Multifidus Atrophy: Causes, Symptoms, and Treatment

Multifidus atrophy is the shrinking and fatty replacement of the multifidus, a deep spinal muscle that accounts for more than two-thirds of your lumbar spine’s stiffness. It develops after disc injuries, nerve damage, and spinal surgery, but it also creeps in with prolonged disuse and chronic pain. What makes it especially frustrating is that it can persist long after back pain resolves, quietly setting the stage for the next episode.

What the Multifidus Actually Does

The multifidus runs along the full length of the spine, but its thickest, most clinically relevant portion is in the lower back. Each segment of the lumbar multifidus spans just two or three vertebrae, giving it the ability to control motion between individual segments rather than moving the whole trunk at once. That architecture makes it the primary stabilizer of the lumbar spine, responsible for resisting shearing forces and compressive loads while you bend, twist, and lift.¹ Its contribution to spinal stability is well established in biomechanical research, though some clinical beliefs about the muscle still outpace the evidence supporting them.²

Think of the multifidus as the fine-tuning system for your spine. Larger muscles like the erector spinae generate the big movements, but the multifidus makes the small, precise adjustments that keep each vertebral segment from sliding or buckling under load. When it wastes away, you lose that segment-by-segment control. The spine can still move, but it does so with less precision and more vulnerability.³

What Causes Multifidus Atrophy

The causes fall into a few broad categories, and they often overlap in a given person.

Disc and Nerve Injury

Animal studies show that damage to a lumbar disc or a spinal nerve root triggers rapid multifidus wasting, sometimes within days. In one experimental model, a disc lesion reduced the multifidus cross-sectional area at the affected level, while nerve root injury caused shrinkage across multiple levels. Both types of injury led to enlarged fat cells and clustering of muscle fibers, hallmarks of the degenerative process.⁴ In people with chronic unilateral radiculopathy (nerve root compression on one side), the multifidus on the affected side is measurably smaller, particularly at the level where the compressed nerve root supplies the muscle.⁵

The nerve supply to the multifidus matters here. Each segment is innervated by a single branch of the dorsal ramus, meaning there is little backup wiring. Damage to that one branch can denervate the muscle at that level, and electrodiagnostic testing can identify this pattern alongside MRI findings of paraspinal atrophy.⁶

Spinal Surgery

Open spinal surgery is one of the most well-documented causes. The traditional posterior approach requires retracting the multifidus to reach the spine, which can stretch or crush the muscle and damage the nerve branches supplying it. In a study comparing minimally invasive surgery to conventional open fusion, the open approach produced about a 37% reduction in multifidus cross-sectional area at the operative level, compared with roughly 12% in the minimally invasive group. Fat infiltration followed the same pattern. Critically, the degree of surgical muscle damage was correlated with long-term atrophy, pain scores, and functional disability.⁷ Minimally invasive techniques preserve the nerve supply and muscle architecture much better.⁸

Pain-Driven Inhibition and Disuse

This is the most insidious cause because it does not require a dramatic injury. When the spine is painful, the nervous system reflexively dials down multifidus activation, a protective mechanism called arthrogenic muscle inhibition. The muscle does not just get weaker from lack of use; its recruitment pattern changes. Instead of the steady, low-level background activity the multifidus normally maintains, the brain switches it to brief bursts of activity that are less effective at stabilizing the spine. Over time, muscle fibers shift from the fatigue-resistant type that suits a postural muscle to a faster-twitch type that fatigues easily.⁹

What Happens Inside the Muscle

Multifidus atrophy is not simply a muscle getting smaller. The tissue undergoes a process where functional muscle fibers are progressively replaced by fat and fibrous tissue. Biopsies from people with degenerative lumbar conditions reveal elevated inflammatory cells, reduced blood vessel density, and widespread degeneration-regeneration cycling, where damaged fibers attempt to repair themselves but cannot keep up with the destruction.¹⁰ In people with disc herniations, microscopic analysis shows “moth-eaten” fibers and clustered fiber atrophy on the affected side, indicating the damage is localized and nerve-mediated.¹¹

A resident population of stem cells in the muscle, called fibro-adipogenic progenitors, plays a central role. In a healthy muscle, these cells support repair after minor damage. But in chronic injury states, they become the main drivers of fat and scar tissue accumulation, gradually converting what was once contractile muscle into inert padding.¹² Broader systemic factors such as obesity and chronic inflammation may also accelerate fat infiltration in the paraspinal muscles.¹³

How Symptoms Show Up

Multifidus atrophy does not announce itself with a sharp, distinctive pain. It tends to manifest as a vague sense that the back “gives way” or feels unstable during certain movements. People often describe deep aching in the lower back that worsens with sustained postures, difficulty controlling trunk movements during bending or lifting, and a feeling that the spine is unprotected. The clinical picture is one of functional instability: the spine can still move through its full range, but it lacks the fine-tuned muscular control to do so safely under load.

One of the most important clinical features is that the dysfunction typically outlasts the original pain. After a first episode of low back pain resolves, the multifidus often remains inhibited and atrophied. That persistent weakness is thought to contribute to the notoriously high recurrence rate of low back pain.¹⁴ You feel better, you resume normal activity, but the stabilizer muscle that was supposed to protect you never came back online.

How It Is Diagnosed

MRI is the gold standard for identifying multifidus atrophy because it can visualize both the size of the muscle and the extent of fat infiltration. A commonly used grading system categorizes the severity into four levels: normal (up to 10% fat), mild (10–30%), moderate (30–50%), and severe (over 50%).¹⁵ Research studies often use semi-automated software to measure the pure fat component and the overall cross-sectional area on axial MRI slices, typically at the L3-L5 levels.¹⁶

In a clinical rehab setting, ultrasound imaging offers a more practical and cheaper alternative. It cannot measure fat infiltration as precisely as MRI, but it can measure muscle thickness changes during contraction, providing a real-time window into whether the multifidus is actually firing. Ultrasound-measured thickness change correlates well with electromyographic activity of the multifidus.¹⁷ It can also be used as a biofeedback tool, letting you watch the muscle thicken on screen as you practice activating it.¹⁸

Why It Matters for Surgical Outcomes

The degree of multifidus atrophy before surgery turns out to be a meaningful predictor of how well things go afterward. In a study of patients undergoing microdiscectomy for disc herniation, those with moderate-to-severe fatty atrophy of the multifidus were roughly 3.4 times more likely to experience a recurrent disc herniation than those with normal or mildly affected muscles. Multifidus fatty atrophy was the single strongest risk factor for recurrence in that analysis.¹⁹ The implication is straightforward: a spine that lacks its primary muscular stabilizer is more vulnerable to re-injury, even after the structural problem has been surgically corrected.

The Relationship With Facet Joint Disease

The multifidus does not deteriorate in isolation. A recent study examined the link between facet joint osteoarthritis and multifidus fat infiltration in patients with lumbar spondylolisthesis. Advancing facet joint disease was associated with significantly more multifidus fat at multiple lumbar levels, and facet joint osteoarthritis emerged as an independent predictor of fat infiltration. Interestingly, disc degeneration, central canal stenosis, and foraminal stenosis did not contribute independently to multifidus fat in that analysis.²⁰ This suggests the facet joints and the multifidus may share a particularly tight mechanical and neurological relationship, one that is somewhat independent of other spinal degenerative processes.

Treatment Through Exercise

Targeted exercise is the first-line approach, and the evidence broadly supports that it can increase multifidus size. The key word is “targeted.” General back strengthening is not the same as specific multifidus reactivation. Most research that has shown improvements in multifidus dimensions has used motor control exercises, which involve learning to isolate and contract the multifidus at low loads before progressing to heavier, more dynamic tasks.²¹

A systematic review of motor control exercise for multifidus morphology found that roughly 36 sessions could produce detectable increases in cross-sectional area in people with chronic low back pain. However, the quality of the evidence was rated low to very low, and here is the uncomfortable part: changes in multifidus size from motor control exercise did not clearly correlate with improvements in pain or disability. The muscle got bigger, but that bigger muscle did not reliably translate into feeling better.²² That disconnect is one of the more honestly confusing findings in the field, and it suggests that multifidus size alone may not be the whole story. Neuromuscular control, the quality and timing of the muscle’s activation, likely matters as much as or more than raw bulk.

In a study of young elite cricketers with low back pain, stabilization training increased multifidus cross-sectional area at L5 and reduced the side-to-side asymmetry that had developed with injury. Pain levels dropped by about half. These effects were specific to the L5 level and were not seen at higher lumbar levels, which is consistent with the segmental nature of multifidus dysfunction.²³

Electrical Stimulation as an Add-On

For people who cannot voluntarily activate their multifidus well enough for exercise to be effective, neuromuscular electrical stimulation offers a way to force the muscle to contract externally. When combined with motor control exercise, electrical stimulation produced a significantly greater improvement in multifidus activation compared to motor control exercise alone in people with recurrent low back pain.²⁴ The intensity of the stimulation matters: people who tolerated higher current amplitudes experienced greater multifidus thickness changes during stimulation.²⁵

A more invasive form of electrical stimulation, called restorative neurostimulation, involves surgically implanting electrodes near the nerve branches that supply the multifidus. The device stimulates the muscle twice a day for 30-minute sessions, essentially forcing it to contract and rebuild over time. Two-year results from the pivotal trial showed that about 71% of participants achieved at least 50% pain relief, and 56% had substantial improvements in both pain and functional disability scores.²⁶ This is intended for people with chronic low back pain that has not responded to conservative treatment, and a 2025 guideline from the International Society for the Advancement of Spine Surgery summarizes the growing evidence base supporting it, including data out to five years.²⁷

Spine-Specific Sarcopenia Is Not the Same as General Muscle Loss

As people age, both general muscle mass and paraspinal muscle mass decline. It would be reasonable to assume these are the same process happening everywhere at once. They are not. Research shows a low correlation between generalized sarcopenia (whole-body muscle loss associated with aging) and spine-specific sarcopenia, suggesting the two conditions have distinct causes. Screening tools designed to detect general sarcopenia can miss spine-specific muscle loss entirely.²⁸ You can have a person whose arms and legs look reasonably muscular but whose multifidus has been quietly replaced by fat for years. Conversely, someone flagged for general sarcopenia might still have a relatively intact multifidus. The local biomechanical and neurological environment of the spine drives its own form of muscle wasting that does not simply mirror what is happening elsewhere in the body.²⁹

The Role of Fear and Avoidance Behavior

Psychology and paraspinal muscle health are more connected than most people expect. Fear-avoidance beliefs, the conviction that physical activity will worsen pain and cause further injury, are associated with greater fat infiltration in the erector spinae muscles in people with low back pain. Interestingly, this association held up even after adjusting for other factors, though it was specific to the erector spinae and did not reach statistical significance for the multifidus in that particular study.³⁰ Still, the broader principle is clear: people who fear movement tend to move less, and less movement means less muscular demand on the paraspinal muscles. Over months and years, that reduced demand contributes to the same fatty replacement seen in more dramatic causes like disc herniation or surgery.

Posture may play a role as well. People with a sway-back posture, where the pelvis shifts forward and the trunk leans back, have been shown to have greater fat deposition in both the multifidus and erector spinae compared to people with more neutral alignment. Pain may amplify the effect, but the postural deviation itself appeared to contribute independently.³¹ This does not mean sway-back posture inevitably causes atrophy, but it suggests that habitual standing and sitting positions can shift the mechanical load on the multifidus over time.

When Fat Replaces Muscle, Can You Get It Back

This is the question that matters most to people living with the condition, and the honest answer is: partially, and it depends on how far the process has gone. Early-stage dysfunction where the muscle is inhibited but still structurally intact responds well to targeted exercise. The muscle is still there; it just needs to be woken up. Ultrasound biofeedback, hands-on cueing from a physiotherapist, and electrical stimulation can all help re-establish the voluntary contraction that the nervous system has been suppressing.

Once significant fat infiltration has occurred, the picture is less optimistic. Fat and fibrous tissue within the muscle are unlikely to fully reverse with exercise alone. You can still strengthen whatever functional muscle remains, and you can improve the neuromuscular control of that muscle, both of which are clinically meaningful. But the MRI may never look like it did before the atrophy set in. That gap between structural restoration and functional recovery is part of why motor control exercise can improve pain and function for some people even when muscle size does not fully normalize. The brain gets better at using what it has, even if what it has is not a pristine muscle.

The practical takeaway is that earlier intervention gives you a larger window of opportunity. If you have had a first episode of acute low back pain, the time to start specific multifidus reactivation exercises is during recovery, not after your third recurrence years later. Clinicians who screen for multifidus dysfunction after disc injuries or spinal surgery are trying to catch the problem during this more reversible window, before the tissue replacement becomes entrenched.