A normal sacroiliac joint X-ray shows two smooth, well-defined joint margins with a visible joint space roughly 2 to 4 millimeters wide, bordered by a thin band of dense bone on each side. An abnormal X-ray may show blurred margins, irregular erosions, excessive sclerosis, widening or narrowing of the joint space, or in advanced cases, complete fusion. The challenge is that a surprising number of findings fall in a gray zone between “clearly normal” and “clearly pathological,” especially because the sacroiliac joint changes substantially with age, physical stress, and pregnancy, and plain radiographs are not particularly good at capturing early disease.
What a Normal Sacroiliac Joint Looks Like on X-ray
On a standard anteroposterior (AP) pelvic X-ray, each sacroiliac joint appears as a narrow, slightly irregular line where the sacrum meets the ilium. In younger adults, the joint space is relatively uniform and measures around 2.3 mm wide on average, with a thin rim of subchondral sclerosis (dense bone) on both the iliac and sacral sides. That iliac-side sclerosis is typically a bit thicker than the sacral side, partly because the cartilage covering the iliac surface is thinner to begin with. A small amount of sclerosis in the upper-front portion of the joint is considered a normal variant and does not indicate disease.
The joint itself has two distinct parts: a lower portion that is a true synovial joint (lined with cartilage, capable of producing inflammatory fluid) and an upper portion that is more of a ligamentous connection. X-rays mostly show the synovial portion, and even in healthy people, the two sides are not always perfectly symmetrical. Slight differences in joint-space width between your left and right sacroiliac joints can be normal, particularly after age 30.
How Age Changes the Picture
The sacroiliac joint is not static. It remodels throughout life, and what counts as “normal” at age 25 looks different from what is normal at age 60. In people over 40, the average joint space narrows to around 1.9 mm and becomes less uniform, while the sclerotic bands on both sides widen and become more irregular.1PubMed. Variation in the appearance of the normal sacroiliac joint on pelvic CT These changes alone can make an aging joint look suspicious on X-ray if you are not expecting them.
A large study of CT scans in people without sacroiliac disease found that nearly half of all subjects showed some degree of sclerosis, and a similar proportion had osteophytes (bony spurs) at the joint margins. Both findings increased steadily with age, peaking above 75 years, where more than three-quarters of people had osteophytes. Joint-space alterations were rare before age 35 but present in about a quarter of those over 75.2Scientific Reports. Impact of age, sex, and joint form on degenerative lesions of the sacroiliac joints on CT in the normal population In other words, a degree of sclerosis and spurring at the sacroiliac joint is the rule in older adults, not the exception.
Actual bony fusion of the joint is also part of normal aging, although its frequency is debated. Radiographic studies of cadaver specimens found evidence of some degree of synostosis (bony bridging) in roughly 60 percent of joints when examined closely, with men fusing more often than women.3Clinical Anatomy. Age changes in the human sacroiliac joint: Joint fusion On a standard AP pelvis X-ray, though, clear-cut fusion is visible far less often because overlapping anatomy hides what thin-slice imaging would reveal. The broader point matters: partial fusion in a 70-year-old is a sign of aging, not necessarily a sign of ankylosing spondylitis.
Because the sacroiliac joint varies so much between individuals, between the left and right sides of the same person, and across decades of life, a review in 2024 emphasized that knowledge of the wide range of normal findings is essential to avoid labeling healthy joints as diseased.4PubMed Central. The sacroiliac joint across ages – what is normal?
The Modified New York Grading System
When radiologists and rheumatologists evaluate sacroiliac joint X-rays for disease, they typically use a 0-to-4 grading scale that has been around since the 1960s and was formalized in the modified New York criteria. The grades break down as follows:
- Grade 0: Normal joint. Clear margins, uniform joint space, no suspicious changes.
- Grade 1: Suspicious changes. Slight blurring of the joint margins that might be early disease or might be nothing.
- Grade 2: Definite mild abnormality. Small erosions, some sclerosis, minor joint-space changes, but the joint is still clearly open.
- Grade 3: Definite advanced abnormality. Severe erosions, marked sclerosis, and joint-space widening or narrowing, possibly with partial fusion.
- Grade 4: Complete ankylosis. The joint has fused entirely and the joint space is no longer visible.
For a diagnosis of ankylosing spondylitis under these criteria, you need at least bilateral grade 2 changes or unilateral grade 3 changes.5PubMed Central. Diagnostics of Sacroiliitis According to ASAS Criteria: A Comparative Evaluation of Conventional Radiographs and MRI in Patients with a Clinical Suspicion of Spondyloarthropathy Grade 1 is deliberately excluded from diagnostic thresholds because it is too unreliable. Even experienced radiologists often disagree on whether a joint is grade 1 or grade 0.
What Abnormal Sacroiliitis Actually Looks Like on X-ray
Inflammatory sacroiliitis from conditions like ankylosing spondylitis follows a recognizable progression on X-ray. The earliest visible change is usually erosion of the iliac side of the joint. Because the cartilage on the iliac surface is thinner, it is more vulnerable to inflammatory damage. As erosions grow and merge, they can create the paradoxical appearance of a widened joint space, even though the cartilage is being destroyed. After that, the body tries to repair the damage by laying down new bone, which shows up as sclerosis. Eventually, the joint space narrows as bone grows across it, and in the final stage, the joint fuses completely into a solid bridge of bone.6European Society of Radiology. Sacroiliitis – Findings on Conventional Radiography, CT and MRI studies
The inflammatory pattern tends to be bilateral and roughly symmetrical, which is one of the features that separates it from infection or mechanical stress. The sclerosis involves both the iliac and sacral sides of the joint, and erosions create an irregular, “moth-eaten” appearance of the subchondral bone. In advanced disease, the so-called “shiny corners” and bridging syndesmophytes along the spine often accompany the sacroiliac fusion, forming the classic “bamboo spine” of ankylosing spondylitis.
Conditions That Mimic Sacroiliitis on X-ray
One of the trickiest parts of reading sacroiliac joint X-rays is that several benign or unrelated conditions produce changes that look disturbingly similar to inflammatory disease. Getting these wrong in either direction has real consequences: a false-positive sacroiliitis diagnosis can lead to years of unnecessary immunosuppressive treatment, while a missed diagnosis delays care.
Osteitis Condensans Ilii
Osteitis condensans ilii (OCI) is a condition marked by dense, triangular sclerosis on the iliac side of the sacroiliac joint, most often in women who have been pregnant. On X-ray, the sclerosis can look quite dramatic, sometimes extending more than 10 mm into the bone. The key distinguishing feature is that OCI affects only the iliac side. The sacral surface and the joint space itself remain intact, with no erosions and no progressive narrowing.7PubMed Central. Osteitis condensans ilii masquerading as sacroiliitis: a case report of an underdiagnosed condition In contrast, true sacroiliitis tends to involve both sides of the joint and is accompanied by erosions and structural damage that worsens over time. OCI is considered a response to mechanical stress rather than inflammation, and it often requires no treatment.
Degenerative Changes
Osteoarthritis of the sacroiliac joint is extremely common in older adults and produces osteophytes, subchondral sclerosis, and joint-space narrowing that can overlap with the appearance of inflammatory sacroiliitis. The main differentiator is location: degenerative changes favor the anterior and superior portions of the joint and tend to include osteophytes (bony spurs projecting from the joint margins), which are not a feature of inflammatory sacroiliitis. Transitional vertebrae at the lumbosacral junction, where the lowest lumbar vertebra is partially fused to the sacrum, can alter the mechanical load on the sacroiliac joint and accelerate degenerative changes on one side, producing an asymmetric pattern that can further confuse the picture.
Pregnancy-Related Changes
Pregnancy places enormous mechanical stress on the sacroiliac joints. Hormonal changes loosen the ligaments, and the altered biomechanics of carrying and delivering a baby leave lasting marks on the joint. On MRI, bone marrow edema around the sacroiliac joint was found in roughly 69 to 80 percent of women at three months postpartum, and it was still present in over half at 12 months after delivery. Strikingly, about 41 percent of these healthy postpartum women met the formal imaging criteria for sacroiliitis at 12 months.8PubMed. Evolution of Magnetic Resonance Imaging Lesions at the Sacroiliac Joints During and After Pregnancy by Serial Magnetic Resonance Imaging From Gestational Week Twenty to Twelve Months Postpartum On X-ray and MRI alike, iliac-side sclerosis increased gradually during pregnancy and continued climbing postpartum, reaching about 46 percent prevalence at 12 months, with expanding depth over time.9PubMed. Evolution of osteitis condensans ilii-like sclerosis in first-time mothers by serial sacroiliac joint MRI during pregnancy and in the postpartum period These changes clustered in the anterior and inferior portions of the joint and were not correlated with pain.10PubMed. Patterns and prevalence of sacroiliac joint bone marrow edema during pregnancy: A retrospective analysis
The clinical takeaway is stark: evaluating sacroiliac joint imaging in women who have been pregnant within the past year or two requires extra caution. What looks like active sacroiliitis may be the joint’s normal response to the physical demands of pregnancy and childbirth.
When Sacroiliitis Is One-Sided
Inflammatory sacroiliitis from spondyloarthritis is usually bilateral, but it can be unilateral, especially early in the disease. Infection of the sacroiliac joint (septic sacroiliitis) is also typically one-sided. Telling them apart matters because the treatments are completely different: antibiotics versus immunosuppressants.
On imaging, certain features help distinguish the two. Infectious sacroiliitis tends to produce thick capsular swelling, fluid collections extending beyond the joint capsule, and swelling of the nearby piriformis and iliac muscles. In one study, fat infiltration in front of the joint was seen in 83 percent of septic sacroiliitis cases, and muscle swelling on both sides appeared in 71 percent.11PubMed Central. Analysis of the early signs of septic sacroiliitis on computed tomography Spondyloarthritis, by contrast, tends to show bone marrow edema predominantly on the iliac side without the dramatic soft-tissue involvement. When periarticular muscle edema was used as the sole distinguishing criterion, infectious sacroiliitis was correctly identified about 91 percent of the time, and spondyloarthritis about 77 percent of the time.12PubMed. Unilateral Sacroiliitis: Differential Diagnosis Between Infectious Sacroiliitis and Spondyloarthritis Based on MRI Findings Plain X-rays, unfortunately, capture soft-tissue changes poorly, which is one reason infection is easy to miss on radiographs alone.
Why Plain X-rays Often Are Not Enough
Plain radiographs have been the default imaging tool for sacroiliac joints for decades, but their limitations are well documented. X-rays cannot detect active inflammation at all. They show only the structural aftermath of disease: erosions, sclerosis, and fusion. By the time these changes are visible on a standard radiograph, the disease has often been present for years. One comparison found that conventional radiography had a sensitivity of only about 19 percent for detecting active sacroiliitis, compared to 95 percent for contrast-enhanced MRI.13PubMed. Magnetic resonance imaging (MRI) for detection of active sacroiliitis–a prospective study comparing conventional radiography, scintigraphy, and contrast enhanced MRI In a study of patients suspected of spondyloarthritis, among those whose X-rays showed no sacroiliitis, about 40 percent turned out to have active disease on MRI.5PubMed Central. Diagnostics of Sacroiliitis According to ASAS Criteria: A Comparative Evaluation of Conventional Radiographs and MRI in Patients with a Clinical Suspicion of Spondyloarthropathy
The sacroiliac joint is also difficult to image with plain films because of overlying bowel gas, the complex three-dimensional shape of the joint, and the oblique angle at which it sits in the pelvis. A dedicated angled view called the Ferguson view was long thought to be superior to a standard AP pelvis, but comparative studies have found no clear advantage of one projection over the other.14PubMed. Analysis of dedicated sacroiliac views to improve reliability of conventional pelvic radiographs The fundamental problem is the modality itself, not the angle.
X-rays remain useful as a screening tool and for monitoring long-term structural progression in patients with an established diagnosis. They are cheap, widely available, and involve a low radiation dose. But for initial diagnosis, especially when the clinical suspicion is high and the X-ray looks normal, MRI has become the preferred follow-up.15The Journal of Rheumatology. Unmet Needs in Spondyloarthritis: Imaging in Axial Spondyloarthritis
How MRI and CT Compare to X-ray
MRI’s main advantage is its ability to see active inflammation, specifically bone marrow edema, which is the hallmark of early sacroiliitis before any structural damage has occurred. It also picks up erosions and joint-space changes with much greater sensitivity than X-rays. In one head-to-head comparison using low-dose CT as the reference standard, MRI detected erosions correctly in about 79 percent of cases compared to 42 percent for X-rays, and detected overall structural abnormalities in 85 percent versus 48 percent. The one area where X-rays held their own was sclerosis, where radiography actually outperformed MRI (70 percent versus 30 percent sensitivity), though the difference did not reach statistical significance.16Annals of the Rheumatic Diseases. Comparison of MRI with radiography for detecting structural lesions of the sacroiliac joint using CT as standard of reference: results from the SIMACT study This makes some intuitive sense: dense bone lights up brightly on X-ray, while MRI is better at spotting subtle bone and soft-tissue changes.
Low-dose CT has been gaining ground as a middle option. It provides excellent structural detail of the joint, clearly showing erosions, sclerosis, joint-space narrowing, and osteophytes without the overlapping anatomy that plagues plain films. Newer protocols deliver radiation doses comparable to or even lower than a standard pelvic X-ray.17PubMed Central. Future of Low-Dose Computed Tomography and Dual-Energy Computed Tomography in Axial Spondyloarthritis One study found low-dose CT vastly outperformed X-rays in soft-tissue delineation, erosion detection, and freedom from overlapping structures.18International Journal of Research in Medical Sciences. Comparison of conventional radiograph with low-dose computed tomography of sacroiliac joints in patients with proven sacroiliitis CT still cannot detect inflammation the way MRI can, but for structural assessment in patients who cannot undergo MRI, it is a strong alternative.
Sacral Stress Fractures and the Diagnostic Overlap
Insufficiency fractures of the sacrum are another source of confusion. These fractures happen when normal forces act on weakened bone, typically in older adults with osteoporosis or in athletes with high-impact training loads. On MRI, the bone marrow edema surrounding a sacral stress fracture can look nearly identical to the edema of active sacroiliitis, creating a genuine diagnostic dilemma. X-rays are frequently negative in the early stages of insufficiency fractures, meaning a patient with sacral pain, a normal X-ray, and then an MRI showing bone edema near the sacroiliac joint could plausibly be given either diagnosis. The fracture line itself, when visible on MRI or CT, is the distinguishing feature, but it is not always apparent on initial imaging. Clinical context, particularly a history of osteoporosis, recent changes in physical activity, or use of corticosteroids, is often what tips the scale.
Artificial Intelligence in Sacroiliac Joint Imaging
Reading sacroiliac joint X-rays is subjective enough that even trained experts disagree on grading, particularly for the ambiguous grade 1 and grade 2 categories. This has led to growing interest in using deep learning algorithms to assist with interpretation. Several research groups have trained convolutional neural networks on sets of normal and abnormal pelvic radiographs, aiming to develop tools that can flag likely sacroiliitis automatically.19PubMed Central. Deep learning for detection of radiographic sacroiliitis: achieving expert-level performance Early results have been promising, with some models approaching expert-level accuracy in distinguishing definite sacroiliitis from normal joints.20PubMed. Deep learning methods in the diagnosis of sacroiliitis from plain pelvic radiographs
These tools are still in the research phase and not yet part of routine clinical workflow. One recent study tested ChatGPT’s ability to grade sacroiliitis on pelvic X-rays using the modified New York criteria, with mixed results, suggesting that general-purpose AI models are not yet reliable for this specific task.21PubMed Central. Evaluation of the Diagnostic Performance of ChatGPT in Radiographic Staging of Sacroiliitis According to the Modified New York Criteria Purpose-built image analysis networks trained specifically on sacroiliac radiographs show more promise, but their real-world utility will depend on how well they handle the gray-zone cases that give human readers the most trouble. If these tools eventually make it into clinical practice, they could help standardize grading and reduce the substantial interobserver disagreement that currently exists, particularly in settings without access to musculoskeletal radiology specialists.