Conventional and kilohertz-frequency spinal cord stimulation produces intensity- and frequency-dependent inhibition of mechanical hypersensitivity in a rat model of neuropathic pain

Abstract

Background: Spinal cord stimulation (SCS) is a useful neuromodulatory technique for treatment of certain neuropathic pain conditions. However, the optimal stimulation parameters remain unclear.

Methods: In rats after L5 spinal nerve ligation, the authors compared the inhibitory effects on mechanical hypersensitivity from bipolar SCS of different intensities (20, 40, and 80% motor threshold) and frequencies (50, 1 kHz, and 10 kHz). The authors then compared the effects of 1 and 50 Hz dorsal column stimulation at high- and low-stimulus intensities on conduction properties of afferent Aα/β-fibers and spinal wide-dynamic-range neuronal excitability.

Results: Three consecutive daily SCS at different frequencies progressively inhibited mechanical hypersensitivity in an intensity-dependent manner. At 80% motor threshold, the ipsilateral paw withdrawal threshold (% preinjury) increased significantly from pre-SCS measures, beginning with the first day of SCS at the frequencies of 1 kHz (50.2 ± 5.7% from 23.9 ± 2.6%, n = 19, mean ± SEM) and 10 kHz (50.8 ± 4.4% from 27.9 ± 2.3%, n = 17), whereas it was significantly increased beginning on the second day in the 50 Hz group (38.9 ± 4.6% from 23.8 ± 2.1%, n = 17). At high intensity, both 1 and 50 Hz dorsal column stimulation reduced Aα/β-compound action potential size recorded at the sciatic nerve, but only 1 kHz stimulation was partially effective at the lower intensity. The number of actions potentials in C-fiber component of wide-dynamic-range neuronal response to windup-inducing stimulation was significantly decreased after 50 Hz (147.4 ± 23.6 from 228.1 ± 39.0, n = 13), but not 1 kHz (n = 15), dorsal column stimulation.

Conclusions: Kilohertz SCS attenuated mechanical hypersensitivity in a time course and amplitude that differed from conventional 50 Hz SCS, and may involve different peripheral and spinal segmental mechanisms.

Fig. 1

Effects on paw withdrawal threshold of different frequencies of spinal cord stimulation (SCS) applied at 20–80% motor threshold in nerve-injured rats. (A) The paw withdrawal thresholds, shown as percent of prespinal nerve ligation (SNL) baseline, before and at different time points after SCS at one of three frequencies (50, 1 kHz, and 10 kHz) delivered at 20, 40, and 80% motor threshold (MoT, 0.024 ms, constant current, 30 min/session, 1 session/day). Data from animals that received sham stimulation in different studies were combined for analysis. Error bars are omitted to improve clarity. (B) On each treatment day, the overall effect of SCS on PWT was compared among groups by averaging together the PWTs at pre-SCS and 15, 30, and 60 min after the initiation of SCS applied at 20% MoT, 40% MoT, and 80% MoT. Data are expressed as mean + SEM in (B). +P < 0.05, ++P < 0.01, +++P < 0.001 versus pre-SCS PWT; *P < 0.05, **P < 0.01, ***P < 0.001 versus sham stimulation; ^†P < 0.05 versus 50 Hz.

Fig 2

Dorsal column conditioning stimulation (CS) of 1 Hz and 50 Hz frequencies changed Aα/β-fiber conduction property in nerveinjured rats. (A) Left: schematic diagram illustrating the experimental setup for recording antidromic sciatic compound action potentials (APs) evoked by graded test electrical stimulation (0.1–2.2 mA, 0.2 ms) applied at the L4 dorsal root in rats that received an L5 spinal nerve ligation (SNL). The CS (5 min, 50 and 1 kHz, 0.024 ms, biphasic pulse) was delivered to the ipsilateral dorsal column at T13–L1 level. Right: examples of different compound AP waveforms corresponding to Aα/β- and Aδ-fiber activation to increasing intensities of dorsal column CS and dorsal root test stimulation. The intensity for CS was calibrated by recording sciatic compound AP to graded dorsal column stimulation (0.01–2.0 mA, 0.024 ms, biphasic pulse): The intensity that resulted in the first detectable Aα/β waveform (Ab0), followed by the peak Aα/β waveform (Ab1, the highest Aα/β waveform intensity without inducing an Aδ waveform), to dorsal column stimulation was determined. (B) Examples of sciatic compound APs evoked by 0.2 and 0.4 mA dorsal root test stimulation before and after dorsal column CS (1 kHz, Ab1, 5 min) were shown. (C) In the off-line analysis, the areas under the Aα/β waveforms generated by graded dorsal root stimulation were measured to establish the stimulus–response (S–R) functions. The S–R functions were not changed after 50 Hz CS of Ab0 intensity, but were significantly depressed at 0–5 min after 50 Hz CS at Ab1 intensity. (D) The size of Aα/β waveform to the lower intensities of dorsal root stimulation (0.2–0.6 mA) was significantly decreased from the prestimulation baseline at 0–5 min after 1 kHz dorsal column CS of Ab0 intensity. The S–R function was significantly depressed by 1 kHz CS of Ab1 intensity. Data are expressed as mean, and error bars are not shown to improve clarity. *P < 0.05, **P < 0.01, ***P < 0.001 versus pre-CS baseline.

Fig 3

Dorsal column conditioning stimulation of 50 Hz, but not of 1 kHz, inhibited windup in wide-dynamic–range (WDR) neurons of nerve-injured rats. (A) Schematic diagram illustrating the experimental setup for recording WDR neuron in the L4 spinal segment of rats that received an L5 spinal nerve ligation. The conditioning stimulation (50 and 1 kHz, 0.024 ms, Ab1, 5 min) was delivered through two tungsten needle electrodes inserted in the ipsilateral dorsal column at T13–L1 level. (B) An analog recording of the WDR neuronal response to the first, fourth, and eighth stimulus of a train of intracutaneous electrical stimuli (0.5 Hz, 16 pulses, 2.0 ms, supra-C threshold) before and 0–15 min after 50 Hz dorsal column stimulation (Ab1, 0.024 ms, 5 min). (C) The number of action potentials (APs) in C-component of WDR neuronal response to 0.5 Hz stimulation and 0.1 Hz stimulation were plotted against the stimulation sequence number of each trial. For clarity, error bars are not shown. (D) At 0–15 min and 30–45 min after 50 Hz (n = 13), but not 1 kHz (n = 15), dorsal column stimulation, the total C-component of WDR neuronal response evoked by 0.5 Hz (16 pulses) windup-inducing stimulation delivered to the skin receptive field (i.e., area under the windup function) was –significantly decreased from the prestimulation baseline. (E) The stimulus–response (S–R) functions of the C-fiber component of WDR neurons to graded intracutaneous electrical stimulation (0.1–10 mA, 2.0 ms) are shown before and after 50 Hz (n = 13) and 1 kHz (n = 15) dorsal column stimulation. (F) The area under the S-R function (i.e., total numbers of APs in C-fiber component) was not significantly decreased from baseline after dorsal column stimulation. Data are presented as mean +SEM. *P < 0.05 versus prestimulation baseline.