Abstract
Introduction
Spinal cord stimulation (SCS) has emerged as a promising modality for managing zoster-associated pain (ZAP). This study aimed to evaluate the efficacy and safety of short-term spinal cord stimulation (SCS) for treating ZAP during the subacute phase.
Methods
Records of patients undergoing transforaminal epidural injection (TFEI), pulsed radiofrequency (PRF), and short-term SCS were retrospectively reviewed. A 1:2:2 propensity-score matching was performed. The primary endpoint was Numeric Rating Scale (NRS) pain scores at 3 months after rash onset. Secondary endpoints included postherpetic neuralgia (PHN) incidence, Hospital Anxiety and Depression (HADS) scores, Pittsburgh Sleep Quality Index (PSQI) scores, quality-adjusted life years (QALY) scores, and adverse events.
Results
At 3 months post-rash onset, mean NRS pain score was 1.59 ± 1.18 in SCS cohort, which was lower than that of 4.80 ± 1.41 in TFEI cohort and 4.54 ± 1.40 in PRF cohort, with the difference exceeding the prespecified minimal clinically important difference of three points (p < 0.001). PHN occurred in 8.9% of patients in SCS cohort at 3 months and 5.9% at 6 months following rash episode, which was lower than the other two cohorts (both p < 0.001). The proportion of patients requesting weak opioids for rescue pain relief was lower in SCS cohort, with less usages, at these time points. SCS cohort had better improvements in terms of HADS, PSQI and QALY scores over time compared to both TFEI and PRF cohorts (all p < 0.05). No severe adverse events were observed.
Conclusions
Among patients with subacute ZAP, short-term SCS was more effective than TFEI or PRF, resulting in greater improvements in PHN occurrence, psychological function, sleep quality, and quality of life.
Keywords: Zoster-associated pain, Postherpetic neuralgia, Spinal cord stimulation, Epidural injection, Pulsed radiofrequency
Key Summary Points
| Why carry out this study? |
| Zoster-associated pain (ZAP) can become chronic and debilitating, progressing to postherpetic neuralgia (PHN). |
| While both transforaminal epidural injection (TFEI) and pulsed radiofrequency (PRF) have been utilized for subacute ZAP, a need existed to directly compare their effectiveness with another promising modality, short-term spinal cord stimulation (SCS), in a real-world context. |
| This study aimed to investigate the comparative effectiveness of short-term SCS versus both TFEI and PRF for treating subacute ZAP using a propensity score-matched analysis. The hypothesis was that short-term SCS would demonstrate better effectiveness. |
| What was learned from the study? |
| Our results indicated that short-term SCS provided more effective pain relief and higher safety, leading to greater improvements in PHN incidence, psychological well-being, sleep quality, and overall quality of life compared to TFEI and PRF. |
| The study provided initial evidence that short-term SCS might be a more effective therapeutic strategy for subacute ZAP, potentially offering multi-dimensional benefits beyond pain control. |
| These positive findings, if confirmed by larger-scale studies, could influence future clinical treatment algorithms for managing subacute ZAP. |
Introduction
Herpes zoster (HZ) occurs when the dormant varicella-zoster virus in the dorsal root ganglion (DRG) is reactivated due to the failure of the immune system to control viral replication [1]. According to a nationwide population-based surgery in China in 2024, the adjusted annual incidence of HZ ranges was 5.85 per 1000 persons per year [2]. The acute eruptive phase is characterized by painful vesicles on the skin innervated by the affected ganglion. Although the rashes are self-limiting, the zoster-associated pain (ZAP) can progress to become chronic and debilitating, leading to the development of postherpetic neuralgia (PHN) [3]. It has been reported that PHN is associated with a heavy burden of illness in urban China, affecting daily activities, sleep, and quality of life [4]. The transition from ZAP to chronic pain states indicates the contribution of neuropathic events in the reactivated ganglion. Furthermore, the neural tissue damage can extend along the ascending sensory pathway to the dorsal horn of the spinal cord [5]. Based on the European consensus-based (S2k) guideline, antiviral therapy combined with analgesics serves as the first-line treatment for HZ, and peripheral nerve root block is effective for moderate-to-severe ZAP in the acute phase [6]. For disabling pain that lasts more than 30 days following rash onset, transforaminal epidural injection (TFEI), pulsed radiofrequency (PRF) on DRG and short-term spinal cord stimulation (SCS) have been widely utilized as promising modalities to prevent PHN in general clinical practice, supported by clinical evidence [7, 8]. Nevertheless, no studies have examined all three treatments simultaneously.
The hypothesis of this study is that the short-term SCS was more effective for the treatment of subacute herpes zoster-associated pain compared to both the transforaminal epidural injection (TFEI) and PRF on DRG under fluoroscopic guidance. It can also prevent the transition of subacute pain to PHN.
Methods
Study Design and Participant
The matched retrospective observational study was conducted at a large medical university-affiliated hospital. The ethical approval was obtained from the Ethics Committee of Capital Medical University-affiliated Beijing Xuanwu Hospital (xw-clincal-2026120) in accordance with the Declaration of Helsinki principles. Written informed consent was obtained from all patients. Results were reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for a retrospective case–control study [9].
Between January 1, 2021, and December 31, 2025, patients who underwent minimally invasive intervention for HZ were retrospectively reviewed from the electronic medical records (EMRs). The inclusion criteria were as follows: (1) meeting the diagnosis criteria of HZ according to the International Classification of Diseases 10th revision (ICD-10) codes [10]; (2) the course of disease within 30 to 90 days after rash onset; (3) pain resistant to antiviral therapy combined with analgesics; (4) having a Numeric Rating Scale (NRS) pain score ≥ 4 points; (5) aged > 18 years. Patients were excluded if they had bilateral HZ, active malignancies, other conditions using immunodeficiency, uncontrolled diabetes mellitus, coagulation dysfunction, severe systemic disorders, active psychiatry disease, alcohol addiction, opioid abuse, history of epidural injections within the past 3 months, pregnancy or lactation, and incomplete medical records.
After consulting with their pain specialists, patients chose the most suitable option among TFEI, PRF on DRG and short-term SCS, considering their preferences, values, and possible outcomes through a shared decision-making procress. To mitigate selection bias and unbalanced covariates, a 1:1:1 propensity score-matched (PSM) analysis was performed among three treatment modalities. Logit-transformed propensity scores were obtained by utilizing the nearest neighbor approach with a caliper of 0.1. Clinical variables including: age, gender, body mass index, affected side/location/dermatomes, hemorrhagic lesion, disease duration, baseline NRS pain score, analgesic history, and other treatments during acute phase, as well as comorbidities were selected as matching factors. A standardized mean difference (SMD) less than 0.1 confirmed balance.
Spinal Cord Stimulation Procedure
All procedures were performed by the same two pain specialists who had over 20 years of experience in minimally invasive pain interventions. Patients were placed in the prone position and monitored using electrocardiography, non-invasive blood pressure, and pulse oximetry in the operating room. A Tuohy needle was inserted into the epidural space under fluoroscopic guidance. An eight-contact stimulation electrode (Model 977A2, Medtronic, Minneapolis, MN, USA) was then implanted into the epidural space through the needle until its tip reached the targeted anatomical location at the dorsal column, with the aim of producing a comfortable paresthesia sensation that covered at least 50% of the painful area as reported by patients (Fig. 2). The short-term stimulation was administered for 14 days.
Fig. 2.
Placement of SCS. a AP image and b lateral image showing placement of one, eight-contact epidural electrode array with tips at the C2 vertebral level; c AP image and d lateral image showing placement of one eight-contact epidural electrode array with tips at the T4 vertebral level. SCS spinal cord stimulation, AP anterior-posterior
Rescue Analgesics
According to the World Health Organization (WHO) analgesic ladder, rescue analgesics were prescribed upon request after procedures. Celecoxib was administered at a dose of 200 mg per tablet, up to twice a day, if patients reported an NRS pain score ranging from 1 to 3 points. Tramadol hydrochloride was provided at a dose of 100 mg per tablet, up to twice daily, for patients with an NRS pain score ≥ 4 points [11].
Outcome Measurement
Patients’ demographic and clinical characteristics were retrieved from the EMRs. Follow-up data were collected from medical records by two specially trained investigators. Intensity of ZAP was measured using the NRS with an 11-point scale ranging from 0 (no pain) to 10 (worst pain) [12]. The Chinese version of the Douleur Neuropathique 4 (DN4) questionnaire was utilized to identify neuropathic pain (NeP). This instrument included ten items: seven sensory descriptors, including burning, painful cold, electric shock, tingling, pins and needles, numbness and itching, as well as three signs assessed through a sensory physical examination, including tactile hypesthesia, pinprick hypesthesia, and allodynia in the affected area. Each positive response was scored as 1 and each negative response as 0, yielding a total score between 0 and 10. A score of ≥ 4 points was considered, as the presence of NeP [13]. PHN was defined as a ZAP pain score ≥ 3 points on the NRS, persisting for at least 90 days after rash onset [14]. Patients’ anxiety and depression levels over the past week were evaluated by the self-reported Hospital Anxiety and Depression (HADS) score, which included two subscales: anxiety (HADS-A) and depression (HADS-D). Each subscale contained seven items, and responses were rated on a four-point scale ranging from 0 (do not apply to me at all) to 3 (applied to me very much). The total score for each subscale ranged from 0 to 21, with higher scores indicating more severe symptoms. Patients were categorized as follows: no significant symptoms (0–7 points), mild symptoms (8–10 points), moderate symptoms (11–14 points) and severe symptoms (15–21 points) [15]. Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI), which included 18 items across seven dimensions: subjective sleep quality, sleep latency, sleep duration, sleep efficiency, sleep disturbances, use of medication and daytime dysfunction. Each item was scored on a four-point Likert scale between 0 and 3, with the total score ranging from 0 to 21. Higher scores indicated worse sleep quality [16]. The five-dimension EuroQoL (EQ-5D) 3L index score was used to measure patients’ health-related quality of life (HR-QoL). Responses to different hypothetical choices were converted into an index score on a scale from 0 (representing death) to 1 (indicating perfect health). This index score was then multiplied by the duration spent in the health state to calculate a quality-adjusted life years (QALY) metric, yielding a total score ranging from − 0.594 to 1 [17].
The primary endpoint was the NRS pain scores at 3 months after rash onset among different cohorts. Secondary endpoints included PHN incidence, HADS scores, PSQI scores, QALY scores and adverse events.
Sample Size Calculation
PASS software, version 22 (NCSS, LLC., Kaysville, UT, USA) was utilized for sample size calculation. The present study was designed to compare the means of three cohorts using the Tukey–Kramer pairwise multiple comparison test with a significance level of 0.05. Previous research indicated that the minimal clinically important difference (MCID) for acute pain management was three points in patients with HZ [18, 19]. The researchers wanted a sample size large enough to detect a minimum three-point difference in NRS scores at 1 month following treatments, with a 1:2:2 ratio. To achieve a 90% power, a total of 230 patients were required, assuming a standard deviation (SD) for NRS pain scores was between 1 and 4. Considering a potential 20% data loss, the final sample size was set at 288 cases (58 for ESI cohort, and 115 for both PRF and SCS cohorts).
Statistical Analysis
Statistical analysis was performed using SPSS software, version 22.0 (SPSS Inc., Chicago, IL, USA) with a significance level set at p < 0.05. The Kolmogorov–Smirnov Z-test was applied to evaluate the normality of the data. Continuous data with a normal distribution were described as mean ± SD and percentages, while categorical data were presented as percentages. Differences among cohorts were compared using one-way analysis of variance (ANOVA) followed by the post hoc Bonferroni correction (adjusted Bonferroni p = 0.017) for continuous data and chi-square test for categorial data, respectively.
Results
Figure 1 shows a flowchart of the entire study cohort. The PSM produced a matched sample of 51 patients in TFEI cohort, 99 in PRF cohort, and 99 in SCS cohort for an outcome comparison. As shown in Table 1, the baseline demographic characteristics and clinical variables were well balanced following PSM (all SMD < 0.1) (Fig. 2).
Fig. 1.
Flowchart of the study cohorts. HZ herpes zoster, TFEI transforaminal epidural injection, PRF pulsed radiofrequency, SCS spinal cord stimulation, PSM propensity score matching
Table 1.
Baseline characteristics of patients with zoster-associated pain among three cohorts
| Variables | Before PSM | SMDmax | After PSM | SMDmax | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ESI (n = 89) | PRF (n = 221) | SCS (n = 115) | P | ESI (N = 51) | PRF (N = 99) | SCS (N = 99) | P | |||
| Age ± SD (years) | 55.98 ± 13.24 | 58.12 ± 14.28 | 54.82 ± 13.19 | 0.095 | 0.240 | 55.97 ± 13.23 | 56.06 ± 13.88 | 55.77 ± 13.35 | 0.988 | 0.021 |
| Sex, n (%) | 0.793 | 0.981 | ||||||||
| Female | 48 (53.9%) | 122 (55.2%) | 59 (51.3%) | 0.078 | 26 (51.0%) | 52 (52.5%) | 51 (51.5%) | – 0.030 | ||
| Male | 41 (46.1%) | 99 (44.8%) | 56 (48.7%) | − 0.078 | 25 (49.0%) | 47 (47.5%) | 48 (48.5%) | 0.030 | ||
| Body mass index ± SD (kg/m2) | 23.46 ± 3.78 | 22.60 ± 2.63 | 23.30 ± 3.33 | 0.264 | 0.036 | 23.35 ± 2.56 | 23.40 ± 2.78 | 23.26 ± 2.63 | – 0.056 | |
| Affected side, n (%) | 0.693 | 0.959 | ||||||||
| Left | 37 (41.6%) | 101 (45.7%) | 56 (48.7%) | − 0.143 | 23 (45.1%) | 46 (46.5%) | 44 (44.4%) | 0.042 | ||
| Right | 52 (58.4%) | 120 (54.3%) | 59 (51.3%) | 0.143 | 28 (54.9%) | 53 (53.5%) | 55 (55.6%) | – 0.042 | ||
| Location, n (%) | 0.056 | 0.979 | ||||||||
| Cervical nerve | 14 (15.7%) | 46 (20.8%) | 20 (17.4%) | 0.132 | 9 (18.0%) | 17 (17.2%) | 18 (18.2%) | – 0.026 | ||
| Thoracic nerve | 33 (37.1%) | 99 (44.8%) | 62 (53.9%) | − 0.342 | 31 (62.0%) | 58 (58.6%) | 60 (60.6%) | 0.069 | ||
| Lumbar-sacral nerve | 42 (47.2%) | 76 (34.4%) | 33 (28.7%) | 0.388 | 10 (20.0%) | 24 (24.2%) | 21 (21.2%) | 0.072 | ||
| Affected dermatomes, n (%) | 0.042 | 0.995 | ||||||||
| 1 level | 11 (12.4%) | 35 (15.8%) | 26 (22.6%) | − 0.271 | 10 (19.6%) | 21 (21.2%) | 20 (20.2%) | – 0.040 | ||
| 2 levels | 46 (51.7%) | 132 (59.7%) | 67 (58.3%) | − 0.162 | 31 (60.8%) | 60 (60.6%) | 62 (62.6%) | – 0.041 | ||
| ≥ 3 levels | 32 (36.0%) | 54 (24.4%) | 22 (19.1%) | 0.385 | 10 (19.6%) | 18 (18.2%) | 17 (17.2%) | 0.036 | ||
| Hemorrhagic lesion, n (%) | 0.683 | 0.938 | ||||||||
| No | 74 (83.1%) | 177 (80.1%) | 90 (78.3%) | 0.121 | 40 (78.4%) | 78 (78.8%) | 76 (76.8%) | 0.048 | ||
| Yes | 15 (16.9%) | 44 (19.9%) | 25 (21.7%) | − 0.121 | 11 (21.6%) | 21 (21.2%) | 23 (32.2%) | – 0.048 | ||
| Rash onset duration ± SD (days) | 40.03 ± 11.90 | 40.89 ± 12.27 | 43.18 ± 12.34 | 0.259 | 0.062 | 41.02 ± 11.35 | 40.98 ± 11.21 | 41.15 ± 11.64 | 0.994 | 0.018 |
| NRS pain score ± SD | 7.78 ± 1.93 | 7.36 ± 1.75 | 8.02 ± 1.77 | 0.375 | 0.125 | 7.86 ± 1.99 | 7.92 ± 1.97 | 7.90 ± 1.88 | 0.984 | 0.021 |
| Presence of neuropathic pain | 52 (58.4%) | 114 (51.6%) | 71 (61.7%) | 0.175 | − 0.205 | 31 (60.8%) | 59 (59.6%) | 60 (60.6%) | 0.986 | 0.025 |
| Compliance with antivirals, n (%) | 0.560 | 0.954 | ||||||||
| No | 8 (9.1%) | 25 (11.3%) | 16 (13.9%) | − 0.151 | 4 (7.8%) | 9 (9.1%) | 8 (8.1%) | − 0.047 | ||
| Yes | 80 (90.9%) | 196 (88.7%) | 99 (86.1%) | 0.151 | 47 (92.2%) | 90 (90.9%) | 91 (91.9%) | 0.047 | ||
| History of analgesics, n (%) | 0.046 | 0.998 | ||||||||
| None | 23 (25.8%) | 59 (26.7%) | 29 (25.2%) | 0.034 | 13 (25.5%) | 25 (25.3%) | 26 (26.3%) | − 0.023 | ||
| NSAID | 27 (30.3%) | 73 (33.0%) | 34 (29.6%) | − 0.174 | 14 (27.5%) | 27 (27.3%) | 30 (30.3%) | − 0.066 | ||
| Anti-epileptic or week opioid | 30 (33.7%) | 77 (34.8%) | 32 (27.8%) | 0.151 | 18 (35.3%) | 34 (34.3%) | 31 (31.3%) | 0.084 | ||
| Others | 9 (10.1%) | 12 (5.4%) | 20 (17.4%) | − 0.385 | 6 (11.8%) | 13 (13.1%) | 12 (12.1%) | − 0.039 | ||
| History of other interventions, n (%) | 0.451 | 0.996 | ||||||||
| None | 45 (50.6%) | 104 (47.1%) | 44 (38.3%) | 0.249 | 19 (37.3%) | 39 (39.4%) | 40 (40.4%) | − 0.063 | ||
| Physical therapy or acupuncture | 18 (20.2%) | 37 (16.7%) | 28 (24.3%) | − 0.189 | 13 (25.5%) | 25 (25.3%) | 23 (23.2%) | 0.054 | ||
| Nerve block | 20 (22.5%) | 56 (25.3%) | 31 (27.0%) | − 0.104 | 15 (29.4%) | 26 (26.3%) | 28 (28.3%) | 0.069 | ||
| Others | 6 (6.7%) | 24 (10.9%) | 12 (10.4%) | − 0.149 | 4 (7.8%) | 9 (9.1%) | 8 (8.1%) | − 0.047 | ||
| Comorbidities, n (%) | ||||||||||
| Hypertension | 18 (20.2%) | 50 (22.6%) | 20 (17.4%) | 0.528 | 0.130 | 10 (19.6%) | 17 (17.2%) | 18 (18.2%) | 0.934 | 0.062 |
| Diabetes mellitus | 9 (10.1%) | 49 (24.4%) | 19 (16.5%) | 0.039 | − 0.385 | 6 (11.8%) | 12 (13.5%) | 11 (12.4%) | 0.952 | − 0.051 |
| Cardiovascular disease | 11 (12.4%) | 33 (14.9%) | 15 (13.0%) | 0.801 | − 0.073 | 7 (13.7%) | 13 (13.1%) | 12 (12.4%) | 0.971 | 0.039 |
1TFEI transforaminal epidural injection, PRF pulsed radiofrequency, SCS spinal cord stimulation, PSM propensity score matching, SD standard deviation, SMD standardized mean difference, NRS Numeric Rating Scale, NSAID non-steroidal anti-inflammatory drugs
2SMDmax referred to the maximum value obtained from all pairwise comparisons, serving as the overall balance indicator when calculating the SMD separately for each pair among the three cohorts
Following treatments, patients across all three cohorts showed significant pain relief at both 3 and 6 months after rash onset compared to their baseline values (all adjusted p < 0.05). According to ANOVA analysis, patients who underwent SCS for ZAP during the subacute phase had a significantly lower mean NRS pain score than those treated with TFEI or PRF at 3 months after rash onset (1.59 ± 1.18 vs. 4.80 ± 1.41 vs. 4.54 ± 1.40, p < 0.001), with the difference exceeding the prespecified MCID of three points. At 6 months post-rash onset, the mean NRS pain score was 1.23 ± 0.96 in the SCS cohort, which was also significantly lower than that of 3.90 ± 1.39 in the TFEI cohort and that of 3.81 ± 1.48 in the PRF cohort (p < 0.001). However, no significant differences between the TEFI or PRF cohorts were observed over time (all adjusted Bonferroni p > 0.05). PHN occurred in 8.9% of patients in the SCS cohort at 3 months following the rash episode, which was significantly lower than that of 25.5% in the TFEI cohort and that of 23.7% in the PRF cohort (p < 0.001). Furthermore, the SCS cohort continued to have a lower PHN incidence at 6 months following rash onset compared to the TFEI and PRF cohorts (5.9% vs. 13.7% vs. 15.5%, p < 0.001) (Fig. 3).
Fig. 3.
Comparison of pain relief during the 6 months following rash onset. a Changes in NRS pain scores among three cohorts; b proportion of patients reporting PHN with NRS pain scores > 3 points. * adjusted Bonferroni p < 0.05 compared to the class 3 cohort; ^ adjusted Bonferroni p < 0.05 compared to the baseline value. NRS Numeric Rating Scale, PHN postherpetic neuralgia
As shown in Table 2, the proportion of patients requesting tramadol hydrochloride for rescue pain relief was significantly lower in the SCS cohort compared to the other two cohorts at both 3 and 6 months (both p < 0.05). Additionally, the mean usage of both celecoxib and tramadol hydrochloride was significantly lower in the SCS cohort at these time points (all p < 0.05). However, there was no statistically significant difference in usages of both rescue analgesics between the TFEI and PRF cohorts (all adjusted Bonferroni p > 0.05).
Table 2.
Comparison of rescue analgesics over time among three cohorts
| Outcomes | TFEI cohort (n = 51) | PRF cohort (n = 99) | SCS cohort (n = 99) | F/χ2 value | p | ||
|---|---|---|---|---|---|---|---|
| Patients taking rescue analgesics, n (%) | Celecoxib | 3 months | 7 (13.7%) | 14 (14.1%) | 10 (10.1%) | 0.837 | 0.658 |
| 6 months | 5 (9.8%) | 9 (9.1%) | 5 (5.1%) | 1.576 | 0.455 | ||
| Tramadol hydrochloride | 3 months | 11 (21.6%) | 23 (23.2%) | 9 (9.1%) | 7.759 | 0.021 | |
| 6 months | 9 (17.6%) | 18 (20.2%) | 6 (6.1%) | 8.704 | 0.013 | ||
| Mean daily dosage, mg (mean ± SD) | Celecoxib | 3 months | 283.51 ± 83.79 | 282.96 ± 92.21 | 251.57 ± 84.61 | 3.879 | 0.022 |
| 6 months | 310.72 ± 95.47 | 314.29 ± 96.90 | 270.38 ± 93.67 | 6.004 | 0.002 | ||
| Tramadol hydrochloride | 3 months | 159.34 ± 55.75 | 156.52 ± 62.15 | 108.46 ± 41.62 | 25.095 | < 0.001 | |
| 6 months | 163.11 ± 43.73 | 161.25 ± 52.17 | 116.25 ± 50.73 | 24.879 | < 0.001 |
TFEI transforaminal epidural injection, PRF pulsed radiofrequency, SCS spinal cord stimulation
At 3 months after onset, the mean HADS-A and HADS-D scores in the SCS cohort decreased to 8.18 ± 1.39 and 7.73 ± 3.94, respectively, which were significantly lower than those in the TFEI cohort (11.16 ± 2.66 for HADS-A scores and 9.93 ± 4.41 for HADS-D scores) and the PRF cohort (11.14 ± 2.67 for HADS-A scores and 9.57 ± 4.13 for HADS-D scores) (both p < 0.001). However, the difference between the TFEI and PRF cohorts was not statistically significant (adjusted Bonferroni p = 0.983 for HADS-A scores and 0.905 for HADS-D scores). Similar results were observed in HADS scores at 6 months after rash onset among the three cohorts, with HADS-A scores of 5.13 ± 1.95 vs. 8.56 ± 3.14 vs. 8.65 ± 3.84 (p < 0.001) and HADS-D scores of 5.20 ± 3.38 vs. 8.93 ± 4.28 vs. 8.86 ± 4.80 (p < 0.001). At 3 months post-rash onset, patients in the SCS cohort reported better sleep quality, with a mean PSQI score of 8.59 ± 1.13, which was significantly lower than the scores in the TFEI cohort (12.07 ± 4.23) and the PRF cohort (11.63 ± 3.18) (p = 0.005). This significant difference in mean PSQI scores persisted at 6 months after rash onset, with the SCS cohort scoring 5.81 ± 2.01 compared to 8.69 ± 1.08 in the TFEI cohort and 9.26 ± 1.73 in the PRF cohort (p < 0.001). Regarding HR-QoL, all cohorts showed significant improvements in QALY scores over time following treatment. Nevertheless, the SCS cohort had significantly higher mean QALY scores at both 3 months (0.61 ± 0.10 vs. 0.41 ± 0.20 vs. 0.43 ± 0.21, p = 0.025) and 6 months (0.86 ± 0.16 vs. 0.67 ± 0.27 vs. 0.61 ± 0.20, p = 0.20) after rash onset compared to the other two cohorts. No significant differences were observed between the TFEI and PRF cohorts at these time points (all adjusted Bonferroni p > 0.05) (Fig. 4).
Fig. 4.
Changes in HADS scores a, PSQI scores b, and QALY scores c over time among three cohorts. * adjusted Bonferroni p < 0.05 compared to the class 3 cohort; ^ adjusted Bonferroni p < 0.05 compared to the baseline value. HADS Hospital Anxiety and Depression, PSQI Pittsburgh Sleep Quality Index, QALY quality-adjusted life years
No severe adverse events such as spinal cord injury, nerve root damage, spinal anesthesia, hypophonia and dysphagia, epidural hematoma, unintentional durotomy, surgical site infection, intravascular injection, and local anesthetic toxicity occurred in any of the three cohorts. Minor side effects including pain during needling, paresthesia upon sensory stimulation, nausea, facial flushing, and temporary dizziness were reported in 11.8% of cases in the TFEI cohort, 10.1% in the PRF cohort, and 8.0% in the SCS cohort (p = 0.754). Only one patient in the SCS cohort reported a transient micturition problem, which resolved within 60 min.
Discussion
With the utilization of a propensity score-matched (PSM) analysis, the present study is the first to investigate the effectiveness of the short-term SCS for the treatment of ZAP during the subacute phase as compared to both the TFEI and PRF methods. It proved to be more effective than the other two treatments and mediated significantly better improvements in outcomes related to PHN incidence, psychological function, sleep quality and patients’ HR-QoL.
Following the implantation of the SCS electrode, the pulse generator delivered stimulations with various parameters such as tonic stimulation, high-frequency stimulation, high-frequency burst stimulation and dorsal root ganglion stimulation [20]. Then, these stimulations suppressed the segmental ascending nociceptive transmission by the release of neurotransmitter including gamma aminobutyric acid (GABA), endocannabinoids, and endogenous opioids in the spinal dorsal horn. Meanwhile, the stimulations facilitated the descending inhibition to promote the release of noradrenaline and serotonin, which modulated the wide dynamic range in the spinal cord. This process also played an important role in the antinociceptive mechanism of SCS [21]. Furthermore, SCS downregulated the levels of pro-inflammatory mediators such as IL-1β, TNF-α and NF-κB and inhibited sustained activation of microglia and astrocytes in the dorsal horn, leading to improvements in the neuroimmune condition [22].
Although SCS is usually implanted permanently for treating chronic neuropathic pain such as PHN, complex regional pain syndrome, and failed surgical pain syndrome, the temporary SCS has been widely used in patients who presented with acute and subacute ZAP that failed to conventional therapy [23, 24]. Dao-Song Dong et al. evaluated the efficacy of short-term SCS in 46 patients with refractory acute/subacute ZAP. Their results showed that the ten-point visual analog scale pain scores significantly decreased from the baseline value of 7.28 ± 0.93 to 3.59 ± 2.40 just before being discharged, with 69.6% of patients obtaining a favorable MCID of three points. At 6 months post-treatment, a total of 67.4% of cases achieved excellent pain relief with VAS pain scores ≤ 2 (p < 0.01) [18]. A recent study retrospectively investigated the efficacy of temporary SCS treatment and prevention of PHN among patients with ZAP during acute (n = 42), subacute (n = 34), and chronic phases (n = 23). They found that patients with herpetic pain in both the acute phase and subacute phase were clinically and statistically improved in ten-point VAS pain scores compared to those in chronic phase (p < 0.001). The incidence of PHN (VAS pain score ≥ 3) was observed in 2.38% in acute cohort, 16.0% in the subacute cohort and 68.42% in the chronic cohort at the 6-month follow-up [25]. Consistent with prior evidence, our results demonstrated that the mean NRS pain scores significantly declined over time, with the reductions exceeding the prespecified MCID of three points compared to the baseline values in patients undergoing short-term SCS to alleviate subacute herpetic pain. PHN defined as NRS pain score ≥ 3 occurred in 5.9% of patients at 6 months after rash onset, indicating that this method effectively prevented the development of PHN.
Epidural administration of local anesthetic and steroid into the affected DRG has been reported to be effectively block the nociceptive signals to the spinal dorsal horn in cases of ZAP beyond the acute phase. It may also help prevent further progression to PHN [26]. PRF utilizes radiofrequency waves to generate an electric field and produce intermittent pulsed currents at the electrode tip to control electric stimulation. This effect influences cytokine expression, cellular responses and gene regulation pathways to provide relief for ZAP [27]. According to a recent systematic review and meta-analysis involving 6 trials with 509 patients, patients undergoing short-term SCS to alleviate ZAP showed a better analgesic effect with lower pian intensity than cases receiving PRF (standardized mean difference (MD) = − 0.83, P = 0.002) [28]. A more recent study retrospectively evaluated the comparative efficacy of short-term SCS and PRF therapy in 275 patients with thoracic ZAP in the acute phase, subacute phase, and postherpetic phase. Their findings suggested that the SCS group demonstrated superior pain relief at 3, 6, and 12 months post-treatment compared to the PRF group, particularly in the subacute and postherpetic phases (p < 0.05). Aligned with previous findings, our results showed that patients in the SCS cohort exhibited lower NRS pain scores at all follow-up time points compared to those in both the TEFI and PRF cohorts, indicating that the short-term SCS provided more effective and sustained pain relief.
In addition to analgesic effects, SCS of the dorsal column can activate a variety of supraspinal brain regions, such as thalamus, insular cortex, limbic network, and hippocampus, which are associated with emotion. As a result, can alleviate not only pain but also anxiety and depression [21]. Evidence from recent research indicated that both short-term SCS and PRF for treating thoracic ZAP significantly reduced HADS-A and HADS-D scores across all time points following treatments (p < 0.05), with SCS method demonstrating superior long-term improvement in depression scores (p < 0.05) [29]. The temporary SCS had better effects in improving sleep quality and quality of life as compared to the conventional nerve block therapy in 135 patients with ZAP [30]. Additionally, patients receiving short-term SCS for ZAP also reported less incidence of adverse events (risk ratio = 0.32, p < 0.05) compared to the PRF modality [28]. In line with earlier findings, our results indicated that at 3 and 6 months after the rash episode, the SCS cohort reported lower HADS scores and PSQI scores, as well as fewer adverse events, resulting in greater improves in QALY scores compared to both the other cohorts.
There were several limitations. Firstly, as a retrospective observational study, our findings were subject to potential selection bias. Secondly, although propensity score matching was employed to balance baseline characteristics among cohorts, it cannot account for unmeasured or unknown confounders that were not captured in the dataset. Thirdly, the single-center design limited its external validity. The patient population and clinical practices at our institution might differ from those in other settings, restricting the generalizability of our findings to broader or different demographic cohorts. Lastly, the retrospective design precluded the establishment of causal relationships, limiting our conclusions to associations rather than definitive cause-and-effect outcomes. Therefore, our findings should be validated by a well-designed randomized controlled trial in the future.
Conclusions
In conclusion, the short-term SCS provided more effective pain relief for patients with subacute ZAP than both TFEI and PRF methods. It also resulted in a lower incidence of PHN and fewer adverse events, along with greater improvements in psychological well-being, sleep quality, and overall quality of life.
Acknowledgements
We thank the participants of the study.
Author Contributions
Liqiang Yang was involved in the conceptualization, investigation, methodology, project administration, supervisor, resources, and writing review and editing. Wenxing Zhao was involved in the conceptualization, investigation, methodology, project administration, formal analysis, visualization, validation, and writing original draft. Liangliang was involved in the conceptualization, investigation, methodology, project administration, data collection and software.
Funding
The study was funded by Beijing Municipal Administration of Hospitalbr/Incubating Program (PX2024035). The Rapid Service Fee was funded by the authors.
Data Availability
The data are available from the corresponding author upon reasonable request.
Declarations
Conflicts of Interest
Liqiang Yang, Wenxing Zhao and Liangliang He have nothing to disclose.
Ethical Approval
The ethical approval was obtained from the Ethics Committee of Capital Medical University-affiliated Beijing Xuanwu Hospital (xw-clincal-2026120) based on the principles of the Declaration of Helsinki. Written informed consent was obtained from all patients.
References
- 1.Nair PA, Patel BC. Herpes Zoster. StatPearls. Treasure Island (FL); 2026.
- 2.Xia Y, Ye X, Zhu W, Ai J, Shen Y, Shi Z, et al. Epidemiology of herpes zoster and post-herpetic neuralgia in China: a nationwide population-based survey. Int J Infect Dis. 2025;159:108005. [DOI] [PubMed] [Google Scholar]
- 3.Warner BE, Goins WF, Kramer PR, Kinchington PR. A guide to preclinical models of zoster-associated pain and postherpetic neuralgia. Curr Top Microbiol Immunol. 2023;438:189–221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yu SY, Fan BF, Yang F, DiBonaventura M, Chen YX, Li RY, et al. Patient and economic burdens of postherpetic neuralgia in China. Clinicoecon Outcomes Res. 2019;11:539–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Opstelten W, McElhaney J, Weinberger B, Oaklander AL, Johnson RW. The impact of varicella zoster virus: chronic pain. J Clin Virol. 2010;48(Suppl 1):S8-13. [DOI] [PubMed] [Google Scholar]
- 6.Gross GE, Eisert L, Doerr HW, Fickenscher H, Knuf M, Maier P, et al. S2k guidelines for the diagnosis and treatment of herpes zoster and postherpetic neuralgia. J Dtsch Dermatol Ges. 2020;18(1):55–78. [DOI] [PubMed] [Google Scholar]
- 7.Sabet H, Abbas A, Ho JS, El-Moslemani M, Essam Abo-Elnour D, Aldehri M, et al. Spinal cord stimulation, peripheral nerve stimulation, and pulsed radiofrequency in the management of herpes zoster-related pain: a network meta-analysis. Pain Med. 2026. 10.1093/pm/pnag049. [DOI] [PubMed] [Google Scholar]
- 8.Kang HY, Lee CH, Kim DH, Kim YD, Kim WJ, Kim E, et al. Clinical practice guidelines for the management of refractory postherpetic neuralgia by the Korean Pain Society. Korean J Pain. 2026;39(2):171–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Int J Surg. 2014;12(12):1500–24. [DOI] [PubMed] [Google Scholar]
- 10.Steindel SJ. International classification of diseases, 10th edition, clinical modification and procedure coding system: descriptive overview of the next generation HIPAA code sets. J Am Med Inform Assoc. 2010;17(3):274–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Anekar AA, Hendrix JM, Cascella M. WHO analgesic ladder. StatPearls. Treasure Island (FL); 2026. [PubMed]
- 12.Ferreira-Valente MA, Pais-Ribeiro JL, Jensen MP. Validity of four pain intensity rating scales. Pain. 2011;152(10):2399–404. [DOI] [PubMed] [Google Scholar]
- 13.Wang YF, Yang CC, Ro LS, Tsai YC, Lin KP, Sun WZ, et al. Development and validation of a Taiwan version of the DN4-T questionnaire. J Chin Med Assoc. 2019;82(8):623–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Johnson RW, Rice AS. Clinical practice. Postherpetic neuralgia. N Engl J Med. 2014;371(16):1526–33. [DOI] [PubMed] [Google Scholar]
- 15.Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res. 2002;52(2):69–77. [DOI] [PubMed] [Google Scholar]
- 16.Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213. [DOI] [PubMed] [Google Scholar]
- 17.EuroQol G. EuroQol–a new facility for the measurement of health-related quality of life. Health Policy. 1990;16(3):199–208. [DOI] [PubMed] [Google Scholar]
- 18.Dong DS, Yu X, Wan CF, Liu Y, Zhao L, Xi Q, et al. Efficacy of short-term spinal cord stimulation in acute/subacute zoster-related pain: a retrospective study. Pain Phys. 2017;20(5):E633–45. [PubMed] [Google Scholar]
- 19.Kinfe TM, Pintea B, Vatter H. Is spinal cord stimulation useful and safe for the treatment of chronic pain of ischemic origin? A review. Clin J Pain. 2016;32(1):7–13. [DOI] [PubMed] [Google Scholar]
- 20.Joosten EA, Franken G. Spinal cord stimulation in chronic neuropathic pain: mechanisms of action, new locations, new paradigms. Pain 2020;161 Suppl 1(1):S104–S13. [DOI] [PMC free article] [PubMed]
- 21.Sun L, Peng C, Joosten E, Cheung CW, Tan F, Jiang W, et al. Spinal cord stimulation and treatment of peripheral or central neuropathic pain: mechanisms and clinical application. Neural Plast. 2021. 10.1155/2021/5607898. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Fang JY, Yamamoto H, Romman A, Koutrouvelis AP, Yamamoto S. Spinal mechanisms of pain modulation by spinal cord stimulation: a systematic review. Cureus. 2025;17(6):e85567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Iseki M, Morita Y, Nakamura Y, Ifuku M, Komatsu S. Efficacy of limited-duration spinal cord stimulation for subacute postherpetic neuralgia. Ann Acad Med Singapore. 2009;38(11):1004–6. [PubMed] [Google Scholar]
- 24.Wang X, Gao Y, Yu J, Yang W, Zhang W, Huo J. The efficacy and prognosis analysis of short-term spinal cord stimulation in the treatment of zoster-associated pain: a retrospective study. Front Neurol. 2025;16:1611447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Huang J, Yang S, Yang J, Sun W, Jiang C, Zhou J, et al. Early treatment with temporary spinal cord stimulation effectively prevents development of postherpetic neuralgia. Pain Phys. 2020;23(2):E219–30. [PubMed] [Google Scholar]
- 26.Nahm FS, Choi E, Han WK, Lee HJ, Gil HY, Kim JH, et al. Transforaminal epidural steroid injection for zoster-related pain: the golden period for the best outcome. Pain Phys. 2021;24(5):E669–76. [PubMed] [Google Scholar]
- 27.Rui M, Han Z, Xu L, Yao M. Effect of CT-guided repeated pulsed radiofrequency on controlling acute/subacute zoster-associated pain: a retrospective cohort study. Pain Ther. 2024;13(1):99–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xue S, Yang WJ, Cao ZX, Sun T. Comparing the efficacy and safety of short-term spinal cord stimulation and pulsed radiofrequency for zoster-related pain: A systematic review and meta-analysis. Medicine (Baltimore) 2022;101(11). [DOI] [PMC free article] [PubMed]
- 29.Feng F, Long Z, Qiu L, Zhang Y, Ren C, Ou C, et al. Comparative efficacy and safety of pulsed radiofrequency versus spinal cord stimulation in thoracic herpes zoster pain: a retrospective study. Sci Rep. 2025;15(1):23280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zuo L, Su A, Xie Y, Yang X. Clinical study of short-term spinal cord stimulation for herpes zoster-associated pain. Eur J Med Res. 2024;29(1):603. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data are available from the corresponding author upon reasonable request.