Serum Apolipoprotein B/Apolipoprotein A1 Ratio In Relation To Intervertebral Disc Herniation: A Cross-Sectional Frequency-Matched Case-Control Study

Study Design:A cross-sectional frequency-matched case-control study Objective: In recent decades, the serum lipid prole of apolipoprotein(a) (Lp(a)) level and apolipoprotein B/apolipoprotein A1 ratio (Apo B/ApoA1) ratio were found more representative for serum lipid level and were recognized as the independent risk factors for various diseases. Although the serum lipid levels of total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) were found associated with symptomatic intervertebral disc herniation (IDH), no studies have been conducted to date for the evaluation of the association of Apo AI, Apo B, Lp(a) and Apo B/Apo AI levels with symptomatic IDH. Method: A total of 1,839 Chinese patients were recruited in the present study. 918 patients were diagnosed as IDH cases and were enrolled in the experimental group. A control group of 921 patients underwent a physical examination during the same period. The serum lipid levels of TC, TG, LDL-C, HDL-C, Lp(a), Apo B and Apo B/Apo AI were examined and analyzed. The patients in the control group were collected randomly from patients who were matched with the baseline levels of the aforementioned lipid molecular. Results: The patients with IDH exhibited signicantly higher TC, TG, LDL, Apo B and Lp(a) levels compared with the control subjects. The percentage of high-TC, high-TG, high-LDL, high-Apo B and high-Lp(a) were signicantly higher in the IDH group. However, hyperlipidaemia was not associated with the degenerated segment of the IDH (P=0.201). The odds ratios (OR) for the incidence of IDH with an elevated LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI were 1.583, 1.74, 1.62, 1.58,

With the increased research on serum lipid, the liporoprotein fraction of apolipoproteinAI (Apo AI) and apolipoproteinB (Apo B) and the ratios of apolipoprotein B/apolipoprotein AI (Apo B/Apo AI), and apolipoprotein(a) (Lp(a)) have received considerable attention in the investigation of dyslipidaemia related diseases in recent years, and Apo B/Apo AI and Lp(a) were recognized as the independent risk factors for various diseases, including osteoarthritis and AS [19][20][21]. However, the possible association between Apo AI, Apo B, Lp(a) and symptomatic IDH remains undiscovered.
In this study, a frequency-matched case-control study of serum lipid levels obtained from patients with symptomatic IDH was conducted in order to evaluate the relationship between serum lipid levels and symptomatic IDH in our study. We found that the patients with IDH exhibited signi cantly higher TC, TG, LDL, Apo B and Lp(a) levels compared with the control subjects. The percentage of high-TC, high-TG, high-LDL, high-Apo B and high-Lp(a) were signi cantly higher in the IDH group. Moreover, the odds ratios (OR) analysis revealed elevated LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI had a higher risk of IDH incidence. The correlative analysis further suggested elevated LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI are signi cantly closely with incidence of IDH. Our ndings revealed that elevated levels of serum TC, TG, LDL, Apo B, Lp(a) and Apo B/Apo AI are associated with a higher risk for IDH.

Methods
All procedures described in the present study were approved by the Ethics Committee of our institution. All patients provided written informed consent. The detailed primary ow-charts were presented in Figure  1. A total of 4,349 patients accepted the MRI scan and were potentially considered for inclusion from 2010 to 2019 in our institution. A total of 3,431 patients were excluded due to failure to meet the inclusion criteria (history of spinal disorders (n=155), multiple IDHs(n=367), spondylolysis(n=98), foraminal or central canal stenosis (n=187), spinal trauma (n=67), primary osteoarthritis of the operated and/or contralateral joint (n=367), in ammatory joint disease (n=211), diabetes (n=311), coronary heart disease (n=516), cerebrovascular disease (n=217), hypertension (n=467), smoking(n=468)). A total of 918 patients (399 men and 519 women; mean age: 60.74±12.69 years, range 18 to 93years) met our inclusion criteria and were enrolled in our study in group 1 (symptomatic IDH group). A total of 921 patients (control group) (401 men and 520 women; mean age: 61.02±12.59 years, range 18-91years), who underwent physical examination and conducted the MRI scan during the same period, were matched with the baseline of the symptomatic IDH group ( Table 1). The patients in the control group were excluded for IVDD by the MRI [22,23]. The procedure of selecting the patients in these two groups and giving the diagnosis were performed by experienced spine surgeons who did not know the purpose of the study. No statistically signi cant differences in the variables age, gender, labor intensity and BMI were noted between the two groups (P>0.05). The work has been reported in line with the STROCSS criteria [24].

Patient Selection
Patients in group 1 were included in the study if they had (a) symptomatic cervical spondylotic myelopathy, the thoracic and lumbar IDH. The diagnosis of cervical spondylotic myelopathy and of thoracic and lumbar IDH was conducted on the basis of clinical presentation, physical examination, radiography, electromyography, computerized tomography (CT) and/or MRI. (b) Absence of symptom relief despite adequate medical treatment.
The inclusion criteria for group 2 were the following: (1) The patients who underwent physical examination in the same period and were excluded for IVDD by spine MRI from 2009 to 2019 [2]. The patients who were frequency-matched by age and gender with the patients of Group 1 (Table1).
The exclusion criteria for groups 1 and 2 were the following: history of spinal disorders, multiple IDHs, spondylolysis, spondylolisthesis, foraminal or central canal stenosis, spinal trauma, spondyloarthritis, primary osteoarthritis of the operated and/or contralateral joint, in ammatory joint disease, diabetes, coronary heart disease, cerebrovascular disease, hypertension, smoking and age lower than 18 years[16].

The de nition of symptomatic IDH
The cervical disc herniation [25] (1) Typical sensory radicular symptoms (pain or paresthesias) were always present. (2) Motor (weakness and atrophy), sensory (hypesthesia or dysesthesia), and re ex (diminution or absence of tendon re exes) symptoms, if present, were con ned to 1 dermatome and/or myotome that corresponded to pain and/or paresthesias; Positive signs: a positive Spurling's test, Eaten test and Hoffman sign. The thoracic disc herniation [26] (1) Localized axial back pain and/or axial back pain with radiation into the lumbar spine; sensory impairment; (2) Special nerve root irritation signs: the pain can even mimic cardiac disease and/or present as abdominal and/or shoulder pain; (3) Neurologic de cit: paraparesis and monoparesis; spasticity and hyperre exia; bladder dysfunction; (knee jerk or ankle re ex).
The lumbar disc herniation [16] (1) Low back pain with unilateral or bilateral lower limb radicular pain; (2) Special nerve root irritation signs: straight leg raising test and strengthen test and/or femoral stretch test depending on the level of injury (3) Neurologic de cit: muscle weakness, numbness, and/or lack of the corresponding re ex (knee jerk or ankle re ex).

Imaging diagnosis
All the patients enrolled in the present study underwent the spine examination by 1.5T MRI. Both T2 and T1-weighted images were combined and used to assess the IVDD from C1/2 to L5/S1 regions by an experienced spine surgeon who was blinded to the study. The degree of IVDD on MRI was based on the P rrmann grade system [25]. The criteria for grade 1 are the following: Its structure is homogeneous and bright white in color; The distinction of nucleus and anulus is clear; The signal intensity for the disc is hyperintense and isointense compared with the cerebrospinal uid; The height of the intervertebral disc is normal. The criteria for grade 2 are the following: Its structure is nonhomogeneous in the absence and/or presence of horizontal bands; The distinction of the nucleus and anulus is clear; The signal intensity for the disc is hyperintense and isointense compared with the cerebrospinal uid; The height of the intervertebral disc is normal. The criteria for grade 3 are as follows: Its structure is progressing to nonhomogeneous and gray in color; The distinction of nucleus and anulus is unclear; The signal intensity for the disc is intermediate; The height of the intervertebral disc is normal to slightly decreased. The criteria for grade 4 are the following: Its structure is progressing to nonhomogeneous and gray to black in color; The distinction of the nucleus and the anulus is lost; The signal intensity for the disc is intermediate to hypo-intense; The height of the intervertebral disc is normal to slightly decreased. The criteria for grade 5 are as follows: Its structure is progressing to nonhomogeneous and black in color; The distinction of the nucleus and anulus is lost; The signal intensity for the disc is hypointense; The disc space is collapsed [27][28][29].

Blood examination
All blood samples were collected in an identical manner between 07.30 and 08.30 a.m. following an overnight fast that started at 12.00 midnight. Biochemical analyses of blood samples were conducted on fresh specimens. Fasting blood samples were collected, and ve milliliters of each sample was centrifuged at 4,000 rpm for 6 min. The serum was extracted from the samples, and the concentrations of TC, TG, LDL-C, HDL-C, Apo AI, Apo B, Lp(a) were measured by an automatic biochemical analyser in an identical manner. The normal levels of the following indexes exhibited the following range TC from 0 to 5.2 mmol/L; TG from 0 to 1.7 mmol/L; LDL-C from 0 to 3.4 mmol/L; HDL-C from 0.7 to 2.0 mmol/L; ApoEA1 from 1 to 1.6 g/L; Lp(a) from 0 to 30 mg/mL; Apo B from 0.6 to 1.1 g/L. The normal value of the ratio of Apo B/Apo AI was 0.87 for the male subjects, and 0.65 for the female subjects [30].

Statistical analysis
Continuous variables were expressed as the mean ± standard deviation (SD) and analyzed with unpaired t-tests. Categorical variables were expressed as a percentage of the number and analyzed with a chisquare test. The normality analysis was carried out for the continuous variables. SPSS (Version 20.0) was used for all statistical analyses. A P value that was lower than 0.05 was regarded as statistically signi cant. Multivariate logistic regression was used to evaluate the effects of serum lipids on symptomatic IDH. The effect indicators were odds ratio (OR) and 95 % con dence interval (CI). The data of continuous variables in the present study followed an ordinary normal distribution. The adjustment for multiple comparisons was conducted in the present study. A P value that was lower than 0.007 (P<0.007) was considered statistically signi cant following the adjustment for multiple comparisons (Table 1 and Figure 2). The correlations analysis was carried out in the present study. A P value that was lower than 0.05 (P <0.05) was considered statistically signi cant, A P value that was lower than 0.01 (P<0.01) was considered extremely statistically signi cant. The percentage of high-TC, high-TG, high-LDL, high-Apo B and high-Lp(a) were signi cantly increased in the IDH group

The association between serum lipid abnormalities and the degree of IVDD
To further investigate correlation for the incidence of a symptomatic IDH with an elevated LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI, the correlation analysis was conducted between serum lipid abnormalities and the degree of IVDD (P rrmann grade). As shown in Fig. 3, the correlation between elevated LDL-C, TC, TG, Apo B, Lp(a) and incidence of IDH were signi cant (R 2 LDL = 0.017, P < 0.001; R 2 TC = 0.004, P < 0.004; R 2 TG = 0.015, P < 0.001; R 2 Apo B = 0.004, P < 0.001; R 2 LP(a) = 0.021, P < 0.008).
These results suggested that patients with higher LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI levels are closely related with disc herniation. Hyperlipidaemia did not affect the degenerated segment of the intervertebral disc The categorical data of the patients with disc herniation were analyzed in order to examine the associations between hyperlipidaemia and the disc segment in the intervertebral disc group. The hyperlipidaemia group (n = 689) exhibited the following percentages of degenerated segments in the cervical, thoracic and lumbar regions: 13.9%, 1.3% and 84.8%, respectively. In contrast to the hyperlipidemic samples, the normal serum lipids group exhibited incidences of 18.8%, 1.3% and 79.9% that corresponded to the cervical, thoracic and lumbar regions, respectively (n = 229). No signi cant differences in the herniation segments were noted between these two groups (p = 0.201) (Fig. 4).
The categorical data were further analyzed in order to identify the association between the serum lipid levels and the segment of disc herniation in the cervical and lumbar regions. Considering the small sample of affected segment in the thoracic disc herniation, we only analyzed the serum lipid levels and segment of disc herniation in the cervical and lumbar regions.
The values of the total segment in the cervical, thoracic and lumbar regions were 137, 12 and 769 respectively. No signi cant differences were noted between serum lipid levels in the C3-C4 (P = 0.282) and C5-6 (P = 0.373) segments with regard to TC levels (Fig. 5A). Similarly, no signi cant differences were observed in the C3-C4 (P = 0.108) and C5-6 segments with regard to LDL-C levels (P = 0.254) (Fig. 5C). With regard to the levels of Apo B, the C5-6 segment in the hyperlipidaemia group (31.9%) was higher than that of the normal group (24.2%), although no signi cant differences were noted (Fig. 5D, P = 0.2). With regard to the levels of ApoA (Fig. 5E), Lp(a) (Fig. 5F) and triglycerides (TG) (Fig. 5B), the distribution herniation segment in the hyperlipidaemia and control groups exhibited similar trends both in the lumbar and cervical segments. However, we do not get the signi cance statistic in our study. Comparing with cervical and lumbar IDH, the incident of thoracic IDH is quite low. The relatively small sample size in our study may contribute to the no-signi cance result.
Patients with elevated LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI levels exhibited a higher risk of disc herniation To further identify risk for the incidence of a symptomatic IDH with an elevated LDL-C, TC, TG, Lp(a), Apo B and Apo B/Apo AI, multivariate logistic regression analysis was performed. As shown in Table 4

Discussion
The association between serum lipid and IVDD related disease has been examined by a multitude of studies [12][13][14][15][16]. Elevated levels of TC, LDL-C and TG have been shown to be associated with sciatica [12], back pain and/or disc herniation [13][14][15][16]28]. In agreement with these results, our data showed IDH patients exhibited signi cantly higher TC, TG, LDL levels and the percentage of high-TC, high-TG, high-LDL were signi cantly increased in the IDH group.
Results from several clinical prospective studies indicate that the Apo B/ApoA1 ratio is a accurate risk factor for cardiovascular, osteoarthritis, rheumatoid arthritis, metabolic syndrome disease [19][20][21]27]. However, as aged related degenerated disease, whether any association between Apo AI, Apo B, Apo B/Apo AI and Lp(a) levels and symptomatic IDH are still unclear.
In the present study, the relationship between Apo AI, Apo B, Lp(a) and symptomatic IDH were examined for the rst time. As a result, the levels of Apo B and Lp(a) were shown to positively associate with the incidence of symptomatic IDH. For the LDL-C, Apo B can facilitate cholesterol delivery to the tissues. However, Apo AI can facilitate the peripheral cells uptake of cholesterol and help transport of cholesterol to the liver for digesting. Thus, the levels of these Apo B/Apo AI can re ect cholesterol transport ability to the peripheral tissues and determine the level of cholesterol in our plasma [31,32]. Elevated levels of the plasma concentrations of Apo B and increased proportion of the Apo B/Apo AI ratio in symptomatic IDH patients suggested a prominent cholesterol transport to the peripheral tissues including IVD in these patients. Lp(a) having the similar protein and lipid structure with LDL-C. Various studies have shown that high levels of Lp(a) in plasma can be a risk factor for cardiovascular disease, osteoarthritis, and RA [33]. In the present study, it was shown that the level of Lp(a) was increased in the intervertebral dis herniation group. The current study is the rst study to report this nding in the eld of symptomatic intervertebral disc herniation.
The precise pathophysiologic mechanism underlying the connection between serum lipid levels and lumbar disc herniation remains unclear. The increasing levels of TC, TG, LDL-C, Apo B and Lp(a) in IDH patients could be mainly caused by several reasons. Firstly, the IVD is a poorly vascularized region, whose nutritional supply is through the blood capillary penetration of endplate cartilage and the aunnual brosis [29]. High levels of serum cholesterol [28], triglycerides [28, [34][35][36], LDL-C [37], Apo B and Lp(a) are considered contribute to atherosclerosis. The presence of atherosclerotic will inhibit the vascular supply to the poorly vascularized IVD and induced IVDD/IDH [38]. In agreement with our hypothesis, the association between symptomatic IDH and atherosclerosis were identi ed by a lot of studies. A study that included 86 people concluded that atherosclerosis in the abdominal aorta and notably stenosis of the ostia of segmental arteries may play a vital role in symptomatic IDH [39]. Another 25-year follow-up study showed the calci c atherosclerotic deposits in the abdominal aorta increased the risk of disc herniation and back pain [40]. Secondly, the activated in ammatory cells induced by high serum lipid levels may comprise an important pathway in the development of symptomatic IDH for hyperlipidaemia patients. The activation of cytokines plays a signi cant role in the development of IVDD/IDH [38,41].
Previous studies have reported that pro-in ammatory cytokines were closely associated with serum lipid levels [5,42]. Therefore, the increased serum lipid levels may enhance the in ammatory response and/or the basic level of systemic in ammation, which can in turn contribute to disc herniation [6]. Thirdly, the oxidized low-density lipoprotein (oxLDL) and the increased expression of lectin-like oxidized low-density lipoprotein receptor 1 (LOX-1) that are caused by dyslipidaemia may also be involved in the development of symptomatic IDH. Our previous study suggested [43] that the levels of the oxLDL and LOX-1 were positively correlated with the extent of IDH. The mechanism of action involved the increase in the expression of MMP3 that was induced by LOX-1. This in turn caused the oxLDL to signi cantly reduce the viability of human nucleus pulpous. The production of oxLDL usually originates from LDL-C that is oxidized under oxidative stress conditions. Therefore, the elevated LDL-C levels can increase level of oxLDL/LOX-1 and accelerate IVDD.
As is known, elevated levels of TC, TG, LDL-C, Apo B, Lp(a) and reduced HDL-C level are atherogenic lipid marker. The management of cardiovascular disease has traditionally focused on reducing serum lipid levels [44], In this study, we found that elevated serum lipid levels were signi cantly correlated with IDH and High serum lipid levels predicted a higher incidence of IDH. This association opens the way for a new approach to reducing the risk of IDH/IVDD disease by controlling serum lipid levels. There were several limitations in the present study: Firstly, no data were provided regarding the levels of very low-density lipoproteins (VLDL). Secondly, this was a retrospective case-control study, thus the causal relationships between serum lipid components and symptomatic IDH remain unclear. In order to prove cause and effect relationships and to nd effective treatments for IVDD/IDH, a large longitudinal follow-up observations and intervention studies is needed[16].

Conclusion
The level of TC, TG, LDL-C, Apo B (Apo B/Apo AI) and Lp (a) were positively correlated with the incidence of symptomatic IDH. However, hyperlipidaemia did not affect the degenerated segment of the intervertebral disc. This association provides novel evidence regarding the reduction of the risk of symptomatic IDH disease by the control of the serum lipid levels.

Declarations
Ethical Approval and consent to participate: All procedures performed in studies involving human participants were approved prospectively by the authors' human subjects Institutional Review Board.
Informed consent All of the participants consented to participate in this study.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Consent for publication
Not applicable

Competing interests
The authors declare that they have no competing interests Funding Tables   Table 1 Basline characteristics of participants (N = 1839) F:female M:male ;A chi-square test and unpaired t-tests were used for analysis Table 2 The concentrations of serum lipids in two groups.