seven surgically treated patients. The MCG characteristics of ventricular preexcitation for different locations of the atrioventricular accessory pathways were described in terms of morphology and field patterns. Three mathematical source models in semi-infinite conducting space were used for localization computations: the current dipole model, the truncated current multipole model and the magnetic dipole model. Finally, the localization results of MCG and invasive mappings and electrocardiograms were compared. The mean three-dimensional distance between the localization results obtained from MCG maps and electrophysiologic study was 3.9 cm for the magnetic dipole model, 4.8 cm for the truncated current multipole model, and 7.3 cm for the current dipole model. The corresponding distances in the seven intraoperatively mapped cases were 2.3 cm for the magnetic dipole model, 5.2 cm for the truncated current multipole model, and 6.3 cm for the current dipole model. In conclusion, noninvasive MCG mapping may significantly contribute to the invasive catheter mapping for optimal preoperative localization of preexcitation site and atrioventricular accessory pathways in Wolff-Parkinson-White syndrome.

Analysis of High Resolution MCG Recordings of Patients with Ventricular Tachycardia

Selection of patients prone to sustained ventricular tachycardia (VT), by detecting abnormal cardiac electric micropotentials called late potentials (LP), from high resolution (HR) ECG recordings has recently become to clinical practice. Observation of magnetic late field (LF) signals, corresponcling to electric LP:s, was first reported by Erne et al. in 1983.1 Since then only few studies have been published on late fields.

High-Resolution Magnetocardiographic Mapping of Pre-Excitation in Wolff-Parkinson-White Syndrome

We have studied 10 Wolff-Parkinson-White (WPW) syndrome patients using high-resolution magnetocardiographic (HR-MCG) mapping, standard 12-lead ECG and invasive electrophysiologic study (EPS). The ventricular pre-excitation caused by abnormal conduction via an accessory pathway (AP), seen as a so-called delta-wave prior to the QRS complex, was characterized by studying the morphology of the QRS and the delta-wave in the measured ECG and MCG signals. For all patients, EPS was performed to determine the conduction properties and the location of the AP. The site of the pre-excitation was also localized using the HR-MCG mapping and the results were compared to the invasive results. For a surgical operation or catheter ablation therapy, the site of the AP should be determined with a reasonable accuracy. Noninvasive methods, such as the HR-MCG, would be helpful in minimizing the duration and discomfort of invasive catheter mapping and in some cases could even replace it. Studies of WPW patients have been previously reported by Erne et al. (1985) and first simultaneous catheter and MCG mapping was done by Fenici et al. (1985).

High resolution electrocardiography and magnetocardiography: clinical application

A high-resolution (HR) surface ECG system is described, suitable for HR real-time (RT) recording and for signal averaging (SA). HR-ECG measurements were done in an electrically shielded room. The signal was amplified with a battery-driven 8-channel preamplifier, digitized with a 12(16)-bit A/D converter and stored in a microcomputer. The overall noise level of the system was below 1 µ Vp-p in the band 0.05–300 Hz. RT- and SA-magnetocardiographic (MCG) measurements were done in a magnetically shielded room using a very sensitive (5 fT/√VHz) DC-SQUID magnetic gradiometer. The output signal was transferred to a minicomputer for analysis. Examples of ECG and MCG recordings of cardiac micropotentials are presented, using both RT- and SA-techniques with special reference to late potentials (LP). The MCG technique was also utilized for localizing the Kent bundle in 6 cases of WPW syndrome. The recovery rate of LP was 37% in patients with recent myocardial infarction (AMI), and 72% in patients with ventricular tachycardia (VT) or ventricular fibrillation (VF). The mean duration of LP in the VT/VF-group was 27 ms, and the mean amplitude 17 µV. The His-Purkinje signal was detected in 60% of cases examined. The RT-ECG proved to be superior to the SA-ECG in detecting intermittent or inconstantly timed signals. Preliminary base-line data of 2 prospective studies are presented (10 patients of which with VT, 10 with recent AMI without VT and 10 healthy controls) using both RT-ECG and SA-ECG, the latter with both time and frequency domain analysis. The initial and terminal QRS notches of the low-gain ECG were most frequent in the VT group, and probably belong to the same category of depolarization abnormalities as LPs.

ECG and MCG Studies on the His-Purkinje Activity in a Case of Congenital Heart Block

Non-invasive detection of the His-Purkinje (HP) activity could be useful in clinical practice, e.g. in determining the actual site of atrio-ventricular (A-V) block or in evaluating the effect of anti-arrhythmic drug therapy on heart conduction system. The HP activity from body surface was first measured electrically by Berbari et al. (1978). First magnetic study of the HP conduction system was reported by Farrell et al. (1980). The HP activity was described as a ramp like structure in the P-Q interval. Fenici et al. (1985) reported magnetic His-Purkinje studies with invasive ECG reference. In normal subjects the overlapping atrial activity causes serious interference with the HP-signal. To avoid this overlapping, Makijarvi et al. (1985) investigated patients with total A-V block. However, in these studies the accurate timing was difficult because intracardial reference ECG was not available.

Sensitivity limits in biomagnetic measurements.

It is important to understand the character and the contribution of thermal magnetic noise in designing the measurement site and the instrumentation for biomagnetic measurements. The ultimate limit of the sensitivity is the thermal noise due to the object under study. In the case of the human body, it has been estimated to be about 0.1/square root of Hz. Magnetically shielded rooms are necessary for ultrasensitive biomagnetic measurements of human subject, but they also generate external noise which in some cases may become detectable. This noise problem can be avoided if the innermost walls are constructed of magnetically soft ferromagnetic material. Close to the conducting walls the thermal noise is higher than at the centre. Thus, the shielded room should be relatively large in size. The gantry and other things inside the room may contain metal parts, which can cause excess noise. The intensity depends on the conductivity, geometry, location and movement of these parts. In comparison to bioelectric studies, this inductive noise coupling demands extra attention. In most biomagnetic measurements performed inside a magnetically shielded room, the limiting factor of the sensitivity is the thermal noise caused by electrically conducting thermal shielding used inside the cryogenic measurement dewar. Fortunately, it is possible to reduce the noise contribution arising from the superinsulation in the dewar by careful design. The properties and dynamics of the SQUIDs are well understood nowadays. Studies of the nonlinear character in the coupling between the SQUID and external detection coil have made it possible to reduce the noise contribution of the sensor itself.(ABSTRACT TRUNCATED AT 250 WORDS)