Neuropathic pain and neurofeedback training

Central neuropathic pain (CNP) has a prevalence of 40 % in patients with spinal cord injury. It is caused by an injury to the somato-sensory system and occurs with a high prevalence in cases of amputation, spinal cord injury, multiple sclerosis, Parkinson disease and stroke.

CNP symptoms do not respond well to medication and the drugs used to treat this type of pain are often associated with problematic effects. This has generated interest in non-pharmacological treatments based on neuromodulation and neurostimulation, such as hypnosis, meditation,  neurofeedback, repetitive Transcranial Magnetic Stimulation (rTMS) and transcranial Direct Current Stimulation (tDCS).

Electroencephalography (EEG) studies have shown that this type of pain has identifiable "signatures" in it's frequency spectrum, that could potentially be targeted by a neuromodulation therapy such as neurofeedback training  This could lead to a safe, non-invasive therapy with no known side effects.

A study by Hassan et al investigated the impact of neurofeedback training on central neuropathic pain and its underlying brain signatures in patients with chronic paraplegia.

 Identifying the Frequency Content of the EEG signal (bands) is easy with NeXus hardware

Identifying the Frequency Content of the EEG signal (bands) is easy with NeXus hardware

Neurofeedback has been widely used for treatment of chronic pain conditions such as complex regional pain syndrome, fibromyalgia and migraine. Neurofeedback studies specifically for treatment of CNP have been considered in the past but were inconclusive with respect to the optimal training protocol; most likely due to the small number of treatment sessions.

Identifying Frequency bands and feedback thresholds

Most neurofeedback protocols for chronic pain, target the temporal or central area of the cortex, up-regulating EEG activity in the lower β or α frequency band and down-regulating the activity in the Ɵ and higher β frequency bands.  By up-or down-regulating we mean using a feedback approach to influence brain activity as indicated by the magnitude of the EEG signal in these specific frequency bands.

Based on feedback information patients can be trained to voluntarily influence the brain activity thought to be associated with pain processing.  During neurofeedback training for the treatment of chronic pain, EEG is typically measured at one "training site" only, providing no evidence of global modulation of EEG during training.

The neurofeeback sessions commenced by measuring "baseline" EEG activity followed by audio neurofeedback provided from the occipatal region with eyes closed for relaxation. Feedback provided to the patient simply indicated the EEG level in the frequency bands of interest as a "bar chart". Full details are presented in the referenced paper.

Studies carried out to date indicate a close relation between the existence of CNP and ‘over activation’ of the sensory-motor cortex.  Neurofeedback can target the motor cortex, resulting in normalisation of evoked responses and a reduction of CNP.

In a study by Hassan et al the patients’ EEG activity was modulated from the sensory-motor cortex, electrode location C3/Cz/C4/P4 in up to 40 training sessions

Six out of seven patients reported immediate reduction of pain during neurofeedback training. Best results were achieved with suppressing Ɵ and higher β (20–30 Hz) power and reinforcing α power at C4.  Four patients reported clinically significant long-term reduction of pain (>30 %) which lasted at least a month beyond the therapy.

EEG during neurofeedback revealed a wide spread modulation of power in all three frequency bands accompanied with changes in the coherence most notable in the beta band. The standardised low resolution electromagnetic tomography analysis of EEG before and after neurofeedback therapy showed the statistically significant reduction of power in beta frequency band in all tested patients. Areas with reduced power included the Dorsolateral Prefrontal Cortex, the Anterior Cingulate Cortex and the Insular Cortex.

The authors conclude that Neurofeedback training produced both immediate and longer term reduction of central neuropathic pain that is accompanied with a measurable short and long term modulation of cortical activity. Controlled trials are required to confirm the efficacy of this neurofeedback protocol on treatment of pain. The study is a registered UKCRN clinical trial Nr 9824.

Muhammad Abul Hassan, Matthew Fraser, Bernard A. Conway, David B. Allan and Aleksandra Vuckovic
“The mechanism of neurofeedback training for treatment of central neuropathic pain in paraplegia: a pilot study”
BMC Neurology (2015) 15:200
DOI 10.1186/s12883-015-0445-7

Related Articles