Sage Journals: Discover world-class research

Abstract

Study Design

Systematic Review and Meta-Analysis.

Objectives

Degenerative cervical myelopathy (DCM), the leading cause of spinal cord injury globally, may clinically manifest with hand numbness and tingling, loss of hand dexterity and impaired balance and gait. Biomarkers have been identified as disease indicators with recent studies exploring the role of structural and inflammatory biomarkers in diagnostic and treatment outcome prediction. The review aims to identify relevant cerebrospinal fluid (CSF) and blood-based biomarkers associated with DCM diagnosis and prognostication.

Methods

The systematic review was conducted as per PRISMA guidelines. A comprehensive search strategy was devised to identify articles evaluating biomarkers in DCM and comparative control cohorts. The studies were screened by two independent reviewers and discrepancies resolved by a third author. Study characteristics were recorded, and data synthesis was performed to allow for pooled analyses. The quality of included studies and risk of bias were assessed using the QUADAS-2 tool.

Results

A total of 17 studies incorporating 699 individuals with DCM and 475 controls were included. 9 studies evaluated CSF biomarkers while 6 and 2 studies reviewed blood-based and CSF-to-serum quotients, respectively. Pooled analyses revealed significantly higher levels of NF-L and IL-6, albumin-Q, IgA-Q, and IgG-Q in the DCM cohort as compared to the control group.

Conclusion

This systematic review highlights the structural and inflammatory biomarkers that can be potentially used as substitutes for diagnosing and prognosticating outcomes for DCM. The findings further suggest the need for larger prospective studies to assess the clinical relevance of these biomarkers in DCM management.

Keywords

degenerative cervical myelopathy biomarkers cerebrospinal fluid serum systematic review meta-analysis

Introduction

Degenerative cervical myelopathy (DCM) is the leading cause of spinal cord impairment worldwide.^1,2 DCM is a progressive condition characterized by a wide range of neurological symptoms that develop due to spinal cord compression. Compression may be caused by several pathologies including disc degeneration, spondylosis, and ossification of supporting ligaments.³ Common symptoms include numbness and tingling of the hands, which can progress to loss of dexterity, making it difficult for patients to perform fine motor tasks such as buttoning clothes or writing. Affected individuals may also experience arm or leg weakness, leading to difficulties with balance and frequent falls.⁴ Gait disturbances including an unsteady, wide-based walking pattern are often reported, reflecting compromised coordination and proprioception.⁵ Pain in the neck, shoulders, or radiating pain extending into the arms may also occur. In advanced stages, bladder and bowel dysfunction can present due to autonomic pathway involvement.⁴ Early detection of these symptoms is critical to minimize irreversible damage, as prompt intervention may prevent the worsening of neurological deficits. Despite the prevalence of DCM, many patients experience a delay of two or more years before receiving an accurate diagnosis, which often results in postponed treatment and worsened outcomes.^6,7 Early detection is thus essential, as delayed intervention can lead to irreversible neurological impairment.^8,9 DCM is primarily treated with decompression surgery, which in most cases halts disease progression and may even significantly improve symptoms.^10-12

A biomarker can be defined as an “objective indication of medical state observed from outside the patient – which can be measured accurately and reproducibly”.¹³ Recent advancements in biomarker research have shown great promise in enabling earlier identification of medical conditions, such as cancer, sepsis, and concussion.^14-17 Biomarkers also offer the potential for a more personalized approach to treatment, enabling clinicians to offer therapeutic decisions based on unique patient-specific characteristics. Certain unique biomarkers may be recognized in DCM, owing to the underlying pathophysiological mechanisms of the disease. Neuroinflammation induced by the mechanical compression of the tissues and vasculature results in an increase in inflammatory markers, like IL-6 (interleukin-6).¹⁸ Neuronal tissue damage may also result in the release of structural biomarkers in the cerebrospinal fluid (CSF) and peripheral vascular system, such as neuron specific enolase (NSE). Further, the dynamic compression caused by the movement of the head and neck can cause variable spinal cord compression and irritation.^3,4 Altogether, these parallel mechanisms can result in a unique DCM biomarker profile that is specific to each patient’s symptomatology.¹⁹

While proposed DCM biomarkers to date are primarily imaging-based,^12,20 our goal in this systematic review is to identify fluid-based biomarkers that can also be used as reliable diagnostic and/or prognostic tools. In our analysis, we include all studies that assess the relevance of blood- and CSF-based biomarkers. Our results will help prioritize ongoing biomarker research and thus will potentially accelerate clinical translation by identifying specific candidate biomarkers that require further validation.

Methods

This systematic review and meta-analysis study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines.²¹ No ethical approvals were required for this study.

Search Strategy

The aim of the study was to identify the complement of relevant fluid-based biomarkers that can be potentially used for DCM diagnosis and outcome prediction. A comprehensive search strategy was developed, incorporating the synonyms of search terms “degenerative cervical myelopathy”, “spondylotic myelopathy”, “spinal cord compression”, “biomarker*”, “serum”, and “cerebrospinal fluid*”. The databases searched included PubMed, Embase, and Web of Science. When available, relevant controlled vocabulary was also implemented, including MeSH (Medical Subject Headings) for PubMed and Emtree terms for Embase. Study language was restricted to English, but publication dates were not limited. The initial literature search was completed on September 30, 2024. Full search strategies for each database are presented in Supplementary Material.

Study Selection

Two independent reviewers (HA and CW) performed initial title and abstract screening, followed by full-text screening based on the study inclusion and exclusion criteria. The inclusion criteria included all human observational studies that evaluated either quantitative serum, blood, or CSF biomarkers in patients with DCM as compared to a control group regardless of the year of publication. Non-human studies, studies focusing on imaging biomarkers, and publications without quantitative data such as case reports, case series, reviews, technical notes, editorials, and correspondence were all excluded. Additionally, references identified from the included studies that met the eligibility criteria were also reviewed and incorporated. Disagreements regarding potential study eligibility were resolved by a third reviewer (HD).

Data Extraction and Outcome Measures

Baseline study data, including author name, year of publication, sample size, study design, and patient demographics were extracted and recorded by the reviewers. In addition, the types of biomarkers assessed, pre-operative biomarker levels in individuals with DCM and in controls, and any reported correlations between pre-operative biomarker levels and clinical outcomes were documented.

Quality Assessment

The quality assessment for the included studies was conducted using the QUADAS-2 tool,²² which evaluates the risk of bias and applicability concerns across four key domains: patient selection, index measure, reference standard, and flow and timing. The two reviewers independently performed the assessments, with discrepancies resolved through discussion. Each domain was categorized as having a low, high, or unclear risk of bias.

Statistical Analysis

Descriptive statistics were employed to summarize the study variables; measures of central tendency were used for continuous variables while frequencies and proportions were used for categorical variables. Data synthesis was conducted after extraction of biomarker levels from the included studies. Ratios comparing biomarker concentrations between the DCM and control cohorts were calculated where ratios >1 indicated higher biomarker levels in the DCM cohort and ratios <1 signified higher biomarker levels in the control cohort. For studies stratifying subjects based on symptom severity or duration, DCM subgroup biomarker levels were individually compared with control subjects, generating multiple ratios per study. Meta-analyses were conducted using the DerSimonian and Larid random effects method and forest plots were generated to display effect estimates. Heterogeneity was assessed using I² statistic, with thresholds of <50%, 51%–75% and >75% indicating low, moderate, and high heterogeneity, respectively.^23,24 The alpha level was set at 0.05. All statistical analyses were performed using R Studio 4.2.3.²⁵

Results

Literature Search

A total of 3420 abstracts were initially retrieved from the database registries. After excluding duplicates (n = 1086), the remaining articles (n = 2334) underwent title and abstract screening. As a result, 2248 studies were deemed ineligible and were excluded. The remaining 86 articles underwent full-text screening, which resulted in further exclusion of 69 articles. Reasons for exclusion are shown in Figure 1. As such, 17 studies were included in the final analysis.^18,19,26-40

Figure 1.

PRISMA Flowchart Depicting the Abstract and Full-Text Screening Stages and Final Inclusion of 17 Studies From the Initially Identified 3420 Abstracts

Baseline Study Characteristics

Of the 17 included studies, 9 (52.9%) evaluated biomarkers in CSF^26-34 and 6 studies (35.3%) assessed blood-based and serum biomarkers.^18,19,35-38 Two studies focused on blood-spinal cord barrier (BSCB) disruption by analyzing the quotients of albumin, IgA, IgG, and IgM, comparing their respective levels in CSF and serum.^39,40 The majority of the studies were conducted in Japan (n = 8, 47.1%), with additional studies originating in Sweden, China, USA, Germany (n = 2 for each) and Israel (n = 1). A total of 699 individuals with DCM and 475 controls were included across the 17 studies. The mean age of the subjects with DCM and controls ranged between 52.0-70.0 years and 17.9-70.2 years, respectively. The female-to-male ratios across 16 studies were 1:1.44 and 1:0.98 for subjects with DCM and controls, respectively. The mean duration of symptoms ranged between 10 - 28.9 months. Most studies (n = 14) assessed biomarker levels between the overall DCM and control cohorts, while 2 studies stratified subjects with DCM into acute and chronic subgroups based on symptom duration.^30,34 Furthermore, Peng and colleagues classified subjects with DCM into “good” and “fair” prognostic subgroups based on their overall functional improvement (see Table 1).³⁷ The patient cohort for each of the studies was established based on either clinical diagnosis (n = 4), radiological evidence of the disease (n = 3) or a combination of the two (n = 8). Two studies by Imasato and colleagues and Nagashima and colleagues did not mention the diagnostic criteria for patient inclusion within the DCM cohort.^29,33

Table 1.

Characteristics of all 17 Studies Included in the Systematic Review With Descriptions of the CSF and Serum Biomarkers

Author Year	Type of study	Type of Biomarker	DCM: Control n:n	Age (Years)	Sex ratio male: Female	Symptom duration (Months)	Preoperative Clinical assessment	Biomarkers assessed
Country	Type of study	Type of Biomarker	DCM: Control n:n	Mean ± SD	Sex ratio male: Female	Mean ± SD	Preoperative Clinical assessment	Biomarkers assessed
Tsitsopulous	Prospective cohort	CSF	20:63	DCM = 67.7 ± 13	DCM = 12:8	26.8 ± 35.1	-	GFAP, NF-L, Tau, FABP3, sAPP-α and sAPP-β, Amyloid β38, β40 and β42
2021				DCM = 67.7 ± 13	DCM = 12:8
2021				C = 65.2 ± 14.5	C = 28:35
Sweden²⁶				C = 65.2 ± 14.5	C = 28:35
Tsitsopulous	Prospective cohort	CSF	23:7	DCM = 66.4 ± 12.8	DCM = 12:11	25.3 ± 33.4	EQ-5D index = 0.3 ± 0.3	GFAP, pNF-H, NF-L and UCH-L1.
2022				DCM = 66.4 ± 12.8	DCM = 12:11		Preoperative ASIA impairment = 7:16 (C:D)
2022				C = 45.4 ± 5.3	C = 4:3
Sweden²⁷				C = 45.4 ± 5.3	C = 4:3
Tsukahara	Retrospective cohort	CSF	23:10	DCM = 67.4 ± 10.0	DCM = 18:5	14.8 ± 14.4	-	Neuron specific enolase
2021				DCM = 67.4 ± 10.0	DCM = 18:5
2021				C = 67.0 ± 14.5	C = 12:16
Japan²⁸				C = 67.0 ± 14.5	C = 12:16
Nagashima	Prospective cohort	CSF	21:6	DCM = 66.2 ± 8.5	DCM = 17:4	28.9 ± 35.5	JOA = 10.0 ± 2.1	IL-1b, IL-6 and TNF-α
2009				DCM = 66.2 ± 8.5	DCM = 17:4
2009				C = 41.5 ± 16.5	C = 5:1
Japan²⁹				C = 41.5 ± 16.5	C = 5:1
Takahashi	Prospective cohort	CSF	40:29	DCM_A = 65.4 ± 13.0	DCM_A = 11:9	-	Preoperative JOA	pNF-H, Tau, MBP, NSE and S100b
2018							DCM_A = 8.4 ± 2.9
Japan³⁰
					DCM_C = 15:5
							DCM_C = 11.2 ± 2.3
				DCM_C = 65.6 ± 11.8
				DCM_C = 65.6 ± 11.8	C = 11:18
				C = 70.4 ± 7.3	C = 11:18
Tsuji	Prospective cohort	CSF	29:34	DCM = 60 ± 10.5	-	-	-	Chondrex
2002				DCM = 60 ± 10.5
Japan³¹				C = 38 ± 18.3
Ito	Retrospective cohort	CSF	35:12	DCM = 55.9 ±13.4	DCM = 27:8	-	Preoperative JOA - 10.9 ± 3.3	IL-8, IL-10, and TNF-α
2008				DCM = 55.9 ±13.4	DCM = 27:8
2008				C = 53.9 ±17.5	C = 7:5
Japan³²				C = 53.9 ±17.5	C = 7:5
Imasato	Cross-Sectional	CSF	46:12	-	-	-	-	SPLI and β-ENDLI
1997
Japan³³
Takahashi	Prospective cohort	CSF	14:37	DCM_A = 64.9 ± 10.2	DCM_A = 4:4		Preoperative JOA	pNF-H
Takahashi				DCM_A = 64.9 ± 10.2	DCM_A = 4:4		DCM_A = 9.3 ± 2.4
2014				DCM_C = 65.0 ± 13.2	DCM_C = 5:1		DCM_A = 9.3 ± 2.4
2014					C = 14:23		DCM_C = 10.6 ± 0.8
Japan³⁴
Japan³⁴				C = 70.3 ± 7.9
Akshota	Cross-Sectional	Serum	25:13	DCM = 55 ± 13.9	DCM = 18:7	-	mJOA Category –	BDNF, Cathepsin D, MPO, NCAM, PAI-1, PDGFAA, PDGF-AB/BB, RANTES, sICAM-1, sVCAM-1, S100 B, Amyloid β40, sRAGE, and GDNF
2020				DCM = 55 ± 13.9	DCM = 18:7		Mild = 15
				C = 49.6 ± 15.9	C = 7:6		Mild = 15
							Moderate = 7
Israel³⁵							Moderate = 7
Israel³⁵							Severe = 3
Kato	Prospective cohort	Serum	26:8	DCM = 64.4 ± 12	DCM = 19:7	27.7 ± 39.1	-	pNF-H
2015					DCM = 19:7
					C = 1:7
				C = 17.9 ± 4	C = 1:7
Japan³⁶				C = 17.9 ± 4	C = 1:7
Peng	Prospective cohort	Plasma	236:110	DCM_GP = 59.2 ± 3.6	DCM = 135:101	-	Preoperative JOA	VDBP, 25(OH)D and GSH
				DCM_GP = 59.2 ± 3.6			DCM_GP = 8.4 ± 2.3
				DCM_FP = 58.8 ± 4.2
2018
2018					C = 58:52
China³⁷
China³⁷				C = 58.6 ± 4.4			DCM_FP = 9.2 ± 1.7
Du	Prospective cohort	Serum	40:10	DCM = 51.9 ± 7.7	DCM = 18:22	26.0 ± 8.7	JOA = 10.6	IL-6
2018				DCM = 51.9 ± 7.7	DCM = 18:22
2018				C = 57.5 ± 8.4	C = 5:5
China¹⁸				C = 57.5 ± 8.4	C = 5:5
Vedantam	Prospective cohort	Serum	20:10	DCM = 66.3 ± 11.0	DCM = 10:10	-	mJOA = 14 ± 1.5	BDNF, NF-L, GFAP, AB40, Aβ42, Tau, IFN-ϒ, IL-1β, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-22 and TNFα
2024							NDI = 28 ± 17.2
							QuickDASH = 40.8 ± 17.3
				C = 64.4 ± 8.1	C = 5:5
USA¹⁹
USA¹⁹							SF-36 = 36.5 ± 7.8
Divi	Prospective cohort	Serum	36:35	DCM = 61.5 ± 9.5	DCM = 13:23	-	mJOA = 13.3 ± 1.5	TNFα, IL-8, IL-13, IL-12p70, IL-4, IL-1β, IL-2, IL-6, IL-10, IFN-ϒ
2024				DCM = 61.5 ± 9.5	DCM = 13:23		VAS Neck = 5.1 ± 0.3
				C = 57.5 ± 8.9	C = 11:24		VAS Neck = 5.1 ± 0.3
							NDI = 31.9 ± 2.3
USA³⁸							NDI = 31.9 ± 2.3
Schmidt	Prospective cohort	CSF and serum	44:46	DCM = 62.9 ± 11.2	DCM = 29:21	-	mJOA = 10.7 ± 2.8	Albumin-, IgA-, IgG- and IgM-quotient
2023							mJOA = 10.7 ± 2.8
							NDI = 40.4 ± 20.8
				C = 61.1 ± 16.7	C = 35:17
Germany³⁹				C = 61.1 ± 16.7	C = 35:17
Blume	Prospective cohort	CSF and serum	21:33	DCM = 63.3 ± 11.6	DCM = 9:12	-	mJOA = 10.1 ± 3.2	Albumin-, IgA-, IgG- and IgM-quotient
				DCM = 63.3 ± 11.6	C = 21:12		NDI = 46.4 ± 21.0
				C = 62.3 ± 15.2
2020
Germany⁴⁰

Abbreviations: ǂ SD, Standard Deviation; DCM, Myelopathy; C, Control; DCM_A, Acute Myelopathy; DCM_C, Chronic Myelopathy; DCM_GP, Good Prognosis; DCM_FP, Fair Prognosis; EQ-5D, EuroQol 5-Dimension Health Outcome Measure; ASIA, American Spinal Injury Impairment Scale; JOA, Japanese Orthopedic Association; mJOA, modified Japanese Orthopedic Association Scale; NDI, Neck Disability Index; QuickDash, Quick Disabilities of the Arm, Shoulder, and Hand Outcome Measure; SF-36, Short Form (36) Health Survey; VAS, Visual Analog Scale; GFAP, Glial Fibrillary Acidic Protein; NF-L, Neurofilament Light Chain; Tau, Tau Protein; FABP3, Fatty Acid Binding Protein 3; sAPP-α, Soluble Amyloid Precursor Protein - Alpha; sAPP-β, Soluble Amyloid Precursor Protein – Beta; Amyloid β-38, β40, β42, Amyloid Beta Peptides – 38, 40 and 42; pNF-H, Phosphorylated Neurofilament Heavy Chain; UCH-L1, Ubiquitin C-Terminal Hydrolase L1; NSE, Neuron Specific Enolase; IL, Interleukin; TNF-α, Tumor Necrosis Factor Alpha; MBP, Myelin Basic Protein; S100b, S100 Calcium binding Protein B; YLK-40, Chitinase-3-Like Protein1 (HC gp-39); SPLI, Substance P Like Immunoreactivity; β-ENDLI, Beta-Endorphin Like Immunoreactivity; BDNF, Brain Derived Neurotrophic Factor; MPO, Myeloperoxidase; NCAM, Neural Cell Adhesion Molecule; PAI-1, Plasminogen Activator Inhibitor-1; PDGF, Platelet Derived Growth Factor; RANTES, Regulated upon Activation, Norma T Cell Expressed and Secreted; sICAM-1, Soluble Intracellular Adhesion Molecule-1; sVCAM-1, Soluble Vascular Cell Adhesion Molecule-1; sRAGE, Soluble Receptor for Advance Glycation End Products; GDNF, Glial Cell-Derived Neurotrophic Factor; VDBP, Vitamin D Binding Protein; 25(OH)D, 25 Hydroxyvitamin D; GSH, Glutathione; IFN-ϒ, Interferon Gamma; Ig, Immunoglobulin.

Cerebrospinal Fluid Biomarkers

Among 9 studies evaluating CSF, a total of 19 biomarkers were identified and compared between the DCM and control cohorts.^26-34 pNF-H (phosphorylated neurofilament heavy subunit) was assessed in 3 studies, while NF-L (neurofilament light), GFAP (glial fibrillary acidic protein), tau, NSE, and YLK-40 were assessed in 2 studies each. Additionally, S100 B was analyzed in a single study by Takahashi and colleagues, with subjects stratified into acute and chronic subgroups. All other biomarkers were assessed in a single study without any stratification (see Table 2).

Table 2.

Corresponding Biomarker Levels in the DCM and Control Cohorts

Source	Biomarker	Study, Year	DCM Group	Control Group
Source	Biomarker	Study, Year	Preoperative levels	Preoperative levels
CSF	NF-L	Tsitsopulous, 2021²⁶	10 156 ± 14 422.7 pg/mL	1204.4 ± 1212.4 pg/mL
	NF-L	Tsitsopulous, 2022²⁷	1097.7 ± 1360.3 pg/mL	116.4 ± 83.5 pg/mL
	GFAP	Tsitsopulous, 2021²⁶	946.7 ± 606.5 pg/mL	302.3 ± 214.2 pg/mL
	GFAP	Tsitsopulous, 2022²⁷	2593.6 ± 1656.1 pg/mL	1593.3 ± 836.9 pg/mL
	Tau	Tsitsopulous, 2021²⁶	292.9 ± 112.6 pg/mL	358.7 ± 139.5 pg/mL
		Takahashi, 2018³⁰	DCM_A = 287.0 ± 161.8 pg/mL	270.6 ± 105.7 pg/mL
		Takahashi, 2018³⁰	DCM_C = 193.5 ± 81.6 pg/mL	270.6 ± 105.7 pg/mL
	pNF-H	Tsitsopulous, 2022²⁷	1628.3 ± 2765.6 pg/mL	134.6 ± 86.8 pg/mL
		Takahashi, 2018³⁰	DCM_A = 1907.8 ± 730.4 pg/mL	462.9 ± 913.9 pg/mL
		Takahashi, 2018³⁰	DCM_C = 198.6 ± 124.5 pg/mL	462.9 ± 913.9 pg/mL
		Takahashi, 2014³⁴	DCM_A = 2127.1 ± 556.8 pg/mL	518.7 ± 665.7 pg/mL
		Takahashi, 2014³⁴	DCM_C = 175.8 ± 27.5 pg/mL	518.7 ± 665.7 pg/mL
	NSE	Tsukahara, 2021²⁸	13.7 ± 4.3 ng/mL	13.6 ± 4.0 ng/mL
		Takahashi, 2018³⁰	DCM_A = 2.02 ± 0.68 ng/mL	2.97 ± 0.81 ng/mL
		Takahashi, 2018³⁰	DCM_C = 2.04 ± 0.66 ng/mL	2.97 ± 0.81 ng/mL
	S100 B	Takahashi, 2018³⁰	DCM_A = 192.7 ± 104.8 pg/mL	156.3 ± 35.5 pg/mL
	S100 B	Takahashi, 2018³⁰	DCM_C = 144.2 ± 53.0 pg/mL	156.3 ± 35.5 pg/mL
	YKL-40	Tsitsopulous, 2021²⁶	171 167.5 ± 64 013.6 pg/mL	147 804.9 ± 54 066.5 pg/mL
	YKL-40	Tsuji, 2002³¹	245.3 ± 107.2 ng/mL	113.8 ± 48.3 ng/mL
	FABP3	Tsitsopulous, 2021²⁶	2.4 ± 1.5 pg/mL	5.83 ± 2.8 pg/mL
	sAPPα	Tsitsopulous, 2021²⁶	196.7 ± 113.1 pg/mL	806.9 ± 312.4 pg/mL
	sAPPβ	Tsitsopulous, 2021²⁶	122.6 ± 74.6 pg/mL	502.5 ± 189.7 pg/mL
	Aβ38	Tsitsopulous, 2021²⁶	1446.5 ± 553.2 pg/mL	2047.2 ± 565.7 pg/mL
	Aβ40	Tsitsopulous, 2021²⁶	4241.5 ± 1444.5 pg/mL	13 355.5 ± 3667.9 pg/mL
	Aβ42	Tsitsopulous, 2021²⁶	436.0 ± 185.2 pg/mL	1355.7 ± 432.6 pg/mL
	UCH-L1	Tsitsopulous, 2022²⁷	828.4 ± 2093.9 pg/mL	129.9 ± 108.4 pg/mL
	IL-β	Nagashima, 2009²⁹	0.35 pg/mL	1.17 pg/mL
	IL-6	Nagashima, 2009²⁹	4.58 pg/mL	1.96 pg/mL
	IL-8	Ito, 2008³²	69.0 ± 35.2 pg/mL	43.5 ± 20.9 pg/mL
	SPLI	Imasato, 1997³³	33.9 ± 3.1 pmol/l	18.5 ± 3.6 pmol/l
	β-ENDLI	Imasato, 1997³³	8.6 ± 0.8 pmol/l	9.6 ± 0.7 pmol/l
Serum	BDNF	Akshota, 2020³⁵	2.9800 ± 1.2100	1.3900 ± 0.540
	BDNF	Vedantam, 2024¹⁹	18 729 pg/mL	22 969.5 pg/mL
	AB40	Akshota, 2020³⁵	25.68	7.08
	AB40	Vedantam, 2024¹⁹	64.9 pg/mL	67.8 pg/mL
	IL-6	Du, 2018¹⁸	0.8 ± 0.5 pg/mL	0.5 ± 0.3 pg/mL
		Vedantam, 2024¹⁹	2.9 pg/mL	1.2 pg/mL
		Divi, 2024³⁸	2.09 pg/mL	1.24 pg/mL
	IL-8	Vedantam, 2024¹⁹	19.5 pg/mL	17 pg/mL
	IL-8	Divi, 2024³⁸	21.48 pg/mL	10.87 pg/mL
	IL-10	Vedantam, 2024¹⁹	0.5 pg/mL	0.5 pg/mL
	IL-10	Divi, 2024³⁸	1.13 pg/mL	0.34 pg/mL
	TNF-α	Du, 2018¹⁸	2.4 ± 0.3 pg/mL	2.2 ± 0.2 pg/mL
		Vedantam, 2024¹⁹	0.5 pg/mL	0.7 pg/mL
		Divi, 2024³⁸	3.56 pg/mL	2.72 pg/mL
	IFN-ϒ	Du, 2018¹⁸	4.7 ± 0.7 pg/mL	5.0 ± 1.1 pg/mL
	IFN-ϒ	Divi, 2024³⁸	6.33 pg/mL	7.09 pg/mL
	MPO	Akshota, 2020³⁵	654.08 ± 313.11	368.86 ± 283.2
	PDGF-AA	Akshota, 2020³⁵	1.2600 ± 0.4700	0.5500 ± 0.300
	NCAM	Akshota, 2020³⁵	72.100 ± 25.940	24.300 ± 15.55
	Svcam-1	Akshota, 2020³⁵	231.17 ± 61.480	114.44 ± 28.55
	PAI-1	Akshota, 2020³⁵	69.460 ± 29.480	29.480 ± 11.87
	Cathepsin D	Akshota, 2020³⁵	31.92	5.03
	sICAM-1	Akshota, 2020³⁵	54.74	14.69
	RANTES	Akshota, 2020³⁵	9.12	4.72
	PDGF-AB/BB	Akshota, 2020³⁵	0.98	0.06
	S100 B	Akshota, 2020³⁵	34.3	31.83
	sRAGE	Akshota, 2020³⁵	56.64	42.87
	GDNF	Akshota, 2020³⁵	150	60
	Nf-L	Vedantam, 2024¹⁹	30.2 pg/mL	11.2 pg/mL
	GFAP	Vedantam, 2024¹⁹	132 pg/mL	103 pg/mL
	Tau	Vedantam, 2024¹⁹	0.23 pg/mL	0.25 pg/mL
	IL-22	Vedantam, 2024¹⁹	0.8 pg/mL	0.5 pg/mL
	IL-13	Divi, 2024³⁸	1.05 pg/mL	0.6 pg/mL
	IL-12p70	Divi, 2024³⁸	0.22 pg/mL	0.12 pg/mL
	IL-4	Divi, 2024³⁸	0.01 pg/mL	0.02 pg/mL
	IL-2	Divi, 2024³⁸	0.38 pg/mL	0.74 pg/mL
	IL-1β	Divi, 2024³⁸	0.03 pg/mL	0.12 pg/mL
	pNF-H	Kato, 2015³⁶	Detectable in 6 subjects	Non-detectable
Plasma	VDBP	Peng, 2018³⁷	DCM_GP = 108.40 ± 6.70 μg/mL	141.30 ± 7.10 μg/mL
	VDBP	Peng, 2018³⁷	DCM_FP = 106.60 ± 5.60 μg/mL	141.30 ± 7.10 μg/mL
	25(OH) Vitamin D	Peng, 2018³⁷	DCM_GP = 31.60 ± 2.12 ng/mL	35.70 ± 1.73 ng/mL
	25(OH) Vitamin D	Peng, 2018³⁷	DCM_FP = 31.10 ± 1.31 ng/mL	35.70 ± 1.73 ng/mL
	Glutathione	Peng, 2018³⁷	DCM_GP = 4.23 ± 0.34 μmol/mHb	6.44 ± 0.54 μmol/mHb
	Glutathione	Peng, 2018³⁷	DCM_FP = 4.05 ± 0.55 μmol/mHb	6.44 ± 0.54 μmol/mHb
CSF and serum	Albumin-Q	Schmidt, 2023³⁹	13.79	5.14
	Albumin-Q	Blume, 2020⁴⁰	10.0 ± 3.5	4.8 ± 1.9
	IgA-Q	Schmidt, 2023³⁹	5.9	2.63
	IgA-Q	Blume, 2020⁴⁰	2.7 ± 1.2	1.5 ± 0.8
	IgG-Q	Schmidt, 2023³⁹	4.11	1.52
	IgG-Q	Blume, 2020⁴⁰	4.7 ± 1.7	2.4 ± 0.9
	IgM-Q	Schmidt, 2023³⁹	0.7	0.59
	IgM-Q	Blume, 2020⁴⁰	0.6 ± 0.3	0.7 ± 0.6

Abbreviations: ǂ DCM, Myelopathy; C, Control; DCM_A, Acute Myelopathy; DCM_C, Chronic Myelopathy; DCM_GP, Good Prognosis; DCM_FP, Fair Prognosis; GFAP, Glial Fibrillary Acidic Protein; NF-L, Neurofilament Light Chain; Tau, Tau Protein; FABP3, Fatty Acid Binding Protein 3; sAPP-α, Soluble Amyloid Precursor Protein - Alpha; sAPP-β, Soluble Amyloid Precursor Protein – Beta; Amyloid β-38, β40, β42, Amyloid Beta Peptides – 38, 40 and 42; pNF-H, Phosphorylated Neurofilament Heavy Chain; UCH-L1, Ubiquitin C-Terminal Hydrolase L1; NSE, Neuron Specific Enolase; IL, Interleukin; TNF-α, Tumor Necrosis Factor Alpha; MBP, Myelin Basic Protein; S100b, S100 Calcium binding Protein B; YLK-40, Chitinase-3-Like Protein1 (HC gp-39); SPLI, Substance P Like Immunoreactivity; β-ENDLI, Beta-Endorphin Like Immunoreactivity; BDNF, Brain Derived Neurotrophic Factor; MPO, Myeloperoxidase; NCAM, Neural Cell Adhesion Molecule; PAI-1, Plasminogen Activator Inhibitor-1; PDGF, Platelet Derived Growth Factor; RANTES, Regulated upon Activation, Norma T Cell Expressed and Secreted; sICAM-1, Soluble Intracellular Adhesion Molecule-1; sVCAM-1, Soluble Vascular Cell Adhesion Molecule-1; sRAGE, Soluble Receptor for Advance Glycation End Products; GDNF, Glial Cell-Derived Neurotrophic Factor; VDBP, Vitamin D Binding Protein; 25(OH)D, 25 Hydroxyvitamin D; GSH, Glutathione; IFN-ϒ, Interferon Gamma; Ig, Immunoglobulin; pg, picogram; ng, nanogram; μg, microgram; pmol, picomole; μmol, micromole; mHb, millimole of hemoglobin; ml, milliliter; l, liter.

ǂǂ SI units were not provided in the study by Akshota et al.

Pooled analyses of CSF biomarkers revealed significantly higher levels of NF-L (9.01; 95% CI = 1.03-78.62; P = .047) within the diseased population as compared to the control group. No significant differences were observed for GFAP (2.16; 95% CI = 0.63-7.35; P = .219), NSE (0.78; 95% CI = 0.48-1.26; P = .308), pNF-H (1.76; 95% CI = 0.41-7.67; P = .449), S100B (1.02; 95% CI = 0.51-2.02; P = .955), Tau (0.83; 95% CI = 0.34-1.86 P = .595) or YLK-40 (1.66; 95% CI = 0.15-18.93; P = .684) between the two groups (see Figure 2).

Figure 2.

Pooled Analyses of the Ratios of the Cerebrospinal Fluid Biomarkers Amongst the DCM and Control Cohorts. (A) Ratio of 1 Indicates No Difference Amongst the DCM and Control Populations, while Higher or Lower Values Indicate Relatively Increased and Decreased Biomarker Concentrations in the DCM Population. (A) Glial Fibrillary Acidic Protein; B. Neuron Specific Enolase; C. NF-L (Neurofilament Light Chain); D. pNF-H (Phosphorylated Neurofilament Heavy Chain); E. S100 B; F. Tau Protein; G. Chondrex

Blood-Based Biomarkers

Thirty-two distinct biomarkers were evaluated across 6 studies in the serum of subjects with DCM and in controls.^18,19,35-38 IL-6 and TNF-α (tumor necrosis factor – α) were evaluated in 3 studies each. Similarly, BDNF (brain-derived neurotrophic factor), Aβ40, IL-8, IL-10, and IFN-ϒ (interferon-gamma) were each assessed in 2 studies, while all remaining biomarkers were evaluated in single studies (see Table 2).

Pooled analyses revealed significantly higher levels of IL-6 (1.85, 95% CI = 1.26-2.72, P = .002) and nearly significantly higher IL-8 levels (1.65, 95% CI = 1.00-2.73, P = .051) in subjects with DCM as compared to controls. Finally, no significant differences were observed in Aβ40 (1.89; 95% CI = 0.51-6.95; P = .339), BDNF (1.15; 95% CI = 0.46-2.83; P = .768), IL-10 (1.90; 95% CI = 0.59-6.13; P = .285), TNF-α (1.10; 95% CI = 0.86-1.39; P = .460) or IFN-γ (0.91; 95% CI = 0.65-1.29; P = .600) between the two cohorts (see Figure 3).

Figure 3.

Pooled Analyses of the Ratios of Serum Biomarkers Amongst the DCM and Control Cohorts. (A) Ratio of 1 Indicates No Difference Amongst the DCM and Control Populations, while Higher or Lower Values Indicate Relatively Increased or Decreased Biomarker Concentrations. (A) Amyloid Beta 40; B. Brain Derived Neurotrophic Factor; C. Interleukin-6; D. Interleukin-8; E. Interleukin-10; F. Tumor Necrosis Factor Alpha; G. Interferon Gamma

CSF-Serum Quotients

Two studies evaluated the levels of albumin, IgA, IgG, and IgM (immunoglobulin – A, G and M, respectively) in both CSF and serum,^39,40 comparing the corresponding quotients to assess BSCB integrity (see Table 2). Pooled analyses demonstrated significantly higher quotients in subjects with DCM as compared to controls – albumin-Q (2.60, 95%CI = 1.81-3.74, P < .001), IgA-Q (2.20, 95%CI = 1.52-3.19, P < .001), and IgG-Q (2.59, 95%CI = 1.81-3.72, P < .001). However, IgM-Q (1.17, 95%CI = 0.80-1.62, P = .240) levels were comparable between DCM cases and controls (See Figure 4).

Figure 4.

Description of the Pooled Analyses of the Ratios of the Biomarkers Amongst the DCM and Control Cohorts. A Ratio of 1 Indicates No Difference Amongst the DCM and Control Cohorts, while Higher or Lower Levels Indicate Increased or Decreased Biomarker Concentrations. A = Albumin CSF-To-Serum Quotient; B = IgA CSF-To-Serum Quotient; C = IgG CSF-To-Serum Quotient; D = IgM CSF-To-Serum Quotient

Risk of Bias Assessments

An unclear to high risk of bias was identified in the flow and timing component, primarily due to the general lack of follow-up data. Patient selection was largely representative of the myelopathy population, while the index measures were well defined in the majority of studies, utilizing pre-established biomarker assessment methodologies. The reference standard for diagnosing myelopathy consistently incorporated clinical and radiological criteria. Regarding applicability concerns, the variability in control populations introduced uncertainty in the patient selection domain, leading to unclear applicability in a subset of studies (see Figure 5).

Figure 5.

Quality Assessment of the Included Studies Using the QUADAS-2 Tool. The Risk of Bias and Applicability were Evaluated Across Four Key Domains: Patient Selection, Index Measure, Reference Standard, and Flow and Timing

Discussion

SCI is characterized by initial structural trauma followed by several secondary injury processes including progressive hemorrhagic necrosis, programmed cell death, and inflammation. DCM represents a non-traumatic form of SCI involving multiple parallel and chronic pathophysiological processes, including neuronal/axonal degeneration and neuroinflammation, often leading to progressive neurological deterioration.^41,42 Two major categories of biomarkers – structural and inflammatory - have been identified in both traumatic and non-traumatic forms of SCI^43-46 and in other spinal conditions such as in degenerative disc disease.^47-49 While the underlying pathophysiological mechanisms between these conditions are distinct, there is considerable overlap at the cellular and molecular levels in terms of cellular damage.

Our study represents the first comprehensive effort to identify clinically relevant CSF and blood-based biomarkers with potential utility in the diagnosis and severity stratification of DCM. Among the 17 included studies assessing structural and inflammatory biomarkers in CSF or blood, we identified significantly higher levels of NF-L and IL-6, and IL-8 level increases also reached near significance in DCM patients as compared to controls. Additionally, the CSF-to-serum quotients of albumin, IgA, and IgG, were all significantly increased in subjects with DCM, consistent with potential disruption of the BSCB in this condition.^39,40

As an extension of the blood brain barrier, the BSCB is primarily comprised of capillary endothelium, a supporting basement membrane, pericytes, and astrocytes.⁵⁰ Following traumatic SCI, disruption of the BSCB leads to increased permeability,^51,52 initiating a secondary cascade of biochemical reactions leading to spinal cord edema and further disruption of BSCB integrity.^2,42 In DCM, chronic BSCB disruption similarly allows for protein leakage into CSF and serum indicative of neuronal and astrocytic injury and demyelination.^26,53,54

The initial insult to neuronal and glial cells also triggers the release of proinflammatory cytokines that in turn promote substantial immune cell infiltration. In a rat model of traumatic SCI, increased levels of IL-1, IL-6, IL-10, and TNF-α were initially observed within 6 hours of compression injury while only IL-6 and TNF-α showed elevated levels persisting even after a week.^55-57 Elevated serum IL-6 is thought to be indicative of ongoing neural and/or surrounding disco-ligamentous element degeneration, which are also common findings in DCM.¹⁹

We observed significant cross-study variability in CSF and blood-based biomarker concentrations, which is likely attributable to both heterogeneity of disease severity as well as duration. In 2 studies by Takahashi and colleagues, DCM was classified into acute and chronic groups, based on the duration of symptoms. Significantly higher CSF levels of pNF-H and tau proteins were observed in the acute as compared to the chronic DCM groups.^30,34 Similarly, Tsitsopulous and colleagues reported elevated CSF levels of NF-L in patients in the acute phase as compared to patients with chronic symptoms, indicating the potential role of NF-L in reflecting axonal damage as well as neuroplasticity.²⁶ Kato and colleagues observed a similar trend with pNF-H being more elevated in the acute group.³⁶ Taken together, these observations may indicate that chronic compression may render neural tissue nonviable over time, making it increasingly difficult to detect biomarker traces in the CSF and serum. As such, given the heterogenous findings across the pooled studies and the limited ability of CSF and blood-based biomarkers to consistently reflect overall disease severity, our findings underscore the need for subject stratification by symptom severity and duration when assessing biomarker levels.

Recent studies have evaluated the potential role of RNA and microRNA levels in the diagnosis of DCM. Laliberte and colleagues identified an association between plasma MIR21-5p levels, symptom severity, and disease progression in patients with DCM; elevated levels correlated with moderate-severe symptoms and worse prognoses at 1-year follow-up.⁵⁸ Divi and colleagues identified altered expression in 12 of 15 (80%) analyzed serum microRNAs with specific upregulation of miR-223-3p and downregulation of miR-451a, as reported in various neurodegenerative conditions.³⁸ In a separate study, Zheng and colleagues developed a predictive model for diagnosing DCM using a five-gene model that included blood-based levels of TBCD, TPM2, PNKD, EIF4G2, and AP5Z1, achieving an accuracy of 93.5%.⁵⁹ Single-gene models utilizing TCAP or SDHA genes demonstrated accuracies of 83.3% and 76.7%, respectively. Additionally, recent studies have explored the role of ApoE in DCM prognosis. Desmione and colleagues reported poorer postsurgical neurological recovery in carriers of the ApoE4 allele.⁶⁰ Similarly, in another study, Chintapalli and colleagues demonstrated higher odds of DCM prevalence in individuals homozygous for ε4 allele, using a large biobank analysis.⁶¹ These findings provide a foundation for future studies assessing putative DCM molecular markers.

In patients with traumatic brain injury, the concept of a “liquid brain biopsy” has been proposed, wherein protein biomarkers are utilized for stratification with both therapeutic and prognostic implications.⁶² Several studies have also demonstrated the prognostic value of specific CSF and blood-based biomarkers in determining neurologic outcomes after traumatic SCI.^63-66 Similarly, the studies included in this systematic review have also examined the relationship between biomarker levels and outcomes. Tsitsopulous and colleagues observed a negative correlation between CSF GFAP and clinical outcomes, while NF-L and FABP3 (Fatty Acid Binding Protein 3) showed positive correlations with clinical outcomes at 3-months post-operatively.^26,27 Further, elevated pre-operative levels of pNF-H were associated with poorer surgical outcomes at follow-up.^30,35 Peng and colleagues reported favorable post-operative outcomes in DCM patients with either elevated baseline plasma VDBP (vitamin D binding protein), 25(OH)D (25-hydroxyvitamin D), or GSH (glutathione) levels.³⁷ Future research should explore the development of a biomarker panel that not only aids in the diagnosis of DCM but also serves as a prognostication tool of post-operative outcomes.

Although our study provides insights into the differential concentration of various biomarkers between the DCM and control cohorts, it is important to acknowledge that small sample sizes for biomarker concentration measurements may introduce sampling bias, thereby limiting the reliability and generalizability of the findings. In fact, 12 of the 19 candidate CSF and 25 of the 32 serum biomarkers were each assessed in a single study. Moreover, inherent variability of DCM and control cohorts, lack of standardized inclusion criteria, sample type and collection, as well as storage and biomarker measurement techniques across the included studies further complicates the interpretation of the effect sizes due to the underlying heterogeneity (see Table 3). Furthermore, 8 of the 17 included studies originated in Japan, which potentially limits the generalizability of the findings due to putative population-specific variations in biomarker profiles. Of note, stratification or meta-regression based on symptom severity, baseline demographics and control type were not feasible due to the limited number of studies evaluating each biomarker.⁶⁷ Moreover, although QUADAS-2 is designed to evaluate diagnostic accuracy in studies, we also applied it to assess the risk of bias in studies investigating the prognostic role of biomarkers. These limitations underscore the need for future larger prospective studies to explore the baseline and post-operative variability of fluid biomarker levels in subjects with DCM.

Table 3.

Various Control Groups Identified Across the Included Studies

Author, Year	Diagnostic Criteria for DCM Cohort	Control Group
Tsitsopulous 2021²⁶	Progressive clinical symptoms and radiological evidence	Elective orthopedic surgeries under spinal anesthesia
Tsitsopulous 2022²⁷	Progressive clinical symptoms and radiological evidence	Cervical radiculopathy
Tsukahara 2021²⁸	Presence of clinical symptoms and radiological evidence of spinal cord compression	Diagnostic lumbar puncture to rule out neurological conditions
Nagashima 2009²⁹	Not defined	Osteosynthesis
Takahashi 2018³⁰	Clinical and radiological evidence	Lumbar canal stenosis
Tsuji 2002³¹	Radiological evidence of disease on MRI	Post-traumatic leg fractures
Ito 2008³²	Radiological evidence of disease on MRI	Post-traumatic or Myoma uteri
Imasato 1997³³	Not defined	Healthy controls
Takahashi 2014³⁴	Clinical symptomatology	Lumbar canal stenosis
Akshota 2020³⁵	Clinical symptoms and radiological evidence on MRI	Healthy controls
Kato 2015³⁶	Radiological evidence of disease on MRI	Idiopathic scoliosis
Peng 2018³⁷	Clinical symptoms and radiological evidence of disease	Healthy controls
Du 2018¹⁸	Clinical diagnosis and radiological evidence	Healthy controls
Vedantam 2024¹⁹	Clinical symptoms and evidence of disease on MRI	Healthy controls
Divi 2024³⁸	Clinical symptoms of disease	Healthy controls
Schmidt 2023³⁹	Clinical symptoms of disease	Thoracic aortic aneurysm undergoing surgery
Blume 2020⁴⁰	Clinical symptoms of disease	Thoracic aortic aneurysm undergoing surgery

ǂ DCM, Degenerative Cervical Myelopathy; MRI = Magnetic Resonance Imaging.

Several study strengths are noteworthy, including the comprehensive narrative provided by our analysis on the potential utility of various biomarkers in DCM diagnosis, stratification, and prognostication. The meta-analysis further compared relative biomarker levels between DCM and control subjects, offering insights that could enhance the diagnostic accuracy. However, the study was limited by relatively small sample sizes, variable diagnostic criteria for DCM, and the absence of standardized biomarker measurement protocols across studies, which likely further contributed to variability beyond study-specific differences in disease severity and duration. Future larger studies, potentially incorporating randomized designs and stratification by DCM disease duration and severity, could clarify the clinical relevance of CSF and serum-based biomarkers in DCM.

Conclusion

DCM is associated with injury to the spinal cord via chronic compression. Levels of several structural and inflammatory biomarkers have been shown to be significantly altered across studies including NF-L, IL-6, albumin-Q, IgA-Q, and IgG-Q. Nevertheless, marked variability in symptom duration and severity across individuals and studies has thus far limited the ability of these biomarkers to serve as reliable surrogate measures of disease severity and prognosis. By incorporating prospective and unbiased methodologies, we anticipate that future larger studies will identify sensitive and specific biomarker panels with diagnostic and prognostic utility.

Supplemental Material

Supplemental Material - Assessment of Cerebrospinal Fluid and Blood-Based Biomarkers in Degenerative Cervical Myelopathy: Systematic Review and Meta-Analysis

Supplemental Material for Assessment of Cerebrospinal Fluid and Blood-Based Biomarkers in Degenerative Cervical Myelopathy: Systematic Review and Meta-Analysis by Harshit Arora, Hassan Darabi, Cassandra C Wolsh, Kelsey Karnik, John C. Gensel, Francis Farhadi in Systematic Review

Footnotes

Harshit Arora

Hassan Darabi

Francis Farhadi

Author Contributions

HA, HD and CW conceptualized and designed the study and were involved in data acquisition. Statistical analyses were conducted by HA, HD and KK and analysis and interpretation of data was done by HA. Supervision across all stages of the study and critical revision of the manuscript was done by JCG and FF.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data for the current study are made available from the corresponding author on a reasonable request.*

Supplemental Material

Supplemental material for this article is available online.

References

Nouri

Martin

Kato

Reihani-Kermani

Riehm

Fehlings

. The relationship between MRI signal intensity changes, clinical presentation, and surgical outcome in degenerative cervical myelopathy: analysis of a global cohort. Spine (Phila Pa 1976). 2017;42(24):1851-1858.

Kalsi-Ryan

Karadimas

Fehlings

. Cervical spondylotic myelopathy: the clinical phenomenon and the current pathobiology of an increasingly prevalent and devastating disorder. Neuroscientist. 2013;19(4):409-421.

Tetreault

Kalsi-Ryan

Benjamin

, et al. Degenerative cervical myelopathy: a practical approach to diagnosis. Glob Spine J. 2022;12(8):1881-1893.

Milligan

Ryan

Fehlings

Bauman

. Degenerative cervical myelopathy: diagnosis and management in primary care. Can Fam Physician. 2019;65(9):619-624.

Kalsi-Ryan

Rienmueller

Riehm

, et al. Quantitative assessment of gait characteristics in degenerative cervical myelopathy: a prospective clinical study. J Clin Med. 2020;9(3):752.

Badhiwala

Wilson

. The natural history of degenerative cervical myelopathy. Neurosurg Clin N Am. 2018;29(1):21-32.

Karadimas

Erwin

Ely

Dettori

Fehlings

. Pathophysiology and natural history of cervical spondylotic myelopathy. Spine (Phila Pa 1976). 2013;38(22 Suppl 1):S21-S36.

Behrbalk

Salame

Regev

Keynan

Boszczyk

Lidar

. Delayed diagnosis of cervical spondylotic myelopathy by primary care physicians. Neurosurg Focus. 2013;35(1):E1.

Grodzinski

Stubbs

Davies

. Most degenerative cervical myelopathy remains undiagnosed, particularly amongst the elderly: modelling the prevalence of degenerative cervical myelopathy in the United Kingdom. J Neurol. 2023;270(1):311-319.

10.

Cheung

Arvinte

Wong

Luk

Cheung

. Neurological recovery after surgical decompression in patients with cervical spondylotic myelopathy - a prospective study. Int Orthop. 2008;32(2):273-278.

11.

Fehlings

Wilson

Kopjar

, et al. Efficacy and safety of surgical decompression in patients with cervical spondylotic myelopathy: results of the AOSpine North America prospective multi-center study. J Bone Joint Surg Am. 2013;95(18):1651-1658.

12.

Toop

Gifford

McGahan

, et al. Influence of clinical and radiological parameters on the likelihood of neurological improvement after surgery for degenerative cervical myelopathy. J Neurosurg Spine. 2023;38(1):14-23.

13.

Strimbu

Tavel

. What are biomarkers? Curr Opin HIV AIDS. 2010;5(6):463-466.

14.

Wang

Tong

Zhang

, et al. FDA-Approved and emerging next generation predictive biomarkers for immune checkpoint inhibitors in cancer patients. Front Oncol. 2021;11:683419.

15.

Messing

Brenner

. GFAP at 50. ASN Neuro. 2020;12:1759091420949680.

16.

Zhou

Tao

Qiu

, et al. Tumor biomarkers for diagnosis, prognosis and targeted therapy. Signal Transduct Target Ther. 2024;9(1):132.

17.

Elsarraj

Sarani

Enders

, et al. Clinical Utility of Recently FDA Approved IntelliSep test (Sepsis biomarker) for early Diagnosis of Sepsis: comparison with other Biomarkers and von Willebrand Factor/ADAMTS13 Ration. Am J Clin Pathol. 2024;162(Supplement_1):S158-S.

18.

Sun

Zhao

. Interleukin-6 serum levels are elevated in individuals with degenerative cervical myelopathy and are correlated with symptom severity. Med Sci Monit. 2018;24:7405-7413.

19.

Vedantam

Rahman

Salam

, et al. Serum protein biomarkers for degenerative cervical myelopathy: a prospective study. J Neurosurg Spine. 2025;1(aop):1-9.

20.

Khan

Mohammadi

Haynes

, et al. Evaluating tissue injury in cervical spondylotic myelopathy with spinal cord MRI: a systematic review. Eur Spine J. 2024;33(1):133-154.

21.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Bmj. 2021;372:n71.

22.

Whiting

Rutjes

Westwood

, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529-536.

23.

DerSimonian

Laird

. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188.

24.

von Hippel

. The heterogeneity statistic I2 can be biased in small meta-analyses. BMC Med Res Methodol. 2015;15(1):35.

25.

Team

RCR

. A Language and Environment for Statistical Computing. 4.4.0 ed. R Foundation for Statistical Computing; 2024.

26.

Tsitsopoulos

Holmström

Blennow

Zetterberg

Marklund

. Cerebrospinal fluid biomarkers of glial and axonal injury in cervical spondylotic myelopathy. J Neurosurg Spine. 2021;34(4):632-641.

27.

Tsitsopoulos

Mondello

Holmström

Marklund

. Cerebrospinal fluid biomarkers of white matter injury and astrogliosis are associated with the severity and surgical outcome of degenerative cervical spondylotic myelopathy. Spine J. 2022;22(11):1848-1856.

28.

Tsukahara

Hosokawa

Nishioka

, et al. Neuron-specific enolase level is a useful biomarker for distinguishing amyotrophic lateral sclerosis from cervical spondylotic myelopathy. Sci Rep. 2021;11(1):22827.

29.

Nagashima

Morio

Yamane

Nanjo

Teshima

. Tumor necrosis factor-alpha, interleukin-1beta, and interleukin-6 in the cerebrospinal fluid of patients with cervical myelopathy and lumbar radiculopathy. Eur Spine J. 2009;18(12):1946-1950.

30.

Takahashi

Aoki

Nakajima

, et al. Axonal damage is remarkable in patients with acutely worsening symptoms of compression myelopathy: biomarkers in cerebrospinal fluid samples. Eur Spine J. 2018;27(8):1824-1830.

31.

Tsuji

Matsuyama

Natsume

, et al. Analysis of chondrex (YKL-40, HC gp-39) in the cerebrospinal fluid of patients with spine disease. Spine (Phila Pa 1976). 2002;27(7):732-735.

32.

Ito

Matsuyama

Yukawa

Kato

Ishiguro

. Analysis of interleukin-8, interleukin-10, and tumor necrosis factor-alpha in the cerebrospinal fluid of patients with cervical spondylotic myelopathy. J Spinal Disord Tech. 2008;21(2):145-147.

33.

Imasato

Nagata

Hashimoto

Komori

Inoue

. Objective evaluation of pain in various spinal diseases: neuropeptide immunoreactivity in the cerebrospinal fluid. Spinal Cord. 1997;35(11):757-762.

34.

Takahashi

Aoki

Nakajima

, et al. Phosphorylated neurofilament subunit NF-H becomes elevated in the cerebrospinal fluid of patients with acutely worsening symptoms of compression myelopathy. J Clin Neurosci. 2014;21(12):2175-2178.

35.

Akshota

Friedlander

Hershkovich

, et al. Neural cell adhesion molecule (NCAM) a serum biomarker indicative for the severity of cervical spondylotic myelopathy. Clin Spine Surg. 2020;33(4):E178-E184.

36.

Kato

Chikuda

Ohya

, et al. Phosphorylated neurofilament subunit levels in the serum of cervical compressive myelopathy patients. J Clin Neurosci. 2015;22(10):1638-1642.

37.

Peng

Zhan

Liu

Zong

. Plasma VDBP, 25(OH)D, and GSH levels predict surgical outcome in patients with cervical spondylotic myelopathy. Kaohsiung J Med Sci. 2019;35(2):102-110.

38.

Divi

Markova

D'Antonio

, et al. Circulating microRNAs may be predictive of degenerative cervical myelopathy. Spine (Phila Pa 1976). 2024;49(20):1393-1400.

39.

Schmidt

Jütten

Bertram

, et al. Blood spinal cord barrier disruption recovers in patients with degenerative cervical myelopathy after surgical decompression: a prospective cohort study. Sci Rep. 2023;13(1):7389.

40.

Blume

Geiger

Brandenburg

, et al. Patients with degenerative cervical myelopathy have signs of blood spinal cord barrier disruption, and its magnitude correlates with myelopathy severity: a prospective comparative cohort study. Eur Spine J. 2020;29(5):986-993.

41.

Soufi

Nouri

Martin

. Degenerative cervical myelopathy and spinal cord injury: introduction to the special issue. J Clin Med. 2022;11(15):4253.

42.

Shakil

Santaguida

Wilson

Farhadi

Levi

Wilcox

. Pathophysiology and surgical decision-making in central cord syndrome and degenerative cervical myelopathy: correcting the somatotopic fallacy. Front Neurol. 2023;14:1276399.

43.

Yokobori

Zhang

Moghieb

, et al. Acute diagnostic biomarkers for spinal cord injury: review of the literature and preliminary research report. World Neurosurg. 2015;83(5):867-878.

44.

Bang

Kim

Seo

Cho

Sung

Kim

. Curcumin increase the expression of neural stem/progenitor cells and improves functional recovery after spinal cord injury. J Korean Neurosurg Soc. 2018;61(1):10-18.

45.

Kwon

Stammers

Belanger

, et al. Cerebrospinal fluid inflammatory cytokines and biomarkers of injury severity in acute human spinal cord injury. J Neurotrauma. 2010;27(4):669-682.

46.

Rodrigues

Moura-Neto

Tcls

. Biomarkers in spinal cord injury: from prognosis to treatment. Mol Neurobiol. 2018;55(8):6436-6448.

47.

Leite Pereira

Grad

Gonçalves

. Biomarkers for intervertebral disc and associated back pain: from diagnosis to disease prognosis and personalized treatment. JOR Spine. 2023;6(4):e1280.

48.

Weber

Satoh

Alipui

, et al. Exploratory study for identifying systemic biomarkers that correlate with pain response in patients with intervertebral disc disorders. Immunol Res. 2015;63(1-3):170-180.

49.

Brisby

Olmarker

Rosengren

Cederlund

Rydevik

. Markers of nerve tissue injury in the cerebrospinal fluid in patients with lumbar disc herniation and sciatica. Spine (Phila Pa 1976). 1999;24(8):742-746.

50.

Kim

Yong

Shea

GKH

. Blood-spinal cord barrier disruption in degenerative cervical myelopathy. Fluids Barriers CNS. 2023;20(1):68.

51.

Vargas Castillo

Das

Diwan

. Degenerative cervical myelopathy: insights into its pathobiology and molecular mechanisms. J Clin Med. 2021;10(6):1214.

52.

Jin

Wang

, et al. Blood-spinal cord barrier in spinal cord injury: a review. J Neurotrauma. 2021;38(9):1203-1224.

53.

Dalkilic

Fallah

Noonan

, et al. Predicting injury severity and neurological recovery after acute cervical spinal cord injury: a comparison of cerebrospinal fluid and magnetic resonance imaging biomarkers. J Neurotrauma. 2018;35(3):435-445.

54.

Ahadi

Khodagholi

Daneshi

Vafaei

Mafi

Jorjani

. Diagnostic value of serum levels of GFAP, pNF-H, and NSE compared with clinical findings in severity assessment of human traumatic spinal cord injury. Spine (Phila Pa 1976). 2015;40(14):E823-E830.

55.

Sterner

. Immune response following traumatic spinal cord injury: pathophysiology and therapies. Front Immunol. 2022;13:1084101.

56.

Hellenbrand

Quinn

Piper

Morehouse

Fixel

Hanna

. Inflammation after spinal cord injury: a review of the critical timeline of signaling cues and cellular infiltration. J Neuroinflammation. 2021;18(1):284.

57.

Cox

Varma

Barry

Vertegel

Banik

. Nanoparticle estrogen in rat spinal cord injury elicits rapid anti-inflammatory effects in plasma, cerebrospinal fluid, and tissue. J Neurotrauma. 2015;32(18):1413-1421.

58.

Laliberte

Karadimas

Vidal

Satkunendrarajah

Fehlings

. Mir21 modulates inflammation and sensorimotor deficits in cervical myelopathy: data from humans and animal models. Brain Commun. 2021;3(1):fcaa234.

59.

Zheng

Chen

, et al. Peripheral blood RNA biomarkers can predict lesion severity in degenerative cervical myelopathy. Neural Regen Res. 2025;20(6):1764-1775.

60.

Desimone

Hong

Brockie

Laliberte

Fehlings

. The influence of ApoE4 on the clinical outcomes and pathophysiology of degenerative cervical myelopathy. JCI Insight. 2021;6(15):e149227.

61.

Chintapalli

Mowforth

Davies

. APOE4 and prevalence of degenerative cervical myelopathy: a genetic case-control study in the UK Biobank cohort. J Clin Neurosci. 2025;136:111247.

62.

Wang

Yang

Zhu

, et al. An update on diagnostic and prognostic biomarkers for traumatic brain injury. Expert Rev Mol Diagn. 2018;18(2):165-180.

63.

Stukas

Cooper

Gill

, et al. Association of CSF and serum neurofilament light and glial fibrillary acidic protein, injury severity, and outcome in spinal cord injury. Neurology. 2023;100(12):e1221-e33.

64.

Pouw

Hosman

AJF

van Middendorp

Verbeek

Vos

van de Meent

. Biomarkers in spinal cord injury. Spinal Cord. 2009;47(7):519-525.

65.

Leister

Haider

Mattiassich

, et al. Biomarkers in traumatic spinal cord injury-technical and clinical considerations: a systematic review. Neurorehabil Neural Repair. 2020;34(2):95-110.

66.

Wei

Huang

Chen

, et al. Biomarkers for predicting the severity of spinal cord injury by proteomic analysis. Front Mol Neurosci. 2023;16:1153230.

67.

Higgins

JPTTJ

Chandler

Cumpston

Page

Welch

, eds. Cochrane Handbook for Systematic Reviews of Interventions.version 6.5. 6.5 ed; 2024.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB