Evidence-Based Clinical Practice Guidelines for Patients With Lumbar Disc Herniation With Radiculopathy in South Korea
Article information
, Dong Ah Shin3, Jin Hoon Park4, Miyoung Choi5, Hyung-Youl Park6, In Soo Kim7, Jung-Kil Lee8, Chung-Kee Chough9, Seung Hwan Yoon10, Seong-Soo Choi,11, Sung-Woo Choi,12, The Korean Pain Society, The Korean Spinal Pain Society, The Korean Spinal Neurosurgery SocietyAbstract
Objective
In this study, we aimed to develop evidence-based clinical practice guidelines (CPGs) for the treatment of lumbar disc herniation (LDH) with radiculopathy, tailored to South Korean clinical settings.
Methods
The guideline development process used followed the evidence-based medicine principles. Literature searches were conducted across databases, including MEDLINE (PubMed), Cochrane, Embase, and KoreaMed, using predefined search strategies. Titles and abstracts were reviewed to identify the best research evidence. Data extraction and quality assessment were performed using the Cochrane risk of bias tool and the GRADE method. Quantitative meta-analyses or qualitative synthesis were conducted based on data heterogeneity. Recommendations were assigned strength grades (A, B, C, D, I) reflecting evidence reliability.
Results
In these guidelines, comprehensive recommendations for managing LDH with radiculopathy in clinical settings were provided. International evidence and multidisciplinary expert opinions were integrated. Four key clinical questions were identified and divided into sections: surgical treatment, interventional treatment, and physical treatment/exercise. The recommendations for these questions are summarized in this article.
Conclusion
The aim of establishing these CPGs was to enhance treatment outcomes, reduce healthcare costs, and promote public health. By recognizing limitations in domestic data and the dynamic healthcare circumstances, the need for continuous revision was emphasized in these guidelines. Nonetheless, in future updates, the guidelines will be refined to improve their quality and applicability in clinical practice.
INTRODUCTION
Lumbar disc herniation (LDH) is the localized protrusion of disc material beyond the normal boundaries of the intervertebral disc space. Depending on the volume, location, and type of herniated disc, adjacent neural structures may become compressed, causing radiculopathy characterized by pain, numbness, or muscle weakness [1,2]. LDH with associated radiculopathy is a commonly encountered musculoskeletal condition, often prompting most adults to seek medical support. Reports from a domestic study in Korea revealed that over the past decade, the number of patients who received nonsurgical treatment for lumbar radiculopathy increased by more than 30%, and related medical costs also continued to rise [3].
Currently, various treatment methods have been developed and applied for LDH radiculopathy, yet there are still various opinions and controversies on these treatments, and achieving effective pain control or functional recovery can sometimes be challenging [4,5]. Therefore, establishing scientific and rational evidence for LDH, gathering multidisciplinary expert opinions, and creating clinical guidelines applicable to actual practice is crucial. This approach will enhance patients’ quality of life, reduce unnecessary confusion, lower healthcare costs, and promote public health.
Internationally, the North American Spine Society (NASS) released clinical practice guidelines in 2012 for the Diagnosis and Treatment of Lumbar Disc Herniation with Radiculopathy [6,7]. In Korea, the Korean Academy of Rehabilitation Medicine also published clinical guidelines for the nonsurgical treatment of patients with lumbosacral disc herniation [8]. However, these guidelines were based on studies and clinical settings from abroad, which may not fully align with domestic medical practices. Additionally, the primary focus of these guidelines was on nonsurgical treatments, lacking recommendations for surgical approaches. Consequently, the clinical practice guideline committee under the Korean Spinal Neurosurgical Society initially planned the development of the guideline in February 2022. To ensure a multidisciplinary approach, official collaboration requests were sent to the societies of pain and rehabilitation medicine and orthopedics. Subsequently, the guideline development group was formed, comprising committee members recommended by these societies. The clinical guidelines committee for LDH and radiculopathy, consisting of professionals in anesthesiology, pain and rehabilitation medicine, neurosurgery, and orthopedic surgery, has decided to develop evidence-based clinical guidelines on this subject.
According to the Institute of Medicine, clinical practice guidelines are defined as “statements systematically developed to assist healthcare providers and patients in making informed decisions about appropriate healthcare for specific clinical circumstances, based on scientific evidence.” Therefore, clinical guidelines are intended to follow the principles of evidence-based medicine (EBM), establishing a clear association between the final recommendations and the evidence supporting these recommendations. By applying EBM principles, clinical works of literatures are extensively searched to answer questions about specific disease states or medical conditions. The scientific merit of the retrieved literature is assessed based on specific criteria applied to human clinical research, and answers to the questions are derived using the highest level of evidence available among the identified studies. In the final stage, responses to clinical questions are reformulated into recommendations, with each recommendation assigned a strength grade that reflects the reliability of the best clinical evidence available at the time each question was addressed.
It is expected that this clinical practice guideline will be helpful in treating LDH and radicular pain in primary clinical settings. However, considering the limited available domestic data, the reliance on research conducted overseas, and the rapidly changing landscape of healthcare in South Korea, it is essential to continuously revise and improve these guidelines. Nonetheless, in future updates, the quality and applicability of this guideline will be enhanced to better serve clinical practice.
The procedure for the guideline development process for clinical practice guidelines is as follows:
In step 1, guideline development group participants who had been trained in clinical guideline development held several meetings to establish a list of key questions focused on the treatment of LDH with radiculopathy. The questions were divided into 2 for the surgical treatment section, 1 for the interventional treatment section, and 1 for the physical treatment and exercise section, making a total of 4 key questions. Each key question was assigned to members of the working group.
In step 2, a methodology expert from the National Evidencebased Healthcare Collaborating Agency was invited to develop appropriate search terms and literature search strategies for each key question established in each section. Subsequently, a medical research librarian searched the works of literature and completed the search on February 6, 2023 (Supplementary Material 1). To identify literature published in English and Korean, the following databases were searched: MEDLINE (PubMed), Cochrane Database of Systematic Reviews, Embase, and KoreaMed.
In step 3, each working group member initially screened the retrieved articles using their titles and abstracts. The best research evidence was identified and used to answer the targeted clinical questions. During this process, for key questions with sufficient randomized controlled trial (RCT) studies, only RCTs were selected. If no RCTs or only a few were available, prospective observational studies were included. Subsequently, the selected articles were further reviewed to determine the final selection (The literature selection process for each key question is presented in Supplementary Material 1).
In step 4, members independently summarized the characteristics and research conclusions of the finally selected papers and created evidence tables (Supplementary Material 1). To systematically assess for the risk of bias, each selected article was reviewed by at least 2 members of the working group. They independently assessed the selected articles following the Cochrane risk of bias tool [9], and consensus was reached by discussion. Data extraction for outcomes was conducted using a predefined method tailored to the synthesis requirements [10]. For studies in which 2 interventions were compared, extraction methods were chosen to evaluate comparability. Following data extraction and review, meta-analysis was performed when feasible for quantitative synthesis. Quantitative synthesis was deemed feasible when at least 2 studies provided numerical data on the same outcome measure. A qualitative description was provided when metaanalysis was not possible. Meta-analysis was performed using RevMan 5.3 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). Furthermore, a random-effects model was used when substantial heterogeneity among studies was detected. However, when heterogeneity was deemed low, a fixed-effect model was used. Heterogeneity was assessed using the I² statistic, with an I² value exceeding 50% indicating significant heterogeneity. The levels of evidence were assessed using the GRADE method and categorized into high, moderate, low, and very low (Supplementary Material 2) [11].
In step 5, members of the working group participated in meetings to determine and formalize the levels of evidence and strength of recommendations. The strength of recommendations was not solely based on the certainty of evidence; however, it was determined by comprehensively considering the balance of benefits and harms, patient values and preferences, cost-effectiveness and resource utilization, feasibility, and applicability in clinical practice. Expert opinions were integrated, if necessary, but only when evidence was insufficient and the working group deemed the recommendation essential. The strength of recommendations was divided as follows: A, B, C, D, and I (Supplementary Material 3). The determined levels of evidence and strength of recommendations for all key questions were distributed to the entire guideline committee via email, and the Delphi survey method was used to assess acceptance. An item was considered to have reached a final consensus if at least 70% of the entire committee agreed to its acceptance. For items on which less than 70% of the committee agreed, the Delphi survey was repeated until 70% or more agreement was achieved. Over 70% agreement was obtained for all 4 key items in the first round.
In step 6, the completed guideline was submitted to each society’s research committee for review and feedback. Modifications to recommendations were considered only when there was sufficient supporting evidence of an appropriate level. After incorporating evidence-based modifications in step 7, the guideline was submitted to each society for final review and approval.
RECOMMENDATIONS
1. Key Question 1
Is surgery (I) clinically more effective (O) than conservative treatment (C) for patients with low back and radicular pain due to LDH (P)?
Surgical treatment may be performed to relieve pain and improve functional outcomes in patients with low back or radicular pain due to LDH.
Recommendation: Grade B
Level of evidence: Moderate
1) Meta-analysis
(1) In patients with LDH, surgical treatment significantly reduced leg pain (visual analogue scale [VAS] leg) at 6 months postsurgery compared with conservative treatment (mean difference=-2.40; 95% confidence interval [CI], -4.58 to -0.22; p=0.03); however, after 2 years, surgical treatment showed a favorable trend; nevertheless, the difference was not statistically significant (mean difference=-0.95; 95% CI, -3.47 to 1.56; p=0.46), indicating it did not significantly differ from conservative treatment (Fig. 1A).
Forest plots showing the comparison of clinical and functional outcomes between the surgical and the nonsurgical treatments in patients with lumbar disc herniation. (A) Leg pain. (B) Back pain. (C) ODI. VAS, visual analogue scale; ODI, Oswestry Disability Index.
(2) In patients with LDH, surgical treatment significantly reduced back pain (VAS back) at 6 months postsurgery compared with conservative treatment (mean difference=-1.85; 95% Cl, -3.65 to -0.04; p<0.05); however, after 2 years, surgical treatment showed a favorable trend; nevertheless, the difference was not statistically significant (mean difference=-1.47; 95% CI, -3.01 to 0.06; p=0.06), indicating that it did not significantly differ from conservative treatment (Fig. 1B).
(3) In patients with LDH who underwent surgical treatment, there was no significant difference in their Oswestry Disability Index (ODI) compared with those who received conservative treatment at 6 months (mean difference=-1.76; 95% Cl, -4.09 to 0.57; p=0.14) and 2 years postsurgery (mean difference=-1.03; 95% Cl, -2.28 to 0.22; p=0.11) (Fig. 1C).
2) Evidence review
Despite insufficient prior research on the natural course of LDH, existing studies report that LDH generally improves over time with reduced pain and symptoms [12]. Most patients experience improvement within weeks to months, with many of them recovering through conservative treatment without surgery. The immune system is believed to be involved in the mechanism of this improvement because the immune system recognizes the herniated disc material as a foreign substance [13]. This triggers an inflammatory response, wherein immune cells such as macrophages break down and absorb the herniated disc material. Additionally, over time, the herniated disc material gradually dehydrates and shrinks, reducing in size naturally [14]. The damaged area around the disc undergoes a fibrosis process that stabilizes the disc and reduces pain. However, some patients may require surgery owing to persistent pain and neurological damage.
In this study, we adopted a literature search strategy to select 10 studies, which were ultimately used in the final evidence tables. All 10 selected studies were RCTs [12,15-23]. Reportedly, surgical discectomy provides faster and more effective pain relief than nonsurgical treatment [12,15,16,18-20,22]. According to the SPORT (Spine Patient Outcomes Research Trial) study, superior results for rapid pain relief and functional improvement were reported, although long-term outcomes showed no significant difference from nonsurgical treatment [16,18].
The results of this study were similar: among patients with LDH, those who underwent surgical treatment showed significantly reduced leg pain (VAS leg) at 6 months postsurgery compared with those who received conservative treatment. However, after 2 years, surgical treatment still showed a favorable trend; nonetheless, the difference was not statistically significant, indicating that it did not significantly differ from conservative treatment (Fig. 1A). For back pain, surgical treatment significantly reduced pain (VAS back) at 6 months compared with conservative treatment. However, at 2 years, the difference was not statistically significant, showing no difference from conservative treatment (Fig. 1B). Regarding the ODI, there was no significant difference between the surgical and conservative treatment groups at 6 months or 2 years in patients with LDH (Fig. 1C).
In 2 of the 10 RCTs, the same patient groups were evaluated; nevertheless, different follow-up periods were used [16,18]. Based on the effectiveness of surgical vs. conservative treatment for disc herniation, the limitations of RCT studies included nonstandardized conservative treatments and small study sizes. Functional assessment metrics varied among studies, with a different index, such as ODI, Roland, or 36-item Short Form health survey, used in each study, making it difficult to analyze a consistent data set. Therefore, the overall level of evidence for this clinical question was rated as ‘moderate.’
In most studies, the effects of surgical and conservative treatments were compared; nonetheless, detailed information on the side effects of each method was not provided in these studies. Generally, conservative treatments have minimal severe side effects or complications, while surgical treatments may involve risks such as infection or nerve damage, although these are rare [16,18-20].
LDH is known to improve naturally in many cases over time. Therefore, in most cases, conservative treatments such as physical and exercise therapies, medication, and nerve block injections should be initially implemented, as these options are relatively low-cost, noninvasive, and have minimal associated risks [17,21]. However, conservative treatment often requires a prolonged period, and frequent hospital visits may be necessary. Surgical treatment may have higher initial costs owing to hospitalization, preoperative testing, anesthesia, and the procedure itself, with a risk of complications such as infection or nerve damage [16,18-20]. Nonetheless, surgery is essential when neurological symptoms such as muscle weakness, paralysis, or bowel and bladder dysfunction occur owing to severe nerve compression. Surgery can provide immediate symptom relief in such cases.
The most important consideration should be a comprehensive assessment of each patient’s condition, pain severity, neurological status, insurance coverage, and the physician’s experience level performing surgical or conservative treatment. Based on these factors, the appropriate treatment approach should be selected accordingly.
2. Key Question 2
Is endoscopic surgical treatment (I) more clinically effective (O) than nonendoscopic surgical treatment (C) in patients with LDH (P)?
Endoscopic surgery can be considered for patients with LDH presenting with radicular leg pain.
Recommendation: Grade B
Level of Evidence: Low
1) Meta-analysis
(1) In patients with LDH who underwent surgical treatment, endoscopic surgery did not significantly influence leg pain (measured using the VAS) compared with nonendoscopic surgery, which was conducted immediately till 3 months (standard mean difference [SMD]=-0.26; 95% CI, -0.73 to 0.22; p=0.29) and at 6 months postsurgery (SMD=-0.06; 95% CI, -0.22 to 0.11; p=0.51). However, at 12 months, endoscopic surgery resulted in lower leg pain (SMD=-0.31; 95% CI, -0.46 to -0.15; p<0.05) (Fig. 2A).
Forest plots showing the comparison of clinical and functional outcomes between endoscopic surgery and conventional open surgery in patients with lumbar disc herniation. (A) Leg pain. (B) Back pain. (C) ODI. SD, standard deviation; CI, confidence interval; IV, inverse variance; VAS, visual analogue scale; ODI, Oswestry Disability Index.
(2) In patients with LDH who underwent surgical treatment, postoperative back pain (measured using the VAS) did not significantly differ after endoscopic and nonendoscopic procedures at any time point (immediate to 3 months [SMD=-0.31; 95% CI, -0.92 to 0.31; p=0.33], 6 months [SMD=-0.05; 95% CI, -0.62 to 0.52; p=0.86], and 12 months postoperatively [SMD=-0.10; 95% CI, -0.26 to 0.07; p=0.25]); nevertheless, the pain tended to be lower in endoscopic surgery (Fig. 2B).
(3) In patients with LDH who underwent surgical treatment, ODI did not significantly differ between endoscopic and nonendoscopic procedures at any time point (immediate to 3 months [SMD=-0.02; 95% CI, -0.50 to 0.46; p=0.93], 6 months [SMD=0.08; 95% CI, -0.55 to 0.71; p=0.80], and 12 months postoperatively [SMD=-0.09; 95% CI, -0.25 to 0.07; p=0.27]); nonetheless, the index tended to be lower in endoscopic surgery (Fig. 2C).
(4) In patients with LDH who underwent surgical treatment, operative time did not significantly differ between endoscopic and nonendoscopic procedures; however, the time tended to be shorter in endoscopic surgery. (SMD=-0.92; 95% CI, -2.21 to 0.37; p=0.16) (Fig. 3A).
Comparison of operative time (A) and reoperation (B) rates between endoscopic surgery and conventional open surgery in patients with lumbar disc herniation. SD, standard deviation; CI, confidence interval; IV, inverse variance.
(5) In patients with LDH who underwent surgical treatment, the reoperation rate did not significantly differ between endoscopic and nonendoscopic surgery (odds ratio, 1.09; 95% CI, 0.63–1.90; p=0.75) (Fig. 3B).
2) Evidence review
Endoscopic discectomy (ED) is a minimally invasive technique designed to reduce damage to normal tissue compared with open discectomy (nonendoscopic discectomy, NED) [24,25]. Similar to open surgery, the endoscopic approach allows full access to the herniated disc through direct visualization with an endoscope. The advantages of this method include reduced incision size and minimized soft tissue damage, which helps preserve normal tissue, potentially leading to faster postoperative pain relief and recovery [26]. The transforaminal endoscopic approach, in particular, requires less removal of laminar bone and ligamentum flavum, resulting in fewer anatomical changes. However, most surgeons are not familiar with this method because it is technically challenging and requires time to develop proficiency [27].
Since there is no definitive conclusion regarding which form of ED is more effective and safer than NED, the aim of this study was to systematically compare the efficacy and safety of endoscopic and NED in treating symptomatic LDH.
In total, 11 RCT studies were selected through the literature search strategy [24,28-37]. In patients with LDH who underwent surgical treatment, endoscopic surgery resulted in lower leg pain at 12 months postsurgery than that of nonendoscopic surgery. Additionally, back pain was lower at 6 months postsurgery, and ODI was statistically significantly lower from surgery conducted to 3 months. Endoscopic surgery had a shorter operation time than nonendoscopic surgery, and no significant difference was observed in reoperation rates between endoscopic and nonendoscopic surgery.
Gadjradj et al. [30] conducted a multicenter RCT in which they compared 179 patients who received percutaneous transforaminal ED (PTED) and 309 patients who received open microdiscectomy. They reported that at 12 months postsurgery, the PTED group had a statistically significantly lower VAS leg score (group difference of 7.1; 95% CI, 2.8–11.3) than the microdiscectomy group.
In addition, Park et al. [34] compared 32 patients who underwent biportal ED (biportal group) with 32 patients who underwent microscopic discectomy (microscopy group) and reported that at 12 months postsurgery, the ODI score was 11.97 in the microscopy group and 13.89 in the biportal group, demonstrating the noninferiority of the biportal group (mean difference, 1.92; 95% CI, -3.50 to 7.34). The patient-reported outcome measure, surgery-related outcomes, and adverse events showed similar results to the ODI.
Gibson et al. [31] conducted an RCT comparing 70 patients who underwent transforaminal ED (TED) with 70 patients who received microdiscectomy. They reported similar results in pain and disability improvement between the 2 groups. However, the TED group had a shorter average hospital stay than the microdiscectomy group (0.7 vs. 1.4 nights, p<0.001).
The 11 RCT studies had limitations, as they were conducted with a relatively small number of participants. Additionally, the meta-analysis was mainly based on the study by Gadjradj et al. [30], which had by far the largest sample size, potentially affecting the reliability of the findings. Considering the analysis of the literature and the limitations of the included studies, the overall level of evidence for this clinical question was rated as ‘low.’
Both endoscopic and nonendoscopic surgical treatments were effective for patients undergoing surgical treatment for LDH. For surgical outcomes, operation time, complications, and re-operation rates, the endoscopic surgery showed noninferiority compared with the nonendoscopic surgery. These findings reveal that endoscopic surgery may be considered for patients with appropriate indications, balancing its benefits and risks. In previous literature, open, minimally invasive, and endoscopic spinal surgeries were suggested as clinically effective and safe techniques that can yield favorable clinical outcomes.
There is a growing preference among healthcare providers and patients for less invasive procedures and shorter surgery times in patients requiring surgical treatment for LDH. The initial cost of endoscopic spinal surgery can be relatively high because of the need for specialized endoscopic equipment and tools; however, its minimally invasive nature is expected to reduce overall healthcare costs in the long term by shortening recovery times, decreasing complications, and reducing rehabilitation costs. Meanwhile, costs may vary depending on individual health conditions, insurance coverage, hospital policies, and the skill level of the performing surgeon.
3. Key Question 3
Is an epidural nerve block (I) clinically more effective (O) than conservative treatment (C) for patients with radicular pain due to LDH (P)?
Epidural nerve blocks are recommended for pain relief and functional improvement in patients with low back or radicular pain due to LDH.
Recommendation: Grade B
Level of evidence: Low
Condition: Epidural nerve block may be considered for patients who did not improve with general conservative treatment, have no contraindications to the procedure, and have been adequately informed about its effectiveness and potential complications. The procedure should be performed by a trained and experienced clinician in a facility equipped with appropriate medical equipment and environment.
1) Meta-analysis (Fig. 4)
Comparison of the effects on clinical and functional outcomes between epidural nerve block and conservative treatment in patients with lumbar disc herniation. SD, standard deviation; CI, confidence interval; IV, inverse variance.
Both the treatment and control groups showed a reduction in pain intensity following the epidural nerve block; however, at 6 months, there was no statistically significant difference in pain intensity compared with the control group (p=0.88). At 12 months, the epidural nerve block group showed a significantly greater reduction in pain intensity than the control group (mean difference=-1.54; 95% CI, -2.69 to -0.40; p<0.01). Additionally, functional scores assessed at 12 months postprocedure showed significantly greater improvement in the epidural nerve block group than in the control group (mean difference=-3.15; 95% CI, -4.99 to -1.31; p<0.001).
2) Evidence review
Three studies were included in the final evidence table, all of which were RCTs conducted in Germany [38], the Netherlands [39], and the United States [40]. These 3 studies were selected following the literature search strategy as illustrated in (Supplementary Material 1). In the German study, the intervention group received lumbar epidural nerve blocks [38], whereas, in the Dutch study, segmental epidural steroid injections or selective nerve root blocks were used [39]. Transforaminal epidural steroid injections were used in the United States study [40]. The control groups included general conservative treatments (such as medication and exercise, rehabilitation, and physical therapies) in 2 studies and trigger point injections in 1 study.
The patients included in these studies had relatively acute symptoms occurring within 2 months and were generally younger, with an average age of <50 years. It was likely that the aim was to exclude cases where low back or radicular pain was caused by aging-related degenerative changes. While random allocation was used in all 3 studies, only Spijker-Huiges et al. [39] implemented blinding for patients and practitioners [39]. In the other 2 studies, blinding was only employed in the evaluation of outcome variables [38,40]. Pain intensity was analyzed 6 and 12 months postprocedure, while functional outcomes were assessed at 12 months using the Roland-Morris Disability Questionnaire. The combined results are presented in Fig. 4.
Pain intensity decreased in the intervention and control groups following epidural nerve blocks; however, there was no statistically significant difference in pain intensity at 6 months compared with the control group (p=0.88). At 12 months, the epidural nerve block group showed significantly greater pain reduction than the control group (mean difference=-1.54; 95% CI, -2.69 to -0.40; p<0.01). Functional scores assessed 12 months postprocedure also showed significant improvement compared with the control group (mean difference=-3.15; 95% CI, -4.99 to -1.31; p<0.001). Complications related to the procedure were not reported in any of the studies included. The characteristics of the individual studies and the evidence levels of the included studies are summarized and presented, respectively, in the Supplementary Material 1.
From the analysis of the included studies, no statistically significant difference was found in pain intensity at 6 months between the epidural nerve block and control groups. Moreover, the sample size in these studies did not meet optimal requirements, and there was a risk of bias or uncertainty among the included studies. Therefore, the overall evidence level for this clinical question was rated as ‘low.’
In patients with low back or radicular pain due to LDH or protrusion, epidural nerve blocks significantly reduced pain compared with conservative treatment at 12 months postprocedure (-1.54; 95% CI, -2.69 to -0.40; p<0.01). Functional scores assessed using the Roland-Morris Disability Questionnaire at 12 months also significantly improved compared with the control group (-3.15; 95% CI, -4.99 to -1.31; p<0.001). While this evidence level is low, it suggests that epidural nerve blocks may reduce pain and improve function compared with conservative treatments in patients with low back or radicular pain due to LDH or protrusion. No complications were observed in any of the 3 studies included in the analysis.
Notably, epidural nerve blocks, particularly the transforaminal approach, might pose risks of severe complications [41]. Rare but serious complications of lumbar epidural nerve blocks include spinal cord infarction due to vascular embolism, nerve damage, epidural hematoma, and infections such as epidural abscess. The risk of vascular embolism is especially high when particulate steroids are used in the injectate; therefore, careful patient assessment and prudent choice of injectates are required [41,42]. The use of nonparticulate steroids is generally recommended to minimize these risks.
Additionally, the studies included in this analysis involved patients with symptom durations of <2 months. Compared with the control group receiving general conservative treatments, the pain reduction at 6 months was similar. Therefore, epidural nerve blocks should be considered for patients whose symptoms persist despite adequate conservative treatment.
4. Key Question 4
4-1. Is exercise (I) more clinically effective (O) compared with nonexercise (C) groups in patients with LDH (P)?
4-2. Is exercise (I) more clinically effective (O) than nonexercise (C) groups in patients with LDH who have undergone surgical treatment (P)?
4-1. Core muscle stabilization and strengthening exercises are recommended for patients with LDH undergoing conservative treatment.
Recommendation: Grade A
Level of evidence: Moderate
4-2. Core muscle stabilization and strengthening exercises are recommended for patients with LDH who have undergone surgical treatment.
Recommendation: Grade A
Level of evidence: Moderate
1) Meta-analysis
(1) In patients with LDH undergoing conservative treatment, core muscle stabilization and strengthening exercises helped reduce low back pain (measured using the VAS) (SMD=-1.25; 95% CI, -2.43 to -0.06; p<0.05) and improved patient function (measured by ODI) (SMD=-1.85; 95% CI, -2.21 to -1.48; p< 0.05). As measured using SMD, the absolute effect size was greater than 1, indicating a substantial treatment effect from exercise (Fig. 5A).
(A) Comparison of clinical and functional outcomes between exercise treatment and control groups in patients with lumbar disc herniation undergoing conservative treatment. (B1-3) Comparison of clinical and functional outcomes between exercise treatment group and control group in patients with lumbar disc herniation who underwent surgical treatment. SD, standard deviation; CI, confidence interval; IV, inverse variance; VAS, visual analogue scale; ODI, Oswestry Disability Index.
(2) In patients with LDH who underwent surgical treatment, core muscle stabilization and strengthening exercises did not show a significant effect on pain reduction (measured using the VAS). (SMD=-0.50; 95% CI, -1.19 to 0.18; p>0.05) However, within 3 months of starting exercises, significant functional improvement was observed (measured by ODI, Schober test, and Fingertip-floor distance) (SMD=-1.32; 95% CI, -2.06 to -0.59; p<0.05), (SMD=1.77; 95% CI, 1.07 to 2.46; p<0.05), (SMD=-2.00; 95% CI, -2.58 to -1.42; p<0.05). As measured using the SMD, the absolute effect size was >1, indicating a substantial treatment effect from exercise. However, between 4 to 12 months, ODI did not significantly improve from exercise when the exercise was compared with the nonexercise group (Fig. 5B1-3) (SMD=-0.41; 95% CI, -1.92 to 1.10; p>0.05).
2) Evidence review
Exercise therapy strengthens the spine-supporting muscles, enhances endurance and flexibility, and promotes pain reduction and functional recovery [43-53]. While the exact parameters for exercise—such as dosage, progression, and duration—have not been firmly established, side effects from exercise therapy are generally considered rare [43-53].
From a literature search, 11 studies (4 on conservative treatment [42,47,51,52] and 7 on surgical treatment [42,45,46,48-50,53]) were identified and were used in the final evidence table. Among the included studies, 10 were RCTs [43-49,51-53], and 1 was a prospective observational study [50].
In patients with LDH undergoing conservative treatment, core muscle stabilization and strengthening exercises helped reduce lower back pain (measured using the VAS) and improve function (measured using the ODI) [44,47,51,52]. The absolute effect size, measured using the SMD, was >1, indicating a substantial treatment effect from exercise.
In patients with LDH who underwent surgical treatment, core muscle stabilization and strengthening exercises did not significantly affect pain reduction (VAS) [43,45,46,48-50,53]. However, functional improvement was observed within 3 months of starting exercises (measured using the ODI, Schober test, and Fingertip-floor distance) [43,45,46,48-50,53]. The absolute effect size, as measured using the SMD, was >1, indicating a substantial treatment effect. However, ODI measured between 4- and 12-month postsurgery did not significantly improve compared with the nonexercise group.
The 10 RCTs and one prospective observational study had limitations owing to the relatively small sample sizes. In some analyses, only 2–3 studies were included, which may reduce their reliability. Given the limitations of the included studies and past literature analysis, the overall evidence level for this clinical question was rated as ‘moderate.’
Considering that exercise therapy carries minimal risk to patients, it should be actively encouraged [43-53]. However, if specific exercises or postures cause pain, those exercises or positions should be avoided. While exercise therapy in clinical settings is often supervised by therapists and may incur costs, various video resources are now available to guide exercise methods [54]. Furthermore, providing 1 or 2 sessions of exercise education at clinics, followed by independent exercise by patients, could minimize financial burdens.
DISCUSSION
The NASS guideline [6] and the Clinical Practice Guideline from the American College of Physicians [55] were developed quite some time ago. Since their publication, new treatment techniques and approaches, such as endoscopic surgical procedures, have become increasingly popular and have demonstrated favorable outcomes. As such, there is a growing need to incorporate recent clinical evidence that was not reflected in the existing guidelines. Moreover, the necessity for clinical practice guidelines that reflect Korea’s unique healthcare system and clinical environment has been increasingly emphasized. Korea operates a National Health Insurance (NHI) system in which insurance coverage criteria significantly influence treatment selection. Korean medical institutions have relatively high accessibility to advanced imaging technologies such as magnetic resonance imaging; nonsurgical treatments like physical therapy and injection therapy are commonly used. Korean patients tend to expect rapid symptom relief and generally prefer nonsurgical options before considering surgical intervention. In contrast to some countries where surgery may be limited due to resource or insurance constraints, Korea has relatively high accessibility to surgical treatments.
Although comparing this guideline directly with international ones is challenging, reviewing previous guidelines revealed similarities and differences. Key question 1 of this guideline, which addresses surgical treatment, is comparable to questions 22 and 25 in the NASS guideline. Both guidelines assign a recommendation grade B, although the level of evidence differs slightly. However, the overall content remains relatively consistent. Key question 2 reflects recent research findings on endoscopic surgery and upgrades the recommendation grade for endoscopic procedures to grade B, which is higher than that in the NASS guideline. Key question 3 assigns a lower recommendation grade for epidural steroid injections (ESIs) compared to the NASS guideline, although direct comparison is difficult due to methodological differences. Key question 4 assigns a recommendation grade A for exercise therapy, which is higher than the grade given in the NASS guideline. Due to a lack of highquality evidence, the NASS guideline cautiously suggests structured exercise as an option for patients with mild to moderate symptoms. This guideline more strongly recommends exercise therapy. This is based on the assessment that it poses minimal risk to patients and is less costly than other interventions such as injections or surgery. In conclusion, while this guideline shares many similarities with the NASS guideline, it reflects recent research findings, particularly regarding endoscopic surgery, and considers the Korean healthcare context. This has led to differences in some recommendation grades. However, due to different evidence grading systems and the lack of exact overlap in key questions, direct comparisons between the 2 guidelines are limited.
It is also not easy to compare this guideline directly with the American Family Medicine guideline (Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: A Clinical Practice Guideline From the American College of Physicians). However, key question 4, which recommends exercise therapy, is similar to recommendations 1 and 2 of the American College of Physicians guideline. Both guidelines issue a strong recommendation, indicating that the benefits of the treatment outweigh the risks or burdens. This clinical practice guideline emphasizes the importance of incorporating patient preferences, insurance coverage criteria, and resource availability in developing treatment strategies for patients with LDH and radiculopathy. This guideline enhances practical feasibility and economic justification by focusing on treatments realistically applicable to Korea’s NHI system. The prioritization of nonsurgical interventions, integration of the latest evidence for endoscopic procedures, and encouragement of physical therapy and exercise-based approaches are expected to improve patients’ quality of life and promote more efficient use of healthcare resources. This approach may also help establish standardized care for low back pain across primary care settings (Table 1).
Comparison with relevant guidelines
This clinical practice guideline has several limitations. The meta-analyses used in this guideline are based on previously published individual studies. These studies often apply strict inclusion and exclusion criteria, which may exclude patients with comorbidities or atypical symptoms. Therefore, there may be limitations in using the recommendations for diverse populations in clinical practice, such as older adults or patients with chronic illnesses. Patients in these studies may have been recruited from specific settings, such as tertiary hospitals or specialty centers, which may differ from typical primary or secondary care environments.
This study also had limitations in conducting sensitivity analyses and subgroup analyses. Although subgroup analyses are essential for identifying potential differences in treatment effects, they require sufficient data for each subgroup. However, many of the studies in this guideline did not consistently provide stratified data by factors such as age, symptom duration, severity of disc herniation, or surgical technique. As a result, conducting robust subgroup analyses was difficult. Moreover, post hoc subgroup analyses based on small sample sizes may increase the risk of type I error and reduce statistical power, leading to less reliable results.
FUTURE DIRECTIONS
If cost-effectiveness analyses are conducted in the future and more concrete economic data are secured, they could be used to support health insurance coverage decisions or further refinement of clinical practice guidelines. Despite evidence supporting the short-term effectiveness of some interventions, data on long-term outcomes and recurrence rates are limited. Future prospective studies with long-term follow-up will be needed. In response to the rapidly changing healthcare landscape, it is essential to continuously monitor clinical outcomes, analyze health insurance claims data, and incorporate feedback from clinicians and stakeholders to evaluate the guideline’s feasibility, adherence, and effectiveness in real-world settings. This process will help maintain the relevance and appropriateness of the guideline over time.
CONCLUSION
The aim of these clinical practice guidelines was to enhance treatment outcomes, reduce healthcare costs, and promote public health. However, recognizing limitations in domestic data and the dynamic healthcare circumstances, a multifaceted evaluation strategy will be implemented, including clinical outcome monitoring, analysis of healthcare utilization data, and feedback from clinicians and stakeholders to assess the feasibility, adherence, and effectiveness of these guidelines in real-world practice. To maintain the relevance of these guidelines, we will conduct comprehensive reviews every 5 years while also incorporating significant developments such as new clinical trials, changes in international guidelines, and advancements in treatment technologies. Future updating of the guidelines will refine them to improve their quality and applicability in clinical practice.
Supplementary Materials
Supplementary Materials 1-3 are available at https://doi.org/10.14245/ns.2550094.047.
Search and Selection of Evidence
Level of evidence
Grade of recommendation
Notes
Conflict of Interest
The authors have nothing to disclose.
Funding/Support
This work was supported by a grant from the 2022, 2023 Korean Spinal Neurosurgery Society research fund.
Author Contribution
Conceptualization: DAS; Data curation: JJL, MCC, HYP, SSC, SWC; Formal analysis: JJL, MCC, SSC, SWC; Funding acquisition: DAS, JHP, ISK, JKL, CKC, SHY; Methodology: MC; Project administration: DAS, JHP; Writing – original draft: JJL, MCC, SSC, SWC; Writing – review & editing: JJL, JHP, MC.