Clinical practice guidelines for the management of refractory postherpetic neuralgia by the Korean Pain Society

Abstract

Postherpetic neuralgia (PHN) is persistent pain that occurs after the resolution of skin lesions caused by the reactivation of the herpes zoster virus. Several first-line agents, including anticonvulsants, tricyclic antidepressants, and lidocaine patches, are used in treatment. However, systematic, evidence-based, multidisciplinary clinical practice guidelines (CPGs) addressing the use of opioids and pain interventions—such as nerve blocks—for patients with refractory PHN unresponsive to first-line agents have not yet been established. Therefore, the Korean Pain Society, in collaboration with relevant medical specialty societies, has developed de novo CPGs for the management of refractory PHN. The development process followed Cochrane’s systematic literature review methodology and employed the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system to assess the certainty of evidence and strength of recommendations. Key questions, identified through a survey of experts treating patients with refractory PHN, were finalized by the CPGs development panel. A literature search was conducted in four electronic databases: PubMed (MEDLINE), Embase, Cochrane Library, and KoreaMed. Studies were selected based on predefined inclusion and exclusion criteria using the population, intervention, comparison, outcomes, and study design (PICOS) framework. The resulting recommendations integrate evidence from the literature with patient and clinician preferences and values. They also take into account the quality of evidence, the balance of benefits and harms, potential barriers to implementation, and feasibility within various healthcare settings. These evidence-based multidisciplinary CPGs are expected to assist physicians in safely and effectively treating patients with refractory PHN.

INTRODUCTION

1. General information on PHN

Shingles, or herpes zoster, arises from the reactivation of the varicella-zoster virus, which remains dormant in sensory ganglia, such as the dorsal root ganglion or trigeminal ganglion, after an initial infection. Reactivation typically occurs in individuals with weakened T-cell immunity due to factors such as stress, aging, or immunosuppression [1–3]. Symptoms of herpes zoster typically appear as a unilateral dermatomal rash, maculopapular skin lesions, itching, and intense pain. Although skin lesions heal within a month, pain can persist for several months. Chronic pain that continues beyond the resolution of skin lesions is known as postherpetic neuralgia (PHN), the most common complication of herpes zoster [2,3]. Pain persisting up to 30 days after rash onset is classified as acute zoster pain; pain lasting 30–90 days is considered subacute zoster pain; and pain persisting beyond 90 days is defined as PHN. However, this classification is not universally defined, and the treatment methods for subacute zoster pain and PHN, known as the most common and distressing complication of acute zoster, are similar [4–7]. Therefore, incorporating these points and considering real-world clinical situations, the authors defined PHN in the present clinical practice guidelines (CPGs) as zoster pain that persists even 1 month after the rash onset.

The pathophysiology of PHN is not fully understood but is thought to involve two mechanisms. One is sensitization, where inflammatory mediators such as substance P, histamine, and cytokines lower the activation threshold of nociceptors. The other is neural damage and sensory loss, where inflammatory edema compresses the spinal sensory ganglia, causing ischemia and neural tissue damage [2]. Risk factors for PHN include older age, severe rash, weakened immune system, and high intensity of acute zoster pain [1–3].

PHN is characterized by persistent pain, itching, and increased tactile sensitivity which could significantly disrupt daily life. In severe cases, PHN can cause depression, anxiety, chronic fatigue, and sleep disorders [8,9], with long-term cases experiencing a decrease in social activities and quality of life [10]. Therefore, management of PHN aims to effectively relieve pain and thereby improve the patient’s quality of life [2]. Pharmacologic treatments include anticonvulsants, tricyclic antidepressants, lidocaine patches, and analgesics, all of which have demonstrated efficacy in reducing PHN-associated pain. However, in refractory cases, additional pharmacologic agents or pain interventions could be considered [1,7,11–13].

2. Current status and issues in the management of patients with PHN

PHN occurs in approximately 10%–18% of patients with herpes zoster [14]. The risk of PHN may increase significantly when diagnosis and initiation of treatment of the acute zoster are delayed [5]. Even though the skin lesions may heal within a month, the pain can persist for several months, leading to a decrease in the patient’s quality of life and an increase in medical expenses.

Pharmacologic therapy remains the first-line treatment for neuropathic pain, including PHN. According to the 2010 guidelines from the European Federation of Neurological Societies (EFNS) [15] and the 2015 recommendations by the Neuropathic Pain Special Interest Group (NeuPSIG) [16], primary treatment includes anticonvulsants like gabapentin and pregabalin, tricyclic antidepressants, and lidocaine patches. Opioids, including tramadol, are secondary or tertiary options for treating PHN refractory after the first-line agents [15,16]. Owing to the adverse effects, dose titration is essential to achieve an effective therapeutic concentration. This is particularly important among elderly patients who constitute the majority of PHN cases. The older patients often have multiple comorbidities and take numerous medications which raise concerns like drug–drug interactions and reduce the compliance due to the complexities of polypharmacy. Furthermore, PHN often lasts for several years, necessitating long-term pharmacotherapy. This introduces additional challenges, including the risk of cumulative adverse effects and, in the case of opioids, the potential for tolerance and dependence. Although pain interventions are sometimes used in refractory cases, limited evidence supports their efficacy in neuropathic pain, including PHN, and no comprehensive guidelines currently exist.

3. Background of the clinical guidelines development for management of PHN

The lack of standardized guidelines of PHN in South Korea has urged the development of it. Furthermore, no systematic recommendations exist regarding the use of opioids or pain interventions —such as nerve blocks—for the management of refractory PHN. Managing PHN is difficult, with responses varying among patients. This PHN management guideline could help physicians to keep a standardized approach and to provide timely additional interventions. Moreover, PHN management requires multidisciplinary collaboration across fields such as pain medicine, neurology, internal medicine, and psychiatry. The Korean Pain Society recognized the need for evidence-based, multidisciplinary guidelines and proposed participation from related societies, including the Korean Academy of Family Medicine, the Korean Society for Infectious Diseases, the Korean Academy of Medical Sciences, the Korean Neurological Association, the Korean Neurosurgical Society, the Korean Dermatological Association, and the Korean Association of Pain Management Specialists, as well as others. Representatives from each society contributed to the development process, establishing a foundation for a collaborative, multidisciplinary approach.

These guidelines are intended to support clinicians with the safe and effective management of patients with refractory PHN, with the goal of reducing disease burden, enhancing quality of life, and contributing to public health.

4. Purpose and scope of these CPGs

1) Purpose

The primary goal of these guidelines is to provide evidence-based recommendations for treating refractory PHN in patients who, despite ongoing first-line agents, continue to experience moderate-to-severe pain—defined as a Numeric Rating Scale (NRS) score of ≥ 4—for > 1 month following the onset of herpes zoster rashes. These guidelines present a comprehensive evaluation of the efficacy and safety of adjunctive analgesics (used as second- or third-line treatments) and pain interventions for refractory PHN, along with a clear assessment of evidence levels and a balanced benefit-risk analysis. Designed for applicability across diverse healthcare settings—including primary care clinics, general hospitals, and tertiary referral centers—these guidelines will provide for safe and effective treatment of patients with refractory PHN. They have been developed with consideration of patient values and preferences as much as possible. However, when applying individual recommendations, healthcare providers are encouraged to adapt individual recommendations in accordance with institutional conditions, clinical context, and patient-specific needs.

This was also intended to enhance understanding among patients and policymakers regarding treatment choices. The goal of this guideline is to improve pain management and quality of life for patients suffering from PHN through optimal evidence-based treatment strategies and effective implementation.

2) Intended users

The intended users of these guidelines include healthcare providers in primary, secondary, and tertiary care institutions. The guidelines are also relevant for PHN patients seeking the information of PHN treatment options and policymakers involved in health strategies and healthcare resource allocation.

3) Target populations covered

The target population includes patients with refractory PHN who, despite continuous first-line agents, experience moderate or severe pain for > 1 month after the onset of herpes zoster lesions.

4) Scope

This guideline has focused on effective pain management for patients with refractory PHN, evaluating the efficacy and safety of individual drugs and pain interventions. The selection of additional analgesics or pain interventions may vary depending on the patient’s underlying medical conditions, current medications, personal preferences, indications and contraindications of pain intervention, and the unique characteristics of each medical institution. The guidelines address seven key questions, organized broadly into pharmacological and pain interventions. According to the 2010 EFNS [15] and 2015 NeuPSIG [16] guidelines, first-line agents for PHN include gabapentin, pregabalin, tricyclic antidepressants, and lidocaine patches, whereas tramadol and other opioids are classified as secondary or tertiary options. According to the 2015 Cochrane Review [17], nonsteroidal anti-inflammatory drugs (NSAIDs) showed no significant pain reduction in patients with PHN and are therefore excluded from the guidelines based on neuropathic pain treatment guidelines from EFNS (2010) and NeuPSIG (2015). The pain interventions section excludes intrathecal corticosteroid administration owing to concerns about serious side effects and its inability to produce the same therapeutic effect [18]. Additionally, fascial plane blocks, widely used for postoperative pain management, were excluded owing to limited evidence in PHN cases [19].

5. Determining the level of evidence and grade of recommendation

The CPGs, developed using the de novo method, were based on a systematic review conducted according to the methodology proposed by Cochrane [20]. A literature search was conducted across four electronic databases: PubMed (MEDLINE), Embase, Cochrane Library, and KoreaMed. A modified Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach was adopted to assign the level of evidence and grade of recommendation [21]. The levels of evidence listed in Tables 1 and 2 present the grades of recommendation based on several rounds of discussion. The authors surveyed patients with PHN on each key question, analyzing the opinions of a total of 111 respondents. Additionally, they surveyed 287 experts treating patients with PHN, with their responses referenced to draft the recommendations. The final version of the recommendations was confirmed by a vote of the committee, a user opinion survey (142 respondents), and a review by 6 external experts.

Table 1.

Criteria for assessing evidence levels

Evidence level	Definition
High	There is a high degree of confidence that the estimated effect is close to the true effect.
Moderate	There is moderate confidence in the estimated effect. The estimate is likely close to the true effect but may differ significantly.
Low	There is limited confidence in the estimated effect. The true effect may be substantially different from the estimate.
Very low	There is very little confidence in the estimated effect. The true effect is likely to be substantially different from the estimate.

Table 2.

Definition of recommendation grades

Recommendation grade	Strength	Direction	Definition
General use	Strong	Do	When the benefits of treatment or testing clearly outweigh the associated risks, burdens, and costs.
Elective use	Conditional		When the benefits of treatment or testing outweigh the risks, burdens, and costs; however, some uncertainty exists.
Limited use	Conditional	Do not	When the risks, burdens, and costs of treatment or testing outweigh the benefits; however, some uncertainty exists.
Restricted use	Strong		When the risks, burdens, and costs of treatment or testing clearly outweigh the benefits.

MAIN BODY

1. Guidelines 1

Key question 1: For patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash despite first-line agents, does the addition of tramadol or a tramadol/acetaminophen combination provide pain relief?

1) Recommendation

Adding tramadol or a tramadol/acetaminophen combination is recommended for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash, despite first-line agents.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Very Low

Condition: Elective use is recommended for patients after sufficient explanation of side effects such as constipation, dizziness, drowsiness, nausea, and vomiting.

2) Background

First-line agents for PHN typically include neuropathic pain medications such as gabapentinoids or tricyclic antidepressants. However, patients often report moderate or severe pain even with these treatments. The weak opioid tramadol, commonly used to manage moderate-to-severe pain, is frequently considered the second-line agent for pain relief when typical analgesics, such as NSAIDs, are insufficient. However, despite its frequent prescription in clinical settings, evidence is limited on the utility of adding tramadol or a tramadol/acetaminophen combination to first-line agents in patients with refractory PHN, compared to using only first-line agents. Thus, these guidelines compare the efficacy and safety of adjunctive tramadol or a tramadol/acetaminophen combination versus first-line agents alone in managing moderate-to-severe PHN persisting beyond 1 month after shingles onset.

3) Evidence summary

A literature search and selection process for this key question yielded one single-arm study for analysis [22] (Supplementary Fig. 1-1). The search resulted in four randomized controlled studies on the effects of tramadol on refractory PHN. Three of these studies did not mention first-line agents or evaluated tramadol alone after discontinuing first-line agents and were excluded from the analysis [23–25]. Ultimately, one study was included in the analysis [22] (Supplementary Table 1-1).

The study by Kawai et al. [22] included patients who experienced persistent pain of ≥ 4 on the 11-point NRS for > 2 weeks despite taking oral adjuvant analgesics or non-opioids following the onset of herpes zoster. Initially, during the open-label period, all patients were assigned to a single group and were instructed to add tramadol to their existing medications (Supplementary Table 1-2). Therefore, the evidence level initially started as high. However, owing to the risk of bias in the randomization process, the evidence level was downgraded by one level, resulting in a low evaluation (Supplementary Table 1-3).

The dosage of tramadol was adjusted or maintained weekly based on the reduction in pain scores over 4 weeks to determine a fixed dose, followed by an additional week of fixed-dose administration. Subsequently, a randomized, double-blind trial was conducted for 5 weeks among patients who completed the open-label period. Patients were randomly assigned to either the placebo or tramadol group to take a placebo or a fixed dose of tramadol for 4 weeks. During the 4-week fixed-dose determination in the open-label period, the mean pain scores of patients decreased significantly from 5.8 ± 1.3 (mean ± standard deviation [SD]) to 3.0 ± 1.2. In the double-blind phase, the placebo group showed a pain score increase from 3.1 ± 1.5 at baseline to 3.7 ± 1.7 after 1 week, followed by a gradual decrease to 2.9 ± 1.5 by week 4. The tramadol group recorded similar pain intensity scores throughout, starting at 3.3 ± 1.4 and ending at 2.9 ± 1.4 at week 4 (Supplementary Fig. 1-2). During the open-label period, 78.7% (196/249) of participants reported side effects, with constipation being the most common (43.8%). In the double-blind period, side effects were reported by 39.8% (33/83) in the placebo group and 33.7% (28/83) in the tramadol group. Pharyngitis was the most common side effect in the placebo group (10.2%), whereas nausea predominated in the tramadol group (Supplementary Fig. 1-3).

When evaluating tramadol across various neuropathic pain conditions—including PHN—the number needed to treat (NNT) for achieving ≥ 50% pain relief is estimated at 4.73 (95% confidence interval [CI], 3.6–6.7), based on a median and interquartile range. This makes it more effective compared to commonly used first-line agents such as gabapentin (NNT 7.16 [5.9–9.1]) and pregabalin (NNT 7.71 [6.5–9.4]) [16]. Consequently, tramadol is recommended as a second-line agent for neuropathic pain management. Although excluded from this analysis for focusing only on tramadol versus a placebo without maintaining first-line agents, a study by Boureau et al. [23] demonstrated that patients in the tramadol group showed greater pain relief and required fewer rescue medications after 6 weeks compared to the placebo group. Furthermore, a physician survey revealed that 80.4% of experts would consider adding tramadol in cases of moderate-to-severe pain persisting despite treatment with first-line agents. Similarly, a survey of patients with PHN indicated that a majority (60.4%) would be willing to take tramadol if it provided effective pain relief, and concerns regarding addiction and withdrawal were not reported. Thus, adding tramadol for patients with moderate-to-severe PHN persisting despite treatment with first-line agents offers significant benefits compared to maintaining first-line agents alone. The risks are minimal, and the overall benefits outweigh the harms. Although the balance of benefits and harms is not definitively established, the simplicity of this pain intervention and the preferences expressed by both patients and physicians suggest high acceptance and feasibility. No significant barriers exist beyond potential side effects, and the cost burden is low. This pain intervention is unlikely to exacerbate health disparities or equity concerns. To establish a stronger foundation for these guidelines, given the current very low level of evidence and inherent uncertainties, it is imperative to conduct further high-quality randomized controlled trials (RCTs).

All members of the CPG committee agreed on the direction and level of recommendation (100%) during the first round of voting. In the user survey, 97.2% of respondents agreed with the direction of recommendation, 94.3% agreed with the level of recommendation, and 95.7% indicated their intent to implement the recommendations as presented. In the external expert survey, 100% of the respondents also agreed with the direction and level of recommendation.

2. Guidelines 2

Key question 2: For patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash despite first-line agents and/or tramadol therapy, does the addition of strong opioids provide pain relief?

1) Recommendation

Adding strong opioids is recommended for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash, despite first-line agents and/or tramadol therapy.

Recommendation Grade: Limited Use (Do not, Conditional)

Level of Evidence: Very Low

Condition: Use should be limited to patients with inadequate response to existing treatments and no viable alternatives. Treatment should be short-term, initiated only after careful evaluation of potential adverse effects (including addiction), with close monitoring.

2) Background

PHN is a common condition in clinical practice, presenting with varying levels of severity from mild to severe. Although some cases can be managed with monotherapy using first-line agents, others improve only after adding weak opioids. However, many patients continue to experience persistent pain despite combination therapy.

In patients whose PHN remains uncontrolled despite first-line agents and weak opioids such as tramadol, various strategies for pain management are considered. The use of strong opioids is one of these options, and in clinical practice, they are sometimes recommended as second- or third-line therapies despite the risks associated with their side effects.

However, evidence comparing their efficacy in patients with refractory PHN with that of first-line agents or weak opioids such as tramadol remains limited. This recommendation evaluates the effectiveness and safety of strong opioids in patients with PHN with persistent moderate-to-severe pain, despite treatment with first-line agents and tramadol.

3) Evidence summary

A literature search and selection process for this key question identified one prospective single-arm observational study for analysis [26] (Supplementary Fig. 2-1). In the study by Fan et al. [26], patients aged ≥ 40 with PHN who, despite receiving treatment, continued to experience pain of ≥ 5 on the 11-point Visual Analog Scale (VAS) were administered oxycodone hydrochloride, and their pain scores were measured at 1, 4, and 8 weeks (Supplementary Table 2-1, 2-2). Prior to the study, 82.7% of participants had been receiving the World Health Organization (WHO) step 1 or 2 analgesics, including first-line agents and weak opioids, indicating prior exposure to standard treatments. Although concurrent use of previously prescribed analgesics was not recommended, it was not strictly contraindicated. NSAIDs and antidepressants were allowed to be continued. At baseline, the overall mean VAS pain score was 7.4 ± 10.5 (mean ± SD), which significantly decreased to 3.8 ± 2.0 after 1 week of oxycodone hydrochloride administration. After 4 weeks, the pain score further decreased to 1.8 ± 1.2, and after 8 weeks, it dropped to 1.4 ± 1.2. These results included moderate and severe pain levels, with patients who initially reported severe pain (VAS ≥ 7) also showing a continuous decrease in pain without any rebound during the 8-week observation period (baseline 8.2 ± 1.0; 1 week 4.1 ± 2.1; 4 weeks 1.9 ± 1.4; 8 weeks 1.5 ± 1.2) (Supplementary Fig. 2-2). The most frequently observed side effect was nausea, reported by 16.0% of patients (38 out of 237) after 1 week of oxycodone hydrochloride administration. This was followed by dizziness (9.3%), constipation (6.8%), and drowsiness (4.6%). However, these adverse events showed a general trend of reduction in severity among patients who continued treatment (Supplementary Fig. 2-3; Supplementary Table 2-3).

In a survey of 111 patients with PHN, 76.8% of respondents cited adverse effects as the main reason for not preferring opioid treatment as the next step after persistent moderate-to-severe pain. When informed about the relatively low adverse effects of weak opioids, 64.5% indicated a willingness to use them. However, 66.4% indicated they would not consent to strong opioids, even if significant pain relief was expected. Similarly, physicians demonstrated a reluctance to prescribe strong opioids, even when significant pain reduction was anticipated; 75.4% indicated that they would avoid adding strong opioids. According to a patient survey regarding undesirable adverse effects of strong opioid use, the most frequently reported concerns were addiction and withdrawal (56.9%), followed by constipation (13.7%), drowsiness and dizziness (12.7%), and nausea and vomiting (11.8%). Physicians shared similar concerns, emphasizing addiction and withdrawal as their primary concerns. Thus, careful handling and thorough evaluation are required during prescription, with continuous monitoring for adverse effects post-prescription. When initiating strong opioid therapy, clinicians should collaborate closely with the patient to define the goals of strong opioid therapy which were the pain control, daily functioning, and quality of life before starting it.

Based on the evidence-to-decision framework, the use of strong opioids in patients with PHN with moderate-to-severe pain persisting despite first-line agents and/or tramadol may provide pain relief but carries risks such as nausea, vomiting, and constipation. The evidence level supporting their use is very low and subject to uncertainty and variability. Furthermore, potential risks such as dependence and addiction cannot be excluded. Thus, although strong opioids can be considered when other pain interventions are unavailable or ineffective, they should not be prioritized for pain management. This pain intervention's acceptability varies depending on barriers from side effects and the preferences of patients and physicians. However, strong opioids are feasible to prescribe, unlikely to impose significant financial burdens, and are not expected to significantly impact health equity.

In the first voting round, 92.3% of the CPG committee members supported the recommendation direction and grade. In the user survey, 78.7% of respondents agreed with the direction of recommendation, 90.1% agreed with the level of recommendation, and 85.8% expressed willingness to implement the recommendation as proposed. In the external expert survey, 100% of the respondents agreed with the recommendation direction (not to prioritize the use of strong opioids) and grade of limited use.

3. Guidelines 3

Key question 3: For patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash despite first-line agents, does epidural block provide pain reduction compared to no pain intervention?

1) Recommendation

Epidural block is recommended for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash, despite first-line agents.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Very Low

Condition: Elective use is recommended for patients without contraindications to epidural block. It must be performed by a clinician with adequate training and experience, in a setting equipped with appropriate medical resources. Patients must be fully informed about the procedure’s potential benefits and risks and must provide informed consent prior to treatment.

2) Background

PHN is a representative refractory neuropathic pain condition encountered by pain physicians [27]. Although first-line agents for neuropathic pain could be typically used for PHN management, patients often continue to complain of moderate-to-severe pain, despite these medications [28]. PHN often affects the neck and trunk regions (neck: 13.1%, thoracic: 52.9%, and lumbar: 13.8%) [29]. In such cases, an epidural block is frequently performed as an adjunctive pain intervention [30]. Although epidural blocks are commonly used in conjunction with medication therapy for PHN, the evidence supporting their added benefit over medication therapy alone is limited [31]. Therefore, this guideline seeks to assess the efficacy and safety of epidural block as an adjunct to pharmacologic therapy in patients with persistent moderate-to-severe PHN pain.

3) Evidence summary

Through the literature search and selection process for this key question, two case-control studies and three cohort studies were included in the analysis [14,32–35] (Supplementary Fig. 3-1). The pain interventions evaluated included various epidural block techniques, such as interlaminar and transforaminal approaches, performed at cervical, thoracic, lumbar, and sacral levels. Control groups received either medication therapy or sham pain interventions (Supplementary Table 3-1). The included studies exhibited a high risk of bias owing to insufficient reporting of baseline factors, small sample sizes, and significant variability in disease duration (Supplementary Table 3-2). The overall quality of evidence was rated as very low owing to multiple factors, including small sample sizes, high risk of bias, and wide CIs (Supplementary Table 3-3).

A literature search identified one retrospective study and one RCT on a single epidural block in patients with refractory PHN [36,37]. Although two studies demonstrated significant pain reduction 3 months post-procedure, they compared different epidural block techniques or variations in adjunctive medications. Most published research on epidural block for PHN has focused on continuous, rather than single blocks. Consequently, studies utilizing continuous epidural blocks were selected for these guidelines. Two case-control studies were included, comparing first-line agents alone to medication therapy with an additional epidural block. To investigate complications, one prospective study and two retrospective studies were also included. Although their comparison groups did not fully meet the inclusion criteria, their populations and pain interventions aligned with the inclusion criteria, and they provided relevant data on adverse effects. These studies were analyzed as single-arm studies [14,32,33].

Li et al. [34] showed that pain score and frequency, as well as the rate of inability to work in the experimental group, decreased compared to the control group, showing a significant difference at 6 months after epidural block. Seo et al. [35] compared 17 patients receiving medication therapy with 83 patients who received continuous epidural block for 2 weeks. They reported that the number of patients in the experimental group who experienced a 50% or 100% reduction in their existing pain was 5.4 times and 4.5 times higher, respectively, compared to the control group at 6 months after epidural block (Supplementary Fig. 3-2). Studies including groups of 6, 44, and 22 patients, reported adverse events during epidural block procedures, complication rates ranging from 2%–36%, depending on the study (Supplementary Fig. 3-3). However, only mild and temporary symptoms, such as pain at the catheter insertion site, headache, dizziness, urinary retention, and constipation, were observed. Serious complications, such as nerve injury, infection, epidural hematoma, total spinal anesthesia, cardiopulmonary collapse, pneumothorax, or permanent paralysis, were not reported in the studies, except for a single case of temporary lower limb paralysis. Recent clinical guidelines recommend the use of fluoroscopic guidance during the epidural block to prevent serious complications and ensure effective procedures. This is particularly important for continuous epidural blocks, where proper catheter placement is critical.

Based on the meta-analysis of the two studies, which compared pain reduction and rates of pain relief, adding an epidural block to first-line agents in patients with PHN who continue to experience moderate-to-severe pain appears to offer greater benefits than medication alone. Additionally, an expert survey revealed that 75.9% of respondents recommended the use of an epidural block for patients with refractory PHN. A survey of 111 patients with PHN revealed that when moderate-to-severe pain persisted despite first-line agents, patients preferred nerve block procedures, including epidural block, over opioid therapy as the next treatment option (67.6%). When informed about the evidence supporting epidural block, 74.5% of patients agreed to the procedure. In summary, both patients and physicians demonstrated a high preference for epidural block in patients with refractory PHN (67.6%–75.9%).

Reported complications in the studies were rare or mild, typically resolving with conservative management. However, according to a patient survey, 48.9% of patients who declined the procedure cited concerns about complications. If the procedure is performed by a trained pain physician using fluoroscopic guidance, the risk of adverse events can be minimized. Thus, clinicians fully proficient in the technique and medication should provide patients with accurate information about potential side effects to help them make informed decisions about the procedure.

Based on these considerations, when the epidural block is performed by a trained pain physician using fluoroscopic guidance, the benefits outweigh the risks for patients with PHN who continue to experience moderate-to-severe pain despite first-line agents. Therefore, the epidural block is recommended. Following committee discussions, 90.8% agreed with the recommendation on the direction and level of recommendation for the epidural block during the first round of voting. Among the respondents of the user survey, 95.1% agreed with the recommendation to perform epidural nerve block, 90.8% approved the level of recommendation, and 92.1% expressed willingness to implement the recommendation. In the external review, 83.3% of the respondents agreed with the recommendation grade of its use. Thus, the recommendation and its grade were finalized as originally drafted.

Given the favorable benefit–risk profile, this pain intervention is likely to be widely accepted and utilized in clinical practice. To ensure safety and efficacy, the procedure should only be performed by clinicians with appropriate training and experience, in well-equipped medical facilities, following thorough patient counseling. To establish a stronger foundation for these guidelines, given the current very low level of evidence and inherent uncertainties, it is imperative to conduct further high-quality RCTs.

4. Guidelines 4

Key question 4: For patients with PHN affecting the facial region, cervical spine, or upper limbs who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash despite first-line agents, is stellate ganglion block (SGB) effective?

1) Recommendation

SGB is recommended for patients with PHN affecting the facial region, cervical spine, or upper limbs who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash, despite first-line agents.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Very Low

Condition: Elective use is recommended for patients without contraindications to SGB. The procedure should be performed by a clinician with proper training and sufficient experience, in facilities equipped with appropriate medical equipment and environment. The patient must be thoroughly informed about the potential benefits and risks and must provide consent before proceeding.

2) Background

PHN is a persistent neuropathic pain that can last for months to years after the rash has cleared, severely impacting the function and quality of life [38]. PHN most commonly affects the thoracic region but frequently involves the cranial nerves, especially the ophthalmic branch of the trigeminal nerve, as well as the cervical spine [39]. SGB is often performed in cases of refractory PHN affecting these regions [40]. Despite its widespread clinical use, evidence regarding the utility of SGB, when added to first-line agents and compared to conservative treatment, is limited. These guidelines compare the efficacy and safety of SGB, in addition to first-line agents, versus first-line agents alone, in patients with PHN with moderate-to-severe pain affecting the facial region, cervical spine, or upper limbs.

3) Evidence summary

A literature search identified two prospective studies and one retrospective study for analysis [41–43] (Supplementary Fig. 4-1). No studies compared SGB with conservative treatment. Instead, studies analyzing the effects of SGB were conducted as single-arm studies, comparing pain scores before and after the pain intervention (Supplementary Table 4-1). All three studies were rated as having a high risk of bias owing to confounding factors, including lack of reporting baseline characteristics such as comorbidities and affected regions, and small sample size (Supplementary Table 4-2). The level of evidence was assessed as very low owing to the high risk of bias from confounding factors, as well as imprecision resulting from the small sample size and wide CIs (Supplementary Table 4-3).

Ding et al. [41] compared pain scores and quality of life at 1 and 2 weeks, as well as 1, 3, and 6 months following SGB in 42 patients or pulsed radiofrequency ablation of the stellate ganglion. Fan et al. [42] evaluated the effects of different local anesthetics used in SGB. Pain and anxiety scores were compared at 1 week, 1 month, and 3 months after the procedure. Lidocaine was used in 118 patients and ropivacaine in 134 patients. Wang et al. [43] assessed pain scores following SGB in 12 patients, comparing the results with extracorporeal shock wave therapy conducted along the ganglia with the tenderness point at 2, 4, and 6 weeks. All three studies consistently demonstrated significant pain reduction at 1 and 3 months post-procedure compared to pre-procedure levels (Supplementary Fig. 4-2).

Reports on complication rates related to SGB in the selected studies did not include any serious complications. In the study by Ding et al. [41], complications were reported in 52.4% of cases; however, these included minor and temporary symptoms such as hoarseness, dizziness, and a foreign body sensation in the neck due to the sympathetic block. Fan et al. [42] reported no significant differences in complications based on the type of local anesthetic used. However, a 2019 systematic review of complications associated with SGB revealed that 48.5% of reported complications occurred in procedures performed without image guidance and based solely on anatomical landmarks [44]. Reported severe complications included death and limb paralysis. Therefore, recent clinical guidelines for pain interventions recommend the use of imaging devices to prevent serious complications and to perform effective procedures.

Based on a meta-analysis of the three studies, the addition of SGB to first-line agents for patients with moderate-to-severe PHN in the face, cervical region, or upper extremities appears to provide benefits compared to prescribing medication alone. In the expert opinion survey, 90.2% of clinicians recommended SGB for refractory PHN, and 81.5% of clinicians reported that they could perform the procedure. A survey of 99 patients with PHN revealed that 72.7% agreed to undergo SGB if moderate-to-severe pain persisted despite first-line agents. The survey identified concerns about complications (41.1%), lack of effectiveness (36.8%), pain during the procedure (14.7%), and cost (7.4%) as reasons for patient hesitation. Nonetheless, the survey indicated that both patients and clinicians showed a high preference for SGB in patients with refractory PHN (72.7%–90.2%). Although evidence is limited, these factors suggest that SGB can offer significant benefits for patients with refractory PHN. The complications reported in the selected studies were minor and temporary, often incidental to sympathetic nerve block. If the procedure is performed by well-trained clinicians using imaging devices, the risk of complications can be minimized. For these CPGs, 100% of the CPG committee agreed on the direction and level of recommendation through two-round voting. The user survey results showed that 93.7% agreed with the recommendation direction to perform SGB, 94.3% agreed with the “conditional use” recommendation level, and 89.9% expressed willingness to implement the recommendation. After consultation and review by external experts, the external reviewers showed 83.3% agreement with the direction and level of recommendation. Therefore, these CPGs were finalized by reflecting on the results of voting by the CPG committee, external expert consultation, and user consultation and voting.

Considering the benefits and risks, this pain intervention is likely to be preferred and widely accepted by both patients and clinicians. While studies support its effectiveness, the high risk of bias, small sample sizes, and lack of comparative trials warrant caution. Furthermore, high-quality RCTs are needed to strengthen the evidence base.

5. Guidelines 5

Key question 5: Is peripheral nerve block effective in patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for >1 month after the onset of herpes zoster rash?

1) Recommendation 5-1

Key question 5-1: Is botulinum toxin type A injection effective in patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for > 1 month after the onset of herpes zoster rash?

Subcutaneous botulinum toxin type A injection is recommended for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for > 1 month after the onset of herpes zoster rash.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Low

Conditions: The literature indicates that the risk/benefit ratio for using subcutaneous botulinum toxin A injection appears to be low, with no significant safety concerns reported. However, in South Korea, the subcutaneous injection of botulinum toxin A in patients with PHN is not currently approved by the Ministry of Food and Drug Safety. Therefore, thorough discussion with the patient, careful consideration, and regulatory review and improvement are necessary before its clinical application.

2) Recommendation 5-2

Key question 5-2: Is paravertebral nerve block effective in patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for > 1 month after the onset of herpes zoster rash?

Paravertebral nerve block is recommended for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for > 1 month after the onset of herpes zoster rash.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Very low

Conditions: To perform paravertebral nerve blocks effectively and safely, the use of imaging guidance is essential. This requires appropriate imaging equipment, a dedicated procedural space, and adequately trained medical personnel.

3) Recommendation 5-3

Key question 5-3: Is intercostal nerve block effective in patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for > 1 month after the onset of herpes zoster rash?

Intercostal nerve block is recommended as an equivalent alternative to epidural nerve block for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) despite first-line agents for > 1 month after the onset of herpes zoster rash.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Very low

Conditions: Ultrasound guidance is critical for the effective and safe performance of intercostal nerve block. This requires appropriate ultrasound equipment, a dedicated procedural space, and medical personnel adequately trained in its use.

4) Background

Although first-line agents are effective for many patients with PHN, some patients continue to report persistent moderate-to-severe pain despite these treatments. In such cases, pain interventions, including epidural nerve blocks or peripheral nerve blocks, are considered. Peripheral nerve blocks are particularly valuable when pain occurs in areas where epidural blocks are unsuitable or not an option. Representative examples include subcutaneous botulinum toxin type A injections, paravertebral nerve blocks, and intercostal nerve blocks. Although peripheral nerve blocks are widely used in clinical practice, the evidence regarding their efficacy and safety remains limited. Herein, the authors compared the efficacy and safety of subcutaneous botulinum toxin A injections, paravertebral nerve blocks, and intercostal nerve blocks in patients with PHN who continue to experience moderate or severe pain for > 1 month after the onset of the herpes zoster rash despite ongoing first-line agents.

5) Evidence summary

A literature search identified four single-arm studies [45–48] and two RCTs [49,50] investigating the use of subcutaneous botulinum toxin A injections in patients with refractory PHN. However, the four single-arm studies [45–48] were excluded owing to the absence of a control group, and one RCT [50] was excluded because of inconsistencies in the observation period. Thus, only one RCT [49] was included, in which outcomes such as pain reduction, sleep quality, and the occurrence of adverse events were evaluated at 1 month following subcutaneous botulinum toxin A injection (Supplementary Fig. 5-1).

Regarding paravertebral nerve blocks in patients with refractory PHN, one prospective observational study [51] was excluded because its focus was on changes in laboratory parameters after the procedure rather than clinical symptom improvement, which did not align with the key research question. The study by Liu et al. [52], a single-arm, prospective observational study involving 21 patients with refractory PHN, assessed pain reduction following ultrasound-guided paravertebral nerve block and evaluated improvements in pain and sleep quality at 1 day, 3 months, and 6 months post-procedure. This study was the only one that met the inclusion criteria and was included in the analysis (Supplementary Fig. 5-1).

Among studies investigating intercostal nerve blocks in patients with refractory PHN, the study by Lee et al. [53] was the only one that directly compared the efficacy of intercostal nerve blocks with that of epidural nerve blocks. In this study, outcomes such as pain reduction, the number of nerve block procedures performed, and the total duration of treatment were assessed at 1 month following the intercostal nerve block. Another study [54] was excluded, which was a case report (Supplementary Fig. 5-1).

Thus, three studies were included in the final analysis: one RCT on subcutaneous botulinum toxin type A injection [49], one prospective patient-controlled study on paravertebral nerve block [52], and one cohort study on intercostal nerve block [53] (Supplementary Table 5-1).

The study by Apalla et al. [49], which investigated subcutaneous botulinum toxin type A injections, was assessed for risk of bias using the ROB2 tool (Supplementary Table 5-2). The initial level of evidence was considered high. However, owing to the small sample size and wide CIs indicating substantial imprecision, the level of evidence was downgraded by two levels, yielding a rating of low certainty (Supplementary Table 5-3).

The studies by Liu et al. [52] on the paravertebral nerve block and Lee et al. [53] on the intercostal nerve block were non-randomized controlled studies and were evaluated using the ROBINS-I tool (Supplementary Table 5-4). Although the initial level of evidence was considered high, both studies demonstrated a high risk of bias owing to confounding variables and selection bias. Moreover, issues of imprecision were significant. Therefore, the certainty of evidence for these studies was downgraded by three levels, resulting in a rating of very low certainty (Supplementary Table 5-5, 5-6).

According to the study by Apalla et al. [49], subcutaneous injections of botulinum toxin type A in 15 patients with refractory PHN significantly relieved pain and improved sleep quality at 4 weeks post-injection compared to the saline injection control group. Specifically, the experimental group showed a reduction in the mean VAS pain score from 8.8 to 4.1 (95% CI, 3.5–4.7), whereas the control group showed a reduction from 8.7 to 8.1 (95% CI, 7.5–8.7) (P < 0.05). In terms of sleep quality, the experimental group improved from a mean score of 8.7 to 4.2 (95% CI, 3.2–5.3), whereas the control group showed a change from 9.0 to 8.5 (95% CI, 7.5–9.6) (P < 0.05) (Supplementary Fig. 5-2). However, in Korea, subcutaneous botulinum toxin A injection for PHN is not currently approved by the Ministry of Food and Drug Safety. Furthermore, given that only 16.4% of surveyed physicians reported having performed this procedure, clinical experience remains limited, and its efficacy and potential complications are not yet fully characterized, leading to cautious and restricted use in practice.

In the study by Liu et al. [52], among 21 patients who underwent paravertebral nerve block, the VAS score decreased by approximately 2.3 points at 6 months post-procedure, demonstrating significant pain reduction. Additionally, sleep quality, measured on a 10-point scale, improved by 0.9 points compared to baseline (Supplementary Fig. 5-3).

For the intercostal nerve block, the study by Lee et al. [53] compared outcomes between 18 patients who received intercostal nerve blocks and 20 patients who underwent epidural nerve blocks. Both groups exhibited significant reductions in pain scores at 1 month post-procedure, with decreases of approximately 1.6 points and 1.5 points, respectively, and no statistically significant difference between the two groups. The mean number of procedures over 1 month was 2.3 in the intercostal nerve block group and 2.1 in the epidural group, with no significant difference. The treatment duration also did not differ significantly, at 28.3 days and 34.8 days, respectively (Supplementary Fig. 5-4).

The study involving 15 patients receiving subcutaneous botulinum toxin A injections reported no serious adverse effects [49]. Furthermore, existing literature indicates that the risk-benefit profile of subcutaneous botulinum toxin A injection is favorable, with a very low likelihood of serious safety concerns [55]. The most frequently reported adverse event is transient pain at the injection site, which resolves spontaneously within a few hours, with no significant difference in incidence between the treatment and placebo groups [55]. The paravertebral nerve block resulted in adverse events such as dizziness (9.5%), nausea (4.8%), and drowsiness (9.5%); however, no severe complications such as nerve injury, infection, pneumothorax, or permanent paralysis were reported. Similarly, Lee et al. [53] reported no adverse events. In the study by Liu et al. [52], only minor side effects following intercostal nerve block were reported.

When performing paravertebral or intercostal nerve blocks in patients with PHN, the use of fluoroscopic or ultrasound guidance is essential to ensure procedural safety and efficacy. Both studies employed ultrasound-guided techniques, supporting the conclusion that these procedures entail minimal risk when performed by well-trained clinicians with adequate experience and appropriate medical equipment.

According to a survey of 111 patients with PHN, 73.6% of patients expressed willingness to undergo peripheral nerve blocks if moderate-to-severe pain persisted despite continued first-line agents. When asked about the most important factor in deciding whether to receive a peripheral nerve block, 51% of patients indicated the efficacy of the procedure as their top priority, followed by concerns regarding potential adverse effects (31%), procedural pain (12%), and cost (6%). These findings suggest that patients are generally inclined to prefer peripheral nerve blocks if sufficient pain reduction can be expected. In a separate survey of 287 medical professionals, 80.5% responded that peripheral nerve blocks are effective for patients with PHN who continue to experience moderate-to-severe pain despite ongoing first-line agents. However, the effective and safe performance of peripheral nerve blocks requires the use of ultrasonographic or fluoroscopic guidance. This necessitates an appropriate ultrasound and C-Arm fluoroscopy equipment, a dedicated procedural space, and trained medical personnel capable of performing the procedure safely and accurately. Additionally, in the case of fluoroscopy-guided procedures, a radiation-shielded room and personnel skilled in operating fluoroscopic equipment are essential to ensure safety and optimal outcomes.

During the second round of voting, 96.7% of CPG committee members agreed with the recommendation on the direction and level of recommendation for botulinum toxin A. In the user survey for botulinum toxin A, 69.3% of respondents agreed with the direction of recommendation, 73.8% agreed with the level of recommendation, and 60.3% expressed willingness to implement the recommendation. For paravertebral nerve block, 100% of the committee members agreed with the recommendation direction and 91.7% with the grade during the first vote. In the user survey for paravertebral block, 91.4% of respondents agreed with both direction and level of recommendation, and 88.7% expressed willingness to implement the recommendation. In the first vote on the recommendation direction and grade for intercostal nerve block, 91.7% of committee members agreed with both the direction and grade of recommendation. In the user survey for intercostal nerve block, 92.3% of respondents agreed with the recommendation direction, 92.9% agreed with the level of recommendation, and 89.4% expressed willingness to implement the recommendation. In the external review, all respondents (100%) agreed with the grade and level of recommendation for botulinum toxin A, paravertebral nerve block, and intercostal block.

6. Guidelines 6

Key question 6: For patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash despite first-line agents, is pulsed radiofrequency treatment in addition to medication effective?

1) Recommendation

Pulsed radiofrequency, combined with medications, is recommended for patients with PHN experiencing moderate-to-severe pain for > 1 month despite first-line agents.

Recommendation Grade: Strong (Do, Strong)

Level of Evidence: Low

2) Background

Although first-line agents are typically used in patients with PHN, many patients report persistent, moderate-to-severe pain despite these treatments. For such patients, pain interventions, including pulsed radiofrequency treatment, may be considered. However, the specific utility of pulsed radiofrequency treatment as an adjunct to pharmacotherapy has not been well-established. Herein, the efficacy and safety of pulsed radiofrequency treatment are evaluated compared to first-line agents alone in patients with refractory PHN.

3) Evidence summary

A total of seven RCTs were included after the literature review and selection [56–62] (Supplementary Fig. 6-1). In the study by Gu et al. [56], only patients who received intravenous acyclovir and supraorbital nerve block were compared with those who additionally received pulsed radiofrequency. In the study by Wang et al. [62] and Huang et al. [57], patients who received only pharmacotherapy were compared with patients who received pharmacotherapy plus pulsed radiofrequency. Moreover, the remaining four studies compared patients who received first-line agents plus sham treatment with patients who received first-line agents plus pulsed radiofrequency [58–61] (Supplementary Table 6-1). The included RCTs were assessed using ROB2, and the quality of evidence initially started as high. However, the quality was downgraded owing to the following reasons: 1) risks of bias from missing outcome data, measurement of outcomes, and selection of reported results; 2) few participants in each study led to wide CIs, resulting in significant imprecision (Supplementary Table 6-2). Therefore, the quality of evidence was downgraded by two levels and assessed as low (Supplementary Table 6-3).

The meta-analysis demonstrated that mean pain intensity measured by the VAS (0 to 10) was lower in the pulsed radiofrequency group compared with the control group (mean difference [MD] –1.91 [95% CI, –1.15 to –2.67], –1.70 [95% CI, –1.05 to –2.35], and –1.97 [95% CI, –0.39 to –3.56] at 1, 3, and 6 months after the treatment, respectively) (Supplementary Fig. 6-2). The benefits related to SF-36 scores were analyzed in three studies [58,59,61]. In groups that underwent pulsed radiofrequency treatment, bodily pain scores of SF-36 were significantly higher at both 3 and 6 months post-treatment compared to control groups (MD 15.78 [95% CI, 10.78 to 20.77] and 18.75 [9.07 to 28.42], respectively). SF-36 emotional role scores of the pulsed radiofrequency group were significantly higher at 3 and 6 months post-treatment compared to control groups (MD 15.89 [95% CI, 4.93 to 26.85] and 15.53 [2.53 to 28.54], respectively). Moreover, general health scores of the pulsed radiofrequency group were significantly higher at 3 and 6 months post-treatment compared to control groups (MD 22.81 [95% CI, 12.74 to 32.89] and 20.59 [8.67 to 32.51], respectively). SF-36 mental health scores of SF-36 were significantly higher at 3 and 6 months post-treatment compared to control groups (MD 20.82 [95% CI, 11.13 to 30.52] and 17.15 [9.50 to 24.80], respectively). Two studies [59,61] analyzed the pregabalin dose: patients who received pulsed radiofrequency treatment showed significantly lower medication consumption at 1, 3, and 6 months post-treatment compared to control groups (Supplementary Fig. 6-3). The studies included in the analysis reported no severe complications such as pneumothorax, infections, nerve injury, or paresthesia (Supplementary Fig. 6-4). However, other studies have reported complications such as pneumothorax and hematoma, emphasizing the need for caution, especially when performing the procedure in the thoracic region [63,64]. Given these results, pulsed radiofrequency treatment is recommended for patients with PHN experiencing moderate-to-severe pain despite first-line agents, as the benefits outweigh the risks.

A survey of 111 patients with PHN revealed that 53.3% agreed to undergo pulsed radiofrequency treatment as the next step if their pain persisted despite first-line agents. Among 288 physicians, 86.1% supported pulsed radiofrequency treatment for patients with refractory PHN who did not respond to first-line agents. However, only 29.3% could perform the procedure, reflecting challenges related to access to equipment, training, and clinical resources. In South Korea, the cost of pulsed radiofrequency treatment is typically covered by National Health Insurance, often resulting in minimal or no out-of-pocket expenses. As such, reluctance to undergo this treatment owing to financial reasons is relatively rare. However, pulsed radiofrequency treatment requires additional equipment, such as radiofrequency generator devices, trained personnel, and more procedural time compared to nerve blocks. Moreover, 38.3% of surveyed experts stated that they lacked sufficient training or experience with pulsed radiofrequency. These requirements for equipment, facilities, and expertise are not universally available in all hospitals providing pain management. Moreover, to minimize possible complications associated with pulsed radiofrequency treatment, such as infection, bleeding, hematoma, nerve injury, and pneumothorax, the procedure should be performed under C-Arm fluoroscopy or ultrasound guidance.

The CPG committee unanimously agreed on the direction of recommendation (100%), with 91.7% supporting the level of recommendation in the first round of voting. In the user survey, 81.4% of respondents agreed with the direction of recommendation, 79.4% agreed with the level of recommendation, and 74.5% expressed willingness to implement the recommendation as proposed. In the external expert survey, 100% of respondents agreed with the direction and level of recommendation.

7. Guidelines 7

Key Question 7: Is spinal cord stimulation (SCS) effective for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash despite first-line agents?

1) Recommendation

SCS is recommended over pulsed radiofrequency or nerve block for patients with PHN who continue to experience moderate-to-severe pain (NRS ≥ 4) for > 1 month after the onset of herpes zoster rash, despite first-line agents.

Recommendation Grade: Elective Use (Do, Conditional)

Level of Evidence: Moderate

Conditions: SCS may reduce pain intensity and frequency and improve functionality for patients with refractory PHN who did not achieve adequate pain relief with pulsed radiofrequency or nerve block. However, SCS implantation requires adequate training and experience, and the relatively high costs of SCS implantation must be considered. SCS also requires specialized equipment and personnel within the healthcare institution.

2) Background

In clinical settings, many patients with refractory PHN fail to achieve significant pain relief with first-line agents or experience side effects that limit their use. In these cases, more invasive pain interventions may be considered, such as nerve blocks or pulsed radiofrequency. For patients with refractory PHN with limited response to these pain interventions, SCS may offer an additional therapeutic option. However, the comparative effectiveness and safety of SCS for refractory PHN remain inadequately studied. These guidelines evaluate the efficacy and safety of SCS for patients with refractory PHN with inadequate response to other pain interventional treatments, including nerve blocks and pulsed radiofrequency.

3) Evidence summary

A literature search identified five prospective RCTs for analysis (Supplementary Fig. 7-1). These trials included in the analysis employed SCS temporarily for 1–2 weeks during hospitalization, without permanent implantation. Among the included studies, one compared SCS with nerve blocks (epidural block, intercostal nerve block, or SGB) as a control group [65], whereas the other four studies used pulsed radiofrequency as the control group [63,64,66,67]. The study by Wan and Song [67] selected patients within 3 months of herpes zoster onset; however, since all enrolled patients had already passed 1 month post-herpes zoster onset, this study was included in the analysis. Only Liu et al. [65] investigated pain intensity and sleep quality up to 3 months post-procedure. Conversely, the other four studies assessed pain intensity, functional scales, outcomes, and rates of 50% and 100% pain reduction at 6 months post-procedure, with odds ratios provided [63–65,67] (Supplementary Table 7-1). All included studies were RCTs and were assessed using the ROB2 tool (Supplementary Table 7-2). Therefore, the evidence level initially started as high. However, owing to the risk of bias in the randomization process, the evidence level was downgraded by one level, resulting in a moderate evaluation (Supplementary Table 7-3).

Meta-analysis focused on pain intensity, functional evaluation, and the proportion of patients whose baseline pain decreased by 50% and 100%. At 6 months post-procedure, the pain intensity in the experimental group was significantly lower than in the control group (MD –1.69 [95% CI, –2.06 to –1.33]) (Supplementary Fig. 7-2). Additionally, the SF-36 Bodily Pain score showed a significant improvement in the SCS group compared to the control group at 6 months (MD 14.75 [95% CI, 12.22 to 17.27]). At 6 months post-procedure, the SF-36 Physical Function score was also significantly improved in the experimental group compared to the control group (MD 11.71 [95% CI, 7.91 to 15.49]) (Supplementary Fig. 7-2). The proportion of patients who achieved complete remission of baseline pain 6 months after SCS was higher in the experimental group than in the control group (risk ratio [RR] 2.03 [95% CI, 1.26 to 3.28]) (Supplementary Fig. 7-3). At the same time point, the proportion of patients who experienced a 50% reduction in baseline pain was also higher in the SCS group than in the control group (RR 1.41 [95% CI, 1.14 to 1.76]) (Supplementary Fig. 7-3). SCS implantation is an invasive surgical procedure that may cause bleeding, infection, nerve damage, dural damage, or electrode displacement. However, the five studies included in the analysis did not report complications related to SCS implantation. Nevertheless, two cases of pneumothorax [63] and two cases of hematoma [64] were reported in the pulsed radiofrequency group. The potential complications of SCS implantation, including bleeding, infection, nerve damage, dural damage, and electrode displacement, were not observed in the included studies (Supplementary Fig. 7-4). Reports of complications in other literature were mild and resolved with conservative treatment [68,69]. It is believed that the complication rate associated with SCS implantation would not be high if the procedure is performed by a well-trained and experienced physician at a facility equipped with adequate preparation and patient monitoring devices.

Meta-analysis of five prospective RCTs suggests that SCS significantly benefits patients with moderate-to-severe PHN unresponsive to first-line agents. These benefits include pain reduction and functional improvement. Consistent findings in studies excluded from the analysis, such as single-arm or retrospective comparative studies, also report that SCS is effective in reducing pain and improving function in patients with refractory PHN [68–71].

A survey conducted on 111 patients with PHN revealed that 83.5% identified pain relief as their primary concern. Additionally, 67.6% preferred pain interventions. Notably, only 29.1% of patients with refractory PHN expressed willingness to try SCS treatment. This is in contrast to the 70.2% of 285 surveyed clinicians who agreed with the use of SCS for patients with refractory PHN experiencing moderate-to-severe pain despite first-line agents. Therefore, when considering SCS for patients with refractory PHN, clinicians should provide comprehensive explanations regarding its benefits and potential complications to ensure patients’ understanding and consent. SCS procedures are typically performed in inpatient settings and require higher costs than other pain interventions such as nerve blocks, although it is covered under National Health Insurance in South Korea. According to epidemiological studies of PHN in Korea, the incidence of PHN in 2013 was 2.5 cases per 1,000 person-years. The prevalence has steadily increased with the aging population, and the annual treatment cost per patient with PHN was estimated at approximately 13 million KRW (about 9,600 in USD), excluding additional costs related to pain management and supportive therapies [72]. The procedure also requires specialized skills, including anatomical knowledge, an understanding of pain transmission, and proficiency in SCS operation. Therefore, sufficient training and experience are required for SCS treatment. Moreover, a well-equipped facility (e.g., operating rooms, ward for admission, and imaging devices) and an environment where collaboration with other medical professionals and other specialists are critical for SCS treatment. In a survey of experts, 51% reported insufficient training or experience as a barrier to performing SCS. Additionally, 17.4% indicated that while they were capable of performing the procedure, institutional limitations prevented them from doing so. Only 16.7% of respondents stated that they could perform SCS treatment, highlighting potential resource constraints.

During the first vote, 91.7% of the CPG committee members agreed on the direction and level of the recommendation. The user survey results showed that 75.7% agreed with the recommendation direction, 85.0% agreed with the recommendation level, and 64.5% expressed willingness to implement the recommendation. All external reviewers (100%) agreed with the direction and level of recommendation for SCS treatment in patients with refractory PHN. However, only 50% of reviewers believed the recommendation had high feasibility in clinical practice. One external reviewer raised concerns about practical barriers to implementation.

CONCLUSIONS

The present work, the Korean CPGs for the management of refractory PHN, was developed to address the clinical needs associated with the management of patients with refractory PHN. The development method was “de novo,” and 7 key questions were selected based on the opinions of various experts, and 9 recommendations were made. To increase the convenience of using these clinical guidelines, the contents of the recommendations were presented in the form of an algorithm (Fig. 1).

Fig. 1.

Pain management algorithm for patients with refractory postherpetic neuralgia (PHN). For patients experiencing moderate to severe pain for > 1 month after the onset of herpes zoster rashes despite first-line agents like anticonvulsants or antidepressants, additional pharmacological or pain interventional treatments can be considered. Treatment should be tailored based on patient preferences and physician judgment, and combinations of therapies may be utilized when appropriate. However, strong opioids are not recommended as a first option and should only be used for a limited duration if necessary. Botulinum toxin A injections are not currently approved by the Ministry of Food and Drug Safety in Korea. Although randomized studies comparing pulsed radiofrequency with nerve block were not available for inclusion in this analysis, expert recommendations and local circumstances guided the decision-making process. Spinal cord stimulations may be considered for patients whose pain remains inadequately controlled despite other pain interventional treatments, depending on institutional resources and cost considerations. aFor cases involving the face, cervical region, or upper extremities. bLimited use (Do not, Conditional). cGeneral use (Do, Strong). Other treatments are elective use (Do, Conditional).

These guidelines provide evidence-based, multidisciplinary recommendations for the safe and effective management of refractory PHN in clinical practice. The existing guidelines merely listed the treatment methods for PHN by compiling various studies. In contrast, these guidelines stand out by conducting a meta-analysis and additionally compiling the opinions of doctors and patients to present an algorithm for treatment strategy. Although the proposed guidelines are not mandatory, the authors hope to assist physicians and patients in making decisions regarding the management of refractory PHN.

SUPPLEMENTARY MATERIALS

Supplementary materials can be found via https://doi.org/10.3344/kjp.25161.