Association of central sensitization, visceral fat, and surgical outcomes in lumbar spinal stenosis

We enrolled consecutive patients with LSS aged between 41 and 79 years. The surgeons performed physical testing on consecutive patients who reported with symptoms induced or exacerbated with walking or prolonged standing and relieved with lumbar flexion, sitting, and recumbency. These symptoms included pain, numbness, and neurological deficits in the lower extremities and buttocks as well as bladder or bowel dysfunction. Subsequently, radiographic and magnetic resonance imaging (MRI) findings suggestive of degenerative stenosis of the spinal canal or intervertebral foramen were correlated with the reported symptoms and clinical findings, thereby confirming the diagnosis of LSS. The same surgeons made the final diagnosis of symptomatic LSS, which necessitated the presentation both of clinical symptoms and radiographic findings of LSS. Symptoms of leg pain/numbness and intermittent claudication in patients with LSS who did not respond to conservative therapies for more than 3 months were considered to be indications for decompression surgery [21]. The exclusion criteria were as follows: age < 40 and > 80 years, acute trauma, infection, neoplasm, history of spinal surgery or spinal fracture, leg symptoms presentation for < 3 months, and spondylolisthesis with obvious intervertebral instability, identified as a pain generator causing LBP that may be improved by fusion surgery [22]. Intervertebral instability was defined as sagittal translation of > 3 mm, segmental motion of > 20°, or a posterior opening of > 5° on flexion/extension radiographs [23]. All participants rated their LBP, leg pain, and leg numbness using a VAS (0–100 mm). VASs are commonly used assessment tool for musculoskeletal pain intensity and have been proven reliable and valid [24]. Spinous process splitting laminectomy was performed as decompression surgery [25]. The participants completed clinical outcome assessments preoperatively and 12 months postoperatively. Finally, 60 patients were enrolled in the study (28 men and 32 women; mean age: 62.1 ± 2.8 years).

The CSI measures the extent to which an individual’s symptoms are likely attributable to CS [17, 18]. Part A comprises 25 symptom-related items scored on a 5-point Likert scale (0–4 for each item; total score range of 0–100). Part B identified patients with concurrent fibromyalgia. The CSI has been established as valid and reliable [17] with test–retest reliability of 0.82, Cronbach’s alpha of 0.88, sensitivity of 81%, and specificity of 75% [18]. The CSI has also been translated into Japanese and validated [26]. Neblett et al. [18] investigated patients referred to a multidisciplinary pain centre, specializing in the assessment and treatment of chronic pain, including central sensitivity syndromes (CSSs), such as fibromyalgia, chronic fatigue, and irritable bowel, for which CS may be a common aetiology. A receiver operating characteristic analysis determined that a CSI score of 40 out of 100 best distinguished patients with CSS from their non-patient comparators (area under the curve = 0.86, sensitivity = 81%, specificity = 75%). Thus, a cut-off score of 40 was used to identify low- and high-CS symptoms [27]. Therefore, in the present study, we divided the participants into low- (CSI < 40) and high- (CSI ≥ 40) CSI subgroups. CSI and VAS score measurements were conducted at the same time points for all patients.

Visceral and subcutaneous fat cross-sectional areas were calculated semiautomatically using tissue-specific attenuation thresholds including adipose tissue (− 190 to − 30 Hounsfield unit [HU]) and skeletal muscle (− 29 to 150 HU) [28, 29]. Lee et al. [30] demonstrated excellent intra- and inter-observer reproducibility of adipose tissue measurements on CT images and reported the intra- and inter-observer agreements for the volume of visceral fat (intraclass correlation coefficient [ICC], 0.998 and 0.999; 95% confidence interval [CI] 0.996–0.999 and 0.999–1.000, respectively) and subcutaneous fat (ICC for intra- and inter-observer agreements, 0.998 and 0.997; 95% CI 0.996–0.999 and 0.994–0.998, respectively). CT-measured abdominal circumference referred to the waist perimeter, defined as the circumferential length of the outer margin of abdominal skin on axial CT images. Joo et al. [31] showed that CT-measured abdominal circumference was closely correlated with manually-measured abdominal circumference (r = 0.919; 95% CI 0.908–0.930], p < 0.001), and the ICC of CT-measured and manually-measured abdominal circumference was 0.954 (95% CI 0.947–0.960). The patients underwent CT imaging (Aquilion, Toshiba, Japan), which was performed simultaneously with myelography. We measured the visceral fat area, subcutaneous fat area, and abdominal circumference at the level of the umbilicus. The respiratory phase was at the end of exhalation, and the parameters were set as follows: tube voltage, 120 kV; tube current, 360 Ma; spin time, 0.5 s/rotation; and slice thickness, 0.6 mm. Typical CT images of a low-CSI patient and a high-CSI patient are presented in Fig. 1a and b, respectively.

All numerical data are expressed as the mean ± standard error of the mean. Differences between men and women were analysed using the chi-square test. Additionally, differences between the groups regarding age, BMI, and VAS scores were compared using the Mann–Whitney U test. Analysis of covariance (ANCOVA) was used to compare the CT measurements between the low- and high-CSI groups, adjusted for age and sex. Differences in VAS scores between the pre- and post-12 months periods were analysed using the Mann–Whitney U test. All statistical analyses were performed using SPSS (version 27.0; IBM Corp., Armonk, NY, USA). Statistical significance was set at p < 0.05.