FormalPara Key Summary Points

There was a statistically significant difference in pain scores with the use of temporary peripheral nerve stimulation (PNS) as a treatment for chronic pain at various sites with an average pre-procedure visual analog scale (VAS) pain score of 6.4 and an average post-procedure VAS pain score of 4.2.

Temporary PNS may be useful in the treatment of chronic pain refractory to more conventional treatment measures

More research regarding specific target sites and device settings should be conducted to identify the most efficacious approach for temporary PNS treatment.

Introduction

Chronic pain is very prevalent in the health care system, affecting over 20.4% of adults in the United States (US) and approximately 8% of US adults reporting high-impact chronic pain in 2016 [1]. Chronic pain is defined as persistent or recurrent pain for ≥ 3 months, by the International Association for Study of Pain, and is an underlying driver for negative health outcomes, opioid dependency, mental illness, reduced life expectancy, and poor quality of life in adults who suffer from it [1, 2]. Furthermore, the negative effects of chronic pain disproportionally affect patients of lower socioeconomic status [3, 4]. Chronic pain prevalence is predicted to increase further as obesity rates increase and as the US population ages and shifts towards more sedentary lifestyles [3].

Conventional treatments for chronic pain include non-invasive methods such as oral analgesics and non-steroidal anti-inflammatory drugs (NSAIDs), anti-epileptic drugs and antidepressants, physical therapy, and exercise [5]. In refractory chronic pain, where the pain does not respond to conventional treatments, more invasive approaches are indicated, such as nerve block injections, denervation surgery, implantable drug delivery systems, and nerve stimulators [6]. Given the risk associated with invasive approaches, as well as the burden of mounting cost and pain medication addiction in patients managed with medication therapy, there remains a need to identify methods to treat chronic pain that are effective, minimally invasive, and non-habit-forming [6].

The first modern use of electrical neurostimulation to temporarily treat pain was reported by Wall & Sweet in 1967. In recent years, the progression of peripheral nerve stimulation (PNS) therapy has led to developments in potential improvements in post-operative pain control, variability in lead placement depending on the location of need, breakthroughs concerning PNS influence on local inflammatory mediators that are linked to increased pain threshold and pain neurotransmission inhibition, and overall improved patient satisfaction with treatment and quality of life [6, 7]. PNS involves the implantation of an electrode with 8–16 contact leads in close proximity to the target peripheral nerve in addition to an implanted or external battery power source for stimulation of the electrode [8]. The electrode then delivers electrical stimulation for 60 days, after which the electrode and device are removed.

Recently, PNS therapy development has demonstrated promising results regarding increased physical activity, pain control, increased employment, reduced medication intake (including reduced opioid use), sleep pattern improvements, relief from pain and depression, and improved quality of life [7,8,9,10,11,12,13]. Two mechanisms have been proposed underlying the attenuation of pain by peripheral nerve stimulation—direct inhibition of pain neurotransmission through changes to local inflammatory mediators and elevation of pain thresholds via stimulation [7]. The use of PNS therapy to modulate pain was inspired by the “gate control” theory of pain, wherein non-painful stimuli can override downstream transmission of pain signaling, as described in Deer et al. [13, 14]. More specifically, it is believed that stimulation of non-painful sensory fibers causes a cascade of inhibitory signals to be transmitted to dorsal horn interneurons that would then inhibit pain nociceptive signals from A-delta and C fibers [15]. However, the exact mechanisms leading to the therapeutic effects remain unclear and are widely debated.

This retrospective study aimed to assess the efficacy of PNS in patients suffering from chronic pain affecting various parts of the body including areas innervated by cervical, lumbar, thoracic, suprascapular, brachial, radial, inguinal, sciatic, saphenous, tibial, and peroneal nerves. The absence of a control group or placebo group was not necessary, as this is a retrospective study and not a randomized control trial. Additionally, there would be ethical concern regarding the intentional implantation of patients without providing the planned therapy or withholding potential therapy from patients with refractory chronic pain. A decrease in pain scores was expected following peripheral nerve stimulator therapy.

Methods

This retrospective analysis consists of data collected from electronic health records for patients who underwent temporary peripheral nerve stimulation (PNS) therapy on various anatomical locations from January 2015 until June 2022 at the University of Wisconsin Hospital and Clinics in Madison, WI. A total of 89 patients are included in the data analysis. An institutional review board application was submitted with a minimal risk research exemption granted due to the studies retrospective nature and lack of contact with the patients who underwent the procedure. Data collected relate to patient age, sex, BMI, diagnosis, targeted nerves, visual analog scale (VAS) pain scores from pre and post PNS therapy, subjective reported percent of pain improvement post PNS therapy, and duration of improvement recorded in days.

The targeted nerves include the C2 medial branch nerve, T9 nerve, T10 intercostal nerve, L3 nerve, L4 nerve, L5 medial branch nerve, the brachial plexus along the interscalene groove, the supraclavicular nerve, suprascapular nerve, radial nerve, median nerve, ilioinguinal nerve, genitofemoral nerve, iliohypogastric nerve, cluneal nerve, lateral femoral nerve, popliteal nerve, tibial nerve, posterior tibial nerve, saphenous nerve, the infrapatellar branch of the saphenous nerve, peroneal nerve, the superficial peroneal nerve, and the sural nerve. The variety and quantity of nerves targeted in each case was determined by the managing physician, based on each patient’s clinical presentation.

Statistical Methods

Statistical analysis used SPSS software, version 26 (IBM), using a paired t test to assess significance between pre and post PNS therapy pain scores. Pearson’s correlation analysis regarding the impact of age and BMI on VAS scale improvement and subjective percentage improvement was also conducted using SPSS software, version 26 (IBM). P values are significant if they were found to be ≤ 0.05. Analysis includes consideration of degree of freedom and confidence intervals.

Peripheral Nerve Stimulator Therapy Procedure

All patients who underwent PNS implant suffered from chronic pain refractory to conservative management, which is described as having a natural course of disease with unremitting pain for ≥ 3 months. Patients were considered eligible for intervention after they were determined to have chronic, unremitting pain that demonstrated characteristics refractory to conservative management. All patients received a diagnosis consistent with neuralgia or neuropathy, and a nerve-specific diagnostic block was performed. Improvement in pain exceeding 50% was considered confirmatory.

Procedures were performed using both fluoroscopic and ultrasound guidance; 1% lidocaine was used to anesthetize the skin and subcutaneous tissue. An 18-g stimulating needle was advanced towards the targeted points under imaging guidance. When the needle approached the target nerve, stimulation (through a stimulating needle) was used to identify the appropriate placement. Then the lead was deployed at the target location. Two leads were placed for every case, and programming of the stimulating device was titrated to patient-specific tolerance during lead placement. Intensity values utilized varied from 0.2 to 30 mA and 10–200 µs. The frequency was set to either 12 Hz for motor stimulation program (usually 6–12 h per day) or 100 Hz for the sensory stimulation program (usually up to 24 h per day). Patients were allowed to increase or decrease the stimulation setting throughout the 60-day treatment period. For patients who completed therapy, leads were removed in the clinic 60 days after placement.

Results

The mean age of patients included in this study was 62.21 and the mean BMI was 31.03 (Table 1). All patients were diagnosed with various chronic pain syndromes that were refractory to conventional pain treatments and the nerves targeted for PNS device placements were based on the location of pain (Table 2). Four patients (4.5%) had a diagnosis of cervical facet arthropathy and bilateral occipital neuralgia. Two patients (2.2%) had a diagnosis of low back pain. Two patients (2.2%) had a diagnosis of thoracic back pain. One patient (1.1%) had a diagnosis of lumbar disc disease. One patient (1.1%) had a diagnosis of lumbar neuropathy. Nine patients (10.1%) had a diagnosis of lumbar spondylosis without myelopathy. Six patients (6.7%) had a diagnosis of lumbar facet arthropathy. Nineteen patients (21.3%) had a diagnosis of suprascapular neuropathy. One patient (1.1%) had a diagnosis of chronic chemotherapy-induced neuropathy. One patient (1.1%) had a diagnosis of chronic headaches. One patient (1.1%) had a diagnosis of diabetic polyneuropathy. Five patients (5.6%) had a diagnosis of brachial plexopathy. Three patients (3.4%) had a diagnosis of radial neuropathy. Two patients (2.2%) had a diagnosis of median neuropathy of the hands. One patient (1.1%) had a diagnosis of neuroma of upper extremity amputation. One patient (1.1%) had a diagnosis of intercostal neuralgia. Two patients (2.2%) had a diagnosis of ilioinguinal and iliohypogastric neuropathies. Six patients (6.7%) had a diagnosis of peroneal neuropathy. Two patients (2.2%) had a diagnosis of both peroneal and tibial neuropathies. Two patients (2.2%) had a diagnosis of both peroneal and sural neuropathies. Five patients (5.6%) had a diagnosis of sciatic neuropathy. One patient (1.1%) had a diagnosis of cluneal neuralgia. One patient (1.1%) had a diagnosis of femoral artery injury. One patient (1.1%) had a diagnosis of femoral cutaneous neuropathy. Six patients (6.7%) had a diagnosis of saphenous neuropathy. One patient (1.1%) had a diagnosis of knee osteoarthritis. Two patients (2.2%) had a diagnosis of posterior tibial neuropathy. One patient (1.1%) had a diagnosis of foot neuropathy (Table 2).

Table 1 Demographics of patients receiving PNS therapy
Full size table
Table 2 Summary of patient diagnoses
Full size table

The mean pre-op pain score before PNS therapy was 6.36 (SD = 2.18, SEM = 0.23) and the mean post-op pain score after PNS therapy was 4.19 (SD = 2.70, SEM = 0.29) (Table 3). The mean patient-reported percent improvement in pain following PNS therapy was 49.0% (min = 0.00, max = 100.00, SD = 34.8). The improvement in pain scores between pre-op and post-op was statistically significant with a p value < 0.001 (M = 2.2, SD = 2.8), 95% CI [1.6, 2.8]. Overall, 23.6% (21) patients reported no improvement in their VAS pain scores; of these 21 patients who reported no VAS pain score reduction, nine reported improvement in subjective percent pain improvement and five did not successfully complete their PNS treatment. Seven (7.9%) patients with completed treatment of 60 days reported no improvement by VAS or subjective percent pain improvement. Duration of improvement was recorded for 77 patients. The mean duration of improvement for patients was 123 days after initiation of therapy (min = 6, max = 683, SD = 126) (Table 3).

Table 3 Mean pre-op pain scores and post-op pain scores following PNS therapy
Full size table

Pearson’s correlation coefficient assessing age and VAS improvement was r = 0.14, sig (two-tailed) = 0.20. Pearson’s correlation coefficient assessing age and subjective percent improvement was r = − 0.15, sig (two-tailed) = 0.18. Pearson’s correlation coefficient assessing BMI and VAS improvement was r = − 0.05, sig (two-tailed) = 0.62. Pearson’s correlation coefficient assessing BMI and subjective percent improvement was r = − 0.03, sig (two-tailed) = 0.77 (Table 4). Of the 89 patients who underwent temporary PNS therapy, there were a total of five adverse events. Four of these events were related to lead migration and one was an infection (Table 5).

Table 4 Pearson’s coefficient analysis of co-variables
Full size table
Table 5 Distribution of adverse events
Full size table

Discussion

Brief Summary and Published Data Comparison

This study investigated the safety and efficacy of peripheral nerve stimulation (PNS) in treating a wide range of chronic pain conditions and aimed to provide some initial evidence for which pain conditions may be most conducive to intervention with PNS. Similar to the previous studies, the results of this retrospective analysis suggest the efficacious role of peripheral nerve stimulation in management of chronic pain that is refractory to conventional treatment measures [7,8,9,10,11,12]. Our study found a statistically significant improvement in pain scores following PNS therapy with an average pre-procedure pain score of 6.36 and an average post-procedure pain score of 4.19 with a P value < 0.001. In addition to the reduction of patient-reported pain levels, patients reported improved sleep, mood, and ability to perform daily activities. In this study, the distinction between standardized visual analog scale (VAS) pain score relief and subjective percent pain improvement is necessary to address determined pain relief. VAS pain score improvement and subjective percent pain improvement were used to establish chronic pain relief. Patients who experienced no percent improvement and reported improvement in VAS pain scores were designated as successful therapy. Of the patients who underwent PNS therapy, 23.6% (21) experienced no pain relief designated by VAS pain score improvement. Nine of the 21 patients who reported no VAS pain score improvement had experienced subjective percent improvement and were designated as successful therapy. Five of the remaining 12 patients with no reported VAS pain score improvement nor subjective percent pain improvement experienced adverse events such as lead migration or infection, which can contribute to the absence of pain relief. The remaining seven patients with no pain improvement without adverse therapy effects were determined to have no true relief. The evaluation of co-variable influence on VAS pain improvement and subjective percent improvement revealed non-significant results with all the Pearson’s correlation p values determined to be greater than the ≤ 0.05 significance standard. This strengthens the argument for future research to focus on the influence of which nerves are targeted and where the leads are located when determining more successful methods of PNS therapy, as well as further identify the potential role of age, body mass index (BMI), and other demographics that may impact the procedure’s success.

Evaluation of the effectiveness of PNS within diagnostic subgroup was limited by the number of patients who presented with each indication; however, the results of this study provide some preliminary evidence for diagnostic indications in which PNS may be greatly effective as well as indications that may not see as great of a benefit, requiring further investigation into the utility of PNS for these cases. Diagnoses which saw no improvement in VAS pain score nor subjective percent improvement included lumbar neuropathy, diabetic polyneuropathy, intercostal neuralgia, and femoral cutaneous neuropathy. Additional diagnoses which saw no improvement in VAS pain scale, though some subjective improvement, include knee osteoarthritis, femoral artery injury, peroneal and tibial neuropathies, and neuroma of upper extremity amputation. Each of these diagnoses was limited in subject number; however, given the minimal to no benefit observed from PNS intervention in these patients, these diagnoses may represent areas of interest for future investigation. Additionally, without additional case reports of patients with these diagnoses who saw benefit from the intervention, it is difficult to broadly state that PNS is an effective intervention for all chronic pain conditions related to neuropathic pain. Notable distinguishment of the limitations regarding the patients who did not receive relief from the device is also necessary. Our study did not include the investigation and analysis of those who potentially developed increased pain on our pain scoring and percent improvement scale systems during the study period.

Several studies have demonstrated a significant improvement in pain scores post-PNS implantation with a growing variety of implant locations being evaluated. This includes PNS therapy for low back pain, lower extremities, headaches, and even phantom limb pain [16,17,18,19,20,21]. In a retrospective study of 72 patients who underwent PNS therapy, pain scores and opioid utilization to manage pain improved within 6 months after implantation [22]. In another PNS therapy study, which utilized randomization and sham controls, pain scores and opioid utilization were likewise found to be improved within 7 days of orthopedic surgery in the PNS therapy group [23]. While the results of these studies indicate growing evidence for the efficacy of PNS in treating chronic pain, there is still limited research regarding the efficacy of PNS at various nerve targets and how ubiquitous an intervention PNS is across various chronic pain syndromes.

Mechanism of Action of PNS

As mentioned in the Introduction, there have been two mechanisms—changes to local inflammatory mediators and adjustments of pain thresholds—proposed which function under the concept of the “gate control” theory of pain [7, 13, 14]. The influence of endogenous molecular signaling cascades such as serotonergic (5HT2 and 5HT3), GABAergic, and glycinergic are thought to be the responsible neurotransmitters for the pain pathways, which can be altered with PNS technology [24, 25]. The purpose of this paper does not explore the driving mechanisms of action behind PNS therapy and instead highlights the clinical success of these devices. The complexity of the nervous system physiology and the duality of peripheral and central nervous system fibers has generated much of the dissonance in the current literature regarding alternative theories for pain management mechanisms [25]. Continued experimental studies with animal and human models is necessary to further develop our understanding of these mechanisms.

Stimulation Waveforms of PNS

Future research endeavors regarding the protocol of PNS therapy are an aspect of this study that must be acknowledged. Personalization and adjustment regarding the titration of the electrical stimulation patients received could be better isolated and analyzed to assess the correlation to patient outcomes. Generally, patients received stimulation during lead placement to confirm comfortable sensations. This determined the waveform and parameters the patients received as treatment throughout the 60-day treatment. After placement of the leads, patients also have the ability to increase and decrease the intensity as needed throughout the stimulation period with a wireless remote to maintain comfortable stimulation coverage [26]. Intensity values correspond to amplitude and pulse duration range. These vary from 0.2 to 30 mA and 10 to 200 µs. The frequency is set to either 12 Hz for motor stimulation program (usually 6–12 h per day) or 100 Hz for the sensory stimulation program (usually up to 24 h per day). Our dual-lead placement allowed for variety regarding the received stimulation programming, with some patients receiving a sensory stimulation schedule through both leads and some patients receiving a sensory stimulation schedule through a single lead while receiving a motor stimulation schedule through the other implanted lead. Stimulation schedules were presented to the patients in a manner that utilized their discretion and the freedom to receive both programs. Both programs were reconciled with current published program designs [27, 28]. The 12-Hz program additionally permitted the adjustment of electrical intensity levels within an appropriate range that was deemed comfortable, cyclic activation of the multifidi for chronic low back pain [27]. Devices were additionally used at least 6 h per day, with a maximum of 12 h per day for 60 days [27]. The 100-Hz program demonstrated the deliverance of comfortable, stimulation-evoked sensations that could be locally applied for post-amputation pain [28]. Participants were permitted to adjust waveform and parameter values while maintaining the established test stimulation ranges set by the study staff during the 60-day treatment [28]. The necessity for future investigation pertaining to the frequency and details of the electrical parameters is clear in order to develop parameter programming that optimizes therapy and would assist in the overall advancement and efficacy of these PNS devices.

Adverse Events and Limitations

There were a total of five adverse events out of the 89 patients who underwent PNS therapy, all five of which resulted in no pain relief. Of the five adverse events to result in lack of pain improvement, four of the five patients struggled with lead migration, with two patients experiencing lead loosening to the point of lead dislodgement. This suggests that improvements can be made concerning the security of PNS devices at various locations, with necessary specialized adaptations being developed and adopted for the specific locations to improve therapy efficacy. With the mean duration of improvement lasting 123 days, these devices demonstrated an ability to provide pain relief after the removal of stimulation. For some patients, pain returned before the conclusion of the temporary PNS therapy at 60 days, prompting for further investigation to determine the reasoning for such drastic differences in duration of improvement, from 6 days post implant and therapy initiation to nearly 2 years of relief.

The potential complications seen in this study are consistent with previous studies that have demonstrated a beneficial role for PNS therapy in the management of pain. In a review looking at literature involving spinal cord, occipital, sacral, and peripheral nerve field stimulation, complications were reported to be between 30% and 40%, with most common complications involving hardware issues instead of biological complications [29]. The review reported that the most common hardware-related complication involved lead migration [29]. This retrospective study similarly had hardware issues occurring more frequently than biological complications, however, the complication rate was lower than that reported in previous studies. Other historically reported complications include lead failure or fracture, and biological complications such as infection and pain over the implant, and rarely neurological damage [29]. Complications are most associated with location of lead placement and the experience level of the surgeon [29]. Given that some complications are easily corrected by re-operation and that complications are often minor when they occur, PNS therapy remains an effective and safe therapy for chronic pain management.

Given the retrospective nature of this research, there are some inherent limitations to this work. One limitation of this work is the relatively small number of patients who each had a specific diagnosis prior to PNS therapy as well as the various stimulation parameters used for each patient which were not recorded as they were changing for each patient overtime with their ability to adjust their device settings. Overall, this study provides additional evidence for the promising role for PNS therapy in improving patient-reported pain levels in those suffering from chronic pain. The various target sites shed light on the potential use for temporary PNS therapy at various targeted nerves for numerous diagnoses. Limited research has been conducted on PNS therapy for many of the nerves targeted in this work. Further research including randomized control trials directly comparing PNS therapy to other treatment modalities should be conducted to evaluate the efficacy of PNS therapy for chronic pain refractory to conventional treatment measures versus other procedures. This future work can identify which pain diagnoses and targeted nerves are best treated by PNS and who the ideal patient candidate is for the procedure as well to identify potential improvement in the procedure to lessen complications such as lead migration. Furthermore, research assessing the predictive value of diagnostic nerve blocks pain relief could also potentially help identify which patients will experience the greatest pain relief with PNS therapy for their chronic pain.

Conclusions

In conclusion, our study suggests that PNS therapy may be an effective treatment for patients experiencing chronic pain that is refractory to conventional treatments. While PNS provided statistically significant improvements in pain to the general patient population investigated, there remain some diagnoses and nerve targets that may be better suited for PNS intervention than others. Further studies assessing temporary PNS therapy as treatment for chronic pain refractory to conventional treatment measures need to be conducted to help identify which target locations are best suited for this therapy. Furthermore, while our study did not find a correlation between pain relief and BMI or age, further demographics should be identified and assessed to help locate which patients are the ideal population for this therapy, such as the effectiveness of their diagnostic nerve blocks.