Joe D. Bourland

Review of Patient Safety in Time-Varying Gradient Fields

In magnetic resonance, time‐varying gradient magnetic fields (dB/dt) may stimulate nerves or muscles by inducing electric fields in patients. Models predicted mean peripheral nerve and cardiac stimulation thresholds. For gradient ramp durations of less than a few milliseconds, mean peripheral nerve stimulation is a safe indicator of high dB/dt. At sufficient amplitudes, peripheral nerve stimulation is perceptible (ie, tingling or tapping sensations). Magnetic fields from simultaneous gradient axes combine almost as a vector sum to produce stimulation. Patients may become uncomfortable at amplitudes 50%–100% above perception thresholds. In dogs, respiratory stimulation has been induced at about 300% of mean peripheral nerve thresholds. Cardiac stimulation has been induced in dogs by small gradient coils at thresholds near Reillys predictions. Cardiac stimulation required nearly 80 times the energy needed to produce nerve stimulation in dogs. Nerve and cardiac stimulation thresholds for dogs were unaffected by 1.5‐T magnetic fields. J. Magn. Reson. Imaging 2000;12:20–29.

Electrophysiological control of ventricular rate during atrial fibrillation.

Thirteen anesthetized canine subjects (17–29 kg) were used to demonstrate that mild cervical left vagal stimulation could control ventricular rate effectively during atrial fibrillation (AF). Two studies are presented. The first study used six subjects to demonstrate the inverse relationship between (manually applied) left vagal stimulation and ventricular excitation (R wave) rate during AF. As left vagal stimulation frequency was increased, ventricular excitation rate decreased. In these studies, a left vagal stimulus frequency of 0–10 per second reduced the ventricular excitation rate from > 200/min to < 50/min. The decreasing ventricular excitation rate with increasing left vagal stimulation frequency was universal, occurring in all 26 trials with the six subjects. This fundamental principle was used to construct an automatic controller for use in the second study, in which seven subjects were used to demonstrate that ventricular rate can be brought to and maintained within a targeted range with the use of an automatic (closed‐loop) controller. A 45‐minute record of automatic ventricular rate control is presented. Similar records were obtained in all seven subjects.

The effect of newer antiarrhythmic drugs on defibrillation threshold.

This study was conducted to determine the effects of clofilium phosphate and bretylium tosylate on ventricular defibrillation threshold. Dogs were anesthetized with pentobarbital and subjected to repeated fibrillation-defibrillation episodes. Defibrillation thresholds were determined at 15-min intervals, using underdamped 5–6 msec sinusoidal current shocks, from 30 min before drug injection to 120 min after injection. Eight dogs were given clofilium phosphate (0.34 mg/kg, iv). Another 10 dogs were given bretylium tosylate (10.0 mg/kg. iv). Both drugs lowered defibrillation threshold from 15–90 min after injection. The maximum clofilium effect was a 31% decrease in threshold current and a 54% decrease in threshold energy. The greatest decrease in defibrillation threshold produced by bretylium was 16% for current and 31% for energy. These drug induced changes in defibrillation threshold are of potential clinical benefit if they occur in human subjects at doses which are effective for control of ventricular arrhythmias.

The Strength-Duration Curve

The strength-duration curve is a plot of the lowest (threshold) current (I) required for stimulation versus pulse duration (d); it forms the basis for describing the excitability of a given tissue. It is extremely useful in all manner of studies in which excitable tissues are stimulated because it describes the manner in which the current required is changed when the pulse duration is changed. Moreover, it can be used to determine the charge and energy-duration relationships. It will be the objective of this communication to show how the essential constants can be obtained.

Bipolar catheter defibrillation in dogs using trapezoidal waveforms of various tilts

The choice of defibrillating waveform is critical in determining the size, battery life, and effectiveness of an automatic implantable defibrillator (AID). The trapezoidal (truncated exponential) waveform is well suited for the AID and its use can be optimized by the selection of appropriate values of pulse duration and tilt. The purpose of this study was to determine the dependence of the threshold peak current (the minimum peak current necessary to defibrillate the ventricles) on pulse duration and tilt for a bipolar catheter electrode configuration. Successive fibrillation-defibrillation trials were performed in 30 dogs anesthetized with sodium pentobarbital (30 mg/kg). The defibrillating pulse was applied via a bipolar-electrode catheter positioned such that the electrodes were located in the right ventricle at the apex and in the superior vena cava. The threshold peak current was determined in each dog for trapezoidal waveforms with 80%, 65%, 50%, and less than 5% tilt and with pulse durations of 2, 5, 10, 15, and 20 milliseconds. From a total of 600 threshold peak-current values, a strength-duration curve was derived for each value of tilt. The threshold peak current dose (peak current divided by body weight) increased with increasing tilt and decreasing duration. The threshold average current dose (average current over the duration of the defibrillating pulse divided by body weight) was IAV = 0.26 + 0.47/d, where d is the pulse duration in milliseconds and IAV is the average current in amperes per kilogram. If catheter apparent impedance is known, the minimum capacitance and output voltage necessary for defibrillation can be inferred from the strength-duration curves. From these data one can quantitatively assess the effect of trapezoidal waveform shape on the design criteria for the AID.

Myocardial Stimulation with Ultrashort Duration Current Pulses

In order to identify a practical short‐duration limit for stimulating myocardium, theoretical and experimental studies were carried out using dog and turtle hearts The strength‐duration curves for current, charge and energy were derived from the standard excitable membrane model which employs a parallel resistance and capacitance. From these derivations, the predicted duration for minimum energy was identified. The experimentally measured strength‐duration curves for two types of myocardium followed the predicted vaiues cioseiy The duration for minimum energy was calculated to be 1.25 times the membrane time constant. The practical short‐duration limit for a pacemaking stimulus is about 30% of the membrane time constant. For dog myocardium the average time constant was 2.4 ms Therefore, a practical stimulus duration for minimum charge in the dog should be no longer than about two‐tenths of a millisecond, although shorter duration stimuli are equally effective This minimum charge criterion provides the minimum drain on the stimulator power supply

Tissue stimulation: theoretical considerations and practical applications.

The three electrical characteristics of a stimulus (current, charge and energy) are related and can be predicted from the nature of the tissue stimulated. The older empirically derived Weiss-Lapicque concepts of stimulation are compared with those predicted by membrane theory. It is shown that, from the strength/duration curve for charge, it is possible to determine the membrane time constant of the tissue being stimulated. Two practical cases of cardiac muscle stimulation (pacing and defibrillation) were chosen to illustrate that although the goals are different in these two cases, the common denominator is excitation of cardiac muscle and that the membrane time constants so determined are similar to direct-heart and transchest electrode locations.

Relationship between pulse-wave velocity and arterial elasticity

Pulse wave velocity (PWV) was measuredin situ in 11 isolated canine common carotid arteries. Seven arteries exhibited a linear PWV/pressure function at pressures ranging from 0 to 200 mm Hg. Four arteries yielded a linear relationship between PWV and pressure between 1 and 100 mm Hg; for these vessels the relationship was nonlinear at higher pressures. Seven arteries (five from the group which was linear up to 200 mm Hg and two from the group which was linear up to 100 mm Hg) were excised and presure/volume measurements were madein vivo. Using pressure/volume data, the Moens-Korteweg equation was evaluated as a predictor of the PWV/pressure relationship over the linear region. An expression was developed to anable prediction of the pressure/volume relationship using the coefficients at the linear PWV/pressure function; these predictions were evaluated. We found that, for this range, the Moens-Korteweg equation provides a very good basis for predicting the increase in PWV with increasing bias pressure. In addition, we found that the pressure/volume relationship of common carotid arteries is well represented by an exponential of the form V/Vo=Keαf(P), which was derived as the inverse solution to the Moens-Korteweg equation.

Measurement of pulse-wave velocity using a beat-sampling technique

The relationship between pulse-wave velocity (PWV) and blood pressure was investigated using a new method in which multiple pulse-wave velocities were determined within each blood-pressure pulse. The technique employed measuring pressure at two sites within the aorta and measuring multiple time differences between each pair of pressure waves. Blood pressure was manipulated with drugs. The technique of obtaining multiple PWVs within a beat dramatically reduced the variability of the data in the linear and nonlinear region of PWV versus pressure relationship.

The relationship between arterial pulse-wave velocity and pulse frequency at different pressures

Pulse-wave velocity was measured in isolated canine common carotid arteries using sinusoidal frequency pulses of 1, 2, 5, 10, 15 and 20 Hz at 50, 100 and 150 mmHg. It was found that the pulse-wave velocity was independent of frequency and dependent on pressure. Using the Moens-Korteweg equation, the predicted pulse-wave velocity (y) was compared with measured pulse-wave velocity (x). A good correspondence was found (y = 1.063 x - 0.337, with a correlation coefficient of 0.963). The propagation velocity of the significant harmonic components of the pulsatile pressure waveform is the same for heart rates up to 120 beats/min.