Preoperative erector spinae muscle quality is more important than leg muscle quality for postoperative walking recovery in lumbar spinal stenosis: a retrospective study in Japanese patients

Article information

Asian Spine J. 2026;.asj.2025.0537

Publication date (electronic) : 2026 March 16

doi : https://doi.org/10.31616/asj.2025.0537

Tomoyoshi Sakaguchi¹, Masato Tanaka², Mandar Borde², Jein Boon Chen², Shinya Arataki², Tadashi Komatsubara², Akiyoshi Miyamoto², Masato Tanaka³, Kazuhiko Takamatsu¹, Yosuke Yasuda¹, Adriano Doană-Prodan¹

¹Department of Rehabilitation, Okayama Rosai Hospital, Okayama, Japan

²Department of Orthopaedic Surgery, Okayama Rosai Hospital, Okayama, Japan

³Department of Radiology, Okayama Rosai Hospital, Okayama, Japan

Corresponding author: Masato Tanaka, Department of Orthopaedic Surgery, Okayama Rosai Hospital, 1-10-25 Chikkomidorimachi, Minami Ward, Okayama City, Okayama, Japan, Tel: +81-86-262-0131, Fax: +81-86-262-3391, E-mail: tanaka0896@gmail.com

Received 2025 September 2; Revised 2025 October 16; Accepted 2025 October 19.

Abstract

Introduction

Lumbar spinal stenosis (LSS) is a prevalent degenerative spinal disorder characterized by narrowing of the spinal canal, which leads to compression of neural elements [1]. This condition often results in symptoms, such as low back pain, leg pain, and neurogenic intermittent claudication, significantly impairing patients’ mobility and quality of life [2]. Patients with severe LSS symptoms may require surgical interventions, such as simple decompression or decompression and fusion [3]. The Zurich Claudication Questionnaire (ZCQ) and the 6-minute walk distance (6MWD) test are among the most commonly used clinical evaluation tools for assessing the functional status of LSS [4–6].

The 6MWD is a reliable measure of walking endurance; however, it is crucial to determine whether observed changes are clinically meaningful. The minimal clinically important difference (MCID) represents the smallest change considered meaningful by patients and sufficient to influence clinical decision-making [7]. Achievement of the MCID in the 6MWD following spinal surgery is regarded as a key indicator of functional improvement. In patients with LSS, MCID values for the 6MWD have been reported across various postoperative time points [6,8]. Nevertheless, not all patients attain clinically significant improvements in walking capacity, and the factors influencing the achievement of the MCID have yet to be fully elucidated.

Preoperative muscle quality has been reported to be a significant predictor of postoperative physical function, with lower muscle quality being associated with poor recovery and an increased risk of disability [9,10]. Recent systematic reviews have reported that computed tomography (CT)–based assessments of paraspinal muscle morphology can predict surgical outcomes in patients with LSS [11,12]. Evaluation of the quality of muscles, particularly the trunk and lower limb muscles, has been shown to correlate with clinical outcomes and the progression of disability [12,13]. Despite accumulating evidence on muscle quality and clinical outcomes, the relationship between specific muscle quality parameters and the achievement of the MCID in the 6MWD among patients with LSS remains unclear. A deeper understanding of this relationship may aid in identifying patients at risk for suboptimal recovery and could inform tailored rehabilitation or surgical planning.

Therefore, this study aimed to determine whether preoperative trunk and lower limb muscle quality, assessed by CT, predicts the achievement of the MCID in the 6MWD after LSS surgery.

Materials and Methods

Study population

Consecutive patients who underwent decompression or fusion surgery between April 2021 and April 2023 were assessed for eligibility (n=183). The inclusion criteria were as follows: (1) age ≥60 years; (2) presence of intermittent claudication due to leg pain and/or numbness; (3) availability of patient-reported outcomes using the ZCQ; and (4) provision of written informed consent. The exclusion criteria were as follows: (1) cervical or thoracic spinal lesions; (2) severe osteoarthritis of the hip or knee; (3) history of neurological disease, pulmonary disease, cardiac disease, or dementia; (4) preoperative ZCQ score of <2 points; (5) preoperative 6MWD of ≥500 m; and (6) incomplete data. Based on these criteria, 150 patients were included in the final analysis (Fig. 1). The following variables were assessed: patient demographics, preoperative surgical factors, ZCQ findings, Visual Analog Scale (VAS) scores, the 6MWD, and the functional cross-sectional areas of the trunk and lower limb muscles.

Fig. 1

Patient selection. LSS, lumbar spinal stenosis.

Functional cross-sectional areas of the trunk and lower limb muscles

Preoperative lumbar spine and lower limb CT scans were performed with patients positioned supine and the knees slightly flexed, using either an Aquilion PRIME or Aquilion Lightning scanner (Toshiba Medical Systems, Tochigi, Japan). Images were reconstructed in 5-mm slices under a soft-tissue window. Muscle cross-sectional areas were quantified with SYNAPSE VINCENT software ver. 7.0 (Fujifilm, Tokyo, Japan). Measurements included the psoas major (PM) and erector spinae (ES) at the L3 level, the multifidus (MF) at L4, and the gluteus maximus (GM) and gluteus medius (GMed) at S2 (Fig. 2). The functional muscle area was determined by categorizing voxels according to Hounsfield unit (HU) ranges as follows: low-attenuation muscle area (LAMA; −29 to 29 HU) representing lipid-infiltrated tissue, and normal-attenuation muscle area (NAMA; 30 to 150 HU) corresponding to healthy muscle [15,16]. To normalize for differences in body size and more accurately reflect muscle quality deterioration, the LAMA ratio was calculated as “LAMA/(LAMA+NAMA),” with a higher ratio indicating greater fatty degeneration.

Fig. 2

Measurement of functional muscle cross-sectional areas. (A) The cross-sectional areas of the psoas major (green) and erector spinae (purple) at the L3 vertebral level. (B) The cross-sectional area of the multifidus (blue) at the L4 vertebral level. (C) The cross-sectional areas of the gluteus maximus (orange) and gluteus medius (pink) at the S2 vertebral level.

Results

A total of 150 patients were analyzed, among whom 86 (57%) achieved the MCID in the postoperative 6MWD. The MCID-achieved group was significantly younger than the non-achieved group (71.4±8.0 years vs. 76.0±6.6 years, p<0.001) and had a higher preoperative leg pain score (53.8±30.7 mm vs. 43.8±30.9 mm, p=0.033), ZCQ SSS (3.2±0.5 vs. 3.0±0.7, p=0.019), and ZCQ PFS (2.8±0.6 vs. 2.5±0.6, p<0.001) but a shorter preoperative 6MWD (262.8±105.8 m vs. 303.3±104.2 m, p=0.031) (Table 1).

Table 1

Comparison of preoperative patient factors, surgical factors, and clinical characteristics between the MCID-achieved and non-achieved groups

Analysis of muscle composition showed significantly lower LAMA ratios in the MCID-achieved group than in the non-achieved group across all examined muscles, including the PM (0.1±0.1 vs. 0.2±0.1, p<0.001), ES (0.2±0.2 vs. 0.3±0.2, p<0.001), MF (0.4±0.2 vs. 0.5±0.2, p<0.001), GM (0.4±0.2 vs. 0.5±0.2, p<0.001), and GMed (0.1±0.1 vs. 0.2±0.1, p<0.001), indicating less intramuscular fat infiltration (Table 2).

Table 2

Comparison of preoperative muscle composition between MCID-achieved and MCID-not-achieved groups

Univariate logistic regression identified several variables that were significantly associated with the achievement of the MCID, including younger age (odds ratio [OR], 0.92; 95% confidence interval [CI], 0.88–0.96; p<0.001), a higher ZCQ SSS (OR, 2.02; 95% CI, 1.07–3.84; p=0.030), a higher ZCQ PFS (OR, 2.29; 95% CI, 1.28–4.12; p<0.001), and a shorter preoperative 6MWD (OR, 1.00; 95% CI, 0.99–1.00; p=0.022). In addition, higher LAMA ratios across all muscles were inversely associated with the achievement of the MCID (all p<0.001) (Table 3).

Table 3

Univariate logistic regression analysis for MCID achievement

In the multivariate logistic regression analysis, younger age (OR, 0.90; 95% CI, 0.84–0.95; p<0.001), a shorter preoperative 6MWD (OR, 0.99; 95% CI, 0.98–0.99; p=0.016), and a lower ES LAMA ratio (OR, 0.69; 95% CI, 0.53–0.86; p<0.001) were identified as independent predictors of the achievement of the MCID (Table 4).

Table 4

Logistic regression analysis for MCID achievement

Discussion

In this study, we identified lower preoperative age, a shorter preoperative 6MWD, and a smaller ES LAMA ratio as independent predictors of achieving the MCID in the 6MWD at 12 months after surgery for LSS. The main strength and novelty of this study lie in demonstrating that preoperative trunk muscle quality, rather than lower limb muscle quality, is a key determinant for achieving a significant improvement in walking ability after LSS surgery. Specifically, a lower degree of fat infiltration in the ES, reflected by a smaller LAMA ratio, was associated with achieving the MCID in the postoperative 6MWD among patients with LSS. ES stabilizes and controls lumbar spine movement and may be a crucial factor in improving the walking ability of LSS patients.

LAMA reflects intramuscular fat infiltration and thereby indicates reduced muscle quality [18]. Healthy ES muscles are characterized by a high proportion of type I fibers, which are fatigue-resistant and play an essential role in postural control and spinal stability [19]. In patients with lumbar disc herniation, fatty infiltration of the ES has been shown to cause structural alterations in the paraspinal muscles, including atrophy of type I fibers and a shift toward type II fibers [19]. Moreover, ES muscle quality has been identified as an independent predictor of poor long-term outcomes following lumbar disc herniation surgery [20]. Collectively, an increased ES LAMA ratio may lead to diminished muscle endurance and postural stability, which can adversely affect postoperative improvements in the 6MWD. These findings underscore the crucial role of preoperative ES muscle quality in determining functional recovery after surgery. In this context, preoperative physiotherapy may help promote the achievement of the MCID in patients with LSS by improving ES muscle composition. Previous studies have shown that resistance and endurance training can enhance muscle quality in the trunk and thigh muscles by increasing NAMA and reducing LAMA [21]. Based on these findings, structured preoperative interventions targeting the ES are beneficial.

Age is known to strongly influence improvements in the 6MWD [22]. Aging contributes to sarcopenia, muscle weakness, balance impairment, and cardiopulmonary limitations, all of which can hinder postoperative gait recovery [23]. Consistent with findings in the study by Takase et al. [24], our results demonstrated that younger patients were more likely to achieve the MCID in the postoperative 6MWD, underscoring the importance of considering age-related variability when planning surgical interventions for LSS.

Furthermore, patients with a shorter preoperative 6MWD showed higher MCID achievement rates, suggesting that preoperative walking capacity strongly influences the potential for improvement. This may, in part, reflect a floor effect, as patients with greater preoperative disability have more room for improvement [25]. The preoperative 6MWD of the MCID-achieved group was below 300 m, a proposed threshold for severe gait impairment [26], suggesting that this group had a higher proportion of patients with pronounced intermittent claudication. Therefore, patients with more severe preoperative gait limitations may have a greater potential to achieve meaningful postoperative improvement.

Several limitations of this study should be noted. First, this was a single-center retrospective study, which may limit the generalizability of the findings. Second, CT-based muscle analysis was limited to selected axial slices and specific muscles, which may not fully represent the overall muscle condition. Third, potential confounding factors, such as physical activity, nutrition, and comorbidities, were not fully controlled. Fourth, the follow-up period was restricted to 12 months. Finally, although the 6MWD was used as a functional outcome, incorporating gait analysis and patient-reported outcomes could provide a more comprehensive evaluation of recovery.

Conclusions

The findings of this study suggest that muscle characteristics (muscle quantity and quality) and preoperative patient functional status are important determinants of postoperative walking recovery. Reduced fatty infiltration of the ES was found to be associated with better postoperative walking recovery, highlighting the importance of preoperative assessment and optimization of trunk muscle composition. Interventions aimed at improving ES muscle quality before surgery may represent effective strategies to enhance gait recovery and long-term functional outcomes in patients with LSS.

Article information Continued