Evaluating optimized temporal patterns of spinal cord stimulation (SCS)

Abstract

Background: Temporal patterns of stimulation represent a novel dimension for improving the efficacy of spinal cord stimulation to treat chronic neuropathic pain. Objective: We hypothesized that nonregular temporal patterns of stimulation designed using a computational model would be superior to conventional stimulation at constant frequencies or completely random patterns of stimulation. Methods: Using a computational model of the dorsal horn network and an optimization algorithm based on biological evolution, we designed an optimized pattern of spinal cord stimulation with comparable efficacy and increased efficiency relative to constant frequency (CF) stimulation. We evaluated the effect of different temporal patterns on individual neurons recorded in the dorsal horn of urethane-anesthetized rats. Results: The optimized pattern and 50 Hz CF stimulation produced greater inhibition of spontaneously firing neurons recorded in vivo than random 50 Hz stimulation or a pattern designed intentionally with poor fitness. Spinal Cord Stimulation (SCS) led to significant changes in the firing patterns of recorded units, and stimulation patterns that generated significant inhibition also tended to reduce entropy and regularize the firing patterns of units, suggesting that patterns of dorsal horn neuron activity may be important for pain perception in addition to the firing rate. Conclusions: These results demonstrate that the computational model can be used as a tool for optimizing stimulation parameters and suggest that optimized temporal patterns may increase the efficacy of spinal cord stimulation.

Keywords

• Spinal cord stimulation
• Genetic algorithm
• Neuropathic pain
• Computational model
• Spared nerve injury