Sage Journals: Discover world-class research

Abstract

Study Design

Systematic Review.

Objectives

To systematically review the current literature and perform a meta-analysis on patients with cervical or lumbar spine Modic changes to determine if their baseline axial back pain and disability are comparable to patients without Modic changes.

Methods

A systematic review of the PubMed database was conducted in accordance with PRISMA guidelines. A meta-analysis was performed to compare the mean differences in back pain, leg pain, and disability based on the presence of cervical or lumbar spine Modic changes. A subgroup analysis of the different types of Modic changes was conducted to determine if Modic type affected back pain or disability.

Results

– After review of 259 articles, 17 studies were included for meta-analysis and ten studies were included for qualitative synthesis. In the lumbar spine, 10 high-quality studies analyzed Visual Analog Scale (VAS) back pain, 10 evaluated VAS leg pain, and 8 analyzed Oswestry Disability Index. VAS back pain (mean difference (MD), −.38; 95% CI, -.61 – .16) and Oswestry disability index (MD −2.52; 95% CI, −3.93 – −1.12) were significantly lower in patients without Modic changes. Modic change subtype was not associated with differences in patient-reported outcomes. Patients with cervical spine Modic changes did not experience more severe pain than those without MC.

Conclusions

Modic changes in the lumbar spine are not associated with clinically significant axial low back pain severity or patient disability. Similar to the lumbar spine, Modic changes in the cervical spine are not associated with symptom severity, but they are associated with pain duration.

Keywords

modic changes back pain neck pain end-plate signal changes patient reported outcome measures systematic review

Introduction

Nearly 28% of US adults report back pain lasting more than 1 day during the last 3 months.¹ Neck pain, although less common, has a global age-standardized prevalence of 3.6%.² Both contribute significantly to global disability, poor quality of life, and significant economic costs.^2-5 For most patients, the etiology of their spinal pain is undiagnosed. However, for those with a clear etiology of neck or back pain, treatment can provide significant improvement in quality of life.^6-9 Vertebral end plate signal changes, specifically Modic changes (MC), are one frequently suggested cause of back and neck pain.

Modic elaborated on previous work by de Roos et al who first described subchondral vertebral body endplate signal changes in 1987.¹⁰ The three classic MC subtypes were subsequently described by Modic as distinct magnetic resonance imaging (MRI) signaling phenotypes.^11,12 Type I Modic changes (MC-I) are an inflammatory subtype that appear hypointense on T1-weighted MRI and hyperintense on T2-weighted MRI. Type II Modic changes (MC-II) represent fatty replacement of the bone marrow and appear hyperintense on T1-weighted MRI and either hyperintense or isointense on T2-weighted imaging. Type III Modic (MC-III) are the rarest subtype, representing sclerosis of the vertebral endplate.¹³ MC-III appear hypointense on both T1- and T2-weighted MRI.

While the underlying cause of MCs are not fully elucidated, they are theorized to have one of the following etiologies: degenerative disc disease, infection, trauma and/or inflammation.¹⁴ Moreover, the three primary subtypes of Modic changes appear to represent different phases of disease process. MC-I are evoked by inflammatory changes wherein vertebral endplates are disrupted and edematous changes are apparent in the bone. This hypothesis is supported by preclinical research, which has found that modulation of the local inflammatory environment occurs through cytokine signaling in the subchondral endplate.^15,16 MC-I may progress to MC-II, MC-III, or resolve, although they do not always progress in a sequential order.¹⁷ MC-II represents a more chronic disease phase highlighted by fatty replacement of subchondral bone, while MC-III are typically the final phase of MC disease. Although MC-II rarely progress to MC-III, MC-III are identified by continuous progression of the disease until subchondral bone marrow is absent and it is instead replaced by endplate sclerosis.^11,17-19

While the exact etiology of MCs remains unclear, several studies have linked these lesions to both back and neck pain.^20-24 These most frequently occur in the lumbar region. Studies evaluating cervical changes remain limited but have increased in recent years.²⁵ The literature has experienced significant growth in identifying the prevalence of and symptoms related to MCs. Therefore, the objective of this systematic review is to provide an overview of the literature evaluating the association of both lumbar and cervical MCs with patients’ baseline patient-reported outcome measures (PROMs) and to provide a meta-analysis evaluating how MCs affect PROMs.

Methods

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement. This study was deemed Institutional Review Board exempt since no identifiable health information was collected and only data from published studies were incorporated. We conducted a query of the PubMed database from inception until May 1, 2022, using combinations of the words “Modic changes,” “endplate signal changes,” and “outcomes.” We also conducted a reference review of included studies to identify additional studies that met inclusion criteria.

Study Eligibility

Studies were included if they described clinical pain or disability prior to any intervention in a population of patients with identifiable MCs or endplate defects in the lumbar or cervical spine on MRI. Studies were excluded if (1) a full text of the article could not be obtained, (2) the article was written in a language other than English, or (3) they were review papers. Two authors independently screened the abstracts and titles of identified articles. Disagreements on any articles at this stage were automatically included for full text review. At the full-text stage, articles were screened for inclusion based on the predetermined inclusion criteria. Discrepancies were reconciled by a more senior author.

Data Collection Quality Assessment

Authors extracted data from the included studies with a self-designed table. Contents of data selection included type and prevalence of Modic signal changes, baseline patient reported outcome measures, quality of life measures, and basic patient population statistics. A meta-analysis was conducted on any PROM that was compared by a minimum of 6 studies. This criterion was set to ensure adequate analysis of data heterogeneity. Meta-analyses were performed to compare the available PROMs among patients with and without Modic changes. Additionally, a subgroup analysis was done evaluating PROMs based on MC subtype (no-MC, MC-I, or MC-II). Due to the low number of studies evaluating cervical MCs, a meta-analysis of was not conducted.

Weighted means were evaluated for all continuous variables using a random effects model. Results are reported as weighted mean difference and 95% confidence interval. To assess the heterogeneity of studies (I²), comparisons were displayed in a forest plot. To examine publication bias, the existence of asymmetry between studies was verified and displayed in a funnel plot. Studies with data outside the funnel plot were considered biased and were removed from the meta-analysis. Studies were also assessed for bias using the validated Newcastle-Ottawa Score with a high-quality score defined as a greater than or equal to 6.²⁶ All statistical analyses were done using R Studio (Version 4.1.2, Vienna, Austria).

Results

Literature Search Results

The initial literature review identified 259 articles meeting our search criteria. After title and abstract screen, 56 full text articles were reviewed. Four studies were included from citation review.^22,24,27,28 Twenty-eight studies evaluated patients with MCs in the lumbar spine and ten studies evaluated patients with MCs in the cervical spine. The PRISMA search strategy is provided in (Figure 1). However, of the 28 lumbar spine studies meeting potential inclusion criteria, only 17 were included in our meta-analysis (since meta-analysis inclusion required a minimum of six studies to evaluate the same outcome measure). Thus, a total of 3889 patients were ultimately included (Table 1). No PROM in the cervical spine was evaluated by a minimum of 6 studies. Therefore, a meta-analysis was not performed, but an overview of the studies evaluating cervical spine MCs are described in Table 2. The majority (25 of 27) of studies included were considered high quality. Individual risk of bias scores are presented in Table 1 and Table 2.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram indicating study inclusion criteria and the reason for study exclusion.

Table 1.

Characteristics of Studies Evaluating Modic Changes in the Lumbar Spine.

Study ID	Authors	Patient Population	Modic subtypes	Patients	Age (SD or range)	Gender (F)	Any MC(%)	MC-I (%)	MC-II (%)	Risk of Bias
1	Kumarasamy et al (2021)	Patients with chronic LBP due to single level lumbar disc herniation	I, II	309	42.5 (12.6)	107	86 (27.8%)	6 (1.9%)	68 (22.0%)	7
2	Jiao et al (2021)	Patients with LDH, spinal stenosis, or spondylolisthesis (LS)	I, II	89	56.7 (8.9)	49	51	20 (22.5%)	31 (34.8%)	6
3	Annen et al (2018)	Patients with LBP without LDH	I, II	112	45.3 (16.2)	47	57 (50.9%)	28 (25%)	29 (25.9%)	7
4	Annen et al (2016)	Patients with LBP with LDH	I, II	72	41.9 (11.4)	17	40 (55.6%)	16 (22.2%)	24 (33.3%)	6
5	Schistad et al (2014)	Patients with radicular pain due to lumbar disc herniation	I, II, III	243	41.3 (10.4)	114	184 (74.7%)	30 (12.3%)	147 (60.5%)	8
6	MacLean et al (2021)	Patients who underwent discectomy or transforaminal lumbar interbody fusion for single level LDH	I, II, III	179	53 (13)	101	110 (61.5%)	28 (15.6%)	63 (35.2%)	7
7	Udby et al (2020)	Patients undergoing discectomy for bilateral or unilateral radiculopathy	I, II	620	50.5 (19-75)	310	290 (46.8%)	73 (11.8%)	217 (35%)	7
8	Chen et al (2019)	Patients with 3-12 months of LBP without radicular pain	I, II	129	28.3 (21-67)	73	79 (61.2%)	31 (24%)	48 (37.2%)	6
9	Sørlie et al (2012)	Patients with one-level lumbar disc herniation	I, II	178	41.2 (12.1)	66	104 (58.4%)	36 (20.2%)	68 (38.2%)	8
10	Xu et al (2019)	Patients with one-level intracanal disc herniation and unilateral radicular pain non-responsive to 6 weeks of conservative treatment	I, II	276	39.97 (12.5)	104	94 (34.1%)	44 (15.9%)	50 (18.2%)	6
11	Bianchi et al (2015)	Patients with LBP	I, II	226	61.6 (13.3)	138	141 (62.4%)	83 (36.7%)	58 (25.7%)	6
12	Ohtori_a et al (2010)	Patients with one-level lumbar disc herniation	I	45	35.5 (5.0)	19	23 (51.1%)	23 (51.1%)	N/A	6
13	Blondel et al (2011)	Patients with chronic LBP due to single-level lumbar or lumbosacral disc degeneration	I, II	221	41.1 (18-60)	101	114 (51.6%)	65 (29.4%)	49 (22.2%)	8
14	Maata et al (2014)	Patients who were referred for spine surgery for LBP	I, II, III	181	42.8 (11.3)	84	84 (46.4%)	31 (17.1%)	53 (29.2%)	6
15	Ohtori_b et al (2010)	Patients with LBP and leg pain due to lumbar spinal canal stenosis	I, II	33	65.4 (4.7)	16	33 (100%)	21 (63.6%)	12 (36.4%)	6
16	Shahmohammadi et al (2019)	Patients with LBP and DDD	I, II	162	54.8 (9.2)	74	162 (100%)	75 (46.3%)	87 (53.7%)	5
17	Mera et al (2021)	Population study of all patients without history of inflammatory conditions or prior spine fusion	I,II,III	814	63.6 (13.1)	568	516 (63.5%)	151 (18.6%)	343 (42.1%)	7

Abbrevitions: LBP = low back pain, LDH = lumbar disc herniation, TLIF = transforaminal lumbar interbody fusion, DDD = degenerative disc disease.

Table 2.

Patient Characteristics of Studies Evaluating Modic Changes in the Cervical Spine.

Authors	Patient Population	Modic subtypes	Patients	Age (SD)	Gender	Any MC (%)	MC-I (%)	MC-II (%)
Yang et al (2019)	Patients with cervical radiculopathy due to single-level disc herniation	I,II	223	45.2 (7.3)	131	41 (18.4%)	10 (4.5%)	29 (13%)
Baker et al (2020)	Patients with symptomatic degenerative pathology	I,II,III	861	53.1 (11.2)	409	356 (41.3%)	70 (8.1%)	218 (25.3%)
Qiao et al (2018)	Patients with cervical spondylotic myelopathy	II	539	46 (5.2)	265	70 (13.0%)	20 (3.7%)	41 (7.6%)
Li et al (2015)	Patients with radiculopathy or myelopathy	II	248	46.4 (7.2)	134	35 (14.1%)	N/A	35 (14.1%)
Matsumoto et al (2012)	Healthy patients without known cervical spine disease	I,II,III	223	50.5	36	31 (13.9%)	9 (4%)	18 (18.1%)
Sheng-Yun et al (2014)	Patients with low-back or neck pain with or without radiculopathy	I,II,III	1520	—	—	132 (8.7%)	N/S	N/S
Kressig et al (2016)	Patients with cervical radiculopathy due to single-level disc herniation	I,II	44	44.7 (7.9)	—	13 (29.5%)	N/S	N/S
Li et al (2015)	Patients with chronic axial symptoms resulting from single-level cervical disc degeneration	I,II	106	55.8 (6.5)	49	62 (58.5%)	23 (21.7%)	39 (36.8%)
Kong et al (2017)	Patients with cervical disc degenerative disease with radiculopathy or myelopathy	I,II	381	50.7 (12.9)	188	47 (12.3%)	21 (5.5%)	26 (6.8%)
An et al	Patients with cervical kyphosis	I,II	286	54.2 (12.2)	146	102 (35.7%	38 (13.3%)	75 (26.2%)

Visual Analogue Score Back Pain

Pooled reports of back pain intensity among patients with and without MC were reported in 10 studies.^24,29-39 Low back pain intensity scores were standardized on a 0-10 scale. An initial funnel plot revealed possible publication bias in three studies^15,29,30 After these studies were excluded, a forest plot revealed low data heterogeneity (I² = 21%, P = .25) (Figure 2A). Publication bias was also deemed unlikely in the remaining studies (Figure 2B). When evaluating a random-effects model of the pooled studies, there was no significant difference in VAS back scores in patients with Modic changes vs those without MCs (weighted MD, −.38; 95% CI, −.61 – .16).

Figure 2.

A. Forest Plot of included studies evaluating VAS Back Pain, which are stratified based on the presence of Modic Changes. The Forest plot indicates the relative lack of back pain heterogeneity between patients with and without Modic changes. Although back pain is significantly lower in patients without MCs, the values do not meet clinical significance. 2B. Funnel plot indicates there is no obvious publication bias in any of the included studies.

On MC subtype analysis, 12 studies specifically reported on back pain intensity in non-MC patients,^{15,24,29,30,32-39} 14 studies were performed in patients with MC-I,^{15,24,29,30,32-41} and 12 were examined patients with MC-II.^{15,24,29,30,32-34,36,37,39-41} Subgroup analysis showed no significant differences in a weighted random-effects model for patients without MC (5.35; 95% CI: 4.43-6.27) vs patients with MC-I (5.73; 95% CI: 4.91-6.54) or MC-II (5.53; 95% CI: 4.45-6.54) (P = .83) (Supplemental Figure 2). However, there was high data heterogeneity among included studies (I² = 99%).

Visual Analogue Score Leg Pain

Ten studies reported on leg pain scores in their analysis of patients with MCs compared to those without MCs.^{29-33,35,36,38,39,42} Random effects analysis of all studies demonstrated no significant association between baseline VAS leg pain score and Modic changes (MD .01, 95% CI: −.18 - .20; P = .68). Among these studies, low heterogeneity was observed (I² = 0%). (Figure 3A). Among these studies, sensitivity analysis revealed no studies exhibiting publication bias (Figure 3B).

Figure 3.

A. Forest Plot of included studies evaluating VAS Leg Pain based on the presence of Modic Changes. The Forest plot indicates the relative lack of leg pain heterogeneity between patients with and without Modic changes. 3B. Funnel plot indicates there is no obvious publication bias in any of the included studies.

After stratification by MC subtype, nine studies were included that reported on leg pain in patients without Modic changes,^{29,30,32,33,35,36,38,39,42} ten examined patients with MC-I,^{29,30,32,33,35,36,38,39,41,42} and eight reported on leg pain in patients with MC-II.^{29,30,32,33,36,39,41,42} Forest plots revealed a high degree of data heterogeneity (I² = 61%, P < .01). MC subtypes were not significantly associated with difference in VAS leg scores (P = .38) (Supplemental Figure 3).

Oswestry Disability Index

Twelve studies reported on Oswestry Disability Index (ODI) in patients with and without Modic changes.^{29,31-33,35,36,38,39,42,43} Two studies were at high risk of publication bias and were excluded^38,42 Forest plot analysis of the eight included studies demonstrated low data heterogeneity when evaluating ODI scores between patients with Modic changes compared to those without Modic changes (I² = 0%) (Figure 4A). A random effects comparison between patients with and without MCs demonstrated a significantly greater disability score in patients with MC (MD −2.52; 95% CI, −3.93 - −1.12) (Figure 4A). There was a low risk of publication bias among these studies (Figure 4B). Subgroup analysis was performed comparing ODI between patients with no-MC, MC-I, and MC-II, respectively. Weighted random effects modeling demonstrated no significant difference in ODI between groups (35.42 in no MC vs 36.80 in MC-I vs 34.82 in MC-II, P = .91). However, there was significant data heterogeneity between groups (I² = 98%).

Figure 4.

A. Forest Plot of included studies evaluating Oswestry Disability Index (ODI) based on the presence of Modic Changes. The Forest plot indicates the relative lack of heterogeneity in disability between studies with Modic changes having a statitistically significant greater amount of disability. B. Funnel plot indicates there is no obvious publication bias in any of the included studies.

Cervical Modic Changes

Ten studies evaluating MC were identified, which evaluated validated PROMs or patient symptoms in patients with MCs.^{22,25,27,28,44-49} The key findings and baseline outcome measures evaluated are summarized in Table 3. Three studies broadly evaluated any patient with degenerative disc disease, two studies specifically evaluated patients with disc herniations, one study examined patients with cervical myelopathy, one evaluated patients with cervical kyphosis, and two studies included patients with axial neck pain and/or radiculopathy.

Table 3.

Key Findings of Studies Evaluating Cervical Modic Changes.

Authors	Clinical Outcome measures	Key findings
Yang et al (2019)	NDI, MCS-12, PCS-12, Neck VAS, Arm VAS	MCs were not associated with worse baseline NDI, PCS-12/MCS-12, or VAS Arm/Neck
Baker et al (2020)	Pre-operative symptoms (Neck pain, radiculopathy, arm pain, sensory deficits, weakness, myelopathy) PROMs (VAS Neck, Vas arm, SF12, VR12)	Patients with MCs had longer duration of symptoms. Simply, having MC, regardless of stratification by type, did not ultimately yield any baseline differences in PROMs
Qiao et al (2018)	Duration of symptoms	Patients with MCs were more likely to have symptom durations greater than 18 months
Li et al (2015)	JOA, NDI, VAS neck	No significant differences in baseline PROMs were identified between patients based on the presence of MCs
Matsumoto et al (2012)	Questionnaires of neck pain, shoulder pain/stiffness, arm pain/numbness	Only MC-II were associated with arm pain/numbness No MC type was associated with neck pain or shoulder pain/stiffness
Sheng-Yun et al (2014)	Neck pain, age, disc degeneration, spinal level, and the presence or absence of kyphosis	MCs were associated with neck pain, kyphosis, greater disc degeneration, and older age
Kressig et al (2016)	Neck and arm pain scores, Neck Disability Index	Patients with MCs had greater baseline NDI, but similar neck and arm pain scores
Li et al (2015)	Japanese Orthopaedic Association (JOA) score, VAS neck	No significant difference in baseline PROMs between patients with or without MCs
Kong et al (2017)	NRS Neck pain, persistent neck pain (3+ on scale for 12+ months), severe pain (5+)	MCs were associated with persistent neck pain but its severity was not different. Patients with MC 1 have worse neck pain than MC 2
An et al (2017)	Presence of axial neck pain symptoms	Patients with MCs were more likely to have axial neck pain.

Abbrevitions: NDI, neck disability index; MCS-12, mental component of the short form 12 health survey; PCS-12, physical component of the short form 12 health survey; VAS, visual analog scale; NRS, numeric rating scale; MC, Modic change; VR12, Veterans Rand 12 item survey.

Matsumoto et al⁴⁴ identified a population of people without neck pain. Of the 223 study participants, 31 (13.9%) had Modic changes on MRI. Modic changes were not associated with neck or shoulder pain (given that the population was derived from asymptomatic patients), but MC-II lesions were associated with recent symptoms of arm pain/numbness.

Baker et al⁴⁵ identified 861 patients with symptomatic degenerative disc disease and found that 356 (41.3%) patients demonstrated some form of MCs. They found that MCs had no clear relationship with patient reported outcomes, although patients with MCs did have a longer duration of symptoms. Similarly, Kong et al⁴⁹ evaluated patients with degenerative disease causing radiculopathy and/or myelopathy and found no association between MCs and neck pain severity. However, patients with MCs were more likely to have persistent neck pain, defined as a score of at least 3/10 on neck pain score for at least 12 months. Li et al⁴⁸ similarly found no association between MCs and baseline pain or Japanese Orthopaedic Association score.

Of the two studies evaluating patients with radiculopathy due to disc herniation, Yang et al²⁵ included 223 patients, of whom 41 (18.4%) displayed MCs. They found no association between Modic changes and patient-reported disability, function or pain. Kressig et al⁴⁶ examined a smaller cohort of 44 patients, although they found that patients with MCs had worse baseline disability, they suggest they likely have similar pain scores compared to patients without MCs.

Qiao et al²⁸ evaluated 529 patients with cervical spondylotic myelopathy, 13% of whom displayed MCs. MCs were significantly associated with axial neck pain persisting greater than 18 months. However, they did not evaluate baseline differences in pain scores or PROMs between patients with and without MCs.

Sheng-Yun et al²² identified 1520 patients with neck pain, regardless of etiology, and compared them to 120 asymptomatic patients with cervical spine MRI. Nine percent of symptomatic patients had MCs compared to only 3% of asymptomatic patients. Patients with MCs were also more likely to have concomitant disc degeneration and cervical kyphosis. Li et al⁴⁷ evaluated any patients with radiculopathy or myelopathy requiring anterior cervical discectomy and fusion (ACDF). They only evaluated patients with MC-II lesions; however, they found no significant association between baseline PROMs and the presence of MCs.

In one study evaluating patients with cervical kyphosis, An et al.²⁷ identified that patients with Modic changes were 5.4 times more likely to experience axial neck pain than patients with kyphosis and no MCs. However, they did not evaluate specific patient reported outcome measures or VAS scores.

Discussion

Low back pain is the leading cause of disability worldwide.^3,50 Patients with identifiable sources of pain may benefit from targeted intervention, although unequivocally identifying spinal pain generators can be challenging since most asymptomatic patients have abnormalities on MRI including disc degeneration and central canal stenosis.^51,52 If MCs are identified as a frequent source of pain across multiple studies, targeted treatment of these abnormalities may be one avenue to reduce patients’ time off work, while also lowering disability claims.^24,53 This systematic review and meta-analysis evaluated the effects of Modic changes, a theorized etiology of axial pain, on the presence of patient reported disability and neck/low back pain. Our pooled analysis suggests that patients with MCs in the lumbar spine have significantly worse back pain and disability; however, these findings appear to be clinically insignificant based on previously established minimal clinically important differences (MCIDs). While the literature evaluating cervical spine MCs and their effect on baseline pain is limited, patients with cervical spine MCs tend to have longer symptom duration defined as persistent pain, but similar pain intensity and neck disability.

The most recent prior meta-analysis attempting to correlate MCs and low back pain found that MCs increased the likelihood of symptomatic back pain by a factor of 3.2-6.1.⁵⁴ However, that study only pooled patients undergoing provocative discography, which itself may be a significant pain generator and lead to significant long-term disc disease.⁵⁵ Further, while their question centered on identifying whether patients with MCs were more likely to have back pain, the question appears somewhat moot in a population of patients who required a lumbar spine MRI due to persistent lower back pain. Therefore, our study attempted to address a more prudent question, which is whether MCs may produce a significantly altered disease course defined by greater symptomatic back pain intensity or disability.

When aggregating all lumbar spine studies, regardless of MC subtype, Modic changes were associated with statistically worse back pain and disability, but the mean differences were small indicating they were clinically insignificant. This can easily be appreciated when evaluating well-established MCIDs for VAS back pain (1.2) and ODI (12.8), which serves as the threshold for the smallest change perceived to be important to patients.^56,57 Our meta-analysis found that the weighted mean differences for VAS back pain (.38) and ODI (2.52) were substantially lower than the required threshold needed to be considered clinically important to patients. Therefore, the current literature shows, with some level of confidence, that Modic changes are not a clinically significant source of back pain severity or disability. This finding is strengthened by the low data heterogeneity present (I² = 21%) indicating that this was a consistent finding across all studies evaluated. However, there is some evidence that MCs are linked to symptom chronicity as seen by studies suggesting that patients with MCs are more likely to have persistent symptoms.^28,45,49

Since recent evidence supports MC-I phenotypes may be greater pain generators than MC-II or MC-III subtypes, we analyzed whether any correlation existed between back pain severity or disability across MC subtypes.²⁴ While this analysis was fraught with significant data heterogeneity (I² = >97%, both), there was no clear evidence supporting different MC subtypes resulted in greater low back pain intensity or disability ([no MCs – VAS Back: 5.35; ODI: 35.4]; [MC-I – VAS Back: 5.73; ODI: 36.8] or [MC-II – VAS Back: 5.53; ODI: 34.8]).

When evaluating MCs’ effects on radicular leg pain, VAS Leg pain scores were not significantly different between groups with or without MCs. Since Modic changes represent inflammatory changes or endplate sclerosis confined to the vertebral body, nerve root compression and resultant radiculopathies are not an expected finding. Therefore, this finding may provide some additional validity to the data obtained from our meta-analysis. Although there was insufficient data to perform a meta-analysis evaluating a relationship between MCs and radiculopathy in the cervical spine, there were two studies that found no significant relationship.^25,46

Literature evaluating Modic changes in the cervical spine remains limited with only ten studies meeting inclusion criteria. None of the included studies identified differences in axial neck pain severity based on the presence or absence of Modic changes. However, five studies demonstrated an association with the presence of pain,^22,27 persistence of pain,⁴⁹ or symptom duration.^28,45 Clinically, this may indicate that patients with Modic changes present similarly to other patients with neck pain, but they may have experienced a prolonged disease course. Qiao et al²⁸ identified MC-II as a primary cause of longer symptom duration, which supports the theory that MC-II is indicative of long-standing disease course compared to MC-I, which is more likely to be identified in the acute to subacute setting.

Vertebral end plate signal changes (VESCs) are another potential source of back pain, but they have only recently been recognized as a distinct phenomenon from Modic changes.^58,59 VESCs include Schmorl’s nodes, insufficiency fractures, endplate erosions, and endplate calcifications, all of which display distinct phenotypes on imaging.⁶⁰ Historically, these lesions were thought to be a common finding in asymptomatic patients.⁵² However, several recent studies suggest that the combined Modic-endplate disruption phenotype results in significantly greater back and neck pain compared to Modic changes independently.^20,45,61 The literature on this phenotype remains extremely limited. However, the clinical impact of concomitant endplate defects and Modic changes may significantly impact patient presentation and serve as a prognostic factor for patient outcomes.⁶¹ Future research should evaluate this phenotype and its potential impact on axial back pain and disability with a particular emphasis on whether any potential increase in pain or disability reaches the MCID threshold.

Patients with greater pain intensity or disability may be more likely to require additional treatment including non-operative or operative management. Several studies and reviews have evaluated both operative and non-operative treatment for Modic changes.^62-65 A recent systematic review and meta-analysis of operative treatment for Modic changes demonstrated that surgical intervention for discal or vertebral pathologies successfully treat patient symptoms regardless of the presence of Modic changes.⁶³ Several non-surgical interventions have been evaluated in attempts to address discogenic back pain secondary to Modic changes including antibiotics, corticosteroids, basivertebral nerve ablation, physical therapy, and bisphosphonates.^33,34,64-68 However, all of these therapies have varying levels of efficacy without conclusive evidence pointing towards superiority of a specific treatment. Based on the results of the current meta-analysis, the similar baseline PROMs between patients with or without Modic changes may add to the complexities of treatment algorithms for multifactorial discogenic back pain. Moreover, the various operative and non-operative interventions may be dependent on concomitant spinal pathologies and specific patient symptoms rather than Modic changes.

Several limitations of this systematic review and meta-analysis exist. Significant heterogeneity in studies reporting Modic subtypes made it inappropriate to draw strong conclusions regarding the impact of specific Modic subtypes, particularly differences between MC-I vs MC-II. However, our analysis is strengthened by the low differences in pain or disability between patients with or without Modic changes after they were aggregated irrespective of Modic type, something that has not been previously performed. While this may be subject to study bias, efforts to account for bias by excluding studies outside of our funnel plot further validate these findings. It should be noted that each author group may have defined and classified MCs differently. This may have accounted for some study heterogeneity, but this could not be accounted for or controlled for in our study. Additionally, we were unable to conduct a meta-analysis of axial neck pain or disability in patients with Modic changes of the cervical spine due to the limited literature evaluating these patients, but all available evidence suggests that Modic changes do not significantly increase neck pain intensity.^22,28,44-49

Conclusion

Previously reported associations between Modic changes and axial back pain and patient disability do not meet clinical relevance in either the lumbar or cervical spine. Moreover, this review suggests that Modic changes do not serve as prognostic indicators of disease severity. Future studies evaluating the connections between VESCs and Modic changes are needed given emerging evidence linking the combined disease process with greater back pain severity. Despite the expanding literature on Modic changes and PROMs, additional studies should evaluate the impact of MC size and duration on symptom intensity and disability.

Supplemental Material

Supplemental material - Modic Changes of the Cervical and Lumbar Spine and Their Effect on Neck and Back Pain: A Systematic Review and Meta-Analysis

Supplemental material for Modic Changes of the Cervical and Lumbar Spine and Their Effect on Neck and Back Pain: A Systematic Review and Meta-Analysis by Mark J. Lambrechts, Tariq Z. Issa, Gregory R. Toci, Meghan Schilken, Jose A. Canseco, Alan S. Hilibrand, Gregory D. Schroeder, Alexander R. Vaccaro, and Christopher K. Kepler in Global Spine Journal.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Mark J. Lambrechts

Tariq Z. Issa

Gregory R. Toci

Jose A. Canseco

Alan S. Hilibrand

Supplemental Material

Supplemental material for this article is available online.

References

Blackwell

Lucas

Clarke

. Summary health statistics for US adults: national health interview survey, 2012. Vital Health Stat 10. 2014; 260:1-161.

Safiri

Kolahi

Hoy

, et al. Global, regional, and national burden of neck pain in the general population, 1990-2017: systematic analysis of the Global Burden of Disease Study 2017. The BMJ. 2020;368:m791. doi:10.1136/bmj.m791.

Hartvigsen

Hancock

Kongsted

, et al. What low back pain is and why we need to pay attention. The Lancet. 2018;391(10137):2356-2367. doi:10.1016/S0140-6736(18)30480-X.

Kazeminasab

Nejadghaderi

Amiri

, et al. Neck pain: global epidemiology, trends and risk factors. BMC Musculoskelet Disord. 2022;23(1):26. doi:10.1186/s12891-021-04957-4.

Dieleman

Cao

Chapin

, et al. US health care spending by payer and health condition, 1996-2016. JAMA. 2020;323(9):863-884. doi:10.1001/jama.2020.0734.

Stull

Goyal

DKC

Mangan

, et al. The outcomes of patients with neck pain following ACDF: A comparison of patients with radiculopathy, myelopathy, or mixed symptomatology. Spine. 2020;45(21):1485-1490. doi:10.1097/BRS.0000000000003613.

Sasso

Smucker

Sasso

. Long-term clinical outcomes of cervical disc arthroplasty: a prospective, randomized, controlled trial. Spine. 2017;42(4):209-216. doi:10.1097/BRS.0000000000001746.

Kunder

SL de

Kuijk

SMJ van

Rijkers

, et al. Transforaminal lumbar interbody fusion (TLIF) versus posterior lumbar interbody fusion (PLIF) in lumbar spondylolisthesis: a systematic review and meta-analysis. Spine J. 2017;17(11):1712-1721. doi:10.1016/j.spinee.2017.06.018.

Bydon

Alvi

Goyal

. Degenerative lumbar spondylolisthesis: definition, natural history, conservative management, and surgical treatment. Neurosurg Clin N Am. 2019;30(3):299-304. doi:10.1016/j.nec.2019.02.003.

10.

de Roos

Kressel

Spritzer

Dalinka

. MR imaging of marrow changes adjacent to end plates in degenerative lumbar disk disease. AJR Am J Roentgenol. 1987;149(3):531-534. doi:10.2214/ajr.149.3.531.

11.

Modic

Masaryk

Ross

Carter

. Imaging of degenerative disk disease. Radiology. 1988;168(1):177-186.

12.

Modic

Steinberg

Ross

Masaryk

Carter

. Degenerative disk disease: assessment of changes in vertebral body marrow with MR imaging. Radiology. 1988;166(1 Pt 1):193-199. doi:10.1148/radiology.166.1.3336678.

13.

Mann

Peterson

Hodler

Pfirrmann

CWA

. The evolution of degenerative marrow (Modic) changes in the cervical spine in neck pain patients. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc. 2014;23(3):584-589. doi:10.1007/s00586-013-2882-6.

14.

Din

Cheng

Yang

. Diagnostic role of magnetic resonance imaging in low back pain caused by vertebral endplate degeneration. J Magn Reson Imaging. 2022;55(3):755-771. doi:10.1002/jmri.27858.

15.

Chen

Yang

Zhang

Wang

Zou

. Analyzing the influence of modic changes on patients with lower back pain undergoing conservative treatment. Pain Res Manag. 2019;2019:8185316. doi:10.1155/2019/8185316.

16.

Kerttula

Luoma

Vehmas

Grönblad

Kääpä

. Modic type I change may predict rapid progressive, deforming disc degeneration: a prospective 1-year follow-up study. Eur Spine J. 2012;21(6):1135-1142. doi:10.1007/s00586-012-2147-9.

17.

Hutton

Bayer

Powell

. Modic vertebral body changes: the natural history as assessed by consecutive magnetic resonance imaging. Spine. 2011;36(26):2304-2307. doi:10.1097/BRS.0b013e31821604b6.

18.

Teichtahl

Finnin

Wang

, et al. The natural history of modic changes in a community-based cohort. Joint Bone Spine. 2017;84(2):197-202. doi:10.1016/j.jbspin.2016.03.011.

19.

Tamai

Teraguchi

Hashizume

, et al. A prospective, 3-year longitudinal study of modic changes of the lumbar spine in a population-based cohort: the wakayama spine study. Spine. 2022;47(6):490-497. doi:10.1097/BRS.0000000000004301.

20.

Baker

Sayari

Tao

, et al. Endplate abnormalities, Modic changes and their relationship to alignment parameters and surgical outcomes in the cervical spine. J Orthop Res Off Publ Orthop Res Soc. Published online April;10:2022. doi:10.1002/jor.25333.

21.

Jensen

Leboeuf-Yde

. Is the presence of Modic changes associated with the outcomes of different treatments? A systematic critical review. BMC Musculoskelet Disord. 2011;12:183. doi:10.1186/1471-2474-12-183.

22.

Sheng-yun

Letu

Jian

, et al. Comparison of Modic Changes in the lumbar and cervical spine, in 3167 patients with and without spinal pain. PLoS One. 2014;9(12):e114993. doi:10.1371/journal.pone.0114993.

23.

Sun

Yan

Shen

. Relation between neck pain and modic changes in cervical spondylotic myelopathy. Med Sci Monit Int Med J Exp Clin Res. 2020;26:e923908. doi:10.12659/MSM.923908.

24.

Mera

Teraguchi

Hashizume

, et al. Association between types of Modic changes in the lumbar region and low back pain in a large cohort: the Wakayama spine study. Eur Spine J. 2021;30(4):1011-1017. doi:10.1007/s00586-020-06618-x.

25.

Yang

Donk

Arts

Vleggeert-Lankamp

CLA

. Are modic vertebral end-plate signal changes associated with degeneration or clinical outcomes in the cervical spine? World Neurosurg. 2019;129:e881-e889. doi:10.1016/j.wneu.2019.06.067.

26.

Wells

Shea

O’Connell

, et al. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses. Oxford; 2000.

27.

Shen

. Characteristics of Modic changes in cervical kyphosis and their association with axial neck pain. J Pain Res. 2017;10:1657-1661. doi:10.2147/JPR.S138361.

28.

Qiao

Zhang

Tian

. Modic changes in the cervical endplate of patients suffering from cervical spondylotic myelopathy. J Orthop Surg. 2018;13:90. doi:10.1186/s13018-018-0805-2.

29.

Schistad

Espeland

Rygh

Røe

Gjerstad

. The association between Modic changes and pain during 1-year follow-up in patients with lumbar radicular pain. Skeletal Radiol. 2014;43(9):1271-1279. doi:10.1007/s00256-014-1928-0.

30.

Udby

Ohrt-Nissen

Bendix

, et al. Are modic changes associated with health-related quality of life after discectomy: a study on 620 patients with two-year follow-up. Spine. 2020;45(21):1491-1497. doi:10.1097/BRS.0000000000003618.

31.

Kumarasamy

Rajasekaran

Anand

K S SV

, et al. Lumbar disc herniation and preoperative modic changes: a prospective analysis of the clinical outcomes after microdiscectomy. Glob Spine J. 2022;12(5):940-951. doi:10.1177/2192568220976089.

32.

Jiao

Luo

Zhang

. Clinical and radiographic outcomes of hybrid graft in patients with Modic changes undergoing transforaminal lumbar interbody fusion. J Orthop Surg. 2021;16(1):486. doi:10.1186/s13018-021-02652-7.

33.

Annen

Peterson

Leemann

Schmid

Anklin

Humphreys

. Comparison of outcomes in MRI confirmed lumbar disc herniation patients with and without modic changes treated with high velocity, low amplitude spinal manipulation. J Manipulative Physiol Ther. 2016;39(3):200-209. doi:10.1016/j.jmpt.2016.02.012.

34.

Annen

Peterson

Humphreys

. Comparison of treatment outcomes in nonspecific low-back pain patients with and without modic changes who receive chiropractic treatment. J Manipulative Physiol Ther. 2018;41(7):561-570. doi:10.1016/j.jmpt.2018.01.008.

35.

Sørlie

Moholdt

Kvistad

, et al. Modic type I changes and recovery of back pain after lumbar microdiscectomy. Eur Spine J. 2012;21(11):2252-2258. doi:10.1007/s00586-012-2419-4.

36.

Wang

, et al. Percutaneous endoscopic lumbar discectomy for lumbar disc herniation with modic changes via a transforaminal approach: a retrospective study. Pain Physician. 2019;22(6):E601-E608.

37.

Bianchi

Peterson

Pfirrmann

CWA

Hodler

Bolton

. Are the presence of MODIC changes on MRI scans related to “improvement” in low back pain patients treated with lumbar facet joint injections? BMC Musculoskelet Disord. 2015;16:234. doi:10.1186/s12891-015-0688-x.

38.

Ohtori

Yamashita

Yamauchi

, et al. Change in Modic type 1 and 2 signals after posterolateral fusion surgery. Spine. 2010;35(12):1231-1235. doi:10.1097/BRS.0b013e3181bde562.

39.

Blondel

Tropiano

Gaudart

Huang

Marnay

. Clinical results of lumbar total disc arthroplasty in accordance with Modic signs, with a 2-year-minimum follow-up. Spine. 2011;36(26):2309-2315. doi:10.1097/BRS.0b013e31820f7372.

40.

Shahmohammadi

Behrouzian

. Effect of preoperative Modic change in the outcome of patients with low back pain following posterior spinal fusion or laminectomy. Asian J Neurosurg. 2019;14(2):432-435. doi:10.4103/ajns.AJNS_41_18.

41.

Ohtori

Yamashita

Yamauchi

, et al. Low back pain after lumbar discectomy in patients showing endplate modic type 1 change. Spine. 2010;35(13):E596-600. doi:10.1097/BRS.0b013e3181cd2cb8.

42.

MacLean

Kureshi

Shankar

Stewart

Christie

. Modic change and clinical assessment scores in patients undergoing lumbar surgery for disk herniation. Clin Spine Surg. 2021;34(4):E205-E210. doi:10.1097/BSD.0000000000001076.

43.

Määttä

Kautiainen

Leinonen

, et al. Association of Modic changes with health-related quality of life among patients referred to spine surgery. Scand J Pain. 2014;5(1):36-40. doi:10.1016/j.sjpain.2013.08.003.

44.

Matsumoto

Okada

Ichihara

, et al. Modic changes in the cervical spine: Prospective 10-year follow-up study in asymptomatic subjects. J Bone Joint Surg Br. 2012;94-B(5):678-683. doi:10.1302/0301-620x.94b5.28519.

45.

Baker

Harada

Tao

, et al. The impact of modic changes on preoperative symptoms and clinical outcomes in anterior cervical discectomy and fusion patients. Neurospine. 2020;17(1):190-203. doi:10.14245/ns.2040062.031.

46.

Kressig

Peterson

McChurch

, et al. Relationship of Modic changes, disk herniation morphology, and axial location to outcomes in symptomatic cervical disk herniation patients treated with high-velocity, low-amplitude spinal manipulation: a prospective study. J Manipulative Physiol Ther. 2016;39(8):565-575. doi:10.1016/j.jmpt.2016.08.004.

47.

Lei

Shen

. The impact of Modic-2 changes on the clinical outcomes of single-level anterior cervical discectomy and fusion. Eur Spine J. 2015;24(12):2936-2940. doi:10.1007/s00586-015-4161-1.

48.

Wei

Shen

. A study on the cervical spondylotic myelopathy treated by anterior cervical diskectomy and fusion in accordance with Modic changes with a 2-year minimum follow-up. J Orthop Surg. 2015;10:11. doi:10.1186/s13018-014-0146-8.

49.

Kong

Tian

Cao

Wang

Zhang

Shen

. Predictive factors associated with neck pain in patients with cervical disc degeneration: A cross-sectional study focusing on Modic changes. Medicine (Baltimore). 2017;96(43):e8447. doi:10.1097/MD.0000000000008447.

50.

Global Burden of Disease Study 2013 Collaborators . Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet Lond Engl. 2015;386(9995):743-800. doi:10.1016/S0140-6736(15)60692-4.

51.

Boden

Davis

Dina

Patronas

Wiesel

. Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg Am. 1990;72(3):403-408.

52.

Jensen

Brant-Zawadzki

Obuchowski

Modic

Malkasian

Ross

. Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med. 1994;331(2):69-73. doi:10.1056/NEJM199407143310201.

53.

Kjaer

Leboeuf-Yde

Korsholm

Sorensen

Bendix

. Magnetic resonance imaging and low back pain in adults: a diagnostic imaging study of 40-year-old men and women. Spine. 2005;30(10):1173-1180. doi:10.1097/01.brs.0000162396.97739.76.

54.

Herlin

Kjaer

Espeland

, et al. Modic changes-Their associations with low back pain and activity limitation: A systematic literature review and meta-analysis. PloS One. 2018;13(8):e0200677. doi:10.1371/journal.pone.0200677.

55.

Cuellar

Stauff

Herzog

Carrino

Baker

Carragee

. Does provocative discography cause clinically important injury to the lumbar intervertebral disc? A 10-year matched cohort study. Spine J. 2016;16(3):273-280. doi:10.1016/j.spinee.2015.06.051.

56.

Copay

Glassman

Subach

Berven

Schuler

Carreon

. Minimum clinically important difference in lumbar spine surgery patients: a choice of methods using the Oswestry Disability Index, Medical Outcomes Study questionnaire Short Form 36, and Pain Scales. Spine J. 2008;8(6):968-974. doi:10.1016/j.spinee.2007.11.006.

57.

Stratford

Binkley

Riddle

Guyatt

. Sensitivity to change of the Roland-Morris back pain questionnaire: part 1. Phys Ther. 1998;78(11):1186-1196.

58.

Minetama

Kawakami

Teraguchi

, et al. Endplate defects, not the severity of spinal stenosis, contribute to low back pain in patients with lumbar spinal stenosis. Spine J. 2022;22(3):370-378. doi:10.1016/j.spinee.2021.09.008.

59.

Feng

Liu

Yang

Battié

Wang

. Lumbar vertebral endplate defects on magnetic resonance images: classification, distribution patterns, and associations with modic changes and disc degeneration. Spine. 2018;43(13):919-927. doi:10.1097/BRS.0000000000002450.

60.

Wang

Videman

Battié

. Lumbar vertebral endplate lesions: prevalence, classification, and association with age. Spine. 2012;37(17):1432-1439. doi:10.1097/BRS.0b013e31824dd20a.

61.

Baker

Sayari

Harada

, et al. The Modic-endplate-complex phenotype in cervical spine patients: Association with symptoms and outcomes. J Orthop Res. 2022;40(2):449-459. doi:10.1002/jor.25042.

62.

Laustsen

Bech-Azeddine

. Do Modic changes have an impact on clinical outcome in lumbar spine surgery? A systematic literature review. Eur Spine J. 2016;25(11):3735-3745. doi:10.1007/s00586-016-4609-y.

63.

Lambrechts

Brush

Issa

, et al. Evaluating the impact of Modic changes on operative treatment in the cervical and lumbar spine: a systematic review and meta-analysis. Int J Environ Res Public Health. 2022;19(16):10158. doi:10.3390/ijerph191610158.

64.

Bråten

LCH

Rolfsen

Espeland

, et al. Efficacy of antibiotic treatment in patients with chronic low back pain and Modic changes (the AIM study): double blind, randomised, placebo controlled, multicentre trial. BMJ. 2019;367:l5654. doi:10.1136/bmj.l5654.

65.

Shea

GKH

Zhang

Suen

, et al. Oral Zoledronic acid bisphosphonate for the treatment of chronic low back pain with associated Modic changes: A pilot randomized controlled trial. J Orthop Res. Published online. 2022. doi:10.1002/jor.25304.

66.

Khalil

Smuck

Koreckij

, et al. A prospective, randomized, multicenter study of intraosseous basivertebral nerve ablation for the treatment of chronic low back pain. Spine J. 2019;19(10):1620-1632. doi:10.1016/j.spinee.2019.05.598.

67.

Becker

Hadjipavlou

Heggeness

. Ablation of the basivertebral nerve for treatment of back pain: a clinical study. Spine J. 2017;17(2):218-223. doi:10.1016/j.spinee.2016.08.032.

68.

Fayad

Lefevre-Colau

Rannou

, et al. Relation of inflammatory modic changes to intradiscal steroid injection outcome in chronic low back pain. Eur Spine J. 2007;16(7):925-931. doi:10.1007/s00586-006-0301-y.

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

1.24 MB