Spinal neuromodulation using ultra low frequency waveform inhibits sensory signaling to the thalamus and preferentially reduces aberrant firing of thalamic neurons in a model of neuropathic pain

Abstract

Introduction: Many forms of chronic pain remain refractory to existing pharmacotherapies and electrical neuromodulation. We have recently reported the clinical efficacy of a novel form of analgesic electrical neuromodulation that uses ultra low frequency (ULF^™) biphasic current and studied its effects on sensory nerve fibers. Here, we show that in anesthetized rats, epidural ULF current reversibly inhibits activation of neurons in the thalamus receiving sensory spinothalamic input.

Methods: In naïve, neuropathic and sham-operated rats, recordings of ongoing and evoked activity were made from thalamic neurons, targeting the ventral posterolateral (VPL) nucleus.

Results: Responses to electrical stimulation of hind limb receptive fields were reduced in 25 of 32 (78%) neurons tested with lumbar epidural ULF neuromodulation. Cells preferentially responsive to low intensity stimulation were more likely to be found than cells responding to a range of stimulus intensities, or high intensity only; and low threshold responses were more likely to be inhibited by ULF than high threshold responses. On-going activity unrelated to hindlimb stimulation, observed in 17 of 39 neurons in naïve animals (44%), was reduced by lumbar epidural ULF current in only 3 of 14 (21%) neurons tested with ULF. By contrast, in rats with a well-characterized neuropathic injury, spinal nerve ligation (SNL), we found a much higher incidence of on-going activity in thalamic neurons: 53 of 55 neurons (96%) displayed firing unrelated to hindlimb stimulation. In this group, ULF current reduced thalamic neurone discharge rate in 19 of 29 (66%) neurons tested. In sham-operated animals, the incidence of such activity in thalamic neurons and the effect of ULF current were not significantly different from the naïve group.

Discussion: We conclude firstly that ULF current can acutely and reversibly interrupt signaling between sensory afferent fibers and relay neurons of the thalamus. Second, ongoing activity of thalamic neurons increases dramatically in the early stages following neuropathic injury. Third, this novel form of neuromodulation preferentially attenuates pathological thalamic activity in this neuropathic model compared to normal activity in naïve and sham-operated animals. This study, therefore, demonstrates that epidural ULF current can reduce nerve injury-related abnormal activity reaching the brain. These findings help advance understanding of possible mechanisms for the analgesic effects of ULF neuromodulation.

Keywords: ULF™ neuromodulation; inhibition; neuropathic; pain; spinal cord stimulation; thalamus.

Figure 1

Schematic drawing showing experimental setup. (A) Stimuli applied to the hind paw activate sensory fibers (B) projecting to the dorsal horn of the spinal cord. Some of these fibers ascend in the ipsilateral dorsal column to contact 2nd order neurones in the gracile nucleus of the brainstem. Axons of those neurons cross and ascend in the medial lemniscal pathway to the VPL nucleus of the thalamus. Other fibers synapse with dorsal horn neurones (C) that project via the lateral or ventrolateral spinothalamic tracts to VPL. (D) ULF current is applied via a pair of epidural electrodes positioned just lateral to the spinal midline. (E) Tungsten microelectrodes were stereotactically inserted into the VPL to record thalamic neuron responses (F) to stimulation of peripheral receptive fields or spontaneous activity (red flash mark in F indicates stimulus timing). (G) Detail of ULF current waveform.

Figure 2

Epidural ULF applied at the level of the L4-L5 spinal segments reversibly inhibits foot-shock evoked low-threshold activity of a thalamic neurone. (A) Continuous recording from a thalamic unit responsive only to low-intensity electrical and mechanical (not shown) stimulation, with some ongoing spontaneous activity. Above the neurogram, the timing of electrical stimuli is indicated, with the ULF current waveform shown at the top of the panel. Prior to ULF current (Aa), stimuli were applied at a constant rate of 1 Hz, usually evoking a pair of spikes. After 1 min of ULF current at 200 μA, stimuli were applied manually during the cathodal (Ab) and anodal (Ac) plateau phase of the ULF current. After switching off ULF, the constant 1 Hz stimulation was resumed (Ad). Ongoing spontaneous activity of the neurone seen at baseline (green shaded box) after the foot stimulus is discontinued, and throughout the recording, was unaffected by the ULF current. (B) Example sweeps showing details of the foot stimulus-evoked spikes at (Aa–Ad). At baseline, before ULF current (Ba), the neurone fires a pair of spikes in response to an electrical foot shock delivered to its receptive field (shaded area in inset). A foot stimulus applied during the cathodal (Bb) and anodal (Bc) plateau phases of the ULF current (200 μA), respectively, evokes no response of the thalamic neurone. When the ULF current is discontinued, the neurone again fires a pair of spikes in response to the resumed foot stimulus (Bd). (C) Raster plot compiled from a previous ULF application in the same cell where the current was applied at 100 μA and increased to 200 μA. Dots at the top (blue arrow) indicate stimulus artifacts; dots below indicate the occurrence of spikes, usually in pairs, with some jitter in the interspike interval. Plot shows absence of effect of ULF current at 100 μA, instant and complete inhibition of response at 200 μA, and immediate recovery when current is switched off (vertical scale at left, 0-25 msec post-stimulus).

Figure 3

ULF current reversibly attenuates cumulative response of an HT thalamic neurone to a sequence of electrical stimuli. The neurone was unresponsive at low intensity stimulation. High intensity pulses (2.0 msec, 3.4 mA, 2 pulses at 200 Hz) applied percutaneously via a pair of pin electrodes placed in the receptive field (lateral ankle) at a rate of 1 Hz evoked a cumulative response with many spikes at longer latency (100–1000 msec post-stimulus) and an after-discharge of variable duration. (A) Bars in chart show the cumulative responses of the neurone (spike count at 0–1000 msec post-stimulus plus 20 s after-discharge) to a sequence of 10 electrical stimuli. After stable baseline responses were recorded, ULF current was applied epidurally at spinal level L5-L6. No reduction in the cumulative response was seen at currents of 100–300 μA. After 12 min at 400 μA, the spike count was progressively reduced to 8% of the mean baseline. On discontinuation of the ULF current, response magnitude recovered fully over a period of 30 min. (B–D) Raster plots showing example responses at baseline (B), at 400 μA ULF (C), and at 25 min post-ULF (D). Insets on (C) show receptive field of neurone near ankle and overlay of 20 spikes recorded during a baseline stimulus sequence.

Figure 4

Incidence of spontaneously active thalamic neurons in each experimental group as a percentage of the whole group sample. Bars and significance levels are shown to indicate group comparisons with Fisher’s Exact test, using cell numbers from Table 2. ***p < 0.0001.

Figure 5

Effect of ULF current on spontaneous firing of thalamic neurons in each experimental group. Bars and significance levels are shown to indicate group comparisons with Fisher’s Exact test, using cell numbers from Table 3. **p < 0.01; ns, not significant (p = 0.128).

Figure 6

ULF current reversibly inhibits spontaneous firing in a thalamic neurone following L5 spinal nerve injury. Bars represent 1 min epochs. Numbers below chart indicate ULF current in μA. Firing rate recovered rapidly, and this cell appeared to exhibit a rebound increase in firing rate after ULF current was discontinued. Rate returned to pre-ULF baseline 9 min after ULF current was switched off.