Sacroiliac Joint Pain: Diagnosis, Injections & Evidence-Based Treatment

The sacroiliac joints (SIJs) connect the pelvis to the base of the spine. Together with the joint capsule and the strong posterior ligaments they form the sacroiliac joint complex, which transfers large forces between your trunk and legs. International guidance suggests the SIJ complex may account for roughly 15–30% of chronic lower-back pain felt mainly below the L5 level (McCormick et al., 2025). But pain in this area doesn't automatically mean the SIJ is the problem, which is why I treat assessment, not injection, as the first and most important step.

Understanding SIJ pain

SIJ pain isn't one single condition. It can arise from inside the joint itself, the capsule, the posterior sacroiliac ligaments, surrounding soft tissue, degenerative change, inflammatory sacroiliitis, or altered loading across the pelvis. Modern evidence suggests that much of what's traditionally called "SI joint pain" often involves the posterior SIJ complex rather than only the joint space itself (McCormick et al., 2025), which helps explain why some people respond better to treatment placed around the joint than strictly inside it.

The symptoms also overlap heavily with other problems, so I always weigh up competing sources:

• The lumbar spine
• The hip
• The gluteal region and proximal hamstring
• The surrounding pelvic ligaments
• Occasionally, peripheral nerves

How I assess sacroiliac joint pain

There's no single test, scan or symptom that definitively diagnoses SIJ pain. My assessment starts with a detailed history, where the pain sits, what aggravates it, whether it's linked to prolonged sitting, standing, walking or turning in bed, any previous lumbar/hip/pelvic problems, trauma, pregnancy-related change, and how you've responded to any previous treatment. I then examine the lumbar spine and hip as competing sources.

SIJ provocation tests

I use a cluster of established SIJ pain-provocation tests, thigh thrust, compression, distraction, sacral thrust and Gaenslen-type stress testing, which apply controlled forces across the complex to see whether they reproduce your familiar pain. Using several together is far more useful than any one manoeuvre. But I'm careful not to overstate them: a 2021 systematic review found that a positive cluster raises suspicion of SIJ pain but doesn't have the accuracy to definitively confirm the SIJ as the pain generator (Saueressig et al., 2021). So I use the examination to estimate the likelihood that the SIJ complex is involved, not to claim certainty.

Treatment: conservative care first

Initial management is usually conservative, activity modification, appropriate pain relief, and physiotherapy focused on hip and gluteal strengthening, lumbopelvic control, progressive loading and a graded return to normal activity. Many people improve at this stage. When pain persists despite appropriate rehabilitation, an image-guided injection becomes useful, and I don't see it merely as a way to dull pain temporarily. It's part of a combined diagnostic and treatment pathway.

Corticosteroid injection

Corticosteroid is the most extensively studied injectable treatment for SIJ pain. The aim isn't to "reposition" the pelvis, it's to reduce local inflammation and pain so you can walk, sit, sleep, exercise and progress with rehab more comfortably.

The most useful recent evidence is the 2024 systematic review and meta-analysis by Ruffilli and colleagues, covering 43 studies and 2,431 patients (63% corticosteroid, 16% PRP). Corticosteroid injections gave an average pain improvement of about 3.4 points on the VAS at mid-term and 3.0 points at longer-term follow-up, both statistically significant, with a failure rate around 26%. The authors concluded corticosteroid remains the best-documented SIJ injectable, effective and safe overall, while being honest that a meaningful minority don't respond (Ruffilli et al., 2024).

How long does it last?

There's no single answer, some people get a few weeks, others several months. International consensus supports short-term benefit from intra- and extra-articular corticosteroid in well-selected patients (McCormick et al., 2025). The point is often not to permanently remove the problem, but to open a window of reduced pain in which strengthening and progressive loading become possible.

The injection also helps clarify the diagnosis

This is the part I find most valuable. The injection includes local anaesthetic, which works far faster than the steroid. If your familiar pain drops substantially straight after an accurately targeted injection, that raises confidence the treated structures are genuinely relevant. So the pathway isn't simply "pain → injection". I think of it as:

Consensus guidance reports that the threshold most commonly used for a "positive" diagnostic or prognostic block is around 50% pain reduction; higher thresholds haven't reliably improved the outcome of later procedures (McCormick et al., 2025). It isn't a perfect test, but when history, examination and the response to targeted anaesthetic all point the same way, confidence in the diagnosis becomes much stronger.

If relief is good but temporary: sacral lateral branch radiofrequency

If you have convincing SIJ pain, several positive provocation tests, a clear response to injection, but the pain gradually returns, repeating steroid injections indefinitely usually isn't the best long-term strategy. That good-but-temporary response is actually very useful: it can support moving to sacral lateral branch radiofrequency. Rather than injecting anti-inflammatory medication, radiofrequency targets the small sensory nerve branches that carry pain from the posterior SIJ complex, reducing their ability to transmit pain signals.

For appropriately selected patients with posterior-complex pain, the evidence is increasingly strong: the 2025 international multispecialty guideline concluded there is strong evidence that sacral lateral branch radiofrequency provides pain relief for at least six months in suitable patients with extra-articular SIJ pathology (McCormick et al., 2025), and a network meta-analysis identified cooled radiofrequency among the more effective interventional options (Park et al., 2026). Here, the injection has done more than reduce symptoms, the response has helped guide the next stage of treatment.

What about PRP?

Platelet-rich plasma (PRP), made from your own blood, has attracted interest as an alternative to steroid, intended to influence tissue healing and inflammatory signalling rather than simply suppress inflammation. But the SIJ evidence is inconsistent. In the Ruffilli review, PRP produced significant improvements too (about 2.2 VAS points mid-term and 2.3 long-term), but only 16% of the included studies looked at PRP versus 63% for corticosteroid, so the authors called it promising while unable to determine the optimal injectate (Ruffilli et al., 2024). Consensus guidance likewise rates the PRP evidence as considerably weaker than for corticosteroid or radiofrequency (McCormick et al., 2025). Individual trials have split both ways, some favouring PRP, others steroid (Chen et al., 2022; Singla et al., 2017; Burnham et al., 2020). For most people needing an initial injectable, corticosteroid therefore remains the more established option. I'd currently regard PRP as an emerging alternative, not a proven superior treatment.

Corticosteroid vs PRP vs radiofrequency

Does the injection need imaging guidance?

Yes. The SIJ is deep and anatomically irregular, so accurate treatment by feel alone is difficult. For a true intra-articular SIJ injection, fluoroscopy (X-ray) with contrast remains the most established way to prove needle placement inside the joint. Ultrasound has different strengths: real-time views of the posterior SIJ landmarks, ligaments, the needle, surrounding muscle and blood vessels, and no radiation. Current guidance doesn't yet consider ultrasound proven non-inferior to fluoroscopy specifically for confirming purely intra-articular placement (McCormick et al., 2025; Cohen et al., 2019). For that reason I prefer the clinically broader term ultrasound-guided SIJ complex injection, which also fits the growing recognition that clinically relevant SIJ pain often comes from the posterior ligamentous, extra-articular structures.

My usual SIJ treatment pathway

1. Detailed clinical assessment, is this genuinely compatible with SIJ complex pain? The lumbar spine, hip and other buttock-pain causes are considered.
2. SIJ provocation testing, several specific tests, to see whether loading the complex consistently reproduces your familiar pain.
3. Image-guided SIJ complex injection, if the picture strongly suggests SIJ pain and conservative care hasn't worked. It has both a therapeutic role (reducing pain/inflammation) and a diagnostic role (the immediate anaesthetic response).
4. Reassess the response, none, partial, substantial immediate relief, weeks of benefit or months? The pattern guides what happens next.
5. Rehabilitation, if pain drops, that's the window to progress gluteal strengthening, trunk/pelvic control, loading and walking tolerance.
6. Consider radiofrequency if relief is good but temporary, sacral lateral branch radiofrequency may give more durable control than repeated steroid injections (McCormick et al., 2025).

When I consider an injection

I consider an SIJ injection when the symptoms fit SIJ complex pain, provocation tests support it, hip and lumbar causes have been weighed up, symptoms significantly limit daily life or rehab, conservative treatment hasn't done enough, and the expected benefit outweighs the risks. It's especially useful when pain is severe enough to block meaningful rehabilitation.

What if the injection doesn't work?

A poor response is also useful information. If accurately targeted SIJ treatment barely changes your familiar pain, that lowers the likelihood the SIJ complex is the dominant source, and I'd reconsider other possibilities (lumbar spine, hip, gluteal or proximal-hamstring problems, nerve-related pain) rather than automatically repeating the procedure or escalating to something more invasive.

Frequently asked questions

References