SI Joint Out of Place? What the Evidence Actually Shows

Key Takeaways: SI Joint Out of Place?

• Fascial connections between the pelvis and upper body are anatomically real, but the claim that an SI joint positional fault routinely causes distant symptoms remains an unproven and inadequately tested hypothesis.
• SI joint pain is real, but pain near the joint does not prove that it is subluxated, misaligned or that it is “SI joint out of place.”
• The SI joint normally moves only small amounts, making repeated clinically meaningful displacement difficult to reconcile with its known biomechanics.
• Palpation and positional tests cannot reliably identify a verifiable SI joint positional fault. Pelvic asymmetry may instead reflect pain-related posture, protective muscle tension or altered movement.
• Provocation-test clusters are more useful for ruling SI joint pain out than confirming it. Even image-guided anesthetic injections—the closest available reference test—have important limitations.
• Suspected inflammatory sacroiliitis in axial spondyloarthritis is a separate clinical situation. Patrick–FABER may support further assessment, but it cannot establish the diagnosis without the broader clinical picture and appropriate imaging.
• MRI findings require clinical context. Degenerative spinal changes are common in people without pain, while unnecessary imaging may lead to anxiety, further interventions and potentially avoidable surgery without improving outcomes.
• Manual therapy may provide genuine short-term relief, but improvement does not prove that the joint was physically repositioned. Passive treatment should not indefinitely replace active rehabilitation when the latter is appropriate.

Introduction: SI Joint Out of Place?

If you have searched online for the cause of low back or buttock pain, you have probably encountered the idea that your sacroiliac (SI) joint has slipped “out of place” and needs to be manually put back. It is a compelling explanation: the problem sounds mechanical, the proposed solution is straightforward, and some patients do report feeling better after manual therapy. However, research into how the SI joint moves, whether clinicians can reliably detect a “misalignment” by hand, and why manual therapy may help presents a considerably more complicated picture.

I see a large number of patients with low back pain in my clinical work, particularly in occupational healthcare. Pain around the buttock and posterior pelvic region is one of the most common patterns I encounter. My patient population naturally influences this observation: many of the people I assess have office-based jobs and describe symptoms associated with prolonged sitting and standing. This is therefore a clinical impression from a selected population, not an estimate of how common this pain pattern is in the general public.

What has increasingly caught my attention is not simply the location of the pain, but the explanation patients have already been given—or have found online. Some arrive convinced that they have “SI joint dysfunction,” a subluxation, or a joint that is somehow sitting in the wrong position. They may ask specifically for imaging of the SI joint or describe previous manipulation as having “put it back into place.” I have encountered this explanatory model not only in online health content and manual-therapy settings, but also in discussions within conventional musculoskeletal care.

From a clinical perspective, pain in the anatomical region of the SI joint does not by itself establish that the joint has shifted, become subluxated, or represents the true source of the symptoms. This distinction matters. A persuasive mechanical explanation may shape how patients understand their bodies and can encourage repeated reliance on passive treatment, even when the proposed structural fault has not been demonstrated.

That is why I wanted to examine the theory more closely. This article separates what current evidence supports about SI joint pain from what remains uncertain, speculative, or unsupported. Using systematic reviews and meta-analyses, I examine how much the SI joint actually moves, whether a positional fault can be identified reliably, what provocation tests can and cannot tell us, and whether improvement after manual therapy confirms that the joint was ever “out of place.” The aim is to offer patients and clinicians a more defensible framework for interpreting this common pain presentation—one that does not mistake the location of pain or a temporary response to treatment for proof of a specific mechanical diagnosis.

The SI Joint Barely Moves — So How Could It Be “Out of Place”?

The sacroiliac joint connects the sacrum to the ilium and transfers load between the spine and the legs. A 2021 systematic review pooling data from 13 peer-reviewed biomechanical studies found the average range of motion of an intact SI joint is 1.88° in flexion/extension, 0.85° in lateral bending, and 1.26° in axial rotation [1]. Other biomechanical reviews using three-dimensional coordinate methods report a wider spread depending on axis and study — one systematic review found rotation ranging up to roughly 8° and translation up to roughly 8 mm on some axes, while other axes showed sub-degree, sub-millimeter movement — but concluded that motion of the SI joint overall is limited to minute amounts of rotation and translation [2].

This matters because it sets the ceiling on what “out of place” could even mean biomechanically. The joint is not a hinge with several centimeters of freedom that can swing into a locked position the way a subluxated shoulder or a dislocated finger can. Whatever is happening when someone describes their SI joint as “jammed,” it is happening within an extremely narrow physiological range of motion.

This does not mean that the relief patients feel after manipulation is imaginary. Some patients genuinely report less pain, tightness, or restriction immediately afterward. I have also had my fair share of visits to chiropractors, and from my own experience, manipulation can genuinely feel helpful—at least in the short term. When that improvement follows a treatment described as “mobilizing” or “putting the joint back,” it is easy to conclude that the SI joint must previously have been displaced and was then restored to its correct position.

In my view, this is where the treatment response and the explanation for that response need to be separated. A person can feel better after manipulation without the proposed structural mechanism being correct. Given the SI joint’s very limited physiological movement, it is difficult to explain how an otherwise intact joint could repeatedly move into a clinically meaningful “lower,” “rotated,” or subluxated position and then be manually returned to place. The improvement may be real, but it does not by itself demonstrate that the joint was ever out of position. That distinction is central to understanding the evidence examined in the rest of this article.

Can Clinicians Actually Feel an SI Joint That’s Out of Place?

This is where the popular narrative runs into trouble. The core diagnostic claim behind “SI joint out of place” is that an experienced clinician can, through palpation and positional testing, detect asymmetry or restricted movement in the joint and correct it manually.

A 2021 systematic review and meta-analysis evaluated 15 different palpatory clinical tests used to assess SI joint mobility. Across the pooled studies, only one test — the sitting flexion test — achieved good, statistically significant intra-examiner reliability. No test in the review had its diagnostic validity established against any reference standard [3].

A separate systematic review from 2020, specifically designed to update older reviews with more recent data, reached a similar conclusion: most individual SI joint mobility tests showed only slight-to-fair inter-tester reliability, and confidence intervals around the reliability estimates were wide even for the better-performing tests [4]. A related systematic review focused specifically on manual palpation for low back pain assessment found that SI joint motion tests were not even associated with SI joint pain provocation tests — meaning a clinician detecting “restricted movement” by feel did not reliably correspond to the patient’s actual pain source [5].

In plain terms: the evidence does not support the idea that a clinician’s hands can reliably detect that an SI joint is out of place, let alone that two different examiners would agree on what they’re feeling.

In my clinical work, I sometimes see patients adopt an antalgic posture because of pain, with asymmetrical muscle tension making the pelvis appear tilted or rotated. It is easy to interpret this visible or palpable asymmetry as evidence that the SI joint itself has shifted or become subluxated. However, an asymmetrical posture during a painful episode may reflect protective muscle activity and altered positioning rather than structural displacement of the joint.

The same issue can arise after manipulation. If muscle tension eases and the patient stands or moves more symmetrically afterward, the pelvis may appear to have returned to a more neutral position. That visible change can easily be interpreted as the SI joint having been put “back into place.” Yet an improvement in muscle tone, posture or movement does not by itself show that the joint was structurally displaced before treatment—or that manipulation physically repositioned it.

This distinction matters because posture, muscle tone, pain and joint position are not interchangeable findings. A clinician may genuinely observe asymmetry before treatment and a visible improvement afterward without being able to conclude that the SI joint was ever “out of place.”

SI Joint Pain Is Real — But It’s Diagnosed Differently Than You Might Think

None of this means that SI joint pain is a myth. It is a recognized contributor to low back pain, with published estimates generally ranging from approximately 15% to 30% of axial low back complaints, depending on how SI joint pain is defined and identified [6]. Much of the underlying diagnostic literature has used pain relief following an image-guided intra-articular local-anesthetic injection as the reference standard. Although imperfect and not fully objective, this remains the closest available method for identifying pain arising from within the SI joint. Injection protocols and thresholds for a positive response have varied, however, and both false-positive and false-negative results are possible. An intra-articular injection may also fail to identify pain arising from ligaments or other extra-articular structures surrounding the SI joint [6]. Importantly, even a positive anesthetic response supports the joint as a pain source—it does not demonstrate that the joint was subluxated or “out of place.”

But the diagnostic approach that current evidence actually supports is not confirming the joint is out of place through palpation — it’s a cluster of pain provocation tests. The most commonly used tests are thigh thrust, compression, distraction, sacral thrust, and the Gaenslen test, with three or more positive tests conventionally used as the threshold suggesting SI joint involvement [6][7].

Even so, a 2021 systematic review and meta-analysis of these provocation test clusters found their diagnostic accuracy is limited. Assuming a 20% pretest probability of SI joint pain — the review’s own working assumption, not a universal constant — a positive cluster result gives only about 35% certainty that SI joint pain has been correctly identified. The review also reported a false-positive rate of 37.4%. Under the same 20% pretest-probability assumption, a negative cluster result corresponded to an 8% residual probability, or around 92% certainty that SI joint pain had been correctly ruled out [7]. The certainty of this evidence was rated very low using the GRADE framework [7].

Evidence summary: SI joint pain is a recognized clinical entity, but the diagnostic-accuracy evidence for provocation test clusters was rated very low certainty. These clusters are more useful for ruling SI joint pain out than for confidently ruling it in.

An important exception is suspected inflammatory sacroiliitis in axial spondyloarthritis. This is a different clinical population from patients being assessed for mechanical SI joint pain. In medical school, the Patrick–FABER test is commonly introduced as a simple bedside “screening test” for sacroiliitis and ankylosing spondylitis. In this context, the test may be more informative: one study comparing it with SI joint MRI reported a sensitivity of 76.2% and a specificity of 66.2% for MRI-detected sacroiliitis [12].

A positive result may therefore add to the clinical suspicion, but it cannot confirm axial spondyloarthritis or sacroiliitis on its own. In practice, when inflammatory sacroiliitis is genuinely suspected, SI joint MRI can become relevant relatively early to clarify the situation. The diagnosis still depends on the broader clinical presentation together with appropriate laboratory and imaging findings.

This inflammatory context should not be confused with using palpation to diagnose a mechanical SI joint “dysfunction” or positional fault. Patrick–FABER may help identify a reason to investigate possible sacroiliitis further; it does not show that the SI joint is rotated, subluxated or “out of place.”

Why Imaging Findings Need Clinical Context

A recurring argument in favor of this “lockup” framing is that MRI and X-ray often show nothing, so the pain must be coming from somewhere imaging can’t detect. Imaging findings do not always map neatly onto symptoms — but the reason why is more specific than “imaging shows nothing.”

A widely cited 2015 systematic review found that degenerative spinal findings are common even in people without back pain: disk degeneration was estimated to be present in 37% of asymptomatic 20-year-olds and 96% of asymptomatic 80-year-olds [8]. This does not mean that imaging is unhelpful or that structural findings never matter. It means that a degenerative finding should not automatically be assumed to explain a person’s pain simply because it appears on a scan. Equally, uncertainty on imaging does not validate a manually detected SI joint “misalignment” — that claim would still require reliable and valid evidence from the palpation tests used to identify it, which, per the section above, doesn’t currently exist.

In my experience, the threshold for imaging uncomplicated low back pain is relatively high in the Finnish public healthcare system. Unless the clinical picture raises concern about inflammatory sacroiliitis, serious pathology or a condition for which imaging could meaningfully change management, MRI is not usually the first step. Access may be easier in private healthcare, particularly through insurance or occupational healthcare agreements, but availability alone is not a sufficient clinical indication.

This caution is justified. Routine or premature imaging of nonspecific low back pain does not improve outcomes and may initiate a chain of further referrals and interventions based on findings that are also common in people without symptoms. In a randomized trial, rapid MRI produced clinical outcomes similar to radiography but was associated with more spinal operations and higher costs [13]. A large observational study likewise found that early MRI was associated with more back surgery, although residual confounding means that the study cannot prove imaging itself caused the operations [14].

There is also limited evidence that unnecessary imaging may adversely affect symptom perception in some patients. In one randomized trial, patients assigned to lumbar radiography were more likely to report persistent low back pain at three months, although the difference was no longer present at nine months [15]. Across randomized trials as a whole, however, routine imaging has not been shown to consistently worsen pain—it has mainly been shown to provide no clinical benefit while increasing downstream healthcare use. The concern is therefore not that the scan itself biologically causes pain, but that diagnostic labelling, patient expectations and subsequent interventions may influence how symptoms are understood and managed.

This is something I try to discuss with patients before ordering a scan: what specific clinical question are we asking, and would the result realistically change what we do next? Imaging can be valuable when it answers a relevant question. When it does not, an incidental description of “degeneration” may create anxiety and reinforce the impression that the spine is damaged, even when the finding may be broadly compatible with age and unrelated to the current pain.

Direct-access MRI adds another layer to this problem. In Finland, for example, SYNLAB currently allows patients to book an MRI without a physician’s referral. The service includes a radiologist’s written report, but when there is no referring clinician, the responsibility for arranging a clinical interpretation may effectively fall to the patient. The report is written primarily for medical use and may contain terminology that is difficult to place in context without an appropriate clinical assessment.

This can become problematic if the patient does not discuss the findings with a healthcare professional—or seeks an interpretation from someone who is not familiar with the high prevalence and limited clinical significance of many spinal imaging findings. A radiologist can describe what is visible, but the report alone cannot determine whether a finding explains the patient’s pain. That requires the imaging to be considered alongside the history, examination and original reason for performing the scan.

In my clinical experience, a poorly contextualized imaging report can sometimes make recovery more difficult by changing how the patient understands and responds to normal movement and pain. The scan itself is not the problem. The problem is imaging without a clear clinical question and interpreting common findings without appropriate clinical context.

Does Manual Therapy “Fix” the Joint — or Just Help the Pain?

Manual therapy does appear to help some patients with SI joint pain, though the picture is more mixed than it might first seem. A 2024 systematic review and meta-analysis of 16 randomized controlled trials (421 adults total, 11 meta-analyzed) evaluated manual therapy’s efficacy for SI joint pain syndrome and found that manual therapy did not produce a statistically significant reduction in pain compared to non-manual physiotherapy or sham treatment, while it did produce a statistically significant, moderate improvement in disability [9]. The certainty of the pain finding was rated very low and the disability finding low, using GRADE [9]. A separate 2024 conference abstract reporting a systematic review and meta-analysis similarly concluded that SI manual therapy with or without exercise may improve pain and disability in the short term, although the reported disability improvements were small and may not have been clinically significant [10] — a conclusion worth weighting somewhat more cautiously than ref [9] above, since a conference abstract’s methods and results can’t be scrutinized to the same depth as a full peer-reviewed article.

Note that these two reviews don’t fully agree on the pain outcome specifically — a reminder that “manual therapy helps” is not a settled, uniform finding across the literature even within 2024 alone.

What this evidence does not establish is the mechanism behind whatever benefit does occur — particularly for disability, where the effect was more consistently positive across reviews than for pain specifically. The trials themselves did not test whether any benefit arose from biomechanical, neurophysiological, contextual, or natural-history effects — that wasn’t what they were designed to measure. What they do show is that a real, measurable clinical improvement cannot, by itself, confirm that a joint was physically repositioned; the improvement and its underlying mechanism are separate questions, and only the first one has been tested here.

My own experience with manual therapy reflects this distinction. I visited a chiropractor regularly for several years, and the treatment did genuinely feel helpful. The relief was noticeable enough to keep me returning, sometimes repeatedly. Looking back, however, I can also see that the benefit was temporary. In my case, more lasting improvement came only after I shifted toward active rehabilitation, particularly isometric exercises. With hindsight, that was the part I would have introduced earlier.

This personal experience does not prove that the same sequence applies to every patient, but it influences how I now think about short-term relief. A treatment can feel effective and still fail to change the longer-term course of the problem. The immediate improvement may be precisely what makes repeated passive treatment so appealing, even when the patient has not yet developed confidence in active rehabilitation.

I encounter a similar tension in clinical practice. Some patients begin an exercise-based approach, experience a temporary increase in pain or discomfort, and conclude that movement is making the condition worse. They may then return to the passive treatment that previously provided faster relief. This does not mean that every increase in pain during rehabilitation is harmless, nor that manual therapy has no place. It means that short-term symptom relief and long-term recovery are not always the same outcome.

In my view, the potential problem is not manual therapy itself but allowing it to replace active rehabilitation indefinitely. However, I cannot generalize from my clinical observations alone that manual therapy generally reduces adherence to exercise, but it may do so when repeated passive relief reinforces the belief that recovery depends on someone else correcting the body. For some of my patients, this can become a cycle: temporary relief, recurrence, another treatment and continued uncertainty about movement. Breaking that cycle may require helping the patient understand that a transient symptom response does not necessarily indicate damage—and that gradual active rehabilitation may initially feel different from a treatment designed to provide immediate relief.

The Fascia Claim: Where Real Anatomy Ends and Speculation Begins

One of the more far-reaching claims in the popular SI joint narrative is that a “locked” joint can, via the fascial system, cause problems well outside the pelvis — neck and shoulder tightness, one-sided glute weakness, trochanteric bursitis, piriformis symptoms, and pain “all over the body.”

The anatomical starting point here is genuine. The thoracolumbar fascia (TLF) is a real, well-studied structure that connects the sacrum region to muscles extending toward the upper body, and a comprehensive anatomical review has detailed its layered structure, its connections to the latissimus dorsi and gluteal muscles, and its relevance to low back and pelvic pain [11]. The TLF also contains nociceptive nerve endings, meaning it’s a tissue genuinely capable of contributing to pain in its own right [11].

I can understand why the broader fascial explanation is compelling. In clinical practice, some patients describe a sequence in which pain begins in the lower back or buttock and is later accompanied by tightness around the shoulder on the same side—or the sequence occurs in the opposite direction. When symptoms appear to follow the same side of the body, it is easy to interpret them as evidence that the two regions are mechanically linked through the fascial system. The sequence may be entirely real, but temporal order and same-sided symptoms do not establish the mechanism connecting them. Other explanations may also be possible, and an individual clinical examination cannot determine that fascia is responsible simply from the symptom pattern.

What this anatomical continuity does not establish is that a specific SI joint positional fault—one that cannot currently be identified reliably—routinely causes distant, one-sided symptoms through the fascial system. To my knowledge, this causal chain has not been adequately tested in clinical trials, and there is no established interventional evidence showing that a verifiable SI joint fault produces symptoms in areas such as the neck or shoulder, or that “correcting” it reliably resolves them.

This is therefore better described as an unproven and inadequately tested hypothesis than as a theory that research has definitively disproved. Anatomical connectivity makes some interaction between body regions biologically conceivable, but anatomy alone cannot establish the direction, clinical significance or cause of an individual patient’s symptoms. The fact that two regions are connected does not justify concluding that an SI joint is “out of place” and causing pain elsewhere in the body.

Evidence summary: The anatomy and nociceptive innervation of the TLF are well documented. The claim that an SI joint positional fault routinely produces distant unilateral symptoms through fascial connections remains speculative

Conclusion: SI Joint Out of Place?

SI joint pain is real, but the idea that the joint routinely slips “out of place” is a much less defensible explanation. Current evidence does not show that clinicians can reliably identify a positional fault by palpation, that visible pelvic asymmetry proves subluxation, or that improvement after manipulation means the joint was physically repositioned. Manual therapy may still provide genuine short-term relief, just as anatomical connections through the thoracolumbar fascia may allow some interaction between different regions of the body—but neither observation confirms the larger mechanical story often built around it.

In my view, the most useful approach is to separate the patient’s experience from the explanation attached to it. Pain relief can be real even when the proposed mechanism is wrong, and pain near the SI joint does not automatically identify the joint as its source. Imaging, provocation tests and diagnostic injections may each add information when used for a specific clinical question, but none demonstrates that an otherwise intact SI joint is simply “out” and needs to be put back. Patients deserve an explanation that reduces fear and supports recovery rather than one that portrays the pelvis as repeatedly unstable or dependent on correction. Manual therapy can have a role, but it should not become a substitute for appropriate clinical assessment, a broader understanding of the pain, and—when suitable—an active path toward restoring confidence in movement and function.

A Simple Tool for Active Movement

The THERABAND Resistance Band Set is a simple option for readers who want versatile equipment for general strength and mobility exercises at home or while travelling. It does not diagnose SI joint pain, identify a joint “misalignment,” or replace an individual assessment when pain is persistent or concerning.

Some of the links on this site are affiliate links. If you purchase a product through these links, I may earn a commission. This doesn’t cost you anything, but it helps me maintain this page so I can continue to share free information with the world. As an Amazon Associate, I earn from qualifying purchases. Product links are provided for informational and convenience purposes only. Their inclusion does not constitute medical advice or a personalized product recommendation.

View the THERABAND Resistance Band Set on Amazon →

References

1. https://pubmed.ncbi.nlm.nih.gov/33940477/
2. https://pubmed.ncbi.nlm.nih.gov/19119382/
3. https://pubmed.ncbi.nlm.nih.gov/33896601/
4. https://www.sciencedirect.com/science/article/abs/pii/S2468781219302735
5. https://link.springer.com/article/10.1186/s12998-021-00384-3
6. https://onlinelibrary.wiley.com/doi/10.1111/papr.13338
7. https://www.jospt.org/doi/10.2519/jospt.2021.10469
8. https://pubmed.ncbi.nlm.nih.gov/25430861/
9. https://pubmed.ncbi.nlm.nih.gov/38353102/
10. https://www.sciencedirect.com/science/article/abs/pii/S0003999324007871
11. https://pubmed.ncbi.nlm.nih.gov/22630613/
12. https://pubmed.ncbi.nlm.nih.gov/37650318/
13. https://jamanetwork.com/journals/jama/fullarticle/196680
14. https://pmc.ncbi.nlm.nih.gov/articles/PMC7728897/
15. https://pubmed.ncbi.nlm.nih.gov/11179160/