C-RFA of Residual Limb Neuroma

Residual limb (RLP) and phantom limb pain (PLP) affects most amputees at some point in their life1. The incidence of PLP has been estimated to range between 50 - 80%. RLP prevalence has been estimated to be 43%. The peak of onset is bimodal and often appears within the first month and second year after amputation. RLP is more common in the first year after amputation, with PLP becoming the predominate amputee pain complaint after one-year post-amputation.

Both RLP and PLP fall under the umbrella term "post-amputation pain." While these conditions are frequently found in combination, their clinical features and underlying causes are distinct. PLP is a painful sensation in the distribution of the missing limb. Following amputation, abnormalities at multiple levels of the neural axis have been implicated in the development of PLP; changes include cortical reorganization, reduced inhibitory processes at the spinal cord, synaptic response changes and hyperexcitability at the dorsal root ganglion, and retrograde peripheral nerves shrinkage.

Residual limb pain has been called "neuroma pain" and is mechanistically distinct from PLP11. Neuromas may form as early 6-10 weeks after nerve transection, and are thought the produce ectopic neural discharges resulting in severe pain. Evidence suggests RLP and PLP commonly co-occur and patients may struggle to differentiate between these pain types. Risk factors include female sex, upper extremity amputation, pre-amputation pain, residual pain in contralateral limb, and time since amputation.

Depression, anxiety, and stress are known to exacerbate PLP / RLP. Patients experiencing PLP and RLP also experience a higher incidence of indecisiveness, suicidal ideation, and thoughts of self-harm8. Current guidelines for treatment of PLP and RLP are not standardized. Treatments includes pre-operative analgesia, neuromodulation mirror therapy, imagery, acupuncture, transcranial stimulation, deep brain stimulation, and medications (including, but not limited to: TCAs, SSRIs, gabapentinoids, sodium channel blockers, ketamine, opioids, and NSAIDs). Many agents have been injected in neuromas. These include local anesthetic, phenol, alcohol, and botulinum toxin. These oral, intravenous, and nonpharmacological modalities have demonstrated limited success in the treatment of PLP / RLP. Neuroma cryoablation has been used, but this method of neural destruction poses technical challenges related to cumbersome needle placement and the requirement for time-intensive freeze-thaw cycles.

Conventional RFA has been studied on RLP. Zhang et. al treated 13 patients with painful stump neuromas. The study started with alcohol neurolysis before using ultrasound-guided RFA for refractory cases. The frequency of sharp pain was reduced in all RFA-treated patients. Kim et. al described a case in which ultrasound-guided RFA was successfully used to treat a sciatic neuroma of an above-knee amputee.

No outcome literature on the effectiveness of C-RFA technology has been published. C-RFA is similar in mechanism to conventional RFA: a thermal lesion is created by applying radiofrequency energy through an electrode placed at a target structure. In C-RFA, a constant flow of ambient water is circulated through the electrode via a peristaltic pump, maintaining a lowered tissue temperature by creating a heat sink. By removing heat from tissues immediately adjacent to the electrode tip, a lower lesioning temperature is maintained, resulting in less tissue charring adjacent to the electrode, less tissue impedance and more efficient heating of target tissue. The volume of tissue heated, and the resultant thermal lesion size is substantially larger with C-RFA, conferring an advantage over conventional RFA. Further, given the spherical geometry and forward projection the C-RFA lesions beyond the distal end of the electrode, the RFA probe can be positioned at a range of possible angles and still capture the target neural structure, whereas more fastidious, parallel positioning is required with conventional RFA. These technical advantages increase the probability of successful denervation of neural pain generators that have variability in anatomic location. Additionally, a longer lesion of the RLP-generating nerve may be more reliably achieved with C-RFA compared to conventional RFA.

As such, the present study aims to define the attributable effect of cooled RFA on pain, physical function, and health-related quality of life in patients with post-amputation neuroma-associated residual limb pain. This prospective single-arm pilot study is intended to inform a future properly powered randomized controlled trial.