Introduction

Herpes zoster (HZ) is a neurocutaneous disease caused by the varicella-zoster virus (VZV), with 80% of patients experiencing pain. Zoster-associated pain (ZAP) can be divided into three phases based on duration: the acute phase occurs within one month of the onset of herpes; the subacute phase is from one to three months; and postherpetic neuralgia (PHN) is pain that lasts for more than three months1. Data show that about 10–20% of patients will develop PHN after skin lesion healing, with the risk being higher in the elderly2,3. Once PHN occurs, the disease may last for months or even years and can lead to serious complications, such as secondary bacterial infection, disseminated rash, encephalitis, and herpes zoster ophthalmicus4. Patients often experience anxiety, depression, and sleep disorders, causing a significant burden on their mental health, family, and social economy. Effective treatment of ZAP has become a major focus and challenge in clinical practice5.

The treatment of ZAP includes medications such as antivirals, tricyclic antidepressants, antiepileptic drugs, opioids, and local analgesics6,7. For refractory cases that do not respond to conventional drug treatment, invasive interventional treatments such as nerve block, dorsal root ganglion (DRG) pulsed radiofrequency (PRF), and spinal cord stimulation (SCS) are often required8,9,10. SCS, a widely used interventional therapy for chronic pain, has been proven to effectively alleviate neuropathic pain, including ZAP11. In SCS, electrodes are placed in the epidural space of the spinal canal at the segments corresponding to the herpes zoster skin lesion. These electrodes are connected to a pulse generator that generates and outputs a continuous current between the cathode and anode, covering the patient’s pain area12. However, for patients with severe pain, SCS may not be fully effective, and some patients continue to experience moderate or severe pain after discharge. A recent study showed that 41.7% of ZAP patients treated with SCS had less than 50% pain reduction at discharge13. Another review summarized the long-term follow-up of various types of SCS in the treatment of PHN, noting that 32.1% of patients still had unsatisfactory pain relief14. Addressing residual pain after SCS treatment of ZAP remains an important clinical challenge.

DRG PRF is a type of peripheral neuromodulation technology that uses radio waves to generate electric fields, producing brief electrical stimulation and intermittent alternating pulse currents at the tip of the radiofrequency needle to regulate neural activity and provide pain relief15,16. DRG PRF has been shown to be an effective option for chronic refractory neuropathic pain and has been successfully applied to treat various types of neuropathic pain, including ZAP17,18,19. To address the problem of inadequate pain relief, multimodal analgesia is often used in clinical practice. The generation mechanism of ZAP involves both central and peripheral levels20,21.However, there is no research report on the effectiveness of neuromodulation therapy for ZAP at these two levels. This study aims to analyze the efficacy and safety of this dual nerve stimulation therapy for ZAP and to recommend a more effective treatment strategy for the comprehensive management of ZAP in clinical practice.

Methods

Participants

This study followed the International Conference on Harmonisation of Good Clinical Practice guidelines and the Declaration of Helsinki (2008).With the approval of the Ethics Committee of the Third Xiangya Hospital of Central South University (No. 23863) to waive the requirement for signed informed consent, a retrospective medical record analysis was conducted on patients with ZAP who were hospitalized in the Department of Pain at The Third Xiangya Hospital, Central South University, from October 2019 to October 2023. Inclusion criteria were as follows: age over 18 years, meeting the diagnostic criteria for ZAP, lesions affecting the cervical, thoracic, or lumbar segments, inability to control pain with conventional treatments, and undergoing either SCS treatment or a combination of SCS and PRF treatment. Exclusion criteria included: insufficient medical records (lack of baseline data or follow-up records at 6 months post-treatment), experiencing other significant medical conditions during the follow-up period, and receiving other interventional therapies during the follow-up period.

Patients who received only SCS treatment were defined as the SCS group, while those who received both SCS and PRF treatments were classified as the SCS + PRF group. According to the course of the disease, acute pain was defined as pain occurring within 1 month, pain persisting for more than 3 months was considered as PHN, and subacute pain was defined as lasting between 1 and 3 months1.

Surgery

Spinal cord stimulation

The surgical procedure for SCS has been detailed previously22. Briefly, the target spinal segment for electrode implantation was determined based on the affected pain area, and the electrode position was confirmed by intraoperative fluoroscopy. The patient was placed in a prone position, and after local anesthesia, an epidural puncture was performed with a 14G Tuohy needle. The core was removed once the needle entered the epidural space, and an 8-contact lead (3873; Medtronic, Minneapolis, MN, USA) was inserted through the cannula. The lead was advanced under fluoroscopic anteroposterior view, and a sensory test was conducted to ensure that the electrical stimulation covered the patient’s pain area. Patients were asked to remain in bed for 2 days to avoid potential lead migration. Those with lead displacement and dislocation were excluded from the study. The stimulation frequency was set to 50 Hz, and the pulse width was 500 µs. The electrical stimulation voltage was adjusted according to the degree of pain. Stimulation leads were removed within 2 weeks after surgery to prevent infection.

DRG PRF

PRF treatment was performed about one week after SCS. As described23, for DRG PRF, the patient was placed prone on the operating table with a comfortable pillow under their chest. The needle was guided into the thoracic paraspinal space using B-scan ultrasound (Fujifilm Sonosite, Bothell, WA, USA). The needle tip was fine-tuned based on the ultrasound probe scan to the target segment. X-ray imaging confirmed that the needle tip was directly below the lateral border of the pedicle in the anteroposterior view and in the superior quadrant dorsal to the foramina in the lateral view (Fig. 1). The internal needle was replaced by a pulsed radiofrequency electrode, connected to a standard clinical specification radiofrequency generator (Beiqi, R-2000BA1, Beijing, China). The position of the needle tip was controlled by sensory and motor nerve stimulation before proceeding. DRG PRF treatment was set at 2 Hz (20 ms pulse width) three times for 240 s. Impedance was maintained at less than 300Ω throughout the procedure.

Fig. 1
figure 1

PRF cannulas were accurately located in the intervertebral foramens confirmed by the guidance of C-arm fluoroscopy. (a) Anterior-posterior. (b) lateral view.

Full size image

Clinical outcomes and follow-up

The primary data analyzed included changes in the intensity of patients’ pain, evaluated using the Visual Analog Scale (VAS), which ranges from 0 (“no pain”) to 10 (“the worst pain imaginable”). Preoperative information included age, gender, duration of disease, and baseline VAS score upon admission. Postoperative data included VAS scores at discharge, VAS scores 6 months after surgery, whether patients in the acute and subacute phases progressed to PHN, and the occurrence of treatment-related complications (including pneumothorax, bleeding, infection, nerve injury, and electrode displacement). During postoperative follow-up, a reduction in VAS score of 50% or more compared to the baseline score upon admission was defined as a responder13.

Statistical analysis

Prism 9.0 software (GraphPad, San Diego, CA, USA) was used for statistical analysis. Continuous variables were expressed as mean ± standard deviation. Two-way analysis of variance with repeated measures and post-hoc multiple pairwise comparison using Sidak’s test was employed to assess changes in pain scores between the two groups over time. Differences in response rates were compared using χ2 tests (including possible χ2-corrected tests and Fisher’s exact test). A P value of less than 0.05 was considered statistically significant.

Results

A total of 39 patient medical records were reviewed. Four patients had missing data for the 6-month postoperative follow-up, three patients lacked sufficient preoperative medical records, three patients underwent other interventional procedures, and one patient reported other significant medical conditions during follow-up. The medical records of these patients were excluded from the analysis. Finally, the medical records of 28 patients were analyzed. Thirteen patients received SCS treatment during hospitalization (SCS group), and the other fifteen patients received both SCS and PRF treatment (SCS + PRF group) (Fig. 2).

Fig. 2
figure 2

Flow diagram for patient selection.

Full size image

Demographics

The demographic and clinical characteristics of the enrolled patients are summarized in Table 1. The mean age of patients in the SCS group was 70.92 ± 9.74 years, while in the SCS + PRF group, it was 68.53 ± 6.45 years, with no statistically significant difference (P = 0.476). There were no significant differences in gender distribution, duration of illness, hospital stay, or electrode implantation between the two groups. Among the participants, 71.43% (20/28) had acute or subacute herpetic lesions with a disease duration of less than 3 months. Additionally, 8 patients were diagnosed with PHN. All enrolled patients had failed to achieve pain control with conventional treatments before undergoing neuromodulation therapy.

Table 1 General characteristics of patients.
Full size table

Long-term analgesic effect

The baseline VAS upon admission showed no significant difference between the SCS and SCS + PRF groups (P = 0.934). Following treatment, both groups exhibited significant pain relief at discharge, with VAS scores of 2.62 ± 1.33 and 2.60 ± 1.06 for the SCS and SCS + PRF groups, respectively, showing no significant statistical difference (P>0.99). At the 6-month follow-up, both groups demonstrated effective long-term pain relief. The VAS score of the SCS + PRF group (1.53 ± 1.06) was significantly lower than that of the SCS group (3.23 ± 1.50) (P < 0.001) (Table 1). Additionally, the response rates at discharge were 84.62% and 93.33% for the SCS and SCS + PRF groups, respectively, with no significant difference (P = 0.583) (Table 1). However, at the 6-month follow-up, the response rate in the SCS + PRF group remained at 93.33%, significantly higher than the 53.85% observed in the SCS group (P = 0.029) (Table 1). Furthermore, at the 6-month follow-up, 6.67% of the SCS + PRF group still had moderate pain (Fig. 3b), which was significantly lower than the 38.4% in the SCS group (Fig. 3a) (P = 0.041). These findings suggest the superior long-term efficacy of the dual-modulation therapy.

Fig. 3
figure 3

6 months postoperation, the proportion of patients with different degrees of pain in the two groups was compared. (a) 6 months postoperation, 38.46% of the patients in the SCS group had moderate pain. (b)6 months postoperation, 6.67% of the patients in the SCS + PRF group had moderate pain. χ2 test, P = 0.041.

Full size image

Prevention of PHN

We conducted efficacy analyses on patients in the acute and subacute phases (non-PHN) of ZAP and found no significant differences in preoperative and discharge VAS scores between the acute and subacute patient groups in both the SCS and SCS + PRF groups (all P > 0.05). However, at the 6-month follow-up, the VAS scores in the SCS + PRF group were significantly lower than those in the SCS group (P < 0.001) (Fig. 4a). Furthermore, at 6 months post-operation, 45.45% (5/11) of patients in the SCS group in the acute and subacute phases had VAS scores higher than 3, whereas no patients in the SCS + PRF group had VAS scores exceeding 3, indicating a significant difference (P = 0.038) (Fig. 4b). These findings suggest that central and peripheral dual modulation therapy may help prevent the development of PHN in patients in the acute and subacute phases of ZAP.

Fig. 4
figure 4

Analysis of the therapeutic effects of patients with acute and subacute in ZAP. (a) Changes of VAS scores at baseline, discharge and 6 months postoperative in Non-PHN patients.Two-way analysis of variance with repeated measures and post-hocmultiple pairwise comparison Sidak’s testing was used to assess the alteration of pain scores between two groups over time.***P<0.001. (b) Numbers of Non-PHN progressing to PHN at 6 months postoperative.χ2 test, P = 0.038.

Full size image

Safety

No complications related to the treatment regimens used (including pneumothorax, bleeding, infection, nerve injury, or stimulation lead migration) were observed during hospitalization or follow-up.

Discussion

Patients with ZAP often suffer from severe and persistent neuropathic pain, and there is no widely accepted optimal treatment in clinical practice. For refractory ZAP, clinicians aim to provide adequate pain relief, prevent recurrence, and enhance the patient’s quality of life. Drug therapy alone often has limited effectiveness and comes with various side effects24,25. Neuromodulation therapy plays an important role in managing ZAP when conventional therapies are ineffective26. Here, we present preliminary evidence suggesting that a dual neuromodulation approach, combining peripheral nerve stimulation with central stimulation, is an effective option for pain management in patients with ZAP.

We observed that the dual stimulation strategy of central and peripheral neuromodulation significantly relieved severe pain in patients with ZAP. At the final follow-up, patients in the SCS + PRF group showed a significant reduction in VAS scores and a notably lower proportion of residual moderate pain compared to the SCS group. This indicates that combining SCS with DRG PRF provides superior long-term pain relief. Moreover, among patients with subacute pain, the dual stimulation protocol effectively prevented the progression to PHN compared to single central stimulation alone. This finding aligns with current trends in pain management, suggesting that multimodal and multitarget therapeutic approaches often yield better outcomes27. The analgesic effects of SCS and DRG PRF are well-established; however, there exist significant differences in their mechanisms of action.Research on the analgesic mechanisms of SCS has predominantly focused on the spinal cord and supraspinal central nervous system levels. The earliest theory, proposed by Melzack and Wall in 1965, is the gate control theory, serves as the foundation for understanding how the central stimulation generated by SCS may function by selectively activating large, rapidly conducting fibers and influencing pain signal transmission at the spinal cord level28.As research progresses, it has been discovered that SCS can also modulate pain pathways, primarily by regulating the lateral pain upstream pathways and inducing analgesia through interference with the electrical and metabolic activities in the cingulate gyrus, lateral sensory thalamic nucleus, prefrontal cortex, and posterior central gyrus29.Furthermore, studies have also demonstrated that SCS exerts analgesic effects by enhancing the release of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) in the spinal cord30.Additionally, SCS inhibits pain by elevating the levels of 5-hydroxytryptamine receptors, acetylcholine, and norepinephrine in the spinal cord31,32. Beyond these mechanisms, the opioid system and the endogenous cannabinoid system are also integral to the analgesic actions of SCS33,34.Currently, the theories regarding the mechanism of DRG PRF primarily focus on the spinal cord and peripheral nerves.When PRF was first introduced, it was supposed to reduce pain by long-term depression(LTD) of pain signaling from the peripheral nerve to the central nervous system35. Furthermore, deactivation of microglia at the level of the spinal dorsal horn, reduction of proinflammatory cytokines, increased endogenous opioid precursor messenger ribonucleic acid, enhancement of noradrenergic and serotonergic descending pain inhibitory pathways, suppression of excitation of C-afferent fibers, and microscopic damage of nociceptive C- and A-delta fibers have been found to contribute to pain reduction after PRF application36,37,38,39,40.The multifaceted pathogenesis of ZAP involves sensitization of central and peripheral nerves and abnormal ion channel expression20,21. In our study, regarding the treatment of ZAP patients, within the SCS group, SCS primarily influenced the conduction of pain signals by broadly modulating spinal neural activity at the central level. In comparison to SCS alone, the dual stimulation strategy employed in the SCS + PRF group targeted pain signal processing at both the central and peripheral levels. This dual stimulation approach likely offers a more comprehensive coverage of pain signal transmission pathways, thereby enhancing therapeutic efficacy. This could potentially account for the superior long-term treatment outcomes observed in the SCS + PRF group and a reduced propensity of progressing to PHN.ZAP originates from VZV latent in peripheral ganglia, evolving from peripheral sensitization to central sensitization over time, with the two mechanisms influencing each other27,41,42. We describe this phenomenon of layered and integrated sensitization as the “layer-integrate” sensitization theory (LIST). The dual stimulation strategy of SCS combined with DRG PRF offers comprehensive pain management at both central and peripheral levels.

Neuromodulation therapy is a key component of multimodal analgesia for refractory neuropathic pain9. A review summarized that SCS combined with conventional medical management reduces chronic pain intensity, decreases the dose of analgesics needed, and improves long-term quality of life and physical function43. A recent study indicated that the efficacy of combined DRG and SCS stimulation in treating chronic focal neuropathic pain was 78.9%44. Ji et al. also reported that, after 6 months of PRF combined with nerve block for ZAP, only 16.67% of patients had an unsatisfactory prognosis45. These findings consistently suggest that integrated management may be a better approach for treating refractory neuropathic pain. Research on dual neuromodulation therapy for patients with ZAP is still extremely limited. Ma et al. found that a dual stimulation regimen of peripheral nerve electrical stimulation combined with PRF in the trigeminal ganglion achieved better clinical outcomes compared with peripheral nerve electrical stimulation alone in the management of herpes zoster ophthalmicus46. Additionally, Wang et al. reported the analgesic effect of dual neuromodulation in the dorsal root ganglia in a patient with extremely painful ZAP47. They performed dual nerve stimulation on a single level of the peripheral nervous system. Our study supported the efficacy of dual electrical stimulation in treating refractory ZAP at both the central and peripheral nervous system levels. This suggests that neuromodulation of both the peripheral and central nervous systems may have significant advantages in the comprehensive treatment of ZAP.

Our study has several limitations. Firstly, as a retrospective study, the data is sourced from existing medical records, which may introduce certain biases due to variations in data recording by different doctors. Secondly, the single-center nature of the study resulted in a small sample size. Finally, long-term follow-up results are not yet available. Despite these limitations, our results preliminarily demonstrate the excellent effect of this central and peripheral hierarchical and integrated neuromodulation for pain management. Therefore, it is necessary to conduct multi-center, large sample-size randomized controlled clinical trials to verify the efficacy of SCS combined with DRG PRF dual stimulation in the treatment of ZAP in the future. This will help explore the best neuromodulation scheme for treating ZAP.

Conclusions

SCS combined with DRG PRF is feasible and effective for treating ZAP. This new central and peripheral dual nerve stimulation strategy may be a more appropriate regimen for ZAP treatment.