MS lesions can appear on the spinal cord while the brain MRI looks normal, though this pattern is relatively uncommon. In one study of patients with a first episode of optic neuritis, spinal cord lesions showed up in about 12% of those whose brain MRI was completely clean.1Journal of Neurology, Neurosurgery & Psychiatry. Spinal cord MRI in clinically isolated optic neuritis The scenario raises real diagnostic questions, because spinal-only lesions can steer neurologists toward MS, away from it, or toward a different condition entirely.
How Often Spinal-Only Lesions Actually Occur
Most people with MS have lesions in both the brain and spinal cord. In a study comparing primary progressive MS (PPMS) and relapsing-remitting MS (RRMS), only about 15% of PPMS patients and roughly 8% of RRMS patients had lesions confined to the spinal cord alone.2PubMed Central. Brain and spinal cord MRI lesions in primary progressive vs. relapsing-remitting multiple sclerosis The vast majority in both groups had lesions in both locations. So while a spinal-only presentation is real, it sits at the edges of typical MS imaging patterns.
That said, “normal brain MRI” can be misleading. Spinal cord lesions are harder to detect than brain lesions because the cord is much smaller, moves with breathing and heartbeats, and standard MRI protocols sometimes skip it altogether. A brain MRI that looks clean at one point in time does not guarantee it will still look clean six months later. And a spinal MRI may miss small lesions if the technique is not optimized for cord imaging. The takeaway is that spinal-only lesions in MS exist, but they are the exception and they warrant a careful second look at the diagnosis.
How Spinal Cord Lesions Fit Into an MS Diagnosis
The 2017 revision of the McDonald criteria, which is the standard framework neurologists use to diagnose MS, treats spinal cord lesions as one of several locations that count toward meeting the requirement for disease spread across the nervous system. Specifically, a symptomatic spinal cord lesion can serve as evidence of dissemination in space, meaning the disease has affected more than one region of the central nervous system.3The Lancet Neurology. Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria But MS diagnosis also requires evidence that the disease has occurred at more than one point in time, called dissemination in time. A single spinal cord lesion on a single scan does not meet that requirement on its own.
This means a person presenting with spinal cord symptoms and a cord lesion but a clean brain MRI faces a diagnostic gray zone. If cerebrospinal fluid testing shows oligoclonal bands, those bands can substitute for dissemination in time under the current criteria. About two-thirds of MS patients test positive for these bands, though the rate varies depending on how early in the disease the test is performed.4PubMed Central. The Diagnostic Utility of Oligoclonal Bands in Multiple Sclerosis: A Time-Course Analysis So even without brain lesions, it is technically possible to meet MS diagnostic criteria through a combination of spinal imaging and lab findings. In practice, though, neurologists tend to monitor closely and repeat imaging rather than rushing to a definitive MS diagnosis when the brain is clear.
When Spinal-Only Lesions Suggest a Different Condition
A spinal cord lesion with a normal brain MRI should raise a red flag for conditions that are not MS. Two of the most important to rule out are neuromyelitis optica spectrum disorder (NMOSD) and myelin oligodendrocyte glycoprotein antibody disease (MOGAD). Both preferentially attack the spinal cord and optic nerves, and both can easily be mistaken for MS early on. Getting the right diagnosis matters enormously because the treatments differ, and some MS therapies can actually worsen NMOSD.
On MRI, several features help tell these conditions apart. MS spinal cord lesions tend to be short, patchy, and often sit toward the edges of the cord. In a multi-center Japanese study, the median MS spinal cord lesion was about 13 millimeters long, and lesions were distributed fairly evenly between the cervical and thoracic spine. NMOSD lesions, by contrast, had a median length of 47 millimeters and were more likely to be centrally located within the cord and concentrated in the thoracic region.5American Journal of Neuroradiology. Imaging Differences between Neuromyelitis Optica Spectrum Disorders and Multiple Sclerosis: A Multi-Institutional Study in Japan A blinded review of spinal MRIs confirmed that the distinction between MS and NMO can be made with moderately high confidence based on lesion characteristics alone, particularly during an acute attack.6PubMed Central. Differentiation of neuromyelitis optica from multiple sclerosis on spinal magnetic resonance imaging
In children, the distinction follows a similar logic. Longitudinally extensive lesions spanning three or more vertebral segments were found in only about 5% of pediatric MS cases, compared to 75% in children with MOGAD.7JAMA Network Open. Comparison of Spinal Cord Magnetic Resonance Imaging Features Among Children With Acquired Demyelinating Syndromes MOGAD lesions also showed a distinctive pattern of leptomeningeal enhancement and were far more likely to resolve completely on follow-up imaging. If a child or adult presents with a long spinal cord lesion, a clean brain, and no oligoclonal bands, the working diagnosis should lean away from MS and toward antibody-mediated disease until blood tests prove otherwise.
Where in the Spine MS Lesions Tend to Appear
When MS does affect the spinal cord, lesions cluster in the cervical region, particularly around the C3 and C4 vertebral levels. A study mapping lesion frequency across the entire cord in a mixed cohort of MS subtypes found the highest concentration at that level, with lesion volume increasing as the disease progressed from an initial episode through relapsing-remitting and then progressive phases.8PubMed Central. Multiple sclerosis lesions and atrophy in the spinal cord: Distribution across vertebral levels and correlation with disability The distribution itself did not change across subtypes, just the total amount of damage.
Within the cervical cord, higher-resolution mapping has revealed that PPMS patients have a higher density of lesions than RRMS patients, particularly in the lateral and central regions of the cord at C3.9Brain. Spatial distribution of multiple sclerosis lesions in the cervical spinal cord This matters because PPMS is the subtype most associated with gradual motor decline, and the lateral columns of the spinal cord carry the nerve fibers responsible for movement. The observation fits with what clinicians see at the bedside: people with progressive MS tend to develop walking difficulty as a prominent early feature.
Spinal Cord Lesions in Primary Progressive MS
PPMS deserves special mention because it behaves differently from the more common relapsing-remitting form. People with PPMS often have relatively few brain lesions compared to the severity of their symptoms. The spinal cord appears to carry a larger share of the disease burden. Research using advanced imaging techniques has shown that cord atrophy and diffuse tissue damage in PPMS correlate with disability even when brain lesion counts are low.10Brain. In vivo assessment of the brain and cervical cord pathology of patients with primary progressive multiple sclerosis
This disconnect between brain MRI appearance and clinical severity is one of the reasons PPMS was historically difficult to diagnose. Someone whose brain MRI looks relatively quiet might not meet the older imaging-based criteria, even though their spinal cord tells a very different story. The 2017 McDonald criteria partially addressed this by allowing spinal cord lesions to count toward dissemination in space, but the mismatch between brain and cord findings remains a real diagnostic puzzle in PPMS.
Spinal Lesions as an Early Warning Sign
Some people undergo brain MRI for unrelated reasons, like a headache or a head injury, and incidentally turn up findings suggestive of demyelination even though they have never had neurological symptoms. This situation is called radiologically isolated syndrome (RIS). The critical question for these individuals is whether they will go on to develop MS. The answer depends heavily on whether the spinal cord is also involved.
In a study of 71 people with RIS, those with asymptomatic spinal cord lesions had dramatically higher odds of eventually developing clinical symptoms. About 84% of RIS subjects who had cervical spine lesions went on to experience a first demyelinating event or develop progressive MS, typically within about a year and a half. The odds of clinical progression with a spinal cord lesion were very high, with sensitivity and specificity both above 85%.11PubMed Central. Asymptomatic spinal cord lesions predict disease progression in radiologically isolated syndrome This risk was independent of brain lesion findings, meaning the spinal cord lesion itself added unique predictive information.
A similar pattern holds after a first clinical event like optic neuritis or a single episode of numbness, known as clinically isolated syndrome (CIS). In CIS patients who did not meet brain MRI criteria for MS, the presence of a spinal cord lesion was associated with a substantially higher risk of converting to definite MS.12Neurology. Spinal cord lesions in patients with clinically isolated syndrome: a powerful tool in diagnosis and prognosis The spinal cord, in other words, often telegraphs where the disease is heading before the brain catches up.
How Spinal Cord Lesion Location Relates to Pain
One of the most debilitating symptoms in MS is central neuropathic pain, a burning or shooting pain that originates from damage to the nervous system itself rather than from an injury to body tissues. Not everyone with spinal cord lesions develops this kind of pain, and the reason appears to come down to exactly where within the cord the lesion sits.
A study comparing MS patients with and without central neuropathic pain found a striking pattern. Patients with pain almost universally had a centrally located lesion in the upper-to-mid thoracic cord, between T1 and T6. The sensitivity of this finding was about 97%, and the specificity was above 83%. The association was powerful: having a centrally placed thoracic lesion increased the odds of developing central neuropathic pain roughly 155-fold compared to patients without that lesion pattern.13PubMed Central. Central neuropathic pain in MS is due to distinct thoracic spinal cord lesions Patients whose thoracic lesions were off-center or in the lower thoracic spine were far less likely to have this symptom. The finding suggests that certain tracts running through the middle of the upper thoracic cord are especially vulnerable to producing chronic pain when damaged.
Blood Biomarkers and What They Add
MRI captures the structural damage of MS, but blood tests are increasingly used to measure the biological activity underneath. Neurofilament light chain (NfL) is a protein released when nerve fibers are damaged, and it can be measured with a simple blood draw. In MS patients, NfL levels rise when there is active inflammation in the brain, the spinal cord, or both.
A study measuring serum NfL found that patients with gadolinium-enhancing lesions in the brain or spinal cord had higher NfL levels than patients without active lesions. Patients who had active lesions in both locations showed the highest levels.14PubMed Central. Serum Neurofilament light: A biomarker of neuronal damage in multiple sclerosis For someone with spinal-only lesions, elevated NfL provides an additional signal that active nerve damage is occurring, which may help guide decisions about starting or escalating treatment even if the brain MRI does not look alarming.
Starting Treatment After a First Spinal Event
When someone has a first demyelinating event affecting the spinal cord, whether it is diagnosed as CIS or as a first MS attack, the question of when to begin disease-modifying therapy looms large. Evidence from clinical trials consistently shows that starting treatment at the CIS stage reduces the likelihood of converting to definite MS. A systematic review and network comparison of several disease-modifying therapies found that all of them significantly reduced conversion risk compared to placebo.15Medicina. The efficacy of disease-modifying therapies in patients with clinically isolated syndrome: a systematic review and network meta-analysis
For people with spinal cord involvement specifically, the case for early treatment is arguably stronger than for those with brain-only findings. The spinal cord is a compact structure where even a small lesion can cause meaningful motor or sensory problems, and the prognostic data discussed earlier show that spinal lesions predict a faster trajectory toward definite disease. Waiting to see if brain lesions eventually appear before starting therapy means accepting a period of risk when new cord damage could cause lasting disability.
Pediatric Spinal Cord Findings
Children with MS do develop spinal cord lesions, though the imaging picture can look different from adults. In a study of early pediatric MS, about a third of children had at least one cervical cord lesion. Those with lesions tended to be older and were more likely to have had a previous episode of cervical myelitis. None of the lesions in the study showed active gadolinium enhancement at the time of imaging.16Annals of Clinical and Translational Neurology. Spinal Cord Abnormalities in Early Pediatric Multiple Sclerosis
The differential diagnosis in children is even more important than in adults, because MOGAD is relatively more common in pediatric populations. As noted earlier, children with MOGAD are far more likely to have long, extensive spinal lesions than children with MS, and MOGAD lesions tend to resolve more completely over time.7JAMA Network Open. Comparison of Spinal Cord Magnetic Resonance Imaging Features Among Children With Acquired Demyelinating Syndromes A child with a long cord lesion and a clean brain should be tested for MOG antibodies before anyone assumes MS.
Progressive Solitary Sclerosis
At the far end of the spectrum sits a rare condition called progressive solitary sclerosis, where a single demyelinating lesion causes gradually worsening neurological impairment without ever spreading to other parts of the nervous system. In a case series, the most common location for these solitary lesions was the cervical spinal cord, followed by the junction of the cervical cord and brainstem.17Neurology. Progressive solitary sclerosis: Gradual motor impairment from a single CNS demyelinating lesion These patients do not meet criteria for MS because the disease never disseminates, yet they experience real and progressive disability from that one lesion. The condition is a reminder that the spinal cord’s tightly packed anatomy means a single lesion in the right spot can do significant harm, and that not every demyelinating lesion fits neatly into the MS framework.