Introduction

Lumbar spinal disorders (LSDs) such as lumbar spinal canal stenosis and lumbar disk herniation are globally common musculoskeletal conditions. With societal aging, the number of patients with LSDs has increased [1]. LSDs are involved in mobility reduction and social life decline because of low back pain, leg pain, motor weakness, and sensory disturbances of lower extremities, and intermittent claudication [2, 3]. In addition, older patients with LSDs were prescribed multiple medications because of multiple comorbidities [4,5,6]. Among comorbidities, metabolic syndrome was reported as a disorder in which lumbar spinal canal stenosis is potentially involved [7]. Decreasing physical activity due to lumbar spinal stenosis was speculated to mediate the incidence of metabolic syndrome [7].

According to guidelines [8, 9], those with visceral fat (VF) accumulation accompanied by any two of dyslipidemia, hypertension, and diabetes mellitus (DM) are diagnosed with metabolic syndrome. VF accumulation can be defined as the cross-sectional area of viscous adipose tissue measured at the level of the navel using computed tomography (CT) ≥ 100 cm2 [9]. Serum levels of triglycerides, high-density lipoprotein (HDL), and fasting plasma glucose are parameters commonly used for dyslipidemia and type 2 DM, respectively [8, 9]. Lifestyle guidance including exercise is the first line of treatment for metabolic syndrome [8, 9].

Although conservative therapy is recommended as the first-line treatment for LSDs according to the guidelines, it should not be continued unnecessarily in cases with severe and long-standing symptoms due to polypharmacy and medical costs as well as treatment resistance to surgery [2, 3, 6, 10]. In surgery for LSDs, the 2-year outcomes of surgery have been reported to be better than conservative therapy, although the choice of decompression alone or combination of decompression and fusion should be considered depending on the case [10]. In general, lumbar spinal surgery for patients with LSDs has shown efficacy on pain, motor function, social life, psychology, and healthy life expectancy [11,12,13,14,15,16]; however, it is still unclear how lumbar spinal surgery affects the lipid metabolism of patients with LSDs. Thus, this study primarily aimed to assess the impact of lumbar spinal surgery on lipid metabolism by measuring pre- and postoperative VF accumulation and serum levels of triglycerides and HDL in patients with lumbar spinal surgery for LSDs. The secondary aim was to clarify the factors associated with changes in lipid metabolism before and after lumbar spinal surgery.

Materials and methods

Study participants

We prospectively enrolled the patients aged ≥ 30 years who were scheduled to undergo surgical treatment for LSDs at a single hospital between April 2021 and March 2022. The follow-up period was 1 year. LSDs included lumbar spinal canal stenosis, lumbar degenerative spondylolisthesis, lumbar spondylolytic spondylolisthesis, and lumbar disk herniation. Lumbar spinal surgery was indicated for patients with apparent symptoms and who had no efficacy of conservative therapy [2, 3, 10]. The diagnosis of LSDs was confirmed by magnetic resonance imaging, CT, or myelography. Degenerative lumbar scoliosis was defined as a Cobb angle ≥ 10°. Lumbar spinal fusion was basically recommended for patients with lumbar degenerative spondylolisthesis and spondylolytic spondylolisthesis. Patients with an upper instrumented level of thoracic spine or lower instrumented level of the pelvis were excluded.

Data collection

Spinal CT images of participants were examined before surgery and at 6 months and 1 year after surgery. The cross-sectional VF area (VFA) at the level of the navel on CT images was calculated using Vitrea (Canon Medical Systems Corporation, Tochigi, Japan) (Fig. 1A). Because one of the criteria for metabolic syndrome is VFA ≥ 100 cm2 [9], we used 100 cm2 of VFA on CT as the cut-off value for grouping in this study. Blood examination of each participant was also performed before surgery, and at 6 months and 1 year after surgery. Blood examination items included triglycerides and HDL. The participants completed the following clinical outcome assessments before surgery, and at 6 months and 1 year after surgery: the Zurich Claudication Questionnaire (ZCQ) and the Japanese Orthopaedic Association Back Pain Evaluation Questionnaire (JOABPEQ). The following patient data were collected: age; sex; body mass index (BMI); duration of conservative therapy before surgery; medical history, including DM, hypertension, dyslipidemia, and degenerative lumbar scoliosis; and perioperative factors including surgical procedures, surgical time, and surgical blood loss. Surgical treatment was considered effective in each domain of the JOABPEQ according to a previous study [17].

Fig. 1
figure 1

Caluculation of visceral fat area (VFA). A Computed tomography image of the VFA (red) at the level of the navel. B Correlation between body mass index (BMI) and VFA. R, correlation coefficient

Full size image

CT examinations

All CT examinations were performed as spinal CT examinations, which were covered entire abdomen from diaphragm to pelvis and reconstructed as spinal CT with bone algorithm and bone window setting and abdominal CT with abdominal window setting. CT data were obtained with a 64-detector row CT scanner (Revolution Maxima: GE Healthcare, Chicago, IL, USA). CT examinations were conducted as unenhanced CT with helical scanning using the following parameters: 64 × 0.625 mm collimation, auto mA with image noise, 120 kVp, 0.516 beam pitch, 0.7 s gantry rotation time, 512 × 512 matrix, and 320-mm field of view. All CT data were then reconstructed with the filtered back projection or hybrid iterative reconstruction (ASIR-V4, GE) methods in contiguous section thickness of 5 mm and used for generating the lung reconstruction kernel as a soft tissue kernel (GE). All CT examinations were performed with breath-holding at end expiration.

Statistical analyses

Data among groups were compared using the chi-squared test, Wilcoxon signed-rank test, or Kruskal–Wallis test as appropriate. Correlations between BMI and VFA were evaluated by Spearman correlation coefficients. We constructed a Poisson regression model to explore the factors, including age, sex, BMI, medical history including hypertension, dyslipidemia, and diabetes mellitus, surgical procedure, and preoperative score in 5 domains of JOABPEQ, associated with changes in visceral fat accumulation before and after lumbar spinal surgery. Considering that Poisson regression model have the advantage to estimate direct relative risk [18], it was selected to directly estimate relative risk, which adjusted for age and sex, for preoperative VFA ≥ 100 cm2 and 1-year postoperative decrease of VFA. In this model, the JOABPEQ scores were categorized in tertiles to account for the number of samples according to a previous study [6]. Poisson regression was performed using the STATA16 software (Stata Corporation, College Station, TX, USA). P values < 0.05 to indicate statistical significance.

Results

The study enrolled a total of 138 patients. Table 1 summarizes the characteristics of participants, and Table 2 shows the preoperative and 6-month and 1-year postoperative ZCQ scores including symptom severity and physical function and JOABPEQ scores. The 6-month and 1-year postoperative ZCQ and JOABPEQ scores were significantly improved compared with the baseline data in all domains (Table 2).

Table 1 Baseline Characteristics
Full size table
Table 2 Valuables of patient reported outcome at baseline and follow-up after surgery (n = 138)
Full size table

First, to examine the association between VFA and BMI, we calculated the correlation coefficient between preoperative VFA and preoperative BMI for participants and found a correlation coefficient of 0.62, indicating a moderate correlation (Fig. 1B). Table 3 shows preoperative and 6-month and 1-year postoperative data for VFA on CT and serum levels of triglycerides and HDL in blood examination. VFA tended to increase after surgery as a whole (Table 3). No significant changes were seen in men; however, significant changes were seen in women 6 months and 1 year after surgery compared with before surgery (Table 3). In blood examination, triglyceride levels also tended to increase after surgery as a whole (Table 3). While men showed no significant changes, women showed significant changes 6 months and 1 year after surgery compared with before surgery (Table 3). Meanwhile, HDL levels tended to decrease both as a whole and by sex; however, none of these changes were significant (Table 3).

Table 3 Visceral fat area on CT and serum levels of triglyceride and high-density lipoprotein cholesterol at baseline and follow-up after surgery (n = 138)
Full size table

When the participants were divided into two groups based on whether VFA increased or decreased 1 year after surgery, each group included 55.8% (n = 77) and 44.2% (n = 61) of the patients, respectively. Figure 2 shows the percentage of the four groups that were further divided based on whether the preoperative VFA was greater or less than 100 cm2; group A, postoperative increase in VFA and preoperative VFA < 100 cm2; group B, postoperative increase in VFA and preoperative VFA ≥ 100 cm2; group C, postoperative decrease in VFA and preoperative VFA < 100 cm2; group D, postoperative decrease in VFA and preoperative VFA ≥ 100 cm2 (Fig. 2). In the present study, we examined whether the preoperative JOABPEQ scores for each domain differed among these four groups (Table 4). However, no significant difference in the preoperative score of each JOABPEQ domain was found among the four groups (Table 4). Then, the frequency of effective cases of surgical treatment in each domain for JOABPEQ was compared among the four groups (Fig. 3). However, they also showed no significant differences among the four groups (Fig. 3). Next, the participants were divided into two groups based on whether surgery was effective or ineffective in each domain of the JOABPEQ, and the frequency of postoperative VFA increase or decrease was compared between the two groups (Fig. 4). In this analysis, there were no significant differences in the frequency of postoperative VFA increase or decrease between the two groups in all domains.

Fig. 2
figure 2

Distribution of the four groups

Full size image
Table 4 Comparison of preoperative scores of each domain for JOABPEQ among four groups
Full size table
Fig. 3
figure 3

Comparison of the frequency of effective cases of surgical treatment among the four groups. Group A, postoperative increase in the visceral fat area (VFA) and reoperative VFA < 100 cm2 (n = 44); group B, postoperative increase in the VFA and preoperative VFA ≥ 100 cm2 (n = 33); group C, postoperative decrease in the VFA and preoperative VFA < 100 cm2 (n = 27); and group D, postoperative decrease in the VFA and preoperative VFA ≥ 100 cm2 (n = 34)

Full size image
Fig. 4
figure 4

Comparison of the frequency of postoperative visceral fat area (VFA) increase or decrease between the two groups based on whether surgery was effective or ineffective in each domain of Japanese Orthopaedic Association Back Pain Evaluation Questionnaire (JOABPEQ)

Full size image

Finally, to identify the characteristics of those with a preoperative VFA ≥ 100 cm2 and a decrease in VFA 1 year after surgery, we used a Poisson regression model to examine the factors associated with this group. After adjusting for age and sex, a lower score in the walking ability of preoperative JOABPEQ was associated with an increased risk of this group (RR 2.1; 95% CI 1.1–4.1) (Table 5).

Table 5 Poisson regression model of a preoperative visceral fat area  ≥ 100 cm2 and a decrease in visceral fat area 1 year after surgery
Full size table

Discussion

This study followed the lipid metabolism of patients with LSDs who underwent lumbar spinal surgery. Overall, patients with LSDs tended to have increased VFA and increased triglyceride and decreased HDL levels on blood examination after lumbar spinal surgery, indicating that the lipid metabolism of patients with LSDs tended to deteriorate after lumbar spinal surgery. In particular, female patients with LSDs had significantly increased VFA and increased serum levels of triglycerides after lumbar spinal surgery. The multivariable analysis showed that the group with > 100 cm2 of preoperative VFA and a postoperative decrease in VFA had significantly worse preoperative walking ability.

Considering that lumbar spinal surgery improves motor function weakness and impaired social life in patients with LSDs [11, 12, 14], the lipid metabolism of patients with LSDs may have improved. Meanwhile, considering that improvement in psychological disorders could lead to increased appetite [19], lumbar spinal surgery has the potential to deteriorate lipid metabolism as psychological disorders improved [15]. The results of this study indicated that lipid metabolism in patients with LSDs tends to deteriorate with lumbar spinal surgery. Separate analyses for men and women showed that this was true for women. Because women were reported to have a less daily exercise [20], lipid metabolism in female patients with LSDs may more strongly reflect improved psychological impairment than improvements in motor function following lumbar spinal surgery than male patients with LSDs. However, patients with LSDs who are not diagnosed with metabolic syndrome or obesity do not need to improve their lipid metabolism after lumbar spinal surgery. Therefore, most importantly, LSD patients with a preoperative diagnosis of metabolic syndrome or obesity will have their lipid metabolism improved by lumbar spinal surgery. In this study, we attempted to identify associated factors using multivariable analysis, with a preoperative VFA of ≥ 100 cm2 and a decrease in postoperative VFA as objective variables. On multivariable analysis, a lower walking ability before lumbar spinal surgery was associated with VFA improvement. Given that LSD patients with lower walking ability before lumbar surgery are more likely to cause metabolic syndrome because of decreased physical activity and that such patients may have improved lipid metabolism because of improved walking ability following lumbar spinal surgery, these results are to some extent consistent with our clinical impression. Conversely, our results also indicate that lumbar surgery alone does not improve the lipid metabolism of patients with LSDs. According to the previous study [21], patients who underwent total knee replacement for knee osteoarthritis, another common musculoskeletal degenerative disease, showed no change in the frequency of metabolic syndrome 2 years postoperatively. These findings may suggest the importance of postoperative lifestyle guidance for the improvement of lipid metabolism in patients with musculoskeletal degenerative diseases. Regarding clinical research, the results of this study were not conclusive because the study participants did not receive uniform postoperative lifestyle guidance, including exercise. To accurately assess changes in lipid metabolism after lumbar spinal surgery in patients with LSDs, all patients with LSDs should receive the similar postoperative lifestyle guidance.

This study had several limitations. First, the number of participants is limited. A larger sample is needed to gain a more comprehensive understanding of lipid metabolism in patients with surgical LSDs. In particular, multivariable analysis for identifying factors involved in VFA improvement may yield more meaningful results if conducted in a larger number of patients. Second, this study did not include patients with LSDs receiving conservative therapy as a control group. VF has been reported to likely increase with age because of a decrease in muscle mass and a decrease in basal metabolic rate with age [22, 23]. Thus, whether our findings are specific to lumbar spinal surgery remains unclear. Third, the follow-up duration of 1 year was insufficient to assess the impact of lumbar spinal surgery on lipid metabolism. In the longer term, the impact of lumbar surgery on lipid metabolism in patients with LSDs may be different. Fourth, given that metabolic syndrome is more likely to occur in middle-aged individuals than in older people [24, 25] and that a significant relationship between lumbar spinal canal stenosis and metabolic syndrome has been observed in people aged < 65 years [7], this study should have focused primarily on those aged < 65 years. Fifth, this study only analyzed VFA on CT images and serum triglyceride and HDL levels in blood examination and did not examine metabolic syndrome based on diagnostic criteria. Finally, we have not examined potential variations in exercise levels or dietary habits that could underlie the observed changes in VFA and serum triglycerides levels in female patients after surgery. The absence of evaluation of these variables was one of limitations in this study. To the best of our knowledge, however, this is the first prospective study to assess lipid metabolism in patients with LSDs who underwent surgery. This study could provide useful information for both physicians and surgeons.

In conclusion, the results of this study demonstrated that patients with LSDs did not necessarily have improved lipid metabolism after lumbar spinal surgery. Instead, female patients with LSDs had significantly deteriorated the status after lumbar spinal surgery. Finally, the multivariable analysis indicated that a worse preoperative walking ability was associated with the improvement in excess VF accumulation after lumbar spinal surgery.