Neuropathic pain and sarcopenia based on skeletal muscle mass assessment following lumbar spinal stenosis surgery: a prospective observational study in Japan
Article information
Abstract
Study Design
Prospective cohort study.
Purpose
To investigate the relationship between neuropathic pain (NeP) and skeletal muscle mass following lumbar spinal stenosis (LSS) surgery.
Overview of Literature
Sarcopenia generally worsens clinical outcomes in degenerative diseases. Although a high prevalence of sarcopenia is observed in patients with LSS, it is reported that sarcopenia does not significantly affect the clinical symptoms of lumbar degenerative diseases; thus, its role must be reconsidered in LSS alongside other factors such as spinal alignment and degree of stenosis.
Methods
We prospectively evaluated improvements in leg pain and lower limb skeletal muscle mass in 421 patients who underwent surgical treatment for LSS. Bilateral limb muscle mass was compared in 260 patients with unilateral leg pain. The association between postoperative muscle mass increase and improvement in leg pain (Visual Analog Scale [VAS] score <3) was assessed using multivariate logistic regression analysis.
Results
Preoperatively, the affected limb exhibited significantly reduced muscle mass compared to the unaffected limb. Postoperatively, the affected side gained muscle mass, which was significantly greater than the unaffected side. Leg pain VAS score <3 and the epidural sac area were significantly associated with increased muscle mass postoperatively. Patients with postoperative improvement in leg pain showed a significant increase in muscle mass compared to those without improvement.
Conclusions
In unilateral LSS, the affected limb undergoes significant muscle mass loss, which improves postoperatively with rehabilitation. The increase in muscle mass correlates with improvements in leg pain, indicating a potential relationship between NeP and sarcopenia.
Introduction
Neuropathic pain (NeP) is one of the most incapacitating types of pain caused by primary lesions or diseases of the somatosensory system. According to the International Association for the Study of Pain, NeP affects 6.9%–10% of the global population, significantly diminishing an individual’s health and quality of life (QoL) and putting immense strain on their families and healthcare resources [1]. Classical etiologies of peripheral NeP include painful peripheral neuropathies, postherpetic neuralgia, and traumatic nerve injuries. Many patients present with mixed pain syndromes involving both neuropathic and nonneuropathic mechanisms, such as lumbar or cervical spondylotic radiculopathy, which represent a frequent cause of peripheral NeP in the general population. The first-line pharmacological treatment for NeP includes gabapentinoids, which interact with α2δ subunits of voltage-gated calcium channels; however, the economic burden and side effects of gabapentinoid use on the patient’s daily life are not commensurate with treatment effectiveness [2], prompting the need for developing new therapies.
Lumbar spinal stenosis (LSS) is a condition characterized by degenerative changes in the lumbar spine. Degenerative thickening of the spinal joints and ligaments in LSS increases pressure on the neurovascular structures within the spinal canal, which may cause compression of the exiting nerve roots or the cauda equina, leading to neural ischemia, nerve damage, and NeP development. In LSS, the leg pain component is caused through the NeP mechanisms, whereas back pain is caused by nociceptive mechanisms; approximately 20% of patients with low back pain in LSS have an NeP component [3]. As a result, LSS is the most common reason for lumbar spine surgery in middle-aged to older populations. Although generally favorable results are anticipated from surgery, some studies show increased adverse events and reoperation [4]. To address the shortcomings associated with surgical treatments, a comprehensive examination of factors influencing the chances of postoperative NeP improvement is essential. Predicting the likelihood of postoperative recovery is challenging; therefore, other causes should be identified before surgery [5].
Park et al. [6] reported a high prevalence of sarcopenia in patients with LSS. Interestingly, a recent systematic review and meta-analysis concluded that sarcopenia does not significantly influence the symptoms of lumbar degenerative disease and is unlikely to act as the sole moderator of leg pain intensity in individuals with LSS; it is rather associated with worse clinical outcomes, such as QoL [7]. However, it is noteworthy that this systematic review and meta-analysis included only two studies involving participants without lumbar pathology as controls, and matched controls likely shared risk factors, like sedentary lifestyle and age-related comorbidities, resulting in no significant differences in sarcopenia prevalence between groups with and without LSS [7]. Furthermore, clinical symptoms of LSS, such as pain, are influenced by multiple factors, including spinal alignment and the degree of stenosis, suggesting that the conclusions from two-group comparisons may not be adequately representative. Therefore, a longitudinal analysis of NeP and an imaging-based objective quantification of skeletal muscle mass is essential to elucidate this association.
In this study, we aimed to assess the cross-sectional relationship between NeP and skeletal muscle mass by comparing the lower limb skeletal muscle mass between the affected and unaffected sides in patients with unilateral lower limb symptoms in LSS. Additionally, we aimed to examine causality through a prospective longitudinal analysis of NeP improvement and skeletal muscle mass changes following treatment.
Materials and Methods
Ethical considerations
The prospective cohort study protocol was approved by the Institutional Review Board of the National Center for Geriatrics and Gerontology (approval number: 2021). This study adhered to the principles of the Declaration of Helsinki. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed throughout the study. All participants were informed of the study objectives and provided written informed consent prior to enrollment.
Study design and participants
For this prospective cohort study, we screened 524 nonparalyzed patients treated surgically for LSS between April 2018 and March 2023. After excluding four cases of death within 1 year of surgery, 20 of reoperation, five of postoperative infection, and 74 of missed follow-up, a total of 421 patients (mean age=74.68±5.60 years; 231 males, 190 females) were included in the study. These 421 patients were prospectively tracked for lower limb skeletal muscle mass to evaluate the causal relationship between NeP and skeletal muscle mass (Fig. 1).
Schematic diagram illustrating enrollment of patients with lumbar spinal stenosis (LSS) and 1-year postoperative follow-up. This study comprises a prospective analysis of 524 cases involving surgical treatment for LSS and a longitudinal analysis of 421 patients who underwent 1-year postoperative follow-up. The final analysis included 260 patients with LSS and clear unilateral symptoms. PLIF, posterior lumbar interbody fusion.
For most cases, a posterior lumbar interbody fusion (PLIF) was performed for cases with vertebral slippage and/or segmental instability; in the absence of these pathologies, a partial laminectomy with ligamentum flavum resection for decompression was done. Adequate decompression of the affected intervertebral space was achieved in both procedures. The patient was usually able to ambulate on the second postoperative day and underwent lower limb muscle strengthening and gait training as part of a standard postoperative rehabilitation program.
Two board-certified spinal surgeons diagnosed NeP and unilateral LSS based on symptoms of neurogenic claudication, regardless of the presence of low back pain, magnetic resonance imaging (MRI) findings, neurological examination, and bilateral or unilateral lower limb neurological symptoms. NeP was diagnosed according to the diagnostic algorithm established by the International Association for the Study of Pain [8], which states that the pain distribution should be neuroanatomically plausible, and objective diagnostic tests should be able to confirm the lesion or disease as corresponding to the somatosensory nervous system. For unilateral LSS, skeletal muscle mass was compared between the affected and unaffected sides. Exclusion criteria were as follows: (1) previous lumbar surgery; (2) inability to walk; (3) patients with lower limb muscle strength of grade ≤4; (4) gabapentin use; (5) vertebral fractures; (6) systemic diseases, such as rheumatoid arthritis, infections, and malignancies; and (7) dementia.
Physical findings
In the physical examination, walking ability (independent or requiring assistance), leg and lower back pain (using the Visual Analog Scale [VAS]), and pre-LSS independence in activities of daily living (ADLs) (using the Barthel index [9]) were evaluated. In unilateral LSS, the affected sides were identified, and LSS severity was evaluated using the Japanese Orthopaedic Surgery Association score [10]. Health and clinical outcomes were assessed using the EuroQol 5-Dimension questionnaire [11], the Geriatric Depression Scale [12], and the Central Sensitization Inventory [13].
Patients were followed up for 1 year; based on the cutoff of VAS ≤4, which indicates pain-free status [14], patients were categorized into the improvement or nonimprovement groups if they had VAS scores of <3 or ≥3, respectively.
Blood sample analysis
Venous blood samples were obtained from patients using standard venipuncture techniques before surgery. The samples were used to assess complete blood cell counts and albumin, total cholesterol, hemoglobin A1c, fasting insulin, fasting plasma glucose, D-dimer, and vitamin D (25-hydroxyvitamin D [25(OH)D]) levels in the peripheral blood. Serum 25(OH)D levels were determined using electrochemiluminescence immunoassay. Insulin resistance was calculated using the standard formula, which is the homeostatic model assessment for insulin resistance [15].
Radiographic findings
Lateral radiographs of the lumbar spine were obtained with the participants standing naturally, with forearms crossed, and hands on the chest. Spinopelvic parameters of lumbar lordosis (L1–S1), sacral tilt, lumbar range of motion (L1–S1), sagittal vertical axis, thoracic kyphosis, pelvic tilt, and pelvic incidence minus LL were evaluated on whole-spine standing lateral radiographs. Radiographic assessments were performed by two board-certified orthopedic surgeons in Japan.
MRI findings
To evaluate LSS severity, the cross-sectional area of the dural sac from L1/2 to L5/S1 was measured on T2 sagittal images using the SYNAPSE image analysis software (Fujifilm Medical, Tokyo, Japan). The minimum dural sac area was used as the target for evaluating the narrowest spinal level.
Body composition analysis
We used dual-energy X-ray absorptiometry (Lunar iDXA; GE Healthcare, Chicago, IL, USA) before surgery and 1-year post-surgery to assess body composition. Bone mineral density was evaluated in g/cm2 and T-score for the lumbar vertebrae (L2–L4) and femoral neck. Skeletal muscle mass of the extremities was measured separately for bilateral lower limbs, and the skeletal muscle mass index was calculated by dividing the muscle mass of the upper and lower limbs by the square of the height [16]. Body fat was calculated as the percentage of total body fat divided by body weight.
Statistical analysis
Power analyses were performed using G*Power ver. 3.1.9.7 (Heinrich-Heine-Universität, Düsseldorf, Germany). A linear multiple regression was performed using F-testing with fixed mode and R2 deviation from zero (effect size=0.05, α error probability=0.05, power [1–β error probability]=0.9, and number of predictors=10); accordingly, the minimum sample size required was calculated as 420 patients (actual power=0.9001023).
Statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan). The Kolmogorov-Smirnov test was used for normality testing. Proportions and means (with standard deviations) were calculated for covariates and demographic information as applicable. Categorical variables were expressed as frequencies or percentages, and compared using chi-square or Fisher’s exact tests. Mean values were compared using an independent t-test. A p-value of <0.05 was considered statistically significant. The muscle mass between the unaffected and affected sides in unilateral LSS was compared using a paired t-test.
A multivariate logistic regression analysis was used to detect the risk factors for muscle mass reduction 1-year postoperatively. In addition to the factors that showed a significant difference (p<0.05) between postoperative change in muscle mass, we introduced age, sex, and factors considered clinically relevant to changes in lower limb skeletal muscle mass. Multicollinearity was examined by calculating the variance inflation factor (VIF); by convention, a VIF value of >5 suggests the presence of multicollinearity.
Results
Among the 421 patients with LSS, 260 had unilateral symptoms, and 161 had bilateral. Surgical treatment included PLIF in 180 patients (64 bilateral and 97 unilateral) and decompression in 241 patients (97 bilateral and 144 unilateral). Table 1 presents a comparison of clinicodemographic characteristics between LSS patients with unilateral versus bilateral symptoms. Except for a significant reduction in the MRI-detected dural canal cross-sectional area, compared with bilateral LSS, the unilateral symptom group showed no significant differences in terms of body composition (including skeletal muscle mass), spinal alignment, preonset ADL, clinical symptoms, ADL at onset, or QOL. Since patients with unilateral LSS symptoms showed no significant clinical differences compared to bilateral cases, the unilateral group (n=260) was subjected to further analysis regarding NeP and lower limb skeletal muscle mass.
Comparison between bilateral and unilateral LSS
Comparison of skeletal muscle mass between the affected and unaffected sides
In terms of preoperative skeletal muscle mass of the bilateral upper and lower limbs for patients with unilateral symptoms, the mean upper limb muscle mass was 2,116.4 g on the affected side and 2,120.4 g on the unaffected side, with no statistically significant difference. In comparison, the mean lower limb muscle mass was 5,917.1 g on the affected side, which was significant lesser compared to 6,001.3 g on the unaffected side (p<0.05) (Fig. 2). When we compared preoperative and 1-year postoperative lower limb skeletal muscle mass, although the differences were not statistically significant, there was a tendency toward increased muscle mass in the affected side (p=0.6296), whereas no change was observed on the unaffected side. The difference in change in skeletal muscle mass between the affected and unaffected sides was statistically significant (p<0.05) (Fig. 3).
Comparison of muscle mass in upper and lower extremity between affected and contralateral limbs. A comparison of skeletal muscle mass between the affected and unaffected sides showed no significant difference in the upper limb muscle mass, whereas the lower limb muscle mass exhibited a significant reduction on the affected side. a)By unpaired t-test.
Comparison of pre- and postoperative muscle mass in lower extremity between affected and contralateral limbs. Comparison of the lower limb skeletal muscle mass before and after surgery showed no significant differences. However, there was a tendency toward increased muscle mass on the affected side, whereas no change was observed on the unaffected side. A significant difference was found between the changes on the affected and unaffected sides (p<0.0001). a)By paired t-test. b)By repeated measure analysis of variance.
Factors affecting the change in postoperative lower limb skeletal muscle mass
To assess the factors affecting change in postoperative skeletal muscle mass, we calculated the difference in preoperative versus postoperative skeletal muscle mass for the entire cohort (n=421). The data were normally distributed (Kolmogorov-Smirnov test, p=0.177) with a mean of 0.967±819.7 g and a median of 34.5 g (−2,979.0 to 2,872.0 g) (Appendix 1). Accordingly, patients demonstrating a postoperative skeletal muscle mass change of >0 were grouped as the muscle mass gain group and those with a change value of <0 as the muscle mass loss group (Kolmogorov-Smirnov test) (Appendix 1). Table 2 shows a comparison between the lower limb muscle mass gain and loss groups.
Frequency distribution of differences in lower limb skeletal muscle mass before and after lumbar spine surgery. SD, standard deviation.
Comparison between postoperative increase and decrease in lower limb muscle mass
Differences between the two groups were statistically significant only for the postoperative lower limb VAS score and the frequency of postoperative lower limb pain VAS score of <3. The muscle mass gain group showed a reduction in postoperative leg pain VAS scores. Thus, we added the following explanatory variables considered clinically relevant to skeletal muscle mass increase in the subsequent multivariate logistic regression analysis: duration of illness, preonset Barthel index, vitamin D levels, insulin resistance, dural sac area, and body fat percentage. Additionally, the lower limb pain VAS score of <3, which showed a significant difference between the groups, and age and sex were included. Of these variables, a lower limb pain VAS score of <3 and the dural sac area were found to be significantly associated with increased lower limb skeletal muscle mass after LSS surgery (Table 3).
Logistic regression analysis for the risk of postoperative sarcopenia
Association between improvement in postoperative leg pain and increased lower limb muscle mass
In the 260 cases with unilateral LSS, postoperative changes in lower limb muscle mass were compared between two groups: postoperative leg pain VAS <3 (n=177) and VAS ≥3 (n=83). In patients with postoperative leg pain VAS <3, a significant increase in lower limb muscle mass was observed postoperatively in both the affected and unaffected limbs. Furthermore, in this group, a comparison between the affected and unaffected limbs revealed a significant increase in muscle mass in the affected limb compared to that in the unaffected limb (Fig. 4).
(A–C) Changes in skeletal muscle mass of the lower limbs according to postoperative lower limb pain improvement. Changes in lower-limb skeletal muscle mass before and after surgery in 260 patients with unilateral lumbar spinal stenosis (LSS). In 177 patients whose postoperative lower limb pain improved to Visual Analog Scale (VAS) <3, a significant increase in lower limb muscle mass was observed in both affected and unaffected limbs. The increase in postoperative lower-limb skeletal muscle mass was significantly greater in the affected limb than in the unaffected limb. a)By paired t-test. b)By repeated measure analysis of variance.
Discussion
Patients with chronic musculoskeletal pain with neuropathic components generally report higher pain intensity, whereas physically active individuals exhibit lower chronic pain incidence [17]. A previous study on the pathophysiology of chronic pain reported that increasing muscle pain leads to decreased motor unit firing rates [18]. However, several physiological mechanisms, including reductions in supraspinal excitatory drive, α-motoneuron excitability, motor unit recruitment, and rate coding, have been proposed to explain age-related loss of muscle strength [19]. These findings suggest potential connections between NeP, skeletal muscle loss, and muscle weakness with aging [20]. In this context, the present study demonstrated a cross-sectional association between NeP and lower limb skeletal muscle mass by longitudinally comparing the unaffected and affected lower limbs of patients with unilateral LSS.
Skeletal muscle mass is a crucial component for sarcopenia diagnosis and requires body size adjustments. The Asian Working Group of Sarcopenia 2025 consensus update recommends adjusting the skeletal muscle mass measurements by dividing them by height squared or body mass index [21]. However, the impact of this adjusted definition on the decline in ADL ability of older adults remains unclear. Otsuka et al. [22] longitudinally analyzed appropriate methods for adjusting skeletal muscle mass in older Japanese adults for predicting mortality and disability, and concluded that unadjusted appendicular lean mass is the primary variable measurable regardless of sex. This also supports the fact that comparing skeletal muscle mass between individuals is challenging due to body size differences. Given the current difficulty in defining the optimal skeletal muscle mass for ADL independence, using the unilateral LSS framework provides a suitable model for evaluating the relationship between NeP and skeletal muscle mass while avoiding body size adjustments using the nonpainful side comparison method. Based on this model, our results revealed that NeP reduced skeletal muscle mass in the lower limbs.
Despite many studies reporting positive findings regarding the relationship between NeP and skeletal muscle mass, pathophysiological mechanisms underlying its causality have not been elucidated [8,20]. It is known that age-related changes in peripheral nerves and neuromuscular junctions elicit a partial denervation response within motor units [23]. In addition, the gradual decline in axonal sprouting capacity in old age leads to the progressive loss of motor end plates, accelerating the process of muscle fiber loss [24]. However, a peripheral nerve injury activates microglia in the dorsal horn, releasing inflammatory and proinflammatory mediators, such as interleukin (IL)-1β, IL-6, and tumor necrosis factor (TNF)-α, that sensitize neurons, leading to NeP [25,26]. Spinal microglia further promote NeP by releasing several glial mediators that sensitize spinal neurons [26]. Presumably, this inflammatory signal from microglia acts on spinal dorsal horn neurons, increasing excitatory synaptic transmission and contributing to pain generation. Based on this mechanism, the concept of “microgliopathic pain” has been proposed, which describes a pain syndrome mediated by dysfunctional microglia [27]. Although senescence and NeP represent distinct pathologies, the effects of neuroinflammation on skeletal muscles may serve as a suitable foundation to explain sarcopenia in both pathologies. Notably, Imagama et al. [28] reported that skeletal muscle loss increases the risk of NeP based on a 5-year study of healthy older adults, aligning with the geriatric pain hypothesis, which suggests that chronic inflammation seen with aging underlies the pathogenesis of chronic pain and sarcopenia [29].
While clinical evidence linking muscle loss to NeP remains inconclusive, the findings of the present study demonstrate that surgical treatment of LSS increases lower limb muscle mass through pain alleviation, and highlight the need to address sarcopenia in NeP treatment. While the results of this study hint at a potential correlation between increased postoperative lower-limb muscle mass and improved lower-limb pain in LSS, the exact influence of maintaining or enhancing skeletal muscle mass to accelerate NeP improvement following LSS treatment must be examined. Nevertheless, the findings of this study suggest that NeP and sarcopenia may be causally related; consequently, modifying conventional approaches, such as exercise therapy and nutritional guidance, targeting skeletal muscles, could serve as important modalities in the treatment of NeP. Conversely, developing pain management strategies targeting peripheral nerves, neuromuscular junctions, or the dorsal horn of the spinal cord may prove useful for sarcopenia treatment.
While a significant strength of this study is its longitudinal assessment of skeletal muscle mass, enabling the identification of a possible causal relationship between NeP improvement and increased muscle mass. However, the evaluation period of 1 year may be relatively short and may not adequately reflect age-related changes in muscle mass. The sample size in this study provided sufficient statistical power for detection; however, analyzing larger changes in skeletal muscle mass that could potentially influence ADL independence in older adults would require a larger sample size and reassessment of skeletal muscle mass affecting ADL decline. Furthermore, this study analyzed postoperative outcomes and muscle mass changes at a single institution, potentially introducing selection bias regarding LSS treatment methods and data collection. As this was not a randomized controlled prospective trial, the choice of surgical procedure reflects the attending surgeon’s preference. Another limitation was that the standard screening tools for NeP were not used during the diagnostic process for LSS, and the possibility of coexisting nociceptive pain could not be completely excluded. Given that the study included different surgical procedures, such as decompression and PLIF, the differences in postoperative mobility, skeletal muscle recovery, activity levels, and nutritional status resulting from the degree of invasiveness in each surgery may influence postoperative skeletal muscle mass. Regarding postoperative changes in skeletal muscle mass, we did not assess osteoporosis medication use [30], which is known to affect skeletal muscle mass; hence, the interpretation of the results concerning changes in lower limb skeletal muscle mass is limited. Finally, it is important to acknowledge that this study included patients who experienced difficulty walking or maintaining a seated posture due to leg pain; consequently, the inability to fulfill the diagnostic criteria for sarcopenia, such as grip strength and walking speed, was a significant limitation of the study.
Conclusions
This study provides strong cross-sectional evidence demonstrating skeletal muscle loss due to NeP by comparing skeletal muscle mass in the unaffected and affected lower limbs of patients with unilateral LSS, with particularly reduced skeletal muscle mass in the affected limb. Improvement in leg pain was associated with increased lower limb muscle mass following LSS surgery. By longitudinally proving that leg pain improvement associated with LSS surgery leads to increased lower limb skeletal muscle mass, we verified the relationship between NeP and sarcopenia.
Key Points
Patients with lumbar spinal stenosis presenting with unilateral neuropathic pain experience loss of skeletal muscle mass on the affected side.
Postoperatively, the affected side significantly regains skeletal muscle mass, accompanied by improvement in leg pain, indicating a potential relationship between neuropathic pain and sarcopenia.
The increase in skeletal muscle mass is greater with greater improvement in pain, indicating that treatment strategies targeting the skeletal muscle may effectively address pain.
Data Availability
The data that support the findings of this study are not publicly available as they contained information that could compromise the privacy of the research participants; however, they are available from the corresponding author (Yoshihito Sakai) upon reasonable request.
Notes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Declaration of Generative AI in Scientific Writing
During the preparation of this work, the corresponding author, whose native language is Japanese, used DeepL Translate to assist in translating short passages of the text. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Acknowledgments
We express our sincere gratitude to all the patients who participated in the study, the staff who delivered the intervention, and the surgeons who assisted with patient recruitment. We appreciate the administrative support provided by Junko Suzuki, Miki Morita, and Reina Yamamoto.
Funding
This study was supported by the National Center for Geriatrics and Gerontology (26–10). The funders were not involved in the design or conduct of the study; collection, management, analysis, or interpretation of the data; or in the preparation, review, or approval of the manuscript. The funders were not involved in the design or conduct of the study; collection, management, analysis, or interpretation of the data; or in the preparation, review, or approval of the manuscript.
Author Contributions
YS and KU conceived the idea for the work, designed the study, interpreted the data, and wrote the final version of the article. YS was involved in data analysis and data management. TW, HM, RS, YA, YT, and TT collected data, prepared all tables and figures, and revised the manuscript. All authors provided input into the editing of the manuscript for publication. The corresponding author had full access to all data in the study and final responsibility for publication.