NEUROIMAGING CENTER

Brain Imaging Center

LFMS: Low Field Magnetic Stimulation

A Test of a Low Field Magnetic Stimulation Device for Antidepressant Effects Using the Rodent Forced Swim Test
Michael L Rohan, MS^(1,2), William A Carlezon, PhD^(1,2), Kenroy Cayetano, MEng⁽¹⁾, Stephen Mague, BS⁽¹⁾, Hilarie C Tomasiewicz, BS⁽¹⁾, Elizabeth D Rouse, BS⁽¹⁾, Bruce M Cohen, MD, PhD^(1,2), Perry F Renshaw, MD, PhD^(1,2). ⁽¹⁾McLean Hospital, Belmont, MA, ⁽²⁾Department of Psychiatry, Harvard Medical School, Boston, MA.

Background

At the McLean Hospital Brain Imaging Center we have recently demonstrated anti-depressant effects in a novel echo-planar magnetic resonance spectroscopic imaging exam (EP-MRSI)[1]. Immediate mood enhancement effects comparable to or exceeding those of rTMS in subjects with bipolar disorder were demonstrated in a study of 30 subjects with bipolar disorder, 10 subjects with bipolar disorder who were given sham treatments, and 14 comparison subjects.

In the current study these effects are further demonstrated with a table-top low field magnetic stimulation (LFMS) system using a head-sized magnetic coil that mimics the read gradient field of the EP-MRSI exam. This follow-up trial used Sprague-Dawley rats in the Porsolt Forced Swim Test (FST)[2]. Anti-depressant effects, this time comparable to those of desipramine and fluoxetine, were demonstrated in this study.

LFMS provides magnetic stimulation at field strengths two orders of magnitude weaker than previously used in rTMS, allowing for coils to be built that provide full head penetration. These effects occurred using stimulation waveforms of a very different type than are used in rTMS. The potential applications of this effect are widespread, and the window into the mechanisms of magnetic and electric field neural stimulation mechanisms make the full development of this device and investigation of the mechanisms of its interaction with the brain a promising pursuit.

We have proposed that the mood enhancement was caused by electric fields in the brain, which were induced by the changing gradient magnetic field that was a part of the EP-MRSI exam. The details of these fields are described below.

Methods: Forced Swim Test

FST studies were conducted using our previously described methodology [3]. Briefly, on the first day of the FST, male Sprague-Dawley rats (325-350 g at testing) were placed in clear, 65 cm tall - 25 cm diameter cylinders filled to 48 cm with 25°C water. After 15 min of forced swimming, the rats were removed from the water, dried with towels, and placed in a warmed enclosure for 30 min. The cylinders were emptied and cleaned between rats. At 1hr, 19hr and 23 hr after exposure to forced swimming, the rats (10 per group) are restrained in plastic tubes (Decapicones; Braintree Scientific, Braintree MA) and exposed to LFMS or sham treatment for 20 min (i.e., 60 min total exposure over 23 hr).

At 24 hr after the forced swim, rats were re-tested for 5 min under identical swim conditions. Re-test sessions were videotaped from the side of the cylinders and scored using the behavioral sampling methodology [4] by raters unaware of the treatment condition. For behavioral sampling, rats were rated at 5 sec intervals throughout the duration of the re-test session. At each 5 sec interval, the predominant behavior was assigned to one of 4 categories: immobility, swimming, climbing, or diving. A rat was judged to be immobile if it made only movements necessary to keep its head above water, climbing if it made forceful thrashing movements with its forelimbs directed against the walls of the cylinder, swimming if it made swimming movements that caused it to move within the center of the cylinder, and diving if it swam below the water, toward the bottom of the cylinder.

Methods Male Sprague-Dawley rats (46 rats, 325-350 g) (N = 7-10 rats per group) were used to examine if exposure to LFMS has antidepressant-like effects in the FST. There were 6 groups of rats used to test 6 LFMS or pharmaceutical conditions. Between the first and second exposure to forced swimming, rats received on each of 3 occasions 20 min of exposure to either

LFMS within the focal point of the field (N=9),
LFMS outside the focal point of the field (N=6),
sham 1: no fields (N=9),
sham 2: constant Y magnetic field gradient (1.5 G/cm) (N=9).

These tests mimicked antidepressant administration regimens typically used with pharmacological treatments [3-5]. For comparison, the effects of

desipramine (DMI) (N=6) and
fluoxetine (FLX) (N=7)

were examined in parallel, using all of the same conditions as were used for LFMS. We used doses of these drugs that we have previously shown to be effective in the FST [3].

Methods: Statistics

Statistics Mean latency to immobility and mean number of samples for immobility, swimming and climbing behaviors were compared between all groups using ANOVA. Fisher's post hoc tests were used to examine specific LFMS, DMI and FLX measures in comparison to the sham groups.

Figure 1 Electric field magnitude contour plots for LFMS and rTMS. These contour plots represent a 20cm field of view in a coronal plane in the head. Contours of different colors show order of magnitude for the electric field strength. Green contours show electric fields less than 1 V/m at intervals of 0.01 V/m (these contours are not shown for the rTMS plot); blue contours show fields between 1 V/m and 10 V/m at 1 V/m intervals; red contours show fields between 10 and 50 V/m in intervals of 10 V/m. Fields greater than 50 V/m are not shown; fields in the rTMS coil exceed 500 V/m in the 1-2cm surrounding the coil. Both plots show an oval to represent a coronal section of the brain for context. rTMS contours were obtained by modeling the rTMS coil as a figure eight made of two 4cm diameter loops, tangent and coplanar. The loops are placed at a 45 degree angle and are shown as a thick diagonal line. The rTMS coil was modeled with 60,000 Amp-turns, producing a magnetic field of 20,000G at a distance of 1cm from the tangent point.

Figure 2 Magnetic field and electric field waveforms for LFMS and rTMS. Plots have a range of 10 msec. (a) The LFMS magnetic field is 512 alternating trapezoids, each 1 msec long, repeated for 20 min; LFMS magnetic fields range from 6 G to 0 G in the head. (b) The rTMS magnetic field is a single cycle sine pulse, period 0.280 msec, repeated at 20Hz to 1 Hz; rTMS magnetic fields range from 20,000G to 10 G in the head. Note that magnetic field plot magnitudes are differently scaled. (c) The LFMS electric field is a series of 512 alternating square pulses, each 0.25 msec long; the series is repeated for 20 min. (d) The rTMS electric field is a single cycle cosine pulse, period 0.280 msec, repeated at 20Hz to 1Hz. Note that electric field plot magnitudes are differently scaled.

Figure 3 Summary of behavior in the FST.

Figure 4 The LFMS Device

Low Field Magnetic Stimulation Device

A tabletop, stand-alone system was designed to deliver the LFMS electric field pulses to a human or animal subject. This system used an existing unshielded MRI head gradient coil, amplifier, liquid cooling system and personal computer to reproduce the gradient pulses delivered by the EP-MRSI exam in the MRI system. This 1st prototype system was used for the preliminary FST study described below.

Methods The head gradient coil is a cylinder about 14" in diameter and 18" long and 5/8" thick; coils that are embedded in this cylinder will deliver the same type of gradient magnetic field to the head as the gradient coils that are embedded in the MRI system. The gradient is a water-cut plate design for the transverse layers (X and Y) and a wire design for the Z layer. A copper cooling tube was embedded in the coil. The gradient strength for the 3 coils was 0.04 G/cm/A, and the vector potential strength per amp at the coil center was a0=2.5 mVs/m/A. The gradient coil was designed for limited use in an MRI system, and so it needed cooling to support the longer duty cycle required for LFMS use; this coil heating prevents its further use at desired future experimental fields and waveforms. The gradient coil was a surplus development prototype made by Advanced NMR Systems Inc., previously of Wilmington MA.

Design information was available for both the MRI gradient coil and the head sized gradient coils so that the electric field delivered with the smaller system could be matched to that in the original system. Electric field distributions for the LFMS coil and the MRI gradient coil were calculated using a Biot-Savart [6] style integration of simulated coil design fields containing 3D wire information with a 0.5cm linear resolution. Calculations were done using in-house gradient design and analysis software at McLean Hospital.

The power supply is a QDCM 380/135/220-LN switching amplifier made by MTS Automation Inc (Horsham, PA) which provides electrical power at peak values of 380V and 135A.

Control of the LFMS waveform was provided by commercial waveform generation software (LabVIEW, National Instruments Inc., Austin TX) installed on a standard personal computer. A user interface for operator control of the LFMS system was written using LabVIEW.

For use in the FST using rats a clinical and patient interface was not necessary, and so this prototype system was constructed without cosmetic or patient comfort considerations.

LFMS Magnetic and Electric Fields

Magnetic Field The LFMS system was used to deliver a Y magnetic gradient field. The Y gradient magnetic field is a field whose z component changes linearly in strength in the y direction. Spatially, the Y gradient magnetic field for the LFMS procedure can be expressed as

where G is the gradient magnetic field strength in Gauss/cm and "z hat" indicates field in the z direction. This field distribution is accurate over the region of the head and has differing fields outside this region which are not relevant to this experiment.

The magnetic field waveform provided by this LFMS system was a continuous series of alternating trapezoid pulses that were each 1.0 msec long with 125 μsec (0 to peak) ramp times and a 750 μsec flat top. The trapezoids were delivered continuously for 20 min.

Electric Field Any time-changing magnetic field is necessarily accompanied by an electric field. The electric field induced by these changing magnetic fields has a spatial distribution and waveform that are different from but related to the magnetic field distribution and waveform. Spatially, the induced electric field delivered by the Y magnetic gradient coil can be shown to have the form

Where "A0dot" and "G dot" are the time rate of change for the vector potential of the coil at the center of the coil and the time rate of change for the gradient magnetic field, respectively. Note that this electric field is solely in the right-to-left (X) direction. The vector potential at the center of the coil is obtained by simple Biot-Savart style integration over the coil design which determines the vector potential per amp (a0), and depends mostly on the length of wire and the size of the coil. This value is evaluated during the design phase of the coil and is the primary measure of induced electric field efficiency for a magnetic stimulation coil. For the purposes of clarity, it can be assumed that each magnetic coil has an a0 value associated with it. Time rates of change are simply given by the magnetic field waveform (specifically, by the rate of change of electric current in the magnetic coil). For the LFMS system used in this study a0= 2.5 mVs/m/A, G = 1.52 G/cm = 0.0152 Vs/m3. The coil was operated at 38A, with a ramp time of 125 μsec so the electric field in the center of the coil was about 0.75 V/m.

The electric field has a shape which is spatially saddle shaped with a significant offset in the middle of the coil; the magnitude of the saddle shaped variation is small compared to the offset and so the field can be considered spatially uniform to about 20%. The field points in the x direction (right-to-left in the subject).

The electric field delivered by this coil was a series of alternating square pulses that were each 250 μsec long. The electric field magnitude in each pulse was 0.75 V/m. The electric field pulses were delivered for 20 minutes for each treatment.

The sham LFMS condition (DC field) is the same as the full LFMS treatment except that the current in the coil is a DC current that produces the same coil heating and system noise as the full LFMS treatment, but with no induced electric fields.

Comparison: rTMS Electric Fields

Another device that induces electric fields into the brain has shown some antidepressant effects. This treatment is called repetitive transcranial magnetic stimulation (rTMS). The rTMS device uses a small coil shaped like a figure-8 or single loop a few inches across, held by hand near the scalp. Fast, biphasic (reversed) magnetic field pulses are delivered to the coil producing magnetic field pulses that are over 20,000G, and last about ½ msec; these pulses are delivered at about 20 Hz. The magnetic field varies from 20,000G near to coil to less than 50G on the opposite side of the head. The induced electric field pulses are also biphasic (reversing) pulses lasting ½ msec, occurring at about 20Hz, and varying in strength from 500 V/m near the coil to less than 1 V/m on the opposite side of the head. This treatment differs from the LFMS treatment with respect to the electric field pulse characteristics in several aspects. The pulses for rTMS are biphasic (reversing sign within the pulse) vs monophasic pulses for LFMS (square shape pulse); the pulses for rTMS are delivered at a frequency of at 20Hz vs a frequency of 1 kHz for r\LFMS; electric field strength for rTMS varies greatly over the head (120 V/m to <1 V/m) vs a relatively uniform electric field (±10%) for LFMS; and field direction for rTMS (Roth et al, 1991) is not well defined (roughly circular) vs LFMS uniformly in the right-to-left direction.

rTMS shows some antidepressant effects in human trials, which are ongoing [7] [8-10]. We feel that we may be exploiting a similar mechanism, but the time scale and fields that we use make it difficult to speculate on any link in more detail; in particular, the differing spatial dependence of the electric fields and the pulse rates may suggest different cellular mechanisms and differing anatomic regions might be involved.

Special note should be made of the applicability of rodent models in the investigation of rTMS [11-13] and of LFMS. The rTMS field is highly non-uniform, changing from maximum to 1/1000 of maximum over the space of the human head, while the change over the space of the head of a rat is much less. Thus, there is concern that the relative distribution of fields in the human and rat brain is very different, when using the same device for testing. The LFMS device provides a uniform electric field over the human head, so that the free electric field for both human and rodent trials is comparable. There will, of course, be a difference due to the different tissue structure and size of the two brains when the effect of the brain on the field is considered.

Results

LFMS dramatically reduced immobility, suggesting antidepressant-like effects. Immobility was affected in this test (F(4,40)=5.0, p<0.01) with LFMS (p<0.01) and DMI (p<0.05) treatments having reduced immobility. Swimming was affected in this trial (F(4,40)=4.24, p<0.01) with LFMS (p<0.01) and FLX (P<0.05) showing increased swimming activity. Climbing was affected (F(4,40), p<0.01) with DMI (p<0.01) and FLX (p<0.05) showing increased swimming activity. The effects of LFMS were qualitatively similar to, but larger than, those of FLX, the effects of which may have been reduced by the additional stress associated with repeatedly restraining the rats for the LFM.

Additional tests were done to rule out potential confounds of memory or learning impairment caused by the LFMS. Rats were trained in a standard fear-potentiated startle (FPS) procedure (Tsvetkov et al., 2002). Beginning 1 hr after FPS, rats received LFMS or sham treatment according to the same schedule used for the FST studies. Upon testing for FPS, no differences were observed between rats that received LFMS or sham treatment. These data suggest that the antidepressant-like effects of LFMS in the FST are not due to impairments in learning or memory.

Discussion

From this study we have confirmed the existence of a significant magnetic stimulation effect using the LFMS field and waveform, using rats in the Porsolt FST. In this test it has shown antidepressant effects resembling those of fluoxetine. This confirms earlier, serendipitous discovery in bipolar human subjects in a MRI system during an atypical magnetic resonance spectroscopic imaging exam. This effect is surprising because of the very low magnetic (<10G) and electric (< 1V/m) fields involved. Further investigations are being planned.

>This work was supported in part by MH-58681, the Poitras Foundation, the Stanley Foundation Bipolar Disorders Research Center at McLean Hospital, and gifts from John and Virginia Taplin.

Rohan, in press. 2003.
Porsolt, R.D., et al., Behavioural despair in rats: a new model sensitive to antidepressant treatments. Eur J Pharmacol, 1978. 47(4): p. 379-91.
Carlezon, W.A., et al., Antidepressant-like effects of cytidine in the forced swim test in rats. Biol Psychiatry, 2002. 51(11): p. 882-9.
Detke, M.J., M. Rickels, and I. Lucki, Active behaviors in the rat forced swimming test differentially produced by serotonergic and noradrenergic antidepressants. Psychopharmacology (Berl), 1995. 121(1): p. 66-72.
Porsolt, R.D., M. Le Pichon, and M. Jalfre, Depression: a new animal model sensitive to antidepressant treatments. Nature, 1977. 266(5604): p. 730-2.
Jackson, J.D., Classical Electrodynamics. 2nd Edition ed. 1975, New York: John Wiley and Sons.
George, M.S., et al., Daily repetitive transcranial magnetic stimulation (rTMS) improves mood in depression. Neuroreport, 1995. 6(14): p. 1853-6.
Holtzheimer, P.E., 3rd, J. Russo, and D.H. Avery, A meta-analysis of repetitive transcranial magnetic stimulation in the treatment of depression. Psychopharmacol Bull, 2001. 35(4): p. 149-69.
9. Burt, T., S.H. Lisanby, and H.A. Sackeim, Neuropsychiatric applications of transcranial magnetic stimulation: a meta analysis. Int J Neuropsychopharmacol, 2002. 5(1): p. 73-103.
Schlaepfer, T.E., M. Kosel, and C.B. Nemeroff, Efficacy of Repetitive Transcranial Magnetic Stimulation (rTMS) in the Treatment of Affective Disorders. Neuropsychopharmacology, 2003. 28(2): p. 201-5.
11. Sachdev, P.S., et al., Effects of different frequencies of transcranial magnetic stimulation (TMS) on the forced swim test model of depression in rats. Biol Psychiatry, 2002. 51(6): p. 474-9.
Keck, M.E., et al., Repetitive transcranial magnetic stimulation induces active coping strategies and attenuates the neuroendocrine stress response in rats. J Psychiatr Res, 2000. 34(4-5): p. 265-76.
Keck, M.E., et al., Neuroendocrine and behavioral effects of repetitive transcranial magnetic stimulation in a psychopathological animal model are suggestive of antidepressant-like effects. Neuropsychopharmacology, 2001. 24(4): p. 337-49.