Yes — scoliosis can cause SI joint pain, but usually through mechanics rather than inflammation. The curve tilts the sacral base and loads one sacroiliac joint more than the other; over years, that asymmetry can provoke pain deep in the buttock and posterior pelvis. The link is real but weaker than most online explainers imply. Curve size alone predicts little, and many people with scoliosis have SI pain from an unrelated cause. What imaging and injection studies actually show is separated below from what remains plausible theory.
How a Curved Spine Loads the SI Joint
The sacroiliac joint is the only bony bridge between the spine and the legs. Each joint allows 2 to 4 degrees of rotation and less than 2 mm of translation, which sounds trivial until you consider what it transmits. That restraint is the design: the joint behaves as a force transmitter, not a hinge.
In relaxed standing, model and instrumented estimates put the reaction force through the sacroiliac joints near 400 to 500 N. Single-leg stance, stair climbing, and rising from a low chair push the loaded joint past 2,000 N. The joint tolerates that because load arrives symmetrically and spreads across a broad, textured articular surface.
Scoliosis breaks the symmetry. A thoracic curve shifts the trunk's center of mass sideways, and the lumbar spine compensates beneath it. The sacrum then meets the pelvis at an angle, so the sacral base is no longer level in the frontal plane. Load follows the tilt.
Axial rotation matters as much as the sideways shift. Vertebral bodies rotate toward the convexity, and apical rotation of 15 degrees or more is common once a curve passes 30 degrees. That rotation travels down the lumbar spine and twists the sacrum inside the pelvic ring. One sacroiliac joint ends up compressed while the other is strained.
The restraining ligaments absorb the difference. The long posterior sacroiliac ligament and the iliolumbar ligament are densely innervated, and they are the structures most likely to generate pain when load becomes one-sided. In plain terms: the curve rarely damages the joint directly, it teaches the pelvis to stand crookedly for decades.
The asymmetry stays quiet at first. Ligaments adapt, muscles hypertrophy, and the nervous system recalibrates. Symptoms usually appear when a second factor arrives — a leg-length difference, a pregnancy, a fall, or simple age-related stiffening of the joint capsule.
Can Scoliosis Cause SI Joint Pain? What Studies Show
The honest answer is that the link is plausible, documented, and modest. No large prospective study has measured injection-confirmed sacroiliac pain in a defined scoliosis population. What exists is a stack of indirect evidence, and it points in one direction without settling the size of the effect.
Start with the base rate. Injection-confirmed sacroiliac joint pain accounts for roughly 15 to 30 percent of chronic mechanical low back pain in adults, a range established by double-blockade studies summarized by Cohen in 2005. Any adult with scoliosis therefore carries that background risk before the curve is considered at all.
Adult deformity cohorts push the number higher. Series of adults with degenerative and idiopathic spinal deformity generally place sacroiliac pain somewhere between one in five and one in three patients. Comparison groups are small and inconsistently matched, so treat that range as directional rather than precise.
Imaging supports a mechanical story. Computed tomography studies of scoliosis patients show sacral and iliac subchondral sclerosis, osteophytes, and joint-space narrowing more often than in matched controls. Degeneration clusters on the concave side of a lumbar curve, which is where compressive load concentrates.
Correlation with curve size is weak. Across adolescent and adult series, Cobb angle explains only a small fraction of pain variance, with reported correlation coefficients clustered around 0.1 to 0.3. A 60-degree curve is not reliably more painful than a 25-degree one.
The Scoliosis Research Society's clinical position reinforces that caution. Adolescent idiopathic scoliosis is typically painless, and pain in a teenager is a reason to look for another cause rather than to assume the curve is responsible. That single point corrects most of what the internet says on this question.
Which Curve Patterns Predict SI Pain
Not every curve loads the pelvis the same way. The features that matter are location, rotation, and what happens at the lumbosacral junction.
Thoracolumbar and lumbar curves carry the strongest mechanical case. Lenke type 5 curves, in which the lumbar or thoracolumbar region carries the major curve, sit directly above the sacrum. Curves confined to the thoracic spine can be large and still leave the sacral base nearly level.
Sacral obliquity is the measurable intermediate. A lumbar curve of 30 degrees or more usually produces several degrees of frontal tilt at the sacral base, and the tilt tracks the curve. Radiologists can measure it on a standing full-spine film, but rarely report it unless the ordering clinician asks.
Curve direction does not reliably predict which side hurts. Some case series find pain on the concave side, others on the convex side, and that inconsistency is worth stating plainly rather than dressing up as a rule. In plain terms: assume the mechanics are asymmetric, not that the pain must land on a predictable side.
Leg length is the most overlooked contributor. A true or functional difference of 5 mm to 10 mm is enough to tilt the pelvis and create a compensatory curve that mimics scoliosis. The distinction matters because a heel lift trial costs little and can resolve the pain entirely when the difference is real.
Pelvic incidence mismatch is a separate axis of loading. When pelvic incidence exceeds lumbar lordosis by more than 10 degrees, the spine is mechanically obliged to lean forward and the sacroiliac joints absorb the compensation. Schwab's adult deformity classification places 10 to 20 degrees in a moderate band and above 20 degrees in a marked one.
Spinal fusion rewrites the map. Long constructs ending at S1 or the pelvis transfer stress to the sacroiliac joint below, and new buttock pain appears in a meaningful minority of patients within 1 to 5 years. Reported rates vary widely, from roughly 10 to 40 percent, depending on construct length and follow-up.
Pregnancy is a second, temporary amplifier. Relaxin increases sacroiliac laxity from about the 10th week of gestation, and parous women report sacroiliac pain more often than nulliparous women. A pre-existing scoliotic asymmetry stacks directly on top of that laxity.
Telling SI Pain Apart From Scoliosis Pain
Three misconceptions do most of the damage in this area, and each has a practical counter.
The first is that scoliosis explains all back pain in a person who happens to have it. It does not. Among adults over 60, hip osteoarthritis and lumbar spinal stenosis are common enough to compete directly, and both refer pain into the buttock and thigh.
The second is that imaging identifies the pain source. It identifies degeneration. Sacroiliac sclerosis and osteophytes are near-universal after age 50, so a scan read as degenerative change cannot by itself prove that the joint is generating the pain.
The third is that curve size determines symptoms. Weinstein's long-term Iowa follow-up found that thoracic curves above 30 degrees at skeletal maturity progress roughly 1 degree per year, while smaller curves often do not progress at all. Pain does not follow that line.
These thresholds serve different decisions and do not conflict: 10 degrees defines a curve, 25 to 40 degrees opens bracing in a growing child, and 45 to 50 degrees opens the surgical discussion.
| Feature | SI joint pain | Scoliosis-related spinal pain | Hip osteoarthritis | Lumbar radiculopathy |
|---|---|---|---|---|
| Typical location | Buttock below the belt line, sometimes groin or posterior thigh | Paraspinal or midline, along the curve | Groin and anterolateral thigh | Below the knee, in a dermatome |
| Provoked by | Single-leg standing, stairs, rolling over in bed, prolonged sitting | Prolonged standing or walking, end-of-day fatigue | Hip flexion, internal rotation, walking more than 15 minutes | Sitting, coughing, straight-leg raise |
| Relieved by | Lying flat, shifting weight off the leg | Sitting, changing position | Rest, leaning forward on a cart | Position change, sometimes walking |
| Key exam finding | Three or more positive provocation tests | Focal paraspinal tenderness, rib hump | Internal rotation limited below 15 degrees | Positive straight-leg raise, reflex change |
| Imaging signal | Sclerosis, vacuum phenomenon, subchondral cysts | Cobb angle, apical rotation, pelvic parameters | Joint-space narrowing under 2 mm | Disc herniation matching the dermatome |
The Time Course: From Curve to Chronic Pain
The table below is a population average, not a forecast for any one person. Most people with a moderate curve never develop sacroiliac pain at all, and the sequence described here reflects the minority who do.
| Stage | What changes at the pelvis | Typical symptom | Useful test |
|---|---|---|---|
| Age 10–14, curve passes 10 degrees | Sacral base begins to tilt marginally | Usually none | Standing PA radiograph |
| Age 14–18, curve 20–40 degrees | Apical rotation reaches the sacrum; several degrees of sacral obliquity | Intermittent ache after sport | Full-spine standing film with pelvic parameters |
| Age 18–30 | Curve stable in most; obliquity now fixed | Buttock pain after 60 or more minutes of standing | Sacroiliac provocation cluster |
| Age 30–45 | Asymmetric sclerosis appears, worse on the concave side | Pain with single-leg loading such as stairs or exiting a car | CT or MRI to grade degeneration |
| Age 45–60 | Disc height loss below the curve adds a second generator | Mixed midline and buttock pain | Image-guided diagnostic injection |
| Age 60 and older | Degenerative scoliosis; sacroiliac degeneration near-universal; hip arthritis competes | Pain at rest, night pain | Hip radiograph plus sacroiliac imaging |
| 1–5 years after long fusion | Construct ending at S1 or pelvis concentrates stress below | New buttock pain where none existed before | Provocation tests and injection |
Two features of that sequence deserve emphasis. First, the interval between the curve and the pain is usually measured in decades, not months. Second, the arrival of symptoms often coincides with a discrete event that changed load — a new job requiring standing, a pregnancy, a knee injury, or a fall.
What to Actually Do
Begin with imaging that answers the mechanical question. A standing full-spine posteroanterior and lateral film with pelvic parameters gives Cobb angle, apical rotation, pelvic incidence, pelvic tilt, and sacral slope in one study. Add a scanogram or a block test when a leg-length difference is plausible, because a 5 mm to 10 mm difference changes the treatment plan more than most curve findings do.
Examine the joint directly rather than inferring it from the spine. The Fortin finger test, Gaenslen maneuver, thigh thrust, compression, distraction, and sacral thrust form a reasonable cluster. Three or more positive tests raise the probability meaningfully; no single test is worth much on its own.
Confirm with an image-guided injection when the diagnosis changes management. An intra-articular local anesthetic injection producing 75 percent or greater relief is the current reference standard, ideally repeated with a different agent on a separate visit. Relief that lasts only for the duration of the anesthetic is a diagnostic result, not a treatment failure.
Run a conservative trial of 8 to 12 weeks before escalating. Target hip abductor strength, trunk control, and gait symmetry, and reduce habitual unilateral loading such as standing with weight on one leg. Evidence specific to scoliosis is thin, but this trial length is standard in sacroiliac pain generally and carries little downside.
For a documented leg-length difference, try a 5 mm to 10 mm heel lift for 6 to 8 weeks and reassess. If the pain resolves, the curve was at least partly compensatory and no spinal intervention is warranted.
For a growing child with a curve between 25 and 40 degrees, bracing remains the standard option. For a curve of 45 to 50 degrees or more, surgical consultation is reasonable, though the conversation should include the 10 to 40 percent rate of new sacroiliac pain after long fusions.
If a fusion has already been performed and buttock pain is new, evaluate the sacroiliac joint before labeling the problem adjacent segment disease. The two look alike on history and demand different treatments.
Seek urgent evaluation for any of the following:
- Fever, night pain that wakes you, or unexplained weight loss over 5 percent in six months
- New bowel or bladder dysfunction, saddle numbness, or weakness in both legs
- Pain after a fall, or an inability to bear weight on one leg
- Curve progression of more than 10 degrees within a year in a skeletally mature adult
- A left thoracic curve, a short sharply angulated curve, or any curve in a child under 10 years old
- Progressive neurologic deficit in the legs
What the Evidence Still Cannot Say
Three gaps are worth naming, because each determines how confidently anyone can answer the title question.
The first is the absence of a dedicated trial. No randomized study has compared sacroiliac treatment with sham treatment in a scoliosis population, and the injection literature that does exist enrolls patients selected for suspected sacroiliac pain rather than for spinal deformity. The 15 to 30 percent base rate comes from that selected population, not from an unselected scoliosis cohort.
The second is confounding by age. Scoliosis prevalence climbs steeply after 60, and so does sacroiliac degeneration. Separating the two requires age-matched controls that most published series lack, so the reported excess in deformity cohorts may be smaller than it appears.
The third is whether correcting the curve relieves the joint. Results after surgery are mixed. Some patients report buttock pain resolution, others develop it for the first time, and the 10 to 40 percent figure after long fusion suggests the relationship is not simply one of unloading a compressed joint.
What remains defensible is a mechanical sequence grounded in anatomy and supported by imaging patterns. Treat it as a plausible contributor rather than a proven cause, and treat the sacroiliac joint as a testable diagnosis rather than an assumption. The injection response, the heel lift trial, and the physical therapy block each cost relatively little and each answers a question the imaging cannot.