How the Erector Spinae Muscles Support the Spine and Core

The erector spinae are a group of long muscles that run alongside the spine from the pelvis to the base of the skull, forming the primary extensors of the back. They keep you upright when you stand, control your descent when you bend forward, and fire rhythmically with every step you take. Despite being among the most heavily used muscles in the body, they tend to get far less attention than the “mirror muscles” people see at the gym. That obscurity is unfortunate, because the erector spinae sit at the center of posture, back pain, surgical planning, and even modern pain-management techniques.

What the Erector Spinae Actually Are

The name “erector spinae” is a collective label for three parallel columns of muscle that run vertically along the spine. From closest to the spine outward, they are the spinalis, longissimus, and iliocostalis. Each column has regional subdivisions named for the part of the spine they serve. In the lumbar region, the longissimus thoracis and iliocostalis lumborum are separated by the erector spinae aponeurosis, a tough sheet of connective tissue. The lumbar fibers of longissimus arise from the ilium and insert into the accessory processes and transverse processes of the lumbar vertebrae, while the iliocostalis fibers insert into the costal elements of the first four lumbar vertebrae.¹ This layered arrangement, with a superficial part corresponding to the iliocostalis lumborum and upper longissimus and a deep part corresponding to the lower longissimus, means that different portions of what feels like one big muscle mass can be activated somewhat independently.²

That selective activation matters more than most people realize. Studies placing electrodes at different locations along the thoracic erector spinae show that the body can shift the workload between upper and lower regions depending on the task. The position of the trunk and the edge of a support surface influence which segments light up on electromyography.³ After spinal cord injury, this regional control changes markedly: activation concentrates in a smaller area, and the normal difference between extending forward and returning upright disappears.⁴ The takeaway is that the erector spinae are not a simple on-off switch. They are a set of regional units that the nervous system orchestrates with surprising precision.

Built for Endurance, Not Speed

If you’ve ever wondered why your back muscles can hold you upright for hours but cramp when you try to do explosive movements with them, the answer lies in their fiber makeup. The erector spinae have a clear predominance of relatively large type I (slow-twitch) fibers, which is unusual compared with most other skeletal muscles and reflects their primary job as postural muscles.⁵ Slow-twitch fibers resist fatigue and rely heavily on oxygen-dependent energy production, which is why training the erector spinae effectively tends to involve sustained or repeated efforts rather than single explosive reps.

This endurance-oriented profile also explains why blood flow to the muscle matters so much. In people with chronic low back pain, the erector spinae show impaired oxygenation during exercise. An exercise therapy program was able to significantly improve oxygen delivery and blood volume in the muscle, with oxygenation parameters jumping by roughly 20 to 60 percent after the training period.⁶ When the oxygen supply to a slow-twitch-dominant muscle improves, the muscle can do its job longer before fatigue sets in.

How Spinal Posture Changes Their Leverage

The erector spinae do not generate the same force in every position. Their mechanical advantage depends on how far they sit from the center of the spinal joints they are extending, and that distance changes as the spine curves. When the lumbar spine is in its normal lordotic curve, the lever arm lengths average around 60 to 63 mm. Flatten or round the spine into kyphosis, and those lever arms shrink to roughly 49 to 57 mm, a drop of 10 to 24 percent.⁷ A shorter lever arm means the muscle has to generate more force to produce the same extension torque, which in turn increases compression on the lumbar discs.

This effect is most pronounced at the lowest spinal levels. Going from a neutral standing position to 45 degrees of trunk flexion reduces the erector spinae moment arm at L5-S1 by about 9 to 10 percent, and the reduction correlates with body size and spinal curvature.⁸ This is one reason lifting coaches emphasize keeping a neutral lumbar curve: not because a rounded back is automatically dangerous, but because a rounded posture demands more muscle force and produces more joint compression for the same external load.

The Flexion-Relaxation Phenomenon

One of the more counterintuitive things the erector spinae do is shut off. When you bend all the way forward to touch your toes, the muscle activity in the erector spinae drops to near zero at a certain point in the bend, a phenomenon researchers call flexion-relaxation. The electrical silence kicks in at roughly 84 percent of maximum vertebral flexion and at about 57 percent of maximum hip flexion, then snaps back on at nearly the same angle during the return to standing.⁹

What holds you up when the muscles go quiet? The passive structures of the posterior spine, mainly the ligaments, joint capsules, and the stretched muscle tissue itself, take over the extension torque. The posterior disc-ligament complex and contributions from deeper muscles like the quadratus lumborum share the load while the erector spinae rest.¹⁰ People with chronic low back pain often lose this normal relaxation response, meaning their erector spinae stay active through the full range of bending. Clinicians sometimes use the absence of flexion-relaxation as an assessment finding: if the muscles never turn off, the system’s load-sharing strategy has broken down.

Walking, Running, and the Spine

You might think of the erector spinae as muscles that matter mainly when lifting or bending, but they are active during every stride. As gait speed increases, the activation of the erector spinae and the deeper multifidus rises, although the rate of increase tapers off at higher speeds.¹¹ Unlike the deep transversus abdominis, which stays tonically active during most walking and running, the erector spinae fire in bursts timed to specific phases of the gait cycle, and the relative duration of those bursts shifts with speed and mode of locomotion.¹²

From an evolutionary standpoint, the transition to habitual bipedal walking in humans did not require a wholesale reinvention of back muscle function. Comparative research on bipedal walking in chimpanzees and gibbons suggests that the shift mainly involved improving the mechanical advantage and architecture of the existing erector spinae rather than changing what they fundamentally do.¹³ Our upright posture is less a feat of novel musculature and more an optimization of the same basic toolkit other primates already had.

How the Erector Spinae Coordinate with the Core

The erector spinae do not stabilize the spine alone. They work in concert with the thoracolumbar fascia (a diamond-shaped sheet of connective tissue covering the lumbar back muscles), the abdominal muscles, and intra-abdominal pressure. Research modeling how these tissues share the workload shows a clear division of labor: the muscles act as antagonists to external forces, the thoracolumbar fascia dissipates and absorbs excessive loads, and intra-abdominal pressure limits unwanted movement. Different tasks call for different combinations, with the tissues optimizing their coordination to minimize individual strain.¹⁴

Tensioning the thoracolumbar fascia can also alter the stiffness of the muscles underneath it. Stretching the latissimus dorsi, which attaches to the fascia, changes the measurable stiffness of the erector spinae and multifidus as detected by shear-wave elastography.¹⁵ This interconnection is why “core training” that focuses only on crunches and planks misses part of the picture. The posterior muscles and the fascia wrapping them are as central to trunk stability as the abdominals.

Erector Spinae and Low Back Pain

The relationship between the erector spinae and low back pain is one of the most studied topics in musculoskeletal research, and the picture is messier than simple “weak muscles cause back pain” narratives suggest. A systematic review and meta-analysis found that while the multifidus muscle shows consistent atrophy and fatty infiltration in people with chronic low back pain, differences in the erector spinae were less consistent, often limited to specific lumbar levels or detectable only with more sensitive analysis methods.¹⁶

That said, the erector spinae are not off the hook. One study found that patients with low back pain had notably severe fatty infiltration in the erector spinae specifically at the upper lumbar levels, even while the multifidus was fattiest at the lower levels.¹⁷ Another cross-sectional study found that fat content in both the multifidus and erector spinae was higher in people with continuous chronic low back pain than in those with recurrent episodes, even though total muscle size did not differ between groups.¹⁸ And a systematic review with attention to blinding found that paraspinal muscles generally were smaller in chronic low back pain patients compared with controls, and smaller on the symptomatic side in people with one-sided pain.¹⁹

The pattern of fatigue also differs. In people with low back pain, the thoracic part of the erector spinae shows lower median frequency on electromyography than the lumbar part, a sign that the upper segments are fatiguing differently from the lower ones, and in a pattern distinct from pain-free individuals.²⁰ The overall picture is that chronic low back pain does not simply shrink the erector spinae in a uniform way. It alters their composition, their fatigue behavior, and their regional activation, often in patterns that vary by spinal level.

Training the Erector Spinae

When it comes to exercise selection, the erector spinae activate reliably during any movement that involves hip extension with trunk stabilization. Comparing three single-joint hip extension exercises in trained women, researchers found no difference in erector spinae activation between them, meaning all three were equally effective at working the low back.²¹ Among compound movements, deadlifts and their variants consistently produce high erector spinae activation, often greater than the activation seen in the glutes or hamstrings during the same lift.²² One exception: the Romanian deadlift tends to bias the hamstrings over the erector spinae.

Equipment choice matters, too. The conventional straight-barbell deadlift produces greater erector spinae activation during the lowering phase than the hexagonal (trap-bar) deadlift.²³ If you specifically want to load the erector spinae, a straight bar and a conventional stance give you more back-muscle demand, while a hex bar shifts some of that work toward the legs.

For rehabilitation, both stabilization exercises and dynamic strengthening produce significant improvements in lumbar extension strength after about eight weeks, and both reduce pain. One study found that the stabilization approach led to greater functional improvements in daily-life disability scores compared with dynamic strengthening alone.²⁴ The practical message is that there is no single best exercise. A mix of controlled stabilization work and progressive loading tends to cover both the endurance and strength deficits that develop in the erector spinae during chronic pain.

What Happens with Aging and Sarcopenia

As people age, the erector spinae and multifidus lose cross-sectional area and accumulate fat, mirroring what happens in limb muscles but with consequences specific to spinal loading. Modeling studies suggest that normal age-related decline down to about 60 percent of peak strength has only a minor impact on spinal loads and requires only slightly elevated muscle activity to compensate. The problems escalate with sarcopenia. At severe levels, with muscle strength reduced to about a third of normal, compression forces on the upper thoracic spine can jump by up to roughly 36 percent, and shear loads along the whole spine can increase by as much as 75 percent.²⁵ Those added loads help explain why fragility fractures in the thoracic spine become more common in advanced age, because weakened erector spinae cannot distribute forces evenly, and the vertebrae bear the excess.

The Erector Spinae Plane Block

One of the more surprising clinical applications of the erector spinae in recent years has nothing to do with movement or rehabilitation. It involves injecting local anesthetic deep to the muscle. The erector spinae plane (ESP) block, introduced in 2016, has become a widely used regional anesthesia technique for pain control after surgeries on the chest, abdomen, and spine. The primary mechanism appears to be the physical spread and diffusion of anesthetic through the fascial plane beneath the erector spinae to the ventral rami of spinal nerves, though the extent of spread varies. Dorsal rami are consistently reached, which relieves posterior trunk and incision pain, while epidural spread occurs less commonly.²⁶

Cadaver studies confirm that dye injected into the ESP reaches the paravertebral space and intercostal spaces, staining a mean of about 4 to 5 intercostal levels per injection, with a range of 3 to 7.²⁷ The block’s appeal for anesthesiologists is that it is technically simpler and lower-risk than paravertebral or epidural injections, because the needle target is a superficial muscular plane rather than a space adjacent to the spinal cord. The erector spinae, in this context, are not the treatment target but the anatomical landmark that makes the procedure safe.

Surgical Vulnerability

The same anatomy that makes the erector spinae useful as a landmark during nerve blocks also makes them vulnerable during posterior spinal surgery. The medial multifidus, which sits just inside the erector spinae mass, is innervated only by the medial branch of the dorsal ramus and has no backup nerve supply from adjacent segments. Surgical dissection, prolonged retraction, and denervation during posterior lumbar approaches can all injure the erector spinae and multifidus.²⁸ MRI studies after lumbar fusion have shown that erector spinae cross-sectional area distal to the fusion site can decrease by about 27 percent, and attempts to use muscle-relaxing agents during surgery did not prevent the damage.²⁹ This post-surgical muscle wasting is one of the recognized contributors to “failed back surgery syndrome,” where patients have persistent pain and stiffness after an operation that appeared successful on imaging. Minimally invasive spine surgery techniques aim in part to reduce this collateral damage to the erector spinae.

Measuring Erector Spinae Health

Clinicians increasingly want to measure erector spinae condition objectively rather than relying only on strength tests or patient-reported pain. Shear-wave elastography, an ultrasound-based technique that measures tissue stiffness, has emerged as a reliable tool for assessing the erector spinae, with reliability estimates ranging from fair to excellent. Accuracy improves when clinicians use the average of three measurements and when the subject is seated rather than lying flat.³⁰ Interestingly, while a five-week course of therapeutic massage might be expected to change muscle stiffness, one randomized controlled trial found no statistically significant effect on erector spinae stiffness compared with a control group.³¹ Massage may feel good, but the underlying mechanical properties of the erector spinae do not appear to shift measurably with manual therapy alone, at least over that time frame.

MRI-based measurements of fatty infiltration and cross-sectional area remain the gold standard for research, though they are expensive for routine clinical use. The disc degeneration grading at a given spinal level turns out to be one of the strongest predictors of how much fat has infiltrated the adjacent paraspinal muscles, suggesting the disc and the muscles deteriorate in tandem rather than independently.³² For people wondering whether their back pain is a “muscle problem” or a “disc problem,” the answer, frustratingly, is often both.