Effects of pulsed magnetic fields on neurite outgrowth from chick embryo dorsal root ganglia.

We have previously shown that neurite outgrowth from 6-day chick embryo dorsal root ganglia (DRG) in vitro was stimulated when nerve growth factor (NGF) and pulsed magnetic fields (PMF) are used in combination. 392 DRGs were studied in a field excited by a commercial PMF generator. We have now analyzed an additional 416 DRGs exposed to very similar PMF's produced by an arbitrary wavefrom generator and power amplifier. We reproduced our previous findings that combination of NGF and bursts of asymmetric, 220 microsecond-wide, 4.0 mT-peak pulses induced significantly (p < 0.05) greater outgrowth than NGF alone, that fields without NGF do not significantly alter outgrowth, and that, unlike NGF alone, 4.0 mT fields and NGF can induce asymmetric outgrowth. The asymmetry does not seem to have a preferred orientation with respect to the induced electric field. Analysis of the data for the entire 808 DRGs confirms these findings. Importantly, we find similar results for pulse bursts repeated at 15 or 25 Hz.

Electrical field distribution within the injured cat spinal cord: injury potentials and field distribution.

This study investigated the spontaneous injury potentials measured after contusion or transection injury to the cat spinal cord. In addition, the distribution of electrical field potentials on the surface and within the spinal cord were measured following applied electrical fields after transection and contusion injuries. After transection of the spinal cord, the injury potentials were -19.8 +/- 2.6 mV; after contusion of the spinal cord, the injury potentials were -9.5 +/- 2.2 mV. These potentials returned to control values within 2.5-4h after injury. The electrical field distribution measured on the dorsal surface, as well as within the spinal cord, after the application of a 10 microA current, showed little difference between contusion and transection injuries. Scalar potential fields were measured using two configurations of stimulating electrodes: dorsal to dorsal (D-D), in which both electrodes were placed epidurally on the dorsal surface of the spinal cord, and ventral to dorsal (V-D), in which one electrode was placed dorsally and one ventrally. As reported in normal uninjured cats, the total current in the midsagittal plane for the D-D configuration was largely confined to the dorsal portion of the spinal cord; with the V-D configuration, the current distribution was uniform throughout the spinal cord. In the injured spinal cord, the equipotential lines midway between the stimulating electrodes have a wider separation than in the uninjured spinal cord. Because the magnitude of the electrical field E is equal to the current density J multiplied by the resistivity r, this suggests that either the current density is reduced or that the resistivity is reduced.

Electric field distribution within normal cat spinal cord.

Electric currents of small magnitude have been used successfully to induce regrowth of injured spinal cord fibers. The purpose of this study was to determine the potentials and current density distributions on the surface, as well as within the spinal cord, after the application of exogenous electric fields. A 10 microA DC current was applied epidurally to the spinal cord using two different electrode configurations. The two electrode configurations studied were: anode and cathode dorsal (D-D) and anode ventral and cathode dorsal (V-D). Two types of recording electrodes were used to map the potentials on the surface and within the spinal cord. The recording system consisted of glass microelectrodes connected to differential amplifiers. The output was recorded on a polygraph. The current density was more localized on the dorsal surface of the spinal cord for the D-D configuration. In contrast, in the V-D configuration, the current density was greater near the anode on the ventral surface and near the cathode on the dorsal surface of the spinal cord. As a result of the anode being located ventrally, there was a more uniform current density distribution within the spinal cord.

A probe for measuring current density during magnetic stimulation.

Time-varying magnetic fields induce currents in conductive media, and when the induced current is large enough in excitable tissue, stimulation occurs. This phenomenon has been applied to the human brain and peripheral nerves for diagnostic evaluation of the neural system. One important aspect that is presently unknown is the current level necessary in tissue for stimulation induced by magnetic fields. This study presents a method of measuring the induced current density from pulsed magnetic fields in vitro and in vivo. The current-density probe was inserted into three concentrations of saline and into the brains of ten anesthetized cats. Two stimulation systems with coils 9 cm and 5 cm in diameter were used. The two systems provided sinusoidal and pulsatile coil currents. Measurements made in saline were compared with those calculated theoretically for a semi-infinite medium. The measured values were within 5% of the calculated values. Measurements made in the cat brain showed a 67% decrease compared with the theoretic model. This variance is attributed to the finite bounds of the skull. The results indicate that direct measurement of current density is possible. Subsequent measurements will aid in the design of improved magnetic stimulation systems.

Determination of tissue viability in experimental electrical injuries.

Electrical burns or ischemia (induced by vascular ligation) were produced in the legs of 15 anesthetized dogs to study evolution of tissue changes compared with impedance alterations. After the application of 1-ampere currents at 60 Hz, animals were monitored from 1 to 4 days. Muscle impendance was measured with frequency sweeping to determine tissue destruction. Nuclear magnetic resonance spectroscopy (phosphorus 31) was used to assess metabolic activity, and results were compared to impedance measurements. In burned limbs, 70% reduction in muscle impedance was seen, which corresponds to decreased metabolic activity (absent organic phosphates) and suggests necrosis. Visually viable tissue had impedance decreases of 25% and levels of organic phosphates slightly lower than normal. Relaxation frequencies in dogs with severe burns exceeded 80 kHz; in viable tissue, 30 to 40 kHz (normal: 30 kHz). In ischemic muscle, organic phosphates decreased rapidly (1 to 2 hours); impedance changes evolved more slowly (1 day), but they ultimately reached the same degree of severity. Measurement of impedance may be a valuable adjunct in the evaluation of electrical burns, since significant changes strongly suggest nonviability.

Fibrillation induced at powerline current levels.

Electrical fibrillation of the human heart results in many unfortunate deaths. Because little information is available on short duration high current fibrillation, current levels below 1 and 50 A were used to induce ventricular fibrillation in hogs. Application times ranged between 16 ms and 3 s. Fibrillation was only produced when currents were applied during the T-wave period of the cardiac cycle. However, only 50 percent of the current application during the T-wave caused fibrillation. The total body resistance of the hogs was also measured at the high voltages and currents. The average resistance for 90 current applications was 284 omega. Trends in the data show that the total resistance decreases for increasing voltage, for increasing electrode size, and for current applications following the first current application.

Measure of tissue resistivity in experimental electrical burns.

Studies were conducted in 14 mongrel dogs to compare resistivities in normal muscle with those from muscle subjected to electrical burns. One-ampere, 60-Hz currents were passed between the hind limbs of the dogs producing injury in three measurement regions of the gracilis muscle. Histology, heart rate, body temperature, arterial and pulmonary artery pressure, cardiac output, hematocrit, leukocyte counts, fibrinogen levels, and platelet levels were determined. Muscle resistivity associated with severe tissue necrosis was 70% lower than control values. Resistivity in tissue showing edema and minimal necrosis decreased 20 to 40% from control values. Muscle showing only edema had a 10 to 30% decrease in resistivity.

Experimental electrical injury studies.

Voltages from 10 to 14,000 volts demonstrated currents up to 70 amperes with resistances of approximately 200 ohms in studies in hogs. Below 1,000 volts, a current reduction is observed following arcing and skin necrosis. At the higher voltages, this phenomenon was not observed. The energy required for tissue damage was dependent upon the voltage and time of application. The tissue electrode resistance with stainless steel disc was proportional to the diameter. Skin buring commenced at the periphery of the electrodes and moved inwards. For application of currents between the hindlimbs of the hog, the current per tissue cross-section was greatest in artery and nerve, followed by muscle, fat, bone marrow, and bone cortex.

Effect of electrical current on temperature and pH in cerebellum and spinal cord.

Electrical currents similar to those used for relief of pain and spasticity were applied to the spinal cord and cerebellum of monkeys and cats. Direct coupled monopolar pulses increased pH at the tissue surface near a negative electrode and correspondingly decreased the pH near a positive electrode. Cerebrospinal fluid (CSF) circulation around the electrodes neutralized these changes. Capacitively coupled currents up to a 10 mA peak and 2.5 mucoul per peak failed to change tissue pH measurably. Current levels with power densities of 100 mW per cm2 did not cause any change in tissue temperature.

Evaluation of electrode configurations in cerebellar implants.

Capacitively coupled currents of 100 Hz, 0.25 msec duration, were applied to multielectrode arrays implanted upon the superior and posterior surfaces of the chimpanzee cerebellum. The current required for 90% reduction in the amplitude of the evoked potential was inversely proportional to the number of electrodes upon the cerebellar surface. A study of various waveforms showed that z Hz, 0.25 msec pulse duration is near optimal for reduction of amplitude of the somatosensory evoked potential. The current densities per electrode were 5--11 mA/cm2 with a charge per pulse of 0.04--0.08 muC in humans with 15--20 electrodes on each superior surface and 10 electrodes on each posterior cerebellar surface.

Innovations in neurologic implant systems.

Physiologic findings and clinical observations with several new implant systems are discussed. One system studied is designed with electrodes implanted over the anterior region of the spinal cord. Another system is designed with an anterior-posterior spinal cord electrode passing current across the spinal cord. Also given are observation and description of cortical and cerebellar implant systems. Current-density plots of the electrode arrays are included.

A comparison between anterior and posterior spinal implant systems.

In four patients with intractable pain from metastatic cancer, application of current through electrodes placed on the anterior surface of the cord produced analgesia and pain relief below the level of implant without the development of paresthesias. Application of current through electrodes placed on the dorsal columns in these patients also relieved pain, but to a lesser degree and with the development of associated paresthesias. In one patient, application of current from anterior electrodes to posterior electrodes produced a zone of dissociated sensory loss. While it is simpler to implant electrodes over the dorsal columns, the anterior location may be superior when currents are to be applied for the pain relief in the lower lumbar and sacral dermatomes.