Background
Cervical compression myelopathy is one of the most common spinal cord disorders affecting the elderly. It is well known that the mechanism of compression myelopathy is chronic compression of the spinal cord by degenerated discs, osteophytes, thickened ligamentum flavum, and ossification of the posterior longitudinal ligament [
1]. Although a slow and stepwise decline in function is usually observed in compression myelopathy, a rapid progression of motor paralysis and paresthesia with mild or no trauma is occasionally observed. Acutely worsening symptoms of compression myelopathy result in severe neurological deficits with poor functional recovery [
2]. To date, early surgical treatment is recommended as the only valid therapy for compression myelopathy, and the surgical outcome is relatively favorable [
3]. However, in some cases, sufficient improvement in neurological function is not achieved [
4]. Currently, an accurate prediction of the recovery rate of neurological function before surgical treatment is not possible. Furthermore, a subjective evaluation using the JOA score for example, is the only procedure available to assess the severity of neurological dysfunction [
5]. Therefore, several biomarkers that reflect the degree of damage to the spinal cord and the severity of neurological symptoms would be useful.
Results of a recent study show that phosphorylated neurofilament subunit H (pNF-H) and tau protein, which reflect axonal damage to the spinal cord, become elevated in the cerebrospinal fluid (CSF) of patients with acutely worsening compression myelopathy [
6,
7]. It was suggested that pNF-H in CSF may act as a biomarker to predict neurological recovery after surgical treatment. However, because CSF sample collection is strictly limited to patients who elect to undergo a myelogram, CSF samples are unsuitable for evaluating changes over time given associated ethical limitations. Therefore, more easily obtained serum biomarker is preferred.
Recent reports indicate that oxidative stress markers in serum are elevated in various neurodegenerative disorders such as Alzheimer’s disease, amyotrophic lateral sclerosis, and Parkinson’s disease [
8,
9]. For orthopaedic lesions, it was reported that the ROM in serum reflected the severity of rheumatoid arthritis (RA) [
10]. However, to our knowledge, no report has analyzed oxidative stress markers based on ROMs in the serum of patients with compression myelopathy. Therefore, we conducted a pilot cross-sectional study to determine the levels of oxidative stress in patients with compression myelopathy and to investigate any relationship of oxidative stress levels with clinical outcome.
Discussion
To our knowledge, the present study is the first to determine the levels of oxidative stress markers in the serum of patients with compression myelopathy. Generally, biological markers of oxidative stress are molecules that are modified via interactions with ROS. As previously stated, the direct measurement of ROS in serum is difficult because of their biochemical instability [
14]. Thus, an assay to detect a more stable class of ROM, or hydroperoxides, has been developed. In fact, measuring ROM, a proxy for ROS production, has already been recognized as useful for the evaluation of oxidative stress levels in a multitude of diseases including, but not limited to, obesity, metabolic syndrome, asthma, and RA [
10,
16,
17]. In addition, although the pathogenesis of compression to the spinal cord is a little different between CSM and OPLL, there were no significant difference by the cause of disorders (CSM or OPLL) in this study that suggest that OPLL does not concern to the elevation of ROM.
According to past reports, markers of oxidative stress increase with increasing age, female gender, obesity, and diabetes [
12,
14,
16] and endogenous oxidative stress has been implicated. The markers of oxidative stress in serum were increased in females in the AM group compared with males in the AM, males in the CM, and females in the CM group, a finding also in partly supported by a previous study [
12]. However, there were no significant difference between the sex and the recovery rate of JOA score. To date, we have no idea to speculate this phenomenon. Further investigation will be needed to explain this phenomenon. On the other hands, in our study, there were no correlation between ROM and age, BMI, history of smoking, and existence of diabetes that may indicates those factors less concern to the elevation of oxidative stress.
In contrast, the level of oxidative stress based on serum analyses increases when there is an acute spinal cord injury, Alzheimer’s disease, amyotrophic lateral sclerosis, and Parkinson’s disease [
9,
18,
19]. Reports of oxidative stress in the presence of neurodegenerative diseases suggests that there exists the neurogenic oxidative stress in which neuronal cell apoptosis may be involved. In this study, although ROM in the serum of patients in the CM group showed evidence of a moderate level of oxidative stress, patients in the AM group had ROM indicative of a significantly more severe level of oxidative stress. This may, at least in part, reflect more oxidative stress resulting from worsening neurodegeneration in the AM group. Furthermore, in the present study, high ROM in the AM group are reflective of a poor surgical outcome. Neuronal apoptosis may be in partly more serious in patients in the AM group with high ROM considering the favorable neurological recovery of CM. In addition, the recent report has indicated that the ischemia-reperfusion phenomenon concerns the surgical outcome in AM as the major pathogenesis [
20]. Our results may reflect that the ischemia-reperfusion injury is also more serious in patients in the AM group with high ROM. Ultimately, the results of the present study suggest that the surgical outcome may be poor in patients with high ROM, like those in the AM group or the representative case. Thus, the results help us to explain during the time of informed consent, before surgery, that the surgical outcome may be inferior in such patients. Nevertheless, the ratio of endogenous oxidative stress to neurogenic oxidative stress is unclear. In addition, it is also unclear as to whether endogenous or neurogenic factors are strongly influential in compression myelopathy. Further basic and clinical investigations will be needed.
The present study has some limitations. First, the sample size was small, and the severity of myelopathy and the selection of the surgical procedure could contain some bias. Although our previous report has indicated that the recovery rate based on the JOA score was significantly higher in the AM group than in the CM group, unexpectedly, there were no significant differences in our study [
6]. In that study, the main surgical treatments were posterior procedures such as LMP and PDF. However, we mostly chose ADF for the patients in our study. For K-line (−) OPLL, ADF or PDF was considered suitable surgical treatment based on JOA scores to assess recovery compared with LMP [
15]. In addition, in cases of CSM, ADF is recommended for the treatment of multilevel cervical myelopathy compared with LMP [
21]. In fact, the result of this study indicated that the recovery rate of JOA score is significantly higher in ADF. The choice of anterior surgery in our study may have resulted in favorable surgical outcomes in patients in the CM group. Further investigations with single surgical procedures will be needed to clarify the comparison of surgical outcome between AM and CM groups. Second, the definition of AM and CM is subjective because the distinction is only due to the neurological findings. CM contain both the sub-acute phase and real chronic phase. That is the other reason that there was no significant difference in the recovery rate of JOA score between AM and CM. It was unable to speculate surgical outcome of AM/CM only by ROM before surgery because oxidative stress also increased in CM. The definition of CM also may explain this unexpected phenomenon. We speculate that ROM increases until the sub-acute phase and gradually decreases to the normal level if the patient become the real chronic phase. Nevertheless, to our knowledge, there are no biomarkers that distinguish acute, sub-acute, and chronic phase of myelopathy accurately. Further investigation about ROM in combination with the other useful biomarkers will be expected. Third, we lacked data on ROM in serum over time. In this pilot study, we corrected the serum samples only at the time before surgery to clarify whether the ROM before surgery can speculate the surgical outcome. However, the collection of serum samples after surgery is easier than sampling CSF [
6]. In the near future, we plan to conduct the observational study to investigate longitudinal changes in ROM after surgery. Such an approach could be used to determine the ratio of neurogenic oxidative stress associated with compression myelopathy. We speculate that if ROM improve postoperatively, and there is a correlation between them and the recovery rate assessed using JOA scores, oxidative stress levels with a neurogenic origin may be improved by surgery in patients with compression myelopathy. If our hypothesis is proven, it could result in the more effective use of antioxidant medications. Fourth, we don’t investigate the patient-based outcomes like SF-36, EQ-5D, and JOA CMEQ in this study. Although the patient-based outcome is important, the major purpose of this study is to evaluate the recovery of myelopathy. However, those patient-based outcomes may include the residual neck pain or psychogenic factor after surgery. To simplify the result, we investigated only the JOA score. Fifth, details of the mechanism underpinning the elevation of oxidative stress levels in patients with compression myelopathy remain unclear. In addition to neuronal apoptosis [
8] and neurodegenerative changes that can result in increased oxidative stress levels, a recent report has indicated that modic type endplate changes in vertebrae of the lumbar spine and disc degeneration could also cause inflammation and play a pathologic role in compression myelopathy [
22].
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.